View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

TREAS.
HJ
10
.A13P4
v.313

U.S. Department of the Treasury

PRESS RELEASES

fJ while II(T~~ ~d SAeef jJde Ir/f~
ht.{j ~ lite btf ilul'l1 /'1 Jer;{, C tjjJt!~ of(,
IAntLbl-e, {b f---tf/~e..e

FOR IMMEDIATE RELEASE
January 3, 1992

CONTACT: Claire Buchan
(202)566-8773

statement of Secretary Brady
on
Membership of the New States of the Former Soviet Union
in the IMF and World Bank

The dramatic developments in the former Soviet union have
created new opportunities and challenges for
international
financial
cooperation.
The United States supports early
consideration by the IMF and World Bank of membership for new
states of the former Soviet Union with whom we are establishing
diplomatic relations
(Russia,
Ukraine,
Kazakhstan,
Belarus,
Kyrgyzstan, Armenia).
Membership in the IMF and World Bank will further marketoriented economic reform in these newly independent nations.
We
will work with them to ensure that their applications are
considered as quickly as possible.
We are also prepared to
consider the membership of the other new states of the former
Soviet Union once diplomatic relations are established with them
(Azerbaijan, Turkmenistan, Uzbekistan, Tajikistan, Georgia, and
Moldova) .
The benefits of technical assistance and expertise provided by
the IMF and World Bank, pending full membership, should continue to
be available to all twelve states of the former Soviet Union.

000

NB-1606

FOR RELEASE AT 2:30 P.M.
January 3, 1992

CONTACT:

8ffice of Financing
202-219-3350

TREASURY'S 52-WEEK BILL OFFERING
The Department of the Treasury, by this public notice,
invites tenders for approximately $12,500 million of 364-day
Treasury bills to be dated January 16, 1992
and to mature
January 14, 1993
(CUSIP No. 912794 ZZ 0). This issue will
provide about $ 700
million of new cash for the Treasury,
as the maturing 52-week bill is outstanding in the amount of
$ 11,803 million. Tenders will be received at Federal Reserve
Banks and Branches and at the Bureau of the Public Debt, Washington, D. C. 20239-1500,
Thursday, January 9, 1992,
prior to
12:00 noon for noncompetitive tenders and prior to 1:00 p.m.,
Eastern
Standard
time, for competitive tenders.
The bills will be issued on a discount basis under ~ompet~­
tive and noncompetitive bidding, and at maturity their ~2C ~illuunt
will be payable without interest. This series of bills wlll be
issued entirely in book-entry form in a minimum amount of S10,000
and in any higher $5,000 multiple, on the records either of the
Federal Reserve Banks and Branches, or of the Department of the
Treasury.
The bills will be issued for cash and in exchange for
Treasury bills maturing
January 16, 1992.
In addition to the
maturing 52-week bills, there are $ 22,069 million of maturing
bills which were originally issued as 13-week and 26-week bills.
The disposition of this latter amount will be announced next
week.
Federal Reserve Banks currently hold $ 977
million as
agents for foreign and international monetary authorities, and
$7,885 million for their own account. These amounts represent
the combined holdings of such accounts for the three issues of
maturing billS. Tenders from Federal Reserve Banks for their
own account and as agents for foreign and international monetary authorities will be accepted at the weighted average bank
discount rate of accepted competitive tenders. Additional
amounts of the bills may be issued to Federal Reserve Banks,
as agents for foreign and international monetary authorities,
to the extent that the aggregate amount of tenders for such
accounts exceeds the aggregate amount of maturing bills held
by them.
For purposes of determining such additional amounts,
foreign and international monetary authorities are considered to
hold S 130
million of the original 52-week issue. Tenders for
bills to be maintained on the book-entry records of the Department of the Treasury should be submitted on Form PD 5176-3.
NB-j60]

TREASURY'S 13-, 26-, AND 52-WEEK BILL OFFERINGS, Page 2
Each tender must state the par amount of bills bid for,
which must be a minimum of $10,000.
Tenders over $10,000 must
be in multiples of $5,000.
Competitive tenders must also show
the yield desired, expressed on a bank discount rate basis with
two decimals, e.g., 7.15%.
Fractions may not be used. A single
bidder, as defined in Treasury's single bidder guidelines, shall
not submit noncompetitive tenders totaling more than $1,000,000.
The following institutions may submit tenders for accounts
of customers if the names of the customers and the amount for
each customer are furnished:
depository institutions, as
described in Section 19(b)(1)(A), excluding those institutions
described in subparagraph (vii), of the Federal Reserve Act
(12 U.S.C. 461(b»; and government securities broker/dealers
registered with the Securities and Exchange Commission that are
registered or noticed as government securities broker/dealers
pursuant to Section 15C(a)(1) of the Securities and Exchange
Act of 1934, as amended by the Government Securities Act of
1986. Others are only permitted to submit tenders for their
own account.
Each tender must state the amount of any net long
position in the bills being offered if such position is in excess
of $200 million.
This information should reflect positions held
as of one-half hour prior to the closing time for receipt of competitive tenders on the day of the auction.
Such positions would
include bills acquired through "when issued" trading, and futures
and forward contracts as well as holdings of outstanding bills
with the same CUSIP number as the new offering.
Those who submit
tenders for the accounts of customers must submit a separate
tender for each customer whose net long position in the bill
being offered exceeds $200 million.
A noncompetitive bidder may not have entered into an
agreement, nor make an agreement to purchase or sell or otherwise dispose of any noncompetitive awards of this issue being
auctioned prior to the designated closing time for receipt of
competitive tenders.
Tenders from bidders who are making payment by charge
to a funds account at a Federal Reserve Bank and tenders from
bidders who have an approved autocharge agreement on file at a
Federal Reserve Bank will be received without deposit.
Tenders
from all others must be accompanied by full payment for the
amount of bills applied for.
A cash adjustment will be made
on all accepted tenders, accompanied by payment in full, for
the difference between the par payment submitted and the actual
issue price as determined in the auction.
11/5/91

TREASURY'S 13-, 26-, AND 52-WEEK BILL OFFERINGS, Page 3
Public announcement will be made by the Department of the
Treasury of the amount and yield range of accepted bids. Competitive bidders will be advised of the acceptance or rejection
of their tenders. The Secretary of the Treasury expressly
reserves the right to accept ·or reject any or all tenders, in
whole or in part, and the Secretary's action shall be final.
Subject to these reservations, noncompetitive tenders for each
issue for $1,000,000 or less without stated yield from anyone
bidder will be accepted in full at the weighted average bank
discount rate (in two decimals) of accepted competitive bids
for the respective issues. The calculation of purchase prices
for accepted bids will be carried to three decimal places on
the basis of price per hundred, e.g., 99.923, and the determinations of the Secretary of the Treasury shall be final.
Settlement for accepted tenders for bills to be maintained
on the book-entry records of Federal Reserve Banks and Branches
must be made or completed at the Federal Reserve Bank or Branch
by the issue date, by a charge to a funds account or pursuant to
an approved autocharge agreement, in cash or other immediatelyavailable funds, or in definitive Treasury securities maturing
on or before the settlement date but which are not overdue as
defined in the general regulations governing United States
securities. Cash adjustments will be made for differences
between the par value of the maturing definitive securities
accepted in exchange and the issue price of the new bills.
Department of the Treasury Circulars, Public Debt Series Nos. 26-76, 27-76, and 2-86, as applicable, Treasury's single
bidder guidelines, and this notice prescribe the terms of these
Treasury bills and govern the conditions of their issue. Copies
of the circulars, guidelines, and tender forms may be obtained
from any Federal Reserve Bank or Branch, or from the Bureau
of the Public Debt.
11/5/91

UBLIC DEBT NEWS
Department of the Treasury •

Bureau of the Public Debt • Washington, DC 20239

FOR IMMEDIATE RELEASE
January 6, 1992

CONTACT: Office of Financing
202-219-3350

RESULTS OF TREASURY'S AUCTION OF 13-WEEK BILLS
Tenders for $10,262 million of 13-week bills to be issued.
January 9, 1992 and to mature April 9, 1992 were
accepted today (CUSIP: 912794YH1).
RANGE OF ACCEPTED
COMPETITIVE BIDS:
Low
Higl1
Average

Discount
Rate
3.84%
3.85%
3.85%

Investment
Rate
3.94%
3.95%
3.95%

Price
99.029
99.027
99.027

Tenders at the high discount rate were allotted 32%.
The investment rate is the equivalent coupon-issue yield.
TENDERS RECEIVED AND ACCEPTED (in thousands)
Location
Boston
New York
Philadelphia
Cleveland
Richmond
Atlanta
Chicago
st. Louis
Minneapolis
Kansas city
Dallas
San Francisco
Treasury
TOTALS

Received
23,440
40,210,100
38,715
49,245
76,570
37,740
1,054,295
52,940
7,690
26,745
2.6,070
718,730
1,026,745
$43,349,025

Acce12ted
23,440
8,883,110
38,590
37,235
45,855
34,740
45,700
12,940
7,690
26,745
26,070
53,030
1,026,745
$10,261,890

Type
Competitive
Noncompetitive
subtotal, Public

$38,821,115
1,675,440
$40,496,555

$5,733,980
1,675,440
$7,409,420

2,693,230

2,693,230

159,240
$43,349,025

159,240
$10,261,890

Federal Reserve
Foreign Official
Institutions
TOTALS

An additional $165,160 thousand of bills will be
issued to foreign official institutions for new cash.
N8-1-608

UBLIC DEBT NEWS
Department of the Treasury •

Bureau of the Public Debt • Washington, DC 20239

FOR IMMEDIATE RELEASE
January 6, 1992

CONTACT: Office of Financing
202-219-3350

RESULTS OF TREASURY'S AUCTION OF 26-WEEK BILLS
Tenders for $10,308 million of 26-week bills to be issued
January 9, 1992 and to mature July 9, 1992 were
accepted today (CUSIP: 912794ZC1).
RANGE OF ACCEPTED
COMPETITIVE BIDS:
Low
High
Average

Discount
Rate
3.85%
3.86%
3.86%

Investment
__Rate
3.99%
4.00%
4.00%

Price
98.054
98.049
98.049

Tenders at the high discount rate were allotted 42%.
The investment rate is the equivalent coupon-issue yield.
TENDERS RECEIVED AND ACCEPTED (in thousands)
Location
Boston
New York
Philadelphia
Cleveland
Richmond
Atlanta
Chicago
st. Louis
Minneapolis
Kansas City
Dallas
San Francisco
Treasury
TOTALS

Received
25,580
44,774,310
11,715
27,955
546,760
20,325
1,082,985
34,395
8,100
29,280
17,380
601,320
742,565
$47,922,670

Accel2ted
25,580
9,287,725
11,715
27,955
30,960
19,325
58,705
12,395
8,100
29,280
17,380
36,320
742,565
$10,308,005

Type
Competitive
Noncompetitive
Subtotal, Public

$43,960,280
1,140,830
$45,101,110

$6,345,615
1,140,830
$7,486,445

2,450,000

2,450,000

371,560
$47,922,670

371,560
$10,308,005

Federal Reserve
Foreign Offi~ial
Institutions
TOTALS

An additional $390,840 thousand of bills will be
issued to foreign official institutions for new cash.
NB-1609

Report of

THE DEPARTMENT OF THE TREASURY

on

Integration
of
The Individual and Corporate
Tax Systems
Taxing Business Income Once

January 1992

For sale by the U.s. Government Prmting Office
Superintendent of Documents, Mail Stop: SSOP, Washington, DC 20C(02·lJ.12H

ISBN 0-16-036045-5

DEPARTMENT OF THE TREASURY
WASHINGTON

January 1992

ASSISTANT SECRETARY

The Honorable Dan Rostenkowski
Chairman
Committee on Ways and Means
United States House of Representatives
Washington, D.C. 20515
Dear Mr. Chairman:
Section 634 of Public Law 99-514, the Tax Reform Act of 1986 directed the
Secretary of the Treasury or his delegate to study reforms of the taxation of corporate income
under Subchapter C of the Internal Revenue Code. This mandate is quite broad. We concluded
that a comprehensive study of the issues presented by integration of the corporate and individual
income tax would address fundamental questions concerning how the corporate income tax might
be restructured to reduce tax distortions of important corporate financial decisions and to achieve
a more efficient system. Given the prevalence of integrated corporate income tax systems in the
world today, we believe that an examination of these substantial issues should precede
consideration of other, less fundamental, approaches to corporate income tax reform.
Accordingly, this Report is submitted pursuant to the statutory directive cited above.
I am sending a similar letter to Representative Bill Archer.
Sincerely,

1~:net:: :~~
Assistant Secretary
(Tax Policy)

"( !:t. .
8
\

n---1)

~----

DEPARTMENT OF THE TREASURY

·1

WASHINGTON

January 1992

/

ASSISTANT SECRETARY

The Honorable Lloyd Bentsen
Chairman
Committee on Finance
United States Senate
Washington, D.C. 20515
Dear Mr. Chairman:
Section 634 of Public Law 99-514, the Tax Reform Act of 1986 directed the
Secretary of the Treasury or his delegate to study reforms of the taxation of corporate income
under Subchapter C of the Internal Revenue Code. This mandate is quite broad. We concluded
that a comprehensive study of the issues presented by integration of the corporate and individual
income tax would address fundamental questions concerning how the corporate income tax might
be restructured to reduce tax distortions of important corporate financial decisions and to achieve
a more efficient system. Given the prevalence of integrated corporate income tax systems in the
world today, we believe that an examination of these substantial issues should precede
consideration of other, less fundamental, approaches to corporate income tax reform.
Accordingly, this Report is submitted pursuant to the statutory directive cited above.
I am sending a similar letter to Senator Bob Packwood.
Sincerely,

7f~0.~
Kenneth W. Gideon
Assistant Secretary
(Tax Policy)

PREFACE
The so-called classical system of current U.S. tax law treats corporations
and their investors as separate entities and levies tax at both the corporate and
shareholder levels on earnings from investments in corporate equity. Corporate
earnings distributed to lenders as interest are generally deductible by the
corporation and taxed, if at all, to the lender. Investors who conduct business
activity in noncorporate fonn, such as a sole proprietorship or partnership, are
taxed once on their earnings at the owners' tax rate.
As a result, despite the critical role played by corporations as a vehicle for
economic growth, the United States tax law often perversely penalizes the
corporate fonn of organization. The current system of taxation also distorts
corporate fmancial decisions-in particular by encouraging debt and discouraging
new equity fmancing of corporate investments. The tax system also prejudices
corporate decisions about whether to retain earnings or pay dividends and
encourages corporations to distribute earnings in a manner to avoid the doublelevel tax.
Integration of the individual and corporate tax system would tax corporate
income once and reduce or eliminate these economic distortions. Most trading
partners of the United States have integrated their corporate tax systems. The
potential economic gains from integration are substantial.
This Report examines in detail several different integration prototypes,
although it does not attempt an exhaustive discussion of all possible integration
systems or of all the technical issues raised by the alternative prototypes.
This Report does not contain legislative recommendations. Rather, it is
intended to stimulate discussion of the various prototypes and issues they raise.
By advancing the opportunity for such debate, this Report should encourage
serious consideration of proposals for integrating the individual and corporate tax
systems in the United States.

v

EXECUTIVE SUMMARY
WHAT IS INTEGRATION AND WHY SHOULD IT BE BENEFICIAL?
Currently, our tax system taxes corporate profits distributed to shareholders at least
twice;-once at the shareholder level and once at the corporate level. If the distribution is
made through multiple unrelated corporations, profits may be taxed more than twice. If, on
the other hand, the corporation succeeds in distributing profits in the form of interest on
bonds to a tax-exempt or foreign lender, no U.S. tax at all is paid.
The two-tier tax system (Le., imposing tax on distributed profits in the hands of
shareholders after taxation at the corporate level) is often referred to as a classical tax
system. Over the past two decades, most of our trading partners have modified their
corporate tax systems to "integrate" the corporate and shareholder taxes to mitigate the
impact of imposing two levels of tax on distributed corporate profits. Most typically, this has
been accomplished by providing the shareholder with a full or partial credit for taxes paid
at the corporate level.
Integration would reduce three distortions inherent in the classical system:
(a)

The incentive to invest in noncorporate rather than corporate
businesses. Current law's double tax on corporations creates a
higher effective tax rate on corporate equity than on noncorporate equity. The additional tax burden encourages "selfhelp" integration through disincorporation.

(b)

The incentive to [fiance corporate investments with debt rather
than new equity. Particularly in the 1980s, corporations issued
substantial amounts of debt. By 1990, net interest expense
reached a postwar high of 19 percent of corporate cash flow.

(c)

The incentive to retain earnings or to structure distributions of
corporate profits in a manner to avoid the double tax. Between
1970 and 1990, corporations' repurchases of their own shares
grew from $1.2 billion (or 5.4 percent of dividends) to $47.9
billion (or 34 percent of dividends). By 1990, over one-quarter
of corporate interest payments were attributable to the substitution of debt for equity through share repurchases.

These distortions raise the cost of capital for corporate investments; integration could
be expected to reduce it. To the extent that an integrated system reduces incentives for
highly-leveraged corporate capital structures, it would provide important non-tax benefits by
encouraging the adoption of capital structures less vulnerable to instability in times of
economic downturn. The Report contains estimates of substantial potential economic gains
from integration. Depending on its form, the Report estimates that integration could increase
the capital stock in the corporate sector by $125 billion to $500 billion, could decrease the

Vll

Executive Summary

Vlll

debt -asset ratio in the corporate sector by 1 to 7 percentage points and could produce an
annual gain to the U.S. economy as a whole from $2.5 billion to $25 billion.

PROTOTYPES
This Report deftnes four integration prototypes and provides speciftcations for how
each would work. Three prototypes are described in Part II: (1) the dividend exclusion
prototype, (2) the shareholder allocation prototype, and (3) the Comprehensive Business
Income Tax (CBIT) prototype. In addition, in Part IV, titled "Roads Not Taken," the Report
describes the imputation credit prototype and a dividend deduction alternative. For
administrative reasons that the Report details, we have not recommended the shareholder
allocation prototype (a system in which all corporate income is allocated to shareholders and
taxed in a manner similar to partnership income under current law). Simplification concerns
led us to prefer the dividend exclusion to any fonn of the imputation credit prototype.
In the dividend exclusion prototype, shareholders exclude dividends from income
because they have already been taxed at the corporate level. Dividend exclusion provides
signiftcant integration beneftts and requires little structural change in the Internal Revenue
Code. When fully phased in, dividend exclusion would cost approximately $13.1 billion per
year.
CBIT is, as its name implies, a much more comprehensive and larger scale prototype
and will require signiftcant statutory revision. CBIT represents a long-tenn, comprehensive
option for equalizing the tax treatment of debt and equity. It is not expected that implementation of CBIT would begin in the short tenn, and full implementation would likely be phased
in over a period of about 10 years. In CBIT, shareholders and bondholders exclude dividends
and interest received from corporations from income, but neither type of payment is
deductible by the corporation. Because debt and equity receive identical treatment in CBIT,
CBIT better achieves tax neutrality goals than does the dividend exclusion prototype. CBIT
is self-fmancing and would pennit lowering the corporate rate to the maximum individual
rate of 31 percent on a revenue neutral basis, even if capital gains on corporate stock were
fully exempt from tax to shareholders.

POLICY RECOMMENDATIONS
In addition to describing prototypes, the Report makes several basic policy
recommendations which we believe should apply to any integration proposal ultimately
adopted:
(a)

Integration should not result in the extension of corporate tax
preferences to shareholders. This stricture is grounded in both
policy and revenue concerns and has been adopted by every
country with an integrated system. The mechanism for
preventing passthrough of preferences varies; some countries
utilize a compensatory tax mechanism and others simply tax
preference-sheltered income when distributed (as we recommend in the dividend exclusion prototype). Both of these
mechanisms are discussed in the Report.

ix

Executive Summary

(b)

Integration should not reduce the total tax collected on co(porate income allocable to tax-exempt investors. Absent this
restriction, business profits paid to tax-exempt entities could
escape all taxation in an integrated system. This revenue loss
would prove difficult to fmance and would exacerbate distortions between taxable and tax-exempt investors~

(c)

Integration should be extended to foreign shareholders only
through treaty negotiations. not by statute. This is required to
assure that U.S. shareholders receive reciprocal concessions
from foreign tax jurisdictions.

(d)

Foreign taxes paid by U.S. co(porations should not be treated.
by statute. identically to taxes paid to the U. S. Government.
Absent this limitation, integration could eliminate all U.S.
taxes on foreign source profits in many cases.

A table summarizing the characteristics of each of the prototypes follows.

OBJECTIVES OF THE REPORT
This Report is not a legislative proposal but rather a source document to begin the
debate on the desirability of integration. This Report concludes that integration is desirable
and presents a variety of integration mechanisms. A major reform such as integration should
be undertaken only after appropriate deliberation and consideration of public comments. In
light of the increasing isolation of the United States as one of the few remaining countries
with a classical tax system, serious consideration of integration is now appropriate.

Executive Summary

x

Comparison of the four principal integration prototypes
Prototype
Dividend
Exclusion
Protot),pe

Shareholder
Allocation
Prototype

CBIT
Prototype

Imputation
Credit
Prototype

Distributed
Income

Corporate rate

Shareholder rate l

CBIT rate (31 percent)

Shareholder rate l

Retained

Corporate rate
(additional shareholder
level tax depends on
the treatment of capital gains; see Chapter

Shareholder rate

CBIT rate (additional
investor level tax
depends on the
treatment of capital
gains; see Chapter 8)

Corporate rate
(additional shareholder level tax
depends on the
treatment of capital
gains; see Chapter 8)

Issues
Rates
a)
b)

Income~

8)

Treatment of
non-corporate
hw;inesses

Unaffected

Unaffected

CBIT applies to noncorporate businesses
as well as corporations, except for very
small businesses.

Unaffected

Corporate tax
preferences

Does not extend preferences to shareholders. Preference income is subject to
shareholder tax when
distributed.

Extends preferences to shareholders.

Does not extend preferences to investors.
Preference income is
subject to compensatory tax or investor level
tax when distributed.

Does not extend
preferences to shareholders. Preference
income is subject to
shareholder tax when
distributed.

Tax-exempt

Corporate equity income continues to
bear one level of tax.

Corporate equity
income continues
to bear one level
of tax.

A CBIT entity's equity
income and income
used to pay interest
bear one level of tax.

Corporate equity
income continues to
bear one level of tax.

Foreign source
income

Foreign taxes are
creditable at the corporate level, but
shielded income is
subject to shareholder
tax when distributed.

Foreign taxes are
creditable at the
corporate level
and at the shareholder level.

Foreign taxes are
creditable at the entity
level, but shielded
income is subject to
compensatory tax or
an investor level tax
when distributed.

Foreign taxes are
creditable at the
corporate level, but
shielded income is
subject to shareholder
tax when distributed.

Foreign
investors

Corporate equity income continues to
bear tax at the corporate level and current withholding taxes
(eligible for treaty
reduction) continue to
apply to distributions.

Corporate equity
income continues
to bear tax at the
corporate level
and current withholding taxes
(eligible for treaty
reduction) con tinue to apply to
distributions.

A CBIT entity's equity
income and income
used to pay interest
bear tax only at the
entity level, and no
withholding taxes are
imposed on distributions to equity holders
or on payments of
interest.

Corporate equity
income continues to
bear tax at the
corporate level and
current withholding
taxes (eligible for
treaty reduction)
continue to apply to
distributions.

Treatment of
debt

Unaffected

Unaffected

Equalizes treatment of
debt and equity

Unaffected (unless
bondholder credit
system adopted)

in\'t~stors

IPlus 3 percentage points of corporate level tax not creditable because the prototype retains the 34 percent corporate
rate but provides credits at the 31 percent shareholder rate.
2Assuming no DRIP. See Chapter 9.

Table of Contents
PART I: THE CASE FOR INTEGRATION . . . . . . . . . . . . . . . . . .. 1
Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 1
1.A The Corporate Tax: Need for Change. . . . . . . . . . . .. 1
1. B The Corporate Tax and Economic Distortions . . . . . . ., 3
1. C Neutrality as the Goal of Integration . . . . . . . . . . . . .. 12
PART II: PROTOTYPES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 2: Dividend Exclusion Prototype . . . . . . . . . . . . . . . . ..
2.A Introduction and Overview of Prototype . . . . . . . . . . .
2.B The Need for a Limitation on Excludable Dividends . . ..
2. C Foreign Source Income . . . . . . . . . . . . . . . . . . . . . .
2.D Low-Bracket Shareholders . . . . . . . . . . . . . . . . . . . .
2.E Individual Alternative Minimum Tax . . . . . . . . . . . . .
2.F Structural Issues . . . . . . . . . . . . . . . . . . . . . . . . . .
2.G Pension Funds . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 3: Shareholder Allocation Prototype . . . . . . . . . . . . . . .
3.A Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.B Overview of the Shareholder Allocation Prototype .. . ..
3.C Corporate Level Payment of Tax . . . . . . . . . . . . . . . .
3.D Passthrough of Corporate Losses to Shareholders . . . . "
3. E Tax Treatment of Preferences . . . . . . . . . . . . . . . . . .
3.F Allocating Income Among Different Classes of Stock . "
3.G Change of Stock Ownership During the Year . . . . . . . .
3.H Reporting and Auditing Considerations . . . . . . . . . . ..
3.1
Treatment of Tax-Exempt and Foreign Shareholders .. "
3.J
Foreign Source Income . . . . . . . . . . . . . . . . . . . . . .
Chapter 4: Comprehensive Business Income Tax Prototype . . . . . "
4.A Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. B Overview of CBIT Prototype . . . . . . . . . . . . . . . . . .
4.C Entities Not Subject to CBIT . . . . . . . . . . . . . . . . . .
4.D Tax Preferences . . . . . . . . . . . . . . . . . . . . . . . . . .
4.E International Considerations . . . . . . . . . . . . . . . . . . .
4.F Impact of CBIT on Investment Behavior of Low-Bracket,
Tax-Exempt, and Foreign Investors . . . . . . . . . . . . . .
4.G Structural Issues . . . . . . . . . . . . . . . . . . . . . . . . . .
4.H Conduits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1
Financial Intermediaries Under CBIT . . . . . . . . . . . . .

15
15
17
17
18
21
22
23
23
24
27
27
27
29
30
30
32
33
35
36
36
39
39
40
41
43
45
49
52
56
58

PART III: PRINCIPAL ISSUES . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Chapter 5: Treatment of Tax Preferences . . . . . . . . . . . . . . . . . . 63

XI

Chapter 6: Tax-Exempt and Tax-Favored Investors. . . . . . . . . . ..
6.A Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.B Distortions Under Current Law . . . . . . . . . . . . . . . . .
6.C Neutrality Under an Integrated Tax System . . . . . . . . .
6.D General Recommendations . . . . . . . . . . . . . . . . . . . .
Chapter 7: Treatment of Foreign Income and Shareholders . . . . . "
7. A Introduction.............................
7.B Overview of U.S. International Tax Policy . . . . . . . . "
7. C International Tax Policy and Integration . . . . . . . . . . .
Chapter 8: The Treatment of Capital Gains in an
Integrated Tax System . . . . . . . . . . . . . . . . . ." . . . . . . . .
8. A Taxation of Capital Gains Attributable to
Retained Taxable Earnings . . . . . . . . . . . . . . . . . . .
8.B Sources of Capital Gains Other Than
Taxable Retained Earnings . . . . . . . . . . . . . . . . . . .
8.C Adjustments to Eliminate Double Taxation of
Retained Corporate Earnings . . . . . . . . . . . . . . . . . .
8. D Other Countries . . . . . . . . . . . . . . . . . . . . . . . . . .
8.B Share Repurchases . . . . . . . . . . . . . . . . . . . . . . . .
8. F Capital Losses . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chapter 9: Dividend Reinvestment Plans . . . . . . . . . . . . . . . . . .
9.A Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9.B Design Considerations . . . . . . . . . . . . . . . . . . . . . .
Chapter 10: Transition Considerations . . . . . . . . . . . . . . . . . . . .
10.A Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10.B Taxation of Transitional Gains and Losses . . . . . . . . ..
10.C Phase-In of Integration . . . . . . . . . . . . . . . . . . . . . .
10.D Mechanics of a Phase-In . . . . . . . . . . . . . . . . . . . . .

67
67
69
69
70
73
73
74
77
81
81
82
82
84
84
86
87
87
88
89
89
89
90
91

PART IV: TIlE ROADS NOT TAKEN . . . . . . . . . . . . . . . . . . . . . 93
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Chapter 11: Imputation Credit System . . . . . . . . . . . . . . . . . . . . 95
II.A Overview of Imputation Credit Prototype . . . . . . . . . . . 95
lI.B Choice Between a Credit Limitation System and a
Compensatory Tax System . . . . . . . . . . . . . . . . . . . . 97
Il.C Role of the Corporate Alternative Minimum Tax .... , 101
11. D Foreign Source Income . . . . . . . . . . . . . . . . . . . .. 102
11. E Choices Required Because of Shareholders with
Different Rates . . . . . . . . . . . . . . . . . . . . . . . . .. 103
11.F Anti-Abuse Rules . . . . . . . . . . . . . . . . . . . . . . . . 103
11. G Structural Issues . . . . . . . . . . . . . . . . . . . . . . . .. 104
Il.H Extending the Imputation Credit Prototype to Debt. . .. 105
II.I Dividend Reinvestment Plans (DRIPs) . . . . . . . . . . . 106

xii

Chapter 12: Other Proposals to Reduce the Bias Against
Corporate Equity . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12.A Dividend Deduction. . . . . . . . . . . . . . . . . . . . . ..
12.B Institute for Fiscal Studies Proposal . . . . . . . . . . . ..
12.C American Law Institute Reporter's Study Draft . . . . ..
PART V: ECONOMIC ANALYSIS OF INTEGRATION. . . . . . . . ..
Chapter 13: Economic Effects of Integration . . . . . . . . . . . . . ..
13.A Introduction and Summary. . . . . . . . . . . . . . . . . ..
13.B Corporate Tax Distortions: Economic Issues. . . . . . ..
13.C Methodological Issues in Analyzing the
Allocation Effects of Integration . . . . . . . . . . . . . . .
13.D Overview of the Integration Prototypes . . . . . . . . . ..
13.E Integration, Corporate Financial Policy, and the
Cost of Capital . . . . . . . . . . . . . . . . . . . . . . . . ..
13.F Integration and the Allocation of Resources .........
13. G Distributional Effects of Integration . . . . . . . . . . . ..
13.H Revenue Estimates for Integration Prototypes . . . . . . .

107
107
108
108
111
111
111
112
118
120
121
128
146
150

APPENDICES
Appendix A: The Corporate Income Tax in the United States . . . . . . .
A.1 Brief Description of the Corporate Income Tax . . . . . .
A.2 Overview of U.S. Corporate Tax Receipts . . . . . . . . .
Appendix B: Experience of Other Countries with
Distribution-Related Integration Systems . . . . . . . . . . . . . .
B.l Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B.2 Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B.3 France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B.4 Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B.5 New Zealand . . . . . . . . . . . . . . . . . . . . . . . . . . .
B.6 United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . .
Appendix C: Equivalence of Distribution-Related Integration Systems
C.l Equivalence of Systems If Tax Rates Were Equal . . . .
C.2 Effects of Rate Differences, Preference Income, and
Exempt Shareholders . . . . . . . . . . . . . . . . . . . . . .

153
153
153
156

NOTES

189

................................................................................

159
159
163
167
172
177
181
185
185
185

GLOSSARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
BIBUOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

xiii

List of Figures
Figure 1.1

Distortions Under the Classical System

4

Figure 1.2

Ratio of Corporate Investment Relative to
Residential Investment in Four Countries, 1976-1989 ...

5

Ratio of Corporate Investment Relative to Noncorporate
(including Household) Investment in
Five Countries, 1976-1989 . . . . . . . . . . . . . . . . . . .

6

Ratio of Credit Market Debt to the Book Value of
Tangible Assets: Nonfmancial Corporations . . . . . . . . .

7

Ratio of Market Value of Debt to Market Value of the
Finn: Nonfmancial Corporations . . . . . . . . . . . . . . ..

8

Figure 1.6

Changing Sources of Funds for the Corporate 'Sector

8

Figure 1.7

Ratio of Net Interest to Cash Flow, 1948-1990:
Nonfmancial Corporations . . . . . . . . . . . . . . . . . ..

Figure 1.3

Figure 1.4
Figure 1.5

10

Figure 4.1

Comparison of CBIT and Current Law . . . . . . . . . . . . 39

Figure 6.1

Pension Fund Holdings of Corporate Capital, 1950-1990

68

Figure 13.1

Profits of Nonfmancial Corporations,
Proprietors' Income, and Net Interest as a
Percentage of Net National Product, 1950-1990 . . . . ..

113

Figure 13.2

Measures of Corporate Activity in the Economy,
1950-1990 . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 114

Figure A.l

Corporate Receipts as a Percentage of Total Receipts and
Gross National Product, 1940-1991 . . . . . . . . . . . . . 156

xv

List of Tables
Table 1.1

Corporate Tax Wedges for New Investments in
Manufacturing, 1991 . . . . . . . . . . . . . . . . . . . . . . .

6

Table 1.2

Sources of Funds, Nonfmancial Corporations, 1946-1990

9

Table 1.3

Estimates of Maximum Amount of Interest Attributable to
Increased Share Repurchases, 1980-1990 . . . . . . . . .

10

Table 2.1

Total U.S. Tax Rate on a Dollar of NonPreference, U.S.
Source Income from a U.S. Business Under Current Law
and the Dividend Exclusion Prototype . . . . . . . . . . . . 18

Table 3.1

Total U.S. Tax Rate on a Dollar of NonPreference, U.S.
Source Income from a U.S. Business Under Current Law
and the Shareholder Allocation Prototype . . . . . . . . . . 29

Table 4.1

Total U.S. Tax Rate on a Dollar of NonPreference, U.S.
Source Income from a U.S. Business Under Current Law
and the CBIT Prototype . . . . . . . . . . . . . . . . . . . . . 42

Table 6.1

Financial Assets of the Tax-Exempt Sector,
End of Year 1990 . . . . . . . . . . . . . . . . . . . . . . . . 68

Table 8.1

Taxation of Individuals on Long-Term Gains on
Securities: Select Foreign Countries . . . . . . . . . . . . . 85

Table 11.1

Total U.S. Tax Rate on a Dollar of NonPreference, U.S.
Source Income from a U.S. Business Under Current Law
and the Imputation Credit Prototype . . . . . . . . . . . . . 97

Table 13.1

Total U.S. Tax Rate on a Dollar of NonPreference, U.S.
Source Income from a U.S. Business Under Current Law
and the Integration Prototypes . . . . . . . . . . . . . . . . 122

Table 13.2

Effect of Integration on Corporate Financial Policy

Table 13.3

Cost of Capital Under Current Law . . . . . . . . . . . .. 126

Table 13.4

The Cost of Capital Under Current Law and the
Integration Prototypes: With Financial Distortions

128

The Cost of Capital Under Current Law and the
Integration Prototypes: No Financial Distortions . . . . .

129

Table 13.5

...

125

Table 13.6

General Equilibrium Results, Augmented Harberger
Model: With Financial Distortions . . . . . . . . . . . . . 130

Table 13.7

General Equilibrium Results, Augmented Harberger
Model: No Financial Distortions . . . . . . . . . . . . . . 131

xvii

Table 13.8

General Equilibrium Results, Mutual Production Model:
With Financial Distortions . . . . . . . . . . . . . . . . . .

134

Table 13.9

The Effect of Integration on the Allocation of
Physical Capital, Wealth, and Corporate Financial Policy:
Results from the Portfolio Allocation Model . . . . . . .. 137

Table 13.10

Summary of the Effects of Integration on
Real and Financial Decisions: Results from the
Portfolio Allocation Model (Lump Sum Replacement)

.

138

Table 13.11

General Equilibrium Results: International Model,
Projected Long-Run Effects of
Tax Integration Alternatives . . . . . . . . . . . . . . . . . 144

Table 13.12

Effective Tax Rates on Individuals: Current Law and
Integration Prototypes Standard Incidence Assumption

Table 13.13

Effective Tax Rates on Individuals: Current Law and
Integration Prototypes Alternative Incidence Assumption

xviii

.

148
148

PART I: THE CASE FOR INTEGRATION
CHAPTER 1: INTRODUCTION
1.A THE CORPORATE TAX:
NEED FOR CHANGE

however, did not address tax-related distortions of
business organizational and fmancing decisions. In
fact, the 1986 reforms may have increased the
pressure to select noncotporate organizational
forms by imposing a higher marginal rate on
cotpOrations than on individuals and by repealing
the General Utilitiet doctrine, which had protected cotpOrations from cotporate level tax on
liquidating dispositions of cotporate assets. Corporate integration can thus be regarded as a
second phase of tax reform in the United States,
extending the goal of neutral taxation to the
choice of business organization and fmancial
policy.

Issues
Current U.S. tax law treats cotporations and
their investors as separate taxable entities. Under
this classical system of cotporate income taxation,
two levels of income tax are generally imposed on
earnings from investments in cOtporate equity.
First, cotporate earnings are taxed at the cotporate level. Second, if the cotporation distributes
earnings to shareholders, the earnings are taxed
again at the shareholder level. In contrast, investors in business activities conducted in noncotporate form, such as sole proprietorships or
partnerships, are generally taxed only once on the
earnings, and this tax is imposed at the individual
level. Cotporate earnings distributed as interest to
suppliers of debt capital also are taxed only once
because interest is deductible by the cotporation
and generally taxed to lenders as ordinary income.

The current two-tier system of cotporate
taxation discourages the use of the cotporate form
even when incotporation would provide nontax
benefits, such as limited liability for the owners,
centralized management, free transferability of
interests, and continuity of life. The two-tier tax
also discourages new equity fmancing of cotporate
investment, encourages debt fmancing of such
investment, distorts decisions with respect to the
payment of dividends, and encourages cotporations to distribute earnings in a manner designed
to avoid the double-level tax.

Despite its long history, considerable debate
surrounds the role of the cotporate income tax in
the Federal tax structure. The central issue is
whether cotporate earnings should be taxed once
rather than taxed both when earned and when
distributed to shareholders. Integration of the
individual and cotporate income tax refers to the
taxation of cotporate income once. This Report
discusses and evaluates several integration
alternatives. I

These distortions have economic costs. The
classical cotporate tax system reduces the level of
investment and interferes with the efficient allocation of resources. In addition, the tax bias against
cotporate equity can encourage cotporations to
increase debt fmancing beyond levels supported
by nontax considerations, thereby increasing risks
of fmancial distress and bankruptcy.

Despite their differences, the methods of
integration studied in this Report reflect a common goal: where practical, fundamental economic
considerations, rather than tax considerations,
should guide business investment, organization,
and fmandal decisions. The Tax Reform Act of
1986 (the 1986 Act)2 made the tax system significantly more neutral in its impact on business
decisions about capital investment by reducing tax
rates and tax preferences. The 1986 Act,

Historically, the cotporation has been an
important vehicle for economic growth in the
United States, but the classical cotporate tax
system often perven~ely penalizes the cotporate
form of organization. With the increasing integration of international markets for products and
capital, one must consider effects of the cotporate
1

2

The Case for Integration

tax system on the competitiveness of

u.s. fmns.

Most of the major trading partners of the United
States have revised their tax systems to provide
for some integration of the corporate and
individual tax systems.
This Report provides a comprehensive study
of integration, including both the legal and economic foundations for implementing integration in
the United States. We present three prototypes
representing a range of integration systems and
recommend two prototypes that implement our
policy goals. One prototype, a dividend exclusion
system, can be implemented with minimal changes to current law. The second, the Comprehensive
Business Income Tax (CBIT) , extends the dividend exclusion model to debt. CBIT achieves the
important goal of equating the treatment of debt
and equity, but because it represents a greater
departure from current law, it would require a
longer transition period. We have included, albeit
with substantial reservations as to feasibility a
third prototype-a shareholder allocation system,
often referred to as full integration. We considered it necessary to examine such a prototype
because this system is so frequently viewed as
ideal by proponents of integration, although we
ultimately reject it on both policy and
administrative grounds.
l

The Report also documents the substantial
economic benefits of integration . We estimate that
any of the three prototypes would increase the
capital stock in the corporate sector by $125 to
$500 billion and would decrease the debt to asset
ratio in the corporate sector from 1 to 7 percentage points. Further, efficiency gains from integration would be equivalent to annual welfare gain
for the U.S. economy as a whole of 0.07 to 0.7
percent of annual consumption (or $2.5 to $25
billion (in 1991 dollars).4 See Chapter 13.

Brief Description of Current Law
Under current law, income earned by corporations is taxed at the corporate level, generally at
a marginal rate of 34 percent. 5 When the corporation distributes earnings to shareholders in the
fonn of dividends, the income is generally taxed

again at the shareholder level. 6 If corporations
retain earnings, the value of their stock will
generally increase to reflect those earnings. When
shareholders sell their stock, gains from the sale
are taxed also. Thus, like income distributed as
dividends, retained corporate income generally is
taxed twice. In contrast, investors who conduct
business activity in noncorporate fonn, such as
through a sole proprietorship or partnership, are
taxed once on their earnings at their individual tax
rate.
Dividends distributed to individual U.S.
citizens and residents are taxed generally at
marginal rates of 15, 28, or 31 percent. 7 Dividends distributed to nonresident aliens and foreign
corporations by U. S. corporations are generally
subject to a nonrefundable "withholding" tax,
currently set by statute at 30 percent. United
States treaties with trading partners frequently
reduce the rate to 15 or 5 percent on a reciprocal
basis. Dividends received by U.S. cotporate
shareholders generally qualify for a dividends
received deduction of 70, 80 or 100 percent,
depending on the degree of affiliation between the
corporations. Shareholders' gains from sales of
corporate stock are taxed also, typically as capital
gains, although capital gains of foreign shareholders generally are exempt from U.S. tax.
Unlike dividends, interest is generally deductible by corporations. Interest income received by
domestic lenders is generally taxed at their marginal tax rates. Interest income received by foreign lenders from U.S. corporations, however,
generally is not subject to U.S. tax. 8
Tax-exempt entities supply a substantial
portion of the corporate capital in the United
States. These tax-exempt entities include pension
funds and educational, religious and other charitable organizations. These entities are generally not
taxed on interest, dividends or gains from the sale
of their investments. However, the corporate level
tax applies to corporate income attributable to the
equity capital they supply. Tax-exempt entities
may be subject to the unrelated business income
tax (UBIT) on earnings from equity investments
in partnerships.

3

l.B

THE CORPORATE TAX AND
ECONOMIC DISTORTIONS

The classical corporate income tax system
distorts three economic and fmancial decisions:
(1) whether to invest in noncorporate rather than
corporate fonn, (2) whether to fmance investments with debt rather than equity, and (3) whether to retain rather than distribute earnings. Apart
from corporate and investor level tax considerations, nontax benefits and costs also influence
these decisions. To the extent that the classical tax
system distorts the choice of organizational fonn,
fmancial structure, and dividend policy, economic
resources can be misallocated. 9

The Cost of Capital As a Measure of
Investment Incentives
This Report examines distortions resulting
from the corporate income tax in tenns of effects
on the cost of capital. In deciding whether to
undertake an investment, fmns require that the
investment provide a sufficient after-tax return to
compensate investors. The cost of capital is the
pre-tax rate of return that is sufficient to cover
operating expenses, taxes, economic depreciation,
and the investor's required after-tax rate of return.
Thus, the cost of capital depends in part on the
return fmns must pay to suppliers of debt or
equity capital to attract funds. The cost of capital
also depends on such factors as tax rates, the
investment's economic depreciation rate, the
capital cost recovery deductions allowed on the
investment, the inflation rate, and the source of
fmancing for the investment. Because a higher
cost of capital makes certain investments unprofitable, corporate and individual income taxes
reduce investment incentives by raising the cost of
capital.
This section uses the cost of capital as a
framework for analyzing the effects of the current
classical corporate tax system on the business
decisions described above (i.e., fonn of business
organization, fonn of fmancing, and retention of
earnings). The fmal part of this section discusses
the effect of the corporate income tax on savings
and investment in the economy as a whole.

The Case for Integration

Organizational Form
The waste of economic resources from taxdistorted misallocation of capital between the
noncorporate and corporate sectors was the
original focus of criticism of the corporate income
tax. Beginning with Harberger,IO economists
have argued that a classical corporate tax system
misallocates capital 'between the corporate and
noncorporate sectors. Over the years, more
sophisticated models have been developed to
examine more carefully the efficiency costs of
corporate taxation. Contemporary approaches
suggest that these costs are significant. See
Chapter 13.
A simple example illustrates the effect of the
current corporate tax system on investment decisions. Suppose that an investor requires an aftertax rate of return of 8 percent and the investor's
effective tax rate is 20 percent. An equity investment in a noncorporate enterprise must earn a
return high enough to pay tax at the investor's
rate (20 percent) and still yield the required 8
percent after-tax returnY The noncorporate
investment must therefore earn a 10 percent pretax rate of return (net of depreciation) in order to
cover the investor's .income taxes and meet the
required return (0.10 X(l-0.20) = 0.08). However, if the corporate tax rate is 34 percent and
the corporation distributes all of its income, the
cost of capital of an equity fmanced investment in
the corporate sector in the above example is 15.2
percent. This 15.2 percent pre-tax return yields an
8 percent return after paying both the corporate
tax and the investor level tax on dividends
(0.152x(l-0.34)x(l-0.20) = 0.08). Since
fewer investments can earn the higher required
return (15.2 percent as opposed to 10 percent),
the corporate tax discourages investment in the
corporate sector by raising the cost of capital.
More complex calculations support this result.
For example, a Congressional Research Service
report estimates, under realistic assumptions, the
total effective Federal income tax rate on corporate equity (taking into account both corporate
level and shareholder level taxes) to be 48 percent, compared to 28 percent for noncorporate

The Case for Integration

4

equity.12 Therefore, some corporations fail to
undertake investments that would be profitable if
the tax burden on corporate and noncOlporate
investments were the same. Moreover, for some
business enterprises, the added corporate taxes
exceed the nontax benefits of incorporation,
causing such businesses to forego those benefits
and to operate instead in noncorporate form.
Figure 1.1 illustrates the differences in taxation of
equity investments in corporate and non-corporate
businesses.
The bias against corporate sector investments
compared with investments in the noncorporate
sector reduces the productivity of the nation's
capital investments and reduces potential national
income. See Chapter 13. This reduction in productivity is a hidden cost of the corporate tax. In
addition, the classical system encourages corporations to convert to noncorporate form, thereby
abandoning the benefits of incorporation. 13

the relative tax advantage of noncOIporate fmns.
Considering only tax costs, corporate and noncorporate entities face the same cost of debt
fmanced capital, because interest paid is deductible. Thus, corporations can reduce the difference
in tax burdens for total investment by fmancing
new investment with debt. Increases in debt may,
however, increase the risk of fmanciaI distress or
bankruptcy. Second, accelerated cost recovery
deductions provide, in effect, an interest-free
government loan to fmance new investment.
These deductions lower the total cost of capital
for both corporate and noncorporate fmns, but
because corporate tax rates generally exceed
individual tax rates, corporations realize greater
tax benefits from accelerated depreciation. Thus,
such deductions reduce, but do not eliminate, the
additional tax burden on corporate equity
investments.

Corporations also can reduce the distortion between corporate and noncorporate investments by
Certain tax provisions mitigate this tax bias
distributing corporate income to shareholders
against corporate investment. First, by using debt
through share repurchases and other nondividend
to fmance investments, corporations can reduce
distributions. The advantage of a nondividend
distribution is that it
allows shareholders to
Figure 1.1
recover the cost (or basis)
Distortions Under the Classical System 1
of their shares, with any
.
Equity Holders
excess generally taxed as
T3ltj Taxable
Corporation
capital gains. Current law
provides a slight rate
>
preference for capital
gains of individuals (a
Debt Holders
No Tax
maximum rate of 28
~ ·>N?~ax< .. ~=:
percent compared with a
No Tax
Tax-Exempt
maximum of 31 percent
Investment
on other income). Capital
Equity Holders
gains also benefit from
......... T
bl
.:::::::Tax><::> axa e
the deferral permitted
••..••• :... T
Foreign
under
current law, be... ··ax· ..... Tax-Exempt
cause shareholders do not
No Tax
Debt Holders
Return.
recognize gain until stock
Taxable
is sold, and capital assets
Foreign
receive a tax-free step-up
Tax-Exempt
in
basis at death. The
Non-Corporate Form
preferential tax treatment
IThe figure does not take into account tax preferences or taxes
of capital gains reduces ,
imposed by other countries.
but does not eliminate,

e

~ • N!f"" i:~empt

~

ax «
~

The Case for Integration

5

the distorting effect of the current corporate tax
system on corporate level investment.
International comparisons add perspective on
the effect of the corporate tax on the U. S. corporate sector. One measure is the ratio of corporate
investment to investment in housing, which
provides a comparison of resource allocation in
different economies. Figure 1.2 presents the ratio
of corporate gross ftxed investment relative to
private residential investment in the United States
and three other industrialized countries for which
data are available since 1976. Throughout the
period, the United States had a lower ratio than
the United Kingdom. Although the U.S. ratio
exceeded that for Japan and Australia until the
early 1980s, corporate investment relative to
housing investment has tended upwards over the
whole period for Japan and Australia while the
ratio for the United States has remained fairly
stable, except for the 2 years following the Economic Recovery Tax Act of 1981. Indeed, for the
last 4 years for which data are available, the
United States has had essentially the lowest
corporate investment per dollar of housing investment of any of the four nations. A similar picture

of relatively low corporate investment in the
United States is depicted in Figure 1.3, which
presents the ratio of investment (net of depreciation) in the corporate sector relative to the total
noncorporate sector (households and unincorporated businesses combined) during the same period
for the same four countries plus France. By this
measure, the United States had the lowest ratio of
corporate to noncorporate investment during the
last 3 years for which data are available for any
of the five nations.

Another useful international comparison is the
spread between the pre-tax return on corporate
investment and the cost of funds in the U oited
States and other countries. This spread, or corporate "tax wedge," generally depends upon the type
of asset acquired, the corporate tax rate, the
capital recovery allowances, the rate of inflation,
and various other country specific factors. Table
1.1 presents a listing of preliminary OECD
calculations of the 1991 corporate tax wedge
based on a standardized mix of assets and sources
of funding for a manufacturer located in several
OECD member countries. According to these
data, the corporate tax wedge in the United States
is higher than in France or
Germany, is approximately the
Figure 1.2
same as in the U.K., and is
Ratio of Corporate Investment Relative to
lower than the tax wedge in
Residential Investment in Four Countries, 1976-1989
Canada and Japan.

Corporate Capital
Structure

4

3.S

Corporations have three
alternatives for fmancing new
investments: (1) issuing new
equity, (2) using retained
earnings, or (3) issuing debt.
There can be important nontax
benefits and costs of alternative corporate financing
arrangements, and the tax
system should avoid prejudicing fmandal decisions.

3

1.5

1

1976

1980

1984

1988

Year
Source: Organisation for Economic Q)..()peration and Development,
National Accounts (1976-1989).

The current classical corporate tax system discriminates

6

The Case for Integration

Figure 1.3
Ratio of Coryorate Investment Relative to
Noncorporate (mcluding Housebold) Investment
in Five Countries, 1976-1989

3.5
3

/

/

2.5

\

\

\

i

\

i

v

\

\

--United Kingdom
--"

Japan
/\

1.5

I

/

/

\

1
0.5

1980

1976

Year

1988

1984

Source: Organisation for Economic co-operation and Development,
National Accounts (1976-1989).

Table 1.1
Corporate Tax Wedges for
New Investments in Manufacturing
1991
Country
Canada
France
Gennany
Japan
United Kingdom
United States

COlporate Tax
Wedge l
1.2

0.4
0.6
1.4

0.9
0.8

Department of the Treasury
Office of Tax Policy
lThe difference between the pre-corporate tax real rate of return and 5
percent (the real interest rate). The calculations assume no personal taxes
and an inflation rate of 4.5 percent for all countries. The weights for the
proportion of investment in each type of asset and the proportion of
finance from each source of funds are assumed to be the same for each
country: 50 percent for machinery, 27 percent for buildings, and 23
percent for inventories and 35 percent for debt, 10 percent for new
equity, and 55 percent for retentions.

against equity fmancing of
new corporate investment.
See Figure 1.1. Because of
the two levels of taxation of
corporate profits, the cost of
equity capital generally exceeds the cost of debt capital.
The Congressional Research
Service estimates, under
realistic assumptions, the
total effective Federnl income
tax rate on corporate debt to
be 20 percent, compared with
48 percent for corporate equity.14 Moreover, the total
effective tax rate on debt can
be negative. The lower effective tax rate for debt fmanced
corporate investment than for
equity fmanced corporate
mvestment encourages the
use of debt by corporations,
assuming nontax factors that
affect fmancing decisions do
not change.

If a corporation borrows
from an individual to fmance
an investment, the corporation deducts the interest
payments from its taxable
income and is therefore not
taxed on the investment's
pre-tax return to the extent of
interest payments, although
the lender is taxable on the
interest at the individual tax
rate. 15 Consequently, to the
extent that corporations finance investment with debt,
current law does not distort
the choice between investment in the corporate and
noncotporate sectors. Using
the assumptions in the numerical example set forth under
"Organizational Form,
above, for a 100 percent
debt financed
corporate
If

Source: Organisation for Economic Co-operation and Development,
preliminary unpUblished estimates.

7

investment, the cost of capital is 10 percent
(0.10X(I-0.2) = 0.08, the required rate of
return). This cost is well below the 15.2 percent
cost of capital for equity fmanced investments for
corporations that distribute income as dividends,
and is the same as the cost of capital for a noocorporate investment.

Recent Trends in Corporate Debt
Historical data show U. S. corporate debt to be
at relatively high levels by postwar standards,
with some, but not all, measures growing at an
unusually rapid pace in the 1980s. Because there
is no single, universally agreed-upon measure of
debt, the discussion below considers trends based
on alternative measures.

The Case

fOT

Integration

ratio of credit market debt to the book value of
tangible assets for nonfmancial corporations,
based on Federal Reserve Board data. This ratio
grew from 43 percent in 1948 to 61 percent in
1989. Although the ratio generally increased over
the postwar period, it declined sharply beginning
in 1975 and continuing through the mid 1980s.
Following that decrease, the ratio began to rise
again and by 1989 had reached a postwar high of
61 percent. In 1989, this book-value debt to asset
ratio was more than 17 percentage points higher
than in 1980, but only 10 percentage points higher
than the pre-1980s peak of 51 percent reached in
1973.

Figure 1.5 presents Federal Reserve Board
data showing the ratio of the market value of debt
to the market value of the ftnn (debt plus equity)
One group of debt measures focuses on corpo- for nonfmancial corporations from 1961 through
1989. Like the book-value measure, the marketrate balance sheets: the ratio of debt to total
assets. The debt to asset ratio can be computed value ratio indicates that corporate debt has
using either book value (the par value of debt and generally increased since 1961. In 1961 , debt
represented 26 percent of the total market value of
the historical cost of assets as reported for
the capital stock of nonfmancial corporations
fmancial accounting purposes) or market value.
Figure 1.4 displays one book value measure, the compared to 38 percent of total market value in
1989. The market-value data,
however, suggest that the
dramatic increase in corpoFigure 1.4
Ratio of Credit Market Debt to the
rations' use of debt occurred in
Book Value of Tangible Assets
the middle 1970s. Indeed, the
Nonfmancial Corporations
market-value ratio peaked at 47
0.7
percent in 1974, a year in
which the stock market fell
0.65
sharply. During the 1980s, the
market-value ratio does not
0.6
show a discernible upward
0.55
trend because rising stock
market prices largely offset the
..=0 0.5
growth
in the dollar amount of
~
debt during this period. In
0.45
contrast, the book-value mea0.4
sure described in the preceding
paragraph shows a large in0.35
crease during the 1980s, because stock market growth is
0.3
not reflected directly in the
1945 1950 1955 1960 1965 1970 1975 1980 1985
book-value
measure, and thus
Year
does not offset the rising dollar
Source: Federal Reserve Board, EWw of ~ Accounts (various issues).
volume of debt. 16

8

The Case for Integration

A second measure of
leverage focuses on the
importance of debt in corporations' sources of additional
funds rather than corporations' total outstanding
debt. See Table 1.2. Over the
entire postwar period, equity
fmance was dominant. For
nonfinancial corporations,
retained earnings and net new
equity issues accounted for
roughly 78 percent of funds
raised. Debt provided the
balance, divided about
equally between private
issues (bank loans and private
placements) and public issues
(bonds). Relative fmancing
patterns changed during the
1980s. While corporations
continue to rely heavily on
retained earnings, they have
sharply adjusted the composition of external fmance. Most
notably, corporations have
undertaken substantial repurchases of equity, fmanced
mainly with debt. 17 In
(current) dollar tenns, this
pattern is illustrated in the
left panel of Figure 1.6. The
increase in nonfinancial
corporate debt during the
early and middle 1980s was
largely matched by a reduction in outstanding equity. As
shown in the right panel of
Figure 1.6, nonfmancial
corporations relied significantly more on internal funds
(retained earnings) during the
1980s than was the case for
the postwar period as a
whole.
Recent evidence suggests
that share repurchases have
contributed to the increase in

Figure 1.5
Ratio of Market Value of Debt to
Market Value of the Firm
Nonfmancial Corporations
0.5
0.45

0.4
0.35

0.3
0.25
0
'::2

~

0.2
0.15
0.1
0.05
0
1961

1965

1969

1973

1977

1981

1985

1989

Year

Source: Federal Reserve Board, unpublished estimates.

Figure 1.6
Changing Sources of Funds for the Corporate Sector
Nonfinancial Corporate
Debt and Equity
(billions of current dollars)

300.----------

Nonfinancial Net Funds Raise(
Four Quarter Moving Average
(percent of nominal GNP)

6.---------------

5+--~-----------

Debt
100 +--------rt+-I-J.---l-----lA--

4+-~~-------------

Equity
-100 +------~'--'I--.:H..+-rl--

-200 + - - - - - - - - - - - - - ' -

lr-------------~~

-300 -'rmm-mmrmTmlTTTIrrnmrnnrnmnTIn"mn"""""mmmrnmmrmmrr

O~~~~~==~==TIn"mn~

1969 1973 1977 1981 1985 1989
Year
Source: Strongin (1991).

1969 1973 1977 1981 1985 1989
Year

The Case for Integration

9

Table 1.2
Sources of Funds, Nonimancial Corporations, 1946-1990

Year

Internal
Funds

1946
$8,503
1947
13,335
1948
19,651
1949
20,024
18,539
1950
1951
20,761
1952
22,457
1953
22,334
1954
24,403
1955
29,943
1956
30,045
1957
31,983
30,659
1958
1959
36,434
35,842
1960
1961
36,895
1962
43,219
1963
46,967
52,309
1964
1965
59,098
63,274
1966
1967
64,250
1968
65,766
1969
65,195
62,693
1970
74,614
1971
86,214
1972
93,704
1973
1974
88,972
124,249
1975
141,272
1976
164,401
1977
181,914
1978
197,206
1979
1980
199,772
239,098
1981
1982
241,901
1983
285,217
335,885
1984
351,815
1985
344,294
1986
372,448
1987
391,371
1988
380,010
1989
369,458
1990
Department of the Treasury
Office of Tax Policy

Amount
(millions of dollars)
New Debt
Net New
Issues
Equity Issues

$6,103
7,306
6,398
1,826
6,772
8,770
6,852
4,022
4,714
8,557
10,397
9,587
8,395
10,150
9,976
9,853
12,591
12,245
12,667
18,931
23,451
24,924
27,677
28,995
28,484
25,986
31,463
68,439
50,835
13,171
40,138
66,695
70,970
68,142
58,206
104,085
46,567
56,521
170,828
134,260
209,718
123,749
184,633
159,537
86,186

$1,018
1,093
1,000
1,212
1,288
2,107
2,320
1,766
1,583
1,719
2,250
2,441
1,968
2,078
1,365
2,121
369
(341)
1,145
(28)
1,259
2,397
(159)
3,406
5,694
11 ,435
10,922
7,883
4,097
9,908
10,524
2,727
(101)
(7,836)
10,375
(13,450)
1,900
20,000
(78,975)
(84,500)
(84,975)
(75,500)
(129,500)
(124,150)
(63,000)

Shares
Total
Funds

$15,624
21,734
27,049
23,062
26,599
31,638
31,629
28,122
30,700
40,219
42,692
44,011
41,022
48,662
47,183
48,869
56,179
58,871
66,121
78,001
87,984
91,571
93,284
97,596
96,871
112,035
128,599
170,026
143,904
147,328
191,934
233,823
252,783
257,512
268,353
329,733
290,368
361,738
427,738
401,575
469,037
420,697
446,504
415,397
392,644

Source: Federal Reserve Board, Flow of Funds Accounts (various issues).

Internal New Debt
Net New
Funds
Issues Equity Issues
54.4%
6.5%
39.1%
61.4%
5.0%
33.6%
72.6%
23.7%
3.7%
86.8%
7.9%
5.3%
69.7%
4.8%
25.5%
65.6%
27.7%
6.7%
71.0%
21.7%
7.3%
79.4%
14.3%
6.3%
79.5%
15.4%
5.2%
74.4%
21.3%
4.3%
70.4%
24.4%
5.3%
5.5%
72.7%
21.8%
74.7%
20.5%
4.8%
20.9%
4.3%
74.9%
76.0%
2.9%
21.1 %
75.5%
20.2%
4.3%
76.9%
22.4%
0.7%
20.8%
-0.6%
79.8%
19.2%
79.1%
1.7%
75.8%
24.3%
-0.0%
26.7%
1.4%
71.9%
70.2%
27.2%
2.6%
70.5%
29.7%
-0.2%
29.7%
3.5%
66.8%
64.7%
29.4%
5.9%
66.6%
23.2%
10.2%
67.0%
24.5%
8.5%
40.3%
4.6%
55.1 %
61.8%
35.3%
2.8%
84.3%
8.9%
6.7%
73.6%
20.9%
5.5%
70.3%
28.5%
1.2%
72.0%
28.1%
-0.0%
76.6%
26.5%
-3.0%
74.4%
21.7%
3.9%
4.1%
72.5%
31.6%
83.3%
16.0%
0.7%
78.8%
15.6%
5.5%
78.5%
39.9%
-18.5%
87.6%
33.4%
-21.0%
73.4%
44.7%
-18.1 %
88.5%
29.4%
-17.9%
87.7%
41.4%
-29.0%
91.5%
38.4%
-29.9%
22.0%
94.1 %
-16.0%

The Case for Integration

corporate debt. Rather than simply replacing
dividends, repurchases have been fmanced primarily by debt, which results in higher interest
costs,u Increased share repurchases, therefore,
accounts for part of the recent increases in net
interest payments, and may be viewed as one
method that flnns have used to reduce their
corporate tax liabilities. Table 1.3 presents estimates of the portion of net interest payments of
nonfinancial corporations that might be attributable to "excess" share repurchases of the 1980s,
where the excess is the difference between actual
repurchases and the levels that would have
occurred if the ratio of repurchases to dividends
had continued at its average for the 1970s. 19 The
table shows that, by 1990, over one quarter of the
interest payments of nonfmancial corporations was
attributable to increased share repurchases. 20
A third measure of corporate debt focuses on

10

Table 1.3
&timates of Maximum Amount of
Interest Attributable to
Increased Share Repurcbases
1980-1990
Year

Percentage of Net Interest
of Nonfinancial Corporations

1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990

1.0
0.9
1.3
1.8

5.4
11.2
12.4
18.2
23.6
23.4

25.5

Department of the Treasury
Office of Tax Policy
Source: Office of Tax Policy calculations based on
Standard and Poor's COMPUSTAT data and on information in Poterba (1987).

the ability of corporations to service their debt.
Corporations meet their interest payments out of
the cash available after other payments, such as
those for labor, materials, energy, and taxes.
Cash flow, calculated as after-tax proflts plus
depreciation, serves as a
measure of funds from which
Figure 1.7
a corporation can cover its
Ratio
of
Net
Interest
to Cash Flow, 1948-1990
interest payments. Figure 1.7
Nonfmancial Corporations
shows the ratio of net interest
0.2
to cash flow for nonfmancial
corporations from 1948
through 1990. These data
show a generally upward
0.15
trend over time with substantial increases in the late
1960s and early 1970s, again ~ 0
in the early 1980s, and in the ~ .1
last 2 years (1989 and 1990).
After reaching 19 percent in
1982, the ratio of net interest
0.05
to cash flow showed little
upward movement through
1988 but has increased in
1989 and 1990. By 1990, it
1966
1972
1978
1984
1960
1948
1954
1990
reached a postwar high of 19
Year
percent. Finn level data
Sources: Department of Commerce (1986) and Department of
document a similar
Commerce, Survey of Current Business (July, various years).
pattern. 21

11

Some economists also are concerned that high
debt-service burdens (by postwar standards)
during the 1980s have been associated with an
increase in corporate bankruptcies. While bankruptcies followed a cyclical pattern over most of
the postwar period, they remained high (relative
to postwar standards) throughout the expansion
following the 1981-1982 recession. 22
Benefits and Costs of Corporate Debt

Debt fmance may have nontax benefits.
Analysts most sanguine about high levels of
corporate debt and debt-service burdens typically
maintain that the discipline of debt is desirable
because it gives lenders indirect means to monitor
the activities of managers. This need for supervision owes to the separation between ownership
and management that is characteristic of the
traditional corporate structure. 23

The Case for Integration

relatively high taxes on such dividends compared
to capital gains. This Report assumes that corporate dividends offer special nontax benefits to
shareholders that offset their tax disadvantage,26
and, accordingly, that corporations set dividend
payments so that the incremental nontax benefit of
dividends paid equals their incremental tax cost.
Under this assumption, the amount of dividends
paid out is expected to decrease as the tax burden
on dividends relative to capital gains increases;
empirical studies are consistent with this prediction. 27 Investor level taxes on dividends also
raise the cost of capital (and thereby reduce
investment) to the extent that corporations payout
earnings as dividends. Thus, under the assumptions used in this Report, dividend taxes reduce
the payout ratio and real investment incentives.

A disadvantage of higher debt levels is that
they can increase nontax costs of debt, including
costs associated with fmancial distress. Even
when corporations avoid formal bankruptcy
proceedings, they incur costs when they cannot
meet their interest obligations or when debt
covenants restrict operating flexibility. The costs
include extra demands on executives' time, supply
disruptions, declines in customers' confidence,
and, frequently, significant legal fees. Corporations therefore must evaluate the tax and nontax
benefits of additional debt relative to these costs.
Tax-induced distortions in capital structure can
entail significant efficiency costs. 24

The growth in share repurchases in the last
decade supports this view of the linkage between
the corporate tax and corporate dividends. Share
repurchases provide a means of distributing
corporate earnings with, in many cases, more
favorable shareholder level tax treatment than
dividend distributions. While a shareholder pays
tax on the full amount of a dividend at ordinary
income rates, the shareholder generally pays tax
on the proceeds of a share repurchase only to the
extent they exceed share basis and, in some cases,
at a preferential capital gains rate. Share repurchases increased substantially from 1970 to 1990,
growing from $1. 2 billion (or 5.4 percent of
dividends) to $47.9 billion (or 34 percent of
dividends), and peaking in 1989 at $65.8 billion
(or 47 percent of dividends).28

Corporate Dividend Distributions

Savings and Investment

The current system of corporate taxation also
may distort a corporation's choice between distributing or retaining earnings and, if amounts are
distributed, whether they are paid in the form of
a nondividend distribution, such as a share repurchase. Differences in effective tax rates on dividends and retained earnings are significant. 25

The corporate tax. increases the tax burden on
the returns from saving and investing. The magnitudes of tax-induced distortions of investment and
savings decisions depend on two factors: the size
of the spread (or wedge) between pre-tax and
after-tax returns and the responsiveness of savers
and investors to changes in after-tax returns. The
more responsive savers and investors are to
changes in rates of return, the larger the effect of
a tax wedge of a given size. 29 The Report documents significant wedges between pre-tax and

Assessing the efficiency costs of such tax
differentials requires an analysis of motives for
cOIpOrate dividend distributions in the presence of

The Case for Integration

after-tax returns to saving and investment. While
empirical evidence on the effect of changes in the
after-tax return on savings is in conflict, there is
substantial empirical evidence documenting important effects of capital taxation on investment. 30
See Chapter 13.
In the presence of international capital flows,
the U. S. corporate income tax can reduce incentives to invest in the United States, even if it has
a relatively small effect on saving by U.S.
citizens.

I.e

NEUTRALITY AS THE GOAL
OF INTEGRATION

Integration would reduce and in some cases
eliminate the distortions of business decisions
under the current tax system by coordinating the
individual and corporate income tax systems so
corporate income is taxed only once. Broadly
speaking, corporate tax integration seeks to reduce
tax-induced distortions in the allocation of capital
by taxing corporate income once, rather than
zero, once, or multiple times as under the current
regime. Integration has attracted the attention of
tax policymakers for many years. The Department
of the Treasury and the Congress have considered
integration on several occasions, most recently in
1984 and 1985. 31 Many industrial countries have
long had integrated systems; several others have
recently adopted integration. 32
The classical system of corporate taxation is
inefficient because it creates differences in the
taxation of alternative sources of income from
capital. Under the classical system, a taxpayer
conducting business in corporate form faces a
different tax burden on equity fmancing than a
taxpayer conducting the same business in noncorporate form. A corporation that raises capital
in the form of equity faces a different tax burden
than a corporation that raises the same amount of
capital from debt. A similar disparity exists in the
treatment of corporations that fmance with retained earnings and those that pay dividends and
fmance with new equity. This Report provides
evidence that these distortions impose significant
economic costs, including reduced fmancial

12

flexibility of corporations and an inefficient
allocation of capital.
A traditional goal of integration proposals has
been to tax corporate income only once at the tax
rate of the shareholder to whom the income is
attributed or distributed. 33 Under the traditional
approach, corporate income ideally would be
taken into account when earned in determining
each individual's economic income and would be
taxed at each individual's marginal tax rate.34 To
illustrate, assume that a cOlporation has $100 of
income on which it pays $34 in corporate tax.
The corporation's shareholder has a marginal rate
of 28 percent. Traditional proposals would typically treat the shareholder as having received
income of $100, but credit the shareholder with a
tax payment of $34. Since the shareholder owes
only $28 in tax on $100 of income, traditional
proposals typically provide that the shareholder is
entitled to a $6 refund or credit against other
taxes.
Assuring that corporate income is taxed once,
but only once, does not require that corporate
income be taxed at individual rates, however.
Attaining a single level of tax-with the most
significant efficiency gains we project from any
system of integration-can be achieved with a
schedular system in which all corporate income is
taxed at a unifonn rate at the corporate level
without regard to the tax rate of the corporate
shareholder. Under the current rate structure, in
which the corporate rate is slightly higher than the
maximum individual rate, there seems little reason
to tax corporate income at shareholder rates. In
contrast, an integration proposal developed in the
late 1970s, when the maximum individual rate on
capital income of 70 percent exceeded the corporate rate of 46 percent, might well have required
taxation at shareholder rates in order to prevent
avoidance of the higher shareholder rates. 35
Neutral taxation of capital income will reduce
the distortions under the current system. 36 Economic efficiency suggests that all capital income
should be taxed at the same rate. Accordingly, we
place less emphasis than some advocates of
integration on either trying to tax corporate

13

The Case for Integration

That is, current law results in too little capital in
the corporate sector relative to that elsewhere in the
economy. Integration seeks to reduce this
distortion.

income at shareholder tax rates or on simply
trying to eliminate one level of tax on distributed
corporate income.
The prototypes advanced in this Report use the
corporation not as a withholding agent for individual shareholders (which implies ultimate taxation
at shareholder rates), but rather as a means of
collecting a single level of tax on capital income
at a uniform rate. Nevertheless, Chapter 3 discusses a shareholder allocation prototype, which
closely resembles the traditional passthrough
methods of integration. We do not recommend
adopting shareholder allocation, but it illustrates
the problems presented by an integration mechanism that imputes corporate income to shareholders and taxes it at individual rates.
A decision to adopt a schedular system for
taxation of business capital is not irreversible.
Future policymakers can, if they wish, add refund
and crediting mechanisms to achieve the traditional objective of taxing corporate income at the
individual shareholder's marginal rate, or they can
address the issue by adjusting the corporate rate to
more precisely approximate individual rates. 37
Our judgment is that neither of these courses is
necessary to achieve the principal benefits of an
integrated tax system. They are options that can
be added once the complexities of transition have
been mastered. Deferring them makes the integration prototypes examined in this Report simpler to
implement and conserves revenues.
We approach integration primarily as a means
of reducing the distortions of the classical system
and improving economic efficiency. This Report's
emphasis on enhancing neutrality in the taxation
of capital income can be summarized in four goals
for the design of an integrated tax system:
•

Integration should make more uniform the taxation
of investment across sectors of the economy. The
U.S. corporate system discourages investment in
the corporate sector relative to investment in the
noncorporate sector and owner-occupied housing.

•

Integration should make more uniform the taxation
of returns earned on alternative financial instruments, particularly debt and equity. The U.S.
corporate tax system discourages corporations from
financing investments with equity as opposed to
debt. Such a system violates the goal of neutral
taxation. Although equalizing the tax treatment of
debt and equity need not be the overriding goal of
integration, equal treatment follows from the goal
of attaining neutral taxation of capital income.

•

Integration should distort as little as possible the
choice between retaining and distributing earnings.
The U.S. corporate system discourages the payment of dividends and encourages corporations to
retain earnings or to make nondividend
distributions.

•

Integration should create a system that taxes capital
income once. Imposing double or triple taxation on
some forms of capital income while not taxing
others violates the objective of achieving neutrality
between corporate and noncorporate forms of
investment.

Integration is not a cure-all. Even an integrated system cannot attain complete neutrality with
respect to the taxation of capital income. One
reason is that integration fails to address an
important category of tax distortions: distortions
in allocating investment capital among assets.
These inter-asset distortions are important, and
reducing such distortions was an important impetus and goal of the 1986 Act. Because a corporate
income tax per se does not cause inter-asset
distortions, this Report does not directly address
them. 38
The integration prototypes analyzed in this
Report are income tax systems. The Report does
not consider non-income tax reform of corporate
taxation. For example, some economists have advocated a corporate cash-flow tax. 39 In 1984, the
Department of the Treasury rejected substitution
of a consumption-based tax for the income tax,40
and in the 1986 Act, Congress moved decisively

The Case for Integration

in the direction of strengthening the individual
income tax. So long as the individual tax base is
income, we do not believe a corporate cash-flow
tax would enhance the neutral treatment of capital
income relative to the refonns discussed here.
Revenue concerns also may prevent integration
from fully equalizing the taxation of alternative
investments. Some integration proposals would
reduce government revenue from income taxes.
Lost tax revenue must be made up either by
increasing other taxes or by reducing government
spending. Replacement taxes may create distortions and alter the distribution of tax burdens. See
Chapter 13.

14

Finally, integration does not directly address
the general question of whether the overall tax
rate on capital income, and hence the overall cost
of capital, is too high. If integration eliminates
double taxation of corporate source income, the
overall tax rate on capital income would fall,
other things being the same. Integration must be
fmanced, however, and taxes on other types of
capital income might rise. Thus, integration primarily focuses on improving the allocation of the
Nation's capital stock, but not necessarily on
reducing the overall tax rate on capital income.
As Chapter 13 documents, the benefits associated
with such improvements are nonetheless
substantial.

PART D: PROTOTYPES
INTRODUCTION
This Part presents three prototypes for implementing integration in the United States: (1) a
dividend exclusion prototype, (2) a shareholder
allocation prototype, and (3) the Comprehensive
Business Income Tax (CBIT) prototype. 1

Chapter 3 describes a shareholder allocation
integration prototype, which would extend integration to retained earnings by taxing both distributed
and retained corporate earnings at the shareholder's tax rate. Chapter 4 describes the CBIT
prototype, which, in effect, would extend a
dividend exclusion system to payments of interest
in order to equalize the treatment of debt and
equity and would tax corporate and noncorporate
businesses in the same manner. This Report
recommends the dividend exclusion prototype and
CBIT for further study. While we do not recommend adopting the shareholder allocation prototype, we include it here to illustrate how a traditional full integration or passthrough model might
be implemented and the problems it presents.

Our trading partners that have integrated their
corporate tax systems, including most European
countries, as well as Canada and Australia, have
all adopted distribution-related integration systems. Such integrated systems retain a separate
corporate level tax on undistributed earnings but
eliminate part or all of the corporate level tax on
corporate earnings distributed to shareholders as
dividends. Distribution-related integration can be
accomplished by excluding dividends from shareholders' income (a dividend exclusion system), by
allowing shareholders a credit for corporate level
taxes (an imputation credit system), or by allowing corporations a deduction for dividends (a
dividend deduction system).

Each of these prototypes would move the U. S.
tax system in the direction of more neutral taxation of corporate income and, in so doing, would
reduce significantly tax-induced distortions in the
allocation of capital. The prototypes generally are
structured to implement OUf recommendations on
four major issues:

After considering each of these three alternatives, we determined that a dividend exclusion
system would implement in a relatively simple
and straightforward manner our policy recommendations. The flexibility of an imputation credit
system in responding to important policy issues,
such as the treatment of tax preferences, foreign
taxes, and tax-exempt and foreign shareholders
under integration, does not, in our view, outweigh
its complexity in implementation. A dividend
deduction system would produce results in many
cases contrary to our policy recommendations.
Chapter 2 outlines a dividend exclusion prototype,
and Chapters 11 and 12 discuss the imputation
credit and dividend deduction alternatives. Because an imputation credit system is the mechanism of corporate tax integration most frequently
used abroad, we discuss an imputation credit
prototype in considerable detail in Chapter 11. 2
The Report also examines two integration
systems that are not distribution-related.
15

•

The benefit of comorate level tax preferences
should not be extended to shareholders. Tax preferences, e. g., exempt state and local bond interest
and accelerated depreciation, may reduce the
corporate level tax, but current law does not extend
corporate level tax preferences to shareholders.
When corporate earnings sheltered by preferences
are distributed to shareholders, they are currently
taxed. Integration of the corporate income tax need
not become an occasion for expanding the benefits
of tax preferences. Therefore, we do not recommend extending corporate level tax preferences to
shareholders under integration, and we have attempted to develop administrable rules to reach this
result whenever we could do so in a manner
compatible with the prototype. See Chapter 5.

•

Integration should not reduce the total tax collected
on corporate income allocable to tax-exempt investors. Under current law, tax-exempt organizations
holding corporate stock, in fact, are not exempt
from the corporate level tax imposed on corporate
equity investments. Because corporate income is

Prototypes

subject to tax at the corporate level regardless of
the exempt status of a shareholder, a tax-exempt
organization is exempt only from the shareholder
level tax. Integration presents the fundamental
question whether under an integrated tax this
treatment should continue, or whether integration
should reduce the total taxes paid on corporate
income allocable to tax-exempt entities. This
Report recommends, in general, retaining the
current level of taxation of corporate equity income
allocable to tax-exempt shareholders. See Chapter 6. The CBIT prototype would introduce a
corporate level tax on income allocable to taxexempt bondholders as well. See Chapter 4.
•

Integration should be extended to foreign shareholders only through treaty negotiations, not by
statute. The United States generally imposes two
levels of tax on foreign equity investment in U.S.
corporations (inbound investment). Thus, the
United States taxes the business profits of foreign
owned domestic companies similarly to the profits
of U.S. owned companies and also imposes significant withholding taxes on dividends paid to foreign
investors. The basic issue that an integration
proposal must resolve for inbound investment is
whether, by statute, the United States should
continue to collect two levels of tax on foreign
owned corporate profits or whether foreign investors should receive benefits of integration similar to
those received by domestic investors. This Report
generally recommends that foreign shareholders not
be granted integration benefits by statute, but

16

instead that this issue be addressed through treaty
negotiations in order to achieve reciprocity. Most
of the major trading partners of the United States
that have adopted integrated corporate tax regimes
have followed this approach. See Chapter 7 and
Appendix B.
•

Foreign taxes paid by U.S. comorations should not
be treated, by statute, identically to taxes paid to
the U.S. Government. The United States permits
U. S. corporations to credit foreign taxes against
U.S. taxes on foreign source income (outbound
investment) but taxes shareholders on the distribution of such income without regard to the foreign
taxes paid on that income. Treating foreign and
U.S. corporate level taxes equally under an integrated system by statute would significantly reduce
the current U. S. tax claim against foreign source
corporate profits and often would completely
exempt such profits from u.S. taxation at both the
corporate and shareholder levels. Such unilateral
action would result in a significant departure from
the current allocation of tax revenues between the
source and residence country. We therefore recommend that foreign taxes not be treated, by statute,
the same as U.S. taxes. As a consequence, the
prototypes generally would retain the foreign tax
credit at the corporate level but would continue to
tax foreign source income when it is distributed to
shareholders. Extending the benefits of integration
to foreign source income is more properly accomplished in the context of bilateral treaty
negotiations. See Chapter 7.

CHAPTER

2:

DIVIDEND EXCLUSION PROTOTYPE

2.A INTRODUCTION AND
OVERVIEW OF PROTOTYPE

rules but, when distributed, would be exempt
from tax at the shareholder level. 4

The dividend exclusion prototype set forth in
this chapter would, with few changes in current
law, implement many of this Report's key policy
recommendations. 1 The principal advantage of
the dividend exclusion prototype is its simplicity
and relative ease of implementation. We considered an imputation credit prototype that would
achieve results similar to the dividend exclusion
prototype but at the cost of additional complexity,
including an entirely new regime for taxing
corporate distributions. Although we do not
recommend an imputation credit system, such a
system is described in Chapter 11 because it
provides useful background for understanding the
dividend exclusion prototype. A summary of the
prototype follows.

Corporate Shareholders. A corporate shareholder would exclude from income excludable
dividends received and would add the amount of
such dividends to its EDA. The prototype retains
the current dividends received deduction for
taxable dividends.
Tax Preferences.· The prototype retains the
corporate tax preferences available under current
law and the corporate alternative minimum tax.
To avoid extending corporate tax preferences to
shareholders, the prototype permits shareholders
to exclude only those dividends deemed made out
of income that has been taxed fully at the corporate level. Thus, corporate dividends paid to
shareholders out of preference income would
continue to be taxable as under current law.
Mechanically, this is accomplished once the
corporation's supply of fully-taxed income (as
reflected in the EDA) is exhausted, by making
additional dividends taxable to shareholders. 5 See
Section 2.B. As under current law, preference
income distributed to tax-exempt shareholders
would escape taxation at both the corporate and
shareholder levels.

Mechanics. Under the dividend exclusion
prototype, corporations would continue to calculate their income under current law rules and pay
tax at a 34 percent rate. 2 Shareholders receiving
corporate distributions treated as dividends under
current law, however, generally would exclude
the dividends from gross income. The prototype
requires corporations to keep an Excludable
Distributions Account (EDA) to measure the
amount of dividends that can be excluded by
shareholders--essentially an amount on which
corporate taxes have been paid. Thus, the dividend exclusion prototype would apply the corporate tax rate of 34 percent to both distributed and
retained income but would eliminate the shareholder level tax on dividends paid from fullytaxed corporate income. 3 All other distributions,
e.g., interest and returns of capital, would be
taxed in the same manner as under current law.

Foreign Source Income. The prototype retains
the current foreign tax credit system, including
the corporate level indirect foreign tax credit for
taxes paid by foreign subsidiaries. The prototype,
however, does not treat foreign taxes the same as
U.S. taxes in determining the EDA, with the
consequence that, as under current law, distributions of foreign earnings that have been shielded
by the foreign tax credit at the corporate level are
taxable to shareholders when distributed. 6

Tax-Exempt Shareholders. The dividend
exclusion prototype would automatically retain the
current level of taxation of corporate income
earned on equity capital supplied by tax-exempt
shareholders. Income from equity investments by
tax-exempt organizations would be taxed at the
corporate level under the current corporate tax

Foreign Shareholders. The prototype retains
the current 30 percent statutory withholding tax
on dividends. In addition, it retains the branch
profits tax on earnings considered repatriated from
U.S. branches of foreign corporations. Thus, as
under current law, inbound investmerit is subject
to two levels of U. S. tax, with reductions in the
17

Prototypes

18

rate of withholding tax negotiated through tax
treaties.?

tax bias against distributed earnings thus would
remain for foreign investors.)8

Capital Gains and Share Repurchases. Chapter 8 discusses the treatment of capital gains on
sales of corporate stock and the treatment of share
repurchases.

2.B

Structural Issues. The dividend exclusion
prototype does not require any major changes to
current rules concerning the tax treatment of
corporate acquisitions. Adopting the prototype
does, however, require consideration of rules for
the carryover or separation of corporation EDA
balances in liquidations and tax-free corporate
reorganizations.
Impact on Tax Distortions. Table 2.1 illustrates the impact of the dividend exclusion prototype on the three distortions integration seeks to
address: the current law biases in favor of corporate debt over equity fmance, corporate retentions
over distributions, and the noncorporate over the
corporate form. The only difference between the
current law treatment of nonpreference, U.S.
source business income and its treatment under
the dividend exclusion prototype is the taxation of
corporate equity income distributed to individuals.
Since exclusion of dividends by individuals would
remove the individual level tax, the total tax rate
on distributed earnings would be reduced to the
corporate rate (te, generally 34 percent), except
for the influence of investor level taxes on foreign
investors. This reduction would narrow (but not
eliminate) the rate differential between distributed
corporate and noncorporate equity income and
between corporate equity income and interest.
These reductions in differentials would help
reduce the debt-over-corporate-equity-fmance and
noncorporate-over-corporate fonn distortions. The
tax rate on undistributed corporate equity income
would now be higher for individuals than the rate
on distributed corporate equity income, so the tax
bias against corporate distributions would likely
be reversed, in the absence of a DRIP. See
Chapter 9. For tax-exempt and foreign investors,
there would be no change in the tax treatment of
nonpreference, U.S. source business income. (The

THE NEED FOR A
LIMITATION ON
EXCLUDABLE DIVIDENDS

In General
An exclusion from shareholder level tax for all
dividends received not only would eliminate the
Table 2.1
Total U.S. Tax Rate on a Dollar of
NonPreference, U.S. Source Income from a
U.S. Business Under Current Law and the
Dividend Exclusion Prototype

Type of Income

Dividend
Exclusion
Prototype

Current Law

t. Individual Investor is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

to +(l-tJt;.
tt +(l-tJtg

tt
tt+(l-tJtg

~

~

~

~
~

1;

D. Tax Exempt Entity is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

m.

Foreign Investor is Income Recipient

Corporate Equity:
Distributed
tc + (1 - tJt wo
to
Undistributed
Noncorporate Equity
lwN
Interest
twJ
Rents and Royalties
twR.
Department of the Treasury
Office of Tax Policy

tc + (1 - tJtwo
tc

lwN
twJ
lwR

tc = U.S. corporate income tax rate.
= U.S. individual income tax rate.
t, = U.S. effective individual tax rate on capital gains.
two. tWN • !wI. twR = U.S. withholding rates on payments
to foreigners of dividends, noncorporate equity
income. business interest, and rents and royalties,
respectively. Generally varies by recipient, type of
income, and eligibility for treaty benefits and may be
zero.
~

19

double tax on distributed corporate income, but
also would eliminate the current shareholder level
tax that serves as the only U.S. tax on distributed
income that has been sheltered from corporate
level tax by preferences and on distributed foreign
source income that has borne only foreign taxes.
To prevent the dividend exclusion system from
extending preferences to shareholders and to
ensure that foreign source income that has not
borne U. S. tax at the corporate level is subject to
tax at the shareholder level when distributed, the
dividend exclusion prototype limits the amount of
dividends that can be excluded at the corporate
level to an amount that has been subject to U. S.
tax at the corporate level. Thus, as under current
law, corporate preference income would generally
remain free of tax until distributed and, when
distributed, would be taxed at shareholder rates.
Foreign source income sheltered by foreign tax
credits at the corporate level also would continue
to be taxed when distributed to shareholders. See
Chapters 5 and 7.
The prototype treats dividends as made fIrst
from a corporation's fully-taxed income, rather
than from preference or foreign source income.
Stacking dividends fIrst against fully-taxed income
should permit many corporations to continue their
current dividend policy while paying excludable
dividends. Even corporations with substantial
preference or foreign source income can continue
to pay dividends without incurring any additional
corporate level tax, although the dividends would
be taxable at the shareholder level. We considered, but rejected, the alternative of imposing a
nonrefundable compensatory tax" at the corporate level on distributions of preference or foreign
source income. 9 See Chapter 5. A nonrefundable
compensatory tax not only reduces cash available
to pay dividends but also increases the total tax
burden on dividends paid to tax-exempt and
foreign shareholders as well as to any shareholder
taxed at less than a 34 percent rate; on the other
hand, imposition of such a tax would permit
uniform dividend exclusion. On balance, concern
that a compensatory tax would distort the dividend
decisions of corporations, particularly those with
large numbers of tax-exempt or foreign shareholders, by requiring them to pay an extra tax to
II

Prototypes

maintain their current dividend policy, led us to
the alternative described here. Section 11.B
discusses a compensatory tax in more detail.
The prototype retains the corporate alternative
minimum tax (AMT), which functions, as under
current law, to curb the excessive use of tax
preferences at the corporate level. The prototype
treats AMT as taxes paid for purposes of determining the corporation's supply of fully-taxed
income, but effectively converts income taxed at
the 20 percent corporate AMT rate to a smaller
amount of income taxed at the regular 34 percent
rate. 10

Identifying Distributed Preference
Income: the EDA
To determine whether dividends are paid out
of fully-taxed income or preference income, the
prototype requires corporations to maintain an
Excludable Distributions Account (EDA).
Amounts included in the EDA are considered
llfully-taxed income." Dividends paid are stacked
fIrst against fully-taxed income.
As a mechanical matter, the EDA measures a
corporation's supply of fully-taxed income based
on the taxes actually paid by the corporation. The
corporation simply tracks actual corporate taxes
paid and then converts that amount into an equivalent amount of after-tax income taxed at a 34
percent rate, using the following formula:
Annual additions to EDA

u.s.
[
+

=

tax paid for taxable year
.34

u.s.

tax paid for taxable year ]

excludable dividends received

Thus, for each $34 of taxes paid (whether regular
corporate tax or AMT), the corporation may pay
$66 of excludable dividends, i.e., each $1 of
corporate taxes paid supports $1.94 of excludable
dividends or each dollar of excludable dividends
must be supported by at least $0.52 of corporate
taxes paidY The effect of calculating additions
to the EDA at 34 percent is to ensure that distributed income has been taxed at the full corporate
rate, even though, if taxable to shareholders, the

Prototypes

dividend would be taxed, at most, at the 31
percent maximum individual rate.
The EDA increases when a corporation pays
taxes (including estimated taxes) or, as described
under "Corporate Shareholders" below, receives
an excludable dividend from another corporation.
The EDA decreases when a corporation pays a
dividend or receives a refund of taxes paid.
Dividends paid when the EDA has been reduced
to zero are treated as paid from preference income
and are fully includable in shareholder's income.

Example. A corporation with a zero initial EDA
balance earns $75 of taxable income and $25 of
exempt income. The corporation pays $25.50 of
corporate tax and has $74.50 available for distribution to shareholders. The $25.50 of tax supports
the addition of $49.50 to the corporation's EDA
($25.50/.34-$25.50). If the corporation actually
distributes $74.50, only $49.50 of the dividend is
excludable, because the EDA balance is $49.50.
The remaining $25 represents a distribution of
preference income that is fully subject to tax at the
shareholder level.

The prototype requires corporations to report
annually to shareholders and the IRS the excludable and taxable portions of dividends. In the
preceding example, the corporation would report
the ftrst $49.50 distributed as an excludable
dividend and the next $25 distributed as a taxable
dividend. Shareholders would include taxable
dividends in income as under current law. Corporations also would report to the IRS annually the
adjustments to and balance in the EDA.
Adjustments to a corporation's tax liability for
a prior year are reflected as adjustments to the
corporation's EDA in the current year. Making
audit adjustments to the EDA in the current year
avoids the problem of recharacterizing dividends
paid in prior years. 12 An increase in a prior
year's tax liability increases the EDA in the year
the adjustment is made and the additional tax is
paid, and a decrease in a prior year's tax liability,
e.g., through carryback of a net operating loss,
gives rise to a refund and requires a corresponding reduction in the EDA in the year the refund is
received. Refunds would be limited to the balance

20

in the corporation's EDAY Refunds in excess of
the EDA balance would be carried forward to be
applied against future corporate taxes. Similarly,
an NOL carryback would not be permitted to
reduce the EDA below zero; losses in excess of
this amount would be carried forward. 14

Corporate Shareholders
Current law limits the imposition of multiple
levels of corporate taxation by permitting corporate shareholders to deduct some or all of their
dividends received from domestic corporations,
depending on the degree of affiliation with the
distributing corporation.
Under the prototype, distributions from an
EDA are excludable from the income of any
shareholder, including a corporate shareholder.
The recipient corporation adds the amount of
excludable dividends it receives to its EDA. This
prevents the imposition of a second level of tax
when excludable dividends are redistributed to the
shareholders of the recipient corporation.
The prototype retains current law for taxable
dividends (dividends in excess of the distributing
corporation's EDA) received by corporations.
Thus, taxable dividends received from a U.S.
corporation (and a portion of dividends from
certain foreign corporations engaged in business
in the United States) would entitle the recipient to
a dividends received deduction (DRD). A recipient corporation allowed only a 70 or 80 percent
DRD would pay tax on the remainder of the
dividend. Any taxes paid on the dividend would
be added to the EDA, determined in accordance
with the general formula for computing additions
to the EDA set forth above. To the extent the
recipient corporation qualiftes for the DRD, the
prototype defers the investor level tax on preference income until it is ultimately distributed to
individual shareholders. 15

Anti-abuse Rules
We have considered whether special rules are
necessary to limit a corporation's ability to target
(or "stream ") excludable dividends to taxable

21

shareholders and otherwise taxable dividends to
tax-exempt shareholders. Streaming undercuts the
prototype's preservation of the current level of
taxation of cotporate equity income paid to taxexempt and foreign shareholders by denying
refunds of cotporate taxes paid. On the other
hand, tax-exempt and foreign investors may enter
into a variety of ordinary business structures that
enable them to receive income not taxed at the
corporate level, e.g., by holding debt instead of
equity.16 These arrangements are permitted under current law, and they are not limited under
the prototype. The ability to arrange a capital
structure to minimize taxes emphasizes the point
that eliminating the double tax on dividends will
not, by itself, eliminate the tax system's current
bias in favor of debt fmancing. A more comprehensive approach such as CBIT (described in
Chapter 4) is required to address this systemic
bias.
In the dividend exclusion prototype, concerns
about streaming are balanced against the cost of
complexity by restricting only a limited class of
streaming transactions. In the prototype, current
law rules that apply in analogous situations are
extended. 17 First, the prototype adopts a 45 day
holding period requirement for dividends to be
excludable to prevent tax-exempt shareholders
from routinely selling stock to taxable shareholders just before payment of an excludable dividend
and then repurchasing the stock. 18 Second, depending on the treatment of capital gains, the
prototype could extend application of the extraordinary dividend rules of IRC § 1059 to excludable
dividends in order to prevent taxable shareholders
from "stripping" excludable dividends. 19 The
existing rules of IRC § 305 also may be useful in
preventing other kinds of streaming. 20
Rules like those of IRC §§ 382 through 384,
which limit the use of net operating losses and
other corporate attributes after a change in ownership, are not included in the prototype. An EDA
balance represents fully-taxed cotporate income,
and, in general, integration should prevent that
income from being taxed again at the shareholder
level. The issue is difficult, however, because
allowing unlimited use of EDA balances may

Prototypes

permit an acquiror to use a target's EDA balance
to defer or eliminate tax on the acquiror's preference income. 21 On balance, we decided that
extending the rules would create considerable
complexity and may not provide any substantial
benefit in addition to the rules discussed above. 22
If significant evidence of abuse develops, ownership change limitation rules could be adopted at
that time. 23
Policymakers may wish to consider whether
interest expense paid on debt incurred to purchase
cotporate stock should be disallowed under rules
like those of IRC § 265(a). In a dividend exclusion system, cotporate earnings generally bear
only one level of tax. See the example in Section 4.G.24 While the potential for rate arbitrage
exists under current law, it may be less of a
problem where only one of two levels of tax is
eliminated. The issue is a difficult one, however,
because disallowing an interest deduction for
interest paid to a taxable lender will result in the
imposition of two levels of tax. Moreover, in
CBIT, we recommend extending the interest
disallowance rules with respect to CBIT debt and
equity. See Section 4. G. There may be less
pressure to adopt the same rule in the dividend
exclusion prototype, however, because it does not
equate the treatment of debt and equity. 25

2.C

FOREIGN SOURCE INCOME

Under the prototype, U.S. individual shareholders would continue to include in income
dividends received from foreign corporations and
to claim a foreign tax credit for any foreign
withholding taxes imposed on the dividend.
Similarly, U.S. cotporate shareholders owning
less than 10 percent of a foreign corporation's
voting stock (the threshold requirement for the
U. S. cotporation being eligible to claim an indirect foreign tax credit under IRC § 902) would
include in income dividends from the foreign
cotporation and would claim a foreign tax credit
for foreign withholding taxes. The corporate
shareholder would not add any amount to its EDA
to reflect foreign income taxes paid by the foreign
cotporation or foreign withholding taxes on
dividends.

Prototypes

U. S. corporate shareholders owning at least 10
percent of a foreign corporation's voting stock
would continue to include in income dividends
from the foreign corporation and to claim both a
direct credit for foreign withholding taxes and an
indirect foreign tax credit with respect to such
dividends under the rules of IRe § 902 of current
law, subject to the foreign tax credit limitation in
IRe § 904. Under these provisions, the corporate
shareholder receives a credit, subject to certain
limitations, for foreign income taxes paid by the
foreign corporation with respect to earnings out of
which the dividends are paid. A U.S. corporation
would increase its EDA only by an amount that
reflects the residual U.S. tax (if any) imposed on
the dividend income. Thus, absent any residual
U.S. tax (and any EDA balance attributable to
U.S. tax on U.S. source income), distributions
out of foreign source income taxed abroad, in
effect, would be taxed at the shareholder level as
under present law.
U. S. corporations with foreign branch operations, or which receive interest, rents, royalties,
or other income from foreign sources, would
continue to be subject to current U. S. tax on their
foreign source income with a credit under IRe §
901 for foreign income taxes. As with earnings of
foreign subsidiaries, the U.S. corporation would
increase its EDA only to reflect the amount of
any residual U.S. tax imposed on the foreign
source income.
Although we do not recommend a statutory
rule permitting additions to an EDA based on
payment of foreign taxes, consideration might be
given to granting authority to enter into tax
treaties that treat foreign taxes like U.S. taxes,
where reciprocity exists. 26 Treating foreign taxes
like U.S. taxes would allow a U.S. corporation
doing business in a treaty jurisdiction to pay
excludable dividends to its U.S. shareholders even
if its income was entirely shielded from U.S. tax
by foreign tax credits. 27

22

2.D

LOW-BRACKET
SHAREHOLDERS

Taxing corporate income at a uniform rate at
the corporate level significantly reduces the
complexity of the dividend exclusion (and CBIT)
prototypes and reduces the burdens of transition to
a new system because refund and credit provisions
are not required to deal with "overcollections of
tax from individual taxpayers with marginal rates
lower than the 34 percent corporate rate. While
this simplification concern has been a major factor
in our decision to recommend a schedular system,
inspection of the available data also suggests that
the adoption of a schedular system will not result
in significantly higher taxation of corporate
income than the use of individual rates for most
taxable shareholders. The data indicate that
approximately two-thirds of corporate dividends
paid to taxable individual shareholders, i.e.,
shareholders who are U.S. citizens or residents,
are paid to individuals with average marginal tax
rates of more than 25 percent.
It

It might at first appear that corporate income
distributed to individuals with average marginal
tax rates of less than 25 percent should be taxed
at a lower rate, because a lower marginal rate
indicates a lower income and, inferentially, less
ability to pay. On the other hand, low-bracket
shareholders who receive dividends clearly own
some property, i.e., stock, and it is not clear
whether their low taxable incomes accurately
reflect their ability to pay. 28 Accordingly, the
dividend exclusion and CBIT prototypes do not
contain provisions reducing the rate of tax collected on corporate income distributed to low-bracket
shareholders.

If policymakers desired to tax distributed
corporate income at shareholder rates, a dividend
exclusion system could allow a tax credit that
would refund all or part of the excess tax collected at the corporate level. To refund fully the

23

difference between 34 percent and the shareholder
rate, the amount of the tax credit would equal (1)
the amount of the dividend received, grossed up
at the 34 percent rate, multiplied by (2) the
difference between 34 percent and the shareholder's marginal tax rate. Each shareholder
would calculate his own credit based on a formula
(or a set of tables) and his marginal tax rate. 29
Example. A corporation earns $100, pays tax of
$34, and distributes $66 to a shareholder in the 15
percent marginal tax bracket. The shareholder
would owe no tax on the dividend and would be
allowed a tax credit of$19 «$66/.66) X(.34- .15»),
which could be used to offset other income.

Such credits would be allowed only for
excludable dividends. 30 Allowing a shareholder
tax credit for taxable dividends (dividends considered made out of preference income) would
confer a shareholder level benefit for corporate
level tax that had not been paid.

2.E

INDIVIDUAL ALTERNATIVE

MINIMUM TAX
Historically, individuals have been subject to
a minimum tax to ensure that at least a small
amount of tax is paid on an individual's economic
income and to respond to public perceptions that
permitting high-income individuals to pay little or
no income tax undermines the fairness of the tax
system. The exclusion for dividends described
here might result in some high-income individuals
paying little or no tax at the individual level, thus
raising issues of public perception. The EDA,
however, operates to ensure that any dividends
excludable from an individual's gross income
have already been subject to one level of tax at
the corporate level. The investor's income tax has
been prepaid at the corporate level at the 34
percent corporate rate, which exceeds the top
individual rate. Including excludable dividends in
the individual AMT would serve only to reinstitute a double tax on dividends and would

Prototypes

undermine to some extent the basic goals of this
system of integration.

2.F

STRUCTURAL ISSUES

This section discusses several areas of current
law that should be modified to reflect adoption of
the dividend exclusion prototype. This section
does not provide a comprehensive analysis of the
technical changes required but instead raises
issues for further development.

Corporate Acquisitions
The dividend exclusion prototype retains the
basic rules governing the treatment of taxable and
tax-free corporate asset and stock acquisitions.
The prototype permits taxable asset acquisitions to
be made with only a single level of tax. Corporate
tax paid on gain recognized on the sale of assets
would be treated like any other corporate level tax
payment and would support a corresponding
addition to the EDA, thus generally allowing a
tax-free distribution of proceeds to shareholders
when the corporation liquidates. Upon liquidation,
shareholders would, as under current law, generally recognize gain to the extent liquidation
proceeds exceed share basis. A shareholder's gain
would be excludable, however, to the extent of a
proportionate share of the liquidating corporation's EDA.31 Stock acquisitions may face a
higher tax burden than asset acquisitions if capital
gains on corporate stock that are attributable to
retained earnings are taxed in full at shareholder
rates. See Chapter 8.
The prototype retains current law rules that
treat a qualifying corporate reorganization as taxfree at the corporate level (with the target's tax
attributes, including its asset bases, carrying over
to the acquiror) and at the shareholder level. 32
Additional rules would be needed to coordinate
the reorganization provisions with the dividend
exclusion prototype. For example, the EDA of a
corporation acquired in a reorganization should

Prototypes

generally carry over to its successor. In a divisive
reorganization, the EDA should be divided proportionately between the corporations. 33

Earnings and Profits
The prototype retains the current law rules
that treat a distribution as a dividend only to the
extent of current and accumulated earnings and
profits. 34 Distributions that exceed earnings and
profits are treated as a return of capital to the
extent of a shareholder's basis and then as gain on
the disposition of the stock. 35 Under the
prototype, only a distribution that is made out of
the corporation's EDA is eligible for exclusion at
the shareholder level. If a distribution is made
when a corporation has no EDA balance but has
earnings and profits, it is a taxable dividend; if
the corporation has no earnings and profits, the
distribution is treated as a return of capital to the
extent of the shareholder's basis and then as gain.
Some commentators have argued that the
earnings and profits rules should be eliminated
under current law, essentially arguing that the
complexity of the earnings and profits rules
outweigh any benefits that may result. 36 In
general, at least two alternatives to the earnings
and profits rules are possible. All nonliquidating
distributions to shareholders could be treated as
dividends, except where a distribution results in a
reduction in capital (stated or surplus) for
corporate law purposes. Alternatively, all
nonliquidating distributions to shareholders could
be treated as dividends, subject generally to
current rules allowing basis recovery with respect
to transactions where a shareholder's interest in
the corporation is reduced or terminated.
Under the dividend exclusion prototype, as
under current law, replacing the earnings and
profits rules with either of the alternative rules
would simplify the determination of whether a
corporate distribution is a dividend for tax
purposes. 37 However, although the simplification
benefits of eliminating the earnings and profits
rules are important, we conclude that adoption of
the dividend exclusion prototype, by itself, neither

24

compels the elimination of the rules nor demands
their retention. 38 Thus, under the dividend exclusion prototype, earnings and profits would continue to provide a rough measure of whether, for
purposes of determining the shareholder level tax,
a distribution represents income from, or a return
of, a shareholder's investment. 39

Dividend Reinvestment Plans (DRIPs)
Distributed earnings are subject to only one
level of tax under the dividend exclusion prototype, but retained earnings may be subject to a
greater tax burden to the extent that they increase
the value of stock and are taxed as capital gains.
See Chapter 8. A dividend reinvestment plan, or
DRIP, is one way for corporations to extend the
benefits of integration to retained earnings. In a
dividend exclusion system, a DRIP would allow
a corporation to treat its shareholders as if they
had received an excludable cash dividend and had
reinvested it in the corporation. The shareholder's
basis would be increased to reflect the amount of
the deemed dividend, ensuring that the shareholder would not be taxed on appreciation due to retained fully-taxed earnings when the stock is sold.
Example. A corporation earns $100, pays $34 in
tax, and adds $66 to its EOA. The corporation
declares a deemed dividend of $66 and reduces the
EOA by $66, and the shareholders increase their
share basis by $66.

Chapter 9 discusses DRIPs.

2.G PENSION FUNDS
Under current law, contributions to qualified
pension plans are generally deductible by the
employer and are not currently includable by the
employee. The employee is generally taxed only
when distributions of benefits are made. The
deduction provided to the employer combined
with the deferral of income to the employee until
benefits are paid effectively exempts the investment earnings on the contribution from tax. 40
Thus, pension fund income from investments in
stock bear only one 'level of tax-the corporate
tax paid by the corporation.

25

The dividend exclusion prototype does not
change this treatment. Under the prototype, most
dividends are excludable by shareholders. Thus,
if dividends were received directly by plan beneficiaries, they would be tax-free. The earnings of
pension plans would be taxed when distributed,
however, even if the distributions were attributable to excludable dividends received by the plan
on its investments. Just as under current law,

Prototypes

however, the combination of the employer's
deduction for contributions and the deferral of the
beneficiary tax until earnings are distributed
ensures that earnings on pension fund investments
in stock are taxed only once. Although retaining
the current treatment of pension funds in a dividend exclusion system perpetuates some bias
against investments in stock by pension plans, the
disincentive is no greater than under current law.

CHAPfER

3: SHAREHOLDER ALLOCATION PROTOTYPE

3.A INTRODUCTION

We nevertheless discuss the shareholder
allocation prototype in some detail because it is
the integration system advanced by advocates of
traditional full integration proposals, which generally would treat a corporation as a conduit and
allocate income to shareholders as earned. This
chapter shows how a passthrough model of integration might be modifted to confonn as closely
as possible with our policy recommendations and
identiftes some of the most difficult administrative
issues. 4

The dividend exclusion prototype and other
distribution-related systems of integration provide
relief from double taxation only for distributed
income. As a consequence, they may create an
incentive for corporations to distribute, rather
than retain, earnings at least to the extent that
fully-taxed income can be distributed to taxable
shareholders. 1 In contrast, the shareholder allocation prototype would extend integration to retained
earnings by allocating a corporation's income
among its shareholders as the income is earned.
Shareholders would include allocated amounts in
income, with a credit for corporate taxes paid,
and would increase the basis in their shares by the
amount of income allocated, less the amount of
the credit. Distributions would be treated as a
return of capital to the extent of a shareholder's
basis and, thereafter, as a capital gain. 2

In contrast to a pure passthrough model of
integration, the shareholder allocation prototype
(1) does not pass through losses to shareholders,
(2) retains the corporate level tax, which would
assume a function similar to a withholding of
shareholder level tax, (3) requires corporations to
report to shareholders only an aggregate income
amount, rather than separately report all items,
and (4) does not exterid integration benefits to taxexempt shareholders or to foreign shareholders
except by treaty.

Thus, the shareholder allocation prototype
treats retained and distributed earnings equally. We
do not favor adopting the shareholder allocation
prototype, however, because of the policy results
and administrative complexities it produces. As
examples of policy problems, if it is to retain
parity between retained and distributed earnings,
the shareholder allocation prototype must extend
tax preferences to shareholders and exempt from
U. S. tax foreign source income that has borne no
U.S. tax. While the shareholder allocation prototype reduces (but does not eliminate) current
law's bias in favor of debt fInancing, the same is
true of the dividend exclusion prototype, which is
a simpler regime. 3 Administratively, shareholder
allocation integration would require corporations
and shareholders to amend governing instruments
for outstanding corporate stock to provide for
income allocations, would require corporations to
maintain capital accounts similar to those used
under the partnership rules, and could create
signifIcant reporting difficulties for shareholders
who sell stock during a year and for corporations
that own stock.

3.B

OVERVIEW OF THE
SHAREHOLDER
ALLOCATION PROTOTYPE

The shareholder allocation prototype continues
to treat the corporation as a separate entity for
many reporting and auditing purposes. All tax
items, including different types of income, deductions, losses and credits, are aggregated at the
corporate level rather than being passed through
to shareholders. To enhance compliance and
mitigate shareholder cash flow problems, the
prototype requires the corporation to pay income
taxes at regular corporate rates as under current
law. The corporation allocates its taxable income,
as reported for regu~ar tax purposes, among its
shareholders. The shareholders include the allocated amounts in income and credit corporate taxes
paid and corporate tax credits claimed (including
the foreign tax credit and other corporate tax
credits) against their tax liability. Shareholders

27

Prototypes

with marginal tax rates less than the corporate
rate may use excess credits to offset tax liability
on other income but may not obtain refund of the
credit.
Example. A corporation has $100 of taxable
income and owes $31 of corporate level tax. S The
corporation also is entitled to a tax credit (e.g., a
low-income housing credit) of $5. Thus, the corporation pays $26 in tax. The corporation allocates
$100 of taxable income among its shareholders,
together with $31 of tax credits ($26 tax actually
paid plus $5 tax credit).6

Shareholders would increase share basis by (1)
the amount of taxable income allocated to them,
after subtracting corporate taxes paid (including
corporate tax credits), 7 and (2) tax-exempt income. See Section 3.E. Thus, in the examples
noted above, the shareholders' collective basis
increases by $69. Share basis would decrease by
the amount of distributions. Distributions to
shareholders are treated as a nontaxable return of
capital to the extent of a shareholder's basis in his
stock. Distributions in excess of basis would be
treated as gain recognized on the sale of the
stock, which would generally be capital gain. 8
Corporate losses and excess corporate tax
credits would not flow through to shareholders but
could be carried forward at the corporate level.
Losses or excess tax credits could not be carried
back to claim a refund of corporate tax, because
that tax would already have been made available
to offset shareholder tax on allocated income. 9
Current law limitations on the use and transfer of
corporate losses and other tax attributes would
continue to apply at the entity level.
Mechanics. Corporations would allocate
income and taxes paid to the holder of stock on a
quarterly record date. A corporation with multiple
classes of stock would allocate tax items in accordance with the terms of the stock certificate,
which would designate the share of income to be
allocated to each class of stock. See Section 3.F.
AU. S. corporate shareholder would allocate to its
own shareholders its share of the second
corporation's taxable income and tax credits.

28

Intercorporate holdings may create difficult
reporting issues. See Section 3.H.
The mechanics of shareholder allocation
integration can be illustrated with a simple
example.
Example. A corporation has three classes of common stock, the terms of which provide for the
allocation of 30 percent of corporate income to
Class A, 20 percent to Class B, and 50 percent to
Class C. The corporation has taxable income of
$100, pays $31 in corporate tax and pays a $10
dividend with respect to Class C stock. The shareholder integration prototype allocates the income
and the credit to each class of stock based on the
respective percentages (so, for example, Class C
would be allocated income of $50 and credits of
$15.50). Within each class of stock, each share
receives a pro rata amount. 10 Holders of Class A
stock would collectively increase their basis by
$20.70 (.30X($100-$31)), holders of Class B
stock would increase their basis by $13.80 (0.20x
($100-$31)), and holders of Class C stock would
collectively increase their basis by $24.50 (.5 x
($100-$31)-$10).

Tax-Exempt Shareholders. To preserve one
level of tax on corporate income allocable to taxexempt shareholders~ credits for corporate tax
would not be refundable to tax-exempt shareholders. See Section 3.1.
Tax Preferences. The shareholder allocation
prototype would generally extend corporate level
tax preferences to shareholders. See Section 3.E.
Foreign Source Income and Foreign Shareholders. A U.S. corporation would pay corporate
tax on its worldwide income and, where permitted
under current law, could claim a foreign tax
credit for foreign taxes paid directly and by a
foreign subsidiary. The corporation would then
allocate its taxable income to shareholders and the
foreign tax credit would be creditable by shareholders. Section 3.1 discusses the difficulty of
implementing appropriate shareholder level foreign tax credit limitation rules. Income of a
foreign corporation would be includable in income
of U.S. corporate shareholders only as under

29

current law, i.e., generally when distributed. The
shareholder allocation prototype does not pennit
foreign shareholders, except pursuant to tax.
treaties, to claim a refund of the corporate tax or
to use the credit for corporate tax to offset the 30
percent (or lower) withholding tax levied on
dividends (which would continue to apply). Such
treaty benefits should be provided only in return
for reciprocal benefits.
Capital Gains and Share Repurchases. Chapter 8 discusses the treatment of capital gains on
sales of corporate stock and the treatment of share
repurchases.
Structural Issues. Section 3.G discusses the
problems of midyear sales of stock, and Section 3.H discusses the reporting difficulties that
arise in the case of intercorporate stock ownership. We do not discuss further the treatment of
corporate taxable and tax-free acquisitions under
the shareholder allocation prototype.
Impact on Tax Distortions. Table 3.1 illustrates the impact of the shareholder allocation
prototype on the three distortions integration seeks
to address: the current law biases in favor of
corporate debt over equity fmance, corporate
retentions over distributions, and the noncorporate
over the corporate fonn. For nonpreference, U.S.
source income received by individuals, the shareholder allocation prototype is fully successful. All
fonns of income are taxed at the individual rate
(~, which can range from zero to 31 percent).
Equalization of the tax rate across all sources of
income for individuals means that shareholder
allocation reduces all three current law distortions. For tax-exempt and foreign investors,
however, the shareholder allocation prototype
makes no change in the current taxation of nonpreference, U.S. source income.

3.C

CORPORATE LEVEL
PAYMENT OF TAX

In theory, corporate level payment of tax is
not an essential feature of shareholder allocation
integration. 11 Shareholders could have the sole
responsibility for payment of taxes on corporate

Prototypes

level earnings, including retained earnings. Under
such a system, corporations would report income
to shareholders, who would include their allocable
share of corporate income with other income on
their returns and pay tax on their total income.
Partnerships and S corporations follow this approach under current law. However, because tax
is more likely to be collected if paid at the corporate level, the shareholder allocation prototype
retains the current system requiring payment at
the corporate level and then allocates to shareholders the corporation's taxable income and taxes
paid.
Table 3.1
Total U.S. Tax Rate on a Dollar of
NonPreference, U.S. Source Income from a
U.S. Business Under Current Law and the
Shareholder Allocation Prototype

Type of Income

Current Law

Shareholder
Allocation
Integration

I. Individual Investor is Income Recipient
Corporate Equity:
Distributed
tc +(l-tJf:;
f:;
Undistributed
tc + (1 - tJt,
t;
Noncorporate Equity
1;
1,
Interest
1;
Ii
Rents and Royalties
1;
1;
n. Tax Exempt Entity is Income Recipient
Corporate Equity:
Distributed
tc
tc
Undistributed
Noncorporate Equity
Interest
Rents and Royalties
m. Foreign Investor is Income Recipient
Corporate Equity:
tc +(l-tJtwo
to +(l-tJtwo
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties
Department of the Treasury
Office of Tax Policy
tc = U.S. corporate income tax rate.
f:; = U.S. individual income tax rate.
tg = U.S. effective individual tax rate on capital gains.
two, tWN , t Wl , tWR = U.S. withholding rates on payments to
foreigners of dividends, noncorporate equity income,
business interest, and rents and royalties, respectively.
Generally varies by recipient, type of income, and
eligibility for treaty benefits and may be zero.

Prototypes

In addition to increasing compliance, retaining
corporate level payment of tax provides a mechanism for imposing tax on corporate income allocable to tax-exempt and foreign shareholders.
Denying refundability of credits for corporate
level tax to tax-exempt shareholders, in effect,
preserves current law, which taxes corporate
equity income allocable to tax -exempt shareholders at the corporate level. Nonrefundability of
credits also preserves current law for foreign
shareholders. See Section 3.1.

3.D PASSTHROUGH OF
CORPORATE LOSSES TO
SHAREHOLDERS
While it would be possible to pass through to
shareholders aggregate net losses incurred at the
corporate level, the prototype does not do soY
Passthrough of corporate losses would raise a host
of fundamental policy, technical, and administrative issues. For example, one issue is whether, as
for partnerships (but generally not S corporations), shareholders would be permitted to include
entity level debt in their basis to determine the
extent to which losses could be passed through. A
second issue is whether the current at-risk and
passive activity rules would apply at the shareholder level to limit the use of losses incurred by
corporations. Failure to apply these rules could
allow taxpayers to use corporations as tax shelters
and to circumvent current restrictions applicable
to partnerships and S corporations. Passthrough of
corporate losses also would create significant
administrative complexity. Even small shareholders would have to track losses allocated to them,
including losses in excess of basis carried forward
from previous years, and would have to apply the
at-risk rules and the passive activity loss rules.
To avoid the complexity created by applying
additional loss limitations at the shareholder level
and the need for anti-abuse rules, the shareholder
allocation prototype denies passthrough of corporate losses to shareholders. Instead, corporate
losses may be carried forward and used to offset
corporate income in later years. This allows a
reasonable degree of accuracy in measuring

30

corporate income over time while minimizing
complexity and opportunities for abuse.

3.E

TAX TREATMENT OF
PREFERENCES

Integration generally does not require extending the benefits of corporate level tax preferences
to shareholders. Extending preferences to shareholders under integration would increase the value
of corporate preferences relative to current law
and would raise the .revenue cost of integration.
See Chapter 5. Accordingly, the dividend exclusion and CBIT prototypes are structured not to
extend preferences to shareholders. See
Section 2.B and Section 4.D.
In contrast, the shareholder allocation prototype generally extends preferences to shareholders. While we considered modifying the
shareholder allocation prototype in order not to
extend preferences to shareholders, we found such
modifications to be difficult and inconsistent with
the passthrough nature of the prototype. Eliminating preferences by including preference income in
shareholder income as earned would treat COlpOrate preference income more harshly than under
current law. 13 Current law generally taxes corporate preference income at the shareholder level
only when the income is distributed or stock is
sold. While shareholder allocation could be
modified to tax preference income only when
distributed, doing so would effectively convert
shareholder allocation into distribution-related
integration, for which less cumbersome structures
can be used. 14
For these reasons, the shareholder allocation
prototype generally passes through preferences to
shareholders, but that feature is a major reason
we do not favor the adoption of shareholder
allocation. If policy makers were to adopt the
shareholder allocation prototype, serious consideration should be given to restricting the preference
items available to corporations.
The extent to which the shareholder allocation
prototype extends preferences to shareholders

31

depends on the type of preference. An exclusion
preference, e.g., tax-exempt interest on state and
local bonds, allows a cOIporation to earn economic income that is not included in taxable income
and, thus, is not allocated to shareholders. The
prototype provides a shareholder basis increase
for tax-exempt income, similar to the basis increase provided under current partnership rules,
which ensures that such income is not taxed to a
shareholder who sells his stock or receives a
distribution. 15 If such a special basis increase
were not provided, then preference income attributable to an exclusion preference would be taxable
upon distribution or sale of stock.
A credit preference, e.g., the credit for increasing research activities, reduces cotporate
level taxes payable. The shareholder allocation
prototype passes through a credit preference to
shareholders (to the extent it is claimed by the
cotporation) by treating it as cotporate taxes paid,
which are creditable by shareholders. A basis reduction for the amounts of taxable income shielded from tax by credit preferences would make
these amounts taxable either upon the sale of
stock or receipt of distributions in excess of basis.
A deferral preference, e.g., accelerated depreciation, initially reduces corporate taxable income
relative to cOIporate economic income. In later
years, however, as the deferral preference turns
around, the cOIporation's taxable income exceeds
its economic income. Thus, because the shareholder allocation prototype allocates only taxable
income to shareholders, a shareholder who holds
stock throughout the deferral period generally
benefits from a deferral preference to the same
extent as the corporation. As under the partnership rules, however, a shareholder's basis increases only by the amount of taxable income (and taxexempt income) allocated to him. Thus, a shareholder who sells stock or receives a distribution
from the cOIporation may realize taxable gain
because the shareholder's basis does not reflect
the economic income that has been sheltered at
the corporate level by a deferral preference. 16
On the other hand, a distribution that does not

Prototypes

exceed basis before the deferral preference reverses will be treated as a return of basis. In such a
case, the deferral preference will not be taxed to
the shareholder until the stock is sold.
Certain features of shareholder allocation
integration indirectly limit the flowthrough of
preferences. Because the shareholder allocation
prototype does not allow losses to flow through to
shareholders, preferences are not passed through
to the extent they create corporate losses. In
addition, because cOIporate debt is not included in
shareholder basis and inside basis in assets is not
stepped up to reflect the price paid for corporate
shares, there could be disparities between inside
and outside basis that could limit the benefit to
shareholders of corporate level preferences.
A fmal issue involving preferences is the
treatment of the corporate alternative minimum
tax (AMT). In general, the corporate AMT would
be retained under integration to limit use of
preferences at the corporate level. Accordingly,
the dividend exclusion prototype and the CBIT
prototype retain the cotporate AMT. The shareholder allocation prototype does not retain the
cotporate AMT because we found no simple and
administrable mechanism for doing so in the
context of a passthrougb system.
For example, the approach most consistent
with the passthrough nature of the shareholder
allocation prototype would continue to collect
AMT at the cOtporate level, include cOIporate
alternative minimum taxable income (AMTI) in
shareholder AMTI, and credit corporate AMT
against an individual's liability for regular tax and
AMT .17 This approach would treat the cOIporate
AMT as equivalent to a mechanism for withholding shareholder level AMT. 18 However, the
inclusion of cotporate AMTI in shareholder
AMTI would increase unacceptably the complexity of information reporting to shareholders and
the calculation of shareholder tax. We considered
but rejected as unworkable other solutions designed to confme the complexity of the AMT
calculation to the corporate level. 19

32

Prototypes

3.F

ALLOCATING INCOME
AMONG DIFFERENT
CLASSES OF STOCK

Under the shareholder allocation prototype,
once the corporation detennines its taxable income and taxes paid, additional rules are needed
to allocate that amount among different classes of
shares. Both S corporations and partnerships must
make such allocations under current law. However, neither of these models is appropriate for
shareholder allocation integration. The S cOIporation rules, which are designed for corporations
with a single class of stock and a limited number
of shareholders, cannot readily be adapted to
more complex capital structures. 20 The partnership allocation rules are sufficiently flexible, but
generally are too complex, to apply to widely held
corporations. Therefore, the shareholder allocation prototype adopts a modified version of the
partnership approach.
Under current law, a partnership may allocate
its income in any manner that has substantial
economic effect. 1121 Subject to this limitation, a
partnership has great flexibility to allocate income
and loss or particular items of income or deduction to particular partners. In general, an allocation of partnership taxable income or loss can
have substantial economic effect only if such
income or loss is allocated to the partner or
partners that will receive the benefit or bear the
burden of the economic consequences corresponding to the taxable income or loss. The economic
consequences of partnership allocations are reflected in capital accounts maintained by the
partnership m accordance with detailed
regulations. 22
1/

The shareholder allocation prototype approximates the basic approach of the partnership
allocation method while reducing its complexity.
It retains the principal economic advantage of the
partnership system by pennitting allocations of
income to reflect varying economic rights among
different classes of stock.
Under the shareholder allocation prototype, a
corporation can allocate varying amounts of

income to different classes of stock, in accordance
with the tenus of the corporation's governing
instruments. Within each class of stock, a cotpOration allocates every share a pro rata portion of
the income and tax credits allocable to that class.
A corporation could not allocate income separately from credits for taxes paid. Thus, while the
corporation and shareholders may agree on the
amount of income allocated to each class of stock,
all income allocated carries a proportionate share
of credits for cOIporate taxes paid. Allowing
corporations to allocate income and credits disproportionately would allow corporations to
allocate credits to taxable shareholders and income without credits to tax-exempt shareholders.
The shareholder allocation prototype simplifies
the partnership model by (1) imputing to shareholders only a single amount of taxable income,
(2) requiring that tax credits be allocated in
proportion to income, and (3) not allocating
corporate losses to .shareholders. As a consequence, the prototype permits considerable flexibility in corporate capital arrangements but does
not allow corporations to adopt the complex
allocations possible under the partnership rules
(which permit special allocations of items of
income, deduction, and loss).
A substantial disadvantage is that this approach requires corporations to maintain capital
accounts for each class of shares. Although, as
discussed below, these capital accounts are simpler than the capital accounts required to be
maintained for each partner in a partnership under
the regulations under IRC § 704(b), they still add
complexity to the shareholder allocation system.
Capital accounts are needed, however, to help
ensure that allocations of tax consequences follow
allocations of economic income. As the following
simplified example demonstrates, without tax
rules requiring capital accounts, the corporation
could allocate tax liability without regard to the
economic substance of the capital structure.
Example. Two shareholders each contribute $1,000
to a new corporation. One shareholder has a 15
percent marginal rate and enough other tax liability
to absorb excess credits, and the other has a 31
percent marginal rate. The corporation issues Class

33
A stock, which is allocated 100 percent of the
corporation's taxable income, to the low-bracket
shareholder. The corporation issues Class B stock
to the high-bracket shareholder and provides that
no taxable income will be allocated to the Class B
stock. Cash distributions, however, are to be made
pro rata between the Class A stock and the Class B
stock. If these allocations are respected, all the
corporation's taxable income and credits for corporate taxes paid will be allocated to the 15 percent
shareholder. The Class A shareholder's share basis
will increase accordingly, but the Class B shareholder's basis will remain $1,000. Thus, when the
corporation is liquidated, the low-bracket shareholder will realize a loss and the high-bracket
shareholder will realize a gain. In the meantime,
however, the shareholders have arranged for
substantial deferral of tax by having the corporation's income taxed currently at 15 percent
(rather than having half taxed at 15 percent and
half taxed at 31 percent, in accordance with the
economic bargain between the parties).

This strategy would fail if the allocations were
subject to the "substantial economic effect" requirement of IRe § 704(b). The rules under IRC
§ 704(b) would require the allocation of equal
amounts of income to the two shareholders in
order to establish capital accounts that would
permit an equal division of liquidation proceeds.
Thus, some capital account mechanism is
needed in the shareholder allocation prototype.
The remainder of this discussion outlines generally the mechanics of maintaining capital accounts.
Because we do not recommend adoption of shareholder allocation, however, we have not
developed the additional technical analysis needed
for a workable capital account regime. 23
Capital accounts should be easier to maintain
under shareholder allocation than under the
partnership rules because the shareholder allocation prototype passes through only a single item
(net taxable income) and a proportionate amount
of credits for taxes paid. As a consequence,
capital accounts increase by the amount of income
allocated, net of credits for corporate taxes paid,
and decrease by the amount of distributions.
Further, because each share of stock within a
class of stock receives a pro rata share of the
income and taxes allocated, it is not necessary to
keep detailed capital accounts for each

Prototypes

shareholder. Instead, capital accounts can be
maintained for each class of stock. Rules also
would be needed to govern the allocation of losses
to capital accounts. Although losses are not passed
through to shareholders, losses reduce corporate
assets available for distribution and should be
reflected in capital accounts. Special allocations of
losses among classes of stock are permitted, if
appropriately reflected in capital accounts. While
special allocations of losses create additional
complexity, relative to a system in which losses
are required to be allocated in proportion to
income allocations, they seem necessary to preserve corporations' ability to issue preferred
stock. 24 It may be difficult, however, to fashion
practical rules that allow special allocations of
losses to capital accQunts that are liberal enough
to preserve typical corporate capital structures but
are restrictive enough to prevent abuse.
Existing corporations would have to seek
shareholder approval to modify the terms of
outstanding stock to provide for allocations of
corporate income and the maintenance of capital
accounts. This is likely to be a lengthy and
difficult process that would substantially complicate the transition to a shareholder allocation
system of integration. Accordingly, while we do
not recommend shareholder allocation, if it were
adopted, we would recommend a delayed implementation. See Chapter 10. Additional transitional
rules may be needed to provide relief where a
corporation cannot obtain the necessary
shareholder approvals, for example, because of
state law or contractual supermajority
requirements.

3.G CHANGE OF
STOCK OWNERSIDP
DURING THE YEAR
Allocating both a corporation' s retained and
distributed income to shareholders requires a
mechanism to reflect changes in stock ownership
during the period to which such income relates
and thereby apportion income tax consequences
among the corporation's various owners. The
current rules are straightforward: corporations pay
dividends to the shareholder who owns the stock

Prototypes

on the dividend record date and the Code taxes
the person who receives the dividend.
The shareholder allocation prototype requires
that corporate taxable income and corresponding
credits for corporate taxes paid be allocated to
shareholders of record as of the end of each
quarter of the corporation's taxable year. 25 Corporations would not close their books and fue tax
returns and information returns quarterly, but
rather would close their books at year end and
allocate net income ratably to the record holder of
the stock at the end of the four quarters. 26
Closing corporate books at year end and
allocating income pro rata among shareholders of
record unavoidably creates problems in the treatment of shareholders that sell shares before
corporate income and corporate taxes are known
at the end of the year. As long as there is uncertainty concerning a given quarter's income, the
buyer and seller of stock will not be able to price
the stock accurately.
Example. At the beginning of the year, a corporation has assets of $100. Shareholder A owns 100
percent of the single class of stock and has a basis
in the stock of $100. The corporation's taxable
year is the calendar year. On July 1, when the
corporation has earned $25 of taxable income, A
sells all her stock to Shareholder B for $117.25. If
the corporation's books closed on June 30, it would
pay $7.75 of corporate tax and would allocate $25
of income and $7.75 of tax credits to A. If A has
a marginal tax rate of 31 percent, the taxable
income allocated to her will be exactly offset by
the allocated credits. A's basis in her stock would
increase to $117.25, and A would report no gain
on the sale. Because the shareholder allocation
prototype does not determine taxable income until
year end, A's final basis will be determined based
on her pro rata share of the actual earnings and
taxes paid for the year, which will tum on events
subsequent to A's sale of stock and may differ
from estimated earnings as of the date of sale. For
example, if the corporation's taxable income for
the full year is $80, A will be allocated $40 of
income and $12.40 of tax credits and her basis will
increase to $127.60. She will report a capital loss
of $10.35.27

Thus, while a shareholder can tentatively
calculate gain on a sale at the time the sale is
made, that estimate may need to be revised based

34

on more precise or differing information available
only later and may even require the filing of an
amended return. 28 The problem of amended
returns may be particularly acute for shareholders
that hold stock in corporations with taxable years
other than the calendar year. The uncertainty of
income allocations may result in some inefficiency
in pricing sales of stock, although sellers of large
blocks of stock may be able to limit uncertainty
by effectively shifting the tax burden through
contractual mechanisms.
This uncertainty could be reduced by requiring
a quarterly closing of corporate books.29 We
rejected such a requirement, however, as imposing too great a reporting burden at the corporate
level. Requiring quarterly filings of Form 1120
and quarterly information reports to shareholders
would significantly increase the tax reporting
burden on corporations. Although many large
corporations must fue quarterly fmancial statements (lO-Qs) with the Securities and Exchange
Commission (SEC), and most corporations must
make quarterly estimated tax payments, refming
that information to the degree of precision needed
for tax return purposes can be a time-consuming
process. Requiring a true quarterly closing of
books would in effect abandon the taxable year
concept and substitute a "taxable quarter"
regime. 30
Some intermediate solution may be possible.
For example, capital gains and extraordinary
dispositions could be allocated to the quarter in
which they occurred. Large corporations might be
required to provide estimates of each quarter's
income, based on lO-Q filings (if any) and the
kinds of calculations used for estimated taxes.
Shareholders could be permitted to report the
estimated income and tax amounts and make
corrections when fmal reports were issued after
year end. Such a system would, however, allow
a significant degree of latitude to corporations
unless there were rules governing the quarterly
estimating and annual correction process. Such
rules would likely be complex.
This problem would not exist in a pure passthrough integration system with no corporate level

35

tax, no differences in the treatment of capital
gains and losses and ordinary income and full
flow through of corporate losses to shareholders.31 For the policy reasons stated above, however, the shareholder allocation system retains the
corporate level tax and does not require a quarterly closing of books. Accordingly, unless a satisfactory intermediate solution can be devised, the
uncertainty of tax consequences for midyear sales
of stock is unavoidable and is one of the significant obstacles to adoption of the shareholder
allocation prototype.

3.H REPORTING AND AUDITING
CONSIDERATIONS
As the preceding discussion makes clear, any
passthrough integration system would increase the
administrative burden on corporations and their
shareholders. Although the shareholder allocation
prototype includes simplified reporting provisions,
it does require corporations to provide information
reports (not now required) to shareholders showing each shareholder's portion of corporate taxable income and credits for corporate taxes paid
(including other tax credits claimed by the corporation). The information returns also would have
to provide information on appropriate basis
adjustments. Because basis will increase for taxexempt income, the basis adjustment will not
necessarily be the same as the allocated income
less the allocated tax credits. Shareholders, in
turn, must take into account both corporate income and credits for corporate taxes paid in
calculating their own tax liability and will need to
keep detailed records to determine share basis
when stock is sold.
Another administrative problem is the timing
of income reporting. For example, U.S. corporations cannot report taxable income and corporate
level taxes to shareholders until they receive
reports of the taxable income and credits of other
U.S. corporations in which they own stock. We
have been unable to devise a precise solution for
these timing issues. The taxable years of members

Prototypes

of a consolidated group or other closely held and
closely affiliated corporations can be conformed
so that income is calculated at the same time. For
corporate portfolio shareholders, however, timing
difficulties may be severe. Before shareholder
allocation could be. implemented, it would be
necessary to design a reporting system capable of
accommodating corporate cross-ownership. 32
The shareholder allocation system also requires substantial changes in the way corporations
and shareholders are audited. In theory, under a
shareholder allocation system, any increase or
decrease in tax as a result of an adjustment to a
tax return, resulting from an IRS audit or an
amended return, should be reflected in the tax
liability of the shareholders. The current system
for partnerships carries an adjustment back to the
partners' taxable year in which the understatement
arose. Thus, if in 1990, it were determined that a
partnership'S income for 1988 had been understated by $1,000, the increase of $1,000 would be
allocated to those who were partners in 1988.
Extending this regime to corporations under
integration would require the IRS to track and
adjust the returns of shareholders holding stock in
prior years. Furthermore, under such a system an
adjustment in one year may require related
adjustments in other years.
To avoid these problems, the shareholder
allocation integration prototype would treat any
audit or other adjustment to corporate income as
a taxable event in the year of the adjustment.
Under the prototype, it is unnecessary to adjust
returns of prior year shareholders because
adjustments to corporate income would be treated
as an increase or decrease in the corporation's
current year taxes and income. The adjustments
would be passed through to current year shareholders. 33 The IRS would collect deficiencies
directly from the corporation, and the corporation
would pass through the credits for corporate taxes
paid along with the additional income. Shareholders' bases would be adjusted to reflect the
additional income.

Prototypes

3.1

TREATMENT OF TAXEXEMPT AND FOREIGN
SHAREHOLDERS

Tax-Exempt Shareholders
The shareholder allocation prototype maintains
the current taxation of corporate equity income
allocated to tax-exempt shareholders by making
shareholder credits for corporate level taxes
nonrefundable to tax-exempt shareholders. Thus,
tax on corporate income allocable to a tax-exempt
shareholder would be taxed at the corporate level
at the corporate rate. Tax-exempt shareholders
would not be subject to UBIT on corporate
income allocated to them and would not be
allowed to use credits for corporate taxes paid to
offset UBIT liability on other income.

Foreign Shareholders
We believe that foreign shareholders making
investments in the United States should not
receive, by statute, the benefits of integration
received by U.S. shareholders. Thus, the shareholder allocation prototype denies refunds of
corporate level taxes to foreign shareholders and
continues to impose U.S. withholding tax on
dividends. As under current law, corporate tax
would be paid at the corporate level and withholding tax would be imposed at the investor level.
The branch profits tax would continue to apply to
U.S. branches of foreign corporations. Although
in principle, the shareholder level withholding tax
might be imposed on income allocated annually,
the prototype continues to impose withholding tax
only when distributions are made. Annual imposition of both the corporate and the investor level
taxes would increase the tax burden on foreign
investments in U.S. corporations as well as the
disparity in the treatment of debt and equity
owned by foreign investors. Denying integration
benefits to foreign shareholders under the shareholder allocation prototype does not violate U.S.
tax treaty obligations. Refundability of all or a
part of the credit could be considered in treaty
negotiations in exchange for reciprocal benefits.
See Chapter 7.

36

3.J

FOREIGN SOURCE INCOME

We do not believe that an integrated tax
system should, by statute, treat foreign taxes like
taxes paid to the U. S. Government. Extending the
benefits of integration to foreign taxed income, if
appropriate, is more properly achieved through
bilateral tax treaty negotiations. See Chapter 7.
Accordingly, the dividend exclusion and CBIT
prototypes are designed to collect at least one full
level of U. S. tax on foreign source income earned
by U.S. corporations.
In contrast, the shareholder allocation prototype treats foreign taxes paid like U. S. taxes paid.
As a consequence, depending on foreign tax rates,
the United States may collect only a residual U.S.
tax or no tax at all on corporate foreign source
income. We considered modifying the shareholder
allocation prototype to account separately for
foreign taxes and deny foreign tax credits to
shareholders, but such modifi~ations are complex
and fundamentally inconsistent with the passthrough nature of the prototype. 34 Denying a
foreign tax credit would be harsher than current
law, which generally allows a foreign tax credit at
the corporate level and defers the shareholder
level tax on foreign source income until it is
distributed. Modifying the shareholder allocation
prototype to tax foreign source income to shareholders only when distributed would effectively
convert shareholder allocation into distributionrelated integration.
Accordingly, the shareholder allocation prototype allows a foreign tax credit, computed under
current law rules, to offset corporate level tax.
The foreign tax credit, like other corporate tax
credits, is passed through to shareholders. One
issue this approach raises is how, if at all, the
foreign tax credit limitation rules should be
applied at the shareholder level. Although the
foreign tax credit ~itation is computed initially
at the corporate level, additional restrictions
would be necessary to prevent individuals with
marginal tax rates of less than 31 percent from
using foreign tax credits to offset liability for
U.S. tax on other income. 35

37

As under current law, the shareholder
allocation prototype allows an individual U.S.
shareholder holding stock directly in a foreign
corporation to claim a foreign tax credit for
withholding taxes paid on dividends. The prototype does not extend the indirect foreign tax credit
of IRe § 902 to individual shareholders of a
foreign corporation. The indirect credit was
originally intended to prevent multiple taxation of
corporate income earned through a foreign subsidiary. Because the shareholder allocation regime
extends integration to foreign taxes, however,
permitting individuals owning more than 10
percent of the stock of a foreign corporation to
claim an indirect credit may merit consideration.
Extending the indirect credit to U.S. individual
shareholders would remove the disparity that
would otherwise exist between foreign corporate
stock held directly and foreign corporate stock
held through a U.S. corporation. Such a change,
however, would be a significant departure from
current law and would exacetbate the problem of
fashioning an appropriate limitation rule at the
shareholder level.
Another issue for outbound investment in
structuring the shareholder allocation integration
prototype is whether to retain or eliminate the
deferral allowed for profits earned through foreign

Prototypes

subsidiaries. As Chapter 7 explains, the deferral
rule provides that profits of aU. S. investor
earned through a foreign corporation are generally
not subject to U.S. tax until the profits are repatriated. Although theoretical consistency in implementing a shareholder allocation integration
system would require eliminating the deferral
rule, taxing foreign income currently is not
essential to shareholder allocation. As a practical
matter, it would be difficult to end deferral for
U.S. portfolio shareholders, because sufficient
information would not be available from the
foreign corporation to determine the domestic
shareholder's tax -liability on undistributed
income. Even for large shareholders, requiring
annual reporting of income and foreign taxes paid
by foreign subsidiaries would compound the
reporting problems discussed in Section 3.H. A
corporation with foreign subsidiaries could not
accurately report to its shareholders its own
income for the year until its subsidiaries had paid
their own taxes in foreign jurisdictions. Accordingly, the shareholder allocation prototype permits
U. S. shareholders in foreign corporations to
continue to take income into account only when
dividends are received. The same rule applies to
U. S. corporate shareholders, subject to the current
Subpart F and other current inclusion rules.

CHAPTER 4:
COMPREHENSIVE BUSINESS INCOME TAX PROTOTYPE
4.A INTRODUCTION

individual rate of 31 percent rate, regardless of
the lender's actual marginal tax rate and regardless of the lender's status as a tax-exempt or
foreign entity. 3

The Comprehensive Business Income Tax
(CBIT) is the most comprehensive of the integration prototypes developed in this Report. 1 It is
not expected that implementation of CBIT would
begin in the short term, and full implementation
would likely be phased in over a period of about
10 years. 2 The CBIT prototype represents a very
long-term, comprehensive option for equalizing
the tax treatment of debt and equity.
CBIT would equate the treatment of
debt and equity, would tax corporate and
noncorporate businesses alike, and would
significantly reduce the tax distortions
between retained and distributed earnings.
CBIT would accomplish these results by
not allowing deductions for dividends or
interest paid by the corporation, while
excluding from income any dividends or
interest received by shareholders and
debtholders. To ensure consistent treatment of corporate and noncorporate entities, CBIT would apply to all but the
smallest businesses, whether conducted in
corporate form or as partnerships or sole
proprietorships. The result is that
one-but only one-level of tax would be
collected on capital income earned by
businesses. An illustration of taxation
under the current classical corporate tax
and r::BIT is depicted in Figure 4.1.
Under current law, income distributed
on corporate equity generally bears two
levels of tax, while interest paid to suppliers of debt capital bears at most one level
of tax. CBIT not only eliminates the double taxation of corporate equity income,
but also provides equal treatment for debt
income. By denying a deduction for interest, the CBIT prototype subjects interest
income, like dividend income, to a single
level of U. s. tax equal to the top

Without any overall revenue loss, the CBIT
prototype permits a reduction in the rate of tax on
corporations from 34 percent to the top individual
rate of 31 percent. 4 A lower rate of tax on capital
supplied by tax-exempt, foreign or low-income

Figure 4.1
Comparison of CBIT and Current Law l
Current Law

Equity Holders

Corporation

• /I'u.<

CI:;j
••--. ~ ..... No Tax ~::~:empt
Toxabl.

<>

Return

Debt Holders
<.< Taxable

No Tax
L -_ _ _- - - - ' - -

~:::.::::~::; Foreign

-----"----

Tax-Exempt

Investment

Equity Holders

(TIiXG
~ ••.•. . Tax

Taxable
Foreign
.. Tax . .. Tax-Exempt

Debt Holders
>. Taxable

~:::.==~:~::
Non-Corporate Form

Corporation
.>
Return

<Tu::>< .

Equity Holders

Taxable

•.• ~ ~ :::::::;~~~::; Foreign
Tax-Exempt

.................. ,
. ·t1ix

Debt Holders
Taxable

~~~.c.c..~>.~••• ~ ~ ~~~~~::=~Foreign
...-1
•••

------

Investment

Tax-Exempt

Equity Holders

~::::==~~~:=~

Taxable

Non-Corporate Form

IThe figures do not take into account tax preferences or taxes
imposed by other countries.
39

Prototypes

40

investors could be incorporated into a CBIT
regime, but we have chosen not to include these
complicating provisions in the prototype described
in this chapter. S Taxing income from business
capital at a 31 percent rate enhances economic
efficiency and advances the policy goals set forth
in Chapter 1. 6 CBIT taxes corporate and noncorporate businesses (other than very small businesses) under identical rules, thus eliminating the
current tax bias against the corporate form. CBIT
also makes significant progress toward the removal of incentives to retain earnings, although a
compensatory tax on distributions of preference
income, if included in CBIT, would provide some
incentive to retain such income.
Like the other prototypes, the CBIT prototype
is structured to conform as closely as possible to
the policy decisions summarized in the introduction to this part with respect to the treatment of
preferences and tax-exempt and foreign investors.
Since CBIT would be a greater change from
current law than either distribution-related integration or shareholder allocation integration-both of
which would apply only to corporate equity-a
very gradual phase-in of CBIT over a long period
will be necessary in order to reduce the economic
dislocations and the gains and losses that might
result during the transition. See Chapter 10. 7

4.B

OVERVIEW OF CBIT
PROTOTYPE

General Mechanics. Under CBIT, distributions
of business income as dividends or interest are not
generally taxed when received by investors (see
the discussion of tax preferences below). The
income of all business entities, including corporations and unincorporated businesses, is measured
and taxed at the entity level at a 31 percent rate. 8
The CBIT tax base is generally the corporate
income tax base under current law, except that no
deduction is allowed for interest expense, and
dividends and interest received from CBIT entities
are excluded. Losses incurred at the entity level
do not pass through to the equity holders. Unused
losses can be carried over at the entity level,
however, generally in the same manner as under
the current law rules applicable to cOIporations. 9

Small Business Exception. Because it is
difficult to separate returns to capital from returns
to labor in the case of very small businesses,
taxing all capital income from those businesses at
the 31 percent CBIT rate might overtax some
labor income that otherwise would be taxable to
an individual in a lower bracket. The CBIT
prototype includes an exception for very small
businesses. See Section 4.C.
Tax Preferences. Tax preferences available to
corporations generally would be available to CBIT
entities. To implement this Report's general
recommendation that preferences not be extended
to shareholders, a flat rate nonrefundable tax of
31 percent (a compensatory tax) could be imposed
at the entity level on dividends and interest
deemed paid from preference income. Alternatively, investors could be required to include in
income any interest or dividends considered to be
paid out of preference income. The choice
between these two methods is discussed in
Section 4.D. In either case, businesses would
determine which distributions are made out of
preference income by maintaining an Excludable
Distributions Account (EDA), which is similar to
the EDA described in Chapter 2 under the dividend exclusion prototype. The EDA would reflect
taxes paid and the prototype would stack interest
and dividend payments first against fully-taxed
income. to See Section 4.D.
CBIT Entities as Investors. CBIT entities are
governed by the rules applicable to nonCBIT
investors. Income from investments (other than
dividends and interest from CBIT entities) is taxed
to the CBIT entity as under current law. Dividends and interest from CBIT entities are not
taxed in the hands of the recipient CBIT entity
and would result in an appropriate addition to the
recipient entity's EDA (thereby enabling the
recipient CBIT entity to distribute such receipts
without paying additional tax). Additional rules
would be needed for taxable dividends and interest paid by CBIT entities if a compensatory tax
were not adopted. See Section 4.D.
Foreign Source Income. CBIT entities would
be entitled to a foreign tax credit computed as

41

under current law, with modifications to reflect
the nondeductibility of interest under CBIT.
Foreign source income shielded from U. S. tax by
foreign tax credits would be treated in a manner
similar to preference income when distributed and
either would be subject to a compensatory tax or
would be taxable at the investor level at that time.
As with distributions from preference income,
stacking distributions fIrst against fully-taxed
income will limit somewhat application of these
rules.
Low-Bracket Investors. While the CBIT
prototype does not include explicit relief for lowbracket equity holders and debtholders, it is
possible to reduce the effective rate of tax on
CBIT investments from 31 percent to the investor
rate with an investor credit for entity level taxes
paid. See Section 4.F.
Tax-Exempt and Foreign Investors. Interest
and dividends paid to tax-exempt and foreign
investors by a CBIT entity are net of the 31
percent entity level tax; however, in general
neither tax-exempt nor foreign investors are
subject to additional U. S . tax on interest or
dividends received from CBIT entities. If a
compensatory tax is adopted, all dividends and
interest would be excludable. As Section 4.D
discusses, however, the alternative to a compensatory tax is to tax preference and foreign source
income at the investor level.
We recognize that, in imposing one level of
source-based taxation on interest paid to foreign
investors, CBIT would represent a departure from
current policy on inbound debt investment. Any
such departure would have to be the result of
extensive international discussions with tax authorities and market participants. 11
Capital Gains and Share Repurchases. Chapter 8 discusses the treatment of capital gains on
CBIT equity and debt and the treatment of share
repurchases.
NonCBIT Interest and Other Capital Income.
CBIT does not require any change in the current
taxation of interest paid on debt issued by a

Prototypes

borrower other than an entity subject to CBIT.
Thus, for example, home mortgage interest would
continue to be deductible by an individual borrower and includable in the income of the recipient.
State and local bond interest would remain excludable from gross income to the same extent as
under current law. Interest on Treasury debt
would, as under current law, be includable in
income by the recipient. 12 See "Interest Not
Subject to CBIT" in Section 4.G.
Impact on tax distortions. Table 4.1 illustrates
the impact of the CBIT prototype on the three
distortions integration seeks address: the current
law biases in favor of corporate debt over equity
fmance, corporate retentions over distributions,
and the noncorporate over the corporate form. In
general, CBIT is very successful in achieving the
goals of integration because it removes most
differentials in the tax rates on alternative income
sources for domestic and foreign investors and
tax-exempt entities. The near-uniform tax rate on
all nonpreference, U.S. source business income is
the maximum individual income tax rate (~m, 31
percent under current law). For individual investors, the only exceptions to this uniform rate are
for undistributed corporate equity income (if
capital gains on corporate stock continue to be
taxed) and for rent and royalties, which would
continue to be taxed at regular individual rates.
For tax-exempt entities and foreign investors, the
only exception to the uniform rate on nonpreference, U.S. source business income is the rate on
rents and royalties, for which current law rates
would be retained.

4.C

ENTITIES NOT SUBJECT TO
CBIT

In theory, CBIT would apply to all businesses,
without regard to size or legal form of organization. Thus, all sole ptoprietorships, partnerships,
S corporations and other business entities would
be subject to an entity level tax. After the
phase-in of CBIT, current law distortions between
the corporate and noncorporate business sectors
would thus be eliminated, and taxpayers' choice
of business entity would depend entirely upon
nontax considerations. To preserve these

Prototypes

42

Table 4.1
Total U.S. Tax Rate on a Dollar of
NonPreference, U.S. Source Income from a
U.S. Business Under Current Law and the
CBIT Prototype

Type of Income

Current Law

CBIT

I. Individual Investor is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

tc + (l-tJt;
tc +(1-t,A
t;
t;
t;

t;m
tt+(l_t;m)tg

t;m
t;m
t;

ll. Tax Exempt Entity is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

tc

o
o

m. Foreign Investor is Income Recipient
Corporate Equity:
Distributed
tc + (1 - tJtwo
Undistributed
tc
Noncorporate Equity
lwN
Interest
twJ
Rents and Royalties
tWR
Department of the Treasury
Office of Tax Policy

1;m
t; m
t;m
t;m
twp.

tc = U.S. corporate income tax rate.
l; = U.S. individual income tax rate.
t;m = Maximum U.S. individual income tax rate.
tg = U.S. effective individual tax rate on capital gains; is
zero in one version of the prototype.
tWD , tWN • t Wl • tWR = U.S. withholding rates on payments
to foreigners of dividends, noncorporate equity
income, business interest, and rents and royalties,
respectively. Generally varies by recipient, type of
income, and eligibility for treaty benefits and may be
zero.

neutrality benefits, we believe that any small
business exception to CBIT should be limited to
very small entities.
The CBIT prototype includes an exception for
small businesses with gross receipts of less than
$100,000. Such businesses would continue to
deduct their interest expense, and the interest they
pay would be taxable to the recipients. Any wages
or profits distributed by an exempt small business
would be taxable to the recipients at the

recipients' marginal tax rates. CBrr interest and
dividends received by a small business WOUld. be
excludable. We concluded that such an exceptIon
was desirable because of complexities that might
otherwise arise in the transition from current law
to CBIT and difficulties in separating capital
income from labor income for very small businesses (proprietorships, in particular). Although
CBIT generally taxes the income shares of creditors and equityholders at a unifonn 31 percent
rate, it does not alter the current progressive
individual rate structure (with graduated rates
from 15 to 31 percent) for taxing wages or other
labor income and nonCBIT capital income. While
all CBIT taxpayers would be allowed to deduct
reasonable compensa~ion paid for services to the
same extent as under current law, these rules may
be inadequate for small businesses. In many small
businesses, income received by an owner-manager, in fact, may be a mixture of returns on both
physical and human capital. Ignoring the distinction and subjecting all the owner-manager's
income to the unifonn CBIT rate, might overtax
the labor component of the owner-manager's
income. In addition, not allowing losses to flow
through currently might create significant hardship
where the owner-manager draws a salary. With a
small business exception, however defmed, all
returns on capital in such nonCBIT small businesses would be taxed at the investors' separate
rates instead of at the uniform CBIT rate. 13
We concluded that an exclusion based on
annual gross receipts would be the simplest to
structure and estimate at the current conceptual
phase ofthe prototype's development. For pUIpOSes of determining an entity's eligibility for the
exception, dividends and interest received from
CBIT entities would be included (although they
would not be taxable to the receiving entity). Such
a deftnition of the exclusion has several advantages. A gross receipts criterion is objective and
easier to apply from a compliance and enforcement standpoint than the alternatives discussed
below. It can be determined readily from documents currently generated for tax compliance
purposes.1 4 So long as the lower bound of gross
receipts determining CBIT status is low, we

43

believe that aggregation rules for nonCBIT entities should be unnecessary. 1S
Other criteria are possible. Ideally, the criteria
should be related to the potential "blurring" of
owners' capital and labor incomes. For example,
businesses with substantial equity held by individuals who also supply substantial labor to the
entetprise might qualify. Other definitions currently used in the Code or elsewhere include
criteria such as whether the business is closely
held (as measured by the number of shareholders), the value of the business (as measured by the
value of stock, net worth, or the value or adjusted
basis of assets), the annual amount (or average
annual amount) of net income, and the number of
employees. The correlation between blurring of
labor and capital income of owner-managers and
some of these characteristics may depend on the
nature of the business, industry characteristics,
and other factors. We believe the more practical
course, however, is simply to exempt certain
"small businesses" based on size. 16

4.D TAX PREFERENCES
Introduction
We have made a general recommendation in
this Report that integration should not become an
occasion for extending cOlporate level tax preferences to shareholders. Future policymakers seem
likely, however, to retain many of the preferences
currently available to cotporations under the
Code. Absent special rules, CBIT's general
exclusion of dividends and interest from income
would automatically extend those preferences to
shareholders. 17
There are two general mechanisms which
could be used to ensure that one level of tax is
imposed on preference income when it is distributed. First, CBIT entities could be required to
report to shareholders and debtholders the
amount, if any, of each dividend or interest
payment that is made out of preference income.
The investor would then include that amount in
income and pay tax at the investor's tax rate. This
is the mechanism we recommend in the dividend

Prototypes

exclusion prototype. 18 The alternative approach
is to impose a 31 percent compensatory tax at the
entity level on all distributions from preference
income. Such a compensatory tax would not be
refundable to tax-exempt or foreign investors.
Although both systems have advantages, the
dividend exclusion prototype (and the imputation
credit prototype described in Chapter 11) reject a
compensatory tax in favor of shareholder level
taxation of distributed preference income and
foreign source income shielded from U. S. tax by
foreign tax credits. As Section ll.B discusses, in
those prototypes, which are limited to corporate
equity, this Report would tax preference income
and foreign source income at the shareholder level
in order to preserve current tax and dividend
policy for corporations with substantial amounts
of such income.
Under CBIT, however, a compensatory tax
has considerable conceptual and practical appeal.
Adopting a compensatory tax would permit investors to exclude all dividends and interest received
from any CBIT entity. Thus, CBIT would consistently collect tax on capital income, whether
interest or dividends, at the entity level at a 31
percent rate.
A compensatory tax would be simpler at the
investor level. Because all distributions with
respect to CBIT investments would be excludable
by investors, no information reporting to shareholders or debtholders would be required. On the
other hand, if preference income distributed as
interest or dividends were subject to investor level
tax, CBIT entities would have to provide information reports to the IRS and to investors, indicating
the extent to which a distribution is excludable. A
compensatory tax under CBIT also would permit
the complete repeal of the withholding tax on
dividends and interest paid to foreign investors.
See Section 4.E.
The principal disadvantage of a compensatory
tax under CBIT is that our economic analysis
suggests that it would create significant inefficiencies in cotporate payout decisions. Our data
indicate that even if distributions were stacked

Prototypes

44

frrst against fully-taxed income, a compensatory
tax would impose a significant entity level tax
burden on distributions. Our models of corporate
behavior predict that, to avoid this additional tax,
CBIT entities would increase their reliance on
retained earnings as a source of fmance and would
rely less on both new equity and debt. Under the
assumptions of our models, this effect is strong
enough to distort corporate payout decisions as
much as under current law. See Section 13.D.
Accordingly, the remainder of this chapter
describes the differences in treatment necessary
under the CBIT prototype if no compensatory tax
is imposed and distributed preference income and
foreign source income are taxed at the investor
level. 19

Excludable Distributions Account
The prototype identifies distributions out of
preference income and foreign source income
shielded from tax by foreign tax credits by requiring CBIT entities to maintain an Excludable
Distributions Account (EDA). (The EDA is
similar to the EDA described in Chapter 2, except
that interest payments as well as dividend payments are charged against the account.) For each
$1.00 of U.S. tax paid, approximately $2.23
would be credited to the EDA. The annual addition to the EDA is referred to as fully-taxed
income and is calculated using the following
fonnula:

+

u.s.

tax paid for taxable year
.31
-

us
..

To illustrate, assume that a corporation subject
to CBIT earns $100 in taxable income and $100
of preference income, and pays $31 in regular
CBIT taxes but neither pays nor receives dividends or interest. Its EDA is thus $69 [$31/.31$31]. If it then pays $75 in interest and dividends,
it will pay a compensatory tax of $1.86 [.31 x
($75 -$69)] or, alternatively, the $6 of distributions that is attributable to preference income will
be taxable to investors. 21
If a compensatory tax is adopted, all distributions on equity and debt of CBIT entities will be
excludable. A CBIT entity receiving a distribution
would add the amount received to its own EDA.
If, alternatively, distributions of preference income were taxable to investors, the prototype
could either (1) tax CBIT entities currently on
such distributions 22 or (2) provide a deduction,
similar to the current dividends received deduction, for such receipts to defer tax until the income is redistributed to a nonCBIT entity. 23

Alternative Minimum Tax
Consequences of CBIT

Annual additions to EDA =

[

compensatory tax or, alternatively, wo~l.d be
taxable to the investor. 20 As in the diVidend
exclusion prototype, refunds of entity level tax
would not reduce the EDA below zero. Refunds
in excess of the taxes reflected by the EDA
balance would be applied to reduce future entity
level tax payments. Similarly, net operating losses
in excess of the EDA would be carried forward.

'd &
bl
]
tax pal lor taxa e year

equity distributions and interest received from CBrr entities

The EDA is reduced by the amount of all dividend and interest payments, in the order in which
payments are made. The EDA is also reduced by
approximately $2.23 per $1.00 of tax refunded.
Positive EDA balances may be carried forward
without limitation.
The prototype stacks payments flrst against
fully-taxed income. Distributions of interest or
dividends reduce the EDA. When the EDA is
reduced to zero, distributions would be subject to

The CBIT system retains an entity level
alternative minimum tax (AMT) similar to the
corporate AMT under current law. As under
current law, the entity level minimum tax would
ensure that some entity level tax is imposed
currently on a profitable business. In a CBIT
AMT, however, neither interest expense nor
dividends would be deductible and dividends and
interest from CBIT entities would be excluded.
Because the CBIT tax base provides no deduction
for interest paid, it is likely that relatively few
nonfmancial businesses would have regular tax
liabilities low enough to trigger a CBIT AMT
imposed at the current 20 percent rate. As in the

45

dividend exclusion prototype, AMT would be
treated as taxes paid in the same manner as the
regular CBIT tax; however, the divisor in the
EDA formula would still be the regular CBIT tax
rate, 31 percent. Thus, a CBIT entity could not
distribute all of its alternative minimum taxable
income (AMTI) without triggering a compensatory tax or an investor level tax.
Adopting CBIT might permit significant
simplifying modifications to the current individual
AMT. If CBIT applied to all but small business
entities, the individual AMT base would apply
principally to two items: (1) excess itemized
deductions and (2) State and local tax-exempt
bond income treated as a preference under current
law.24 It would be inappropriate, however, to
include excludable CBIT interest or equity income
in an investor's AMTI because any such tax
imposed would be a second level of tax on income
that had already been subjected to tax at the
highest individual rate. 25

4.E

INTERNATIONAL
CONSIDERATIONS

Taxation of Income from
Outbound Investment
This Report recommends that the tax burden
imposed by any integration prototype on income
from U. S. investment in foreign businesses
(outbound investment) be roughly equivalent to
the tax burden imposed on such income under
current law. The shift from two-tier taxation of
cOIporate foreign source income to a single-tier
tax should not result in the collection of a significantly greater or lesser amount of tax revenue
from such income than under current law. See
Chapter 7.
Under current law, foreign source income
earned through a domestic corporation is potentially subject to U.S. tax at both the corporate and
the shareholder levels. At the corporate level,
foreign source income is subject to a 34 percent
tax, which may be reduced substantially or eliminated by foreign tax credits. If the U.S. corporate

Prototypes

tax liability on foreign source income is less than
the foreign tax imposed on the income, excess
foreign tax credits may arise. Upon distribution,
the income generally is subject to full taxation at
the shareholder's marginal tax rate, without a
foreign tax credit. This approach is consistent
with U.S. income tax treaty commitments. No
U. S . treaties require that investors in aU. S.
corporation receive tax relief from foreign taxes
paid by the corporation.

Foreign Source Income of CBIT Entities and
Other Business Entities
Under the CBIT prototype, results comparable
to those under current law are achieved by allowing the foreign tax credit (with a modified limitation, as described below) to offset the regular
CBIT tax in full, but adding no amount to the
EDA to reflect foreign source income sheltered
from U.S. tax by foreign tax credits. 26
The EDA mechanism does not distinguish
between foreign source income shielded from the
regular CBIT tax by the foreign tax credit and
U. S. source preference income. Both benefit from
the stacking rule that treats distributions as arising
first from income subject to the regular CBIT tax.
Accordingly, as with preference income, so long
as foreign source income shielded from CBIT by
the foreign tax credit is not distributed, it will
bear no further tax burden. The CBIT compensatory tax or an investor level tax will be triggered
only when such income is distributed-the same
circumstance that would result in imposition of a
shareholder level tax under current law.

If a compensatory tax is not adopted, this
stacking rule ensures that the total Federal tax
burden on outbound investment by corporations
should not vary sigIiificantly from that imposed
under current law, apart from the effect of the
expanded tax base for foreign branch income
reSUlting from the nondeductibility of interest.
Imposition of a compensatory tax could increase
the tax revenue collected from outbound investment. In either case, the tax burden on outbound
investment by corporations may actually be less

Prototypes

46

for foreign source income subject to foreign tax at
a rate less than the CBIT rate, which will be
subject to only a single level of residual U.S. tax.
CBIT will, however, require modification of
the current rules for computing the foreign tax
credit. Under current law, the foreign tax credit
limitation is equal to the product of (I) the taxpayer's pre-credit U.S. tax liability on worldwide
taxable income and (2) the ratio (j)f the taxpayer's
foreign source taxable income to its worldwide
taxable income. This usually reduces to the
product of the U. S. tax rate and the foreign
source income. The foreign source income of a
U.S. taxpayer is currently computed under U.S.
tax. principles for this purposeY In the case of a
foreign subsidiary, the amount of foreign taxes
that are deemed paid by a 10 percent U. S. corporate shareholder in respect of a particular dividend
distribution is equal to the total foreign taxes paid
by the subsidiary, multiplied by the ratio of the
dividend to the total earnings of the subsidiary.
(This amount is subject to the limitation just
described.)
If foreign source income were computed under
CBIT principles, i.e., with no deduction for
interest, problems would arise. In the case of
foreign branch operations of CBIT entities, the
amount of foreign source income in the limitation
fonnula could increase dramatically. Such an
increase would seriously mismatch the computation of taxable income and tax liability by a
foreign jurisdiction that allowed a deduction for
interest. Assuming that foreign tax rates were
high enough to provide an adequate supply of
credits, no U.S. tax. would be collected currently
on foreign source income used to pay interest.
Instead, U.S. tax. would be collected only when
such income was deemed to have been distributed
by the entity and a compensatory tax (or an
investor level tax) was imposed. In the case of a
foreign subsidiary, the amount of earnings in the
denominator of the indirect credit fraction could
increase dramatically, seriously diluting the
amount of foreign taxes attributed to a particular
distribution of earnings.

Accordingly, the CBIT prototype assumes
that, in computing the foreign tax credit limitation, foreign source income of a branch will be
reduced by interest expense claimed with respect
to the foreign operations. 28 Similarly, in
computing the indirect foreign tax credit, earnings
of the foreign subsidiary will be reduced by
interest expense claimed by the subsidiary. 29
Under this approach, CBIT entities will continue
to enjoy approximately the same level of direct
and indirect foreign tax credits as under current
law. Some reduction will occur, however, by
reason of lowering the regular CBIT tax rate to
31 percent from the current 34 percent.
Several additional effects of CBIT on the
taxation of foreign source income should be
noted. As explained above, CBIT would subject
all business organizations to an entity level tax.
This has at least two possible implications for the
foreign tax credit. First, it suggests that an indirect credit for foreign taxes deemed paid by a
foreign subsidiary should be available to noncorporate domestic shareholders, such as partnerships, that are CBIT entities. Under CBIT, the
purpose of the indirect credit would defer the
additional level of CBIT tax until the time of
distribution (when a compensatory tax or an
investor level tax would be imposed) to avoid the
burden of an immediate tax on foreign source
profits. If the indirect credit were not extended to
partnerships and other noncorporate CBIT entities, there would continue to be a strong bias in
favor of the corporate vehicle for multinational
enterprises.
Second, the equal treatment of all business
entities under CBIT means that foreign tax credits
will not fully relieve CBIT tax in circumstances
where U.S. tax is fully relieved under current
law. If a domestic partnership or S corporation
receives a dividend, interest, or royalty payment
from a foreign corporation (or other foreign
payor) under current law, and the payment has
been subject to a foreign withholding tax, the
recipient is eligible for a foreign tax credit, and
no further U. S. tax is imposed to the extent that

47

Prototypes

the partners or shareholders are individuals.
Under CBIT, however, the credit would only
relieve the regular CBIT tax. A compensatory tax
or an investor level tax would be imposed when
the foreign profits are redistributed to the partner
or shareholder.

regular CBIT liability, such income will bear a
pre-distribution tax rate that is higher than the
CBIT rate applicable to domestic source income.
This disparity, which also exists under current
law, is entirely attributable to higher foreign tax
rates.

Finally, CBIT requires some consideration of
the treatment of foreign business entities. Under
current law, deferral of U.S. tax on foreign
profits is available when the profits are earned
through a foreign corporation. When such profits
are earned through a foreign partnership, the U. S.
tax is not deferred, and the results are essentially
the same as for a foreign branch office of a U.S.
taxpayer. Under the CBIT prototype, foreign
entities would generally be treated as nonCBIT
entities. Thus, interest paid by a foreign entity
would continue to be taxable to aU. S. lender,
and would continue to be deductible by the foreign entity. 30 In addition, deferral would continue to be permitted for profits earned through a
foreign corporation.

Foreign portfolio equity investment (less than
10 percent of total equity) by a CBIT entity.
Foreign source portfolio dividends received by a
CBIT entity would be subject to source country
income taxation at the level of the foreign corporation and to a second level source country withholding tax upon distribution. Regular CBIT
would apply to the foreign source dividend when
received by a CBIT entity, subject to offset by a
foreign tax credit for the source country withholding tax. In most cases, some regular CBIT would
be collected, because regular CBIT liability would
generally exceed the foreign withholding tax by
virtue of treaty rate reductions and by virtue of
the expansion of the CBIT income base to include
income paid out as interest. While such income is
subject to an additional level of taxation (the
foreign corporate level tax) relative to income
earned through investment in aU. S. subsidiary,
the disparity should be approximately the same as
under current law. If distributed by the CBIT
entity, such income, to the extent shielded from
regular CBIT by the foreign tax credit, would be
subject to the CBIT compensatory tax or an
investor level tax. If the CBIT entity is a corporation, this result generally will be comparable to
the result under current law. To the extent residual regular CBIT is paid, the result will be better
than under current law for shareholders now
taxable on dividend income. A CBIT entity that is
a partnership with individual shareholders or an S
corporation may be treated less favorably than
under current law in certain circumstances.

Foreign branches of CBIT entities. In the case
of a foreign branch of aU. S. CBIT entity, the
expanded CBIT income base of the branch would
be included in the U. S. CBIT entity's income
currently. Foreign source income earned by a
CBIT entity through a foreign branch would be
subject to residual regular CBIT tax prior to
distribution. As discussed above, there will
always be a residual regular CBIT tax on the
portion of the foreign source income base that is
excluded from the computation of the foreign tax
credit. Where the foreign jurisdiction's tax is
computed with an interest deduction, such income
will bear, in effect, the same tax that it would
have borne if earned from domestic sources. With
respect to the remaining portion of the foreign
source income base, a residual regular CBIT tax
will be imposed if the foreign income tax liability
is less than the regular CBIT liability, with the
effect that such income also will bear the same
pre-distribution aggregate tax (foreign tax plus
CBIT tax) that it would have borne if it were
earned from domestic sources. 31 If the foreign
income tax liability on the remaining portion of
the foreign source income base is higher than the

Foreign direct equity investment (10 percent
or more of total equity). Foreign source income
earned by a CBIT e~tity through a direct equity
investment would be subject to full source country
corporate level tax and to a second level source
country withholding tax upon distribution of a
dividend from the foreign subsidiary. The CBIT
entity (whether a corporation or partnership)

Prototypes

would receive a credit both for the source country
withholding tax and for the source country cotporate level tax under IRC § 902. Thus, regular
CBIT would be imposed only to the extent that
the regular CBIT liability exceeded the total
amount of foreign taxes paid or deemed paid.
Given the opportunity to defer the CBIT
compensatory tax or investor level tax by
retention of foreign subsidiary profits at the CBIT
entity level, the disparity between direct equity
investment in a foreign subsidiary and investment
in a domestic subsidiary under CBIT should not
vary significantly from current law. If distributed
by the eBIT entity, such income would be subject
to the eBIT compensatory tax or an investor level
tax to the extent it was shielded from regular
eBIT by foreign tax credits. However, as with
portfolio investment, the result will generally be
similar to the result under current law in cases
where such dividends would be taxed fully. To
the extent subject to residual regular CBIT, such
income will be taxed less heavily than under
current law. A eBIT entity that is a partnership
or an S cmporation may be treated less favorably
than under current law (depending on whether the
IRe § 902 credit is extended to such
shareholders) .
Foreign debt investment. Foreign source
income earned by a CBIT entity through a debt
investment in a foreign entity or subsidiary would
escape source country income taxation to the
extent that interest is deductible for foreign income tax pUlposes. While such income potentially
would be subject to a foreign withholding tax
upon distribution as interest, the CBIT entity
would receive a foreign tax credit for the
withholding tax (subject to the foreign tax credit
limitation). Thus, regular CBIT would be imposed
only to the extent that regular CBIT liability
exceeds the foreign withholding tax. Interest
income received from a domestic subsidiary also
would be subject to CBIT, in this case imposed
on the subsidiary. Thus, outbound debt investment
should not be subject to greater entity level tax
than domestic debt investment until such income
is distributed. The CBIT compensatory tax or an
investor level tax then would apply to the extent
the income had been shielded from U. S. tax by

48

foreign tax credits. The impact of the CBrr
compensatory tax or an investor level tax, if and
to the extent imposed, will be similar to the
consequences described for the imposition of such
tax on foreign portfolio equity investment.

Foreign Source Income
Earned Directly by Individuals
Under CBIT, foreign cotporations and other
foreign entities would be treated as nonCBIT
entities. Accordingly, as under current law,
interest and dividend income received directly by
aU. S. resident individual from a foreign corporation would be subject to tax at the individual's
marginal tax rate. CBIT does not require the
modification of the foreign tax credit allowed to
individuals under current law.

Taxation of Income from
Inbound Investment
As noted in Section 4. A, we view CBIT as a
very long-range option for equalizing the treatment of debt and equity. We anticipate that
adoption of CBIT would be preceded by a lengthy
period of consideration and, when implemented,
CBIT would be phased-in over a period of about
10 years. See Chapter 10.
Both the dividend exclusion prototype and the
shareholder allocation prototype retain the current
U. S. withholding tax on dividends paid to foreign
shareholders and the branch profits tax on U.S.
branches of foreign corporations. Retaining the
second level of tax on equity income in those
prototypes simply replicates current law and
permits reduction of the second level of tax
through tax treaty negotiations.
We make a different recommendation in
CBIT, however. Retaining current law in the context of CBIT would require collecting two levels
of tax on dividends and zero or one level of tax
on interest. (Chapter 7 discusses the current law
taxation of foreign investors.) Such treatment
would violate the equality between debt and equity
that is one of the principal goals of CBIT. To
maintain parity between debt and equity, the

49

CBIT prototype removes the remaining withholding taxes on both interest and dividends paid by
CBIT entities. 32 The result is to subject both debt
and equity income to CBIT taxation once at the
entity level.
Elimination of the remaining withholding taxes
on both dividends and nonportfolio interest under
CBIT would clearly affect U. S. income tax treaty
negotiations. While existing U.S. treaties provide
for reciprocal reductions of source country tax
rates on interest and dividends, CBIT might
reduce U. S. treaty partners' incentive to grant a
reciprocal exemption in future negotiations. 33 In
order to obtain a reciprocal exemption, it might
be necessary for the United States to make concessions either with respect to entity level tax
collected on dividends and interest or CBIT
compensatory taxes (if any) imposed on dividends
and interest. For example, a tax credit for CBIT
taxes paid could be made available only on a
bilateral basis. Any such treaty concessions should
be made in a manner to protect CBIT's basic goal
of equating the taxation of debt and equity.
If a compensatory tax were not adopted,
distributed preference income and shielded foreign
source income will be taxable to investors.

We recognize that adoption of CBIT would
represent a departure from current policy on
inbound debt investment and that any such departure would require extensive international discussions with tax authorities and market participants.

Conduct of a U. S. Trade or Business
As under current law, income earned by a
foreign investor through the conduct of aU. S.
trade or business would be taxed in the same
manner as income earned by U.S. residents. CBIT
rules would apply to foreign business activities in
the United States. Thus, interest expense attributable to aU. S. trade or business would be nondeductible, and the current law provisions governing
the allocation of interest expense to effectively
connectpif income would be unnecessary. 34

Prototypes

Small Business Exception
The small business exception would apply to
inbound investment. See Section 4.C. Distributions from small, nonCBIT corporations to foreigners would remain subject to current statutory
withholding at 30 percent, unless that rate is
reduced by treaty provision. 35 In the case of a
U. S. branch of a foreign corporation, the size
criteria would be applied on the basis of the gross
effectively connected receipts of the branch.

4.F

IMPACT OF CBIT ON
INVESTMENT BEHAVIOR
OF LOW-BRACKET,
TAX-EXEMPf, AND
FOREIGN INVESTORS

Overview
Because substantial nontax factors influence
investment behavior, we cannot predict with
certainty CBIT's impact on the manner in which
investors allocate their portfolios. Indeed, if tax
considerations were paramount, there would be a
strong bias under current law against any
investment by low-bracket taxpayers and domestic
tax-exempts in domestic corporate equities (as
opposed to debt). Current experience indicates,
however, that both of these groups invest in
corporate equity. While special statutory withholding provisions, the statutory exemption for
capital gains realize4 by foreign investors on
property investments other than in real property,
and treaty mitigation provisions make it hard to
generalize in the case of foreign investors, the tax
provisions of current law, if given paramount
effect, would direct their investment toward
domestic debt rather than corporate equity in most
instances. Other nontax factors are important,
however, and foreign investment in domestic
equity occurs despite higher tax rates than for
domestic debt.
The United States' stable economic and political climate attracts investment. The size of our
consumer market attracts foreign sellers and

Prototypes

investors. Opportunities for diversification not
available through alternative investments can
override tax disadvantages. These nontax factors
will temper portfolio shifts by these classes of
taxpayers, Considering these countervailing
forces, we believe that the best approach is to
adopt a gradual phase-in of CBIT, rather than
specific measures for low-bracket, tax-exempt and
foreign investors although we discuss such
measures below. To preserve CBIT's neutrality
between debt and equity, the discussion contemplates identical treatment of debt and equity. The
reductions of tax due to these mechanisms, of
course, will have revenue consequences.

Interest Rate Impact of CBIT
The interest rate on CBIT debt will be less
than the interest rate on nonCBIT debt, potentially
by an amount up to the 31 percent entity level
tax, because interest received on CBIT debt
represents an after-tax return. 36 For example, if
market interest rates on nonCBIT debt were 10
percent, a debt instrument issued by a CBIT entity
might bear interest at a rate as low as 6.9 percent.
If this were the case, the after-tax return on the
two instruments would be the same for a taxable
investor with a 31 percent marginal rate. While
predicting the actual rate relationship between
CBIT and nonCBIT debt is impossible, experience
with the ratio of interest on tax-exempt state and
local bonds to that on taxable corporate bonds
suggests that the CBIT interest rate may not
reflect a 31 percent tax rate, because there may be
insufficient demand for CBIT debt by investors
with a marginal rate of 31 percent. Thus, for
example, if a nonCBIT bond bore interest at a 10
percent pre-tax rate, a CBIT bond might bear
interest at 8 percent if it were necessary to attract
lower-bracket investors to CBIT debt. In such a
case, the 8 percent (after-tax) CBIT return would
be more attractive to an investor in the 31 percent
bracket than the 10 percent (pre-tax) nonCBIT
return.
Because interest rates on CBIT debt should be
lower than the rates on nonCBIT debt, low-bracket, tax-exempt, or foreign investors (collectively,
tax-favored investors) can be expected to increase

50

their holdings of nonCBIT debt and decrease their
holdings of CBIT debt. (Overall, these portfolio
shifts may be offset by increased demand for
CBIT debt and equity by taxable investors.)
Depending on their tax rates, tax-favored investors, for example, might prefer a 10 percent
nonCBIT bond to an 8 percent CBIT bond. For
any investor with a marginal rate of less than 20
percent, a 10 percent nonCBIT return is worth
more than an 8 percent CBIT (after-tax) return.
While a rate differential of less than 15 percent
between CBIT and nonCBIT bond rates should
not affect the portfolio choices of low-bracket
individual taxpayers, any rate differential could
affect investment choices by tax-exempt and
foreign investors since, as under current law, all
nonCBIT interest paid to tax-exempt investors
(and portfolio interest paid to foreign investors) is
tax-free at the investor level. Domestic tax-exempt
entities might be expected to decrease holdings of
CBIT debt and increase holdings of governmental
or other nonCBIT debt and CBIT equity.37
The treatment of preference income under
CBIT further complicates the analysis of the
expected rate differential between CBIT and
nonCBIT investments. If a compensatory tax were
imposed, all CBIT investments would pay an
after-tax return, and ·one would generally expect
the risk adjusted return on CBIT investments to
be the same. On the other hand, if payments of
dividends and interest out of preference and
foreign source income are taxable to investors,
issuers with substantial preference and foreign
source income may pay a higher return than
issuers with substantial fully-taxed income.
If CBIT were adopted, special attention would
have to be given to its impact on international
capital flows.

Low-Bracket Investors
As discussed in Chapter 1, we have structured
the CBIT prototype to impose a unifonn 31
percent tax on earnings on capital invested in
CBIT entities. However, the impact of CBIT on
taxable equity holders and bondholders with
marginal rates of less than 31 percent could be

Prototypes

51

lessened by providing those investors with a tax
credit. This credit could be designed to give those
investors a tax benefit equal to all or a portion of
the difference between their marginal rate and the
31 percent CBIT rate. While the credit would not
be refundable, it could offset tax on other income.
The effect would be similar to full refundability
for any investor with enough other tax liability to
absorb the credit. 38 If a compensatory tax were
not imposed, the credit would be available only
for excludable payments.
The credit is essentially the same as the
shareholder credit for low-bracket investors
described in Section 2. D in the context of the
dividend exclusion prototype. Because CBIT
extends to both dividends and interest, the credit
would be available to both equity holders and
bondholders.
Example. Assume that a CBIT entity earns $100 of
income and pays $31 in tax. It then distributes $69
as interest to a bondholder with a marginal tax rate
of 15 percent. Applying the formula set forth in
Section 2.0 (adjusted to reflect the 31 percent
CBIT rate), a bondholder credit of $16 (i.e.,
$69/.69X(.31-.15» would produce a tax benefit
equal to the difference between the bondholder rate
and the CBIT rate.

Tax-Exempt Investors
Under the other prototypes described in this
Report, denying refundability of corporate level
taxes preserves the current law treatment of
corporate equity owned by tax-exempt and foreign
investors. Under CBIT, however, some offset for
corporate level taxes would tend to move CBIT
closer to current law by mitigating the additional
tax burden the prototype places on interest earned
by tax-exempt investors. As with low-bracket
shareholders, the credit could be set at a rate that
would refund either all or a portion of the tax
imposed at the 31 percent CBIT rate. If a compensatory tax is not imposed, the credit would be
available only for excludable payments.
Because tax-exempt investors have little or no
tax liability, they would be unable to benefit from
the nonrefundable investor credit described in the

preceding section. One possibility would make the
investor credit described above refundable. An
alternative approach would combine an investor
level credit with a tax on investment income of
tax-exempt entities. Under this approach, taxexempt and foreign investors would be liable for
tax on all investment income (interest, dividends,
capital gains, rents, royalties, and other investment income). The rate of this tax could be set to
produce overall revenues (taking into account the
investor credit) equivalent to those currently borne
by equity supplied by the tax-exempt sector. A
tax-exempt entity could then use the investor level
credit to offset the tax due on other investment
income. See Section 6.D.39
Imposing a tax .on investment income and
allowing a credit would treat CBIT and nonCBIT
debt instruments alike (although it probably would
not fully compensate for the interest rate differential between CBIT and nonCBIT debt). It generally would encourage tax-exempt entities to hold a
mixture of CBIT and nonCBIT debt and equity,
because the nonrefundable investor credit associated with CBIT debt and equity could be used to
offset the tax due on other kinds of investment
income. This approach would minimize differences between CBIT and nonCBIT investments, just
as it could minimize differences between debt and
equity under distribution-related integration. 40

Foreign Investors
The absence of special relief for foreign debt
investors in the CBIT prototype reflects our
judgment that elimination of the withholding tax
on CBIT dividends and interest and elimination of
the branch tax may balance the CBIT change as to
debt, recognizing that, under CBIT, foreign
investors may prefer nonCBIT debt to CBIT debt
and CBIT equity to equity under current law.
Nevertheless, either of the mechanisms described for tax-exempt investors-a refundable
credit or the investment tax and credit mechanism
described in the preceding section~ould be used
to provide relief for foreign investors. A gradual
phase-in of CBIT also would allow assessment of

52

Prototypes

the need for
expenence.

such

mechanisms

based

on

Impact of Relief Measures for
Low-Bracket, Tax-Exempt and Foreign
Investors on the CBIT Prototype
Our recommended CBIT prototype contains
none of the relief mechanisms discussed in the
preceding sections. Adoption of any of these
mechanisms would result in a revenue loss which
would have to be recovered elsewhere in the
prototype or in other offsetting revenues not now
required by the prototype. For example, a compensatory tax could be imposed. (The estimates
for the CBIT prototype in Section 13. H do not
include a compensatory tax.) In addition, the decisions to eliminate the branch tax and withholding
taxes for foreign investors could be re-examined
(although such a modification would be contrary
to the goal of imposing a single level of U. S .
tax).

•

IRC § 385 (granting Treasury the authority to
define the distinction between debt and equity) and
IRC § 279 (denying deductions for equity-like debt)
would be repealed,

•

IRC § 163(e)(5) and (i) (deferring interest deductions on high-yield discount obligations) and IRC §
163(j) (deferring excessive interest deductions on
certain related-party debt-the anti-earnings stripping provision) would be repealed,

•

IRC § 267(a)(2) (relating to matching of interest
income and deductions between related parties)
would no longer apply to interest paid by CBIT
entities,

•

IRC § 469 (the passive loss rules) and IRC § 465
(the at risk rules) would have no application to
interest paid by a CBIT entity,

•

IRC § 263A(f) (relating to capitalization of interest
with respect to self-constructed assets and inventory) could be repealed, and IRC § 266 (the election
to capitalize interest generally) could be repealed
with respect to CBIT entities,41

•

IRC § 1277 (restricting interest deductions allocable to accrued market discount) and IRC § 1282
(restricting interest deductions allocable to accrued
discount) might no longer apply to interest paid by
CBIT entities,

•

IRC § 263(g) (requiring capitalization of interest
and other costs of carrying a straddle) might no
longer apply to interest paid by a CBIT entity,

•

IRC § 265(a)(2) (disallowing deductions for interest
incurred to purchase obligations bearing tax-exempt
interest) might no longer apply to interest paid by
a CBIT entity,

•

IRC § 265(b) (relating to disallowance of interest
deductions of financial institutions allocable to taxexempt obligations) and IRC § 291(e)(1)(B)(ii) (an
earlier version of IRC § 265(b) applicable for taxexempt obligations acquired by financial institutions
between 1982 and 1986) could be repealed,42 and

•

IRC § 264(a)(2), (3), and (4) (denying interest
deductions on certain debts relating to life insurance policies) might not apply to interest paid by
CBIT entities.

4.G STRUCTURAL ISSUES
Current Law Interest Deduction
Limitations Under CBIT
Under current law, interest paid or incurred
by businesses generally is deductible. In special
circumstances, however, the Code limits business
interest deductions. These limitations serve several purposes, such as treating debt instruments with
equity characteristics as equity, preventing mismatches in the timing of income and expense, and
preventing tax arbitrage by borrowing to purchase
tax -favored investments.
CBIT's elimination of the deduction for
business interest by all but the smallest businesses
could allow a major simplification in the Code by
eliminating (or substantially reducing) the need
for several provisions designed to prevent excessive and mismatched interest deductions. Thin
capitalization will no longer be a tax concern. We
believe the following Code sections could be
repealed or substantially reduced in scope:

CBIT will expand the scope of provisions,
such as IRC § 265(a)(2) (which currently disallows deductions for interest on indebtedness

53

incurred or continued to purchase or carry obligations bearing tax-exempt interest) and IRC §
265 (a) (1) (which currently disallows expense
allocable to tax-exempt income other than interest), to apply to taxpayers who receive CBIT
interest and dividends. While the expanded interest disallowance rules would not apply to CBIT
entities, it would apply to individuals and small
business entities to disallow interest on debt
incurred or continued to purchase or carry equity
or debt of CBIT entities. 43 Absent such expansion, much of the CBIT tax base would erode in
tax arbitrage transactions illustrated by the following hypothetical example:
Example. Assume that, for each year of its operation, CBIT entity X earns $1 million, pays
$310,000 in regular CBIT tax and pays the remaining $690,000 as a dividend to individual A, its sole
shareholder. The $690,000 is not taxable to A.
Assume that A borrowed $6,900,000 from taxexempt entity C at 10 percent interest per year to
purchase the X stock. If A is allowed a deduction
of $690,000 for interest paid, he can shelter up to
$690,000 in income from other sources while using
his excludable CBIT dividends to pay the interest
to C. C will pay no tax on the $690,000 in interest
it receives each year. If the $690,000 deduction
allowed to A shelters income otherwise taxable at
31 percent, $213,900 of the tax paid by X will in
effect be refunded to A. While the interest paid and
dividend received in this example are equal, they
need not be. If C is willing to loan A $10 million
against his X stock on the same terms, A's interest
deduction, if used against other income, would
fully offset the CBIT tax X paid with respect to the
distribution to A.44

Under current law, this is simply one of many
opportunities for rate arbitrage through the issuance of debt by taxable issuers to tax-exempt and
foreign lenders. CBIT, however, generallyeliminates businesses' ability to pay interest to taxexempt and foreign lenders without the payment
of one level of tax. Thus, to prevent the erosion
of the CBIT base, it is also necessary to prevent
investor level rate arbitrage through borrowing.
Application of modified IRC § 265 would be
equally appropriate if a compensatory tax is not

Prototypes

adopted and interest and dividends paid by CBIT
entities out of preference income are taxable to
investors. In either case, the potential for arbitrage is the same. See "Anti-abuse Rules" in
Section 2.B.
Finally, some of the interest deduction limitations CBIT might eliminate may serve policies
that would continue to be important but would
require new mechanisms under CBIT. One example is current law's requirement that debt obligations be issued in registered form. Currently IRC
§ 163(t) denies a deduction for interest on unregistered obligations for which registration is required. This sanction would have no deterrent
effect for CBIT entities because CBIT eliminates
interest deductions. Because interest received
from CBIT entities will not be taxed to the investor, the need for registration of debt instruments
of CBIT entities for tax enforcement purposes will
be greatly reduced. However, registration may be
desirable for nontax law enforcement purposes,
and replacement sanctions may be needed. 45

Identifying Disguised Interest
CBIT entities and their investors will be
indifferent to the characterization of payments to
investors as either interest or dividends, because
neither will be deductible by the CBIT entity and
neither will be taxable to the investor. However,
tax tensions will remain and may be exacerbated
by CBIT with respect to rent and royalty payments and allocations between principal and
interest on the purchase of capital assets.
If the market rate of interest on CBIT debt
does not fully reflect the nondeductibility of
interest payments, it will generally be advantageous to a CBIT entity to restructure such payments, where possible, into deductible rental and
royalty payments. Such a restructuring will
generally be disadvantageous to taxable recipients
since it will convert interest that is not taxed into
taxable rents or royalties. No such tension will
exist, however, if the recipient is a tax-exempt
entity or a CBIT entity that is in a net operating

Prototypes

loss position. Similarly, CBIT entities can be
expected to maximize principal and minimize
interest payments on capital purchases, since asset
basis will give rise to deductible cost recovery
while interest payments are nondeductible. Again,
taxable sellers may have opposing interests depending on how gains on asset sales are taxed. 46
As with rents and royalties, these tensions will not
exist where the seller is tax-exempt or is a CBIT
entity with a net operating loss.
CBIT therefore will put increased pressure on
standards, such as those the Internal Revenue
Service has developed, distinguishing fmance
leases (which are treated for tax purposes as loans
and hence generate nondeductible interest for a
CBIT entity) from true leases (which are
respected as such for tax purposes and hence give
rise to deductible rentals for CBIT entities). 47
We believe that it would be prudent in a CBIT
regime to include standards for distinguishing
interest from rents and royalties in the Code,
modeling them on existing standards, such as
those the Service has developed for leases, or on
IRe § 467, which imputes interest to prevent
uneconomic accruals of rent. 48
Purchase price allocations are inherently
factual and governed by the standards of the
market. While CBIT may change the tax stakes in
such allocations, the problem presented is no
different from that confronting the Internal
Revenue Service in making fair market value
determinations under current law. We do not
contemplate that statutory change will be needed
in this connection to implement CBIT.
The current original issue discount (OlD) and
imputed interest rules may be needed in order to
distinguish interest from principal. For example,
in the case of sales of property in exchange for
debt, these rules are needed to determine the
buyer's basis and the seller's amount realized. 49
Similarly, in the case of debt issued for cash,
these rules are needed to distinguish payments of
interest (which reduce the EDA and, when the
EDA is exhausted, are subject to compensatory
tax or investor level tax) from payments of
principal. 50

54

Interest Not Subject to CBIT
CBIT does not dictate any change in the
current taxation of interest paid on debt issued by
a nonCBIT borrower. Thus, for example, home
mortgage interest and personal investment interest
incurred to carry nonCBIT assets would continue
to be deductible by an individual borrower to the
same extent as under current law and includable
in the income of the recipient. Nonmortgage,
personal interest would continue to be nondeductible by the borrower and includable by the lender.
State and local bond interest would generally
remain excludable from gross income to the same
extent as under current law. Interest on Treasury
debt would, as under current law, be includable in
income by the recipient. 51
One administrative issue raised by nonCBIT
debt is tracking income and deductions related to
such debt. For example, maintaining the current
law treatment for home mortgage interest, interest
on Federal debt, and debt issued by foreign and
tax -exempt entities under CBIT will require
special reporting rules to identify such interest as
includable in income and to permit it to retain its
special character when it is collected and distributed by a REMIC, REIT, or other passthrough
entity.
Under CBIT, interest earned on bonds issued
by State and local governments would retain its
current exemption from tax,s2 but interest income on debt issued by CBIT entities generally
would be exempt. Under CBIT, the rate of interest on exempt state and local obligations may
approximate the interest rate on corporate debt of
similar risk and maturity. Thus, State and local
governments might view CBIT as eliminating the
borrowing advantage they currently enjoy relative
to corporate issuers. State and local debt would,
however, retain its advantage over Treasury and
other nonCBIT debt such as home mortgages.

Pension Funds
As Section 2. G discusses, the immediate
deduction for employer contributions to pension
plans, combined with the deferral of income to

55

the employee until benefits are paid, effectively
exempts the investment earnings on the contribution from tax. As a consequence, under current
law pension fund investment earnings from investments in corporate stock bear only one level of
tax-the corporate tax paid by the corporation.
Investment earnings on pension fund investments
in corporate debt, however, bear no tax at all
under current law, because corporate income used
to pay interest is not taxed at the corporate level. S3 Under CBIT, however, investment earnings
from both CBIT debt and equity will be taxed at
the payor level, with the consequence that pension
plans will earn an after-tax return on such investments. The introduction of CBIT thus eliminates
the deferral of tax on inside buildup.
The position of pension plan trusts under
current law could be replicated in CBIT only by
refunding the CBIT entity level tax on interest
paid to pension trusts. This step would eliminate
the need to revise pension tax rules, but would
undermine CBIT's fundamental goals of treating
debt and equity alike and collecting a uniform tax
on business capital income regardless of the
identity of the investor.
To equate the treatment of CBIT debt and
equity investments by pension funds, we recommend requiring pension trusts to maintain separate
accounts for CBIT income and other amounts,
e.g., contributions and nonCBIT income,S4 to
treat all distributions made each year as made
proportionately from the income of each account,
and to notify pension payees of the amount from
each account included in their pension payments.
Payees would be entitled to exclude from income
pension distributions from the CBIT income
account, thereby reducing the tax burden on corporate equity investments relative to current law.
Because pension trusts will enjoy no inside
build-up advantage over other investors with
respect to the CBIT assets they hold, CBIT might
induce such trusts to alter their portfolio mix
toward nonCBIT assets. The degree to which this
occurs depends on the relationship of CBIT to
nonCBIT yields and the portfolio and diversification advantages of particular investments.

Prototypes

If a compensatory tax were not adopted,

pension funds would add only excludable CBIT
income to the CBIT income account. In general,
taxing distributed preference income at the investor level, rather than imposing a compensatory
tax, would lessen the extent to which adoption of
CBIT removes the tax-free inside build-up on
CBIT investments.

Subchapter C Recognition and Reorganization Rules
As in the dividend exclusion prototype, the
CBIT prototype retains the basic rules of Subchapter C governing the treatment of taxable and
tax-free corporate asset and stock acquisitions.
CBIT entity gain on asset sales would be taxable
to the CBIT entity and payment of tax on the
gains would give rise to additions to the EDA,
thereby permitting distribution of the after tax
proceeds of such asset sales to investors without
further tax. As in the dividend exclusion prototype, the Subchapter C reorganization rules would
be retained, and no special limitations analogous
to IRC §§ 382 and 383 would apply to the EDA.
See Section 2.F. As in the dividend exclusion
prototype, EDAs would be combined in acquisitive reorganizations and allocated in divisive
transactions. Liquidations would generally be
treated as in the dividend exclusion prototype. A
liquidating entity's EDA would generally be
allocated among equity holders in proportion to
the amount of other assets distributed to them,
and any gain would be excludable to the extent of
the allocable EDA.55
In CBIT, however, partnerships are treated as
CBIT entities. Imposing Subchapter C structural
rules on partnerships would change current law
significantly by eliminating the partnership rules
found in IRC §§ 731-732 which permit tax-free
distribution of partnership property to partners. 56
While the CBIT prototype contemplates that the
existing Subchapter C recognition rules for distributions ultimately should be applied to all CBIT
entities, policy makers concerned about the
implications of such a rule on changes in the
organization form of smaller CBIT enterprises
could create carryover basis exceptions to the

Prototypes

56

Subchapter C recognition rules for smaller CBIT
entities. 57

Capital Gains, Dividend Reinvestment
Plans, and Share Repurchases
If a compensatory tax were adopted, a full
exemption of investor level gains and losses on
equity and debt could be viewed as consistent with
CBIT's exemption of investor level tax on dividends and interest. However, the fundamental
problem of capital gains taxation in CBIT is
similar to that encountered in other integration
prototypes and either resolution (to tax or to
exempt capital gains) will be controversial. See
Chapter 8. If capital gains are taxed under CBIT,
corporations might implement a dividend reinvestment plan (see Chapter 9) to reduce the incidence
of double taxation on retained earnings. The
appropriate treatment of share repurchases under
CBIT also depends on treatment of capital gains.
See Section 8.B.

4.H CONDUITS
Treatment of Conduits under CBIT
Current law exempts certain organizations
from entity level tax. These entities function as
tax conduits; they either are granted complete
passthrough status or are taxed only on their
undistributed income. Partnerships generally are
granted passthrough status if they meet certain
classification tests that distinguish them from
corporations. 58 Certain publicly traded partnerships are always treated as corporations. 59 Regulated investment companies (RICs) are taxable
corporations but are allowed a deduction for
dividends paid out of both ordinary income and
capital gains. 6O A typical RIC is a mutual fund
that makes diversified investments for its shareholders. Real estate investment trusts (REITs) are
taxed similarly to RICs but are restricted to
investing predominately in real estate. 61 Real
estate mortgage investment conduits (REMICs)
are entities that hold flXed pools of mortgages and
have both regular interests, providing for flXed,
unconditional payments and taxed as debt, and a

single class of residual interests, taxed essentially
like equity interests in a partnership.62 Holders
of REMIC residual interests are taxed on their pro
rata share of the REMIC's net income.
A cooperative, generally, is an organization
that transacts business with and for its patrons
(owners). Some cooperatives enjoy a limited
exemption from tax: Subchapter T cooperatives
are treated as corporations under current law but
are allowed a special deduction for patronage
dividends and per unit returns allocated to patrons
based on business activity. While this results in
effective conduit treatment of patronage distributions and allocations, other earnings of a cooperative are SUbjected to corporate taxation. 63 Typical
cooperatives include farmers' cooperatives that
purchase farmers' crops, sell them, and remit the
proceeds to the farmers or purchase feed and seed
for resale to farmers. Other cooperatives include
grocery, hardware, drug, book, and clothing
stores that operate on a cooperative basis.
Conduits that are not taxable entities under
current law could continue as such under CBIT or
could be treated as CBIT entities. To the extent
that a conduit holds only CBIT equity or debt, its
status as a conduit is irrelevant. A RIC, for
example, that holds only CBIT bonds would pay
no entity level tax even if it were treated as a
CBIT entity, because all of its interest income and
capital gains would be exempt from tax. Any
dividends paid to shareholders also would be
exempt from tax. Conduit status would be equally
irrelevant, whether CBIT included a compensatory
tax or instead imposed tax at the investor level on
distributions out of preference income. See
Section 4.D.
Thus, the treatment of nonCBIT income
earned by conduits is the principal issue in deciding whether conduits should retain their passthrough status. One of the principal purposes for
conduit status under current law is to provide
relief from the double tax applicable to corporations. Because CBIT subjects corporate income
only to a single level of tax, CBIT might replace
the need for cond~its. In addition, retaining
conduit status for some entities would provide a

57

means for avoiding the CBIT regime. Conduit
status permits income to be taxed at shareholders'
rates (which, for tax-exempt shareholders, may be
zero), rather than at the CBIT rate. Thus, there
would be an incentive to have nonCBIT assets
held through a conduit rather than through a CBIT
entity.

Partnerships
The CBIT prototype treats partnerships as
CBIT entities in order to avoid perpetuating the
bias against doing business in the corporate form.
Exempting partnerships from CBIT would create
incentives for investors to choose the partnership
form whenever the tax benefits of passthrough
treatment outweighed the business costs of operating in partnership rather than corporate form.
Example. A group of investors (including some
tax-exempt organizations) is considering undertaking a business venture. The investors decide to
conduct business through a partnership rather than
a CBIT entity so business income will be taxed at
the investors' rates rather than at the CBIT rate.

By removing taxes from the determinants of
organizational form, CBIT enhances neutrality.
In general, under CBIT, partnerships that do

not qualify for the small business exception
described in Section 4.C would be taxed like
other CBIT entities. Thus, a partnership would be
subject to entity level tax each year on its earnings (computed under the normal corporate tax
rules but without a deduction for interest), but
would not allocate earnings to equity holders.
Like other CBIT entities, a partnership would
maintain an EDA and would track actual distributions (rather than allocations of income) to partners and interest payments on debt. Distributions
and payments in excess of the EDA would be
subject to compensatory tax (or investor level
tax).M

Subjecting partnerships to CBIT may treat
certain types of partnership income less favorably
than under current law. For example, partnership
income would be subject to tax at the CBIT rate,
rather than at the partners' individual rates.

Prototypes

Partnership losses, preference income, and foreign
tax credits would no longer pass through to
partners. Distributed preference income and
sheltered foreign source income would be subject
to compensatory tax (or investor level tax). If
these results are undesirable, policymakers may
wish to expand the class of partnerships that are
exempt from CBIT beyond the small business
exception discussed in Section 4.C. However, the
advantages of doing so should be weighed against
the costs of retaining tax incentives favoring
noncorporate forms of organization.

RICs, REITs, and REMICs
The analysis for these special purpose passthrough entities may be somewhat different,
however. There is an argument that they should
retain conduit status because they serve an important function as pooled investment vehicles for
small investors. To the extent that individuals and
tax-exempt organizations could purchase and hold
nonCBIT investments, e.g., home mortgages,
Treasury securities, and tax-exempt bonds, directly, they should be permitted to do so indirectly
through a RIC or REIT.
Example. A CBIT corporation would like to issue
new shares in order to purchase a new building.
Corporate earnings used to pay dividends on those
shares would, however, bear tax at the CBIT rate.
The corporation decides instead to lease its new
building from a REIT, which issues shares to fund
the purchase. As a consequence, the corporation
can deduct the payments of rent, and dividends
paid by the REIT are taxed at shareholder rates.

While the preceding example might be viewed
as avoidance of CBIT, the incentives to engage in
this form of transaction under current law are as
strong as they would ·be under CBIT. In addition,
given a decision to simplify CBIT by making it a
31 percent tax on all capital income, it might be
considered worthwhile to maintain investment
opportunities for low-bracket investors that will
bear tax at the investor's tax rate rather than the
CBIT rate. 6S Maintaining conduit status for
RICs, REITs, and REMICs will require the
expansion of IRC § 265(a)(3) to deny such conduits the ability to deduct expenditures related to
the purchase or carrying of CBIT assets. With this

Prototypes

modification, however, it should be possible to
retain current rules for such entities. This approach will make enforcement of the leasing
standards discussed under "Identifying Disguised
Interest" in Section 4. G particularly important in
maintaining the CBIT base.
Given the decision to treat partnerships generally as CBIT entities, it may be appropriate to
make changes in the REIT qualification rules to
allow entities with fewer than 100 shareholders
and state law partnerships to qualify as REITs for
tax purposes. This would avoid conferring an
advantage on large, corporate REITs in real estate
investing. Similar relaxation of the RIC qualification rules might be considered.

Cooperatives
We believe the limited conduit status granted
to Subchapter T cooperatives would continue to
be the appropriate model for cooperatives under
CBIT. Cooperatives would thus be CBIT entities
but could deduct patronage dividends. 66 As under
current law, patronage dividends would generally
be includable in the patron's income.

4.1

FINANCIAL
INTERMEDIARIES
UNDERCBIT

Financial intermediaries include depository
institutions, insurance companies, investment
banks, and other fmancial services entities.
Although the specific services provided by these
institutions vary, fmancial intermediaries generally
solicit funds from investors, depositors, and other
lenders and use these funds to make loans or to
acquire the debt and equity issues of other companies. Thus, fmancial intermediaries earn most of
their income in the form of dividends and interest
and tend to have substantial noninterest expense
that is incurred to produce net interest and dividend income and gains on securities.
The following analysis suggests the basic
outlines of the taxation of fmancial intermediaries
under CBIT, although further consideration should

58

be given to these issues during the period CBIT is
under discussion. 67

Financial Institutions Generally
CBIT would exempt from tax much of the
income received by fmanciaI institutions because
it is received in the form of dividends and interest
from CBIT entities. In addition, if fmancial
institutions were treated as CBIT entities, their
interest expense would no longer be deductible.
This raises the question of how other operating
expenses of fmanciaI. institutions should be treated. We have generally recommended that IRC §
265(a)(1) and (2), which operate to disallow
deductions and interest allocable to tax-exempt
income, be extended to cover investment in equity
and debt of CBIT entities. Given the large portion
of fmancial institution income that can be expected to come from CBIT investments as well as
from tax-exempt State and local government
bonds, this general rule would operate to disallow
a significant portion of their operating expenses if
deductions for such expenses were not allowed.
This effect is likely to be less significant for
direct lenders such as banks and fmance companies because they would no doubt begin to charge
fees (rather than interest) to cover the costs of
making a loan (as contrasted with the institution's
cost of funds). Indeed, provisions requiring the
borrower to pay the lender's transaction costs
such as attorney's fees, filing fees, survey and
appraisal expenses, inspection costs and similar
items are already a common feature of negotiated
loan transactions. The advantage of converting
interest income into fee income would be that a
CBIT borrower could deduct fees but not interest.
Although the fee income will be includable in the
income of the CBIT lender, the lender will be
permitted to deduct operating expenses against
such income without disallowance under expanded
IRC § 265. Thus, recharacterizing interest income
as fees may permit better matching of a fmancial
institution's income and expenses. This strategy,
however, is likely to be less successful with
respect to publicly traded instruments of CBIT
entities, where the intermediary, in many

59

instances, will be unable to negotiate borrower fee
payments to cover its operating expenses. Given
the prevalence of commissions and fees in the
compensation paid to investment banks and securities trading entities, however, it may be that
market adjustments in these amounts would solve
the problem for these entities.
For revised IRC § 265(a) rules to function as
described in this section, mechanical provisions
which match operating expenses with related fee,
commission, and reimbursement income will be
necessary. In particular, a proportional allocation
rule such as that found in current IRC § 265(b)
would produce inappropriate results if CBIT
income were included in the fraction. Instead,
fmancial institutions should be allowed to allocate
operating expenses fully to offset fee income. To
the extent that fee income is insufficient to cover
operating expenses, the residual expenses would
be allocated between CBIT and nonCBIT income
under the pro rata rule of IRC § 265 (b) and the
portion allocable to CBIT income could be
disallowed under IRC § 265(a).
Alternatively, fmancial institutions could be
exempted from the disallowance rule of expanded
IRC § 265(a) with respect to their operating
expenses. 68 This approach would increase the
incentive for such institutions to generate sufficient nonCBIT income (through investments in
Treasuries, home mortgages, consumer debt, and
leasing activities) to absorb fully the portion of
their operating expenses in excess of their fee
income. Our analysis indicates that most fmancial
institutions currently hold enough nonCBIT debt
to achieve this result; accordingly, the impact of
such an approach on actual investment patterns is
likely to be minimal. However, there is no relationship between the nonCBIT income and the
expenses related to CBIT investments; hence, the
allowance of a full offset may reduce other income, rather than matching nonCBIT income. 69

Savings and Loan Associations
Savings and loan associations (S&Ls) must
invest heavily in home mortgages to maintain
their qualification for special tax rules. Assuming

Prototypes

these requirements were maintained under CBIT,
S&Ls would receive primarily taxable income but
receive no deduction for interest paid to depositors. There should be a significant spread, however, between the interest rates paid on home mortgages (because recipients will pay tax on such
interest) and the interest rates paid to depositors
(because the depositor will not be subject to tax
on interest received from the S&L as a CBIT
entity). This spread may be sufficient to allow
S&Ls to satisfy their CBIT liabilities, and, if so,
no special rules will be needed. Again, a gradual
transition to CBIT would allow policymakers to
study the observed impact of CBIT before finally
resolving structural decisions. Because the need
for a special rule for S&Ls is not clear, the CBIT
prototype does not include such a rule.
If experience proves that the rate differential

between interest on home mortgages and interest
on CBIT deposits is insufficient to allow S&Ls to
operate successfully, consideration could be given
to allowing S&Ls to' issue certificates of deposit
that would bear taxable interest to the recipient
and deductible interest to the S&L. Even such a
limited provision would undermine somewhat the
tax parity between debt and equity achieved by
CBIT, however, and should be adopted only if it
proves necessary. 70

Insurance Companies
Under the CBIT prototype, insurance companies would be CBIT entities. 71 Like other CBIT
entities, they would not be allowed a deduction
for interest paid, but distributions to shareholders
and creditors would not be taxed to the recipients.n Under CBIT, IRC § 809 (which Congress
intended to equalize the treatment of stock and
mutual companies' equity returns) would be
repealed, since equity returns from both stock and
mutual companies would be exempt to the recipient under CBIT. In both types of companies,
payment of tax on earnings from sutplus would
give rise to an EDA permitting distributions free
of further tax to investors. Distributions in excess
of the EDA would trigger the compensatory tax
or an investor level tax, but would preserve the
equal treatment of investors.

Prototypes

CBIT will, however, require an adjustment in
the deduction permitted insurance companies for
annual additions to reserves. Under current law,
tax reserves are calculated on a discounted basis.
Accordingly, the deduction for reserve additions
each year consists of two components: (1) the
discounted present value of amounts required to
fund future casualty and benefit payments plus (2)
the expected return for the year on reserve funds.
This system permits companies to claim deductions currently rather than deducting the entire
loss or claim when paid. The difference between
the present value of such losses or claims and the
full (or nominal) value of such payments is deducted each year as expected return until the loss
or claim is actually paid. The rate used to compute expected return under current law is based
on the applicable Federal rate (APR), which
reflects a taxable rate of return.
Under CBIT, reserves would be calculated
with a blended market interest rate, which would
be a prorated average of a taxable nonCBIT rate
and a non-taxable CBIT rate, according to the
mixture of assets held by each insurance company. To the extent that reserve assets are invested
in CBIT securities, no deduction to shield expected return on CBIT entity dividends and interest
received by an insurance company would be
appropriate because such amounts would not be

60

included in its income and would increase the
insurance company's EDA. Accordingly, insurance companies would be required to maintain
CBIT and nonCBIT income accounts similar to
those of pension funds under CBIT. As with
pension funds, insurance companies would be
required to treat their expected return on reserves
as arising pro rata from the CBIT and nonCBIT
income accounts. An annual deduction for expected return would be pennitted only to the extent
attributable to nonCBIT income. As a result of
this modification, insurance companies should
neither obtain new benefits nor lose current law
benefits with respect to their nonCBIT investments. While insurance companies would pay no
tax on dividends and interest received from CBIT
entities, they would enjoy no advantage over other
investors in this respect.
The prototype's preservation of reserve deductions to prevent entity level taxation of the inside
build-up (the income earned on reserves held in
nonCBIT assets) may be regarded as inconsistent
with the neutrality principles underlying CBIT,
since the prototype may lead insurance companies
to prefer nonCBIT investments which benefit from
this advantage. We believe, however, that a different rule is not necessary for CBIT to function
effectively and would require reversal of longstanding policies underlying insurance taxation.

PART llI: PRINCIPAL ISSUES
INTRODUCTION
Each of the systems of corporate integration
considered in this Report would move the U.S.
tax system in the direction of more neutral taxation of capital income and, in so doing, reduce
current tax-induced distortions in the allocation of
capital. All the systems of corporate integration
would substitute a single level of tax for the
existing two level classical corporate tax system.
The CBIT prototype also would eliminate tax
distortions in the choice between corporate and
noncorporate forms of business organizations by
taxing all business income uniformly, at entity
level tax rates.

transition period, and the revenue impact of
different rules may affect the feasibility and the
desirability of different integration prototypes.
These issues raise important and controversial
issues of tax policy apart from their effects in
structuring an integrated corporate tax system.
Current law reflects compromises among goals of
economic efficiency, equity in taxation, and other
political, social, or economic policy goals (including furthering, for example, specific categories of
investment) as well as the coordination of taxation
across international borders.

Each of the systems of corporate tax integration is economically equivalent if income earned
by corporations and individuals were taxed at the
same tax rate, all income earned by corporations
were treated the same, and all investors were
taxed at the same tax rates. l But they are not. 2
The existence of differing tax rates among individuals and between corporations and individuals,
tax preferences for a variety of kinds of income
and deductions, domestic tax-exempt and foreign
suppliers of capital, and foreign source income
earned by U. S. corporations create significant
differences among basic systems of integration.
These circumstances also raise fundamental
structural issues that must be addressed within the
context of each of the integration systems. How
these issues are resolved in an integrated corporate tax system significantly affects the choices
among the basic integration alternatives and,
ultimately, the efficacy of the method chosen in
reducing or eliminating the distortions associated
with the classical corporate tax system.

The appropriate connection between such
policy considerations and the construction of an
integrated corporate tax system is further complicated because the Internal Revenue Code to date
has addressed questions concerning tax preferences, tax-exempt suppliers of corporate capital,
international considerations, and tax rates only in
the context of a classical corporate tax system, not
within the structure .of an integrated system.
Indeed, in some cases, provisions of current law
have been enacted, at least in part, to redress the
burdens of the classical corporate tax. Therefore,
the treatment of these specific issues under current
law mayor may not be the appropriate benchmark
for resolving the issue under an integrated system.
On the one hand, current law tax rules have had
a major impact on economic decisions and have
shaped a wide variety of existing fmanciaI arrangements; care must be exercised so unwarranted disruptions do not occur in moving to an
integrated corporate tax system. On the other
hand, the resolution of these issues may have
considerable influence on the degree of success of
an integrated corporate tax system in removing
the distortions of the existing system. Our task,
therefore, has been to approach these issues in a
manner that advances this Report's fundamental

Transition rules also must be addressed in any
integration proposal. The speed and administrative
ease with which integration can be implemented,
the degree of distortion experienced during the

61

Principal Issues

objective-more neutral taxation of capital income-where practical, without demanding that a
move from a classical to an integrated corporate
tax system be accompanied by a comprehensive
reevaluation of such fundamental issues as the
treatment of tax preferences or international
business transactions.

62

Although this part discusses these issues as
discrete topics, they are often interrelated. For
example, decisions regarding the use of tax
preferences may affect decisions concerning the
treatment of tax-exempt shareholders, and decisions concerning tax-exempt shareholders may
influence policies regarding foreign investors.

CHAPTER

5:

TREATMENT OF TAX PREFERENCES

Under current law, the Code provides favorable treatment that is generally recognized as
deviating from standard accounting rules for
particular items of income or expense. 1 These tax
preferences may take the form of exclusions of
income or preferential rates for items of income,
accelerated deductions or deferred income recognition rules or credits. Some preferences (like the
exclusion for interest on certain state and local
bonds) create a permanent reduction of tax liability. However, most corporate preferences (like
accelerated depreciation) offer deferral of tax,
rather than outright exemption.

The expanded tax bases for the AMT and the
relatively narrow rate differentials between the
regular and minimum' taxes make the minimum
tax provisions of current law a powerful revenue
source with widespread impact on the tax planning of both high-income individuals and corporations. If the corporate AMT were repealed, a
significant increase in the corporate tax rate would
be required to offset the revenue loss. The minimum tax provisions not only raise revenue directly but also serve to increase the regular income
tax paid by individual and corporate taxpayers
who limit their use of preferences to avoid being
subject to the AMT.

Under current law, there are two mechanisms
for restricting the use of business tax preferences:
the earnings and profits rules and the corporate
and individual minimum tax provisions. The
earnings and profits rules defme the pool of
corporate earnings that is taxable as dividends
(rather than as a return of basis or as capital gain)
when distributed to shareholders. Earnings and
profits are calculated to include most corporate
tax preferences. Thus, income that is tax-preferred at the corporate level is generally subject to
tax when it is distributed to noncorporate shareholders. 2 Thus, under current law, tax preferences may provide corporations with retainable, but
not necessarily distributable, tax-preferred funds.

In integrating the corporate and shareholder
income tax systems, the fundamental question
about tax preferences is the continuing role of
limitations on corporate tax preferences. Some
commentators have suggested that integration
implies giving to shareholders tax reductions due
to corporate level tax preferences. 3 They argue
that if integration is to achieve tax neutrality
between corporate and noncorporate investments,
extending preferences to shareholders is appropriate. The cost of not extending to shareholders
preferences that are available to noncorporate
businesses is retaining a bias against the corporate
form for any activities that are granted tax preferences. Such activities will tend to be performed
by noncorporate firms. As discussed in Chapter 1,
an economic loss results to the extent that such
activities could be carried on by corporations at
lower costS. 4

A strengthened minimum tax for both individuals and corporations was a central feature of the
Tax Reform Act of 1986. Under current law, the
alternative minimum tax (AMT) is payable only if
the computation of the minimum tax produces a
tax greater than the tax due under the regular
computation. For individuals, the AMT is imposed at a 24 percent rate on an expanded tax
base that includes most tax preference items. In
the case of corporations, the AMT is imposed at
a 20 percent rate on a tax base that includes a
broad list of tax preference items. The corporate
minimum tax serves to limit the capacity of tax
preferences to reduce tax on retained, as well as
distributed, earnings.

With respect to deferral preferences, such as
those permitting rapid depreciation or amortization of capital expenditures, some analysts regard
distribution of the related income to shareholders
as the appropriate occasion for ending tax deferral
and view the earnings and profits provisions of
current law as appropriately serving that function.
Retaining the approach of current law and taxing
preferences when distributed to shareholders
would continue some disadvantages for

63

Principal Issues

distributed, as opposed to retained, earnings, but
this could be mitigated by treating distributions as
coming fIrst from fully-taxed income. Where
corporate tax preferences are intended to alleviate
the classical system's double taxation of equity
income, they serve no function in an integrated
system and, at a minimum, should not be passed
through to shareholders. Some analysts, for
example, consider the reduced rate on the frrst
$100,000 of corporate income as a tax preference
intended to reduce the degree of double taxation
for small corporations that decline to elect (or are
ineligible for) S corporation status.
In addition, there are substantial revenue costs
to extending corporate level preferences to shareholders just as there are in cutting back on the
AMT.5 The revenue cost of extending preferences to shareholders or limiting the impact of the
AMT would increase the cost of corporate integration, require higher tax rates to produce equivalent revenues, and, in effect, increase the value
of tax preferences relative to taxable income.
Maintaining current law restrictions on tax preferences would reduce the need to raise tax rates and
thus reduce the efficiency costs associated with
such rate increases. 6 Hence, the issue of the
proper treatment of preferences involves a comparison of these possible costs with the benefits
provided by the preferences in an integrated
world.

64

detennine whether they are justifiable in an
integrated system, but such a comprehensive
review of tax preferences is beyond the scope of
this Report. This Report concludes, however,
that, where practical, integration of the corporate
tax should not become an occasion for expanding
the scope of tax preferences. Neither equity nor
economic efficiency would be enhanced by such
an expansIOn.
In practice, this conclusion implies that in a
distribution-related integration prototype, specific
mechanisms must be devised to playa role similar
to the earnings and profits provisions of current
law to ensure that preferences are not extended to
shareholders. Similarly, the role and function of
both the corporate and individual AMT must be
reexamined to prevent the extension of the scope
of current tax preferences.

Finally, if a goal of integration is to tax
corporate income once, corporate tax preferences
should not be extended to shareholders. In an
integrated system, extending preferences to shareholders may eliminate both the individual level
and the corporate level tax. Foreign systems
generall y do not allow corporate preference
income to be distributed tax-free to shareholders.
Belgium, Canada, Denmark, and Japan are
exceptions. 7

A simple dividend exclusion or shareholder
imputation credit method of distribution-related
integration will not produce the desired result with
respect to preference income. 8 Integrated tax
systems outside the United States that do not
extend corporate tax preferences to shareholders
have principally relied on either or a combination
of two mechanisms. 9 The frrst is an imposition of
corporate level tax on the distribution of preferences through a compensatory tax system. IO The
second is a tracing mechanism or overall lim itation that restricts the amount of relief from tax at
the shareholder level to actual corporate level
taxes paidY The lirilitation mechanism eliminates the benefit of preferences on distributed
income by imposing tax at the shareholder rate on
distributed preference income. The two methods
can vary significantly when the shareholder tax
rate differs from the corporate tax rate, and
would, for example, impose very different tax
burdens on distributions of corporate preference
income to tax-exempt shareholders. 12

Integration of the corporate and individual tax
systems provides an opportunity to review both
corporate and noncorporate tax preferences to

The choice between the two mechanisms is a
close one and a different alternative may be more
appropriate depending on the method of

65

integration adopted. In the distribution-related
integration prototypes described in this Report, we
have recommended limiting tax relief at the
shareholder level to the amount of corporate taxes
paid and imposing shareholder level tax on
distributed preferences. Under the dividend
exclusion prototype, this is accomplished by
requiring corporations to keep an account limiting

Principal Issues

excludable dividendsY In CBIT, this mechanism
also is possible; on the other hand, since all tax is
paid at the entity level, a compensatory tax may
have more appeal. 14 We conclude that it is not
practical to attempt to retain the current law tax
on distributed preference income under the shareholder allocation prototype. 15

CHAPTER

6:

TAX-EXEMPT AND TAX-FAVORED INvEsTORS

6.A INTRODUCTION

The Code exempts these entities from income
tax on all receipts other than net income from a
business unrelated to the entity's exempt purpose.
Such unrelated income, whether earned directly or
through a partnership, is subject to the unrelated
business income tax (UBIT) , which generally is
calculated under the regular corporate income tax
rules. 3 The tax generally applies only if the
business income is unrelated to the organization's
exempt purpose. Thus, engaging in a particular
activity might result in the imposition of UBIT on
one type of exempt organization but not on another. The Federal Government and State and local
governments or their instrumentalities (except
colleges and universities) are exempt from all tax
including UBIT. The Code explicitly excludes
income from certain passive investments from
UBIT, including dividends, interest, rent from
real property, royalties, and gains from the sale of
capital assets. Despite the general exclusion,
passive income generally is subject to UBIT to the
extent that it is fmanced with debt.

Current law defmes many different types of
tax-exempt entities (including pension funds,
charities, hospitals, educational institutions and
business leagues) and imposes various conditions
in order for them to obtain or retain their taxexempt status (including nondiscrimination rules,
minimum payout requirements, limitations on
maximum contributions and restrictions on investments). Tax exemption is generally limited to
income received by the entity that is either passive
in nature or substantially related to an exempt
function.
Tax-exempt entities may be grouped into two
general categories. One group, which includes
pension funds, 401(k) plans, and similar plans
(collectively, pension funds), is characterized by
an exempt entity that holds claims to property on
behalf of specific individuals, with the earnings of
the fund untaxed as earned but taxed when distributed to the individuals. The second group, which
includes charities, hospitals, educational and
religious institutions, is characterized by investment income that does not inure to the benefit of
any particular individuals. I

The tax-exempt sector plays a major role in
U.S. capital markets, and in the corporate capital
market in particular. At the end of 1990, pension
funds and other exempt organizations held over
one-quarter of total fmancial assets in the United
States (Table 6.1). Holdings of the tax-exempt
sector represented even larger fractions of corporate equity and corporate debt-approximately 37
percent of directly held corporate equity and 46
percent of outstanding corporate debt.

Tax exemption provides both groups with a
higher after-tax rate of return on investment
income than if the earnings were currently taxable. Retirees receive higher after-tax retirement
income than if pension fund earnings were taxed
currently or they had invested in taxable savings
plans themselves, and charities and educational
institutions can provide more services or activities
than if the income on their assets were taxable.
Despite the differences in the mechanics of taxing
pension funds and other exempt entities, the
present value benefit is the same. The pension
fund tax exemption, employer deductibility of
contributions to the fund and deferral of employee
tax is equivalent to simply exempting from income tax the pension fund's investment income. 2

Pension funds dominate tax-exempt sector
corporate investments, holding more than onequarter of all directly held corporate stock and
more than two-fifths of corporate bonds. Figure
6.1 illustrates the dramatic growth in the share of
corporate debt and equity held by pension funds
since the 1950s. As the share of corporate capital
held by pension funds has grown, an increasing
share of the associated corporate income has
avoided the investor level tax.

67

68

Principal Issues

Table 6.1
Financial Assets of the Tax-Exempt Sector
End of Year 1990
Total Credit Market Assets l

Corporate Equity

Corporate Debf

(billions
(billions
(billions
of dollars) (percent) of dollars) (percent) of dollars) (percent)

1,636
Foreigners
2,695
Pension Funds3
!RAs & Keoghs4
560
Nonprofit InstitutionsS
515
Total Tax-Exempt Sector 5,450
Total All Sectors
13,996

12
19
4
4
39
100

218
967
141
130
1,457
3,416

6
28
4
4
43
100

203
722
11
10
946
1,629

12
44
1
1

58
100

Department of the Treasury
Office of Tax Policy
ITotal Credit Market Assets: total credit market debt owed by domestic nonfinancial
sectors plus corporate equities (excluding mutual funds).
2Corporate Debt includes some foreign bonds. The total amount includes bonds held
by the financial sector.
3Pension Funds include private pension funds (including Federal Employees
Retirement Thrift Savings Fund), state and local government employee retirement
funds, and pension fund reserves held by life insurance companies.
4lndividual Retirement Accounts (!RAs) and Keogh accounts: figures estimated.
sNonprofit institutions include charitable, educational, and similar institutions.
Estimated as percent of household holdings in Flow of Funds.
Sources: Federal Reserve Board, Flow of Funds (March 1991 revised); Investment
Company Institute, Mutual Fund Fact Book (1991), p. 60; and Office of Tax Policy
calculations.

Figure 6.1
Pension Fund Holdings of Corporate Capital, 1950-1990

so

J
j

40
Corporate bonds

30
/

:t

/
/

."

~

~

J

20
Corp. debt + equity ..

./

/

/

--- --- ---

---

/

Corporate equity

/

lO

---

OL-~---,--

1950

1955

--- --- ---

--- ---

__~__~__- ,____~__~__- ,__~__

1960

1965

1970

1975

1980

1985

1990

Year

Sources: Hoffman (1989) and calculations based on Federal
Reserve Board, Efuw. of Eunds (March 1991 revised).

Under current law, tax.exempt investors, in fact,
are not exempt from the
corporate level tax on
income from their corporate
equity investments.
Although dividends paid to
tax-exempt shareholders are
not taxed to the recipients,
the earnings attributable to
such investors are taxed at
the corporate level whether
or not distributed. By contrast, corporate earnings
paid to tax-exempt investors
as interest escape both the
corporate level tax and the
investor level tax.
The fundamental question addressed here is
whether under an integrated
tax system this treatment of
corporate income of taxexempt investors should
continue, or, alternatively,
whether tax-exempt investors should be subject to a
tax increase or receive a tax
reduction from integration.
The current level of taxation of corporate equity
income received by taxexempt investors can be
retained under integration
as demonstrated in this
Report. Integration does not
necessarily require either an
increase or a reduction in
tax on income from capital
supplied by tax-exempt
entities to corporations.
On the other hand,
corporate integration presents an opportunity to
reexamine the incentives
under current law for taxexempt investors to prefer

Principal Issues

69

debt rather than equity investments in corporations. The specific question raised by corporate
integration is whether the current distinction in the
treatment of corporate equity investments by taxexempt entities (which bear the corporate, but not
the shareholder level tax) versus corporate debt
investments (which bear neither corporate nor
debtholder level tax) should be retained or decreased. An integration system best fulfills its
goals if it provides uniform treatment of debt and
equity investments by tax-exempt investors.
Equating the tax treatment of debt and equity will
require either an increase or decrease in the taxes
on corporate capital supplied by tax-exempt
investors or the introduction of a separate tax on
investment income of these investors. As Section 6.D discusses, such a tax could be designed
to maintain the current level of tax on income
from corporate capital supplied by tax-exempt
investors while equalizing the treatment of debt
and equity.

6.B

DISTORTIONS UNDER
CURRENT LAW

Current law encourages tax-exempt investors,
like taxable investors, to invest in debt rather than
equity. Only two types of income from capital
supplied to corporations by tax-exempt entities are
actually tax-exempt. Interest paid by corporations
is both deductible by the corporate payor and
exempt from tax in the hands of the tax -exempt
recipient. Corporate preference income distributed
to tax-exempt shareholders also is exempt from
tax at both the corporate and the shareholder
level. 4 Non-preference income is taxed at the
corporate level, but is not taxed at the shareholder
level whether it is received by the exempt investor
as capital gains from the sale of shares or as
dividends from distributions. Thus, under current
law, corporate income paid to tax-exempt investors in the form of interest is not taxed at either
the corporate or investor level, while non-preference income retained or distributed to tax-exempt
shareholders is subject to tax at the corporate
level.
Current law does not, however, encourage
tax-exempt investors to invest in equity of

noncorporate rather than corporate businesses, because, in both cases, the income is subject to one
level of tax. While corporate income (other than
preference income) allocable to tax-exempt shareholders is subject to tax at the corporate level, the
noncorporate unrelated business income of taxexempt investors generally is subject to UBIT.5
For tax-exempt investors who invest in equity,
current law generally also does not affect their
preferences for distributed or retained earnings.
Because corporate income (other than preference
income) is subject to current corporate level tax
and both distributed and retained earnings are
exempt from tax at the shareholder level, a taxexempt shareholder has no tax incentive to prefer
distributed earnings over retained earnings.

6.C

NEUTRALITY UNDER AN
INTEGRATED TAX SYSTEM

Because of the asymmetric treatment of debt
and equity investments by tax-exempt entities
under current law, an integrated system can
achieve neutrality between debt and equity investments for tax-exempt investors only by either
decreasing the tax burden on equity income or
increasing the tax burden on interest. A straightforward decrease in the tax burden on equity
investments might be accomplished by removing
the corporate level tax on earnings distributed as
dividends to tax-exempt investors. A deduction
for corporate divide~ds, for example, would
achieve this result. The contrary approach might
subject interest income on corporate debt earned
by tax-exempt investors to one level of tax (at
either the corporate or the investor level).
The ftrst approach, taxing neither dividends
nor interest paid to tax-exempt investors, would
lose substantial amounts of tax revenue relative to
current law. Extending the beneftts of integration
to tax-exempt investors would add costs of approximately $29 billion annually under distribution-related integration and approximately $42
billion annually under shareholder allocation. This
revenue loss would increase the costs of integration and would require offsetting increases in
other taxes or in tax rates, which might create or
increase other distortions. This approach also

Principal Issues

would distort the choice between corporate and
non corporate investment for tax -exempt investors
if UBIT remained in place for noncorporate
investment. If corporate dividends were taxexempt at both the corporate and investor level,
while earnings from businesses conducted directly
or in partnership form were subject to UBIT, a
tax-exempt investor would always prefer corporate dividends. Indeed, anti-abuse rules might be
required to preclude tax-exempt organizations
from avoiding UBIT altogether simply by incorporating their unrelated businesses.
The second approach, taxing both interest and
dividends at a single rate, would reduce the
current advantage of tax-exempt investors relative
to taxable investors. Tax-exempt investors would
no longer enjoy an after-tax return on a given
corporate equity or debt investment higher than
that available to taxable investors. The principal
advantage of this approach is that it would equate
the treatment of debt and equity while maintaining
the neutrality between corporate and noncorporate
equity for tax-exempt investors. 6

6.D

GENERAL
RECOMMENDATIONS

This Report recommends that a level of
taxation at least equal to the current taxation of
corporate equity income allocated to investments
by the tax-exempt sector be retained under integration. The dividend exclusion prototype, described in Chapter 2, essentially continues present
law treatment of tax-exempt investors under an
integrated tax system, so fully-taxed corporate
profits would continue to bear one level of tax
and preference income would not be taxed at
either the corporate or shareholder level. 7 A
similar result can be accomplished under an
imputation credit system of integration, but a
dividend deduction system would eliminate the
current corporate level tax on distributed earnings
on equity capital supplied by tax-exempt investors. 8 Under the shareholder allocation prototype
described in Chapter 3, taxes are collected at the
corporate level on corporate income allocable to
investment by tax-exempt shareholders and no
refund is provided to nontaxable shareholders.

70

Maintaining one level of tax on equity investments by tax -exempt entities would promote one
of the primary goals of integration: achieving tax
neutrality for all investors between corporate and
noncorporate investments. This choice is consistent with a move to integration for taxable shareholders, because choosing to reduce the double
tax burden on corporate income distributed to
taxable investors does not necessarily dictate a
commensurate reduction in the tax burden on taxexempt investors. Finally, continuing to tax equity
investments by the tax-exempt sector avoids the
revenue loss that would result if such investments
were completely tax-exempt. Increasing other tax
rates to compensate for such a revenue loss would
entail other inefficiencies.
Some countries that have adopted integration
have chosen to tax separately corporate and other
income allocable to tax-exempt investors. For
example, in moving to an integrated corporate
tax, Australia and New Zealand imposed a tax on
the income of pension funds, thus reducing the
number of tax-exempt investors. In both countries, the remaining tax-exempt investor base,
such as charities, is small. Australia imposed a 15
percent tax on investment income earned by
pension funds and made available the full 39
percent imputation credit from dividends as a
nonrefundable offset. Australia did not project
collecting more than a token amount of tax from
this tax on investment income: it devised the
mechanism to remove distortions between investing in domestic corporations (which pay Australian tax) and investing in foreign corporations
(which generally do not). The new Australian
system also removes distortions between investing
in equity and investing" in debt. New Zealand went
further and repealed entirely the tax exemption of
pension funds; they now function basically as
taxable savings accounts. Under the U.K. distribution-related integration system, the corporate
level tax is not completely eliminated, with the
consequence that income distributed to tax-exempt
shareholders bears some tax burden. 9
This Report also encourages an effort to
achieve uniform tax treatment of corporate debt
and equity investments by tax-exempt investors.

71

Because of the important role played by the taxexempt sector in the capital markets, failing to
create neutrality for debt and equity investments
by the tax-exempt sector would limit the extent to
which integration could achieve tax neutrality
between the two kinds of investments. This is
achieved under CBIT by treating tax-exempt
shareholders and debtholders generally like other
suppliers of cOIporate capital, with tax imposed at
the cOIporate level. 10
One potential alternative approach would tax
all cOIporate and noncoIporate income allocable to
investment by the tax-exempt sector at a rate
lower than the rate applicable to taxable investors.11 Such a tax on the investment income,
including dividends and interest income, received
by tax-exempt entities could be set to achieve
overall revenues equivalent to those currently
borne by cOIporate capital supplied by the
tax-exempt sector. Under the imputation credit

Principal Issues

prototype discussed in Chapter 11, for example,
imputation credits for cOIporate taxes paid would
be allowed to tax-exempt shareholders. To the
extent that the credit rate exceeds the tax rate on
investment income, the excess credits could be
used to offset tax on interest or other investment
income. In addition to the substantial advantage of
equating the tax treatment of debt and equity held
by such investors, such an approach would allow
tax-exempt investors to use shareholder level
credits for cOIporate taxes paid to the same extent
as taxable shareholders. 12 By doing so, this
approach would limit both portfolio shifts and
other tax planning techniques that might otherwise
be induced by efforts to distinguish among taxable
and tax-exempt investors in integrating the coIporate income tax. A revenue neutral rate for such
a system would be in the range of 6 to 8 percent
depending on the prototype. 13 This would approximate the current law cOIporate tax burden on
investments by tax-exempt shareholders.

CHAYfER 7:
TREATMENT OF FOREIGN INCOME AND SHAREHOLDERS

7.A

INTRODUCTION

law also allows corporate taxpayers that receive
dividends (or include Subpart F income) from at
least 10-percent owned foreign subsidiaries to
claim a foreign tax credit for a ratable portion of
the qualifying foreign taxes paid by the subsidiary
on the income from which the dividends are paid
(the indirect foreign tax credit). The portion of
the foreign taxes which taxpayers may claim as an
indirect credit is proportional to the fraction of the
earnings of the foreign subsidiary distributed or
deemed distributed. The dividend income for U. S.
tax purposes is grossed up by the amount of the
direct and indirect credits claimed. 3 The indirect
foreign tax credit, like the dividends received
deduction available domestically, prevents multiple taxation of corporate profits at the corporate
level.

International issues are important in designing
an integrated tax system because there is substantial investment by U. S. persons in foreign countries (outbound investment) and investment by
foreign persons in the United States (inbound
investment). At the end of 1990, private U.S.
investors owned direct investments abroad with a
market value of $714 billion, and $910 billion in
foreign portfolio investment, while private foreign
investors owned $530 billion in direct investment
in the United States and $1.34 trillion in U.S.
portfolio investment. U. S. investors received a
total of $54.4 billion of income from their direct
investments abroad in 1990, and $65.7 billion of
income from their foreign portfolio investments,
while foreign investors received $1.8 billion from
their direct investments in the United States in
1990 and $78.5 billion from their U.S. portfolio
investments.

The Code limits the maximum foreign tax
credit to prevent the foreign tax credit from
offsetting taxes on domestic source income.
Separate limitations apply to several different
kinds of foreign source income (baskets) in order
to restrict the use of foreign tax credits from hightaxed foreign source income against low-taxed
foreign source income. For each basket, the Code
limits the amount of foreign taxes paid on income
in that basket which a taxpayer may claim as a
credit in the current year to a fraction of the
taxpayer's pre-credit tax on worldwide income in
the same basket. The fraction is the ratio of the
taxpayer's foreign source taxable income in the
basket to the taxpayer's total worldwide taxable
income in the same basket. Credits that a taxpayer
cannot use in a given year because of the limitations may be carried back two years or forward
five years. Additional limitations apply to taxpayers subject to the alternative minimum tax.

The income from transnational investments
may be taxed by both the country in which the
investment is made (the host or source country)
and the country of residence of the investor (the
residence country). The United States uses two
primary instruments for mitigating the potential
problem of double taxation: the foreign tax credit
and bilateral income tax treaties entered into
between the United States and about 40 other
countries.
Taxation of foreign investment by U. S. investors. The United States taxes the worldwide
income of its residents.l The U.S. tax on income
earned by U.S. corporations or individuals
through foreign corporations is generally deferred
until such income is repatriated through dividend
or interest payments to U.S. shareholders or
creditors. 2

Taxation of foreign investors. The taxation of
U.S. investment income of foreign individuals or
corporations generally depends upon whether they
are engaged in a trade or business in the United
States. Foreign corporations and individuals
engaged in a U.S. trade or business generally are

The United States allows taxpayers to claim a
foreign tax credit for qualifying foreign income
taxes paid (the direct foreign tax credit). Current
73

Principal Issues

taxed on their net business income under the same
rules that apply to a U.S. corporation or citizen
engaged in the same business.
The treatment of domestic and foreign investors differs, however, at the shareholder and
creditor level. Foreign investors not engaged in a
U.S. trade or business are not subject to the
individual or corporate income tax. 4 Instead,
subject to significant exceptions noted below, they
are subject to a 30 percent withholding tax on
their gross dividend, interest and other income.
Capital gains realized by a foreign investor on the
sale of stock or securities (except stock in certain
U.S. corporations owning U.S. real property)
generally are exempt from tax.
The Code exempts from the 30 percent withholding tax qualified portfolio interest and interest
earned by foreign investors on U. S. bank deposits. Interest does not qualify as portfolio interest
if the investor has a 10 percent or greater equity
interest in the borrower or is a controlled foreign
corporation related to the borrower or if the
interest is paid on a bank loan made in the ordinary course of a banking business.
Under bilateral tax treaties, interest (if not
already exempt) and dividends and other income
paid to residents of a treaty country may qualify
for a significantly reduced rate of withholding tax.
The reduced rate of withholding tax applicable to
dividends is often IS percent and may be as low
as 5 percent on dividends distributed by aU. S.
subsidiary to a foreign direct corporate investor.
Tax treaties may reduce the rate of withholding
on otherwise taxable interest income paid to
foreign investors (in particular, related foreign
investors) to 5 or 10 percent or, in many cases,
zero.
The current U. S. tax treatment of cross-border
investment generally reinforces the biases created
by other features of the classical system of corporate taxation: against equity compared to debt and
for retention rather than distribution of corporate
earnings. Statutory exemptions for cross-border
interest payments, together with more favorable
treaty provisions for interest than for dividends,

74

reinforce the bias against equity. Likewise, the
potential for deferral of U. S. tax liability on nonSubpart F income reinforces the bias towards
retention of such income by foreign
subsidiaries.
The major international issues that must be
addressed in any integrated system are:
•

Should foreign taxes paid by U.S. corporations be
treated identically to taxes paid to the U. S. Government? If so, the foreign tax credit for corporate
taxes paid, in effect, would be extended to shareholders. As a consequence, income that is taxed
abroad at a rate equal to or greater than the U.S.
tax rate would not be subject to U.S. tax either at
the corporate level or at the shareholder level.

•

Should the benefits of integration be extended to
foreign shareholders? If so, income allocable to (or
paid to) foreign shareholders would be subject to
only one level of U.S. tax, at either the corporate
or shareholder level. If the tax is imposed only at
the shareholder level, U. S. income tax treaties may
substantially reduce the tax.

This Report recommends that: (1) foreign
income taxes paid with respect to outbound
investment not be treated the same as U. S. taxes
paid for integration purposes, (2) foreign shareholders not receive by statute benefits of integration received by U.S. shareholders, and (3) the
United States' income tax treaties with other
countries be used as the appropriate vehicle for
relaxing either of the preceding rules where
reciprocal benefits are given by the foreign country to U. S. taxes or investors in their integration
systems.

7.B

OVERVIEW OF U.S.
INTERNATIONAL TAX
POLICY

As indicated above, cross-border investments
are potentially taxable in at least two countries:
the residence country (the country where the
investor resides) and" the source country (the
country where the investment is made). Sovereignty unavoidably complicates international tax
policy: a country may set its own tax policies, but
not the policies of other countries, even though
the policies of other countries have a direct

Principal Issues

75

impact on the fIrst country's welfare. As a result,
a residence country generally must respect a
source country's claim to tax income that is
derived within the source country's borders.
However, the source country has little control
over the ultimate level of aggregate taxes paid by
foreign investors on profIts earned in the source
country. By choosing to impose additional tax on
an investor's income from the source country, by
exempting such income from its own tax, or by
choosing some intermediate policy, the residence
country, not the source country, makes the fmal
decision about the tax burden borne by the residence country's investors.

Normative Guidance for
International Tax Policy
No consensus exists about the proper norms
for capital taxation in economies with international capital and labor mobility. Integrating models
of capital taxation and international trade, policymakers have suggested two principles for taxation
of international investments:
•

Principle 1 (Capital Export Neutrality). Investors
should pay equivalent taxes on capital income,
regardless of the country in which that income is
earned.

•

Principle 2 (Capital Import Neutrality). All investments within a country should face the same tax
burden, regardless of whether they are owned by a
domestic or foreign investor.

Maintaining both principles sh-nultaneously is
not a practical option, however, because it would
require that capital income be taxed equally in all
countries. That will never occur as long as sovereign countries establish different tax rates.
National tax systems, such as that of the
United States, can approach capital export neutrality while taxing worldwide income of resident
multinational enterprises (the worldwide method
of taxation), if either the residence country provides credits to its enterprises for taxes remitted
to foreign governments or the source country
surrenders the right to tax income from foreign
investments within its borders. Capital import
neutrality can be achieved if the residence country

decides not to tax income earned from foreign
jurisdictions and allows the source country to be
the sole taxing authority for international
investment income.
Since capital export and capital import
neutrality cannot be attained simultaneously when
international differences exist in capital income
taxation, a clear advantage for one or the other
would be useful. However, analyses of international taxation by eConomists specializing in
international trade generally offer no strong
endorsement of one principle relative to the
other. 5 Capital taxation in open economies (economies in which international borrowing and
lending occur) can distort both the level of saving
within an economy and its allocation among
alternative investments at home and abroad.
Capital import neutrality can enhance worldwide
economic effIciency if domestic savings are
ineffIciently low by reducing the tax burden on
savings.
Capital export neutrality, in contrast, enhances
worldwide effIciency in the allocation of savings.
It may be a guiding principle when effIciency
costs of distortions in the allocation of savings are
significant relative to costs of tax-induced distortion in the level of savings. Most available evidence supports the proposition that the sensitivity
of domestic savings with respect to changes in net
return is small relative to the sensitivity of the
location of investment with respect to changes in
net return. 6 Accordingly, many economists and
policy makers presume that capital export neutrality offers better guidance for international tax
policy. Nonetheless, given the existence of taxinduced distortions in both savings and investment, the complexity of the modern multinational
enterprise (relative to two-country examples often
considered in theory), and the possibility of
international tax competition, some compromise
between capital export and capital import
neutrality is inevitable. 7

Outbound Investment
Since 1918, through the foreign tax credit,
the United States has generally implemented the

Principal Issues

principle of capital export neutrality unilaterally
and without interruption. s Since 1921 the foreign
tax credit has been limited so it does not exceed
the U. S . tax liability incurred on the foreign
source income in the absence of the credit. The
limitation seeks to prevent the credit from offsetting U. S. tax on U. S. source income. However,
because the limitation allows a foreign tax rate
that is higher than the U. S. tax on the relevant
income to go unrelieved, the limitation works
against the policy of capital export neutrality.
A taxpayer generally receives a foreign tax
credit only for income taxes paid to a foreign
government on the taxpayer's own income. Thus,
a shareholder generally may claim a credit for
foreign taxes withheld from a dividend payment
includable in the shareholder's income but may
not claim a credit for the foreign taxes paid by the
corporation on the income out of which the
dividend is paid. The only exception to this
principle is the indirect foreign tax credit allowed
for a domestic 10 percent corporate shareholder of
a foreign corporation for the foreign income taxes
paid by the foreign subsidiary on the income out
of which the dividend is paid. 9
In other respects, however, the U.S. taxation
of outbound investment tends toward capital
import neutrality-the tax rate on foreign source
income of aU. S. investor is detennined by the
tax imposed by the source country. First, the U. S.
tax regime generally allows deferral. That is, the
U. S. tax on foreign source income of U. S. owned
foreign companies is deferred until such profits
are repatriated in the fonn of dividends. Deferral
affects a U.S. investor's initial decision to make
or forgo a foreign investment because, even if the
investor is obligated to pay the residual U.S. tax
(a capital export neutral result), the time for
paying this tax may be postponed indefinitely.
Deferral thus substantially reduces, and under
some conditions virtually eliminates, the present
cost of the residual U.S. tax (a capital import
neutral result).10 Deferral, however, is not significant with respect to dividends paid from
current earnings, or where foreign tax rates equal
or exceed the U.S. corporate rate. In addition,
certain foreign corporations controlled by U. s.

76

residents are subject to current U. S. tax on certain
types of undistributed income under the Code's
Subpart F rules. The advantage of deferral also is
less where the domestic corporate ownership
interest is less than 10 percent of the voting stock
in the foreign corporation. In that case, the indirect foreign tax credit is not available. Thus,
dividends will incur both the foreign corporate
level tax and, after deduction of the foreign tax,
the U. S. corporate level tax.
Second, the U.S. tax regime allows averaging.
That is, in detennining the residual U. S. tax on
foreign profits, a high foreign tax imposed on one
item of foreign income may be averaged against
a low foreign tax imposed on another item of
foreign income, as long as the different items of
income are both within the same statutory basket
for purposes of the foreign tax credit limitation
rules. If the foreign tax rate on an item of foreign
income is higher than the U. S. rate, the U. S.
investor mayor may not bear the cost of the
higher foreign rate, depending on the opportunities for averaging. If the investor must bear the
higher rate, it is placed in parity with local investors in the foreign country, a capital import
neutral result. If, on the other hand, the investor
is able to average the high foreign tax rate on the
income in question against low foreign rates on
other foreign income, then the investor will avoid
the extra burden of the high foreign rate. This
should render the investor capital export neutral
with respect to the highly taxed foreign income
(since averaging will reduce the total tax on such
income to the U. S. rate, but no lower), but also
should render the investor capital import neutral
with respect to the lower taxed foreign income
(because the investor is able to escape some of the
residual U.S. tax on such income). The opportunities for averaging have been reduced since the
1986 Act created separate foreign tax credit
limitation baskets for specific types of income.

Inbound Investment
U. S. tax policy on inbound investment generally asserts a substantial source country claim to
tax on certain types of income coupled with a
policy of nondiscrimination against foreign

77

investors. For foreign owned corporate investment, the United States generally imposes two
levels of tax. Thus, the United States taxes the
business profits of foreign owned domestic corporations or U. S. branches of foreign corporations
similarly to the profits of u.S. owned domestic
companies and imposes significant withholding
taxes on dividends paid to foreign investors. The
U.S. rules for taxing the u.S. branch of a foreign
corporation also are designed to impose on the
branch's profits the same amount of tax that
would be imposed if the branch were a subsidiary
of aU. S. cOlporation. The major exceptions to
the general U. S. policy are the exemption of
much of the interest income that is paid from
u. S. sources to unrelated foreign lenders (other
than banks), the decision to exempt capital gains
not effectively connected with aU. S. business or
attributable to a u.S. real property interest, and
the reduction of withholding taxes on dividends,
non-exempt interest, and royalties paid to foreigners (whether or not related) through bilateral
treaties. 11
The United States's network of bilateral
income tax treaties significantly modifies the
statutory orientation toward source country taxation. In general, tax treaties boost the tax claims
of the residence country, largely by substantially
reducing the withholding rates at source on investment income. In addition, tax treaties may require
higher levels of business activity (a permanent
establishment) before asserting a U.S. claim to tax
business profits. 12

7.C

INTERNATIONAL TAX
POLICY AND INTEGRATION

Outbound InvestmentTreatment of Foreign Taxes
This Report generally recommends that, in an
integrated tax system, the statutory treatment of
foreign taxes paid by corporations should differ
from the treatment of the taxes they pay to the
u. S. Government. Equal statutory treatment of
foreign and U. S. corporate level taxes would
significantly reduce the current U. S. tax claim
against foreign source corporate profits and often

Principal Issues

would completely exempt such profits from U.S.
taxation at both the corporate and shareholder
levels. Such unilateral action would result in a
significant departure from the prevailing allocation
of tax revenues between source and residence
countries. 13
The integration systems recommended in this
Report, therefore, generally retain the corporate
level foreign tax credit but do not extend to
shareholders the benefits of a foreign tax credit
for foreign taxes paid by the corporation. However, where foreign income is taxed at a foreign rate
that is lower than the current U. S. corporate rate,
there would be less double taxation than under
current law, because corporate level residual tax
would be treated identically to any other U.S.
corporate taxes. 14 Foreign source income subject
to tax in the source country at source country
rates higher than the U. S. rate would continue to
be subject to a single level of U.S. tax when
distributed. Thus, although foreign source income
earned by U. S. corporations might be subject to
more tax than domestic income, foreign source
income generally would not be subject to double
taxation to any greater extent than under current
law. Retaining a single level of tax on foreign
income should not harm the ability of U.S. finns
to compete in foreign markets relative to current
law.
Critics of continuing to impose any U.S. tax
on foreign profits might contend that, because the
United States currently is willing to give up
entirely its tax on certain types of foreign profits,
it should be willing to do so generally for foreign
corporate profits in an integrated corporate tax
system. This argument is not compelling, however. To be sure, the United States does not always
currently insist on a single level of tax on foreign
source income, as evinced by its unilateral decision to grant a foreign tax credit to individuals
earning foreign income directly or through a
partnership. Individual profits from foreign
sources, however, have been a small fraction of
the foreign source profits earned by U. S. -based
multinational corporations, and the revenue loss
from such a policy has therefore been small
compared to that which would occur if foreign

Principal Issues

taxes paid by corporations eliminated U. S. tax at
both the corporate and shareholder levels. Moreover, allowing a foreign tax credit to individuals
on the foreign source income directly earned
alleviates the burdensome tax structure that would
otherwise arise under current law, because deferral would not be available and the foreign and
U.S. taxes would both be imposed currently.
Another potential criticism is that failure to
pass through foreign tax credits to shareholders
would violate capital export neutrality and, hence,
would be inconsistent with our underlying goal for
integration: to enhance economic efficiency. As
discussed above, however, it is not apparent that
export neutrality does, in fact, lead to an efficient
allocation of capital. In any case, if foreign tax
credits were available to offset the single level of
tax in an integrated system, the revenue loss
would be serious-approximately $17 billion a
year. Taxes would have to be raised elsewhere,
and that would generate its own inefficiencies.
Finally, passing through foreign tax credits to
shareholders would pose significant administrative
difficulties. The foreign tax credit limitation and
sourcing rules would have to be applied at the
individual shareholder level both to ensure that
taxpayers claimed the proper credit for foreign
taxes and to prevent the U. S. Treasury from
bearing the cost of high foreign tax rates. Without
these rules, shareholders in corporations with
foreign income that is taxed at a rate greater than
the U. S. rate could use the excess credits to offset
tax liability on domestic income, with the consequence that the U.S. Treasury would in effect
provide domestic shareholders with refunds of
corporate taxes paid to foreign countries. IS This
is a particularly serious issue because tax rates in
many foreign jurisdictions are higher than current
U.S. tax rates. The difficulty of ensuring the
availability of adequate infonnation concerning
foreign taxes to both the shareholder and the IRS
would complicate application of these rules at the
shareholder level for widely held, non-U.S.
controlled foreign corporations.
From a legal point of view, continuing to
impose a single shareholder level of residence

78

country taxation on foreign source income would
not violate the United States' treaty commitments
to eliminate double taxation by granting a foreign
tax credit. Because U. S. tax treaties generally
reflect an assumption that treaty partners have
classical systems of corporate-shareholder taxation, the United States' treaty obligations require
that U. S. corporations be allowed a foreign tax
credit against the U. S . tax on foreign source
income received directly by the corporation, and
that individuals be allowed a credit for foreign
source income received by the individual. No
treaty obligation requires the United States to
grant further relief with respect to foreign taxes
paid or deemed paid by a domestic corporation,
e.g., by eliminating the shareholder tax on a
taxable dividend under the dividend exclusion
prototype (or CBIT) or, if a compensatory tax is
imposed under CBIT, refunding the compensatory
tax. In specific circumstances, however, the
United States might agree to extend, by treaty, the
benefits of integration. to foreign taxes on profits
of U.S. multinationals.
Under the dividend exclusion prototype, a
problem with maintaining a single level of U. S.
tax on foreign earnings is a continued bias in
favor of the noncorporate, rather than the corporate, fonn for foreign investment, although, as a
practical matter, this problem may not be very
serious. Individuals would be entitled to a foreign
tax credit for foreign taxes imposed on their direct
investments but not for taxes imposed on the
investments of corporations of which they are
shareholders. Thus, by not treating foreign corporate taxes equivalently to U.S. corporate taxes, an
incentive to structure foreign investment through
partnerships would continue. If the corporate fonn
could not be avoided, there also would continue to
be an incentive to make foreign investments in the
fonn of debt, which would reduce the foreign tax
base and convert foreign profits to domestic
profits. Large investors might achieve similar
effects by using rental or royalty payments or by
aggressive transfer pricing.
The dividend exclusion and imputation credit
prototypes implement our policy recommendations
by maintaining the current foreign tax credit rules

79

and by limiting the amounts of excludable dividends to corporate income on which U.S. taxes
have been paid (or limiting shareholder imputation
credits to U.S. taxes paid).16 In effect, dividends
paid out of foreign source income not previously
subject to U.S. tax because of foreign tax credits
would be taxed fully at the shareholder level, as
under current law. Under CBIT, the U.S. tax may
alternatively be imposed through a compensatory
tax at the corporate level on distributions of
foreign source income shielded from regular
CBIT by the foreign tax credit. 17 In either case,
corporations are allowed to treat dividends as paid
flrst out of U.S. taxed income. Under the shareholder allocation prototype, foreign taxes, in
essence, would be treated as equivalent to U.S.
taxes, and this is among the reasons that this
prototype is not recommended in this Report. 18

Inbound InvestmentTreatment of Foreign Investors
The basic issue that an integration proposal
must resolve for inbound investment is whether,
by statute, the United States should continue to
collect two levels of tax on foreign owned corporate proflts or whether foreign investors should
receive beneflts of integration similar to domestic
investorsY For the reasons set forth below, this
Report recommends that, except in the case of
CBIT, foreign shareholders not be granted integration beneflts by statute, but instead that this
issue be addressed on a bilateral basis through
treaty negotiations. Most of the major trading
partners of the United States that have integrated
their corporate tax regimes have followed this
approach. 20
At least two basic obstacles restrain unilateral
extension of integration beneflts to foreign shareholders. The fIrst is the inherent limitation on any
source country's taxation of foreign investors. The
residence country, not the source country, ultimately decides the tax burden that should be
borne by its resident investors. As a consequence,
if the United States unilaterally extended the
benefIts of integration to foreign shareholders, it
would abandon its right to source country taxation
of dividends with no assurance that the foreign

Principal Issues

investors would not be subject to a second level of
tax in their country of residence. Substantial
revenue would be lost without any necessary
increase in efflciency of capital allocation.
The second obstacle is the interaction between
a U.S. integration system and existing treaty
obligations. For example, extending a refundable
imputation credit to foreign shareholders by
statute, combined with traditionally low treaty
withholding rates on dividends, could signifIcantly
reduce the aggregate U.S. tax on proflts distributed to foreign shareholders, without any comparable reduction in foreign taxes on U. S. investments
in the treaty country. 21
Thus, there is no reason for the United States
by statute unilaterally to extend the beneflts of
integration to foreign shareholders. Integration
seeks to provide relief for investors using the
corporate form, not for foreign governments. If a
second level of tax is to be collected, no obvious
conceptual or practical reason exists why the
source country should sacrifIce its claim to this
tax revenue for the sake of consistency.
Several of our treaty partners adopting imputation credit systems have concluded that refusing
to extend integration beneflts by statute to foreign
shareholders residing in treaty countries would not
violate the provisions of tax treaties that prohibit
discrimination based on capital ownership. These
countries argue that, under an imputation credit
system, all proflts are taxed at the corporate level
at the same rate (34 percent, for example), without regard to "capital ownership," and allowing or
denying the imputation credit to the shareholders
is an issue of how to tax the shareholder, not the
corporation. No treaty requires that foreign
shareholders receive the same tax credits as
domestic shareholders. Thus, there is no treaty
violation. Similar arguments could be made about
the dividend exclusion prototype. 22
As Chapter 2 indicates, the dividend exclusion
prototype generally would not provide any integration benefIts to foreign shareholders, because
current withholding taxes would continue to
apply. 23 Similarly, inbound investment in an

Principal Issues

imputation credit system would remain subject to
two levels of U. S. tax because imputation credits
would not be made available to foreigners and
current withholding taxes would continue to
apply. Neither approach would treat inbound
investment more harshly than under current law,
because deferral of the second level of tax would
continue. 24 A dividend deduction system, on the
other hand, would automatically extend the benefits of integration to foreign shareholders, unless
a rule were adopted to deny the deduction for
dividends paid to foreigners - a rule that would
violate u.s. treaty obligations. The shareholder
allocation prototype avoids extending the benefits
of integration to foreign shareholders by imposing
corporate level tax, continuing to impose withholding tax on dividends, and denying refunds of
corporate taxes paid to foreign shareholders. 25
In contrast, to ensure parity between debt and
equity, the CBIT prototype generally removes the
withholding tax on both dividends and interest of
CBIT entities and repeals the branch profits tax.
The result is that both debt and equity mcome
would be subject to tax once.
The United States may consider extending the
benefits of integration to foreign shareholders
resident in countries that have treaties with the
United States. The fundamental policy issue in
deciding whether and how to extend integration
by treaty to foreign shareholders is how to divide
the tax revenue from corporate profits between
the source country and the residence country. As
noted above, traditional treaty rules reflect an
allocation of revenue based on the classical,

80

two-tier tax system for corporations and shareholders: the source country generally has the
exclusive right to tax business profits earned
therein by a domestic corporation and the two
countries divide the right to tax the profits when
distributed, with the greater share of this revenue
going to the residence country. Integration, of
course, alters the original pool of tax revenue by
decreasing the total (assuming no offsetting rate
increases) and by reallocating it between the
shareholder and corporation. Thus, moving to an
integrated corporate tax system may upset the
balance of interests traditionally reflected in the
treaty rules of the United States.
Various methods can be devised for extending
integration by treaty to inbound and outbound
investment, and these different methods will
produce differing allocations of the taxes collected
from the corporation between the source country
and the residence country. For example, the
dividend exclusion prototype could be adopted to
permit the source country to retain its corporate
tax revenues: the source country would eliminate
its withholding tax on distributions to treaty
residents and the residence country would credit
the source country taxes against the direct and
ultimate shareholders' tax liabilities in the residence country and collect any residual tax. An
alternative approach would impose a tax on
foreign shareholders at a rate that would approximate the current level of revenues now collected
by the United States on U.S. source corporate
income from foreign investments and allow a
credit against this tax for corporate level taxes
paid. 26

CHAPrER

8: THE TREATMENT OF CAPITAL GAINS IN AN
INTEGRATED TAX SYSTEM

Moving from a classical to an integrated
corporate tax system raises issues relating to the
taxation of capital gains on sales of corporate
stock. While each of the integration prototypes
reduces the biases of the classical system, rules
selected for taxation of capital gains on sales of
corporate stock will affect the degree of neutrality
achieved by each prototype. Taxing shareholder
level capital gains on stock attributable to earnings
that have been taxed at the corporate level is not
appropriate in an integrated system. Taxing such
gains on stock could perpetuate the classical
system's biases against the corporate form and
against investments in equity rather than debt. In
addition, a higher effective tax rate on retained
earnings could provide a tax incentive for corporations to distribute earnings as dividends. On the
other hand, a failure to tax shareholder level stock
gains may result in significant deferral or even
elimination of tax attributable to unrealized
corporate asset appreciation. l

again at the shareholder level. Imposition of a
capital gains tax in this case would be a double
tax on the retained earnings of the corporation.
The second level of tax, however, may prove
temporary. If the corporation subsequently distributes the retained earnings, the value of the stock
may decline to reflect the distribution of corporate
assets. As a consequence, the tax on the selling
shareholder's gain may be effectively reversed by
an offsetting capital loss of the purchasing shareholder. The extent to which the capital loss
reverses the double tax will depend on the timing
of the distribution of the retained earnings and of
the realization and treatment of the capital loss.3
When the tax reduction from the later capital
loss precisely offsets the tax on the earlier capital
gain, the system will collect only one tax on
corporate earnings. However, a subsequent capital
loss deduction allowed to a taxpayer different
from the one who originally is taxed on the
capital gain will often be an imperfect offset. For
example, the tax on the gain may occur in a year
earlier than the tax reduction from the capital
loss. The acceleration of tax may even approximate, in present value terms, double taxation if
there is a substantial period between the payment
of capital gains tax by the fITst shareholder and
the recognition of an offsetting capital loss by a
subsequent shareholder. In addition, limits on the
deductibility of capital losses may prevent the
purchasing shareholder from fully using the
offsetting capital loss. The additional burden
imposed by a capital gains tax also depends on the
marginal tax rates of the purchaser and seller of
stock,4 and the fact that shareholders with different marginal tax rates will generally face identical
market prices for their stock further complicates
analysis of the extent of double taxation.

S.A TAXATION OF CAPITAL
GAINS ATTRIBUTABLE TO
RET AINED TAXABLE
EARNINGS
When a corporation retains earnings, its stock
will generally increase in value. There is some
controversy about the extent to which an incremental dollar of retained earnings translates into
share appreciation. 2 In integration prototypes that
tax earnings at the corporate level, e. g., the
dividend exclusion and CBIT prototypes, dividends would not generally be taxed again at the
investor level. Under these prototypes, to preserve
neutrality in the taxation of corporate capital
income, shareholders' capital gains attributable to
retained earnings that have already been taxed
fully at the corporate level should not be taxed

81

Principal Issues

8.B

SOURCES OF CAPITAL
GAINS OTHER THAN
TAXABLE RETAINED
EARNINGS

Not all capital gains from increases in the
value of corporate equity arise from accumulated
retained earnings. Gains from other sources may
imply different tax consequences than those
applicable solely to gains from fully-taxed
retained earnings.
First, capital gains on corporate stock may be
attributable to retained preference income. In that
case, taxing capital gains on corporate stock does
not impose a second level of tax, because no tax
has been paid at the corporate level. Taxing such
capital gains produces a single tax on those
earnings at the shareholder level. If, as we recommend in Chapter 5, integration should not extend
corporate level preferences to shareholders, such
gains should be taxed. Providing relief for capital
gains attributable to retained preference income
would exacerbate the incentive to retain rather
than distribute preference income or to distribute
preference income in a nondividend distribution in
which capital gain treatment might be available. 5
Second, capital gains may be attributable to
real unrealized appreciation in the value of corporate assets. In that case, the unrealized corporate
level gain, in effect, will be realized ftrst at the
shareholder level upon the disposition of the
stock. The gain also will be realized at the corporate level when the corporation disposes of the
asset. Although such gains eventually will be
taxed at the corporate level, in a realization-based
income tax system, taxing the shareholder level
gain seems appropriate, since that is the ftrst
realization event with respect to the appreciation.
It may, however, be appropriate to prevent double
taxation when the corporation subsequently disposes of the appreciated asset. 6
Third, capital gains may be attributable to
changes in the anticipated value of corporate
earnings, due, for example, to management
changes or revised estimates of proftts from new
products or inventions. Tax considerations for

82

gains attributable to such factors are similar to
those concerning unrealized appreciation in tangible corporate assets. Accordingly, taxing the
appreciation when the shareholder sells the stock
seems appropriate.
Finally, taxable capital gains may result from
inflation. In an unindexed system, capital gains
tax liability can result simply because nominal
asset values rise with inflation, although a taxpayer may have no increase in real income. Taxing
such gains can lead to high effective tax rates on
capital gains. Indeed, granting relief to capital
gains to offset the effects of inflation has been one
of the principal justifications advanced for measures such as lower rates on capital gains or
indexation of such gains. 7

8.C

ADJUSTMENTS TO
ELIMINATE DOUBLE
TAXATION OF RETAINED

CORPORATE EARNINGS

Although avoiding the double taxation of
corporate retained earnings is an important factor
to be taken into account, how capital gains are
treated in an integrated corporate tax system will
tum ultimately on the resolution of basic policy
issues that have long been controversial under the
income tax. Considerations such as the desire to
stimulate investment and entrepreneurship and to
avoid the overtaxation of inflationary gains support preferential rates or exclusions for all or a
part of capital gains income. On the other hand,
some analysts will contend that capital gains and
ordinary income should be taxed similarly.
Integration of the corporate income tax can
proceed and will serve to reduce substantially the
distortions of the current system whichever of
these options for taxing capital gains is chosen.
However, in designing an integrated corporate
tax, one must consider the treatment of capital
gains, as well as dividends, in developing rules
that minimize distortions in corporate and
individual fmancial behavior.
As discussed in Ch~pter 3, the shareholder
allocation prototype would allocate corporate

83

taxable income to shareholders each year and
would provide a system of shareholder level basis
adjustments similar to those used for partnerships
or S corporations under current law. 8 Share basis
would increase to reflect the corporation's taxable
income and certain preference income and would
decrease to reflect distributions. Thus, under such
a system, any capital gains on sale of corporate
stock would be attributable to preference items for
which no basis adjustment is allowed, unrealized
appreciation, or inflation.
On the contrary, the dividend exclusion prototype, set forth in Chapter 2, does not provide any
adjustments to share basis to reflect the corporation's retention of income that has been taxed at
the corporate level. As a consequence, taxing
capital gains could impose an additional shareholder level tax on retained earnings that have
already been taxed in full at the corporate level.
Because retained fully-taxed earnings would face
a greater tax burden than distributed earnings,
corporations would have an incentive to distribute
rather than retain fully-taxed earnings. This
problem can be limited by allowing a dividend
reinvestment plan (DRIP), which would permit a
corporation to declare deemed dividends to the
extent of its EDA balance and treat the amount of
dividend as reinvested in the corporation. Under
such a system, a shareholder would be treated as
receiving an excludable dividend and would
increase stock basis to reflect the deemed recontribution. Chapter 9 discusses DRIPs in detail.
If corporations were to use a DRIP to declare
deemed dividends equal to their fully-taxed
income each year, the resulting basis adjustments
would ensure that such income would not be taxed
again as capital gains. If, however, nontax considerations lead corporations not to elect DRIP treatment for all their fully-taxed earnings, an elective
DRIP would not eliminate the potential additional
tax on retained corporate earnings. For example,
a corporation that expects to earn substantial
preference or foreign source income shielded by
foreign tax credits might want to retain some
EDA balance to enable it to continue to pay
excludable cash dividends in future years. If no

Principal Issues

DRIP is allowed, or if it is expected that corporations will not elect to make deemed distributions
of all fully-taxed income, one could reduce or
eliminate the potential disadvantage for retained
earnings by adopting a preferential rate (or,
equivalently, a partial exclusion) for capital gains.
Taxing capital gains on equity and debt investments in business entities creates special issues
under CBIT. If a compensatory tax is imposed
under CBIT, all business income would be taxed
at the entity level, and investors would exclude
from income all dividends and interest payments
received. In that case, taxing capital gains would
create an even greater disparity between retained
and distributed income than under the dividend
exclusion prototype. Thus, if CBIT includes a
compensatory tax, a complete investor level
exemption for capital gains (and nonrecognition of
losses) on equity and debt would be consistent
with CBIT's general exemption from investor
level tax of dividends and interest. If CBIT does
not include a compensatory tax, but instead taxes
dividends and interest considered to be paid out of
corporate preference income at the investor level
(see Section 4.D), the case for relief for capital
gains is essentially the same as under the dividend
exclusion prototype.
If CBIT includes a compensatory tax,
exempting gains and losses from the sale of equity
interests in CBIT entities could be justified on the
ground that those gains and losses either have
been, or will be, taken into account in calculating
the income tax imposed at the entity level.
Retained taxable income has already been subject
to tax, retained preference income will be subject
to compensatory tax under CBIT when
distributed, and unrealized appreciation represents
anticipated higher future earnings that will be
subject to entity level tax if and when they are
realized. 9 Exempting capital gains on CBIT
equity and debt would promote simplicity in the
CBIT prototype. For example, exempting capital
gains on CBIT debt and equity would remove the
need for a DRIP mechanism to allow holders to
increase basis to reflect earnings taxed at the
corporate level.

Principal Issues

The principal disadvantage of exempting gains
on CBIT equity is the potential for deferral of tax
on appreciation in an entity's assets. A realization-based tax system may allow a significant
delay between the realization of gain by an equity
investor (through the sale of his equity interest)
and the realization of future earnings or built-in
gain at the entity level. Foregoing the opportunity
to tax gains realized upon a sale of an equity
interest thus increases the potential for the deferral of tax on unrealized appreciation at the entity
level. 10 Although additional realization rules at
the entity level could limit deferral, 11 sale of an
equity interest traditionally has been viewed as an
appropriate realization event and the more traditional solution to the problem of double taxation
has been to adjust entity level asset basis to reflect
investor level realization. 12
CBIT also raises issues relating to capital
gains on debt. Some, but not all, changes in the
value of debt reflect gains and losses that have
been or will be taxed at the corporate level. 13
For example, one source of capital gains on debt
is an increase in the creditworthiness of the
issuer, which may reflect an increase in the
corporation's expected future earnings. If an
increase in creditworthiness is due to earnings that
will be taxed at the corporate level, the issues
created by taxing capital gains on debt are similar
to those for equity. 14 Capital gains and losses on
debt (and corresponding losses and gains to
issuers) also may arise from unexpected movements in market interest rates. 15 The movement
to a CBIT system does not demand an exclusion
of gains on CBIT debt that are due to changes in
interest rates, and it is impossible as a practical
matter to distinguish between gains attributable to
interest rate movements and gains attributable to
other sources. 16

8.0 OTHER COUNTRIES
Many countries recognize the possible distortion caused by taxing capital gains on sales of
corporate stock and have taken measures to
mitigate this effect. Table 8.1 shows the tax
treatment of capital gains of the G-7 countries
with integrated tax systems. All the countries

84

provide some preferential treatment for capital
gains on corporate stock through a lower effective
tax rate. For example, Canada, France, and
Germany all provide for an alternative or reduced
tax rate applied to such gains. These reductions
can be substantial. In Germany, for example, all
gain on securities held more than 6 months may
be excluded. The United Kingdom does not
permit a reduction in its marginal tax rate, although the tax base is indexed for inflation, but
instead allows a specific "dollar" exemption.
Gains exceeding the exemption are taxed at the
applicable marginal rate.

8.E

SHARE REPURCHASES

The differences in taxation of gains from
similar transactions complicates analysis of the
proper treatment of capital gains on corporate
stock under integration. The treatment of share
repurchases is one example. A shareholder who
sells stock to a person other than the corporation
that issued the stock or who receives a liquidating
distribution generally can recover the basis in the
stock against the amount realized on the sale. In
contrast, current law may treat a redemption of
stock by the issuing corporation as a dividend or
as a sale of stock. A redemption generally qualifies for sale treatment if it is "not essentially
equivalent" to a dividend or is substantially
disproportionate among shareholders. 17 For
redemptions treated as a dividend, no basis
recovery is permitted (although, generally, the
basis in the redeemed stock is allocated to the
remaining stock and will be recovered eventually).
Current law favors share repurchases because
dividends are taxable to shareholders in full,
while redemptions generally permit recovery of
basis by shareholders and may permit taxation of
gain at the maximum rate of 28 percent for longterm capital gains (rather than at the higher
marginal rates for ordinary income).18
In general, each of the integration prototypes
should greatly reduce current law's incentive to
engage in share repurchases. Shareholder allocation integration, which treats both distributions
and sales of stock as· tax free to the extent of

Principal Issues

85

Table 8.1
Taxation of Individuals on
Long-Term Gains on Securities
Select Foreign Countries

Foreign Country Amount of Gain Exempt
France

All, if the sale proceeds do not
exceed FF307,760 ($55,323?

United Kingdom All inflationary gains plus an
annual exemption of £5,000
($8,885) of non-inflationary gains

Maximum Individual
Tax Rate
(Capital GainS)1
16%
40%

pronounced if a compensatory tax is imposed on
dividends but not on share
repurchases. Avoiding the
compensatory tax would
allow preference income
to be distributed to taxexempt and foreign investors without tax at either
the corporate or the shareholder level.

One way to eliminate
the
remaining
incentive for
Canada
25 % exclusion, plus a lifetime
22%
share repurchases under
exemption of C$l 00,000
($88,480)
the dividend exclusion and
CBIT prototypes would be
All gain on securities held more
Gennany
0%
2
to treat redemptions like
than 6 months
dividends. In that case,
Department of the Treasury
share repurchases, like
Office of Tax Policy
dividends, by a corpora1National tax only. Subnational taxes are relevant in Canada only. tion with sufficient earnProvincial taxes (non-deductible) amount to roughly 50 percent of the ings and profits would not
Federal tax.
pennit basis recovery .
Share repurchases would
2The exemption does not apply in certain cases where the seller held
be
tax-free to shareholders
a "substantial interest" in the corporation whose shares are being sold.
to the extent of the corporation's fully-taxed income
(and would reduce the corporation's EDA). Any
share basis and capital gain thereafter, would treat
portion of payments to repurchase shares that
share repurchases and dividends similarly. 19 The
were made out of preference income would be
dividend exclusion prototype, which treats divitaxable to shareholders, in a dividend exclusion
dends paid out of fully-taxed earnings as tax free
system, or subject to compensatory tax or an
to shareholders, generally would encourage
investor level tax, in CBIT.21 This result may be
corporations to distribute fully-taxed earnings to
inappropriate, however, in a system in which
taxable shareholders as dividends rather than
capital gains are subject to tax, because a sharethrough share repurchases. COlporations that had
holder's basis would be taken into account on a
exhausted their EDA balance and could pay only
sale to a third party, but not in a corporate repurtaxable dividends, however, would have an incenchase. In theory, dividend treatment could be
tive to distribute earnings through share repurextended to all sales of shares, including sales to
chases. Even corporations with sufficient EDA
persons other than the" issuing corporation. Howbalances might desire to make selective share
ever, it may be impractical to extend dividend
repurchases from tax-exempt shareholders to
treatment to third-party sales, given the large
distribute earnings without reducing the corpovolume of daily trading in corporate stock. 22
ration's EDA.20 The incentives for share repurLimiting dividend treatment to redemptions
chases under CBIT are generally the same as
would, however, create disparities between sales
those under the dividend exclusion prototype,
of
stock to the issuing corporation and to third
except that the incentive to make share repurchases out of preference income may be more parties.

Principal Issues

The treatment of capital gains also may affect
the desirability of measures to equalize the treatment of dividends and share repurchases under the
dividend exclusion and CBIT prototypes. A
preferential rate for capital gains, for example,
might reduce, but not eliminate, the disincentive
for share repurchases out of fully-taxed income
while increasing the incentive for share repurchases out of preference income. On balance, we
believe that any of the integration prototypes will
sufficiently decrease incentives for share repurchases as compared to current law that policymakers may avoid adopting any additional rules
and let the passage of time demonstrate whether
the shifting of EDA balances among shareholders
requires additional measures. 23

8.F

CAPITAL LOSSES

In general, the treatment of capital losses on
corporate stock under integration should parallel
the treatment of capital gains. As Section 8.A
discusses, a purchaser's capital loss may serve to
reverse the tax imposed on a seller's capital gain
attributable to retained earnings that have

86

previously been taxed at the corporate level.
However, if relief is provided for capital gains on
corporate stock, the corresponding loss need not
be allowed in full as an offset. For example, an
exemption (or partial exclusion) for capital gains
on corporate stock might imply a disallowance (or
partial disallowance) of capital losses on corporate
stock. Policymakers may, however, decide to tax
capital gains on corporate stock, on the grounds
that the second level of tax on retained earnings
may prove temporary and that preferential treatment could exempt from tax other gains (like
some of those discussed in Section 8.B) that may
appropriately be taxed under integration.
Other capital losses on corporate stock may
arise from unrealized depreciation in corporate
assets, just as capital gains may arise from unrealized appreciation. 24 As Section 8.B notes, in a
realization-based tax system, it seems appropriate
to allow such losses, although it may be appropriate to make adjustments to prevent a second loss
at the corporate level, e.g., by adjusting corporate
asset basis. As under current law, the desirability
of such measures must be weighted against their
complexity.25

CHAPTER

9:

DIVIDEND REINVESTMENT PLANS

Under the dividend exclusion and CBIT
prototypes, corporations (and other entities subject
to CBIT) may desire to retain earnings but allow
their shareholders to increase share basis to reflect
earnings which have been taxed at the corporate
level. Allowing basis adjustments would reduce
the extent to which taxes on investor capital gains
would be a second tax on retained earnings and
would reduce the tax incentive for corporations
(and other CBIT entities) to distribute fully-taxed
income. See Chapter 8. We contemplate that this
would be permitted through an elective dividend
reinvestment plan (DRIP).1 DRIPs may be adopted by corporations under current law; such plans
commonly are used by mutual funds and utilities.
Because dividends are taxable to shareholders
under current law, participation in DRIPs generally requires an election by the shareholder. Unlike
existing DRIP arrangements, however, deemed
dividends reinvested under an integration prototype would not be taxable to shareholders and the
DRIP could be adopted by the corporation (or
CBIT entity) without the consent of the individual
shareholder. 2 Adopting a DRIP would simply
represent a corporate decision to reduce the
corporate EDA in order to increase share basis.

choose the amount of deemed dividends and the
classes of stock on which they would be paid. The
corporation's ability to stream deemed dividends
to taxable shareholders would be constrained by
the anti-streaming rules generally applicable under
the prototypes for payments of excludable dividends. 3 The corporation would allocate the
deemed dividends to holders of stock on the
chosen record date and would provide information
reports to those shareholders showing the amount
of the deemed dividend and the associated basis
increase.
Dividends are generally paid on a per share
basis, and the share basis increase under the DRIP
also would be on a per share basis. It would be
desirable to have a uniform convention governing
the allocation of such basis, e.g., equally to each
share or in proportion to the existing basis.
Example 1. Corporation X adopts a DRIP and
makes a deemed distribution of $100 to Shareholder A. The fair market value of X shares on the
date of the deemed distribution is $20 per share. A
owns 10 shares of X which he purchased in two
lots, Lot A (5 shares at $4 each) and Lot B (5
shares at $6 each). If basis is allocated on a per
share basis, the basis of each Lot A share will be
$14 and each Lot B share will be $16.

9.A MECHANICS

Although a shareholder may have purchased
various shares of a corporation's stock for different amounts, the treatment of each share under
current law as having a separate basis may be
questioned. If the shares are economically
equivalent, it may be appropriate to require the
shareholder to recognize the same gain or loss
regardless of which shares are actually sold. For
example, a DRIP could be used to reduce basis
disparities.

By adopting a DRIP, a corporation would
elect to treat shareholders as receiving excludable
dividends in an aggregate amount not to exceed
the balance in the corporation's EDA. The
amount deemed distributed would be deducted
from the EDA. The shareholders would then be
deemed to recontribute the distributed amount,
and their share basis would increase by the
amount of the deemed distribution. Share basis
would increase only by the amount deemed
reinvested (rather than by the corporation's pretax earnings), because that would be the result
had the shareholder actually reinvested a dividend.

Example 2. The facts are the same as in Example
1, except that the fair market value of X shares on
the date of the deemed distribution is $15 per
share. The DRIP basis increase could be allocated
between the Lot A and Lot B shares so that the
shares in each lot have a basis of $15.

Mechanically, the electing corporation would
declare deemed dividends in the same manner that
it declares actual dividends. A corporation would

For some shareholders (particularly those with
recently purchased shares), a DRIP may create
87

Principal Issues

88

share basis in excess of fair market value, with
the result that capital losses will be realized when
the shares are sold. Such losses may serve the
same function as those discussed in Section 8.A,
simply "reversing" the double tax imposed on the
seller of shares. In other cases, however, it may
be appropriate to craft anti-abuse rules to prevent
a DRIP from being used to create basis in excess
of fair market value. 4

dividend account for that class of stock. The
deemed dividend account would be reduced by the
amount of dividends treated as a return of capital
under this rule. Distributions in excess of the
deemed dividend account for a class of stock
would be governed by the prototype's rules
applicable to distributions in excess of the EDA. 6

9.B
The dividend exclusion and CBIT prototypes
generally adopt stacking rules that treat distributions as made flrst from fully-taxed income. If a
DRIP is adopted, further stacking rules would be
necessary to determine whether cash distributions
on a class of stock following deemed dividends on
that class of stock are fITst a recovery of basis
from the DRIP or out of other earnings. Thus,
issuers would keep an account of deemed dividends made on each class of stock (the deemed
dividend account), in addition to the EDA.5 To
simplify the operation of these accounts and
minimize the double taxation of retained earnings,
we recommend that all cash distributions, including cash distributions on shares on which deemed
dividends have previously been paid, be treated
fITst as payments out of any remaining balance in
the corporation's EDA. Then cash distributions on
a class of stock on which deemed dividends had
been paid would be treated as a return of capital
to the extent of the balance in the deemed

DESIGN CONSIDERATIONS

We anticipate that deemed distributions will,
in practice, be made only to holders of common
(or at least participating) equity, because holders
of preferred stock typically require cash dividends. Restrictions limiting DRIP distributions to
common and participating equity could be considered if it were feared that DRIPs could permit
inappropriate losses, e.g., distributions on preferred stock bearing limited dividends and a fIxed
liquidation or redemption value might create such
a result. 7
In addition, DRIPs could be made mandatory
on the theory that double taxation of retained
earnings through capital gains taxation could be
minimized by forcing basis allocations as prompt1y as possible. 8 However, there seems to be little
reason why corporations should not be permitted
to control this, as other aspects, of their
distribution policy.

88

CHAPTER

10:

TRANSITION CONSIDERATIONS

10.A INTRODUCTION

realization than under eXIstrng law. (Absent
specific transitional rules for built-in gains and
losses, the second effect will likely become a part
of the fIrst effect.) Finally, some corporations
may have retained earnings which have been
realized and taxed while others may have distributed such earnings. The former may gain advantage if the retained earnings are not taxed on
distribution. 3

Under current law, investors and corporations
generally have made decisions and commitments
based on the two-tier corporate tax system.
Investors' decisions to invest in corporate or
noncorporate entities or in debt rather than stock,
and corporations' decisions to distribute earnings,
to issue debt or equity, or to recognize gains
inherent in appreciated assets all likely have been
made with an expectation that corporate equity
income will likely continue to be subject to tax at
two levels. Introduction of an integrated system
will alter these expectations. We believe that a
transition period is appropriate to prevent undue
dislocation and to mitigate transitional gains and
losses.

While we favor a phase-in of integration
primarily to allow for gradual portfolio shifting
and to allow assessment of integration's impact as
it is implemented, we do not favor other explicit
transitional rules to deal with transition gain and
loss. Phase-in itself will mitigate the impact of
any change in share values. 4
Built-in gains and losses are likely to be
reflected in share value; in any event, the differing tax consequences that will occur arise primarily by virtue of the realization concept fundamental
to current income tax law. Prior law changes
(including significant rate changes) generally have
not attempted to capture this form of transition
gain (other than through phase-in) and we believe
that result is appropriate in the shift to integration
as well.

We anticipate that shifts in investors' portfolios will occur under any integration proposal and,
in some cases, such shifts may be substantial.
While the magnitude of such shifts will vary with
the degree of difference between the integration
proposal and current law, prudence suggests that
phased-in implementation will permit adjustment
to the new system while mitigating transition
gains and losses. It also will provide an opportunity for midcourse corrections, if needed. A
phase-in appears to be the simplest form of
transition for both taxpayers and administrators to
implement. It will not require complicated rules
of uncertain duration for preenactment assets.

Differences in earnings distribution policies
are likely to be significant only in certain forms
of integration. They could be significant, for
example, in the shareholder allocation prototype.
Because that prototype taxes only current COl-pOrate income and treats distributions as a return of
capital, corporations that retained earnings realized under current law could be signifIcantly
favored over those that distributed such earnings.
In contrast, the dividend exclusion and CBIT
prototypes' EDA mechanisms will cause distributions from earnings retained before the establishment of the EDA to be taxable to the shareholder
when distributed. s Accordingly, both the dividend
exclusion prototype and CBIT will produce results
for pre-integration retained earnings similar to
current law. 6

10.B TAXATION OF
TRANSITIONAL GAINS
AND LOSSES
Some believe that it is important for transitional rules to deal explicitly with gains and losses
arising from the shift to an integrated system. I
Several sources of such transition gains and losses
can be identified. First, the shift to integration
may affect the value of corporate shares. 2 Second, at the time of the shift, corporations may
hold assets with unrealized built-in gains or losses
and hence face different tax consequences upon
89

Principal Issues

As an alternative, some fonn of grandfathering of existing assets or activities could be
used to limit or eliminate transition gains and
losses from the shift to integration. Under such an
approach, current law treatment would be retained
for assets that otherwise would be treated more
favorably under integration to preserve asset
values that reflect the classical corporate tax
system. In moving to integration, however, a
penn anent grandfather rule would require maintaining a distinction between pre-enactment and
post-enactment assets and equity interests and, in
CBIT, old and new debt as well. Making such
distinctions over an extended period would create
difficult, if not impossible, reporting burdens and
administrative complexity and would inevitably
result in uneven enforcement. 7 Such an approach
also could require an extensive array of rules to
prevent transfonnation of old equity into new
equity and to govern conversions of non-corporate
entities to corporate status. 8 More importantly,
preserving a dual system to limit the benefits of
integration to new equity, would thwart the goal
of economic refonn by perpetuating the very
distortions the new system seeks to eliminate. 9
We have rejected such an approach on grounds of
both efficiency and simplicity.

90

changes have typically been ignored in connection
with rate changes that raise similar concerns. A
phase-in also would mitigate the revenue effects
relative to immediate change. A phase-in would
delay application of the new rules, however, and
the delay would reduce the present value of the
desired economic changes.
Under a phase-in approach, integration would
be introduced gradually over a designated period.
This approach would reduce the magnitude of
transition gains and losses. A phase-in would not
distinguish between old and new equity or, in the
CBIT prototype, old and new debt. Although
there would be some delay in full implementation
of integration under a phase-in approach, this
delay would be of limited duration, in contrast to
the virtually indefmite delay that would result
from limiting integration to new equity. The
length of the phase-in period should depend on a
variety of factors, including the particular integration prototype adopted. An appropriate period
should be selected by striking a balance between
the need to mitigate the disruption to the status
quo and the desire to achieve as expeditiously as
possible the full value of the anticipated gains of
the new system, taking into account administrative
costs.

lO.C PHASE-IN OF INTEGRATION
Phase-ins have been used in recent legislation
to moderate the harsh effects of significant
changes in the tax law. For example, the passive
loss disallowance rules, the personal interest
disallowance rules, and the new investment
interest limitations adopted in the Tax Refonn Act
of 1986 all were phased in.10
We generally recommend that a phase-in
approach be used to implement the transition from
the classical system to an integrated corporate tax.
A phase-in approach would moderate the transition effects of integration, while avoiding the
serious drawbacks of limiting integration to new
equity. While some transition gains and losses
may occur, fundamental structural changes in the
tax law, such as those proposed here, simply are
not feasible if substantial changes in values of
taxpayers' assets must be avoided. Indeed, such

The dividend exclusion prototype could readily
be phased in. The EDA would automatically limit
the amounts of dividends excludable by shareholders to the amount of earnings taxed after enactment, although stacking distributions first against
the EDA would tend to accelerate the benefits of
integration. See Section 2.B. Additional rules
distinguishing pre-enactment from post-enactment
earnings would not be necessary. Because the
dividend exclusion prototype requires relatively
few changes to current law, the appropriate phasein period for that prototype might be relatively
short, e.g., 3 to 5 years. Mechanically, a phase-in
approach would allow a corporation to pay
excludable dividends .to the extent of its EDA
balance but would limit additions to the EDA to
reflect the phase-in, e.g., amounts based on 25
percent of corporate taxes paid in the first year
after enactment, 50 percent in the second year,
and so on. 11

91

In contrast, a phase-in of the shareholder
allocation prototype appears complex. Attributing
a portion of cotporate tax to shareholders in a
manner that would increase the portion of cotporate income so taxed over time, would require a
complex system for tracking cotporate income and
making share basis adjustments, for example, to
determine how subsequent distributions of phasein years' earnings would be taxed. On balance, if
a shareholder allocation system were desired, it
might be preferable to enact the system in its
entirety with a delayed effective date. A delayed
effective date would have effects similar to a
phased-in effective date in reducing transition
gains and losses, would allow taxpayers an opportunity to plan for the shift, while avoiding the
complexity of a phase-in of the shareholder
allocation prototype. 12

Principal Issues

Although eliminating the interest deduction
ultimately could make certain limitations on
interest deductibility applicable to CBIT entities
unnecessary,14 they would remain important
during the phase-in period. Indeed, a phase-in of
CBIT may require some strengthening of rules to
prevent acceleration of interest deductions to
earlier years of the phase-in, as well as deferral of
interest income into later years of the phase-in.
Transition rules also would have to address the
timing mismatches that arise where interest has
been deducted by the payor but not yet included
in income by the lender or where interest has
been included by the lender but not yet deducted
by the payor. Alternatively, transition to CBIT
could be accomplished by beginning with implementation of the dividend exclusion prototype.

IO.D MECHANICS OF A PHASE-IN
The CBIT prototype generally eliminates the
investor level tax on dividends and interest and
disallows the interest deduction to cotporations
and other CBIT businesses. In addition to the
transition gains and losses that might occur under
the other integration prototypes, under CBIT
lenders to CBIT entities might enjoy an increase
in the value of existing debt with the elimination
of tax on interest received. The magnitude of the
increase would depend on a variety of factors,
including the remaining term of the debt. From
the borrower's perspective, the disallowance of
interest deductions would effectively increase the
cost of borrowing for cotporations unable to call
their bonds or otherwise refmance their debt. 13
CBIT, therefore, should probably be phased in
over a longer period than would be appropriate
for the dividend exclusion prototype. Longer
phase-ins have greater effect in reducing transition
gains and losses. Because, as detailed in Chapter 4, a CBIT regime will continue to have certain
types of includable interest (such as interest on
Treasury securities) even when fully phased in,
proportionate adjustments during the phase-in
period would add complexity but should not
create insurmountable recordkeeping problems for
investors.

Dividend Exclusion Prototype. A dividend
exclusion could be phased in over 4 years, for
example, by crediting the EDA with an increasing
percentage of the fully phased-in EDA amount in
each transition year, i.e., 25 percent of the
formula amount in the fIrst year, 50 percent in the
second, 75 percent in the third. Offsetting revenues could be phased in on the same schedule. By
limiting additions to the EDA at the corporate
level, shareholder level phase-in will not be
required. However, only 25 percent of income
taxed at the corporate level in the fIrst year could
be distributed tax-free to shareholders. Distributions in excess of this amount, like other distributions in excess of the EDA, would be taxable to
the shareholder.
CBIT. CBIT is self-fmancing through the
disallowance of the entity level interest deduction.
Accordingly, the CBIT phase-in must coordinate
the dividend and interest exclusions for shareholders with entity level interest disallowance. For
each year of the CBIT phase-in, the EDA would
be credited with an increasing percentage of the
fully phased-in EDA amount and the same percentage of corporate interest deductions would be
disallowed, i.e., 10 percent in the fIrst year,

Principal Issues

92

20 percent in the second, etc .. In addition, it
would be necessary to credit the EDA with an
additional amount equal to the phase-in percentage
for the year multiplied by the sum of the allowable interest deduction for the year plus interest
paid during the year but deducted in a year before
phase-in begins. 15 Absent this adjustment, the
CBIT compensatory tax or investor level tax on
distributions in excess of the EDA would treat
allowable interest like a preference and the income it offsets would be taxed when distributed.
Unlike the dividend exclusion prototype, CBIT
requires investor level phase-in to mitigate and
smooth portfolio shifts during the phase-in period.
Thus, debtholders would exclude 10 percent of
interest received from a CBIT entity in the first
year while shareholders would exclude 10 percent
of dividends received.
Example 1. A CBIT entity earns $109 of gross
income and has $10 of interest expense in the first
year of a 10 year phase-in of CBIT. If the CBIT
phase-in percentage were 10 percent, the CBIT
entity would deduct $9 of interest ($10 minus (10
percent of $10». It would thus have taxable income
of $100 and pay CBlT of $31.
The amount added to the entity's EOA is $7.80,
computed as follows: 16
$6.90 (10% of ($311.31-$31»
+ .90 (10% of $9 interest allowed as
deduction)
$7.80

a

Oebtholders would be entitled to exclude $1.00 of
the $10.00 in interest they receive, thereby reducing the EOA to $6.80. 17 If the entity distributed
its remaining after-tax earnings of $68 ($109 minus
$10 interest minus $31 tax) to shareholders, shareholders could exclude $6.80 from income, thereby
reducing the EOA to zero.

Example 2. The facts are the same as in Example
1 except that the entity made no distribution to
shareholders in the first year and it has identical
income and interest in the second year. Thus, it has
$109 of gross income and is allowed an $8 interest
deduction, resulting in $101 of taxable income.

The entity's EOA is computed as follows:
$ 6.80 (balance of EOA from year 1)
13.94 (20% of ($31.311.31-$31.31»
1.60 (20% of $8 interest allowed)
$22.34
Oebtholders in this year would be entitled to
exclude $2.00 of the $10.00 in interest they receive, reducing the EDA to $20.34. If the entity
distributed its $68 in. after-tax earnings from year
1 plus its $67.69 in after-tax earnings from year 2
($109 minus $10 interest minus $31.31 tax),
shareholders would be entitled to exclude 20
percent of the $135.69 dividend or $27.14. This
amount exceeds the EOA balance of $20.34 because only 10 percent of the earnings from year
one are reflected in the EDA. To compensate for
the 20 percent exclusion at the shareholder level, a
31 percent compensatory tax of $2.11 is imposed
on the $6.80 differential. (Thus, the differential
amount is treated like retained earnings from preCBIT years.)
Example 3. The facts are the same as in Example
1, except that the entity earns $20 in preference
income in addition to the $109 in gross income.
Thus, its after-tax earnings available for distribution to shareholders in year 1 would be $88 ($68+
$20). If it distributed the entire $88 in year 1,
shareholders could exclude 10 percent of that
amount, or $8.80. As a result, a 31 percent compensatory tax of $.62 is imposed on the $2.00 by
which the shareholder exclusion exceeded the EDA
balance ($8.80-6.80). This amount also is 10
percent of the entity's preference income.

As the foregoing examples indicate, a unifonn
investor level phase-in of CBIT could be more
easily accomplished if the prototype includes a
compensatory tax. If CBIT does not include a
compensatory tax, and instead investors are
subject to tax on preference and sheltered foreign
source income, a phase-in might be accomplished
by limiting the portion of dividends and interest
that are excludable to the lesser of (1) the phasein percentage multiplied by the amount of the
payment and (2) the EDA balance. As a consequence, all payments would be excludable up to
the phase-in percentage to the extent of the EDA,
and all payments thereafter would be taxable.

PART IV: THE ROADS NOT TAKEN
INTRODUCTION
Under an imputation credit system, a shareholder would be taxed on the gross amount of a
dividend, including both the cash dividend and the
associated tax paid at the corporate level. The
shareholder would receive a credit equal to the
amount of corporate tax associated with the gross
dividend. From an individual shareholder's viewpoint, this system would mean that the corporate
tax on earnings distributed as dividends would
generally resemble the current withholding tax on
wages and salaries. An employee includes gross
wages in his taxable income and receives a credit
against tax liability equal to the amount of tax
withheld by the employer. Because of the prevalence of imputation credit systems abroad, such a
system would facilitate international coordination
of corporate tax regimes, especially in the context
of bilateral treaty negotiations. 1 We therefore had
expected to recommend an imputation credit
system as our preferred form of distributionrelated integration.

most important issues of integration. For example,
an imputation credit can extend the benefits of
integration to tax-exempt and foreign shareholders
by allowing refundability of imputation credits or
it can deny such benefits by denying refunds. Its
major drawback is its complexity in creating an
entirely new regime for taxing corporate
dividends. On balance, we concluded that the
dividend exclusion prototype set forth in Chapter
2 was the preferable distribution-related integration alternative because it would implement our
policy recommendations, including such issues as
the treatment of preferences and tax-exempt and
foreign shareholders, in a substantially simpler
manner.
An imputation credit system may not be the
most straightforward distribution-related integration alternative even if policymakers were to
choose policy goals different from ours. A dividend deduction system, described in Chapter 12,
also would be simpler than an imputation credit
system if policymakers chose to extend the benefits of integration to tax-exempt and foreign
shareholders. 2

After a close examination of the imputation
credit system, reflected in Chapter 11, we determined that its principal advantage is its flexibility
to respond to different policy judgments on the

93

CHAPTER 11: IMPUTATION CREDIT SYSTEM
l1.A OVERVIEW OF IMPUTATION
CREDIT PROTOTYPE

eliminate the corporate level tax. However, it
would generally permit shareholders to pay no
additional tax on distributions of corporate earnings that have already been taxed fully at the
corporate level, while ensuring that shareholders
taxable at the maximum individual rate do not use
excess credits to shelter other income from tax or
to claim refunds.! Section II.B explains how
taxes paid at the corporate rate are converted into
imputation credits at the shareholder rate.

In producing this Report, we looked carefully
at the integration systems of other countries. See
Appendix B. The imputation credit prototype set
forth in this chapter is the one we consider to be
most consistent with our policy recommendations.
It closely resembles the system that New Zealand
adopted in 1988.
Mechanics. Corporations would continue to
determine income under current law rule and pay
tax at a 34 percent rate. Shareholders receiving a
distribution treated as a dividend would include
the grossed-up amount of the dividend in
income-including both the amount of cash
distributed and the imputation credit allocated to
the dividend-and could use the credit to offset
their tax liability. The credit would be nonrefundable; it could reduce tax liability to zero,
but would not produce a refund. Credits would be
allowed only for taxes paid after the effective date
of the proposal.

A corporation would maintain an account of
its cumulative Federal income taxes paid, computed as though its taxable income had been subject
to tax at a rate of 31 percent (the shareholder
credit account or SCA). A corporation could elect
to attach a credit to a dividend (frank the dividend) in any amount, provided it does not exceed
the lesser of (1) the adjusted corporate level tax
(computed at the 31 percent rate) on the pre-tax
earnings that generated the dividend (the grossedup dividend),2 or (2) the balance in the SCA.3
The corporation would reduce its SCA balance by
the amount of credits used to frank dividends and
by refunds of corporate tax. It would increase its
SCA by payments of corporate tax and by credits
attached to dividends received from other
corporations.

Allowing a credit for the full amount of
cOlporate tax paid with respect to distributed
earnings would eliminate the corporate level tax
if the shareholder's tax rate at least equals the
corporate rate. Even if the shareholder rate were
less than the corporate rate, the corporate tax
could be eliminated if the credit were allowed
against tax on other income or as a refund. Currently, the maximum statutory rate for individual
shareholders (31 percent) is less than the corporate rate of 34 percent. Thus, if the credit were
computed at the full corporate rate, most shareholders could shelter other income from tax or
claim refunds. This need not be permitted, however, if the goal of the imputation credit prototype
is simply to ensure that distributed earnings that
are taxed at the corporate level are not taxed
again to shareholders. Accordingly, rather than
allowing a credit for the full amount of corporate
tax paid on a distribution, the prototype computes
the amount of the credit at the 31 percent maximum shareholder rate. This approach does not

Tax-Exempt Shareholders. The prototype
would effectively retain the current level of
taxation of income earned on corporate equity
supplied by tax-exempt shareholders. The credit
would be nonrefundable, and fully-taxed income
distributed to tax-exempt shareholders would
continue to bear one level of tax: the corporate
tax. Preference income distributed to tax-exempt
shareholders generally would continue to be
untaxed both at the corporate and shareholder
level.
Corporate Shareholders. The dividends received deduction would be increased to 100
percent for all intercorporate dividends, and any
imputation credits attached to a dividend would be
added to the recipient corporation's SCA.

95

The Roads Not Taken

Tax Preferences and Foreign Source Income.
By adding only U. S. taxes to the SCA and requiring that imputation credits be paid out of the
SCA, the prototype ensures that the credit is
allowed only to the extent of U.S. corporate tax
payments. By generally allowing corporations to
decide how much credit to attach to a particular
distribution, the prototype allows a corporation to
treat distributions as coming frrst from fully-taxed
income and then from preference income and
foreign source income shielded from U.S. tax by
foreign tax credits. The prototype does not impose
a compensatory tax on distributions out of preference or shielded foreign source income. Therefore, the prototype permits a corporation to make
distributions out of preference or shielded foreign
source income without incurring additional corporate level tax liability. However, shareholders
may not claim credits with respect to such distributions. This results in distributed preference
income and shielded foreign source income
continuing to be subject to the same level of
taxation as under present law.
Foreign Shareholders. The prototype also
retains the current law treatment of foreign shareholders. The credit would be nonrefundable to
foreign shareholders, absent treaty provisions to
the contrary, and dividends would be subject to
U. S. withholding tax to the same extent as under
current law.
Anti-abuse Rules. The imputation credit
prototype generally permits a corporation to frank
dividends in any amount (subject to a maximum),
even if they have a remaining SeA balance. This
treatment is more liberal than the dividend exclusion prototype, which requires corporations to pay
fully excludable dividends (equivalent to fully
franked dividends) until their EDA is exhausted.
Permitting this additional flexibility in the imputation credit prototype may require additional antiabuse rules to prevent corporations from attaching
credits to distributions to taxable shareholders and

96

not attaching credits to distributions to
shareholders with low or zero U.S. tax liability,
such as tax-exempt and foreign shareholders. See
Section 11.F.4
Capital Gains and Share Repurchases. Chapter 8 discusses the treatment of capital gains on
sales of corporate stock and the treatment of share
repurchases.
Structural Issues. The prototype generally
maintains current law rules for corporate acquisitions, although new rules would be needed to
govern the carryover or separation of corporations' SCA balanceasin acquisitive and divisive
reorganizations.
Impact on tax distortions. Table 11.1 illustrates the impact of the imputation credit prototype on the three distortions integration seeks to
address: the current law biases in favor of corporate debt over equity fmance, corporate retentions
over distributions, and the noncorporate over the
corporate form. The only difference between the
current law treatment of nonpreference, U.S.
source business income and its treatment under
the imputation credit prototype is on corporate
equity income distributed to individual investors.
The prototype would reduce the tax rate on such
income to tc (when tj =tjm) or a lower rate (when
~ < 1m) , but as long as tc > tjm, the rate will be
greater than ~. Thus, while the rate on corporate
equity income distributed to individuals would be
reduced, it would still be higher than the rate (10
imposed on noncorporate equity income and on
interest. It would be lower, however, than the rate
on undistributed corporate equity income. Some
bias toward debt fmance and the noncorporate
form would remain, while the bias toward corporate retentions woul~ tend to be reversed, in the
absence of a DRIP. See Chapter 9 and Section
11.1. For tax-exempt and foreign investors, there
would be no change in the tax treatment of nonpreference, U.S. source income.

97

I1.B CHOICE BETWEEN A
CREDIT LIMITATION
SYSTEM AND A
COMPENSATORY
TAX SYSTEM

The Roads Not Taken

Table 11.1
Total U.S. Tax Rate on a Dollar of NonPreference,
U.S. Source Income from a U.S. Business
Under Current Law and an
Imputation Credit Prototype
Imputation
Credit
Prototype

Introduction
Type of Income

As set forth in Chapter 5, this Report
recommends that integration not become
an occasion for extending the benefit of
corporate tax preferences to shareholders. In implementing this decision in an
imputation credit system, the most
significant choice is between a shareholder credit limitation system (in which
tax is collected only at the shareholder
level on distributed preference income)
or a compensatory tax system (in which
a tax, creditable by shareholders, is
collected at the corporate level on
distributed preference income). The
choice between a credit limitation system
and a compensatory tax system also is
influenced by the policy recommendations set forth in Chapters 6 and 7 not to
eliminate the corporate level tax on
earnings distributed to tax-exempt and
foreign shareholders and not to treat
identically U. S. corporate level taxes
paid and foreign taxes on corporations'
foreign source income. These policy
recommendations imply that imputation
credits should not be refundable to taxexempt or foreign shareholders and that
foreign corporate level taxes should not
be creditable by shareholders.

Current Law

I. Individual Investor is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

tc + (l-tJt;
tc + (1 - tc)tg
t;
t;
t;

[(l-t;)tc+ti_tim]/(l_t;m)
tc + (1 - tc)tg
t;
t;
t;

II. Tax Exempt Entity is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

tc
0
0

ill. Foreign Investor is Income Recipient
Corporate Equity:
Distributed
tc + (l-tJtWD
Undistributed
tc
Noncorporate Equity
tWN
Interest
t~
Rents and Royalties
tWR
Department of the Treasury
Office of Tax Policy

tc
0
0

tc + (1 -tJt wn

tc = u.s. corporate income tax rate.
t; = U.s. individual income tax rate.
t;m = Maximum U.S. individual income tax rate.
tg = U.S. effective individual tax rate on capital gains.
two, tWN , t~, tWR = U.S. withholding rates on payments to
foreigners of dividends, noncorporate equity income, business
interest, and rents and royalties, respectively. Generally varies
by recipient, type of income, and eligibility for treaty
benefits, and may be zero.

The choice between a credit limitation system
and a compensatory tax system may differ depending upon the kind of integration mechanism
adopted. For example, in the dividend exclusion
prototype, we chose to follow a credit limitationtype approach and to tax distributed preference
income only at the shareholder level. This allows
adoption of the dividend exclusion prototype with
minimal changes from current law and would
continue current law treatment of dividends paid
out of preference or foreign source income. In

addition, because the dividend exclusion prototype
applies only to corporate equity, a compensatory
tax would tend to increase the incentive for
corporations with preference income to issue debt
rather than equity to tax-exempt and foreign
investors. For similar reasons, we adopt a credit
limitation approach in the imputation credit
prototype.
Experience in other countries makes clear that
an imputation credit system can accommodate

The Roads Not Taken

either a credit limitation or a compensatory tax,
however. Australia and New Zealand, for example, adopted credit limitation systems, while
France, Gennany, and the United Kingdom
adopted compensatory tax systems. 5

Comparison of a Compensatory Tax
and Credit Limitation
Under current law, preference income distributed to tax-exempt shareholders is not subject to
tax at either the corporate or the shareholder
level. If a compensatory tax were imposed on
preference income at the corporate level and not
made refundable to tax-exempt shareholders, a
compensatory tax would impose an additional tax
on such income. 6 Similarly, under current law,
preference income distributed to foreign shareholders is subject only to the 30 percent withholding tax (often reduced to as little as 5 percent by
treaty). If distributed preference income were
subject to a compensatory tax at the corporate
level and the imputation credits could not be used
against the foreign shareholders' withholding tax,
the net tax burden on that income would increase.
A similar problem arises with distributions of
foreign source income earned by a U.S. corporation and taxed abroad. As discussed in Chapter 7,
this Report recommends that foreign taxes remain
creditable at the corporate level, but that foreign
taxes not be treated the same as U. S. taxes paid in
determining imputation credits. Under such a
rule, distribution of foreign source income that
has not borne any residual u.s. tax would be
fully taxable at the shareholder level, as under
current law. A nonrefundable compensatory tax
on distribution of foreign source income shielded
from U. S. corporate tax by foreign tax credits
would increase the tax burden on distributions of
such income to foreign and tax -exempt shareholders relative to the burden on such income under
current law.
Because of the additional corporate level tax
imposed by a nonrefundable compensatory tax on
preference and foreign source income distributed
to tax -exempt or foreign shareholders, the compensatory tax and credit limitation systems have

98

very different implications for corporations that
currently pay little U.S. tax, due either to substantial use of tax preferences or to foreign tax
credits. Under current law these corporations
incur little or no United States corporate level tax,
but the dividends paid do bear a shareholder level
tax (except ill the case of tax -exempt
shareho lders) .
A credit limitation system allows corporations
to continue to pay dividends out of preference or
foreign source income without incurring any
additional corporate level tax. In contrast, a
compensatory tax system would require such
corporations to pay an extra corporate level tax in
order to maintain their current level of dividend
payments. In practical terms, a compensatory tax
may create an extra tax cost for corporations
engaged in tax-favored activities, such as research
and experimentation and oil and gas exploration7
and may affect large multinational corporations
doing business in high-tax foreign jurisdictions,
such as certain European countries. In addition,
U.K. experience with a nonrefundable compensatory tax suggests that corporations that would be
subject to such taxes will engage in tax planning
behavior to avoid its burdens. Nevertheless, a
compensatory tax does promote simpler administration, since it collects tax on distributed corporate preference or foreign source income at the
corporate level. 8
The extent to which additional tax burdens
would be created by a compensatory tax system
depends on the method for determining when a
distribution is made out of income that has not
borne U.S. tax. 9 A stacking rule that treats all
distributions as having borne tax at the full corporate rate (to the extent possible based on total
corporate tax paid) may mitigate the imposition of
a compensatory tax. If distributions do not exceed
fully-taxed income, no compensatory tax is due.
Choice of a particular stacking rule also affects
both the revenue effects of distribution-related
integration and corporate incentives to pay dividends. In this and other prototypes, we have
consistently rejected a stacking rule that would
treat dividends as made fIrst from preference
income, and we have been unable to discover any

The Roads Not Taken

99

country that stacks preferences fIrst in its distribution-related integration system. Although that rule
would reduce the revenue loss from adoption of
distribution-related integration, it also would
discourage payment of dividends. 1o Most foreign
systems stack preferences last. See Appendix B.
A credit limitation system may be somewhat
more complex to administer than a compensatory
tax system, because it requires shareholders to
apply a different rate of gross-up and credit for
each distribution from each corporation. In contrast, under a compensatory tax, all distributions
from all corporations are subject to gross up and
credit at the same rate. From the shareholder's
point of view, however, a credit limitation system
may not be significantly more complicated. Under
either system, the shareholder must compute tax
using two pieces of information-the amount of
the cash dividend and the associated credit (also
used to compute the grossed-up dividend). The
only necessary difference between the two systems is that under a compensatory tax system the
credit rate can be provided by instructions to the
tax form, while under a credit limitation system it
would have to be provided by information returns,
which may reflect differing amounts of credit for
different corporations and in different years.
Both compensatory tax systems and credit
limitation systems have posed problems for
countries that have adopted them. For example,
the United Kingdom imposes a compensatory tax
by collecting Advance Corporation Tax (ACT) on
all distributed earnings at the time of distribution.
ACT is then creditable against regular taxY The
United Kingdom has found that many corporations
with a large amount of preference or foreign
source income have built up substantial excess
ACT accounts rather than reduce their dividend
payments. The likelihood of excess ACT accounts
has led to tax planning efforts to avoid imposition
of compensatory taxes and the existence of excess
ACT accounts promotes efforts at traffIcking in
tax attributes. However, credit limitation systems
have had problems in creating and enforcing
effective anti streaming rules. Both the Australian
and New Zealand systems contain an extensive
network of such rules.

On balance, we believe that a credit limitation
system is preferable to a compensatory tax in both
the imputation credit prototype and the dividend
exclusion prototype. In both cases, a credit limitation system would permit corporations to maintain
their current dividend policy without the imposition of additional corporate level tax.

Mechanics of a Shareholder Credit
Limitation System
Under the imputation credit prototype, corporations would keep track of cumulative taxes paid
by maintaining a Shareholder Credit Account
(SCA)-an account of cumulative creditable taxes
paid. A corporation would be allowed to attach a
credit to a dividend (frank the dividend) in any
amount, up to a limit. The credit attached could
not exceed the lesser of (l) an amount equal to
the product of (a) the distribution and (b) the ratio
of the current maximum shareholder tax rate to 1
minus the current maximum shareholder tax rate,
or (2) the balance in the SCA. The corporation
would reduce the balance in the SCA by the
amount of credits used to frank dividends and
refunds of corporate tax and increase the SCA by
payments of corporate tax (including estimated
tax) and imputation credits attached to dividends
received.
For example, consider a corporation with
taxable income of $100. Assuming a 34 percent
corporate tax rate and a 31 percent shareholder
rate, it would pay a tax of $34 and have $66
available for distribution. The corporation would
add $29.65 to its SCA account. The amount
added to the SCA is determined using the
following formula:
Annual additions ID

seA

=

-!,. _I) [u.s. tax paid for taxable year _ u.S.
[ .0'
.34
+

tax paid for taxable year]

imputation credits on dividends received

This is the amount of tax that would fully frank,
at the 31 percent shareholder rate, the
corporation's actual after-tax income of $66
($100-$34).12

The Roads Not Taken

100

If the corporation distributed a cash dividend
of $33, the corporation could elect to frank the
dividend in any amount up to $14.83 (determined
by multiplying the amount of the distribution by
.4493 (the shareholder rate divided by one minus
the shareholder rate). The corporation would
reduce the SeA by the amount of the credit.
Thus, if the corporation chose to fully frank the
dividend, the shareholder would report as income
the gross dividend of $47.83 ($33 plus $14.83)
and claim a credit of $14.83 against the individual
tax. If the $14.83 credit exceeded the shareholder
level tax imposed on the $47.83 gross dividend,
a low-bracket shareholder could use the excess
credit to offset tax imposed on other income. For
example, a shareholder in the 31 percent bracket
would incur tax liability on the gross dividend of
$14.83 (.31 x$47.83) and would receive a credit
of $14.83, exactly offsetting the tax due. A
shareholder in the 15 percent bracket would incur
tax liability on the gross dividend of $7.17
($47.83XI5 percent) and would receive a credit
of $14.83, leaving an excess credit of $7.66 to
offset other tax liability.13

The imputation credit prototype requires
corporations to report annually to each shareholder and to the IRS the amount of dividend distributions to shareholders and the associated imputation
credits. The imputation credit prototype also
requires corporations annually to report to the IRS
the adjustments to and balance in the SeA. This
would permit the IRS to verify aggregate allowable credits to a corporation based on the amount
of taxes paid and to compare the allowable
amount with credits reported by shareholders.
A liquidating corporation would distribute the
remaining balance in its SeA among shareholders
in proportion to the amount of other assets distributed to them. As with any other distributions for
which imputation credits are allowed, the amount
of the shareholder credit would be included in
income and could be used to offset gain on the
liquidation or, in the case of excess credits, other
income.
The imputation credit prototype, like the
dividend exclusion prototype, treats adjustments to

prior years' tax liability as adjustments made in
the current year. 14 Thus, an increase in corporate
tax liability for a prior year would result in an
increase in the SeA for the year of the audit
adjustment. A decrease in a prior year tax liability
could give rise to a refund, but only to the extent
of the current balance in the SeA. Any excess
amount would be carried forward to be applied
against future corporate taxes. IS
This method ensures that an adjustment that
affects a corporation's prior year tax liability
would not affect shareholders' individual tax
positions for the prior year. Shareholders may
thus claim the credits reported to them as allowable by the corporation, without concern that
subsequent corporate level adjustments might
require them to fue amended returns. 16
The imputation credit prototype allows coIporations to carry back losses to claim refunds only
to the extent of any balance in their SeA, with
the SeA being reduced by the amount of the
refund. This limitation prevents corporations from
carrying back losses in order to obtain a refund of
taxes that already have served to reduce shareholders' taxes through imputation credits attached
to dividends. 17 Any unused losses can be carried
forward as under present law. 18
The prototype generally pennits corporations
to choose the extent to which dividends are
franked, with the consequence that there is no
need for a mandatory stacking rule. This flexibility allows a corporation with preference or foreign
source income to continue to detennine its dividend policy by weighing the business reasons for
maintaining a particular level of cash distributions
against the possible detriment to shareholders of
receiving unfranked dividends. In contrast, the
dividend exclusion prototype requires excludable
dividends to be paid until the EDA balance is
exhausted. This is equivalent to an imputation
credit system that requires corporations to pay
fully franked dividends to the extent of the seA.
Permitting the additional flexibility to pay partially franked dividends requires anti-abuse rules in
addition to those ad<?pted in the dividend exclusion prototype to prevent corporations from

101

paying franked dividends to taxable shareholders
and unfranked dividends to tax-exempt shareholders. See Section 11. F .

Corporate Shareholders
The imputation credit prototype allows a
corporate shareholder a 100 percent dividends
received deduction (DRD) for both franked and
unfranked dividends, regardless of the degree of
affiliation. 19 Moving to a single level of tax
under integration does not require increasing the
DRD to 100 percent for unfranked and partially
franked dividends. The dividend exclusion prototype, for example, retains current law for taxable
dividends. See Section 2.B. The imputation credit
prototype contains a 100 percent DRD for all
dividends, however, because retaining current law
for partially franked dividends would create
unwarranted complexity. 20
As under current law, the DRD would be
available for dividends from domestic corporations and for a portion of dividends from certain
foreign corporations engaged in business in the
United States. Any imputation credit associated
with a dividend would be added to the corporation's SCA. Adding the credit to the corporate
shareholder's SCA preserves imputation credits
for individual shareholders when the earnings are
ultimately distributed out of corporate solution.
Because the 100 percent DRD would be
equally available for fully franked and unfranked
dividends, distributions of corporate preference
income would be taxed only when ultimately
distributed to individual shareholders. Mechanically, this result occurs because unfranked dividends
do not increase the recipient's SCA.21 Retaining
the DRD for preference income is consistent with
the rationale for a credit limitation system discussed above. Requiring immediate taxation in
full of preference income received by corporate
shareholders would represent a significant departure from current law and would increase the cost
of intercorporate dividends. Preserving the DRD
means that the ultimate taxability of preference
income is determined at the individual level. 22

The Roads Not Taken

Other countries adopting distribution-related
integration have dealt with the issues presented by
affiliated groups in a variety of ways. In most
cases, these countries have permitted the extension of preferences while the income remains in
corporate solution, as we suggest here. For
example, New Zealand generally exempts intercorporate dividends from taxation and corporate
shareholders are permitted to add credits from
franked dividends to their own SCA. Similar rules
apply in Australia for dividends received by
public corporations and for franked dividends
received by private corporations from within the
same closely held group. In the United Kingdom,
although the intercorporate dividends are generally subject to ACT, a "group dividend election"
can be made to avoid the ACT and the imputation
of credits with respect to distributions between
closely affiliated corporations. See Appendix B.

l1.C ROLE OF THE
CORPORATE ALTERNATIVE
MINIMUM TAX
Under current law, the corporate alternative
minimum tax (AMT) seeks to ensure that, in each
taxable year, corporations pay a minimum amount
of tax on their economic income. A corporation
must pay the higher of the AMT or the regular
tax liability on its alternative minimum taxable
income (AMTI) for the taxable year. Congress
adopted the corporate AMT system in 1986 partly
in response to widely publicized reports of major
companies not paying taxes in years in which they
reported substantial earnings and, in some cases,
paid substantial dividends to shareholders. 23
The imputation credit prototype retains the
corporate AMT. 24 Because the imputation credit
prototype described here does not substantially
alter the current treatment of either retained or
distributed preference income, the AMT would
continue to serve its .current function of limiting
corporate tax preferences and ensuring that
corporations continue to pay some minimum
amount of tax on retained income. 25
Since some corporations are subject only to
the AMT and pay no regular corporate tax for

The Roads Not Taken

long periods, the question whether the AMT
should be considered taxes paid and added to the
SeA is important. For these taxpayers, the corporate AMT is the only tax paid, and, despite the
current law provisions that allow the AMT to be
credited against regular corporate tax in subsequent years, it would not be realistic to view the
AMT simply as an advance deposit against ultimate corporate tax liability. We therefore treat the
AMT in the same manner as regular corporate
taxes paid. Thus, each dollar of AMT is converted into an SeA balance using the formula set
forth in Section 11.B?6 At the corporate level,
any AMT paid would continue to be carried
forward and credited against regular corporate tax
in subsequent years, but regular corporate tax that
is not paid by reason of the credit allowed for
AMT previously paid would not be treated as tax
paid. Accordingly, under the prototype, both
regular taxes paid and AMT paid would be added
to the SeA, and regular tax that is offset by the
AMT credit would not be added to the SeA. If
the AMT were not treated as taxes paid, distributions attributable to earnings that have been
subject to AMT would be taxed twice, and a
higher rate of tax would be imposed on preference
activiti~s. However, if distributions are made with
shareho ..der credits arising from payments of
AMT, such reductions in the seA will reduce the
corporation's ability to pay franked dividends
when the AMT reverses and the corporate tax is
reduced by AMT credits.

l1.D FOREIGN SOURCE INCOME
In general, the prototype permits a U.S.
corporation to claim foreign tax credits against
corporate tax to the same extent as under current
law. A U.S. corporation, however, would increase its SeA only by the amount of the residual
U.S. tax (if any) imposed on its foreign source
income. Distributions out of foreign source
income shielded from U. S. corporate tax by
foreign tax credits generally would be unfranked
and, therefore, would be taxed at the shareholder
level as under present law.
Thus, U.S. corporate shareholders owning less
than 10 percent of a foreign corporation's voting

102

stock (the threshold requirement for claiming an
indirect foreign tax credit under IRe § 902)
would include in income, as under current law,
dividends from the foreign corporation and claim
a foreign tax credit for foreign withholding taxes.
The corporate shareholder, however, would not
add foreign income taxes paid by the foreign
corporation or foreign withholding taxes on
dividends to its SCA.
U. S. corporate shareholders owning at least 10
percent of a foreign corporation's voting stock
would continue to include in income dividends
from the foreign corporation and to claim a
foreign tax credit for foreign withholding taxes on
the dividend as well as foreign taxes paid by the
foreign corporation. The corporate shareholder
would add to its SCA only the U.S. residual tax,
if any, paid on the dividend. 27
U.S. corporations with foreign branch operations would continue to be subject currently to
U.S. tax on their worldwide income with a credit
for foreign income taxes imposed thereon. 28 As
with earnings of foreign subsidiaries, the U.S.
corporation would increase its SeA only by the
amount of any residual U.S. tax imposed on the
foreign source income.
The imputation credit prototype does not
change the treatment of individuals owning stock
in foreign corporations. U.S. individual shareholders would continue to include in income
dividends received and claim a foreign tax credit
for any foreign withholding taxes imposed on the
dividend. Individual shareholders would not
receive an imputation credit for any income taxes
paid by the foreign corporation.
In connection with treaty negotiations with
countries that have imputation credit systems, the
United States may wish to consider whether
imputation credits for foreign taxes paid could be
extended on a bilateral basis. Serious complexities
would arise, however, in applying at the individual shareholder level the foreign tax credit limitations that are designed to ensure that foreign taxes
paid are not credited against U. S. taxes at tax
rates in excess of the applicable domestic tax rate.

103

On the other hand, ignoring the foreign tax credit
limitation would reduce or eliminate U. S. taxes
on U.S. source income, in effect transferring
domestic revenues to foreign treasuries. A possible approach might be to extend the benefits of
foreign corporate taxes paid to individual U. S.
shareholders in the fonn of a shareholder level
exclusion of foreign source corporate income.
Even in this event, care would need to be taken to
avoid inappropriate results. 29

l1.E CHOICES REQUIRED
BECAUSE OF
SHAREHOLDERS WITH
DIFFERENT RATES
Tax-Exempt Shareholders
As discussed in Chapter 6, this Report recommends that integration retain the current treatment
of corporate income distributed to tax-exempt
shareholders. 30 Corporate taxable income would
continue to bear one level of tax. Corporate
preference income and foreign source income
shielded from U. S. corporate tax by foreign tax
credits would continue to be exempt from U. S.
tax at both the corporate and shareholder level to
the extent distributed to tax-exempt shareholders.
Imputation credits could not be used against UBIT
liability. 31

Foreign Shareholders
Chapter 7 of this Report recommends that
foreign shareholders making inbound investments
should not by statute receive the benefits of
integration available to U.S. shareholders, and
that any such extension of the benefits of integration should occur only through treaties. Accordingly, the imputation credit prototype does not
permit foreign shareholders to claim a refund of
the imputation credit or to use the credit to offset
withholding tax imposed on dividends. The 30
percent statutory withholding tax would continue
to apply to the amount of the dividend without
gross up, subject to applicable treaty reductions.
The branch profits tax would continue to apply to
U.S. branches of foreign corporations. Thus, a
U.S. branch of a foreign corporation would be

The Roads Not Taken

taxable on its income effectively connected with
a U.S. business (subject to any available treaty
exemptions), and the branch's earnings withdrawn
from the U.S. business (the dividend equivalent
amount) would be subject to the branch profits tax
under IRe § 884(a) (as modified by any applicable treaty), without credit for U.S. taxes paid on
effectively connected income.
Denying imputation credits to foreign shareholders follows the approach generally adopted by
our trading partners that have integrated corporate
tax systems. Although the imputation credit would
not be available to foreign shareholders as a
statutory matter, a dividend to a foreign shareholder would reduce the distributing corporation's
SCA by the same amount as if the distribution had
been to a taxable domestic shareholder. 32

Low-Bracket Shareholders
The imputation credit prototype uses a rate of
31 percent to compute the shareholder credit.
Consequently, taxpayers subject to maximum tax
rates below 31 percent would receive imputation
credits on dividends received that may exceed the
shareholder level tax that would otherwise apply
to dividends received. Unlike the dividend exclusion or CBIT prototypes, no additional mechanism
(such as addition of a credit) is required to adjust
the tax burden to the shareholder's rate because
the franking process provides the shareholder with
the data necessary to compute shareholder level
tax (the grossed-up income and credit amounts).
The prototype allows these taxpayers to use
excess imputation credits to offset tax that would
otherwise apply to unfranked dividends or other
sources of income. This feature of the imputation
credit system produces an additional revenue loss
in comparison to the dividend exclusion prototype. Taxpayers who could not fully use such
credits against other income could not claim a
refund of the excess credits. 33

l1.F ANTI-ABUSE RULES
Adopting an imputation credit system in which
imputation credits are not refundable to taxexempt and foreign shareholders rna y create

The Roads Not Taken

104

incentives for taxpayers to "stream" fully franked
dividends to taxable shareholders and unfranked
dividends to tax-exempt shareholders. 34 Similar
incentives arise under the dividend exclusion
prototype, in which corporations would prefer to
pay excludable dividends to taxable shareholders
and taxable dividends to tax-exempt shareholders.
Section 2.B discusses the anti-abuse rules we
consider appropriate to limit streaming in the
dividend exclusion prototype, and we would adopt
similar rules in the imputation credit prototype.
Thus, for example, a holding period requirement
would have to be met for a taxpayer to claim an
imputation credit.
In general, opportunities for streaming would
be reduced if the imputation credit prototype
required corporations to pay fully franked dividends until their SeA balance were exhausted. In
that case, the imputation credit system would be
substantially similar to the dividend exclusion
system, which requires corporations to pay
excludable dividends to the extent of their SeA
balances. 35
Application of this rule in an imputation credit
context, however, could interfere with corporate
dividend practices by making the franking level
(and hence shareholder tax consequences) of
dividend distributions dependent on taxable income. To permit corporations to smooth the
pattern of dividends, including the pattern of
associated credits, the prototype permits corporations to pay partially franked dividends. Using
this flexibility, a corporation could reserve a
portion of its SeA balance to pay future franked
dividends.
Because the imputation credit prototype permits corporations to pay partially franked or
unfranked dividends even when they have an SeA
balance sufficient to frank the dividend fully, two
additional anti-abuse rules would be required.
First, to prevent excessive franking of dividends,
the prototype limits the amount of credit that can
be attached to a dividend. The imputation credits
attached to any dividend should not exceed the
maximum creditable tax on the pre-tax earnings
that generated the dividend. See Section II.B.

Second, the prototype requires corporations to
frank all dividends paid during a year to the same
extent. This rule prevents corporations from
paying unfranked dividends on one class of stock
held by taxable shareholders and unfranked
dividends on another class of stock held by taxexempt shareholders. This rule is essentially the
same as that adopted by New Zealand. 36 This
latter rule, while necessary to avoid distortion of
corporate dividend payment practices, could give
rise to significant complications for a corporation
with multiple classes" of dividend paying stock.

l1.G STRUCTURAL ISSUES
Corporate Acquisitions
The imputation credit prototype retains the
basic rules of current law governing the treatment
of taxable and tax-free corporate asset and stock
acquisitions. Adopting the imputation credit
prototype would permit taxable asset acquisitions
to be made with only a single level of tax. Corporate tax paid on gain recognized on the sale of
assets would be added to the SCA and would
create imputation credits to offset shareholder tax
when the corporation liquidates and distributes the
proceeds from the sale. Stock acquisitions may
face a higher tax burden than asset acquisitions
under distribution-related integration if capital
gains on corporate stock that are attributable to
retained earnings are" taxed in full at shareholder
rates. See Section 8.A. This problem could be
mitigated by a dividend reinvestment option. See
Chapter 9.
Nothing in the movement to distributionrelated integration would require a fundamental
change in the basic pattern of taxing qualifying
corporate reorganizations. Current law treats a
qualifying corporate reorganization as tax-free at
the corporate level (with the target's tax attributes, including its asset basis, carrying over to the
acquiror) and at the shareholder level. The policy
underlying the reorganization provisions is that
imposition of tax is inappropriate where a corporate reorganization merely effects a readjustment
of shareholders' continuing interests in corporate
property under modified corporate forms. This

105

policy applies equally under distribution-relation
integration, because it reflects a judgment about
when income should be recognized under a realization-based tax system that does not require
corporate assets or stock to be marked to market,
not a judgment about whether two levels of tax
should be imposed on recognized corporate
income. 37
Rules would be needed to divide a corporation's SCA when it engages in a divisive reorganization. Rules are needed to discourage the use
of divisive reorganizations to isolate amounts in
the SCA in one corporation for the benefit of one
group of shareholders. 38 Current law rules generally provide that earnings and profits of the
distributing corporation in a divisive reorganization that qualifies as a D reorganization under
IRC § 368(a)(1)(D) are divided between the
distributing corporation and the controlled corporation based on the relative fair market value of
their assets. A similar rule could be adopted to
govern the allocation of SCA balances in divisive
reorganizations.
For the reasons set forth in Chapter 2, we do
not urge any rules limiting the use of SCA balances following an ownership change. See "Antiabuse Rules" in Section 2.B.

The Roads Not Taken

ensures that no more than one level of tax is
collected on corporate earnings distributed as
interest. Accordingly, the introduction of integration, without any change in the rules for taxing
debt, would create greater parity in the taxation of
debt and equity.
Because the dividend exclusion and imputation
credit prototypes are designed to retain the existing level of corporate taxes on equity capital
supplied by foreigners and tax-exempt entities,
however, some disparities will remain in the
treatment of debt and equity capital supplied by
those investors. Retaining the interest deduction in
an integrated system would permit earnings that
are used to pay interest to tax-exempt and certain
foreign bondholders to continue to escape U.S.
tax entirely.
Thus, for tax-exempt and foreign investors at
least, the dividend exclusion and imputation credit
prototypes generally maintain current law's bias in
favor of debt fmancing. Eliminating this bias is a
principal argument for CBIT, which represents a
natural extension of the dividend exclusion prototype to debt and imposes tax once at the entity
level. Equating the treatment of debt and equity in
an imputation credit prototype would require a
different approach-a bondholder imputation
credit system.

Earnings and Profits
The imputation credit prototype, like the
dividend exclusion prototype, retains the current
earnings and profits rules for determining when a
distribution is treated as a dividend rather than a
return of capital. See Section 2.F.

l1.H EXTENDING THE
IMPUTATION CREDIT
PROTOTYPE TO DEBT
Adopting any of the methods of integrating the
corporate and individual income taxes discussed in
this Report would narrow significantly the differences in taxation of debt and equity. Under
integration, only one level of tax generally would
be imposed on corporate earnings distributed as
dividends. Retaining the interest deduction also

Under a bondholder credit system with no
corporate level deduction for interest, the mechanics would generally follow the rules applicable to
dividends. Corporate tax paid on earnings used to
pay interest or dividends would be passed through
to bondholders and shareholders as imputation
credits. Bondholders and shareholders would
include in income the amount of the cash interest
or dividend payments plus the imputation credits
and could use the credits to offset tax on interest
income. 39 Tax-exempt and foreign shareholders
would not be entitled to claim refunds of imputation credits, and taxable shareholders could use
excess credits to offset tax on other income but
not to claim refunds. 40
A bondholder credit system differs in certain
ways from CBIT, which equates the treatment of

The Roads Not Taken

debt and equity at the business, rather than at the
individual, level. An imputation credit system
would tend to impose taxation on the supplier of
business fmancial capital rather than on the entity.
The two approaches are similar when the business
and its suppliers of capital would be taxed at the
same rates but will diverge if the tax rate of the
supplier of capital is different from the CBIT
rate. 41 Thus, for example, if both borrower and
lender are taxable, but the lender's rate is less
than the borrower's rate, CBIT will tax the
interest income at the CBIT rate, while the bondholder credit system will generally tax the income
at the lender's rate. 42
Although the bondholder credit system would
generally mirror the imputation credit prototype
detailed in this chapter, addition of a bondholder
credit may require reexamination of the treatment
of foreign investors. The issues would be similar
to those posed in moving from the dividend
exclusion prototype to CBIT. Retaining current
law would require collecting two levels of tax on
dividends and zero or one level of tax on interest.
Such treatment would, however, violate the
equality between debt and equity that is the goal
of adopting a bondholder credit system. Accordingly, to maintain parity between debt and equity,
imputation credits should not be refundable to
foreign investors, but the 30 percent withholding
tax now applicable to dividends and nonportfolio
interest (and the branch profits tax) should be
repealed. 43

11.1 DIVIDEND REINVESTMENT
PLANS (DRIPs)
Chapter 9 discusses how a corporation might
use an elective DRIP in the dividend exclusion
and CBrr prototypes to allow shareholders to
increase share basis to reflect earnings that have
been taxed at the corporate level. A DRIP minimizes the extent to which taxing capital gains on

106

sales of corporate stock imposes a second level of
tax on such earnings. See Chapter 8.

An elective DRIP could be made a part of an
imputation credit prototype as well. A corporation
would be pennitted to declare deemed dividends
up to the amount that can be fully franked by the
balance in its SCA.44 Shareholders would include
in income the amount of the deemed dividend plus
the associated imputation credit and could use the
credit to offset tax due. 45 Share basis would
increase by the amount of the deemed
dividend. 46
Permitting a DRIP in the imputation credit
prototype requires one additional rule to limit
streaming of credits.
As discussed in
Section 11. F , the prototype limits streaming
through cash dividends by requiring each corporation to frank all cash dividends paid during a year
in the same proportion (the consistency rule).47
The consistency rule is necessary because the
imputation credit prototype, unlike the dividend
exclusion and CBIT prototypes, permits corporations to determine the extent to which dividends
(and interest payments, if a bondholder credit
were adopted) are franked.
Absent additional restrictions, a corporation
could use a DRIP to stream by paying unfranked
cash dividends on classes of stock held by taxexempt shareholders and fully franked deemed
dividends on classes of stock held by taxable
shareholders. To limit this practice, the prototype
permits corporations to use an elective DRIP only
if all cash dividends paid during some defmed
period before and after the deemed dividend are
fully franked. This rule effectively extends the
consistency rule to deemed dividends and limits
the benefits of a DRIP to corporations that pay
insufficient cash dividends to carry out its SeA
balance-not those that underfrank cash dividends
and distribute the remainder of the SCA through
the DRIP.48

CHAPTER 12: OTHER PRoPOSALS TO REDUCE THE
BIAS AGAINST CORPORATE EQUITY
12.A DIVIDEND DEDUCTION

from U.S. tax by foreign tax credits. Allowing
such deductions would not simply eliminate
corporate taxes paid on that income (because, by
definition no U.S. corporate taxes have been paid)
but instead would permit the corporation to shelter
earnings on which U.S. corporate tax would
otherwise be imposed. 6

We have not developed a dividend deduction
prototype in this Report. However, the 1984
Department of the Treasury Report on tax reform
recommended a 50 percent dividends paid deduction and the President's 1985 tax proposals included a 10 percent deduction. 1 A dividend deduction
system produces results contrary to our general
recommendations that integration not be the
occasion for eliminating the corporate level tax
imposed under current law on distributions to taxexempt and foreign shareholders. 2 We view these
general recommendations as important in ensuring
that corporate income distributed to such shareholders continues to bear tax similar to that under
current law. In addition, a dividend deduction
proposal would be substantially more expensive
than either a dividend exclusion or imputation
credit system. 3

It is not altogether clear how a dividend
deduction system would treat foreign shareholders. Presumably, the deduction would be allowed
for dividends paid to foreign shareholders, and the
30 percent withholding tax on dividends would be
retained, although treaty provisions reduce the
withholding tax to as low as 5 percent. Similarly,
the branch profits tax on domestic branches of
foreign corporations presumably would be retained with a modification to provide parity with
the dividend deduction for domestic corporations.
Since dividends would be taxable only to the
recipient in a dividend deduction proposal, there
would be no dividends received deduction for
corporations. 7 A DRIP probably would not be
appropriate in a dividend deduction approach
because it could result in allocation of taxable
income to shareholders without receipt of cash
sufficient to satisfy the shareholder's resulting tax
liability. 8

The primary arguments for a dividend deduction approach are that it results in equivalent
treatment for debt and equity and that it taxes
distributions at the shareholder rate. The first
claim is not strictly accurate to the extent that
interest is deductible as it accrues while dividends
are deductible only when paid. 4 The second claim
is correct but will exacerbate the bias toward
distribution of earnings inherent in any distribution-based system, particularly when, as under
current law, the corporate rate exceeds individual
rates.

While we have not developed a dividend
deduction prototype in this Report, we review
below two proposals for dividend deduction
systems, one made in 1991 by the Capital Taxes
Group of the Institute for Fiscal Studies in the
United Kingdom and one made in 1989 by the
Reporter for the American Law Institute's Federal
Income Tax Project (Subchapter C). These proposals are not presented here as fully as other
integration prototypes but are included as related
proposals intended to improve the neutrality of the
tax treatment of debt and equity finance for
corporations.

If policy makers were to select a dividend
deduction system, it would be important to incorporate a mechanism analogous to the EDA of the
dividend exclusion prototype to limit the amount
of deductible dividends to the amount on which
U.S. corporate tax has been paid. 5 Absent such
a restriction, a dividend deduction system would
allow a deduction for dividends paid out of preference income and foreign source income sheltered

107

108

The Roads Not Taken

12.B INSTITUTE FOR FISCAL
STUDIES PROPOSAL
The Capital Taxes Group of the British Institute for Fiscal Studies (IFS) proposed the introduction of an "Allowance for Corporate Equity"
(AFCE).9 Under this approach, a corporation
would be allowed to deduct in its calculation of
taxable income an allowance based on shareholders' equity employed in the business. The
intent of this proposal is to enhance neutrality by
treating equity flnance like debt fmance. 1o
The deductible AFCE allowance would be
equal to the product of shareholders' funds"
(generally the corporation's total equity capital)1l
and an "appropriate nominal interest rate." The
interest rate used for calculating the AFCE would
be set by the government for all corporations and,
in general, should reflect a normal market rate of
return. The IFS recommends that the rate be
established each month equal to the rate for a
medium-term government security. Because flnns
with risky opportunities or facing informational
imperfections in capital markets would have costs
of funds significantly higher than the allowable
rate for deduction, mature, less risky flnns would
receive a greater relative benefit from the AFCE
system.
tt

The AFCE system prevents double counting of
intercorporate investments by reducing shareholders' funds by the amount of funds invested in
other firms. It also prevents allowance of both an
interest deduction and an AFCE allowance with
respect to intercorporate equity investments
funded by debt by imputing a negative AFCE
adjustment to the borrower. 12
The AFCE proposal is designed to operate in
a classical corporate tax system to reduce the tax
bias against equity fmance. The IFS proposal is
not a true integration proposal. Corporate equity
income in excess of the AFCE allowance would
remain subject to a second level of tax when such
income is distributed or when shareholders are
taxed on carital gains attributable to such income.
As a consequence, the IFS proposal would not

eliminate the bias against the corporate form and
the incentive to retain rather than distribute
corporate equity income in excess of the AFCE
allowance.

12.C AMERICAN LAW INSTITUTE
REPORTER'S STUDY DRAFT
In 1989, the Reporter for the American Law
Institute (AU) Federal Income Tax Project
(Subchapter C) outlined a set of four proposals for
refonn of the corporate tax.l3 The Reporter's
Study Draft proposals are not integration proposals. They are intended to revise the classical
corporate tax system to reduce the tax bias against
new equity fmance and to eliminate the tax bias
against dividend distributions relative to nondividend distributions, e. g. , share repurchases.
The latter goal would be accomplished by increasing tax rates applied to nondividend distributions
rather than by decreasing tax rates applied to
dividend distributions.
The Reporter's Study Draft advances two
proposals to reduce the tax bias against new
equity fmance. First, corporations would receive
a deduction for dividends paid on new equity
capital (Qualified Contributed Capital or
QCC).14 The deduction would be equal to a
prescribed interest rate mUltiplied by net contributed capital less extraordinary dividends and
nondividend distributions. The prescribed interest
rate for deductions would be limited to the longtenn borrowing rate specified under IRe § 1274,
plus 2 percent.
Second, the Reporter's Study Draft would
limit corporate interest deductions to the net
amount of debt capital raised. In particular, no
deduction would be allowed for interest on "converted equity," including debt incurred to fmance
an extraordinary dividend or stock acquisition,
share repurchase, or any other nondividend
distribution. The deduction allowed for interest on
any other type of debt also would be limited to
the long-tenn borrow,ing rate specified under IRe
§ 1274 plus 2 percent.

109

Taken together, these two proposals are
designed to reduce the tax bias against new equity
fmance. 1S
The concern over the tax bias against dividend
distributions relative to nondividend distributions
motivates the other two proposals in the
Reporter's Study Draft. First, the AU Reporter
proposes a "minimum tax on distributions"
(MID) equivalent to 28 percent of the gross
amount of any extraordinary dividend or nondividend distribution, including distributions in
redemption and liquidation and any purchase of
shares. The tax would be collected by the distributing corporation, and would be creditable against
a shareholder's tax on the distribution (but not
against other income).16
Second, in the case of direct investments in a
corporation by another corporation, the Reporter's
Study Draft would treat a purchase of shares in a
corporation by another corporation that owns at
least 20 percent of the shares as a nondividend
distribution subject to the MID and other applicable rules. However, intercorporate dividends

The Roads Not Taken

would not be subject to tax, and basis adjustments
similar to those provided under the current consolidated return regulations would be made. For
portfolio investments, on the other hand, the
investor corporation .would be taxed in full like
any other investor and no dividends received
deduction would be allowed. 17
The Reporter's Study Draft proposals would
reduce the tax bias against new equity fmance,
while maintaining the tax bias against dividend
payments from accumulated equity. The economic
assumptions underpinning the AU proposals seem
to be those of the "new view" of dividend taxation, in which the taxes on dividends from accumulated equity are capitalized into share values
and do not affect dividend decisions. As a result,
extending dividend relief to accumulated equity is
perceived as conferring a windfall gain to "old"
equity, since under the assumptions of the new
view, dividend distributions are unavoidable. As
discussed in Chapter 13, we accept the "traditional view," in which reducing the tax burden on
dividends generally increases dividend payouts
and economic efficiency. 18

PART V: ECONOMIC ANALYSIS OF INTEGRATION
CHAPfER

13: ECONOMIC EFFECTS OF INTEGRATION

13.A INTRODUCTION AND

optimal. Depending on the model, fmancing
assumption, and prototype, nominal dividend
payout ratios would increase by 2 to 6 percentage
points.

SUMMARY
This chapter presents quantitative estimates of
the impact of the integration prototypes developed
in the Report on the allocation of resources,
corporate fmancial policy, portfolio allocation,
and Federal tax revenues.

By shifting resources into the corporate sector,
reducing corporate borrowing, and encouraging
dividends, the integration prototypes generate
changes in economic welfare. Overall, the prototypes improve economic welfare in all calculations, and the improvement ranges from an
amount equivalent to 0.07 percent of annual
consumption (total consumer spending on goods
and services) to an amount equivalent to 0.73
percent of consumption, or from approximately
$2.5 billion to $25 billion per year. CBIT or
shareholder allocation prototypes generally contribute the greatest increases in welfare, but the
distribution-related prototypes also produce significant economic welfare gains. Much of the variation in results reflects differences in the models
used to analyze the prototypes or differences in
fmancing assumptions, rather than differences
among prototypes. Indeed, one striking feature of
the calculations is that within each model, and for
a given fmancing assumption, structurally different prototypes often have similar overall effects
on economic well-being. These results accord
with the general economic equivalence of basic
integration prototypes in the absence of distortions
induced by rate differentials demonstrated in
Appendix C.

We examine the effects of each integration
prototype using four alternative models of the
economy and two assumptions about how integration would be financed. Results differ from model
to model, as well as by fmancing assumption, but,
in general, the integration prototypes reduce the
tax penalty on corporate investment and encourage
capital and other resources to flow into the corporate sector. Depending on the prototype, model,
and fmancing assumption, this capital expansion
ranges from a 2 to 8 percentage point increase in
the capital stock used in the corporate sector. In
dollar terms, this ranges approximately from $125
billion to $500 billion in additional corporate
capital. CBIT generally produces the largest
expansion of corporate capital, but in several of
the calculations, the more traditional integration
prototypes yield a similar expansion.
In addition, each of the integration prototypes
generally encourages corporations to use less
debt. Estimated debt to asset ratios decrease by I
to 7 percentage points, depending upon the model,
fmancing assumption, and prototype. CBIT is the
best prototype for encouraging finns to reduce
their relative use of debt.

The results summarized above are generated
from models of the economy that abstract from
international capital flows. While internationally
mobile capital can cause tax law changes to have
different effects from those predicted by closedeconomy models, there is no consensus among
economists regarding the sensitivity of international flows of debt and equity capital to changes in
net returns, especially for a country such as the
United States with a very large domestic
economy. Consequently, the Report does not

The integration prototypes encourage corporations to increase the portion of earnings distributed as dividends. Both CBIT and the shareholder
allocation prototype promote efficient corporate
dividend policy by almost entirely eliminating
taxes as a consideration. In contrast, the distribution-related prototypes encourage finns to payout
more of their earnings as dividends than may be
111

Economic Analyses

present a detailed quantitative analysis of integration in an international context, although the
effects of the integration prototypes on international capital flows and portfolios are discussed in
Section 13.F. The distribution-related and shareholder allocation prototypes are estimated to have
only a small effect on the net capital flows into
the United States; the effects of CBIT are more
uncertain. Each integration prototype, however,
may change substantially the composition of
international portfolios, even if net flows of
capital are not greatly affected.
Section 13.B analyzes the principal economic
issues surrounding the debate over the benefits of
corporate tax integration, building on the discussion in Chapter 1. Section 13. C describes important methodological issues in modeling effects of
integration on economic efficiency. Section 13. E
evaluates effects of integration on the cost of
capital and corporate fmancial decisions. A more
complete analysis of economic effects of integration using a set of computable general equilibrium
models is provided in Section 13. F. Issues relating to distributional implications of integration are
discussed in Section 13. G. Finally, estimates of
integration prototype's effects on Federal tax
revenue are presented in Section 13.H.

13.B CORPORATE TAX
DISTORTIONS: ECONOMIC
ISSUES
Bias Against Investment in
Corporate Form
The waste of economic resources from the
tax-induced misallocation of capital between the
noncorporate and corporate sectors was the
original focus of economists' criticism of the
classical corporate income tax system. Beginning
with Harberger (1962), economists have argued
that a classical corporate tax system increases the
share of capital allocated to the noncorporate
sector, thereby raising pre-tax required rates of
return in the corporate sector.
Harberger's model divides the economy into
two sectors, a corporate sector and a noncorporate

112

sector. The Harberger model has four central
assumptions. First, in both sectors, output is
produced by combining capital and labor. Second,
the total amounts of capital and labor supplied in
the economy are fixed. Third, although the total
amounts of capital and labor supplied are fixed,
the amounts supplied to each sector can vary.
Fourth, suppliers of capital and labor seek to
maximize their incomes.
Taken together, the third and fourth assumptions above have an important implication: In the
long run, the net return on the last dollar of
capital in each sector must be the same, since
suppliers of capital invest their capital where its
net return is highest. As a result, capital will flow
out of the sector with a low net return and into
the sector with the high return. This flow continues until net rates of return are equalized between
the two sectors.
Over the years, more sophisticated versions of
Harberger's model have been developed to examine more carefully the costs of the economic
distortions related to the corporate income tax.
One important step was the development of more
complex models with many sectors of the economy. 1 Most recently, researchers have noted that
economic distortions from the corporate income
tax are greater than earlier estimates to the extent
that the tax distorts the relative importance of
corporate and noncorporate producers within an
industry. 2 Costs associated with this additional
margin of distortion arise when corporate and
noncorporate producers within an industry have
different advantages, for example, greater ability
to exploit scale economies by corporations or
greater entrepreneurial skill in noncorporate
organizations. 3
Current U.S. tax law distorts the allocation of
investment away from the economy's corporate
sector and into the noncorporate sector whenever
investors require equity to fmance investment.
The corporate cost of equity capital generally
exceeds the noncorporate cost of capital because
of the two-level tax on corporate equity income.
Consequently, corporate investment projects
require a higher pretax rate of return than projects

Economic Analyses

113

of noncorporate business enterprises. Therefore,
some corporations fail to undertake investments
that would be profitable if the tax burden on
corporate and noncorporate investments were the
same. Moreover, for some business entetprises,
the added corporate taxes exceed the benefits of
incorporation, and such businesses forego the
advantages of incorporation and choose instead to
operate as partnerships or sole proprietorships.4

net income accounted for only 3 percent of total
cotporate net income, up only slightly from 2.1
percent in the previous decade. Data for 1987 and
1988, in contrast, indicate a substantial increase in
S cotporation net income to 8.6 percent of all
cotporate income in 1987 and 9.5 percent in
1988. 7 This increased S cotporation activity
seems to be a response to the 1986 Act's inversion of the top individual and cotporate tax rates
and repeal of the capital gains rate preference. 8

While the classical system may encourage
A measure of the bias against equity investcorporations to operate in noncorporate fonn,
ment
in a cotporation that pays dividends is the
aggregate data to date do not document a longextent to which the combination of the corporate
tenn trend of shifting economic activity away
from the corporate sector. Figure 13.1 shows that tax rate on earnings and the individual tax rate on
incomes of owners of noncorporate businesses dividends exceeds the individual tax rate on
have fallen as consistently as a share of net business income. In the case of equity investments
in a cotporation, retained earnings are taxed
national product as have cotporate profits. By
ultimately at the shareholder level as capital gains.
contrast, the total income (profits, interest, rents,
Accordingly,
the measure of the bias against
and wages) generated in the cotporate sector has
increased slightly, from an average of 50 percent equity investment in the cotporate sector in that
of net national product in the 1950s to an average case is the extent to which the combination of the
of 53 percent in the 1980s (Figure 13.2). Other corporate tax rate and the individual capital gains
long-tenn comparisons of corporate activity to the rate exceeds the effective individual tax rate on
business income.
general economy also fail to present any general
pattern of disincorporation. S However, data for
Figure 13.1
the past few years (some of
Profits of Nonimancial Corporations,
it preliminary) does suggest
Proprietors' Income, and Net Interest as a
reduction in the size of the
Percentage of Net National Product, 1950-1990
corporate sector relative to
15%
the overall economy and to
the noncotporate sector. 6
Subchapter S cotporations have accounted for an
increased share of corporate
profits and have contributed
to the declining role of the
corporate income tax,
particularly since 1986. The
Subchapter S Revision Act
of 1982 increased the
attractiveness of S corporations and led to an
expansion of S corporation
activity. However, in the 4
years following the 1982
amendments, S corporation

Nonfinancial Corps. Net Interest

-~,
/,.,-/
~~

1950

1955

1960

1965

1970
Year

1975

1980

'-..--------/

1985

1990

Source: National Income and Product Accounts, Bureau of
Economic Analysis, U.S. Department of Commerce.

114

Economic Analyses

corporate investments. The
overall effect depends upon
whether the combination of
65%
the corporate tax rate and
the effective capital gains
rate is greater than, equal
_.-"
to, or less than the individ60%
ual tax rate on business
income. Even when real'\
ized capital gains are taxed
at the same rate as ordinary
income, the effective capital gains rate is generally
lower than the statutory
rate, because the capital
gains tax can be deferred
until gain is realized
45% II I I i i I I i i i i i i i i I i i i i I I i i i i i i i i I i i I i i i i i i I
through a sale or ex1980
1990
1985
1975
1965
1970
1960
1955
1950
change.1O In an extreme
Year
case, if the combination of
Corporate Gross Domestic Product as a percentage of Gross Domestic Product
the corporate tax rate and
Nonfinancial Corporate Gross Domestic Product as a percentage of Gross Domtstic Product
capital gains rate is lower
than the individual rate, the
Nonfinancial Corporate Gross Domestic Product as a percentage of Gross National Product
classical system may actualCorporate Gross Dommtic Income as a percentage of Net National Product
ly create a bias in favor of
investing in corporate
Source: National Income and Product Accounts, Bureau of
equity.11 Currently, howEconomic Analysis, U.S. Department of Commerce.
ever, even a full exclusion
from tax of capital gains on
corporate shares would
Assuming a positive effective corporate tax
generally not eliminate the tax system's bias
rate, the classical system always creates a bias
against equity investment in the corporate sector
against investing in equity in a corporation that
because the corporate rate exceeds the top
distributes all current earnings relative to a nonindividual rate.
corporate investment. If the corporate tax rate
were zero, corporate earnings would be taxed
Two other features of the tax system currently
only at the shareholder rate, and therefore the bias
reduce the tax bias in favor of noncorporate
against corporate equity would be eliminated. 9
investments. First, the benefits of accelerated
That the corporate rate currently exceeds the
depreciation are somewhat greater for corporaindividual rate does not create a new bias; it
tions, because corporate tax rates tend to exceed
individual tax rates on shareholders and on nonmerely exacerbates a bias that is present whenever
corporate businesses. Second, to the extent that
all current earnings are distributed and the corpocorporations fmance investments through debt, the
rate rate exceeds zero, regardless of its relationrelative tax advantage afforded noncorporate firms
ship to the individual rate.
is diminished. Considering only tax factors,
corporate and noncorporate entities face the same
For equity investments in a corporation that
cost of debt fmanced capital. Thus, to the extent
retains earnings, differences among tax rates may
corporations
fmance new investment with debt,
reduce, eliminate, or even reverse the bias against
Figure 13.2
Measures of Corporate Activity in the Economy
1950-1990

/

/

50%1 ,,, '

\

115

the difference in tax burden for total investment,
both debt and equity fmanced, will be reduced.

Bias Against Equity Finance
The Tax Bias Against Equity

The source of the bias against equity fmancing
is similar to the source of bias against corporate
investment described in the preceding section.12
An investment in corporate equity is subject to tax
once at the corporate rate and again at either the
individual rate or the effective rate on capital
gains. In contrast, interest earned on debt, like
income from an unincorporated business, is
subject to tax only at the investor's rate. Consequently, equity funded projects generally require
a higher pretax rate of return than projects
fmanced with debt. 13
Nontax Benefits and Costs of Debt Finance

Chapter 1 discussed important nontax and tax
considerations in corporate borrowing decisions.
Central to the argument that the tax bias against
equity fmance distorts corporate fmancing decisions is the existence of nontax costs and benefits
associated with corporate debt fmancing. If nontax
costs of debt are significant, losses in economic
efficiency can accompany the greater debt levels
resulting from the tax bias against equity fmance.
As corporate borrowing remained high during
the 1980s, many nontax arguments for high debt
fmancing appeared. Analysts most sanguine about
the rise in debt fmancing typically maintain that
debt is desirable because it gives suppliers of
capital an indirect means to monitor the activities
of managers. Their reasoning is that the need for
supervision results from the separation between
ownership and management that is characteristic
of the traditional corporate structure. A conflict
between ownership and management can emerge
if it is difficult for suppliers of capital to observe
and evaluate the activities of entrenched managers. In this kind of environment, management's
self interest may not always coincide with efficiently operating the business enterpris~with
maximizing value. 14

Economic Analyses

In practice, increased debt fmancing may be
an ineffective way to improve managerial incentives. It works best when most of the variation in
an enterprise's cash flow is specific to the firm. It
works poorly when most of the variation is common across business enterprises (as with industrywide or business cycle fluctuations). 15 Thus,
even when there are incentive benefits from debt,
the most efficient fmancial arrangement will
involve both debt and equity, with equity serving
as a cushion against economywide fluctuations in
profitability.
Many academic and business economists have
stressed the nontax costs of a declining reliance
on equity fmance. One concern is that the costs of
fmandal distress and bankruptcies could be
greater than in the past, more businesses with
high debt fmancing. Firm level data illustrate the
reason for this concern. Warshawsky has calculated weighted average, median, and ninetieth
percentile values of (market-value) debt to asset
ratios for firms in the COMPUSTAT Industrial
and Full Coverage samples, over the period from
1969 to 1988. 16 As with the aggregate data discussed in Chapter 1, all statistics for the subsamples indicate a rising debt to asset ratio,
though much of the increase occurred before
1980. This measure can, of course, be distorted
by large swings in the value of equities (as, for
example, in 1973 and 1974). The debt to asset
ratio has, however, climbed since 1983 in spite of
significant increases in the value of equity. 17
Warshawsky also calculated the ratio of interest
payments to cash flow for the individual business
enterprises. Over the 1969-1988 period, the mean
and median value of the ratio virtually doubled;
the value for the ninetieth percentile fums more
than tripled. Much of the change occurred during
the 1980s. In addition, the average quality of
publicly issued debt (as measured by bond ratings)
declined steadily in the 1980s.
To put the macroeconomic concern in shatper
perspective, Bemanke and Campbell considered
the experiment of imposing a reduction in cash
flows similar to those experienced during the
1974-1975 recession on a sample of fums with
fmandal conditions corresponding to 1986 data.

Economic Analyses

116

The sample was drawn from Standard and Poor's
COMPUSTAT flle, and therefore consisted
primarily of large flnns. The simulations implied
that a downturn like 1974-1975 would force more
than 10 percent of the sampled fmus into bankruptcy. Updates for later years in Bernanke,
Campbell, and Whited and in Warshawsky yielded
similar conclusions. IS
What role have tax distortions played in tilting
the balance between beneflts and costs of different
degrees of debt flnancing?I9 Under a tax system
that treats equity fmance unfavorably, fmus are
induced to have less equity outstanding, thereby
lowering their "equity cushion" against business
cycle risk, and raising the chance of incurring
costs of fmancial distress during a future downturn.20 The tax distortion makes this decision
rational for individual corporations but socially
inefficient.

Bias Against Corporate
Dividend Distributions
The current system of corporate income
taxation also may distort a corporation's choice
between distributing or retaining earnings and, if
amounts are distributed, whether they are paid in
the form of a nondividend distribution, such as a
share repurchase. There are two alternative
explanations in contemporary corporate fmance-commonly known as the "new view" and the
"traditional view"---of why corporations continue
to pay dividends despite the high relative taxation
of dividends compared with capital gains generated by reinvested earnings or share repurchases. 21
The traditional view asserts that dividends offer
special nontax benefits to shareholders that offset
their tax disadvantage. For example, dividends
may provide signals to investors about a corporation's relative fmancial strength or future
prospectS.22 Alternatively, high dividend payouts
may reduce managerial discretion over internal
funds (see the analogous discussion above of the
incentive benefits of corporate debt fmancing).
According to the traditional view, corporations set
dividend payments so that, for the last dollar of
dividends paid, the extra beneflt of dividends
equals their extra tax cost. Thus, the amount of

dividends paid out is expected to decrease as the
tax burden on dividends relative to capital gains
increases. Dividend taxes also raise the cost of
capital (and thereby lower investment) to the
extent that corporations payout earnings as
dividends. Thus, the traditional view argues that
raising dividend taxes will lower the dividend
payout ratio and incentives for real investment.
Moreover, under the traditional view, the need to
maintain dividend payments constrains the use of
retained earnings as corporations' marginal source
of equity fmancing for new investments; instead,
corporations frequently must turn to new equity
issues.
Under the new view, dividend payments offer
no nontax beneflts to shareholders relative to
retentions. 23 The hypothesis further assumes that
corporations have no alternative to dividends for
distributing funds to shareholders. Given these
assumptions, investor level taxes on dividends
reduce the value of the frrm, but do not affect the
finn's dividend or investment policies. Since
dividend taxes must eventually be paid, they are
capitalized in share values, reducing share prices
enough to compensate for the tax burden. In
effect, a dividend tax acts as a lump-sum tax on
equity existing when the tax is imposed, and on
new equity contributions. Therefore, corporations
prefer not to issue new shares to fmance
additional investment opportunities. Retained
earnings and debt are preferred sources of funds.
Dividends are determined as a residual after the
frrm undertakes all profltable investments.
Consequently, a permanent change in the tax rate
on dividends will not change a flnn' s investment
policies or payout decisions.24 Although the
dividend tax does not affect investment incentives,25 the capital gains tax affects investment
incentives because retentions increase the value of
a frrm' s shares and such appreciation is taxable as
a capital gain. 26
The tax policy implications of the traditional
and new views with respect to the taxation of
corporate income are quite different. The new
view assumes that the investor level taxes on
distributions are capitalized into share values,
with the consequence that (1) existing shares are

117

valued below the market value of corporate assets,
so eliminating or reducing taxes on existing
corporate assets would produce gains to current
shareholders and (2) moving to a system that is
more neutral in taxing retentions and distributions
would not encourage corporations to pay more
dividends. 27
In contrast, under the traditional view, where
new funds rather than retained earnings provide
the source of fmance for additional investments by
the corporation (1) shares should not sell at a
price below corporate asset values despite the
existence of the existing two level corporate tax
system, so a major shift in the relative treatment
of dividends and retentions should not create
significant share price increases for current
shareholders and (2) making the tax system more
neutral between retentions and distributions would
increase corporate dividend distributions and
economic efficiency. 28
As discussed above, these different views have
different theoretical implications about whether
corporations will vary payout behavior in
response to changes in the tax rate on dividends
relative to the tax rate on capital gains. The
traditional view regards differences in the tax rate
on dividends relative to the tax rate on capital
gains as a determinant of payout decisions; the
new view does not. One way to resolve the
controversy would be to determine how dividend
payout ratios vary over time with the tax rate.
Poterba has calculated that the average dividend
payout ratio (the ratio of dividends to inflationadjusted after-tax profits) for U.S. corporations
was 0.46 in the 1950s, 0.40 in the 1960s, and
0.45 in the 1970s, but increased to 0.61 in the
period from 1980 to 1986 during which the
taxation of dividends was reduced relative to the
taxation of capital gainS. 29 Although this pattern
tends to support the traditional view, it does not
provide convincing evidence, because nontax
factors also affect a corporation's dividend policy.
Statistical analysis of the determinants of dividend
payment policy is required to determine the
independent effect of dividend taxes on corporate

Economic Analyses

payout behavior, and several studies have undertaken this task. 30 The studies use different data
sources and methodologies, and estimates of the
elasticity of the payout rate with respect to dividend taxation. Nevertheless, all of the studies
conclude that dividend payout ratios do respond to
changes in the tax rate on dividends. 31 Thus, this
type of empirical evidence is consistent with the
traditional view. 32
Corporations also distribute significant
amounts of earnings to shareholders by
repurchasing shares. This is inconsistent with the
assumption underlying corporate fmancial policy
under the new view. The tax consequences of a
nondividend distribution, such as through a share
repurchase, are significant: The shareholder is
able to recover at least a portion of the cost of the
shares free of tax, and gain on the sale is taxed as
capital gain, which may be taxed at a rate lower
than the ordinary income tax rate on dividends.
Share repurchases have increased substantially
in recent years. Shoven presents data suggesting
that aggregate share repurchases increased from
$1.2 billion in 1970 to $27.3 billion in 1985 (5.4
percent and 32.7 percent of dividends, respectively). Data presented by Poterba show a similar
pattern. Share repurchases increased from $1.8
billion in 1976 to $43 billion in 1985 (5.0 percent
of dividends and 50 percent of dividends, respectively).33 Department of the Treasury calculations reveal that share repurchases rose from $5.5
billion in 1980 (10 percent of dividends) to $48.8
billion in 1985 (57 percent of dividends), peaking
at $65.8 billion in 1989 (47 percent of dividends).
In 1990, corporate share repurchases totaled
$47.9 billion (34 percent of dividends).34
To summarize, the principal distinction between the two views of corporate dividend policy
for our purposes relates to their assumptions about
nontax benefits of alternative corporate fmancial
policies. The new view assumes that dividends
offer no nontax value to shareholders relative to
retained earnings. Underlying the traditional view
is the idea that information and incentive

Economic Analyses

problems in fmancial markets make particular
corporate fmancial policies valuable for nontax
purposes. 35
The present U.S. tax system treats retained
earnings more favorably than dividends. Alternatively, given the potential nontax benefits of
dividend distributions, one might consider reversing this bias by impo~ing relatively higher taxes
on retained earnings using, for example, an
undistributed profits tax. However, this approach
would disadvantage corporations facing high costs
of external finance relative to internal fmance for
nontax reasons. Such financing cost differentials
could arise from the transaction costs of issuing
securities or from problems of asymmetric
infonnation between corporations and capital
markets. 36

118

governed by the pre-tax return needed to cover
taxes and the worldwide opportunity cost of
funds. At the same time, domestic saving depends
on the after-tax return to investor, earned from
investing at the world rate of return. Domestic
investment would thus depend on domestic COl'pOrate level taxes, although domestic saving would
depend only on domestic individual level taxes.
More broadly, in the presence of international
capital flows, the U.S. corporate income tax can
reduce incentives to invest in the United States,
even if it has a relatively small effect on saving
by U.S. citizens.

13.C METHODOLOGICAL ISSUES
IN ANALYZING THE
ALLOCATION EFFECTS OF
INTEGRATION

Effects on Savings and Investment
The corporate tax increases the tax burden on
the returns from saving and investing. Taxes on
capital income generally reduce capital fonnation.
Because of the importance of international capital
flows, which reflect the possibility of investing
abroad if U.S. investment opportunities are not
sufficiently attractive (or, conversely, thepossibility of increased investment in the United States by
foreign investors if opportunities are more attractive here), the corporate tax may have a larger
effect on U.S. investment than on U.S. savings.
The magnitudes of tax-induced distortions of
investment and savings decisions depend on (I)
the size of the wedge between pre-tax and aftertax returns and (2) the responsiveness of savers
and investors to changes in after-tax returns. The
more responsive savers and investors are to
changes in taxes, the larger the effect of a tax
wedge of a given size. 37
In a closed economy, domestic saving equals
domestic investment, and the average cost of
capital summarizes tax incentives to save as well
as to invest. International capital flows break the
equivalence of domestic saving and investment,
however. Consider the case of perfect international capital mobility. Domestic investment would be

The Importance of Using a
General Equilibrium Model
By distorting incentives, the classical corporate tax system produces an inefficient allocation
of resources. The size of the inefficiency depends
in part on how the households' and corporations'
decisions respond to changes in the tax system.
For example, the more responsive dividend
distributions are to tax considerations, the greater
the fmancial inefficiency induced by the double
tax on dividends. The analysis of the economic
effects of integration is complicated by behavioral
effects in one market that can affect other markets. For example, if the corporate tax tends to
drive capital out of the corporate sector, prices
and rates of return in the noncorporate sector are
affected.
Thus, to assess the economic consequences of
integration, one must analyze how the various
markets in the econoiny operate and interact with
each other. Economists have responded to this
challenge by constructing computer representations of the economy and using these representations to simulate how the economy would respond
to various changes in the tax system. These
representations of the economy are called computable general equilibrium (CGE) models. 38

119

The Advantage of
Using Several Models
As with all economic models, the results
generated by a CGE model depend on underlying
assumptions about how the economy operates.
Since there is no consensus regarding a single best
set of assumptions, this Report analyzes integration proposals using four different CGE models.
This procedure assures that the fmdings are not
associated with a particular modeling strategy. 39
The general equilibrium models used to
evaluate integration are detailed representations of
the U.S. economy and its actual (and proposed)
tax system. Nonetheless, all the models abstract
from some important details of both the economy
and the tax system. For example, none of the
models captures effects from changes in the
degree to which corporate preferences are passed
through to shareholders. In addition, all the
models focus on long-run results. Various transition issues, which might have important implications for economic behavior and for tax revenues,
are not considered. This focus on the long run is
correct, however, because the goal of achieving
an improved long-term performance of the economy is the prime factor motivating a concern with
integration. Nevertheless, short-run transition
effects can be substantial.

The Importance of Replacement Taxes
Given current budgetary constraints, a complete analysis of the integration prototypes requires viewing integration as a revenue neutral tax
reform, including both direct tax changes and
secondary changes required to maintain the same
total revenue yield for the government.
We do not recommend in this Report specific
changes in the tax system to fmance integration.
Nonetheless, to avoid confusing the results of the
simulation analysis by introducing changes in
government spending on goods and services, some
form of replacement taxes must be specified to
hold government revenue constant after the
introduction of the integration prototypes. In part
because of the arbitrary nature of choosing

Economic Analyses

replacement taxes, we consider two types of replacement taxes: (1) lump-sum taxes and (2)
adjustments to statutory tax rates on capital
income. Both the size of each prototype's economic effects and the ranking of prototypes by
their relative impact may depend on the form of
replacement taxes chosen.
Lump-sum taxes are hypothetical, unavoidable
taxes. That is, taxpayers cannot change their tax
liability under such a tax by changing behavior.
As a consequence, by defmition lump-sum taxes
do not distort economic decisions. Though they
are commonly used in academic studies of economic efficiency, lump-sum replacement taxes
have an important drawback for modeling integration prototypes. They can bias comparisons
among prototypes in favor of the prototype that
loses the most revenue, because the efficiency
gain from replacing distorting taxes on capital
income with nondistorting, lump-sum taxes increases with the amount of revenue that must be
replaced. This effect is important in an analysis of
integration because the prototypes have disparate
revenue costs. Compared to the actual gains that
might be realized from integration, the calculations based on lump-sum replacement taxes can
both overstate the size of the gain realized from
each revenue losing prototype and produce a
misleading ranking of prototypes. However,
because not all distortions are analyzed, e.g., the
"lock in" of capital gains and distortions of
intertemporal consumption decisions are ignored,
the lump-sum calculations do not necessarily
generate efficiency gains that exceed the true
gains. In addition, since CBIT raises revenue,
results from the lump-sum replacement may
understate its true gain.
Because of the problems with lump-sum
replacement taxes, calculations also are performed
holding government revenue constant by proportionately increasing or reducing all tax rates on
capital income. In these calculations, the tax rates
applied to corporate income, noncorporate equity
income, dividends, capital gains, interest, and
home mortgages are increased or reduced by an
amount sufficient to hold government revenue
constant at its current law level. Calculations

Economic Analyses

using scaled tax rates offer an important advantage over those based on lump-sum replacement
taxes: The scaled-tax-rate calculations raise
replacement revenue (and distribute excess revenue) by raising (or lowering) taxes that distort
economic decisions, and so reduce the bias in
favor of revenue losing tax changes. Nonetheless,
these calculations are not defmitive. In particular,
to the extent that the integration prototypes could
be made revenue neutral by more efficient tax
changes, the actual economic welfare gains may
be larger than those obtained in our scaled tax
rate calculations.
Because each of the CGE models provides
only a limited picture of the economy, the ability
of these models is to simulate the revenue consequences of each of the prototypes is somewhat
restricted. In particular, none of the models
provide an adequate treatment of the fmancial
services industry, and indeed only the Portfolio
Allocation model (described in Section l3.F) can
account for shifts in the ownership of the various
fmancial instruments issued by businesses and
governments. Even this model, however, tends to
adopt a mechanical approach to the arbitrage
possibilities possible under the different integration prototypes; in contrast, the revenue estimating models recognize that non-tax factors limit
actual shifts in asset holdings. Thus, requiring that
any loss (or gain) in revenues be made up with a
positive (or negative) replacement tax also reduces
any disparities in the results of the different
models that would otherwise arise from
differences in anticipated revenues.
The analysis presented in this Report focuses
on the scaled-tax-rate calculations, but results
based on the lump-sum replacement mechanism
also are presented.

13.D OVERVIEW OF THE
INTEGRATION PROTOTYPES
The basic features of the integration prototypes that are incorporated in the CGE models are
reviewed below. The actual prototypes are described in more detail in Chapters 2, 3, 4, and 11
of this Report. In particular, it should be noted

120

that the CGE models generally do not capture the
investor level tax imposed when distributions are
made from tax preference or foreign-taxed
income.

Distribution-Related Integration
Under the distribution-related prototypes,
corporate earnings are taxed at the corporate
level, but dividends are excluded at the shareholder level (dividend exclusion system), or shareholders receive a credit for the corporate tax paid
on distributed income (imputation credit system).
Under these prototypes, the bias against corporate
equity investment is reduced to the extent that
returns are paid out as dividends; similarly, the
relative bias against equity relative to debt fmance
is reduced to the extent earnings are distributed as
dividends. Distribution-related integration, in
principle, can create a tax bias for or against
dividends, depending on the values of the corporate tax rate, shareholder tax rate, and accrualequivalent capital gains tax rate. The prototypes
assume that the current corporate and individual
tax rates are maintained. Thus, it is likely that
distribution-related integration would increase
dividend distributions.
Dividend Exclusion. The dividend exclusion
prototype applies the corporate tax rate of 34
percent to both distributed and retained income,
but eliminates the seCond shareholder level tax on
dividends paid from earnings taxed at the
corporate level.
Imputation Credit. Relief from the corporate
income tax is provided to the extent that corporate
earnings are distributed as dividends. This relief
takes the form of a tax credit available to shareholders. The nonrefundable tax credit is calculated
at a 31 percent rate, so that it does not offset
completely the corporate income tax paid on
distributed earnings.

Shareholder Allocation Integration
The shareholder allocation prototype adopts a
"modified conduit" approach. Under a pure
conduit approach, corporations would be treated

121

like partnerships, so the corporate level tax would
be eliminated and all income and expenses would
be imputed to shareholders, who would then
include the income and expenses in their own tax
liability. Shareholders would adjust their basis in
shares upward by the amount of net income
imputed to them, and reduce their basis in shares
downward by the amount of net losses imputed to
them and by the amount distributed to them by the
corporation.
The modified conduit approach taken in the
shareholder allocation prototype differs from the
pure conduit approach. For example, the prototype imputes net income to shareholders, but not
net losses. In addition, the prototype retains the
corporate tax at a rate of 34 percent, but credits
the shareholder with the payment. This tax is
creditable against shareholder tax liability at a rate
of 31 percent, but it is not refundable. The shareholder allocation prototype reduces but does not
eliminate the distortions of organizational form
and corporate fmancial policy under current law.

Economic Analyses

corporate-individual tax rate on distributed earnings with a single tax' levied at the CBIT rate. The
same rate would apply to corporate retentions,
and since, as modeled, capital gains on CBIT
assets are exempt from taxation, CBIT would not
distort corporate dividend policy.

13.E INTEGRATION, CORPORATE
FINANCIAL POLICY, AND
THE COST OF CAPITAL

The CBIT prototype imposes a uniform tax
rate of 31 percent on returns to both debt and
equity generated by all business. Because the tax
would be collected at the business entity level,
interest and dividends would be untaxed to the
recipient. Under CBIT, interest on U.S. Government debt would remain taxable. Home mortgage
interest would remain deductible by the borrower
and taxable to the lender.

Table 13.1 illustrates how successful each
prototype is in reducing the three biases in current
law that integration is meant to reduce: the bias
against investment in corporate form, the bias
against equity fmance, and the bias against corporate dividend distributions. For individuals, all
prototypes would reduce the tax rate on distributions of corporate equity nonpreference, U. S.
source income. This ,reduction would address, at
least in part, the current law biases against the
corporate form and equity fmance. The distribution-related and CBIT prototypes would result in
a lower overall tax rate on distributed than on
undistributed corporate equity income, reversing
the current law bias against corporate dividend
distributions. However, this bias could be removed from the CBIT and dividend exclusion
prototypes by allowing shareholders to adjust
basis of stock for retained earnings through a
Dividend Reinvestment Plan (DRIP). Only the
shareholder allocation prototype, as designed,
would completely remove the bias against corporate dividend distributions.

Investments in corporate equity paying current
dividends would not be penalized under CBIT
because, as modeled, all business entities other
than very small entities, regardless of form,
would be subject to the same tax rate. Under
CBIT, neither interest nor dividends would be
deductible at the business level or taxable in the
hands of the recipient. Thus, the CBIT prototype
would equalize the tax burden on interest and
dividends. The efficiency calculations do not take
into account any compensatory tax (see Chapter 4)
on distributions from preference income. 40
Hence, CBIT would replace the combined

Absent a special provision such as the investment income tax discussed in Chaper 6, the CBIT
prototype alone reduces the current law differentials across business income sources for tax
exempt entities and foreign investors. For both
classes of income recipient, CBIT equalizes the
tax rate on all forms of business incomecorporate equity income (whether or not distributed), noncorporate equity income, and interest.
The only exception is rent and royalty income,
which would be taxed as under current law. Thus,
CBIT would address all three of the current law
biases.

CBIT

122

Economic Analyses

Table 13.1
Total U.S. Tax Rate on a Dollar of NonPreference, U.S. S.ource Income from a U.S. Business
Under Current Law and the Integration Prototypes

Type of Income

Current Law

Shareholder
Allocation
Integration

Distribution-Related Integration
Credit

Exclusion

CBIT

t;
t;
t;
t;
t;

[(1-1;)t. Hi -tt]/(l-t;m)
tc +(l-tJtg
t;
t;
t;

tc
tc +(l-tJtg
t;
t;
t;

t;m
t; m+ (l-t; m)t,
t;m
t;m

tc
tc
tc

tc
tc
to

tc
to
to

0
0

0
0

0
0

t;m
t;m
t;m
t;m

tc +(l-tJtwo
tc

tc + (l-tJt wo
to
tWN

to+(l-tJt wo
to
tWN
tWl
tWR

I. Individual Investor is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

tc + (l-tJt;
tc + (1 - tJtg
t;
t;
t;

t;

II. Tax Exempt Entity is Income Recipient
Corporate Equity:
Distributed
Undistributed
Noncorporate Equity
Interest
Rents and Royalties

tc
tc

0

0

m.

Foreign Investor is Income Recipient
Corporate Equity:
Distributed
to + (1 - tJtwo
Undistributed
tc
Noncorporate Equity
twN
Interest
tWl
Rents and Royalties
1wR
Department of the Treasury
Office of Tax Policy

twN
~

~

tWR

tWR

t;m
t;m
t;m
t;m
tWR

tc = U.S. corporate income tax rate.
t; = U.S. individual income tax rate.
t;m = Maximum U.S. individual income tax rate.
t, = U.S. effective individual tax rate on capital gains.
two, tWN , tWl , tWR = U.S. withholding rates on payments to foreigners of dividends, noncorporate equity income, business
interest, and rents and royalties, respectively. Generally varies by recipient and may be zero.

Tax Distortions in Real and Financial
Investment Decisions
Although the most succinct measure of the
economic benefits possible under each of the
integration prototypes is the estimated welfare
gain resulting from reduction or elimination of the
tax distortions affecting real and fmandal investments, this is not the most descriptive or intuitive
characterization of the effects of integration. In
this section, we thus focus more directly on the
extent of these distortions, relying on a more
commonly used measure of the impact of the tax
system on investment decisions-the cost of
capital. Although the specific results noted are
based on a specific CGE model (the augmented

Harberger model described in Section I3.F), these
results are less sensitive to the model used than
the estimates of the welfare gains, which will be
discussed in the following sections. We therefore
also defer discussion of the various CGE models
used to the following sections.
An important effect of integration is that it
would change the tax cost of real investment in
the corporate sector. We measure the effects of
taxes on investment decisions using the cost of
capital concept described in Chapter 1. Taxes on
capital income generally raise the cost of capital
above investors' required rate of return. All other
things equal, a higher cost of capital reduces
incentives to invest. The cost of capital includes

123

the effects of tax rates, depreciation allowances,
tax credits and inflation. The cost of capital also
can depend on the method of fmancing. Our
calculations are designed to be representative, and
therefore reflect a mix of debt and equity
fmancing.
As Section 13.B discusses, the size of the
distortions created by the classical corporate tax
system depends in part on whether one believes
that there are nontax benefits and costs to alternative corporate fmancial policies so that differential
taxation of fmancial arrangements can distort
fmancing decisions.
Under current law, corporations can reduce
the tax costs of investment by fmancing with debt
rather than with equity and by retaining rather
than distributing profits. Altering fmancial behavior to reduce tax liability may itself cause distortions, and raise the cost of capital. For example,
as a corporation becomes more highly leveraged,
it increases the chances that it will experience
costs associated with fmancial distress. Investors
in the corporation would require compensation for
the expected value of these costs, thereby raising
the return the corporation must earn on its investments. To capture such costs, the model augments
the traditional corporate sector cost of capital to
reflect compensation to investors for the efficiency costs of tax-induced distortions in corporate
debt and dividend policy. Tax distortions in
corporate financial policy raise the cost of capital
for corporate investment, and thereby act as a
disincentive to investment in the corporate sector.
Because economists differ on the appropriate way
to model costs of fmancial distortion, the Report
also presents effects of integration prototypes on
the cost of capital that ignore the efficiency costs
of tax distortions in corporate fmancial behavior.

Economic Analyses

patterns at the outset. More specifically, the
corporation chooses its fmancial policy to minimize its cost of capital. Consider first debt policy.
Under current law a corporation may deduct its
interest expense from its taxable income, so
interest is taxed only to the lender. In contrast,
corporate profits are taxed twice, because they are
(in general) subject to both the corporate income
tax and the individual income tax when distributed
as dividends or recognized as a capital gain on
corporate shares. Consequently, equity fmanced
corporate investment is tax disadvantaged relative
to debt fmanced corporate investment. This
difference induces corporations to increase their
use of debt. Increased use of debt, however, also
carries with it the increased possibility that the
corporation will incur costs associated with fmancial distress. In determining their leverage ratio,
corporations trade off the lower tax cost of fmancing with debt against the nontax costs of debt,
e.g., costs of fmandal distress. In contrast to
some earlier treatments, however, debt is assumed
to offer nontax benefits relative to equity (see the
discussion in Section 13.B). That is, if debt and
equity were taxed equally, we assume that corporations would continue to fmance part of their
capital stock using debt. 41
Consider now corporate dividend policy.
Under current law, the shareholder level taxes on
dividends and retained earnings differ. Dividends
are taxed as ordinary income, while retained
earnings raise share values and are taxed on a
realization basis as a capital gain. Because retained earnings benefit from the deferral of the
second level of tax, they enjoy a tax advantage
over dividends. On the other hand, corporate
distributions may be valued differently by shareholders than retentions. As a result, the determination of optimal dividend distributions reflects a
tradeoff of tax costs and nontax benefits. 42

Corporate Financial Behavior
Description of the Model

Corporate fmancial policy-which affects the
debt to asset (leverage) ratio and the dividend
payout ratio-is determined within the model
rather than assuming leverage and distribution

For modeling purposes, the corporate dividend
payout ratio divides real corporate earnings into
dividends and retentions; all purely inflationary
earnings values are assumed to come in the form
of asset appreciation and to be taxed as a capital
gain upon the sale of corporate shares. Corporations choose the real, dividend payout ratio (ratio

Economic Analyses

of real dividends to real earnings) that minimizes
the cost of equity fmanced investment. Because
the inflationary component of nominal income is
excluded, real payout ratios are higher than
conventional nominal payout ratios. Although real
dividends are the choice variable in the formal
models, nominal dividend payout ratios also are
presented in the results. Taxes are assumed not to
affect fmancial choices in the noncorporate business and the owner-occupied housing sectors of
the augmented Harberger model used in obtaining
the results presented in this section. 43

Corporate Financial Policy Under Current Law
and the Integration Prototypes
Table 13.2 shows a measure of the size of the
tax incentive for a corporation to fmance with
debt rather than with equity and to retain rather
than distribute profits. Results are presented for a
neutral tax system that does not distort these
decisions, for current law, and for each of the
integration prototypes. The table also shows
estimates of the effects of these tax incentives on
corporate borrowing and dividend distribution
policy.
Consider first corporate borrowing policy.
Under a neutral tax system, neither debt nor
equity would be tax favored, so there would be no
tax advantage to debt. The behavioral model
predicts that under such a tax system, corporations on average would finance 30 percent of their
investments using debt. In contrast to the neutral
tax system, current law discriminates against
equity finance. To cover its higher tax cost and
still offer the ultimate investor a 4 percent real
after-tax rate of return, an equity fmanced investment must earn a real pre-tax rate of return that
is 3.7 percentage points higher than would be
required were the same investment instead financed with debt. Given the assumptions used in
the calculation, this is equivalent to a 90 percent
higher real after-tax required rate of return. The
extra 3.7 percentage point return reflects debt's
tax advantage over equity and is the amount
needed to pay the higher taxes on the doubletaxed equity investment. Because of this tax
advantage to debt, or penalty to equity,

124

corporations are induced to use more debt than
under the neutral tax system and choose a 37
percent leverage ratio, 7 percentage points greater
than its value under a neutral tax regime. 44
Compared to current law, all the integration
prototypes would reduce debt's tax advantage over
equity. Consequently, all of the prototypes would
promote more efficient corporate borrowing
decisions by moving the corporate leverage ratio
closer to its undistorted value. As modeled, CBIT
eliminates differences in the taxation of debt and
equity by taxing all corporate income once at the
entity level at a 31 percent statutory rate. Under
CBIT, corporate borrowing decisions would be
undistorted by taxes. The other prototypes reduce
debt's current tax advantage over equity less
significantly.
Consider now corporate dividend policy.
Under a neutral tax system, neither dividends nor
retained earnings are tax-favored, so there is no
tax advantage to retentions, nor penalty on dividends. The behavioral model predicts that under
such a tax system, corporations would distribute
as dividends 80 percent of their real after-corporate tax profits, while retaining and reinvesting
the remaining 20 percent of real after-tax profits.
In contrast to the neutral tax system, current
law favors retained earnings over dividends.
Given the assumptions underlying Table 13.2, this
tax advantage is 1.1 percentage points. That is,
under current law, to provide an equity investor
with a real after-tax rate of return of 4 percent, a
corporation distributing all of its earnings as
dividends must earn a real pre-tax rate of return
that is 1.1 percentage points greater than that
required were the company instead to retain its
earnings. As a result of this tax distortion, corporations payout roughly 73 percent of their aftertax real profits as dividends instead of the fully
efficient 80 percent. Including inflation in the
measure of after-tax corporate profits yields a
corresponding nominal dividend payout ratio
under current law of about 43 percent.
All the integration prototypes reduce the tax
on dividends relative to that on retained earnings.

Economic Analyses

125

Table 13.2
Effect of Integration on Corporate Financial Policy!
Shareholder DistributionCurrent Allocation Related Integration
Undistorted Law Integration Credit Exclusion CBIT

A. Scaled Tax Rate Replacement
Corporate borrowing policy
Tax incentive to borrow
.000
.037
.035
.035 .000
.036
30.0% 36.6%
Leverage rati03
36.5% 36.S%
36.5% 30.0%
Corporate dividend policy
Tax penalty on dividends·
.000
.011
-.OOS .000
.000
-.OlD
Dividend payout ratio
80.0% 72.8%
Real5
80.0% 8S.9%
82.9% 80.0%
6
Nominal
42.8% 46.4%
45.9% 42.7%
42.8%
B. Lump Sum Replacement
Corporate borrowing policy
Tax incentive to borrow
.000
.037
.022
.023
.026 .000
3S.1 % 30.0%
Leverage rati03
30.0% 36.6%
34.6% 34.7%
Corporate dividend policy
Tax penalty on dividends·
-.006
-.003 .000
.000
.011
.000
Dividend payout ratio
82.4% 80.0%
80.0% 84.4%
80.0% 72.8%
Reals
6
45.5%
45.2%
42.7%
Nominal
42.8%
42.8%
Department of the Treasury
Office of Tax Policy
'Calculations are based on the augmented Hargerber Model described in section 13.F.
All calculations assume a 3.5 percent inflation rate and a 4 percent real after-tax rate of
return.
2Calculated as the difference between the cost of capital for an equity financed investment and that for a debt financed investment. The calculations assume that tax depreciation equals economic depreciation and that the corporate tax rate is the maximum statutory
rate. Debtholder and shareholder tax rates are estimates of average effective marginal
rates based on calculations from the Office of Tax Policy Individual Tax Model, adjusted
for the taxation of banks, insurance companies and tax exempt institutions.
3J'he ratio of debt to total assets.
·Calculated as the difference between the cost of capital for an investment whose return
is subject to the dividend tax and one whose return is subject to tax as a capital gain.
5'fhe ratio of (cash) dividends to after-tax real profits.
&fhe ratio of (cash) dividends to after-tax nominal profits.

Therefore, all of the prototypes encourage corporations to raise their dividend payout ratio. Both
the shareholder allocation prototype and CBIT
achieve uniformity in the taxation of real dividends and real capital gains. Under either prototype there is no tax penalty (nor tax advantage) to
dividends, so corporations would choose the
efficient 80 percent real dividend payout ratio defmed by the model. Even when the taxation of
distributions out of tax preference or foreign-taxed
income is considered (this feature is ignored in

the model results), both of
these prototypes are found
to come very close to
eliminating tax distortions
relating to payout decisions.
The distribution-related
prototypes reverse the bias
under current law. They tax
retentions less favorably
than dividends because they
provide relief from the
double tax on corporate
equity only to the extent
that earnings are distributed. This is illustrated in
Table 13.2 by a negative
tax penalty, i.e., a tax
advantage to dividends
relative to retentions for the
distribution-related prototypes. Because of this
favorable tax treatment, this
prototype encourages corporations to pay about 83
percent of real after-tax
profits (or about 46 percent
of nominal after-tax profits)
as dividends, as opposed to
the 72 percent payout ratio
(43 percent of nominal
after-tax profits) under
current law. 45

Table 13.2 also presents
calculations based on lumpsum replacement taxes. In
these calculations, all the
integration prototypes encourage (1) more efficient corporate borrowing decisions by reducing
the tax advantage to debt and the leverage ratio
and (2) higher, generally more efficient, dividend
distributions.

Cost of Capital Under
Integration Prototypes
Tables 13.3, 13.4, and 13.5 summarize the
cost of capital calculations. Current law imposes

Economic Analyses

a tax penalty on investment in the corporate sector
and financial distortions can raise this penalty.
Thus, current law can create important distortions
in the allocation of the U. S. capital stock. To
assess effects of the integration prototypes on the
current tax penalty on corporate investment,
effects on the cost of capital must be calculated.
Table 13.3 presents the effect of the current tax
system on the cost of capital among sectors
calculated both with and without the inclusion of
the costs of the fmancial distortions. Table 13.4
reports calculations of the cost of capital which
include the efficiency cost of tax distortions in
corporate fmancial policy, while the calculations
in Table 13.5 ignore such costs. The estimated
reductions in the costs of capital suggest that the
integration prototypes enhance economic efficiency relative to current law. All of the prototypes
reduce the tax bias against investment in the
corporate sector under current law, thereby
improving the allocation of capital among sectors
in the economy.
These calculations again assume that investors
require a 4 percent real, fmancing distortion
adjusted, after-tax rate of return on all investments, and that the expected inflation rate is 3.5
percent. The summary measures reported in the
table are weighted averages of more detailed
calculations of the cost of capital for each of 38
real assets, including 20 types of equipment, 14
types of nonresidential structures, residential
structures, residential and nonresidentia1land, and
inventories.
Cost of Capital Under Current Law

As noted above, there is no universally agreed
upon model of effects of fmancial distortions on
the cost of capital. The calculations in the first
column of Table 13.3 therefore ignore such
distortions. In these calculations, no premium is
imposed to compensate investors for the deviation
of the leverage and dividend payout ratios from
their undistorted values.
To illustrate the effects of the corporate
income tax on the cost of capital, Panel A shows
both the corporate and noncorporate cost of

126

capital for three paqicular investments: engines
and turbines, industrial buildings, and business
(nonresidential) land. The cost of capital for each
asset is higher if the investment is undertaken by
a corporation, because of the extra tax, than if the
investment is undertaken by a noncorporate
business. An investment in an industrial building,
for example, must earn a real return of 6.5
percent if the investment is made by a corporation, but only 5.1 percent if the investment is
made by a noncorporate business. These estimates
reflect a significant disincentive for corporate
investment; to cover extra taxes, the corporate
investment must earn 27.5 percent more than the
comparable noncorporate investment.
The summary measures in Panel B of Table
13.3 also illustrate the current tax bias against
investment in the corporate sector. On average,
the cost of capital for corporate sector investment
(5.9 percent) exceeds the cost of capital for investment in the noncorporate sector (4.9 percent).
Some of this difference, however, results from a
different mix of capital assets in the corporate and
noncorporate sector, hence only part of the
difference is due to intersectoral tax distortions.

Table 13.3
Cost of Capital Under Current Law
No Financial With Financial
Distortions
Distortions

A. Representative Assets
Engines and turbines
Corporate
Noncorporate
Industrial buildings
Corporate
Noncorporate
Business land
Corporate
Noncorporate
B. Summary Measures
Average Cost of Capital
Corporate
Noncorporate
Owner-occupied housing
Economy wide
Coefficient of Variation
Department of the Treasury
Office of Tax Policy

.051
.044

.052
.044

.065
.051

.066
.051

.061
.049

.063
.049

.059
.049
.040
.050
.155

.060
.049
.040
.051
.165

Economic Analyses

127

Owner-occupied housing has the lowest cost of
capital (4.0 percent). The return on owner-occupied housing is virtually free of tax because (1)
the imputed rental value of the housing is not
taxed to the owner, and (2) interest on debt
fmancing is includable by the lender and deductible by the owner. Unless the lender's tax bracket
is higher than the borrower's, the tax system as a
whole does not collect tax on the return on the
investment. Thus, current law discourages investment in the corporate sector in favor of investment in noncorporate enterprises, and discourages
investment in business enterprises in favor of
investment in owner-occupied housing. Overall,
capital income taxes increase the average cost of
capital for the economy as a whole (5.03 percent)
to a level greater than the investor's required
after-tax real return (4 percent). Current law may
reduce the level of resources devoted to investment and capital formation and distort the allocation of capital across sectors of the economy.
The last line in Panel B shows the coefficient
of variation for the cost of capital. The coefficient
of variation is a summary measure of the degree
of dispersion in the cost of capital. If all investments were taxed equally, all would have the
same cost of capital and the coefficient of variation would be zero. Taxes that distort investment
decisions create dispersion in the cost of capital
and raise the coefficient of variation. Under
current law, the coefficient of variation is 0.155.
The second column of Table 13.3 includes in
the corporate cost of capital a premium for tax
distortions in corporate borrowing and dividend
policies. Tax distortions in corporate fmancial
policies raise the cost of capital for corporate
sector investments by approximately 0.1 percentage point, compared to the prior calculations
which ignore fmancial distortions, while leaving
unchanged the cost of capital for investments in
the noncorporate sector and in owner-occupied
housing. Including fmancial distortions, therefore,
increases the tax-induced disparity in the cost of
capital between corporate and other investments.
With fmancial distortions, current law's
coefficient of variation in the cost of capital is

O. 165, greater than the 0.155 coefficient of
variation obtained when fmancial distortions are
ignored. By raising the cost of investing in the
corporate sector, fmancial distortions also raise
slightly the overall cost of investing in the
economy.
Cost of Capital Under the
Integration Prototypes

Tables 13.4 and 13.5 present summary measures of the cost of capital under current law and
each of the integration prototypes, with and
without fmancial distortions, respectively. Table
13.4 presents calculations assuming scaled tax
rates for replacement revenue (Panel A), and
lump-sum replacement taxes (Panel B). All the
calculations in Table 13.4 assume that corporations vary their borrowing and dividend distributions in response to changes in tax rates, and
include a premium for tax-induced distortions in
corporate borrowing policy.
Table 13.4 presents results from calculations
that include the efficiency cost of tax distortions
in corporate fmancial policy. In these calculations
the integration prototypes change both the corporate leverage ratio and dividend payout ratio from
their values under current law, but also change
the magnitude of the associated fmancial distortions. In the scaled-tax-rate calculations, statutory
tax rates on capital income are increased or
decreased proportionately to hold the overall tax
burden on investment at its current level. Each
prototype reduces the corporate cost of capital
toward the lower average for the rest of the
economy, thereby reducing the coefficient of
variation below its current law level. CBIT reduces the coefficient of variation in the cost of capital
most significantly. Compared to current law,
CBIT reduces the coefficient of variation in the
cost of capital by more than one-third, from 0.165
to 0.104. The other prototypes produce a smaller
reduction in the coefficient of variation, a reduction that is nearly the same for each prototype.
Thus, in these calculations, CBIT provides the
greatest incentive for an efficient allocation of
physical capital. 46

Economic Analyses

128

Table 13.4
The Cost of Capital
Under Current Law and the Integration Prototypes:
With Financial Distortions

In the scaled tax rate
calculations, benefits from
CBIT still exceed those of
other prototypes, but
because CBIT reduces
Shareholder Distribution-Related
fmancial distortions more
Current Allocation
Integration
than other prototypes,
Law Integration Credit Exclusion CBIT
there is less difference
between CBIT and the
A. Scaled tax rate replacement
other prototypes in Table
Average cost of capital
13.5 than in Table 13.4 .
.053
.057 .058
.057
.060
Corporate sector
Nonetheless, the results in
.054
.052 .051
.052
Noncorporate sector
.049
.
042
.040
.040
.040
the two tables are similar .
Owner-occupied housing sector .040
.050
.051
.051
.051
Economy wide
.051
In both tables, each proto.104
.144 .148
.143
Coefficient of variation
.165
type reduces the extra tax
cost of investing in the
B. Lump sum replacement
corporate sector, therefore
Average cost of capital
encouragmg a more
.056
.052 .054
.060
.052
Corporate sector
efficient allocation of
.057
.049 .049
Noncorporate sector
.049
.049
capital. Additionally, in
.043
.040 .040
Owner-occupied housing sector .040
.040
both tables, shareholder
.053
.048 .049
Economy wide
.051
.048
allocation leads to the
.123
.111 .120
.107
Coefficient of variation
.165
greatest reduction in the
Department of the Treasury
Office of Tax Policy
coefficient of variation in
the calculations based on
lump-sum replacement, while CBIT reduces the
The results based on lump-sum replacement
coefficient of variation most in the calculations
taxes presented in Panel B are similar to those in
based on the scaled tax rate replacement
Panel A. All prototypes reduce current tax distortions in the allocation of capital, particularly by
mechanism.
reducing taxes on corporate investment relative to
investment elsewhere in the economy. Thus, all
13.F INTEGRATION AND THE
ALLOCATION OF
prototypes lower the coefficient of variation in the
cost of capital. The lump-sum replacement mechaRESOURCES
nism, however, allows all of the prototypes except
CBIT to benefit from lower taxes on capital
This section reviews the simulated effects of
income. Consequently, the shareholder allocation
each integration prototype on the allocation of
prototype most significantly reduces the coeffiresources and economic efficiency. Results from
cient of variation, and provides the greatest incenthree models are presented. The first is a
tive for an efficient allocation of physical capital.
Harberger-type CGE model modified to account
for tax distortions in corporate fmancial policies.
Table 13.5 presents cost of capital calculations
The two alternative CGE models respond to
important limitations of the Harberger-type modthat abstract from the costs of tax distortions in
el. Overall, the cost of capital calculations providcorporate fmancial policy. In those calculations,
ed in the preceding section are reinforced by the
fmancing is unaffected by tax policy changes, so
results from the more comprehensive CGE
corporations have a 73 percent real dividend
calculations.
payout ratio and a 37 percent leverage ratio under
current law as well as under the integration
prototypes.

Economic Analyses

129

Table 13.5
The Cost of Capital
Under Current Law and The Integration Prototypes:
No Financial Distortions
Shareholder Distribution-Related
Current Allocation
Integration
Law Integration Credit Exclusion CBIT

A. Scaled tax rate replacement
Average cost of capital
Corporate sector
Noncorporate sector
Owner-occupied housing sector
Economy wide
Coefficient of variation

.059
.049
.040
.050
.155

.055
.052
.040
.050
.137

.056
.052
.040
.050
.138

.057
.051
.040
.051
.143

.053
.054
.042
.050
.103

B. Lump sum replacement
Average cost of capital
Corporate sector
.059
.051
.052
.053
.056
Noncorporate sector
.049
.049
.049
.049
.057
Owner-occupied housing sector .040
.040
.040
.040
.043
Economy wide
.050
.047
.048
.048
.053
Coefficient of variation
.155
.103
.108
.115
.123
---...:::::.::::::.:..:::=:....:.:.....:=..:===--------=.::....:.....------------Department of the Treasury
Office of Tax Policy

The Augmented Harberger Model
Model Description
In Harberger's original model, the corporate
tax induces capital to leave the corporate sector,
a migration that continues until after-tax returns
are equalized in the corporate and noncorporate
sectors. Through this adjustment process the
burden of the corporate tax is spread to owners of
noncorporate capital and possibly to labor. 47 The
corporate tax thus causes too much capital to be
allocated to the noncorporate sector and not
enough to the corporate sector, so that an ineffIcient allocation of resources results.
The fIrst model used to study the integration
prototypes is an augmented version of Harberger's
original contribution. 48 While the original
Harberger model had only two sectors, the augmented model embodies a richer depiction of the
economy. It has 18 industries and 35 different
types of assets, and includes both intennediate and

fmal goods. In the original
model, the total supplies of
capital and labor were
fIxed. In the augmented
model, the supplies of labor
and capital can vary depending on their rates of
return, but in the simulations the supply of capital
is held constant. Investment
decisions are based on the
cost of capital described in
the preceding section .
Harberger's approach
implicitly assumed that
corporate fInancial policy
was unaffected by the tax
system. In contrast, the
augmented model incorporates the model of fmancial

behavior discussed above,
and so allows the tax systern to influence corporate
borrowing and dividend
policies. Allowing fInancial
decisions to be influenced by the tax system is
particularly important in the present context,
because previous research has suggested that
ignoring tax-induced distortions in fmancial
behavior can lead to substantial underestimates of
the effIciency costs of the classical income tax
system. 49
As emphasized earlier, the simulation of each
integration prototype holds constant real government spending. As in the discussion of the cost of
capital, we emphasize calculations using scaled
tax rates, though calculations based on lump-sum
replacement taxes are presented for comparison.
The method of estimation proceeds by comparing a single equilibrium representing current
law with a corresponding equilibrium under each
integration prototype. The simulations are static,
in the sense that they abstract from savings and
growth issues by holding constant the economy's
capital stock in the face of each prototype's tax
changes. Thus, the model captures effects from

130

Economic Analyses

the prototype's shifts in the allocation of real
resources across sectors and industries and from
changes in corporate financial decisions, but
abstracts from any tax-induced changes in saving
and capital formation. Since integration generally
is perceived as a way to improve the static allocation of real resources and to improve corporate
fmancial policy, this is appropriate. 50

Simulation Results
Table 13.6 presents the results of simulations
that include the costs of tax distortions in

corporate fmancial policy, and Table 13.7 presents results of calculations excluding such costs.
The results in Table 13.6 that include the costs of
fmancial distortions illustrate most broadly the
costs of tax distortions under current law.
The first three rows of Panel A show each
prototype's effect on the allocation of capital,
based upon the scaled-tax-rate replacement mechanism. In these calculations, CBIT generates the
largest changes in capital allocation. CBIT increases the corporate share of capital by almost 5
percentage points, and decreases the share of

Table 13.6
General Equilibrium Results, Augmented Harberger Model:
With Financial Distortions
Shareholder
Allocation
Integration

Distribution-Related Integration
Credit

Exclusion

CBIT

A. Scaled tax rate replacement
Percentage change in capital allocation I
Corporate sector
2.6
2.3
1.7
4.6
Noncorporate sector
-2.7
-2.4
-1.8
-3.8
Owner-occupied housing
0.1
0.1
0.1
-0.8
Annual change in welfare2, by source of change, as a
percentage of consumption (and as a percentage of tax
revenue from corporate capital)
Consumption
0.10 (2.38)
0.10 (2.38)
0.08 (1.90)
0.20 (4.76)
Corporate debt policy
-0.00 (-0.00)
-0.00 (-0.00)
-0.00 (-0.00)
0.17 (4.05)
Corporate dividend policy
0.03 (0.71)
0.01 (0.24)
0.03 (0.71)
0.03 (0.71)
Total
0.13 (3.09)
0.11 (2.62)
0.11 (2.62)
0.40 (9.52)
B. Lump sum replacement
Percentage change in capital allocation I
Corporate sector
3.4
3.2
2.6
4.3
Noncorporate sector
-2.5
-2.4
-1.9
4.2
Owner-occupied housing
-0.9
-0.8
-0.6
-0.1
Annual change in welfare 2, by source of gain, as a percentage of consumption (and as a percentage of tax
revenue from corporate capital)
Consumption2
0.24 (5.71)
0.20 (4.76)
0.23 (5.47)
0.10 (2.38)
Corporate debt policr
0.08 (1.90)
0.06 (1.43)
0.07 (1.67)
0.16 (3.81)
Corporate dividend policy3
0.03 (0.71)
0.03 (0.71)
0.02 (0.48)
0.03 (0.71)
Total
0.29 (6.90)
0.35 (8.33)
0.29 (6.90)
0.32 (7.62)
Department of the Treasury
Office of Tax Policy
IThese represent changes in each sector's share of total private capital.
2Welfare changes from improvements in real resource allocation are measured as changes in
"expanded" national income, i.e., changes in national income plus changes in the value of leisure.
3Welfare changes from changes in financial policies are measured using an excess burden function
derived from investors' preferences for debt and for equity.

131

Economic Analyses

Table 13.7
General Equilibrium Results, Aup:tented Harberger Model:
No Financial Distortions
Shareholder
Allocation
Integration

Distribution-Related Integration
Credit

Exclusion

CBIT

A. Scaled tax rate replacement
Percentage change in capital allocation'
Corporate sector
2.5
2.1
1.6
4.1
Noncorporate sector
-2.6
-2.2
-1.7
-3.S
Owner-occupied housing
-0.6
0.1
0.1
0.1
2
Annual change in welfare as a percentage
of consumption (and as a percentage of tax
revenue from corporate capital)
0.08 (1.9S)
0.08 (1.71)
0.07 (1.71)
0.17 (4. IS)
B. Lump sum replacement
Percentage change in capital allocation'
Corporate sector
3.3
2.9
2.4
3.8
Noncorporate sector
-2.4
-2.2
-1.8
-3.9
Owner-occupied housing
-0.6
-0.8
-0.7
0.1
Annual change in welfare2 as a percentage
of consumption (and as a percentage of tax
revenue from corporate capital)
0.20 (4.88)
0.17 (4.15)
0.21 (S.12)
0.07 (1.71)
Department of the Treasury
Office of Tax Policy
'These represent changes in each sector's share of total private capital.
2Welfare changes are measured as changes in "expanded" national income, i.e., changes in national income plus
changes in the value of leisure.

capital allocated to other sectors by an equivalent
amount. The other prototypes stimulate somewhat
smaller changes in the allocation of capital across
sectors.
The next set of calculations in Panel A represents effects on economic well-being resulting
from adoption of each prototype. Economic
welfare effects are shown separately for (l) the
gain caused by the improved consumption choices
made possible by integration's improvement in the
allocation of real resources, and (2) the gain due
to improved corporate fmancial policy. These
welfare gains do not reflect gains (or losses)
arising from changes in savings and economic
growth attributable to the prototypes. Two welfare
measures are presented. The fIrst measure
expresses the welfare gain as a percentage of
consumption under current law, and can be
interpreted as the percentage gain in annual
consumption possible under each prototype once

the economy fully adjusts to the change in law
and reaches its new equilibrium. The second
measure (in parentheses) expresses the welfare
gains as a percentage of the annual tax revenue
from corporate capital income.
In this model, the annual economic welfare
gains from the improved allocation of resources
range from 0.08 to 0.20 percent of current consumption or 1.9 to 4.8 percent of tax revenue
from corporate capital income (equivalent to a
range of about $2.3 to $5.7 billion per year).
CBIT produces welfare gains at least twice as
large as that generated by the other prototypes.
The other integration prototypes generate a
smaller improvement from a more effIcient allocation of real resources, equivalent to about 0.10
percent of current consumption for each. Thus,
although these prototypes appear structurally
different, from an economic perspective they may

Economic Analyses

132

be quite similar. Indeed, this result can be anticipated from the above discussion of the cost of
capital, which showed that these prototypes had
nearly identical effects on the coefficient of
variation in the cost of capital.
The next simulated economic welfare gain
represents welfare effects of changes in corporate
debt policy. All the integration prototypes lower
the corporate leverage ratio. CBIT, however,
completely eliminates the tax bias against equity,
thereby producing the largest gain, equivalent to
0.17 percent of consumption, or more than 4
percent of tax revenue from corporate capital
(about $4.8 billion). The dividend exclusion and
shareholder allocation integration prototypes
produce only negligible gains from this source.
Table 13.6 also shows the simulated economic
welfare effects of changes in corporate dividend
policy. With the exception of the imputation credit
prototype, the prototypes yield welfare gains in
this respect that are equivalent to an annual
increase in consumption of 0.03 percent (or 0.71
percent of tax revenue from corporate capital).
Welfare gains accompanying the imputation credit
prototype are smaller at this margin.
Combining the economic welfare effects from
changes in debt policy and changes in dividend
policy, shows that all three prototypes improve
overall corporate fmanciaI policy. These gains are
largest for CBIT. By eliminating distortions in
corporate fmanciaI policy, CBIT produces a
welfare gain equivalent to 0.20 percent of consumption, or 4.76 percent of tax revenue from
corporate capital. The shareholder allocation
prototype and the dividend exclusion prototype
produce much smaller welfare gains from improvements in corporate fmancial policy, roughly
equivalent to 0.03 percent of consumption, (0.71
percent of tax revenue from corporate capital).
Perhaps the most striking feature of these results
is that the CBIT prototype's welfare gains from
improved corporate fmancial policy are as large
as the welfare gains from improved real resource
allocation.

The total improvement in economic welfare
ranges from a high under CBIT of 0.40 percent of
consumption to a low for the imputation credit
and dividend exclusion prototypes of O. 11 percent
of consumption. By contributing most significantly to the efficient allocation of real resources and
to the promotion of efficient corporate fmancial
choices, CBIT stimulates the largest gains in
economic welfare.
Panel B presents results based on lump-sum
replacement taxes. In some respects these calculations are similar to those in Panel A. For example, in both set of calculations, the integration
prototypes expand modestly the size of the corporate sector relative to the rest of the economy. In
addition, in both sets of calculations, all prototypes generate modest economic welfare gains. In
the calculations based on lump-sum replacement
taxes, however, all prototypes except CBIT show
welfare gains from reducing taxes on capital
income (and replacing them with more efficient
lump-sum taxes). In contrast, as modeled, CBIT
raises distorting taxes on corporate capital income
and distributes the excess revenue to consumers
through lump-sum rebates. Consequently, CBIT
compares less favorably with the other prototypes
in the lump-sum calculations than in the scaled tax
rate calculations, although this result is largely an
artifact of the revenue estimate for CBIT obtained
from this model. In the lump-sum calculations,
the shareholder allocation prototype produces the
largest improvement in economic well being,
roughly equivalent to an annual gain of 0.35
percent of consumption.
Table 13.7 presents results of calculations that
do not include the cost of tax-induced distortions
in corporate fmancial policy. In those calculations, the prototypes do not change fmanciaI
variables from current law values, and fmanciaI
distortions do not create welfare costs.
The calculations in Table 13.7 are similar in
several respects to those reported in Table l3.6.
All prototypes continue to shift capital into the
corporate sector and produce overall gains in

Economic Analyses

133

welfare, measured relative to annual consumption
or annual tax revenue from corporate capital. The
shareholder allocation prototype increases
economic welfare the most under the lump-sum
replacement taxes, while CBIT increases economic welfare the most under the scaled-tax
replacement approach.

The Mutual Production Model

respects. First, greater substitution exists between
corporate and noncorporate activity in the MPM
than in the augmented Harberger model. Second,
the MPM assumes a ftxed labor supply, while the
augmented Harberger model allows labor supply
decisions to vary depending upon the after-tax
wage rate. Consequently, one would expect
similar, but not necessarily identical, results from
the two models. Results from the MPM are
presented in Table 13.8.

Model Description
Simulation Results
An important problem with models based on

the original Harberger approach is the implicit
assumption that if a commodity is produced in the
corporate sector, it also cannot be produced in the
noncorporate sector, and vice versa. This conflicts
with empirical evidence of such coexistence. To
address this issue, we use a Mutual Production
Model (MPM) , in which corporate and noncorporate businesses coexist in industries because
each has certain advantages: corporate businesses,
which are relatively large, have the advantage of
economies of scale, and noncorporate businesses,
which are smaller, have the advantage of more
effective managerial skill. 51
This approach has been incorporated in a
large-scale model that contains twelve sectors and
allows for the production of capital goods as well
as intermediate goods (goods used in other businesses). Each industry produces with managerial
input, labor input, and a fIxed capital composite
of 31 different assets. The model is a closed
economy model characterized by a representative
consumer, a fIxed labor supply, and a fIxed
capital stock. Financial decisions about corporate
debt to equity and dividend payout ratios are
affected by the tax system.
In many ways, the analysis of resource allocation in the modilled MPM is structurally similar
to the augmented Harberger model discussed
above. 52 For example, both are disaggregated,
competitive models, which base decisions about
capital allocations on the user cost of capital. In
addition, both are closed economy models that
abstract from international capital flows. The
models differ, however, in at least two key

Panel A of Table 13.8 presents the results of
calculations based on the scaled-tax-rate adjustment approach. The fIrst rows of panel A show
the percentage change in the share of total capital
used in each of the corporate, noncorporate
business, and owner-occupied housing sectors,
respectively. All of the prototypes shift capital
(and other resources) into the corporate sector.
CBIT's 7.1 percentage point increase in the
corporate sector's share of total capital would be
the largest shift, while the dividend exclusion
prototype's 2.9 percentage point increase would
be the smallest. For all prototypes, the resource
flow into the corporate sector come primarily
from a contraction of the noncorporate business
sector, but owner-occupied housing also would
decline slightly in the CBIT and imputation credit
prototypes.
The next two rows of panel A illustrate the
change in corporate fmancial policy attributable to
each prototype. As a point of reference, a 5
percentage point reduction in the corporate leverage ratio would eliminate current law's distortion
in this model. In these calculations, CBIT eliminates the tax incentive to borrow, and thus reduces the corporate leverage ratio to its undistorted
level. The shareholder allocation prototype
achieves only a slight reduction. In contrast, the
distribution-related prototypes do not improve
corporate borrowing policy in this model. 53
Both the shareholder allocation and CBIT
prototypes eliminate the tax penalty on dividends.
Consequently, under both prototypes, corporations
increase their real dividend payout ratio by 9

Economic Analyses

134

Table 13.8
General Equilibrium Results, Mutual Production Model:
With Financial Distortions

Shareholder
Allocation
Integration

Distribution-Related Integration
Credit

Exclusion

CBIT

A. Scaled tax rate replacement
Percentage change in capital allocation I
5.5
2.9
7.1
Corporate sector
4.3
-3.0
-6.7
-5.3
Noncorporate sector
-4.5
0.1
-0.4
-0.2
0.2
Owner-occupied housing
Percentage change in financial policy
relative to current law
-5.0
1.0
Corporate debt to asset ratio
-1.0
2.0
9.0
10.0
Real dividend payout ratio
16.0
9.0
2
Annual change in welfare , by source of
change, as a percentage of consumption
(and as a percentage of tax revenue from
corporate capital)
0.43 (5.69)
Consumption
0.27 (3.57)
0.37 (4.90)
0.22 (2.91)
Corporate debt policy
0.06 (0.79)
-0.22 (-2.91)
-0.10 (-1.32)
0.23 (3.05)
Corporate dividend policy
0.07 (0.94)
0.01 (0.13)
0.07 (0.93)
0.07 (0.93)
Total
0.40 (5.30)
0.16 (2.12)
0.19 (2.52)
0.73 (9.67)
B. Lump sum replacement
Percentage change in capital allocation 3
Corporate sector
6.1
4.2
6.2
7.2
Noncorporate sector
-5.0
-5.0
-3.5
-6.7
Owner-occupied housing
-1.1
-1.2
-0.7
-0.5
Percentage change in financial policy
relative to current law
Corporate debt to asset ratio
-3.0
-1.0
-1.0
-5.0
Real dividend payout ratio
9.0
14.0
10.0
9.0
Annual change in welfare2 , by source of
gain, as a percentage of consumption (and
as a percentage of tax revenue from
corporate capital)
Consumption2
0.54 (7.15)
0.50 (6.62)
0.39 (5.16)
0.44 (5.83)
Corporate debt policy3
0.11 (1.46)
0.11 (1.46)
0.07 (0.93)
0.23 (3.04)
Corporate dividend policy3
0.07 (0.93)
0.04 (0.53)
0.07 (0.93)
0.07 (0.93)
Total
0.72 (9.54)
0.65 (8.61)
0.53 (7.02)
0.74 (9.80)
Department of the Treasury
Office of Tax Policy
IThese represent changes in each sector's share of total private capital.
2The model measures the welfare change from an improved allocation of real resources as the
compensating variation of the change from current law to integration. Compensating variation is a
measure of the dollar value of the change in consumer's utility as a result of integration.
3Welfare changes from changes in financial policies are measured using an excess burden function
derived from investor's preferences for debt and for equity.

135

percentage points to the undistorted value calibrated in the model. COlporations also increase their
dividend payout ratio under the two distributionrelated prototypes. Because distribution-related
prototypes relieve the cOlporate level tax on
corporate equity only to the extent profits are
distributed, cOlporations actually pay an
inefficiently large fraction of their earnings as
dividends under these prototypes. Nonetheless,
compared to current law, both prototypes encourage cOlporations to reduce the difference between
their actual payout ratio and the undistorted
payout ratio.
The fmal four rows of Panel A present each
prototype's welfare changes in total, and a decomposition by the source of change. Annual welfare
gains are expressed as a percentage of consumption under current law and as a percentage of
current revenue from cOlporate capital income (in
parentheses). By improving the allocation of
resources, all of the prototypes generate improved
consumption choices, but CBIT has the largest
improvement, equivalent to 0.43 percent of
consumption. The dividend exclusion prototype
yields the smallest improvement, equivalent to
0.22 percent of consumption.
The shareholder allocation and CBIT prototypes improve corporate borrowing policy. CBIT
generates an economic welfare gain equivalent to
0.23 percent of consumption. While the welfare
gain accompanying the shareholder allocation
prototype is smaller in this dimension, the distribution-related prototypes encourage corporations
to increase borrowing slightly above levels under
current law and thereby generate a small welfare
loss.
The shareholder allocation and CBIT prototypes both eliminate the tax distortion in cOlporate
dividend policy, and in so doing generate a small
welfare gain equivalent to 0.07 percent of consumption. Although the distribution-related prototypes encourage finns to distribute an inefficiently
large fraction of their profits as dividends, by
inducing fmns to move the payout ratio closer to
its undistorted level, both generate welfare gains
at this margin.

Economic Analyses

In total, in the scaled-tax-rate calculations the
prototypes produce annual economic welfare gains
ranging from a low of under 0.2 percent of
consumption for distribution-related integration to
a high of 0.73 percent of consumption for CBIT.
In these calculations, CBIT generates as large or
larger welfare gains than the other prototypes in
every category.
Panel B shows calculations based on lump-sum
replacement. In these calculations, all of the
prototypes promote more efficient consumption,
corporate borrowing, and corporate dividend
policies. The other prototypes compare more
favorably to CBIT than in panel A because, as
modeled, CBIT would raise taxes on capital
income, while the other prototypes would lower
capital income taxes. Consequently, although in
part an artifact of the modeling, the shareholder
allocation prototype would generate an annual
welfare gain equivalent to 0.72, almost as large as
that under CBIT (0.74 percent of consumption).
Annual welfare gains for the imputation credit and
dividend exclusion prototypes would be 0.65 and
0.53 percent of consumption, respectively.

Portfolio Allocation Model
Model DeSCription
Both the augmented Harberger model and the
MPM capture tax distortions in the allocation of
physical capital among the corporate, noncOlporate, and owner-occupied housing sectors.
Both also capture tax distortions in the supply of
corporate debt and dividends. Neither model,
however, is designed to capture tax distortions in
the allocation of fmancial assets across households. The portfolio allocation (PA) model addresses this shortcoming by focusing on tax
distortions in household portfolio decisions. 54
The P A model combines an allocation of capital
across sectors reflecting production characteristics
and consumer preferences with an allocation of
capital across investors and forms of investment
through a portfolio mechanism. In the PA model,
real and fmancial variables are determined simultaneously, and taxes can distort both real and
fmancial decisions.

Economic Analyses

136

The PA model explicitly links individual
fInancial decisions with real variables in the
economy. Households and pension funds acquire
securities in a manner consistent with their riskreturn preferences, while businesses and the
government sector issue securities to meet their
demands for capital. Individuals allocate their
wealth among corporate equity, noncorporate
equity, rental housing, owner-occupied housing
equity, durable goods, tax-exempt bonds, and
taxable debt according to the riskiness as well as
the after-tax rate of return on these assets. Individual households are distinguished by income and
wealth levels, tax filing status, and whether they
rent or own their homes.

Simulation Results
Results from the PA model are displayed in
Tables 13.9 and 13.10. As with the other models,
two sets of calculations are perfonned. In the fIrst
set of calculations, presented in Table 13.9,
statutory tax rates on capital income are increased
or decreased proportionately to satisfy the constraint that revenues remain constant. In an alternative set of calculations, presented in Table
13.10, lump-sum taxes or rebates are used to
satisfy the equal yield constraint.

Scaled Tax Replacement. Table 13.9 presents
integration's aggregate effects on the allocation of
real and fmancial capital and on corporate fmancial policy. The top panel shows changes in the
allocation of real capital. In the portfolio allocation model, all of the prototypes shift capital into
the corporate sector. The CBIT prototype produces the largest increase in corporate capital, equivalent to 2.5 percent of total U.S. real capital,
followed by shareholder allocation integration (1.7
percent expansion) and then by distribution-related
integration (1.6 percent expansion for the dividend
exclusion prototype). In all prototypes, the flow
of capital into the corporate sector comes from a
contraction of other sectors. The prototypes
improve the allocation of capital within the business sector as well as between the business and
nonbusiness sectors.

The middle panel of Table 13.9 presents
changes in holdings of fmancial assets, divided
into changes in households' holdings and changes
in pension funds' holdings. 55 In the PA model,
households can make fmancial investments in
corporate stock, noncorporate equity interests, and
debt. All of the prototypes induce households to
raise their holdings of corporate stock. CBIT
produces the largest such shift, equivalent to 6.5
percent of total wealth, compared to about 3 to 4
percent for the other prototypes. In addition, all
prototypes reduce households' holdings of taxable
bonds. The shareholder allocation and distribution-related prototypes produce a reduction equivalent to between 2.0 percent and 2.5 percent of
total wealth. CBIT generates a larger reduction,
and the household sector becomes a net borrower
in the taxable debt market. Traditional tax-exempt
debt holdings are largely unaffected by integration
(except under CBIT). CBIT debt, which is taxexempt to the lender, accounts for 11.6 percent of
total wealth. To a large extent, CBIT debt substitutes for taxable debt under current law. Thus, it
is useful to compare the sum of taxable and CBIT
debt holdings under CBIT and current law. Combining CBIT's 14.8 percent reduction in taxable
debt with the 11.6 percent of total wealth that
corresponds to CBIT debt shows that CBIT
reduces households' direct holdings of formerly
taxable debt by 3.2 percent of total wealth. The
other prototypes reduce direct household holdings
of currently taxable debt by an amount equivalent
to 2.0 to 2.5 percent of private wealth. Combining all types of debt shows that CBIT generates a
larger reduction in direct debt holdings by households, equivalent to 4.3 percent of total wealth
wbile the other prototypes generate a smaller
reduction, equivalent to between 2.0 and 2.6
percent of wealth. Finally, note that holdings of
noncorporate capital decline under all the
integration prototypes. 56
Pension funds' portfolio shifts are the reverse
of household portfolio shifts. In the PA model,
pension funds allocate assets between debt and
corporate equity. By lowering the tax burden
households face on corporate equity, but not
extending the tax reduction to pension funds, all
prototypes induce pension funds to reduce

Economic Analyses

137

Table 13.9
The Effect of Integration on the Allocation of
Physical Capital, Wealth, and Corporate Financial Policy
Results from the Portfolio Allocation Model
(Scaled Tax Rate Replacement)

Prototype

Shareholder
Allocation
Integration

Distribution-Related
Integration
Credit Exclusion
CBIT

A. Change in the Allocation of Physical Capital
(as a percent of total physical capital)
Corporate Business
2.5%
1.7%
1.3%
1.6%
Noncorporate Business
-0.1 %
-0.1 % -0.1% -0.1%
-0.3%
-0.3%
-0.4% -0.4%
Noncorporate Rental Housing
-0.4%
-0.5% -0.5%
Total Noncorporate Capital
-0.4%
State and Local Government
-0.1 % -0.1%
-0.1%
0.0%
Owner-occupied Housing
-0.7%
-0.4%
-0.5% -0.8%
Consumer Durables
-0.5%
-0.5%
-0.5% -1.2%
Total Household Capital
-1.3%
-0.9%
-1.0% -2.0%
B. Change in The Allocation Of the Household Sector's Portfolio
(as a percent of total wealth)
3.2%
4.0%
6.5%
3.9%
Corporate Stock
Debt
-2.0%
-2.5% -14.8%1
Taxable to Investors
-2.3%
Not Taxable to Investors
-0.1 % -1.2%
-0.1 %
-0.1%
Traditional Tax-Exempt
0.0%
0.0% 11.6%
CBIT
0.0%
-0.1 % -0.1% 10.5%
-0.1 %
Total Tax-Exempt
-2.0%
-2.6% 4.3%
Total
-2.4%
0.0%
-0.1 % -0.1 %
-0.1 %
Noncorporate Business
-0.4% -0.4%
-0.2%
Noncorporate Rental Housing
-0.2%
-0.3%
-0.5%
-0.4%
-0.3%
Noncorporate Total Capital
-0.4%
-0.7%
-0.4%
-0.6%
Owner-occupied Housing
-0.5%
-1.1%
-0.4%
-0.5%
Consumer Durables
-0.9%
-1.8%
-0.8%
-1.1 %
Total Household Capital
Pensions
-2.5%
-0.3%
-1.7%
-2.0%
Corporate stock
1.7%
2.5%
0.3%
2.0%
Debt
C. Change in Corporate Financial Policy (in percentage points)
Leverage Ratio
-3.2%
-2.7%
-2.3% -14.7%
(Nominal) Dividend Payout Ratio
3.2%
3.3%
3.8%
3.0%
Department of the Treasury
Office of Tax Policy
lThe household sector goes from a net lender in the market for taxable bonds
under current law to a net borrower under CBIT.

corporate equity holdings and increase debt holdings. Consequently, for the economy as a whole,
the shift out of debt and into equity is less
pronounced than the change for the household
sector alone. Overall, in their effects on households' direct holdings plus pension fund holdings,

the distribution-related
and shareholder allocation
prototypes stimulate a
move into corporate equity equivalent to between
1.5 and 1.9 percent of
total wealth. CBIT generates a much larger net
increase in holdings of
corporate shares, equivalent to 6.2 percent of total
wealth. The total shift
from debt is equivalent to
-4.0 percent of total
wealth under CBIT, and
to between -0.1 and -0.4
percent of total wealth for
the other prototypes. 57
The bottom panel of
Table 13.9 presents each
prototype's effect on
corporate borrowing and
dividend policy. All prototypes encourage corporations to use less debt,
but CBIT generates a 14.7
percentage point reduction
in the corporate leverage
ratio, much larger than
the reduction generated by
the other integration
prototypes. Dividend
payout ratios increase in
all cases (by between 3.0
and 3.8 percentage
points); not surprisingly,
the largest such increase
accompanies the dividend
exclusion prototype.

Lump-Sum Tax Replacement. Table 13.10 summarizes PA model
results illustrating integration's aggregate effects
on the allocation of real and fmancial capital and
on corporate fmancial policy assuming lump-sum
taxes are used to maintain revenue neutrality. The
allocational impacts of integration are qualitatively
similar to those based on scaled tax rate

Economic Analyses

138

Table 13.10
Summary of the Effects of Integration on
Real and Financial Decisions:
Results from the Portfolio Allocation Model
(Lump Sum Replacement)

Prototype

Shareholder
Allocation
Integration

Distribution-Related
Integration
Credit Exclusion
CBIT

A. Change in the Allocation of Physical Capital
(as a percent of total physical capital)
Corporate Business
2.8%
2.6%
2.6%
2.8%
Total Noncorporate Capital
-0.3%
-0.3%
-0.3%
-0.4%
State and Local Government
-0.1 %
-0.1 %
-0.1%
0.0%
Total Household Capital
-2.3%
-2.2%
-2.2%
-2.3%
B. Change in the Allocation of the Household Sector's Portfolio
(as a percent of total wealth)
6.5%
5.3%
5.5%
Corporate Stock
6.2%
4.1%
-3.1%
-3.3%
Debt
-3.8%
-0.4%
Total Noncorporate Capital
-0.3%
-0.2%
-0.2%
-2.1%
-1.9%
Total Household Capital
-2.0%
-1.9%
Pensions
-1.9%
-0.2%
Corporate Stock
-2.3%
-2.0%
1.9%
0.2%
2.0%
Debt
2.3%
C. Change in Corporate Financial Policy
(in percentage points)
Leverage Ratio
-8.3%
-7.3%
-6.9% -16.6%
Nominal Dividend Payout Ratio
3.25
3.4%
3.8%
3.0%
-D-ep-art-m-e-nt-o-f-th-e-=T-r=-eas-u-ry--------------Office of Tax Policy

replacement: (I) the share of physical capital
allocated to the corporate sector rises while that
allocated to the noncorporate and household
sectors declines, (2) households shift toward
corporate equity and away from debt, while
pension portfolios are reallocated in the opposite
direction, (3) corporations reduce their leverage
ratio and increase their dividend payout ratio, and
(4) CBIT generates shifts in the allocation of
physical capital and fmanciaI assets that are at
least as large as those generated by the other
prototypes.

Summary of Results
There is no universaI agreement about the
most appropriate way to model the effects of the
corporate income tax (and the effects of reforms
of that tax) on real and fmanciaI decisions. This
Report examined three different models of the

domestic economy to
assess the likely effects of
integration. The models
are in general agreement
with respect to the major
effects of integration on
capital allocation, corporate fmancial policy,
portfolio allocation, and
the overall effect on economic welfare.
The results of all the
models indicate that integration will encourage
capital to shift into the
cotporate sector. Most of
this shift comes from the
noncorporate business
sector,58 but in some
cases owner-occupied
housing also is reduced.

With only one exception, the models that
allow for tax-induced
distortions in cotporate
borrowing behavior agree
that the integration prototypes will improve efficiency by reducing corporate borrowing. In general, the models suggest
that because shareholder allocation and CBIT
reduce most significantly the tax penalty on
cotporate equity, they similarly reduce most
significantly tax-motivated cotporate borrowing.

The models also agree that the integration
prototypes will increase cotporate dividend payments relative to current law. Shareholder allocation integration and CBIT promote fully efficient
corporate dividend policy, while the distributionrelated prototypes can encourage corporations to
make inefficiently large dividend payouts. Nonetheless, in some calculations even the distributionrelated prototypes improve corporate dividend
policy relative to current law.
All the models show that the integration
proposals stimulate improvements in overall

139

economic well-being. The exact magnitude of the
improvements can vary from model to model and
from prototype to prototype, so integration's
improvement in welfare ranges between 0.07
percent and 0.73 percent of current consumption.
Importantly, these gains take into account that, for
some of the prototypes, taxes would have to be
raised to fmance integration. Shareholder allocation integration and CBIT tend to produce the
largest welfare gains. In addition to the traditional
welfare improvement from the reallocation of
physical capital (and other real resources) from
the rest of the economy into the corporate sector,
the models also show that, under reasonable
assumptions, integration may stimulate important
welfare gains from improvements in corporate
fmancial policy.

Comparison of Welfare Gain
Among Models
The welfare gains from integration are generally larger in the MPM than in the augmented
Harberger model. This is especially true for the
gain from improved resource allocation, and in
some cases for the gain from changes in corporate
fmancial policy as well. An important explanation
for this difference is the MPM's greater substitutability between corporate and noncorporate
businesses within an industry. Thus, in the MPM,
current law reduces economic efficiency more
than in the augmented Harberger model. Both
models predict a similar range of welfare changes
from changes in corporate debt, ranging from
roughly zero to about 0.20 percent of consumption. Additional reasons for this variation include
(1) slight differences in the underlying behavioral
models in the measurement of the tax advantage
of equity and (2) differences in the tax rates
required for the scaled-tax-rate calculations. 59
The size of the simulated gains are comparable
to, or can be reconciled with, results from simulations of similar tax law changes published in
economic literature. 6o Consider first the gains
from an improved allocation of real resources.
Using a simple two sector model, Harberger
originally estimated that the corporate income
tax's distortion in the allocation of real resources

Economic Analyses

produced a welfare gain roughly equivalent to
between 0.5 percent and 1.0 percent of GNP,
corresponding to between 0.75 percent and 1.5
percent of consumption. Shoven corrected two
errors in Harberger's original analysis, dramatically reducing the size of the corporate tax's
welfare cost. He then expanded the model from
two to twelve industries, increasing the welfare
cost of the tax. On balance, Shoven's estimates of
the welfare costs of the corporate tax ranged
between 0.75 percent and 1.5 percent of consumption. Fullerton, et al. obtained a similar
estimate of the welfare cost of the distortion in the
allocation of real resources under the cOIporate
tax. 61
These studies differ in several respects, but
share a common feature. They all use average
effective tax rates to measure the distortions of
the corporate income tax. Average effective tax
rates are measured for existing assets by taking
the ratio of the observed tax payments from the
existing stock of capital to the income generated
by that stock. While such rates may be useful for
many purposes, they can be crude representations
of the effect of taxes on investment incentives.
For example, they can include tax revenue from
lump-sum features of the tax system, from investments made under tax systems no longer in
existence, from unexpectedly profitable investments, or from pure monopoly profits. In addition, as an empirical matter, they bear little
resemblance to the theoretically preferable concept
of marginal effective tax rates. 62
A better measun; of the effect of taxes on
investment incentives is the marginal effective tax
rate (or, equivalently, the cost of capital), which
relates to incentives for incremental uses of
capital. The marginal effective tax rate is calculated using information on expected fmancing sources, economic depreciation rates, inflation rates,
required rates of return, statutory tax rates,
depreciation allowances, and credits. It represents
taxes that business enterprises would expect to
pay on an additional unit of new investment that
is just profitable at the margin. Thus, in contrast
to the average effective tax rate, it relates closely
to the forward-looking nature of a business

Economic Analyses

140

enterprise's investment decisions. Although such
calculation cannot include every detail of the tax
code, marginal effective tax rates dominate average effective tax rates as a measure of the
incentive to invest.

constant the overall average effective tax rate for
the economy as a whole. Since the tax changes
they consider would otherwise reduce revenue,
their estimated welfare gains are smaller than
those resulting from lump-sum replacement taxes.

Studies using marginal effective tax rates have
found smaller welfare costs for tax distortions in
the allocation of real capital than those using
average effective tax rates. For example,
Fullerton and Henderson adopt this approach and
fmd that eliminating all differences in the taxation
of corporate and noncorporate investments would
produce a very small annual economic welfare
gain, equivalent to about 0.007 percent of expanded national income (national income plus labor),
or roughly 0.014 percent of consumption. 63 They
fmd that eliminating all intersectoral tax distortions, including those between corporate and
noncorporate capital and between business and
housing capital, would produce larger gains.
Depending on the assumed ease with which
capital can migrate across sectors, these annual
gains range from 0.039 percent of consumption
when such migration is relatively difficult to 0.35
percent of consumption when such migration is
relatively easy. For a unitary elasticity of substitution between corporate and noncorporate capital
(as assumed in the augmented Harberger calculations above), the annual gain is roughly equivalent
to 0.11 percent of consumption.

In both the augmented Harberger model and
the MPM used in this Report, we have adopted a
marginal approach to measuring investment
incentives, and so obtain results that are more
comparable to those of Fullerton and Henderson
than to the early results of Harberger and Shoven.
For a variety of reasons, however, one would not
expect identical results in the two models. For
one thing, in several key respects, the modeling
assumptions used in the augmented Harberger
model differ from those in Fullerton and
Henderson. 65 In addition, Fullerton and
Henderson analyze tax policy changes starting
from 1985 law, while this Report analyzes tax
policy changes starting from current law.
Fullerton and Henderson also hold constant the
revenue from capital income taxes by directly
adjusting the cost of capital, while we maintain
revenue neutrality by using lump-sum taxes or by
adjusting statutory tax rates. Finally, this Report
studies integration p~ototypes that differ substantially from the hypothetical effective tax rate
equalization policies considered by Fullerton and
Henderson. Thus, one might expect that the
results presented in this Report should be similar,
though not equivalent, to those presented in
Fullerton and Henderson, if fmancing distortions
are ignored.

Fullerton and Henderson obtain these relatively small gains in part because, at the margin, debt
fmance and favorable individual level taxation of
capital gains on corporate stock eliminate much of
the tax disadvantage to investment in the corporate sector. 64 In addition, Fullerton and
Henderson's calculations are based on the new
view of dividend taxes, which magnifies the
benefit of the favorable taxation of capital gains
on corporate share appreciation, thereby reducing
the welfare cost of the current tax system. Even
under the traditional view adopted in this Report,
the Fullerton-Henderson estimates of the welfare
costs of the corporate tax based on marginal
effective tax rates are likely to remain small
compared to earlier estimates. Finally, in all
calculations, Fullerton and Henderson hold

That is indeed the case, especially for the
calculations based on the scaled tax replacement
mechanism. For the integration prototypes studied
in this Report, the augmented Harberger model
simulates annual welfare gains from improved
consumption choices ranging from 0.07 to 0.17
percent of consumption when fmancial distortions
are ignored, and from 0.08 to 0.20 percent of
consumption when fmancial distortions are captured. The most similar calculation in Fullerton
and Henderson yields a 0.11 percent gain for
complete elimination of intersectoral tax distortions, the same order of magnitude as results
presented in this Report. In part because they

141

adopt the new view of dividends, however, they
estimate smaller welfare gains from eliminating
the corporate-noncorporate tax differential.
The allocational gains in the MPM used in this
Report are substantially larger than most of those
obtained by Fullerton and Henderson; in the
scaled-tax-rate calculations, the annual gains range
from 0.22 percent to 0.43 percent of consumption. Despite the use of marginal effective tax
rates, these gains are almost as large as those
obtained by Harberger and Shoven. The primary
reason for the MPM's relatively large welfare
gain is the greater substitutability of capital and
other resources between the corporate and noncorporate sector of each industry. As a result,
even small tax differences can reduce economic
efficiency. Thus, the MPM calculations can be
compared most fruitfully to the upper range of the
Fullerton-Henderson calculations. Both sets of
calculations assume significant substitutability of
resources across sectors, thereby yielding large
welfare gains associated with reforms at this
margin.
Consider now the size of the gains from
improved corporate debt policy. In the scaled-taxrate calculations, the augmented Harberger model
used in this Report produces annual gains ranging
from negligible improvements under some prototypes to 0.17 percent of consumption for CBIT,
while the modified MPM yields annual gains
ranging from -0.22 percent of consumption for the
distribution-related prototypes to 0.23 percent of
consumption for CBIT. These gains from improved corporate borrowing decisions appear
smaller than those estimated by others. 66 Several
factors account for this Report's somewhat smaller gain. One is that not all the integration prototypes eliminate debt's tax advantage over equity,
while earlier studies considered complete elimination of debt's tax advantage. Second, our scaledtax-rate calculations significantly reduce gains
from improved ftnancial choices by raising the
difference between the statutory corporate tax rate
and the tax rate on interest income for nonCBIT
prototypes. No such effect would be found in
earlier studies that implicitly used lump-sum
replacement taxes or that assumed that integration

Economic Analyses

would eliminate debt's tax advantage. Third,
earlier studies assumed that corporate debt would
decline to zero, absent a tax advantage, while this
Report recognizes potential nontax beneftts of
debt so even without a tax advantage corporations
would continue to fmance a substantial portion (30
percent) of their capital investments with debt.
Thus, there is a much larger scope for improvement from eliminating or reducing the tax advantage of debt in the earlier studies than in the
models used in this Report.
Finally, increases in economic well-being
accompanying integration are similar to those
estimated using CGE models for the Tax Reform
Act of 1986. For example, using lump-sum
replacement taxes, Gravelle (1989) estimated that
the 1986 Act would generate annual welfare gains
ranging from 0.08 to 2.00 percent of consumption. Also using lump-sum replacement taxes,
Fullerton, Henderson, and Mackie (1987) estimated that annual welfare changes attributed to the
1986 Act would range from -0.30 to 0.89 percent
of consumption. In their calculations most similar
to those in this Report, they estimated an annual
welfare gain equivalent to 0.37 percent of consumption. The annual welfare gains presented in
this Report are therefore on the same order of
magnitude as estimates for the 1986 Act. 67

Integration in an International Context
Although the models described in the preceding sections differ in many respects, they all
ignore international trade and capital flows and
treat the United States as if it were a closed
economy. Closed economy effects of tax policies
may be modilled in important ways in an open
economy. For example, in a closed economy, a
successful saving incentive might be expected to
lower the cost of capital and increase domestic
investment. In contrast, in a small, open economy
much of the incremental saving might flow
abroad, leaving the domestic capital stock largely
unaffected. It is desirable in principle, therefore,
to analyze the integration prototypes using a
model incorporating international capital mobility.
Such a model, which is presented in the next
section, permits analysis of effects of tax changes

Economic Analyses

142

on holdings of debt and equity by U. S. and nonU.S. investors.

between the consumption of domestic and imported traded goods depending on relative prices.

Economists have analyzed the degree to which
capital is internationally mobile, but there is no
consensus. 68 Also important in the study of integration is the relative mobility of debt and equity
capital, since the integration prototypes examined
in this Report affect returns from debt and equity
investments differently. 69 While there is controversy over the extent of mobility of debt and
equity capital, this Report analyzes some possible
consequences of the integration prototypes on
capital flows. The effects of integration proposals
on foreign investment in the United States, U.S.
investment abroad, the components of the balance
of payments, and the U.S. domestic capital stock
are examined using an open economy model.
While the Report offers some tentative conclusions based on the model results regarding possible net effects of integration-related changes in
incentives in an open economy setting, more
research IS needed before reaching flfffi
conclusions.

Residents of each country allocate wealth
among four assets: domestic debt, foreign debt,
domestic equity, and foreign equity. The allocation depends on real after-tax rates of return.
Foreign and domestic debt are assumed to be
closer substitutes than foreign and domestic
equity, and, thus, international holdings of debt
are much more responsive to changes in relative
returns. Business enterprises in each country
choose the mix of debt and equity to supply
depending on market interest rates and equity
returns, and on the tax treatment of these payments at the corporate level. The international
model thus has features in common with the
portfolio allocation model presented above.

A Model of Taxation and
Internan'onal Capital Mobility

Introducing trade and capital flows complicates significantly the analysis of corporate taxation. As a consequence, economic models of
international corporate flows typically embody a
much simpler representation of the domestic
economy than the closed economy models described above. This Report uses a model of trade
and capital flows between the United States and
an aggregate of all other countries, viewed as a
single foreign country. 70 While such a representation is stylized, it offers an indication of the
likely importance of internationally mobile debt
and equity capital for assessing economic effects
of integration.
In the model, each country has four production sectors: import-competing goods (from the
U.S. perspective), equipment (producers' durables, such as machines and airplanes), nonequipment export goods, and nontraded goods and
services. Consumers in each country can choose

The model takes into account the relationship
among the three major components of the U.S.
balance of payments: the balance of merchandise
and services trade, the balance of capital inflows
and outflows, and the balance of receipts and
payments of investment income on cross-border
holdings. One possibility is an increase in imports
relative to exports in the long run, and a resulting
fall in the output of the import-competing sector.
The different tax treatment of resident and
nonresident investors also plays an important role
in the model. For example, under current law,
foreign investors in U.S. equity are subject to the
U. S. corporate level tax but not to the investor
level taxes imposed on aU. S. resident. They pay
only withholding taxes on dividends and these are
very low on average because of treaty relief.
Similarly, portfolio interest paid to foreigners is
exempt from U.S. tax under current law. To the
extent that integration prototypes alter the relative
tax treatment of foreign and resident investors,
they can lead to a reallocation of internationally
mobile capital among countries.
Three integration prototypes are modeled
explicitly: the shareholder allocation prototype
and the two distribution-related prototypes. While
potential effects of CBIT are discussed, there is
no explicit modeling of the prototype due to the

143

significant uncertainty surrounding the relative
substitutability of U. S. exempt and taxable debt in
the portfolios of U.S. and non-U.S. investors. As
before, tw'J means of fmancing revenue costs of
integration are presented: lump-sum taxes and
scaled-rate replacement taxes on capital income.
Table 13.11 presents the percentage change in the
U.S. and foreign capital stock, cross-border
holdings of debt and equity, and after-tax returns.
In addition, the three rows at the bottom of the
table present the absolute (constant) dollar changes (constrained to sum to zero) in trade, capital
flows, and net international investment income.
As with the closed economy models, simulation
results refer to effects of integration prototypes on
economic variables in the long run.

Foreign Holdings of u.s. Capital
The shareholder allocation prototype encourages foreign investors to reduce holdings of U. S.
equity and increase holdings of U.S. debt. Pre-tax
returns for foreign investors in U.S. equity, who
concentrate their holdings in the U.S. cOIporate
sector, decline as a result of the shift of capital
into the corporate sector by U.S. residents.
Because they would be denied the credit for the
corporate level withholding tax, their after-tax
returns decline as well. Accordingly, there is a
decline in foreign investment in U. S. equity. The
magnitude of the decline, of course, depends
more generally on how responsive foreigners are
to such price changes. With respect to debt, the
shareholder allocation prototype raises slightly the
U. S. interest rate because of the competition from
newly desirable equity. Foreign holdings of U.S.
debt increase as a result. The overall effect on
foreign holdings of U. S. capital depends on the
relative mobility of debt and equity capital. In the
simulations reported here, equity holdings fall,
while debt holdings increase. Nonetheless, since
debt is assumed to be more internationally mobile
than equity, 71 total foreign investment in the
U.S. increases.
The distribution-related prototypes have a
similar effect on incentives for foreign investment
in the United States. Foreign holdings of U.S.
equity decline, while holdings of U.S. debt

Economic Analyses

increase. Because the separate cOIporate tax is
maintained, however, cOIporations deduct interest
at a higher rate than under the shareholder allocation prototype. Thus, the U.S. interest rate is
higher and incentives for foreigners to shift into
U. S. debt are larger. The calculations presented
in Table 13.11 suggest that distribution-related
prototypes increase (slightly) foreign investment
in the United States. As with the shareholder
allocation prototype, the change in the composition of foreign investment is more significant than
the change in its total amount.
We do not model CBIT's effect on foreign
investment in the United States. CBIT would shift
the tax on business interest from the lender to the
borrower. As a consequence, the market interest
rate on business debt would fall below its current
level. Since non-U.S. investors receive no credit
for the tax that the borrower has paid on interest,
their net return from U. S. lending would fall,
giving them an incentive to shift out of business
debt. To the extent that domestic investors shift
capital into the corporate sector and, thereby,
lower the pre-tax rate of return in that sector,
foreign investors would have an incentive to
reduce their holdings of U. S. equity. However,
under CBIT, substantial amounts of government
and home mortgage debt are taxed identically as
under current law, offering pre-tax interest rates.
Foreign investors may shift out of cOIporate
bonds (and equity) and into these nonCBIT debt
instruments, thereby mitigating any outflow of
capital that might otherwise occur.

u. S. Holdings of Foreign Capital
The shareholder allocation prototype reduces
incentives for U.S. taxpayers to hold foreign debt,
and increases the incentive to hold foreign equity.
For U.S. taxpayers, the shareholder allocation
prototype raises the after-tax return to domestic
investment. The after-tax return on domestic
equity rises because 'of relief from the cOIporate
tax, and the after-tax return on domestic debt rises
because of the likely increase in U.S. intere3t
rates. Consequently, foreign debt is less attractive
relative to both domestic debt and domestic
equity. Foreign equity is more attractive for U. S.

Economic Analyses

144

Table 13.11
General EQuilibrium Results: International Model
Projected Long.:aun Effects of Tax Integration Alternatives
Shareholder Allocation
Financed by

u.s.

Dividend Credit
Financed by

Dividend Exclusion
Financed by

Lump Sum Tax on All Lump Tax on All Lump Tax on All
Capital Sum Tax Capital Sum Tax Capital
Tax
Percentage Changes
.9
2.7
1.5
1.2
1.9
.6
-1.3
-.4
-.9
-.6
-1.2
-.3
-17.6
-24.6
-9.2
-11.9
-26.0
-10.9
30.2
8.6
24.8
10.7
43.7
10.6
7.7
28.4
17.9
10.4
31.8
7.5
-12.9
-24.6
-30.3
-17.1
-46.3
-24.1

Capital Stock
Rest of the Wo rid Capital Stock
U.S. Holdings of Foreign Debt
U.S. Holdings of Foreign Equity
Foreign Holdings of U.S. Debt
Foreign Holdings of U.S. Equity
After-tax Return to U.S. Equity
7.7
10.1
13.7
1.8
(U.S. Residents)
20.1
After-tax Return to U.S. Equity
-15.2
-6.1
-28.3
-8.2
-13.8
(Rest of the World Residents)
1.2
3.8
3.3
1.6
.8
U.S. Interest Rate
After-tax Real U.S. Interest Rate
2.8
-6.9
3.8
-18.0
(U.S. Residents)
2.0
Return to Foreign Equity
.2
.4
(Rest of the World Residents)
.3
.3
.1
Return to Foreign Debt
.4
.4
.6
1.2
(Rest of the World Residents)
1.1
Absolute Changes (in $ billions, 1988 base)
1.0
Change in Annual Net Capital Flows
-.8
1.4
4.5
-1.5
48.8
-25.6
-9.6
Change in Net Trade Balance
-20.7
-12.8
Change in Net Receipts of
Investment Income
22.2
49.6
21.1
11.4
81.6
Department of the Treasury
Office of Tax Policy
Note: Simulations assume all U.S. debt is exempt under CBIT. See discussion in text.

investors because foreign tax credits are passed
through to U.S. shareholders.
Distribution-related integration also reduces
incentives for U.S. investors to hold foreign debt.
In contrast to the shareholder allocation prototype,
however, distribution-related integration has an
uncertain effect on incentives for U. S. investors
to hold foreign equity. Under an imputation credit
system, the dividends earned from equity investments overseas are not entitled to a credit to
offset corporate level taxes, while dividends from
domestic equity investments do receive such a
credit. To the extent that this constraint limits the
typical U. S. multinational's ability to attach
credits to dividends from foreign source income,
there is a tax incentive for U.S. investors to

2.6
-12.4
2.5
-6.5
.2
.8
1.4
-21.7
20.3

switch out of foreign equity and into U. S. equity
(and possibly debt). On the other hand, in practice, the typical U.S. multinational is likely to
have a pool of available credits sufficiently large
to attach a credit to dividends ultimately attributable to marginal investment income from abroad.
As a result, U.S. inv~stors might enjoy the benefits of integration on their foreign equity holdings,
so an increase in these investments might occur.
An imputation credit system, thus, would have an
ambiguous effect on total U. S. holdings of foreign
assets. Debt holdings decline and equity holdings
rise. Because of the greater international mobility
of debt assumed in the simulations and the greater
weight of debt in holdings of foreign assets,
however, total U. S. investment overseas declines
slightly.

Economic Analyses

145

The projected effects of the dividend exclusion
prototypes are similar in character to the imputation credit, but somewhat smaller in magnitude
because dividend exclusion provides a smaller
benefit to U.S. equity investors. Under the dividend exclusion prototype, dividends originating
form overseas investments are not eligible for
exemption at the shareholder level. As in the case
of the imputation credit system, the simulations in
Table 13.11 assume that this limitation does not
seriously restrict the typical U. S. multinational
company's ability to pay excludable dividends. As
a result, U. S. holdings of foreign equity are projected to increase. U.S. investment in foreign debt
declines because of the rise in U.S. interest rates.

cause. Nonetheless, our results suggest that
offsetting changes in incentives produce a small
net effect on capital flows. The calculations
indicate that on balance these prototypes lead to a
very small change in the flow of capital into the
United States. Both prototypes reduce net payments of investment income to foreigners. This
effect arises primarily because of the decline in
the pre-tax return on U. S. equity. Both prototypes
reduce the net trade balance. With capital flows
largely unchanged and reduced net investment
income paid to foreigners, the trade balance must
fall, so the overall balance sums to zero.

CBIT would be unlikely to change
substantially the incentives for U.S. investors to
hold foreign equity, but might reduce substantially
incentives for them to hold foreign debt. In part
because foreigners might shift out of U.S. debt,
an increase in the after-tax return available to
u. S. investors on U. S. debt could accompany
CBIT. The higher return available domestically
would offer an incentive for U.S. investors to
shift out of foreign debt and into U. S. debt. The
extent of the rise in the after-tax interest rate
available to U.S. residents, however, is uncertain
because the extent to which foreign investors
would switch out of U.S. debt is uncertain.

Ascertaining effects of CBIT are again difficult. By reducing incentives for foreigners to hold
CBIT debt, CBIT could encourage some flow of
capital out of CBIT debt. Foreigners would likely
shift their U. S. investment out of corporate bonds
into nonCBIT government and home mortgage
debt, however. The combination of a possible
capital outflow under CBIT and the lower pre-tax
returns available to foreigners on some of their
U.S. investments implies that net payments of
investment income to foreigners would fall, or
U.S. net receipts rise. To the extent that CBIT
shifts capital out of the United States, but raises
U.S. net receipts of investment income, CBIT
would have an ambiguous effect on the trade
balance.

Components of the Balance of Payments

Domestic U. S. Capital Stock

This section discusses each prototype's effects
on the three major components of the balance of
payments: net capital flows, net trade balance,
and net receipt of investment income. These three
components must balance (sum to zero) so a tax
law change cannot affect just one; the other
components must show an offsetting adjustment.

Each prototype's effect on the domestic capital
stock depends on its effect on net capital flows,
combined with its effect on saving out of changes
in real income. Both shareholder allocation and
distribution-related integration have a small,
positive effect on the flow of capital into the
United States in the model. These prototypes also
increase U. S. real income as a result of efficiency
gains from reduced net payments of investment
income to foreigners. Consequently, these prototypes increase very modestly the U.S. capital
stock. We have not attempted to model formally
effects of CBIT on the size of the U. S. domestic
capital stock.

Shareholder allocation and distribution-related
prototypes have similar effects on the balance of
payments in the model. Both would leave net
capital flows largely unchanged. As the discussion
above suggests, there is uncertainty about the size
of the portfolio shifts that the prototypes would

Economic Analyses

146

13.G DISTRIBUTIONAL EFFECTS
OF INTEGRATION
Incidence of the Corporate Tax:
Theoretical Predictions
Like most taxes, the corporate income tax
alters the distribution of real income of individuals. This section discusses the evidence relating to
who bears the burden of the corporate tax and
issues to be resolved in analyzing distributional
effects of integration.

Issues
A basic principle underlying proposals for
integration is that because corporations are owned
by shareholders, corporations have no taxpaying
ability independent of their shareholders. Corporations pay taxes out of the incomes of their shareholders. 72 The economic burden of a tax, however, frequently does not rest with the person or
business who has the statutory liability for paying
the tax to the government. This burden, or incidence, of a tax refers to the change in real incomes that results from the imposition of a change
in a tax. Importantly, the burden of the corporate
tax may not fall on shareholders. A corporate tax
change could induce responses that would alter
other forms of income as well. For example,
some of the burden may be shifted to workers
through lower wages, to consumers of corporate
products through higher prices, to owners of
noncorporate capital through lower rates of return
on their investments, or to landowners through
lower rents. This shifting might not happen
quickly, so the short-run incidence could well
differ from the long-run incidence.
Tax policy analysts have long been concerned
with the incidence of the corporate tax. 73
Although there is no unanimous view, the most
frequent fmding is that, while shareholders are
likely to bear the burden of the tax in the short
run, much of the tax is probably shifted to owners
of all capital in the long run. Some further shifting onto labor or consumers also may be possible,
however, under certain circumstances.

The Basic Harberger Model
An early incidence analysis was offered by
Harberger. 74

Suppose that investors always allocate capital
so as to equalize its net return at the margin
across sectors. Consider the imposition of an
extra tax on corporate capital, starting from an
equilibrium in which net rates of return are
equalized. The immediate effect is to lower the
net rate of return in the corporate sector by the
amount of the tax. In the short run, therefore, the
tax is borne by corporate shareholders. Over time,
however, capital begins to shift out of the corporate sector as investors seek the higher (after-tax)
rates of return available in the noncorporate
sector. As capital moves into the noncorporate
sector, its pre-tax rate of return in that sector
falls, while the pre-tax return in the corporate
sector rises. The migration of capital stops only
when the pre-tax returns change enough that the
after-tax rate of return in the corporate sector
equals the rate of return in the noncorporate
sector. Although the tax is levied only on corporate capital, noncorporate capital also suffers from
the tax in the long run; owners of noncorporate
capital receive a lower net rate of return. Indeed,
Harberger found that under reasonable assumptions, the burden of the corporate income tax is
borne equally by owners of all capital.
As in any model, the outcome depends on
initial assumptions. Much attention in the academic literature has been given to the consequences of
changing various assumptions. 7s For example, if
the marginal investment is fmanced by debt, the
burden of the tax may fall on corporate
shareholders. 76

Incidence in a Dynamic Economy
In principle, the incidence of the corporate tax
in a dynamic economy can be quite different from
the Harberger approach, in which the supply of
capital is fIxed. Intuitively, to the extent that the
corporate tax (and taxes on capital income
generally) reduces saving, the capital stock can

Economic Analyses

147

diminish, thereby decreasing wage rates and
shifting the burden to labor.
Analyzing this point is difficult, however. In
addition to addressing the controversy over the
size of the sensitivity of saving to changes in the
net return, one must specify an increase in some
other tax to compensate for eliminating the corporate tax. For example, in a life-cycle context,
fmancing the elimination of the corporate tax by
increasing taxes on individual income could
increase or decrease the capital stock and income.
(There are offsetting effects here, since the redistribution of income from younger high-savers to
older low-savers would reduce the incentive
effects of the tax.)
While the response of savings to the elimination of the corporate tax (holding total income
taxes constant) is likely to be relatively small,
there are important distributional effects across
individuals within a generation with different
mixes of labor and capital income and across
generations.
Incidence in an Open Economy

Many authors have suggested that the incidence of the corporate tax can be dramatically
different from Harberger's early closed economy
analysis.77 With frictionless international capital
markets for securities and real investment, a
small, open economy is a price-taker in international capital markets. Imposing a corporate tax in
such an economy would cause capital to flow
abroad until net rates of return are once again
equalized internationally. To the extent that labor
cannot emigrate, the incidence of the tax falls on
domestic labor.
While correct, this argument is likely to have
limited applicability to an analysis of the incidence of the corporate tax in the United States.
First, the United States is not a small, open
economy; it owns approximately 30 percent of the
worldwide capital stock. Second, world capitalmarket integration, in practice, is substantially
less than complete, particularly for equity capital. 78 As a result, even if capital is mobile

internationally, owners of domestic capital could
be expected to bear a significant portion of the
long-run burden of the tax. 79
Summary

While there is no finn agreement on the
incidence of the corporate income tax, the literature suggests the following assumptions on which
distributional analyses are conventionally based:
(1) the short-run incidence falls on owners of
corporate stock in proportion to their corporate
income or (2) the long-run burden falls either
completely on owners of all capital, or partly on
owners of capital and partly on workers. 80

Assessing Distributional Impacts of
Integration Prototypes
Distribution of Effective Tax Rates

The preceding dis~ussion highlights the importance of assumptions about incidence for analyzing long-run distributional effects of corporate tax
integration. Effects of integration on the distribution of the tax burden also depend on how integration would be fmanced (discussed below).
Tables 13.12 and 13.13 summarize the distributional consequences of the dividend exclusion,
imputation credit, shareholder allocation, and
CBIT integration prototypes, consistent with our
revenue estimates (see Section 13.H) and the
incidence assumptions discussed above. The tables
describe the long-run distribution of tax burdens
as measured by effective tax rates relative to
current law, after taxpayers have adjusted their
behavior in response to the new regimes. The
calculations represent the combined effects of
changes in individual and corporate taxes, as well
as changes in fiduciary, employment, and excise
taxes. 81
For each prototype, the estimated effective tax
rates in Table 13 .12 reflect our preferred assumption about the long-run incidence of the corporate
tax, that the tax burden is borne by the owners of
all capital. Table 13.13 shows for each prototype
the estimated effective tax rates under the alternative assumption that the corporate income tax is

Economic Analyses

148

Table 13.12
Effective Tax Rates on Individuals:
Current Law and Integration Prototypes
Standard Incidence Assumption1

CBIT: with Tax
<;:BIT: No Tax
Shareholder
on CBIT
on CBIT
Imputation
Dividend
Current Law:
Capital
Gains
Capital
Gains
Allocation
Credit
Exclusion
(1991)
Family
Share of
Witb
With
With
With
Witb
Economic
Total
Income
Taxes Effective Prototype Capital Prototype Capital Prototype Capital Prototype Capital Prototype Capital
Alone
Tax2
Tax2
Alone
Tax2
Alone
Tax2
Alone
Tax2
($1000s)
Paid Tax Rate Alone
(faxes as Percentages of Income)
10.5 10.4
10.6
10.1
10.0 lOA
10.0 10.2
10.0
10.2
0- 10
0.009
10.1
13.8
13.3
13.5
13.5
13.1
12.8
13.0
12.8
12.9
13.1
10- 20
13.0
0.037
16.8
16.7
17.1
16.4
16.5
16.0
16.2
16.0
16.3
16.2 16.3
20- 30
0.061
19.8 19.2
19.5 19.4
18.7 19.2
18.8 19.0
18.9 19.1
30- 50
19.1
0.155
21.6
21.3
21.2
21.1
20.9
2004
20.6
20.8
20.6 20.7
50- 75
0.202 20.8
22.8
22.8
23.1
22.6
22.2
21.8
22.2
22.0
22.0 22.1
75-100
0.162 22.3
24.6
23.8
23.8
23.9
23.3
23.7
22.6
23.4
23.2 23.5
100-200
23.8
0.191
26.0
24.5
22.9
22.8
23.5
22.1
23.9 24.4
23.8 24.3
over 200
24.1
0.183
21.8
21.0
20.9
20.9
20.7
20.1
20.6 20.8
20.5 20.8
Total Individual 1.000 20.9
Department of the Treasury
Office of Tax Policy
ICorporate income tax assumed to be borne 100% by capital income.
2Capital tax change imposed to offset change in revenue from prototype. Capital tax assumed to be distributed uniformly
across all capital income.

Table 13.13
Effective Tax Rates on Individuals:
Current Law and Integration Prototlpes
Alternative Incidence Assumption
CBIT: No Tax CBIT: with Tax
Current Law:
on CBrr
Dividend
Imputation
Shareholder
on CBIT
Capital
(1991)
Gains
Exclusion
Credit
Allocation
Capital Gains
Family
Share of
With
Economic
Total
With
With
With
With
Income
Taxes Effective Prototype Capital Prototype Capital Prototype Capital Prototype Capital Prototype Capital
($1000s)
Paid Tax Rate Alone
Alone Tax2
Tax2
Alone
Tax2
Alone
Alone Tax2
Tax2
(faxes as Percentages of Income)
0- 10
10.8
0.009
10.6
10.6 10.8
10.6 10.9
ll.5
11.3
10.6 11.2
11.2
10- 20
13.5
0.038
13.3
13.3 13.5
13.2 13.4
14.3
14.0
14.0
13.2 13.9
20- 30
16.8
0.062
16.6
16.5 16.7
16.3 16.6
17.3
17.6
16.3 17.0
17.2
30- 50
19.6
0.IS6
19.5
19.3 19.5
20.0
20.3
19.1 19.4
19.1 19.7
20.0
50- 75
0.20S
21.6
21.3
21.1 21.3
22.0
22.3
21.1 21.3
21.9
20.9 21.S
7S-IOO
0.164
22.7
23.4
22.4 22.6
23.8 23.1
23.4
22.4 22.7
22.2 22.8
100-200
23.9
0.190
23.8
23.3 23.6
23.9
24.7 23.9
22.7 23.5
23.S 23.8
over 200
21.5
0.176
23.4
21.4
23.1 23.4
21.3 22.4
24.5 23.3
23.0 23.4
21.1
Total Individual 1.000
21.0
22.0 21.1
21.0
20.2 21.0
20.7 21.0
20.7 21.0
Department of the Treasury
Office of Tax Policy
ICorporate income taxes assumed to be borne SO% by labor, SO% by capital income.
2Capital tax change imposed to offset change in revenue from prototype. Capital tax assumed to be distributed uniformly
across all capital income.

149

borne half by capital income and half by labor
income.
The tables classify individuals according to
their Family Economic Income (FEI). FEI is a
broad concept of income that attempts to capture
family income from all sources, taxed and untaxed, in the current year. The concept is designed to place families into income classes with
others about equally well off, with those in higher
income groups considered consistently better off
than those in lower income groups. 82
When we presented estimates of integration on
economic efficiency earlier in the chapter, we
incorporated explicitly the requirement that revenues lost as a result of integration be compensated
by offsetting tax increases. These we considered
as replacement taxes lump-sum taxes and uniform
increases in taxes on capital income. Since lumpsum taxes are not available to policymakers, we
present distributional information in Tables 13. 12
and 13.13 assuming that tax rates on capital
income are increased to fmance integration.
Dividend Exclusion

The dividend exclusion prototype would
reduce total revenues when fully phased in (see
Section 13.H). All PEl groups would receive a
slight reduction in effective tax rates. With the
capital tax replacement, there would be very small
differences in the effective tax rates under current
law and the dividend exclusion prototype (including a slight increase in the effective tax rate for
the highest income group). Hence, the efficiency
gains made possible by this integration prototype
(see Section 13.F) could be obtained with no loss
in revenue and with only slight changes in the
distribution of tax burdens across income groups.
This conclusion holds irrespective of underlying
assumptions regarding the long-run incidence of
the corporate tax (compare Tables 13.12 and
13.13).
Imputation Credit

The distributional consequences of the imputation credit prototype are qualitatively similar to

Economic Analyses

those for dividend exclusion under both incidence
assumptions. The imputation credit prototype,
described in Chapter 11, would lose revenue
when fully phased in. The revenue neutral version
of the prototype decreases the reduction in effective tax rates for upper income groups, with a tax
increase for the highest FEI group (with FEI
exceeding $200,000 per year).
Shareholder Allocation

The third column of calculations in Tables
13.12 and 13.13 presents the distribution of
effective tax rates under the shareholder allocation
prototype. There would be a significant annual
revenue loss under shareholder allocation when
fully phased in (see Section 13.H), leading to
reductions in effective tax rates larger than under
the distribution-related integration proposals,
particularly for the top two income groups (with
FEI of at least $100,000 per year). With an
offsetting uniform increase in tax rates on capital
income to fmance the revenue loss, tax reductions
for upper-income taxpayers are attenuated, with
slight overall increases in tax burdens for middleincome groups.
CBIT

Unlike the other integration prototypes considered in this Report, CBIT would not lose revenue.
When fully phased iri, the CBIT prototype would
raise a small amount of revenue with no taxation
of capital gains from the sale of CBIT assets, and
a substantial amount of revenue with current law
treatment of capital gains (see Section 13.H). In
the former case, the revenue neutral version
amounts to a very small tax increase for lowerand middle-income groups and a reduction in the
effective tax rate for the highest income group.
The reduction for the highest FEI group more
reflects the distributional implications of the
elimination of the capital gains tax on the sale of
CBIT assets than the characteristics of CBIT as an
integration prototype. To see this, note that the
revenue neutral version of CBIT with current law
treatment of capital gains has only very small
impacts on effective tax rates relative to current
law. These patterns of effective tax rates are

Economic Analyses

qualitatively similar under the two incidence
assumptions we considered.

13.H REVENUE ESTIMATES FOR
INTEGRATION PROTOTYPES
This section presents revenue estimates for
integration prototypes. Below we discuss: the
revenue estimating procedures and the assumptions behind the revenue estimates, long-run
revenue estimates for each prototype, and revenue
estimates for a 5 year budget period under the
assumption that the proposals would be adopted
effective January I, 1992, and phased in over a 5
year period. While the prototypes are not legislative proposals and we do not contemplate that any
would be proposed with so early an effective date,
5 year estimates based on the economic assumptions used ~o estimate other items in the Fiscal
Year 1992 Federal budget are useful to permit
comparison with other proposals.

Procedures and Assumptions
We prepared revenue estimates for the integration prototypes using the Individual Income Tax
Model and the Corporate Income Tax Model of
the Office of Tax Policy. These models are based
on large samples of individual and corporate tax
returns. Detailed computer programs are used to
calculate tax liabilities and simulate changes in tax
law provisions.
Earlier in this chapter, we examined economic
effects of the adoption of the prototype integration
proposals. The revenue estimates presented in this
section are dynamic. That is, the revenue estimates use the changes in economic variables
predicted by a computable general equilibrium
model !o adjust the levels of various components
of income and deductions on the tax models.
Among the important economic changes incorporated in the estimates for corporations are changes
in dividend payout rates, debt to equity ratios, the
share of capital in the corporate sector, and rates
of return to capital in the corporate sector.
Among the important changes in individual taxpayer behavior taken into account are those in

150

levels of interest and dividend income, income
from non-corporate businesses (sole proprietorships, partnerships, fanns, and small business
corporations), capital gains realizations, and
interest deductions. Changes in interest rates
affect the income and deductions of both corporations and individuals. The effects of the proposals
on the incentives of foreigners and tax-exempt
institutions to hold different types of assets in
their portfolios are taken into account.
Following the standard convention of revenue
estimates produced by the Office of Tax Policy,
Gross National Product (GNP) and the overall
inflation rate are assumed to be unchanged as a
result of the adoption of the prototypes. 83 Interest rates, relative prices, and the allocation of resources among sectors of the economy do change
depending on the expected economic effects of the
prototype. The allowance for changes in interest
rates is not strictly in accord with conventional
revenue estimating procedures because of the
nature of the proposals estimated. The integration
proposals are more likely to affect relative interest
rates paid on different types of assets than tax
changes commonly estimated. In particular, the
significant changes introduced by some of the
prototypes make it important to consider changes
in interest rates.
An important additional assumption for the
revenue estimates is that tax provisions other than
those included in the proposal remain the same as
under current law. An actual legislative proposal
would include other changes which could affect
the estimates presented here.

Effects of Integration on
Federal Tax Revenue
We estimated fully phased-in revenue effects
for each of the prototypes (at the 1991 level of
real GNP) incorporating behavioral changes that
would occur in the long run, These behavioral
changes are those which would be expected to
occur after the economy has fully adjusted to the
new tax regime. While these estimates are not
necessarily correct for the short run or the 5 year

Economic Analyses

151

budget period, they are important for understanding the long-run effects of the integration
prototypes.

eliminated, and there would therefore be some
substitution of dividends for wages and interest
payments to owners.

Dividend Exclusion

CBIT

The dividend exclusion prototype taxes corporate income (defmed as under current law) at a
rate of 34 percent. Dividends paid out of taxed
corporate income, i. e., those qualified by an
Excludable Distributions Account (EDA) as
described in Section 2.B, are not taxed at the
individual level. 84 The amount added to the EDA
is based on u.S. corporate taxes paid. 85 This
excludes foreign taxes paid to the extent that they
offset domestic taxes through the foreign tax
credit. 86 Capital gains from the sale of corporate
shares are treated the same as under current law.
Outbound foreign investment is basically treated
the same as under current law. For inbound
investment, the withholding tax on dividends paid
to foreigners is maintained.

The CBIT prototype for integration extends
the logic of the dividend exclusion prototype to
debt. Neither interest nor dividend payments
would be deductible at the corporate level, but
both interest and dividend payments from CBIT
entities generally would be excludable at the
investor level. The entity level CBIT tax rate of
31 percent would apply to both corporate and
noncorporate businesses (except for small businesses, which would be taxed as under current
law). Unlike interest on CBIT debt, home mortgage interest would continue to be deductible by
the borrower and taxable to the lender, as under
current law. Interest on U.S. Government debt
would be taxable to· the recipient. Interest taxexempt under current law would remain taxexempt to recipients under CBIT. We considered
two alternative assumptions for the taxation of
capital gains on CBIT assets: (1) no taxation of
capital gains on CBIT assets and (2) current law
treatment of capital gains on CBIT assets.

The basic principle of the dividend exclusion
prototype is to reduce the double tax on distributed corporate income. We estimate that when fully
phased in, integration through dividend exclusion
loses $13.1 billion annually at 1991 levels of
mcome.
Dynamic changes in the economy would
increase corporate income tax receipts under the
dividend exclusion prototype. Increases in corporate tax receipts would result from the incentive to
shift corporate fmancing from debt to equity. The
reduction in corporate borrowing would decrease
corporate interest deductions. Induced changes in
interest rates also would affect corporate interest
deductions and therefore affect corporate tax
revenues. The increases in corporate tax revenues
would be slightly more than offset by the decrease
in individual income tax receipts from the dividend exclusion. The dividend exclusion, thus,
provides incentives for corporations to increase
excluded dividends. In closely-held corporations,
the incentive under current law to payout profits
as managerial wages or interest would be largely

In contrast to the other integration prototypes,
the CBIT prototype would increase tax receipts
relative to those under current law. Once the
behavioral changes are fully accounted for, the
annual increase in revenues would be $3.2 billion
with no taxation of capital gains on CBIT assets
and $41.5 billion with current law treatment of
capital gains. While overall tax receipts would be
increased under the CBIT prototype, individual
tax payments would be substantially reduced
because dividends, noncorporate business income,
most interest and some capital gains would no
longer be taxable to individual recipients. The
reduction in individual income tax receipts reflects
the taxation of capital income at the entity level.
Noncorporate entities subject to CBIT would now
be taxed at the 31 percent CBIT rate. Much of
this income is currently taxed under the individual
income tax.

Economic Analyses

Shareholder Allocation
The shareholder allocation prototype approximates passthrough integration more closely than
the dividend exclusion or CBIT prototypes. The
prototype would retain a corporate tax rate of 34
percent. Taxable shareholders would receive a 31
percent credit for corporate level taxes paid, while
tax-exempt and foreign shareholders would receive no credit. The credit would accompany the
allocation of corporate income to the shareholder.
Intercorporate dividends would be granted a full
dividends-received deduction in lieu of a credit.
Under this prototype, corporate income tax is
taxed at the individual level as part of corporate
income rather than as a separate income item.
Capital gains on corporate stock due to retained
earnings would not be taxed, since undistributed
corporate income would increase shareholders'
basis. Increases in corporate stock values from
other sources would be taxed as under current
law. For outbound investment, the foreign tax
credit would be passed through at the taxable
investor's rate. For inbound investment, the
withholding tax on dividends paid to foreign
investors would be retained.
Because shareholder allocation integration
would extend distribution-related integration to
retained earnings and shareholders would not be
taxed on untaxed corporate preference income, it
would lose significantly more revenue than would
the dividend exclusion prototype. We estimate that
when fully phased in, shareholder allocation
integration would lose $36.8 billion annually at
1991 levels of income.
Most of the revenue loss would be in the
individual income tax. While taxable income of
individuals would be increased substantially by
including all corporate income (rather than just
dividends received), this would be more than
offset by the revenue loss from the credit for
corporate taxes paid. For taxpayers in the 31
percent tax bracket, the tax on the additional
income and the credit for corporate taxes paid

152

would offset each other and leave taxes approximately unchanged. For taxpayers in lower tax
brackets, however, the additional corporate
income subject to tax would be taxed at a lower
rate than the credit. For example, taxpayers in the
15 percent bracket would be taxed at 15 percent
on the additional income but receive a credit at a
31 percent rate. For lower tax bracket taxpayers,
the corporate credit can be used to offset taxes
against wages and other income.
The other major factor in the large revenue
loss from the shareholder allocation prototype is
the basis adjustment for corporate stock.
Shareholders' basis would rise to reflect income
already taxed at the corporate level, and so revenues from the taxation of capital gains on sales of
stock would be reduced.
Corporate tax receipts would increase, since
dynamic behavioral changes (including the expansion of the corporate sector) are taken into account. As with distribution-related integration, the
increase in corporate tax receipts results primarily
from the reduction in corporate debt and therefore
in interest deductions.

Imputation Credit System
The fmal prototype we considered is distribution-related integration through an imputation
credit system. Under this prototype, corporate
taxes paid are credited to a shareholder credit
account (SCA). Individual shareholders report
dividends grossed-up (by one divided by one
minus 0.31) to reflect corporate taxes paid and
receive a credit for corporate taxes paid. The
prototype calculates the credit and gross-up factor
at the top individual 31 percent tax rate rather
than the top 34 percent corporate tax rate to limit
the credit to no more than the individual income
tax paid by individuals in the highest tax bracket.
We estimate that accomplishing distributionrelated integration through an imputation credit
system would generate a fully phased-in revenue
loss of $14.6 billion per year.

APPENDICES

THE

A.1

ApPENDIX A:
CORPORATE INCOME TAX IN THE UNITED STATES

BRIEF DESCRIPTION OF THE
CORPORATE INCOME TAX

The corporate
enacted in 1909 as
of doing business
individual income
enacted in 1916.

corporate shareholder is entitled to a full or
partial dividends received deduction (DRD) ,
depending on its percentage ownership of the
distributing corporation.

income tax originally was
an excise tax on the privilege
in the corporate form. An
tax on dividend income was

U. S. corporations are subject to tax on foreign
as well as domestic income. Although aU. S.
corporation is required to pay U.S. tax currently
on foreign income earned through a foreign
branch, U.S. tax is generally not imposed on
earnings of a foreign subsidiary until the subsidiary distributes its income to the parent corporation as a dividend. In computing U.S. tax
liability, U.S. taxpayers (including corporations)
are allowed a credit for foreign taxes paid, subject
to certain restrictions. See Chapter 7.

The Corporate Income Tax Base
Corporations are generally taxed at a 34
percent marginal rate on their taxable income. To
compute taxable income, a corporation deducts
from gross income business expenses paid or
incurred during the taxable year. These expenses
include employee compensation, state and local
taxes, depreciation, and interest expense, but not
dividends paid. When deductions exceed income,
a corporation has a net operating loss (NOL).
Corporations generally can carry back net operating losses to offset taxable income for the 3
preceding years and can carry forward any remaining net operating loss to offset taxable
income for 15 years.

In addition to these general rules, special rules
apply to specific types of businesses that conduct
activity in corporate fonn, such as financial
institutions and insurance companies. Other
special rules apply to specific types of activities,
such as the exploration, development, and production of natural resources. Certain types of corporations are granted full or partial relief from
corporate level tax.

Like individuals, corporations generally
include gains on appreciated assets in income (and
deduct losses on depreciated assets from income)
only when the assets are sold or otherwise disposed of (when the gains or losses are realized).
Corporations may deduct capital losses only
against capital gains, and unused capital losses
may be carried back for 3 years and forward for
5 years.

Tax Rates
Corporations are subject to tax at a rate of 15
percent on the first $50,000 of taxable income, 25
percent on the next $25,000 of taxable income,
and 34 percent on tax,able income above $75,000.
The marginal rate on a corporation's taxable
income between $100,000 and $335,000 is increased by 5 percent to phase out the benefit of
the graduated rate structure. Thus, corporations
with incomes in the phaseout range pay tax at a
marginal rate of 39 percent. Corporations with
taxable incomes in excess of $335,000 pay tax at

Because the double tax on corporate earnings
distributed to shareholders might become a triple
or quadruple tax if corporations were taxed in full
on dividends received from other corporations, a

153

Appendices

a flat 34 percent rate. In 1989, over 90 percent of
corporate taxable income was taxed at the 34
percent rate.
Corporations also are subject to an alternative
minimum tax (AMT). Corporations pay AMT
only if their minimum tax liability exceeds their
regu lar tax liability. A corporation's AMT base is
its taxable income, adjusted to eliminate the
benefit of certain deferrals of income, accelerations of deductions, and permanent exclusions.
The resulting amount, alternative minimum taxable income (AMTI), is reduced by an exemption
amount and is taxed at a 20 percent rate. The
basic exemption amount is $40,000, which is
reduced by 25 percent of the amount by which
AMTI exceeds $150,000. A corporation's minimum tax liability can generally be credited against
future regular tax liability.

Entities Subject to the Corporate Tax
A business entity is taxable as a corporation if
it is organized as a corporation under state law. In
addition, Treasury Regulations treat an unincorporated entity as a corporation if it has more
corporate characteristics than noncorporate
characteristics. The four relevant corporate characteristics are: (1) continuity of life, (2) centralization of management, (3) limitation of liability
for debts to property of the entity, and (4) free
transferability of interests. 1 Certain partnerships
also are treated as corporations if their interests
are traded on an established securities market or
are readily tradable on a secondary market (or its
equivalent) and the partnership is not engaged in
a qualifying passive activity. 2
Subchapter C refers to the provisions of the
Code that apply to most corporations. In 1958,
Congress enacted Subchapter S of the Code to
enable certain corporations to elect exemption
from the corporate level tax. S corporations, like
partnerships, are generally treated like conduits
for tax purposes. The income of S corporations is
taxed directly to their shareholders. To qualify for
this passthrough treatment, a corporation must
have no more than 35 shareholders and only one
class of stock, and all of its shareholders must be

154

individuals who are U.S. citizens or residents or
certain trusts and estates. There also are restrictions on an S corporation's affIliations with other
corporations.
In addition to S cOlporations, other entities
that meet certain restrictions on assets, type of
business, and distributions to shareholders qualify
as conduits for all or a portion of their income. A
regulated investment company (RIC), a mutual
fund that makes diversifIed investments for its
shareholders, pays no tax on amounts distributed
to its shareholders if it distributes currently at
least 90 percent of its dividend and interest income and meets certain other conditions. 3 A real
estate investment trust (REIT), a corporation or
association that specializes in investments in real
estate and real estate mortgages, also may receive
passthrough treatment if it meets certain conditions designed to ensure that its assets and income
are primarily related to real estate. 4 A real estate
mortgage investment conduit (REMIC), an entity
that holds a fIxed pool of mortgages and issues
multiple classes of interests to investors, also
qualifIes for passthrough treatment. 5 Qualified
distributions to members of cooperative organizations also are taxed directly to the members and
are not taxed at the entity level.

Treatment of Debt and Equity
Under present law, the tax treatment of the
returns to an investor in a corporation depends
upon whether an investment is considered debt or
equity. A corporation generally can deduct interest on corporate debt. 6 Consequently, corporate
earnings paid to debtholders as interest bear no
tax at the corporate level. In contrast, because
dividends are not deductible, corporate tax must
be paid on the earnings attributable to equity
investments, regardless of whether the earnings
are retained or distributed.
Individual debtholders are taxed on interest
income when received or accrued, in accordance
with their method of accounting. Individuals are
taxed on corporate income when the income is
distributed to them as dividends.? Increases in the
value of corporate stock held by individuals,

Appendices

155

whether due to retained earnings, appreciation of
the cotporation's assets, or other factors, are
generally not taxed until the stock is sold. 8 Such
gains are generally capital gains. Individuals also
may not deduct losses on cotporate stock until the
stock is sold. Such losses are generally capital
losses and may be deducted without limitation
against capital gains. However, capital losses in
excess of capital gains also may be used to offset
only $3,000 of an individual's ordinary income
per year, with any excess carried forward
indeftnitely.
Cotporate debtholders also pay tax on interest
income when received or accrued, in accordance
with their method of accounting. A corporate
shareholder must include all dividends in income
but can deduct a portion of dividends received
from other domestic cotporations. The deduction
for dividends received is 70 percent if the recipient corporation owns less than 20 percent of the
stock of the payor, and 80 percent if the recipient
corporation owns between 20 percent and 80
percent of the stock of the payor.9 Intercotporate
dividends among members of affiliated groups
(each 80 percent or more owned, directly or
indirectly, by a common parent) are generally
fully deductible by the recipient. Thus, the maximum rate of tax on dividends received by cotporate shareholders is generally 10.2 percent (30
percent of dividends received multiplied by the 34
percent corporate tax rate). Corporate capital
gains are currently taxed at the same rate as
ordinary income, and capital losses may offset
capital gains, but not ordinary income, with a
3 year carryback and 5 year carryforward.
Although debt and equity are treated very
differently by the tax system, distinguishing debt
from equity is not straightforward. In 1969,
Congress authorized the Department of the Treasury to issue regulations to determine whether an
interest in a corporation should be treated as stock
or debt for tax purposes. Congress suggested that
Treasury consider the following factors in making
this determination: (1) the existence of a written
unconditional promise to pay on demand or on a
specifted date a sum certain in money at a ftxed

rate of interest, (2) whether the instrument is
subordinated to or has preference over any debt of
the corporation, (3) the issuer's debt to equity
ratio, (4) whether the instrument is convertible
into stock, and (5) the relationship between
holdings of the issuer's stock and holdings of the
instrument in question. to
Although Treasury issued three drafts of
regulations attempting to distinguish debt from
equity, the task of devising simple, workable rules
for distinguishing between debt and equity proved
elusive. Ultimately, Treasury withdrew all of
these regulations.
In the absence of regulations, taxpayers and
the IRS look to judicial opinions and IRS rulings
to determine whether an instrument will be treated
as debt or equity for tax putposes. In addition to
the factors listed in the 1969 statute, the following
factors have been considered relevant: (l) the
holder's rights upon default, (2) the equity features of the instrument, such as voting rights or
participation in earnings, (3) whether the corporation has sufftcient projected cash flow to service
the debt, (4) whether the holder has management
rights, and (5) whether the holder acts like a
reasonable creditor in protecting its rights.
To summarize, it has not proved possible to
develop simple and acceptable guidelines for
distinguishing between debt and equity. As ftnancial markets become more flexible and innovative,
that task becomes more difftcult. The administrative complexity and compliance costs associated
with making the debt-equity distinction are serious
problems in the current system of corporate
taxation.

Cross-Border Investment
The tax treatment of cross-border investment
is discussed in Chapter 7.

Tax-Exempt Organizations
The treatment of tax-exempt organizations is
discussed in Chapter 6.

156

Appendices

A.2

OVERVIEW OF U.S.
CORPORATE TAX RECEIPfS

for example, rules requiring unifonn capitalization
of certain expenditures. As anticipated, the 1986
Act increased corporate income tax receipts (and
lowered individual income tax receipts) as a
percentage of total income tax receipts. The
percentage of income tax receipts accounted for
by corporate taxes increased from 15 percent in
1986 to 19 percent in 1989 and dropped back to
17 percent in 1990. The percentage of income tax
receipts accounted for by individual income taxes
fell from 85 percent to 81 percent, rising to 83
percent in 1990. Current estimates indicate that
the 1986 Act increased corporate income tax
receipts by approximately $130 billion from 1987
to 1991.

In 1990, the corporate tax generated Federal
revenues of $93.5 billion. Federal corporate tax
receipts have generally increased over the past 40
years, but when adjusted for inflation, they have
fallen since the late 1960s. In constant 1982
dollars, corporate tax receipts averaged $85
billion per year in the 1950s, $86 billion per year
in the 1960s, $77 billion per year in the 1970s,
and $56 billion per year from 1980 to 1986. Since
1986, real corporate tax receipts have averaged
$76 billion per year in 1982 dollars. From the
1950s to 1986, corporate receipts also fell as a
percentage of Federal budget receipts and of gross
The level of corporate tax receipts depends
national product (GNP). See Figure A.l. Since
heavily on economic conditions. When the u.s.
1986, however, the decline in the relative importance of the corporate tax has stopped and may economy is growing, corporate profits are strong,
have reversed. From 1987 through 1990, COIpOand corporate tax receipts increase, but when the
rate receipts averaged 9.9 percent of total Federal economy is in recessfon, corporate profits tend to
budget receipts, above the average of 8.9 percent fall, and corporate taxes decrease. During the
for the rest of the 1980s, but less than the 1970s recession of the early 1980s, for example,
average of 15.0 percent. From 1987 to 1990,
corporate taxes as a percentage of total budget
estimated tax liabilities for nonfinancial corporareceipts fell from 10.2 percent in 1981 to 6.2
tions, relative to GNP or gross domestic product,
percent in 1983. This decline was mostly
also slightly exceeded the
Fi~re A.1
average for the early 1980s.
Corporate ReceIpts as a Percentage of
Total Receipts and Gross National Product
The Tax Reform Act of
1940-1991
1986 (the 1986 Act) adopted
base-broadening measures
50
designed to increase the
overall level of corporate
40
taxes, although it reduced the
\
Corporate Receipts as a
maximum marginal corporate
Percentage of Total Receipts
/
tax rate from 46 percent to
34 percent. The base broadening was accomplished
primarily by repealing the
investment tax credit, limiting
depreciation deductions, reCorporate Receipts as a
stricting the use of net oper10
Percentage of Gross National Product
\\ _ / '
ating losses, strengthening the
'/cOIporate alternative mini/
O
, , , , I' , , , Iii iii iii ii' iii Iii i iii , iii ii' iii
mum tax, repealing the
General Utilities doctrine,
1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
Year
and adopting significant
changes in accounting rules,

V\

f30! vJ~

~ 20/

/

~

v\

~---

157

attributable to the a decline in pre-tax corporate
profits, from $202 billion in 1981 to an average
of $178 billion in 1982 and 1983.
Foreign countries have a wide variety of tax
systems, which make it difficult to compare
directly corporate tax burdens across countries,
but some general observations can be made. In
1988, corporate income taxes accounted for an
average of 8 percent of total income tax receipts
for the 22 countries in the OECD. The average in
1988 was the same as in 1980. Although U.S.

Appendices

corporate income taxes were 8 percent of total tax
receipts in 1988, the same as the average for the
22 OECD countries, the U.S. percentage is
expected to be higher in subsequent years if
current trends continue. Countries with percentages higher than the OECD average in 1988 include
Japan at 24 percent, the United Kingdom at 11
percent, and Italy at 9 percent; countries with
percentages below the OECD average include
Germany at 5 percent, France at 5 percent, and
Switzerland at 7 percent. 11

ApPENDIX B: EXPERIENCE OF OTHER COUNTRIES WITH
DISTRIBUTION-RELATED INTEGRATION SYSTEMS
This appendix briefly describes the distribution-related integrated systems of six of the United
States' major trading partners. I The Australian
and New Zealand imputation credit systems most
closely resemble the prototype discussed in Chapter 11. The United Kingdom system is a prominent example of a compensatory tax system. This
appendix also discusses the Canadian, French and
Gennan distribution-related systems. This appendix does not describe the Japanese corporate tax
system, because in 1989 Japan replaced its split
rate tax system with a classical system.

B.1

The shareholder receives an imputation credit
equal to the amount of distributions from the
franking account (franked distributions), grossedup at the corporate rate (currently, 39 percent),
and then multiplied by that rate. 2 The shareholder
includes this amount in his income and receives a
credit in the same amount against his personal tax
liability. Imputation credits generally are not
refundable.
The balance in the franking account represents
the portion of the corporation's after-tax income
that, in effect, has been taxed fully (taxed at the
corporate rate). In general, the franking account
balance derives from the amount of tax the corporation pays. At the current tax rate of 39 percent,
for every AU$39 the corporation pays in tax, it
adds AU$61 to the balance of the account. The
calculation converts after-tax corporate income
that is taxed at various rates into an equivalent
combination of fully-taxed and fully exempt
amounts. 3 Thus, Australia's system accords
shareholders relief only from corporate level tax
actually paid with respect to distributed income,
and distributed preference income is subject to tax
at the shareholder level.

AUSTRALIA

Introduction
Australia's imputation credit system became
effective July 1, 1987. Major changes to related
tax laws have subsequently taken effect, most
notably:
•

a reduction in the top corporate rate from 49
percent to 39 percent,

•

the imposition of a 15 percent tax. on the investment income of pension plans, and the extension to
them of the imputation credit (at the full rate of 39
percent), and

•

An Australian corporation must make entries
in its franking account throughout the year upon
the occurrence of specified events in the assessment, payment, and adjustment of tax. The
franking account is credited when the corporation:
carries forward a franking surplus from the
previous year, pays tax, receives franked dividends from another company, is served with a
detennination reducing the amount of a "franking
deficit tax offset," or has an "estimated debit
detennination" (see "Allocating Credits to Dividends," below) that lapses or substitutes a new
estimated debit detennination.

the exemption of most foreign income from the
corporate tax base.

Description of Mechanics
Imputation Credits
Australia's imputation credit system makes
imputation credits available to taxable shareholders (including pension plans) for distributions
from the corporation's franking account. Imputation credits provide full relief from the corporate
level tax paid with respect to distributed income.
Distributions not paid from the franking account
are considered to be paid from preference income
and are taxed to the shareholder without gross-up
and without credit.

The franking account is debited when the
corporation: pays franked dividends, has tax
refunded, is served with a detennination (or
increase) of a franking deficit tax offset, receives
159

Appendices

(or is deemed to receive) notice of an estimated
debit detennination, e.g., appeals a tax assessment, makes on-market share buybacks; or underfranks a dividend (franks it by less than the
required franking amount, if the required franking
amount is 10 percent or more of the dividend).

Compensatory or Withholding Tax
The Australian system does not have a compensatory or withholding tax on distributions.

160

allocated to the buyback from the share premium
account). With respect to the dividend portion, the
corporation debits its franking account as required
under the general rules and the shareholder receives the imputation credit. The shareholder's
basis in his stock is irrelevant for dividend purposes but is relevant for the portion treated as
return of paid-up capital, so the shareholder could
have a dividend and a capital gain or loss on the
same transaction.
An on-market buyback is treated as a capital

Defined, Bonus Shares,
Shnre Repurchnses

Di~7'dends

In general, dividends include all non-liquidating distributions of money or other property to
shareholders out of profits (under corporate law,
the corporation cannot pay dividends as a return
of capital without a court order). Liquidating
distributions generally are deemed to be dividends
to the extent they represent profits.
A corporation can issue bonus shares as a
mechanism for extending the imputation system to
retained earnings. An issue of bonus shares
distributed to a shareholder is treated as a dividend unless it is paid out of the corporation's
share premium account (which represents amounts
paid on issuance of shares in excess of par value).
Thus, if the corporation has a sufficient balance in
its share premium account, it can choose the tax
treatment of the bonus issue by choosing whether
or not to debit the account, subject to certain rules
for dividend-streaming arrangements. See
"Streaming" below.
The tax treatment of a share repurchase (or
"buyback") differs depending upon whether the
transaction is an "on-market" or an "off-market"
purchase. An on-market buyback occurs in the
ordinary course of business on an official exchange; an off-market buyback (a buyback by an
unlisted company or by a listed company not in
the ordinary course) occurs otherwise.
An off-market buyback is treated as a dividend

to the extent it exceeds paid-up capital for the
shares (share capital plus the amount, if any,

transaction to the shareholder (because he does
not know that his buyer is the corporation). The
corporation has no gain, loss, or deductions.
However, the corporation must treat the buyback
as a dividend to the extent it would be a dividend
if it were off-market and, with respect to such
amount, must debit its franking account under the
allocation rules. See "Allocating Credits to Dividends, " below. (This notional dividend also might
affect any provisional required franking amount
for any actual frankable dividend.) No imputation
credit is available to the shareholder to offset his
capital gain. 4

Allocating Credits to .Dividends
Australia has adopted allocation rules generally designed to assure that a corporation pays
dividends first out of the franking account, and to
prevent corporations from streaming franked
dividends to resident shareholders, who can use
imputation credits, and unfranked dividends to
foreign shareholders (and tax-exempt shareholders), who cannot. The allocation rules impose a
minimum "required franking amount" for a
dividend and provide for adjustments and sometimes penalties if a dividend is overfranked or
underfranked by more than a de minimis amount.
The required franking amount ideally franks
all dividends paid during the year to the extent of
the corporation's after-tax income. To ensure that
the corporation does not underfrank a dividend,
the rules require the company: (1) to take into
account all dividends. to be paid on the same day,
that have been declared but not yet paid, or that
the corporation is committed to pay later in the

Appendices

161

same year (a committed future dividend), such as
dividends on preferred stock, in allocating franking credits to a given dividend, (2) to frank a
dividend that was a committed future dividend at
least to the same extent as the earlier dividend,
and (3) to frank a dividend at least to the same
extent as any other dividend on the same day.s
These rules do not, however, prevent a corporation from franking an earlier dividend on one
class of stock at one rate and franking a later
dividend on another class of stock at a lower rate
where the corporation was not committed to pay
the later dividend or where the later dividend is
paid in the succeeding year. An upper limit on
franking is set by reference to the corporate tax
rate; at current rates, a dividend of AU$61 can
carry no more than AU$39 of imputation credits.
The required franking amount could range
from zero, for a corporation with no taxable
income, to 39 percent of the dividend, for a
corporation with sufficient after-tax income.
However, the required franking amount might not
be readily determinable when a dividend is distributed during the year, where it is not clear
whether the corporation will have sufficient
taxable income for that year. The situation also
could be complicated by later events, such as a
refund of previously paid tax. If, for such a
reason, a year-end deficit were to result, the
corporation would be subject to a franking deficit
tax and possibly a penalty tax. An estimated debit
determination is a procedure for resolving this
problem; if the corporation expects such a later
debit, so the dividends paid would tum out to
have been overfranked, the corporation may
notify the tax authorities and make an anticipatory
debit to its franking account.

Where overfranked dividends (or other adjustments) result in a deficit in the franking account
at the end of the year., the corporation must pay a
franking deficit tax. The franking deficit tax is the
amount of tax sufficient to restore the franking
account to zero. 6 This tax does not result in a
positive credit to the franking account, because it
functions as a prepayment of corporate tax prematurely imputed to shareholders by the payment of
overfranked dividends. The franking deficit tax is
not a penalty, and therefore a corporation may
offset a payment of franking deficit tax against its
future tax liability. However, to discourage more
than de minimis overfranking, a penalty equal to
30 percent of the franking deficit tax is payable
where the franking deficit exceeds 10 percent of
the total of the franking credits arising during the
year and any dividend paid during the year was
overfranked.
Tax Rates

°

The corporate tax· rate currently is 39 percent.
Marginal tax rates for individuals range from
percent to 47 percent. The 47 percent rate applies
to taxpayers with taxable income exceeding
AU$50,OOO. Capital gains on assets acquired after
September 19, 1985 are taxed at ordinary income
rates. However, to determine the amount of gain
recognized on disposition of a capital asset, basis
in the asset is indexed for inflation if the asset
was held for more than 1 year.

Treatment of Preference Income

If a corporation underfranks a dividend (and if

Dividends paid out of preference income
(when the franking account balance is zero) are
taxable when received by shareholders and thus
corporate preferences are not extended to shareholders.

the required franking amount is 10 percent or
more of the dividend), the corporation must debit
its franking account to the extent of the underfranking. Thus, the corporation is treated as
having franked the dividends to the required
amount, but the shareholders forfeit the imputation credit attributable to the underfranking.

The Australian system currently provides
corporations few preferences. In 1988 Australia
reformed its depreciation system and other tax
concessions. For example, depreciation rates for
"plant" were based on 5 or 3 year lives; now they
are based on effective lives (using a 150 percent

Appendices

declining balance or "prime cost") plus a 20
percent "loading." The 150 percent deduction for
research and development expenditures is scheduled to be scaled back to 125 percent in the
mid-1990s.

Treatment of Domestic
Intercorporate Dividends
Dividends received by an Australian cOIporation from another Australian corporation generally
are free of tax because tax is rebated. In addition,
credits attached to intercorporate dividends are
credited to the recipient cOIporation' s franking
account. However, unfranked dividends to private
corporations (generally, unlisted corporations) are
taxed without refund. This exception is designed
to prevent the use of private corporations to defer
tax on distributed preference income. Australia
does not permit consolidation of affiliated coIporations for purposes of its imputation system (or
for its corporate tax generally, although there is
loss transfer for 100 percent related corporations).

Treatment of Foreign Source Income
Beginning July 1, 1990, foreign source income
derived from comparable tax countries through a
branch is generally excludable from corporate
income. An exemption from corporate tax also
applies to dividends received from a corporation
resident in a comparable tax country if the Australian corporation owns at least a 10 percent interest
in that corporation. Dividends received from
portfolio investments (i.e., less than 10 percent)
in corporations resident in comparable tax countries are taxable with a credit allowed for foreign
withholding taxes. However, because foreign
taxes paid with respect to foreign source income
derived from comparable tax countries do not
generate credits to the franking account, dividends
paid by an Australian corporation out of such
income do not carry credits in respect of such
foreign taxes and are exposed to shareholder level
tax. Thus, this foreign source corporate income is
still double-taxed, once when earned in the foreign country and once when the after-foreign-tax
amount is distributed to domestic individual
shareho lders.

162

Income derived from low-tax countries
through a branch or a nonresident corporation
generally is subject to full taxation at the corporate level with a credit for foreign taxes paid on
such income. Where an Australian corporation
owns a 10 percent or more interest in a corporation residing in, or deriving substantial income
from, a low-tax country, the Australian corporation is taxed currently on its share of the nonresident corporation's income and may credit its
share of foreign taxes paid by the nonresident
corporation on an "accruals" basis, provided that
the foreign corporat~on is a controlled foreign
company (that is, 5 or fewer Australian residents
control 50 percent or more of the company). Such
a 10 percent shareholder maintains an "Attribution
Tax Account" (ATA) for every entity in the
chain, in which income is attributed to that entity;
when a dividend is paid between entities, a debit
is made to the ATA of the paying corporation and
a credit is recorded in the ATA of the receiving
cOIporation. 7 Where the Australian cOIporation
owns a lesser percentage, the accruals tax does
not apply, but dividends received are subject to
Australian tax (with a tax credit for foreign
withholding taxes paid on the dividend). Because
foreign taxes paid do not generate credits to the
franking account, dividends paid out of such
income to the shareholders of the Australian
corporation are exposed to shareholder level tax.
The net effect of this system is the equivalent of
allowing a deduction for foreign taxes on distributed foreign source. income earned through an
Australian corporation.

Treatment of Tax-Exempt
Shareholders
Excess imputation credits are not refundable to
any shareholder, including tax-exempt shareholders. Accordingly, income taxed at the corporate
level is subject to one level of tax even where it
is distributed to tax-exempt shareholders.
Until 1988, pension funds were tax-exempt,
although distributions were taxable to beneficiaries. The new statute imposes a tax at a 15 percent rate on the investment income of pension
funds, but allows pension funds an imputation

Appendices

163

credit for franked dividends at the full 39 percent
rate. Thus, a pension fund can use the excess
imputation credits (a 24 percent credit) to shelter
the tax on a large amount of other investment
income (such as interest, rents, royalties, foreign
income, capital gains, and unfranked dividends).
Pension funds also may utilize imputation credits
to reduce tax imposed on contributions. These
changes are designed in part to encourage pension
funds to invest in domestic corporations having
Australian tax liability, thus reducing the tax
arbitrage gains to pension funds from investing in
bonds or in corporations paying unfranked
dividends.

Treatment of Foreign Shareholders
Australia generally imposes a withholding
tax on dividends from Australian corporations to
nonresident shareholders. No distinction is made
between portfolio and nonportfolio investment.
The normal withholding rate is 30 percent, but
treaties may reduce this rate to 15 percent. The
gross-up and imputation credit procedure does not
apply to nonresident shareholders. However, the
franked portion of a dividend is exempt from the
withholding tax. Thus, the franked portion of a
dividend bears Australian tax at the 39 percent
corporate rate. Unfranked dividends are subject to
withholding tax and, thus, bear Australian tax at
the applicable withholding rate.

Treatment of Low-Bracket
Shareholders
Although a shareholder may use excess credits
to offset any other tax liability he may have,
excess credits are not refundable. Unused credits
may not be carried forward or back. The imputation credit (aggregated with other nonrefundable
credits) is stacked so refunds from other sources
cannot impair use of the credit.

stripping operation, imputation credits attached to
the dividend and the tax rebate for intercorporate
dividends may be denied. One effect of the dividend stripping rule is to discourage sales of shares
by tax -exempt or nonresident shareholders in
anticipation of the payment of a franked dividend.
Second, to inhibit streaming through partnerships
and trusts, imputation credits received by a partnership or trust are generally allocated in accordance with a partner's or beneficiary's share of
partnership or trust income. Third, a special debit
to the franking account is required when a
corporation distributes an unfranked dividend or
tax-exempt bonus share to a shareholder in substitution for a franked dividend as part of a dividend
streaming arrangement. Generally, the franking
debit is calculated as if the franked dividend had
been franked to the same extent as the dividend
for which it substituted, thus ensuring equal
franking for all dividends paid on a particular
class of stock as part of the same distribution.

Treatment of Interest
Interest paid by an Australian corporation
generally is deductible. Interest paid to a resident
lender is includable in the lender's income.
Interest paid to a foreign lender (whether or not
resident in a treaty country) is subject to a 10
percent withholding tax. Australia has a thin
capitalization rule that denies a resident corporation a deduction for interest paid to foreign
shareholders where the foreign shareholders own
at least 15 percent of the resident corporation and
the resident corporation's debt to equity ratio with
respect to the nonresident shareholders' investment is in greater than 3 to 1 (6 to 1 for fmancial
institutions). Beginning July I, 1990, this rule
applies even if the foreign controlling shareholder
is in turn controlled by Australian residents.

B.2

CANADA

Streaming

Introduction

In addition to the allocation rules described
above, Australia has adopted several anti-streaming provisions. First, where a dividend is paid to
a corporate shareholder as part of a dividend

Canada introduced distribution-related integration in 1971 with the adoption of a straight credit
system that grants a credit to resident individual
Canadian shareholders with respect to dividends

Appendices

received from Canadian corporations. The credit
is not required to be funded at the corporate level.
That is, the amount of the shareholder credit does
not depend on the payment of tax by the corporation. Excess credits are not refundable.

Description of Mechanics

164

on the rate at which distributed income has been
taxed at the corporate level under the Federal and
provincial tax systems. See "Tax Rates," below.
Combining Federal and Ontario provincial tax, the
system integrates 32 percent of a regular cotpOration's tax, 41 percent of a manufacturing cotpOration's tax, and 86 percent of a small business
corporation's tax. 9

Credits
Compensatory or Withholding Tax
Where a Canadian resident individual shareholder receives a taxable dividend (described
below) from a Canadian corporation, the shareholder grosses up the dividend by 25 percent (i.e.,
includes 125 percent of the dividend in income)
and takes a credit against his Federal individual
income tax for 66.7 percent of the amount of the
gross-up. Provincial individual taxes are calculated as approximately 50 percent of the shareholder's Federal tax liability (after the reduction
for the shareholder tax credit). Thus, the provincial tax is reduced by approximately 33.3 percent
of the amount of the gross-up, and the total value
of the shareholder credit against the combined
Federal and provincial liability of the shareholder
is approximately equal to the amount of the
gross-up. 8
The gross-up and credit are not dependent on
the payment of Canadian tax at the corporate
level. Thus, the shareholder credit may provide
full or partial relief from corporate level tax,
depending upon the tax rate applicable to the
corporation paying the dividend. If no tax is paid
at the corporate level, the shareholder credit
completely or partially offsets the shareholder
level tax, which is the only level of tax that would
otherwise apply to the distributed income. For
example, a dividend that is paid exclusively out of
preference income would carry the full credit, the
same as a dividend paid out of Canadian source
sales income. In the fonner case, the corporation
pays no Canadian corporate tax and, in the latter
case, it pays a net Federal tax of more than 28
percent.
The degree to which the Canadian system
integrates corporate and shareholder tax depends

Canada does not impose a compensatory or
withholding tax on dividends to resident shareholders.

Dividends Defined, Bonus Shares,
Share Repurchases
In general, a taxable dividend includes any
nonliquidating distribution with respect to shares
out of surplus funds. Accordingly, a return of
contributed surplus that has not been converted
into paid-Up capital is a taxable dividend. A
liquidating distribution constitutes a taxable
dividend to the extent it exceeds paid-up capital
(deftned to exclude contributed surplus).
A stock dividend is generally treated as a
taxable dividend. However, subject to certain
exceptions, the amount of the dividend is limited
to the increase in paid-up capital in respect of the
stock dividend.
A share repurchase generally is treated as a
taxable dividend to the extent that the amount paid
exceeds the paid-up capital on the shares repurchased. The amount so treated is excluded in
detennining the shareholder's capital gain or loss.
These rules, however, do not apply to a corporation's open market purchases of its shares.

Allocating Credits to Dividends
Because the shareholder credit is not dependent on the actual payment of corporate tax, the
Canadian system does not require rules allocating
credits to dividends.

165

Tax Rates

The Federal basic corporate rate is 38 percent.
Provincial basic corporate rates generally range
from 14 percent to 17 percent. However, an
abatement of Federal corporate tax is allowed in
respect of provincial tax equal to 10 percent of
taxable income earned in a province. In addition,
a surtax currently is imposed on corporations
equal to 3 percent of a corporation's Federal tax
liability. Thus, effective combined Federal and
provincial corporate tax rates vary from 42.8
percent to 45.8 percent.
For individuals, Federal tax rates are 17
percent for taxable income up to $28,784, 26
percent for taxable income of $28,784 to $57,578,
and 29 percent for taxable income in excess of
$57,578. 10 A Federal surtax of 5 percent is currently in place. Provincial tax is imposed as a
percentage of Federal tax, varying from 46.5
percent to 62 percent. Some provinces impose a
surtax on high-income individuals.
Corporate and individual taxpayers are taxed
at ordinary income rates on 75 percent of their net
capital gain in a taxable year. For individuals, a
lifetime exemption of $100,000 of gain applies.
The lifetime exemption is $500,000 for small
business shares and farm property. For individuals, in addition to actual realized gain, gain is
deemed to be realized with respect to many kinds
of assets at death, at the time of transfer by gift
or at the time the owner gives up Canadian
residence.

Treatment of Preference Income
Because the shareholder credit is not dependent on the payment of tax at the corporate level,
the Canadian system can be described as extending preferences to shareholders. However, because the Canadian system may provide less than
100 percent integration of the corporate and
shareholder taxes on distributed income, the
extension of preferences may be more than offset
by the remaining double tax on taxable income.
For example, for regular corporations the credit
generally equals half of Federal corporate tax.

Appendices

Thus, preferences are not extended to shareholders until preference income exceeds half of total
corporate income. 11
A 5 percentage point reduction in the basic
rate of corporate tax (from 38 percent to 33
percent) applies to manufacturing and processing
income of a resident corporation. For Canadian
small business corporations, a deduction applies
that effectively reduces the basic rate by 16
percentage points (from 38 percent to 22 percent).
Except for a 35 percent research and development
credit, investment tax credits apply only in selected geographic areas. A more generalized investment tax credit was phased out in 1988 as part of
tax reform. As discussed above, only 75 percent
of net realized capital gains are included in income. Certain assets are eligible for accelerated
depreciation.

Treatment of Domestic Intercorporate
Dividends
The gross-up and shareholder credit mechanism does not apply to dividends paid by a
Canadian corporation to a Canadian corporate
shareholder. In general, however, domestic
intercorporate dividends are deductible in com put ing the income of the Canadian shareholder corporation. 12 Thus, preferences generally are not
recaptured when preference income is distributed
to corporate shareholders. However, for Canadian
portfolio dividends received by a private or
privately-controlled Canadian corporation, a tax of
25 percent is payable by the recipient corporation
and is refunded to the corporation when the
dividends are redistributed to shareholders.

Treatment of Foreign Source Income
Resident corporations are taxed on their
worldwide income. This includes current taxation
on an accrual basis of passive income earned
through a controlled foreign affiliate. However,
Canada provides exemptions for certain types of
foreign source income and a foreign tax credit
with respect to certain other types of foreign
source income. For example, dividends received
from a foreign affiliate resident in a prescribed

Appendices

country out of its active business income in that
country or another prescribed country generally
are exempt from Canadian corporate tax. Tax
credits are allowed with respect to portfolio
dividends received from a nonresident corporation, but not for underlying foreign taxes paid by
that corporation on the income distributed. The
effect of these exemptions and credits is to relieve, in whole or in part, corporate level Canadian tax on foreign source income. Because the
shareholder credit does not depend on the extent
to which the underlying corporate income has
been taxed, the Canadian system extends the
benefits of integration to foreign source income to
the extent of the shareholder credit.

Treatment of Tax-Exempt
Shareholders
Certain persons are excluded from Canadian
tax, including charities and pension funds. However, because the shareholder credit is nonrefundable, tax-exempt shareholders do not receive the
benefit of Canadian integration.

166

income. Credits not used in the year received may
not be refunded or carried forward.

Streaming
The Canadian system includes stop-loss rules
that inhibit dividend stripping by requiring that, in
certain circumstances, the amount of a loss
recognized on a sale of shares be reduced by
dividends received on the shares.

In addition, the gross-up and credit mechanism
does not apply where a "dividend rental arrangement" exists. A dividend rental arrangement
essentially is a transfer of shares where the transferee receives the dividend but the transferor
retains the risk of loss and opportunity for gain
with respect to the shares. Finally, under a general anti-abuse rule, Canadian tax authorities may
deny a tax: benefit where there is an avoidance
transaction and a misuse of provisions of tax
laws. An avoidance transaction is a transaction
resulting in a tax benefit unless the transaction
reasonably could be considered to have been
undertaken primarily for non-tax reasons.

Treatment of Foreign Shareholders
Treatment of Interest
The Canadian integration system generally is
not extended to nonresident shareholders because
the gross-up and shareholder credit mechanism
does not apply to dividends paid to nonresident
shareholders. Dividends paid to foreign shareholders are subject to a withholding tax at a statutory
rate of 25 percent. By treaty, Canada typically
reduces the rate to 10 percent for direct investment dividends and to 15 percent for portfolio
dividends. The 1980 U.S. treaty, reflecting this
policy, was the first in which Canada reduced its
dividend withholding rate below 15 percent. This
concession for direct investment dividends in the
U. S. treaty was seen as extending to U. S. direct
investors in Canadian corporations some of the
benefit of Canadian integration.

Low-Bracket Shareholders
Excess shareholder credits are available to
offset income tax liability with respect to other

Interest paid by a Canadian corporation is
deductible if the interest relates to borrowed
money used for the purpose of earning income
from a business or property or for acquiring
property for gain upon resale. A thin capitalization rule prohibits the deduction of interest paid
by a thinly capitalized corporation to nonresident
shareholders owning 25 percent or more of any
class of the corporation's stock.
Interest income generally is taxable to resident
lenders. A withholding tax generally is imposed
on interest paid by Canadian corporations to nonresident lenders at the statutory rate of 25 percent.
No withholding tax is imposed with respect to interest paid on corporate bonds or debentures to an
arm's-length lender if no more than 25 percent of
the principal amount is repayable within 5 years
of issuance. In addition, the withholding rate may
be reduced by treaty to 10 or 15 percent.

167

B.3

FRANCE

Introduction
The French distribution-related integration
system combines three elements: (1) an imputation
credit (avoir fiscal), (2) a compensatory tax
(precompte mobilier), and (3) for 1989 through
1991, a "split" tax rate on corporate profits.
The avoir fiscal credit was enacted in 1965
and, simultaneously, a 24 percent withholding tax
on dividends was repealed. The new system
became fully effective in 1967.
In 1989, the French introduced a split rate
system, which applies a higher tax rate to distributed profits. The split rate system was designed to
provide an incentive for corporate fmancing
through retained earnings and balance the incentive, created by the avoir fiscal, to distribute
earnings and to fmance through new equity capital. This system has been eliminated, however,
beginning in 1992.

Description of Mechanics
Imputation Credits

Upon receipt of an eligible dividend (described
below), a French resident individual or corporate
shareholder is allowed a tax credit (the avoir
fiscal) equal to 50 percent of the amount of the
dividend, or 33.3 percent of the amount of the
dividend plus the avoir fiscal. A shareholder must
include in income both the amount of a dividend
payment and the amount of the avoir fiscal.
The gross-up and avoir fiscal partially integrate corporate tax paid on distributed income.
For 1991, distributed income is subject to a tax
rate of 42 percent at the corporate level. The
avoir fiscal, thus, equals 69 percent of the tax
paid by the corporation on distributed income and
29 percent of the pre-tax amount of such income.
For example, profits of FI00 are subject to
corporate tax of F42 prior to distribution, leaving
a net amount for distribution of F58. A shareholder would include a total of F87 (F58 + F29) in

Appendices

income. The avoir fiscal associated with this F87
dividend is F29. For 1992, distributed income
will be subject to corporate level tax at the rate of
34 percent. The avoir fiscal will thus equal 97
percent of the tax paid by the corporation on
distributed income and 33 percent of the pre-tax
amount of such income. 13
In order to encourage corporate distributions,
the avoir fiscal is not allowed to shareholders in
respect of dividends paid out of profits realized
more than 5 years prior to distribution. In addition, the avoir fiscal is not available to foreign
shareholders, unless specific provision is made in
an income tax treaty: If the amount of the avoir
fiscal exceeds the tax liability of an individual
shareholder, the excess is refunded. The same is
true for some tax-exempt shareholders. No refund
is available to a corporate shareholder.
Split Rate Tax and Compensatory Tax
(Precompte Mobilier)

The French split rate tax system, in effect for
1989 through 1991, is unusual in that it applies a
higher tax rate to distributed profits than to
retained profits. For fiscal years beginning on or
after January 1,1991 and before January 1, 1992,
retained corporate profits are taxed at a rate of 34
percent, and distributed corporate profits are taxed
at a higher rate of 42 percent. The additional 8
percent is imposed as a surtax in the year of
distribution. The application of a higher tax rate
to distributed profits was instituted for 1989
through 1991 to encourage corporate saving and
investment. Taking into account the avoir fiscal
credit allowed to shareholders, the effective
corporate level tax rate on distributed taxable
income is 13 percent for 1991. Consistent with
recent corporate tax rate reductions in the United
States and other EC countries, however, the
French government recently eliminated the ~
percent surtax on distributed income.
The precompte mobilier is imposed on a
distributing corporation in respect of dividends
distributed (l) out of profits that have not borne
regular corporate income tax at the 34 percent
rate, e.g., foreign source income, preference

Appendices

income, and dividends received by a parent
company from a 10 percent owned subsidiary or
(2) from fully-taxed profits earned more than 5
years before the distribution. 14 The precompte
mobilier is imposed at a rate of 50 percent of the
amount of the dividend, or 33.3 percent of the
dividend plus the precompte mobilier. Thus, the
amount of the precompte mobilier is equal to the
amount of the avoir fiscal associated with the
dividend. No distinction is made in calculating
precompte mobilier liability between income that
is not taxed and income that is taxed at a rate
lower than 34 percent. 15
French corporations are required to segregate
fully-taxed income from income potentially subject to the precompte mobilier for tax accounting
purposes. In general, dividends eligible for avoir
fiscal are deemed to be distributed fITst out of
current fully-taxed income, and then out of fullytaxed retained income of each of the immediately
preceding 5 years. Once fully-taxed income for
this 5 year period has been exhausted, a corporation may choose to allocate a dividend distribution
to (1) dividends received from foreign subsidiaries, (2) the long-term capital gains reserve, or (3)
other miscellaneous preference income in any
order. France thus allows stacking of dividends
last against preference income.
A French corporation may elect, alternatively,
to allocate part or all of a distribution eligible for
the avoir fiscal first against dividends received
from a French subsidiary within the last 5 years
(rather than to current taxable income). Dividends
received from French subsidiaries are subject, in
principle, to the precompte mobilier. On redistribution, however, the avoir fiscal associated with
such dividends may be credited against the
precompte mobilier liability.

168

shareholders are considered to be repayments of
share capital only if all of the corporation's
earnings and reserves previously have been
distributed.
Distributions in liquidation are taxed as ordinary dividends to the extent the distribution
exceeds the greater of contributed capital or share
basis, and are eligible for the avoir fiscal. To the
extent that liquidating distributions are deemed
made from preference income, they are subject to
the precompte mobilier.
Stock dividends generally are not subject to
tax in the hands of a recipient. However, if the
distribution of new shares is the result of a reinvestment of cash dividends at the election of the
shareholder, the distribution is taxed as an
ordinary dividend distribution.
Proceeds from share repurchases are treated as
distributions, although only the difference between
the value of consideration received and the shareholder's basis in the shares is subject to tax at the
shareholder level. The amount distributed does
not qualify for the avoir fiscal or trigger the
precompte mobilier unless it is paid on a pro rata
basis to all shareholders in accordance with a
regular decision made by the corporation.

Allocation of Credits· to Dividends
The avoir fiscal applies regardless of the rate
of corporate level tax actually borne by distributed
lllcome.

Tax Rates

Dividends Defined, Bonus Shares,
Shnre Repurchases

For the 1991 tax year, individual marginal
income tax rates range from 5 percent to 56.8
percent. France also imposes a net wealth tax at
rates, for 1991, ranging from 0.5 percent to 1.5
percent.

Distributions are eligible for the avoir fiscal if
they are made from corporate income, are made
pro rata to shareholders, and are based on a
regular decision of the corporation. Repayments
of share capital are not taxable, but payments to

For fiscal years beginning on or after January 1, 1991 but before January 1, 1992, undistributed profits are taxed at a flat rate of 34 percent
and distributed profits at a flat rate of 42 percent.
The higher rate applicable to distributed profits

Appendices

169

does not apply to profits distributed in the fonn of
a stock dividend. For fiscal years beginning on or
after January 1, 1992, all corporate profits (distributed and undistributed) will be taxed at a flat
rate of 34 percent.
Net short-tenn capital gains (generally, gains
on the sale of assets held less than 2 years) are
included in taxable income and taxed at regular
rates in the year realized (subject to certain
exceptions that allow gains arising from mergers
or similar reorganizations to be spread over
periods from 5 to 15 years). Net short-tenn
capital losses are either deductible from operating
profits in the year realized or, for a loss corporation' added to the net operating loss (and thereby
made available for 5 year carryforward or an
elective 3 year carryback).
For dispositions occurring prior to July 1,
1991, net long-tenn capital gains generally are
taxed at a rate of 25 percent. Long-tenn capital
gains on property other than buildings, land and
fmancial instruments are taxed at 19 percent and
long-tenn capital gains on industrial property
(e.g., patents) are taxed at 15 percent. Net longtenn capital losses may not be used to offset
operating profits, but may be carried forward for
10 years to offset future long-tenn capital gains.
The after-tax amount of net long-tenn capital gain
is credited to a special capital gain reserve. When
a dividend is deemed distributed out of the capital
gain reserve, a compensatory tax is imposed at a
rate of 17 percent, equal to the difference between
the long-tenn capital gains tax rate (25 percent)
and the tax rate applicable to distributed profits
(42 percent). For dispositions occurring on or
after July 1, 1991, the French government has
replaced the multiple rates on capital gains with a
single 18 percent rate. Compensatory tax will thus
be imposed at a rate of 16 percent for 1992, equal
to the difference between 18 percent and the 34
percent rate applicable to distributed profits.

certain building companies, software acquired
from third parties, research installations, and air
and water purification installations. Corporations
also may be entitled to a tax credit for research
and development expenditures, a tax holiday for
start-up businesses, and a reduced rate of tax on
French headquarters of multinational corporations.
Preferences are not passed through to shareholders, since the precompte mobilier is imposed
on distributions of preference income. However,
as described above, French law allows preference
income to be stacked last.

Treatment of Domestic
Intercorporate Dividends
Nonparent Companies
"Nonparent companies" are defmed as companies that own less than 10 percent of the issued
share capital of the distributing corporation.
Nonparent companies are eligible for the avoir
fiscal. Like an individual shareholder, a nonparent
company must include in income the entire
amount of a dividend received from another
French company and may use the avoir fiscal
associated with the dividend as a credit against its
income tax liability. If, however, the nonparent
company's income tax liability for the year in
which a dividend is received is less than the
amount of the avoir fiscal, no refund or carryforward is allowed.
Parent Companies

Treatment of Preference Income

"Parent companies" are defmed as companies
that own 10 percent or more of the shares of the
distributing corporation. Parent companies are
eligible for a "participation exemption" as well as
the avoir fiscal. Under the participation exemption, 95 percent of the amount of a dividend
received from a 10 percent-owned subsidiary
(including the amount of the avoir fiscal) is
excludable from taxable income. 16

Tax preferences available at the corporate
level include special accelerated depreciation for
new construction in depressed areas, shares in

The avoir fiscal associated with dividends
received by a parent company from its subsidiaries is passed on to the parent's shareholders when

Appendices

the dividends are redistributed. In principle, the
precompte mobilier applies to such redistribution,
because the subsidiary dividends are almost
entirely exempt from tax. The parent company is
pennitted a deduction, however, for the avoir
fiscal associated with the subsidiary dividends and
this deduction exact! y offsets the parent's
precompte mobilier liability. Any available credit
for foreign withholding tax paid on the subsidiary
dividends also may be used to offset the
precompte mobilier. As a result, the shareholders
of the parent company are placed in the same
position as if they had owned shares in the subsidiaries directly.

Consolidated Groups
A French parent company may consolidate for
tax purposes with its direct and indirect 95 percent-owned French subsidiaries. Dividends paid
within the consolidated group are subject neither
to precompte mobilier nor to corporate income
tax.

Treatment of Foreign Source Income
In general, the French integration system does
not extend the benefits of integration with respect
to foreign income taxes imposed on foreign source
Income.
Profits earned by a French company through
a foreign branch or other penn anent establishment
generally are excluded from taxable income until
they are repatriated to France and distributed to
shareholders. Upon distribution of these profits,
the precompte mobilier is imposed. However, if
a branch profits tax is imposed on the branch
income in addition to foreign income tax, and
provided the branch is located in a treaty country,
the French corporation may credit the branch
profits tax against the precompte mobilier. 17
A French nonparent company is taxed on the
net amount of a dividend received from a foreign
corporation (after deduction of foreign withholding tax) resident in a nontreaty country and may
not credit any foreign withholding tax against its
corporate tax liability. Where the foreign

170

corporation is resident in a treaty country, the
dividend is grossed up for any foreign withholding
tax, which is then allowed as a credit against
French corporate tax. Dividends paid by the
nonparent company out of foreign source dividend
income are subject to the precompte mobilier and
qualify for the avoir fiscal.
Under the participation exemption, 95 percent
of the amount of a dividend received by a French
parent company from a 1 percent -owned foreign
subsidiary (including the amount of the avoir
fiscal) is excludable from taxable income. Foreign
withholding tax is not allowed as a credit against
French corporate tax on the foreign source dividend. The precompte mobilier is imposed on, and
the avoir fiscal applies to, dividends paid by the
French parent company out of foreign source
dividends. However, where the foreign subsidiary
is resident in a treaty country, the amount of the
dividend received by the French parent company
is grossed up by the amount of any foreign withholding tax, which may then be credited against
the precompte mobilier due upon the redistribution of the foreign source dividend (provided the
redistribution occurs within 5 years of the receipt
of the foreign source dividend).

°

As of January 1, 1990, special rules apply to
French holding companies. A French holding
company is exempt from the precompte mobilier
upon redistribution of dividend income received
from foreign subsidiaries to its shareholders, if
the holding company satisfies three requirements:
(l) the exclusive purpose of the holding company
is to hold shares in other companies, (2) at least
two-thirds of the capital assets of the holding
company consist of interests in foreign subsidiaries, and (3) the holdiJ1g company derives at least
two-thirds of its accounting profit (excluding
capital gains) from such foreign interests. Generally, the French holding company must hold at
least a 10 percent interest in a foreign subsidiary.
Dividends distributed by a qualifying French
holding company out of dividends received from
foreign subsidiaries are not eligible for the avoir
fiscal, but give rise to a tax credit equal to any
foreign withholding tax imposed on the foreign

Appendices

171

subsidiary dividends. If such dividends are redistributed to a holding company shareholder residing in a nontreaty jurisdiction, the standard 25
percent withholding tax imposed on dividends is
increased to 50 percent. 18

Treatment of Tax-Exempt
Shareholders
Pension funds, charities, and other tax-exempt
organizations are not taxed on dividends received
from French corporations, but are subject to tax
at a reduced rate of 24 percent with respect to
certain types of investment income, including
dividends received from foreign corporations.
Tax-exempt organizations generally are not
eligible for the avoir fiscal. However, retirement
and disability benefit funds, as well as certain
foundations and associations of "public utility,"
are granted a refundable avoir fiscal with respect
to dividends received from French corporations.

Treatment of Foreign Shareholders
Dividends paid by a French company to a
foreign shareholder are subject to French withholding tax at a rate of 25 percent, subject to
reduction by treaty. The avoir fiscal is not generally available to foreign shareholders (whether
individuals or corporations). This is the case even
if a French corporation distributes income subject
to, and pays, the precompte mobilier.
France has extended the avoir fiscal (by means
of a cash refund) to shareholders of a French
corporation who are resident in some treaty
countries and who (1) are subject to income tax in
their residence country on dividends received
from the French corporation and (2) do not
qualify for exemption or foreign tax credit relief
in respect of deemed-paid foreign corporate taxes,
i.e., individuals and corporate portfolio investors.
The avoir fiscal refund is subject to French
withholding tax at a rate of 25 percent, subject to
reduction by treaty. Under some treaties, 10
percent corporate shareholders (nonportfolio
shareholders) and other nonresident shareholders

not entitled to the avoir fiscal refund are allowed
a refund (subject to withholding tax) of any
precompte mobilier imposed in respect of
dividends received.
Under the United States treaty, for example,
the avoir fiscal is refunded to shareholders who
are either (1) United States resident individuals or
(2) United States corporations that own less than
10 percent of the issued share capital of the
distributing corporation and that do not qualify for
the indirect foreign tax credit under IRe § 902
(corporate portfolio shareholders). The avoir
fiscal is treated as an additional dividend amount
and is subject to a 15 percent withholding tax.
United States corporations that are 10 percent
shareholders of the distributing corporation (nonportfolio shareholders) are not eligible for an
avoir fiscal refund, but are entitled to a reduced 5
percent withholding rate on dividends and to a
refund of the precompte mobilier.

Treatment of Low-Bracket
Shareholders
The avoir fiscal is refundable to low-bracket
shareholders.

Streaming
France does not have specific rules to prevent
streaming, although the avoir fiscal is available
only with respect to ~ distribution made pro rata
to all shareholders.

Treatment of Interest
Interest paid to third parties who are not
shareholders and who do not have legal or effective control over the payor is deductible at the
corporate level.
Interest from corporate indebtedness is generally included in the taxable income of a resident
lender (collected in part by withholding). Resident
individuals holding certain fixed income securities
may elect to have interest taxed at a flat rate
collected by withholding. For 1991, the flat rate
is 27 percent for income from bonds.

Appendices

Interest from corporate indebtedness generally
is subject to a withholding tax imposed at statutory rates from 25 percent to 51 percent. However,
interest on bonds paid abroad is exempt. Reduced
treaty rates also may apply.

B.4

GERMANY

172

distributing corporation is held publicly or privately, or by domestic or foreign shareholders.
(The effect of tax treaties on withhOlding is
discussed below.) In some circumstances the
government will grant an exemption certificate to
the shareholder which, when provided to the
withholding agent, will exempt the shareholder
from withholding.

Introduction
The Gennan integration system has both a
split rate tax and an imputation credit system with
a compensatory tax. The split rate tax applies a
"statutory" rate (currently 50 percent) to retained
income and a lower "distribution" rate (currently
36 percent) to distributed income. The imputation
credit mechanism imputes to shareholders the
corporate level income tax paid on distributed
income. In general, the shareholder receives a
credit based on the distribution rate regardless of
the corporation's actual tax liability. However, as
discussed more fully below, the corporation may
become liable for compensatory tax if it has not
paid tax on distributed income at the full
distribution rate.

Description of Mechanics
Imputation Credits
Imputation credits are available to any shareholder subject to German tax on his worldwide
income. This generally excludes nonresident
aliens, foreign corporations, and domestic entities
not subject to Gennan tax (although imputation
credits are available to a foreign corporation or
nonresident that holds the shares as part of a
permanent establishment in Germany).
In general, dividends are subject at the corporate level to a creditable 36 percent distribution
tax (described below) and to a 25 percent withholding tax at the corporate level. The withholding tax is imposed on the amount of the declared
distribution. Thus, a distribution of DM64 is
reduced by DM16 of withholding tax, leaving a
cash distribution of DM48. The withholding tax
applies without regard to whether the stock of the

The shareholder inust gross up the amount of
the dividend by the amount of the withholding tax
plus the imputation credit (equal to 36/64 of the
declared distribution). Thus, a cash dividend of
DM48 (net of withholding tax) is grossed up to
DM64 (for the withholding tax), and the resulting
DM64 is then grossed up to DMIOO. The shareholder reports the grossed-up distribution as
income and claims a credit equal to the amount of
the total gross-up. If the credit exceeds the shareholder's tax liability, the shareholder receives a
full refund of the excess; if the shareholder's tax
liability exceeds the credit, the shareholder must
pay the excess.

Compensatory Tax
The German system uses an "available net
equity" account to track taxable and preference
income with both the .split rate tax and the imputation credit mechanisms. Available net equity
represents after-tax corporate income and certain
other balance sheet items available for distribution. Available net equity is divided into baskets
representing the rate at which the income was
taxed. These "Eigenkapital" (equity capital)
baskets, abbreviated "EK," are:
•

EK 56, containing available net equity from income
taxed at the pre-1990 statutory rate of 56 percent.
(As of January 1, 1995, the balance in this basket
will be "emptied" into the EK 50 basket at a rate
equal to 56/44 of the amount in the EK 56 basket.)

•

EK 50, containing available net equity from income
taxed at the post-1989 statutory rate of 50 percent.

•

EK 36, containing available net equity from lesser
taxed income that has been converted into an
equivalent amount of income taxed at 36 percent,
and thus matches the distribution rate of 36 percent
(see discussion below).

Appendices

173

EK 0, containing available net equity from income
subject to no corporate tax. EK 0 is further divided
into four categories: EK 01, containing foreign
source income realized after 1976 (the imputation
credit became effective in 1977), EK 02, containing items not included in EK 01, 03 or 04, for
example net operating losses (discussed below) and
distributions made when there is no available net
equity in any category (in the latter case, the
corporation pays the 36 percent compensatory tax
and includes the distribution in EK 02 as a negative
item, permitting the corporation to later distribute
an offsetting amount of EK 0 without a compensatory tax), EK 03, containing available net equity
from years before 1977, and EK 04, containing
shareholder contributions to capital in years after
1976.

Fully-taxed income (EK 56 or 50 income) is
considered distributed flrst, followed next by EK
36 income, and last by EK 0 income.
Germany implements its split rate tax by
refunding to corporations the excess tax paid on
distributions out of EK 50 and EK 56Y Distributions out of EK 36 generate neither a refund
nor extra corporate tax. Distributions out of EK 0
(other than EK 04) are subject to a compensatory
tax of 36 percent. If the corporation has DMI00
in its EK 01 account, for example, it may pay the
shareholder only DM64-the original DMI00 in
the account net of a 36 percent distribution tax.
The additional tax is added to the corporation's
total tax liability for the year to which the distribution is assigned. Distributions out of EK 04
(contributions to capital) generate no tax to the
corporation and are excluded from the shareholder's income (as a return of capital). 20
There are generally no time limits on relief.
Thus, a distribution from EK 56 earned in 1977
produces the same credit for corporate tax paid as
a distribution from EK 56 earned in 1989. This
means that the available net equity accounts need
not be segregated into vintage accounts, and
instead may be kept as "pools." Income earned
prior to 1977, however, is placed in the EK 03
category, and thus there is no imputation relief at
the shareholder level for German corporate taxes
paid on such income.

A corporation might actually pay tax on
certain income at rates other than those for which
corresponding EK categories exist. (A substantial
portion of such income is foreign source income,
discussed below.) The German imputation system
converts income subject to some other effective
tax rate into appropriate amounts of EK 50, EK
36, and EK 0 income. The conversion formula
maximizes the amount of pre-tax income converted into income taxed at the 36 percent distribution
rate, since distributions from EK 36 neither entitle
the corporation to a refund nor require the payment of compensatory tax. If the corporation's
effective tax rate exceeds 36 percent, the remainder of its income is converted to EK 50 income;
but if the corporation's effective tax rate falls
short of 36 percent, the remainder of its income
is converted to EK 0 income. 21 For example, if
the corporation has pre-tax income of DM 100 on
which it pays German tax of DM40, then the
effective tax rate is greater than .36 (40/100 =
.40), and so a portion of the income will be
converted into income taxed at the 50 percent
statutory rate. 22 By contrast, if the corporation
has pre-tax income of DMIOO on which it pays
German tax of DM25, then the effective tax rate
is less than .36 (25/100 = .25), and so a portion
of the income will be converted into income taxed
at a zero rate. 23
The EK accounts are determined at the end of
the taxable year. 24 A distribution is classifled
according to the accounts for the year preceding
the year of the dividend declaration.
Dividends Defined, Bonus Shares,
Share Repurchases

Any distribution of cash or property (whether
liquidating or nonliquidating) is a taxable dividend
for German tax purposes unless it is a distribution
out of EK 04 or otherwise is a repayment of share
capital.
Stock dividends are not subject to the distributions tax and are not taxable to shareholders.
However, in certain circumstances, distributions

Appendices

in reduction of share capital within 5 years of the
stock dividend (to the extent not in excess of the
increase in share capital resulting from the stock
dividend) are taxable as dividends and are subject
to a penalty tax.
Stock corporations generally are prohibited
from making share repurchases under Gennan
corporate law. A GmbH is pennitted to make
share repurchases but is effectively required to
finance them out of retained earnings (as opposed,
for example, to borrowing against unrealized
appreciation in its assets). Share repurchases are
not subject to distribution tax at the corporate
level and are capital gains transactions at the
shareholder level.

Allocation of Credits to Dividends
As discussed above, Germany applies a unifonn rate for purposes of determining the shareholder credit regardless of the rate of corporate
tax that the distributed income has actually borne.

Tax Rates
Before 1990, individual marginal rates ranged
from approximately 22 percent to 56 percent
(effective for income exceeding DMI30,000).
Beginning in 1990, marginal rates range from
approximately 19 percent to 53 percent (effective
for income exceeding DM120,000).

174

Gains from sales of stock by individuals are
exempt unless (1) the sale is connected with a
business, (2) the stock is held 6 months or less, or
(3) the shareholder owned more than 25 percent
of the company's stock at some time during the
preceding 5 years. Business and short-tenn gains
are taxable to individuals at normal rates, except
that short-term gains are exempt up to DM1,000
each year. Short-term losses may be netted against
short-term gains. Gains by substantial individual
shareholders are taxed at one-half the normal rate
up to the first DM30 million of net gain and at
the normal rate thereafter. Gains from exchanges
of stock in a liquidation or redemption are treated
as sales (except for any portion that is taxed as a
dividend distribution).
Gains from sales of stock by corporations are
taxable as ordinary income.
Effective for the period July 1, 1991, through
June 30, 1992, Germany has imposed a tax on
each taxpayer equal to 7.5 percent of the tax that
such person would otherwise pay. The surtax
applies to all individual and corporate taxpayers,
foreign shareholders subject to dividend withholding tax, and German branches of nonresident
corporations. For taxpayers using a calendar
taxable year, the surtax has the effect of a 3.75
percent surtax in each of the 1991 and 1992
taxable years.

Treatment of Preference Income
There is a flat rate of 50 percent for retained
profits (before 1990, the rate was 56 percent).
This rate is reduced to 36 percent for distributed
profits. Certain German "public banks" (banks
generally owned by municipal or other public
bodies) and Gennan branches of foreign corporations are subject to a flat rate of 46 percent (pre1990, 50 percent). (See below for a discussion of
Gennan branches.) Income sourced in the former
western sector of Berlin is subject to a special tax
rate of 38.75 percent (pre-1990, 43.4 percent).
West Berlin branches of foreign corporations are
subject to a special tax rate of 35.65 percent (pre1990, 38.75 percent). The special tax rate for
such income, however, is being phased out over
a number of years as a result of unification. 25

Investment incentives in Germany generally
take the form of accelerated depreciation for
certain industries or regions of the country; there
is no investment tax eredit. There are special low
corporate rates for income derived from the
fonner western sector of Berlin (these rates,
described above, are being phased out). Government "incentive grants" (in the form of cash
awards) are awarded in certain cases (usually
related to research and development and energy
production).
The benefit of preferences that take the fonn
of accelerated depreciation is not extended to
shareholders. The benefit is eliminated through

Appendices

175

the 36 percent compensatory tax applied to distributions out of EK O. Preferences are stacked
according to the EK accounts, as indicated. Thus,
fully-taxed income (EK 56 or 50) is distributed
first, and EK 0 is distributed last.
The benefit of the current reduced tax rate for
West Berlin income is extended to shareholders.
Such income is deemed to have borne the full
imputation burden (EK 50 or EK 56).

Treatment of Domestic
Intercorporate Dividends

were subject to a rate of German tax equal to the
residual tax divided by the net profits.

Treatment of Tax-Exempt
Shareholders
In Germany, the tax-exempt sector is divided
into two separate groups for tax purposes: (1)
public law corporations or bodies, e.g., the
government and certain central banks, and (2)
charitable organizations, including religious
groups. Charitable organizations are exempt at the
shareholder level, but public corporations are
subject to one-half of the normal withholding tax
of 25 percent.

Dividends paid to domestic corporations are
treated exactly the same as dividends paid to
resident individuals. The dividends are subject to
the 36 percent distributions tax. The recipient
cOJporation must include the grossed-up distribution in income and is entitled to claim the imputation credit. Therefore, preferences are recaptured
at the corporate level on intercorporate dividends.
No exemption from these rules is provided even
where the distributing corporation is a subsidiary
of the recipient corporation.

In general, neither group is entitled to the
imputation credit. However, the imputation credits
are refunded where the dividend is paid out of EK
01 (foreign source income that has not borne
German tax) or EK 03 (pre-1977 profits). In
addition, all shareholders (except shareholders of
foreign corporations with German branches)
benefit from the 36 percent distribution rate on
distributed profits.

Treatment of Foreign Source Income

Treatment of Foreign Shareholders

German corporations are subject to German
corporate tax on their worldwide income. However, Germany has two methods for relieving double
taxation with respect to foreign profits: by statute,
it gives a foreign tax credit and, by treaty, it
exempts foreign business profits earned by a
domestic corporation (and gives no credit).

Income distributed to foreign shareholders,
like all other income, is taxed at the corporate
level at the distribution rate rather than at the
statutory rate. No distinction is made, for this
purpose,
between portfolio and direct
shareholders.

The foreign tax credit is not treated as tax
paid for purposes of the imputation credit. In
effect, foreign taxes are treated as deductible
expenses for purposes of applying the imputation
system. If the profits are covered by a treaty
exemption, then the profits (net of foreign tax) are
simply placed in EK 01 and are subject to the 36
percent distribution tax when paid to shareholders.
If the profits are not covered by a treaty exemption, they are subject to a residual German corporate tax, as in the United States. In applying the
imputation system to this latter class of profits, it
is assumed that the profits (net of foreign tax)

Dividends to direct and portfolio foreign
shareholders are subject to the statutory withholding tax of 25 percent, except where reduced by
treaty. Treaties frequently reduce the rate from 25
percent to 15 percent for direct corporate shareholders that are residents of the treaty partner. In
some cases, the reduction applies to all residents
of the treaty partner. (This was, for example, the
treatment provided in the 1954 U.S.-Gennany
treaty).26 Foreign shareholders also will be
subject to the 7.5 percent surtax previously described. The surtax will be refunded to shareholders entitled to limited withholding under a tax
treaty.

176

Appendices

In general, foreign shareholders are not entitled to the imputation credit, and the withholding
tax applies to the dividend without gross-up.
Although Gennany has not extended the imputation credit to foreign shareholders, it has been
willing to reduce withholding rates by treaty, in
part in recognition of the benefits of its imputation
system to resident shareholders. In the new U.S.Gennany treaty that entered into force on August
21, 1991 (generally effective retroactive to January I, 1990), Gennany grants a 5 percent withholding rate for direct corporate shareholders (1
percent prior to 1992) and a 1 percent withholding rate for U.S. portfolio shareholders. Under
the treaty, the United States agreed to treat the
additional relief for portfolio investors as a dividend resulting from a refund of Gennan corporate
tax equal to 5.88 percent of the declared dividend;
the entire amount (declared dividend plus refund)
is considered to have been subject to a 15 percent
Gennan withholding tax. Thus, for U.S. tax
purposes, if a Gennan corporation declares a
dividend of DMI00 payable to a U.S. individual
shareholder, the dividend will, in effect, be
grossed up to DMI05.88. After application of a
15 percent withholding rate, the shareholder will
receive a net amount of DM90 and be eligible for
a foreign tax credit of DM15.88.

°

°

Foreign shareholders are entitled to a refund
(subject to withholding tax) of the 36 percent
distribution tax imposed on two types of distributions: (1) distributions out of foreign source
income and (2) distributions out of domestic
source income earned prior to the adoption of
integration in 1977. The refund is only for the 36
percent distribution tax, not for the foreign or
pre-1977 taxes. Refunds paid to foreign shareholders with respect to such distributions are
subject to 25 percent withholding unless a treaty
provides for a reduced rate. In the latter case, the
reduction is granted directly by the government,
eliminating the need to apply for a refund of
excess withholding.
Gennan branches of foreign corporations are
subject to a corporate tax rate of 46 percent (pre1990, 50 percent). There is no reduction in the

corporate rate when the profits are remitted to the
home office or distributed to the foreign corporation's shareholders (nor is there imposed a
branch tax, as under IRe § 884); the distribution
of the profits to the shareholders is not subject to
Gennan withholding; and the shareholders are not
entitled to any imputation credit with respect to
the Gennan corporate taX.27

Treatment of Low-Bracket
Shareholders
As discussed above, excess credits are fully
refundable to low-bracket shareholders.

Streaming
An anti-streaming rule applies where (1) a
shareholder sells a substantial interest in a German corporation (i. e., shares with a value of more
than DMIOO,OOO), (2) the shareholder is not
entitled to the shareholder credit (i.e., a taxexempt or foreign shareholder), (3) the shareholder sells the shares to a person entitled to the credit
(i.e., a Gennan resident), and (4) the gain realized on the sale is not subject to German tax. In
such case, the acquiror is not allowed to recognize
loss on disposition of shares within 1 years to
the extent the loss is attributable to dividends paid
by the Gennan corporation.

°

Treatment of Interest
Interest paid by German corporations on
indebtedness incurred for business purposes
generally is deductible. However, interest paid by
an undercapitalized subsidiary to a related party
may be recharacterized as a hidden dividend.
Interest paid by Gennan corporations to
resident lenders is includable in income. Interest
paid by German corporations to nonresident
lenders generally is not subject to any Gennan
withholding tax. Interest paid on participatory or
convertible bonds, however, is subject to withholding at a statutory rate of 25 percent rate.
Lower treaty rates or treaty exemptions may
apply.

Appendices

177

B.5

NEW ZEALAND

Introduction
New Zealand adopted an imputation credit
system beginning with the tax year starting
April 1, 1988.

"Allocating Credits to Dividends, below. A
closing debit must be cleared within two months
by making a "further income tax" paymentavailable to offset future income tax liabilities, but
not arrears-and also results in a 10 percent
penalty.28 See also "Streaming" below.
II

Compensatory or Withholding Tax

Description of General Mechanics
Imputation Credits
For purposes of shareholder level taxation, the
amount of a dividend includes the amount of
imputation credits that the corporation allocates to
the dividend (see "Allocating Credits to Dividends," below) from its "imputation credit account" (lCA). The imputation credits are then
creditable against shareholder tax liability. Excess
credits are not refundable but do convert into an
equivalent loss carryforward.
The New Zealand system requires every
taxable domestic corporation to maintain an ICA.
The ICA is a memorandum account that runs
from April 1 to March 31, regardless of the
corporation's fiscal year. The first imputation year
ran from April 1, 1988, to March 31, 1989.
Unlike Australia's year-to-year franking account,
the ICA is a continuing account, and so a negative
year-end balance in the ICA results in a tax levy.
The ICA is credited when the corporation pays
New Zealand income tax or receives imputation
credits attached to dividends paid by another
corporation. Where a refund of tax becomes due
because of a revised tax assessment, the amount
of the refund available is limited to the closing
balance of the ICA for the previous year. The
amount of a refund in excess of the balance is
carried forward and may be used to reduce future
tax liability of the corporation.
The ICA is debited when the corporation
attaches imputation credits to dividends paid to
shareholders, receives refunds of New Zealand
mcome tax, or alters its credit ratio without
making a ratio change declaration. See

New Zealand does not impose a compensatory
tax. Recently, New Zealand introduced a withholding tax for dividends paid to residents that do
not carry imputation credits. Technically, the
resident withholding tax is imposed on all dividends at a rate of 33 percent (the higher individual marginal rate), but an offset is allowed to the
extent the corporation is passing through imputation and foreign source dividend withholding
payment credits allocated to the dividend. (See
"Treatment of Foreign Source Income" below for
a discussion of the "dividend withholding payment" relating to foreign source dividends.) As
with imputation credits, the amount of the dividend includes the resident withholding tax paid
and the withholding tax is creditable against
shareholder tax liability. However, excess resident
withholding tax credits are refundable.

Dividends Defined, Bonus Shares,
Share Repurchases
In general, all nonliquidating distributions to
shareholders are treated as taxable dividends
(under corporate law, the corporation cannot pay
dividends as a return of capital without a court
order); on liquidation, the amount in excess of
paid-up capital is a dividend.
A taxable bonus issue, although technically not
a dividend, may carry imputation credits. A
corporation with profits essentially may elect
whether to treat a bonus issue as taxable. In
addition, a bonus issue is taxable if shareholders
may elect to receive cash in lieu of stock. Howev~
er, the importance of bonus issues as a mechanism for extending the imputation system to
retained earnings is reduced, because, as
described under "Tax Rates," below, New

Appendices

Zealand does not impose tax on capital gains
(including gains on sales of stock of New Zealand
corporations) .
In the case of share repurchases, the amount
tr~ated as a dividend is limited to the excess of
the amount paid over the sum of the stated capital
and qualifying premium with respect to the share.
The qualifying premium is equal to the proportionate share of the subscription premium paid on
issuance of the class. The limitation applies,
however, only if the Inland Revenue Department
is satisfied that the shares are not being redeemed
pursuant to an arrangement to redeem shares in
lieu of the payment of dividends.

Allocating Credits to Dividends
New Zealand's imputation statute does not
require a corporation to allocate any credit to a
dividend, but certain allocation rules significantly
limit a corporation's flexibility to reduce opportunities to stream imputation credits to shareholders
who can best use them. The maximum amount
that can be allocated to a dividend is determined
by multiplying the dividend by a fraction, the
numerator of which is the corporate tax rate and
the denominator of which is one minus the corporate tax rate. Once the corporation allocates
credits to a dividend, the corporation has established the "benchmark" imputation ratio, and the
corporation must generally use the same ratio in
allocating credits to any other dividend paid in the
same imputation year on any class of stock. The
corporation may change its ratio, if it files with
the Inland Revenue Department a "ratio change
declaration" showing that the change is made for
commercial reasons and not to convey an imputation credit benefit to one group of shareholders
over another. If the corporation uses a ratio
different from the benchmark and has not filed a
ratio change declaration, it must debit its ICA by
the amount by which the account would have been
debited if all dividends that year had been credited
at the highest rate used that year. Additional tax
and penalties are due if, as a result, the closing
balance is negative.

178

Tax Rates
The corporate tax rate is currently 33 percent. Individuals pay tax at two marginal rates: 24
and 33 percent. The 33 percent rate applies to
individuals with taxable incomes exceeding NZ
$30,875, adjusted for inflation. New Zealand
currently imposes no tax on capital gains.

Treatment of Preference Income
Because a corporation may attach credits to
dividends only to the extent of taxes actually paid
by it, corporate tax preferences generally are not
extended to shareholders. When preference
income is distributed as an uncredited dividend,
the amount of the dividend, in general, is subject
to resident withholding tax. However, subject to
the credit allocation limitations described above,
a corporation may choose the order in which
taxable income and preference income are considered distributed. In addition, New Zealand recently attempted to eradicate most tax preferences.
Various concessions remain for certain industries,
most relating to timber, livestock, farming and
fishing. New Zealand also offers certain export
incentives. The research and development deduction is 100 percent, with special rules for
depreciable property.

Treatment of Domestic
Intercorporate Dividends
Until April 1, 1992, corporations are exempt
from tax on the receipt of domestic source dividends. Any imputati.on credits attached to such
dividends are credited to the recipient's leA and
may be used to frank dividends to its shareholders. The effect of this system is preserve corporate tax preferences until preference income is
distributed out of corporate solution.
Under a recent decision of the New Zealand
Government, domestic source dividends are not
exempt from tax when received by a corporate
shareholder on or after April 1, 1992. Instead, the
normal gross-up and credit rules apply and a

179

corporate shareholder thus will be taxed on the
receipt of an unfranked, domestic source dividend. The reason for this change is to prevent
corporations with tax losses from effectively
transferring the losses to corporate shareholders
through the issuance of redeemable preference
shares and using the proceeds to invest in interestbearing securities. Another effect of the change is
to recapture preferences on the distribution of
preference income to a corporate shareholder.
To mitigate the effect of the repeal on affiliated groups of corporations and for other reasons,
a group of corporations with 100 percent common
ownership is allowed to consolidate for tax purposes. A consolidated group would maintain a
single ICA and intercorporate dividends would be
ignored.

Treatment of Foreign Source Income
Foreign source income other than dividends is
includable in income, and New Zealand allows a
credit for foreign taxes paid. Because a corporation credits its ICA only with any additional New
Zealand corporate tax paid, foreign taxes do not
give rise to imputation credits, and dividends to
shareholders of a New Zealand corporation paid
out of foreign source nondividend income are
exposed to a second level of tax. Foreign source
dividends received by New Zealand corporations
are exempt from tax but are subject to a "dividend
withholding payment" as described below. Foreign taxes paid on the dividend generally are not
added to the ICA and, accordingly, dividends paid
to shareholders of the New Zealand corporation
out of foreign source dividend income also are
subject to shareholder level tax. Special rules
apply to income derived from controlled foreign
corporations (CFCs). The net effect of the New
Zealand system is the equivalent of allowing a
deduction for foreign taxes on distributed foreign
source income earned through a New Zealand
corporation.
Dividend WzthholLiing Payment Account (WPA)

New Zealand enacted a withholding payment
system (at the 33 percent corporate rate) that

Appendices

applies to all foreign source dividends received by
New Zealand resident corporations. The payment
is designed to approximate the income tax that a
New Zealand individual shareholder would pay on
a dividend from a nonresident company. The
corporation makes dividend withholding payments
only to the extent the New Zealand corporate tax
rate exceeds the foreign withholding tax rate.
Although styled a withholding payment, the
payment is imposed when the corporation receives
the foreign dividend, regardless of whether it
makes a distribution to its own shareholders.
However, the corporation records the dividend
withholding payments in its ICA, and thus can
pass through a credit to its shareholders when it
pays dividends. Alternatively, the corporation may
establish a sep1rate Withholding Payment Account
(WPA) and allocate dividend withholding payment
credits from the WPA to its shareholders. A WPA
might be desirable because the imputation credit
is nonrefundable and can only be converted into
a loss, but the dividend withholding payment
credit is refundable to shareholders. In addition,
only dividend withholding credits are creditable
against the withholding tax that applies to dividends paid to nonresident shareholders. Accordingly, a corporation that owns significant interests
in nonresident companies and that is owned in
significant part by tax-exempt or foreign shareholders will find the additional paperwork of a
separate WPA worthwhile.
The WP A is maintained under rules similar to
the leA rules. The WPA is credited when the
corporation pays dividend withholding payments,
and when it receives dividends bearing dividend
withholding payment credits. The WPA is debited
when dividend withholding payment credits attach
to dividends paid to shareholders, and when the
corporation chooses to transfer any part of a WPA
closing credit balance to its ICA. If the corporation has an income tax loss carryforward, or
expects to generate one, it may reduce that loss to
satisfy all or part of the dividend withholding
amount payable (or obtain a refund of payments).
A closing negative balance in the WPA must be
satisfied with a "further" dividend withholding
payment (which may be credited against future

Appendices

dividend withholding payments due, but cannot be
refunded). A debit closing balance, in addition,
automatically incurs a 10 percent penalty.
Dividend withholding payment credits may be
allocated to dividends paid to shareholders under
rules similar to and coordinated with the allocation rules for imputation credits.
Branch Equivalent Tax Account (BETA)

The Branch Equivalent Tax Account (BETA)
regime is designed to reduce the potential for
deferring New Zealand tax by accumulating
income in low-tax countries. A CFC is a foreign
corporation (not resident in Australia, the United
States, the United Kingdom, Japan, France,
Germany or Canada) in which five or fewer New
Zealand residents have a controlling (50 percent
or more) interest. 29 Any New Zealand resident
with a 10 percent interest in a CFC must include
in income its proportionate share of the CFC' s
income and receives credit for its proportionate
share of foreign income taxes paid by the CFC.
Any New Zealand tax paid is then credited to the
BETA (or to the ICA if the corporation does not
elect to maintain a separate BETA). Credits from
a BETA can be used to satisfy the dividend
withholding payment liability on later dividends
actually received from the CFC. When BETA
credits are so used to satisfy the WPA liability, a
corresponding credit to the ICA is made.

Treatment of Tax-Exempt
Shareholders
New Zealand has a small tax-favored investor
sector. Under recent reforms, New Zealand fully
taxes pension plans. At the same time the new
imputation scheme went into effect, New Zealand
confonned the treatment of Maori authorities to
that of corporations (or, in appropriate cases, to
that of trusts). In addition, New Zealand repealed
the income tax exemption on "qualifying activities" enjoyed by certain cooperatives dealing in
primary products.
For tax-exempt charitable and governmental
shareholders, imputation credits in excess of tax

180

liability are not refundable. However, such taxexempt shareholders are exempt from resident
withholding tax so preferences are not recaptured
where preference income is distributed to them.

Treatment of Foreign Shareholders
In general, the benefits of the imputation
credit system generally are not extended to foreign shareholders. New Zealand imposes a nonresident withholding tax at the rate of 30 percent
for dividends, with no difference in treatment of
portfolio and nonportfolio investors. In some
cases, treaties reduce that rate, but to no less than
15 percent. Imputation credits are not creditable
against nonresident withholding tax (although
dividend withholding payment credits are
creditable against such tax).

Low-Bracket Shareholders
Excess imputation credits are available to
offset any other tax liability of the taxpayer, but
are not refundable. Imputation credits not used in
the year that they are received convert into a loss,
which carries forward indefInitely. Excess dividend withholding payment credits and resident
withholding tax credits are refundable.

Streaming
In addition to the allocation rules discussed
above, New Zealand's imputation system contains
several anti-streaming provisions. The ICA,
WPA, and BETA must be debited to reverse a
credit where, after the credit arises, the corporation undergoes a change of beneficial ownership
of more than 25 percent (34 percent after April I,
1992).30 In addition, the ICA and WPA are
debited if there is a "shareholder or company tax
advantage arrangement" (a streaming arrangement). The use of credits by shareholders is
denied if the shareholders are party to such an
arrangement of if there is an arrangement for the
shareholder to be paid a dividend by another
company. The latter provision applies, for example, where streaming is accomplished through
stapled share arrangements.

Appendices

181

Treatment of Interest
Interest paid by a New Zealand corporation is
generally deductible. Interest paid to a resident
lender is includable in the lender's income and,
with certain exceptions, is subject to a withholding tax imposed at a rate of 24 percent. Withholding tax at a statutory rate of 15 percent is imposed
on interest paid to a foreign lender. The New
Zealand Government recently announced its
decision to exempt from withholding tax interest
paid on debt issued on or after August 1, 1991 by
"Approved Issuers" (issuers that agree to pay a
levy equal to 2 percent of the amount of the
interest paid for the right to pay exempt interest).
In addition, in some cases, treaties reduce the
withholding rate, but to no less than 10 percent.

B.6

UNITED KINGDOM

Introduction
The United Kingdom provides for distributionrelated integration of the individual and corporate
income tax systems by allowing a credit for
corporate tax paid with respect to distributed
earnings. The amount of the credit is determined
as though the corporation had paid tax at the
"basic" individual rate, currently 25 percent,
rather than at the corporate rate, currently 33
percent (except for small corporations, which may
be taxable at a 25 percent rate). Thus, the credit
provides only partial relief (except for small
corporations) from corporate level tax because
actual corporate tax paid with respect to distributed earnings is not fully creditable at the shareholder level.

Description of Mechanics

imputation credit equals the amount of net qualifying distributions, grossed up at the basic personal
rate (25 percent), and then multiplied by that
rate. 31 Accordingly, if the shareholder's actual
marginal tax rate equals the basic rate, then the
shareholder owes no tax on the distribution.
Generally, the imputation credit is refundable to
all resident, non-corporate shareholders, including
tax-exempt shareholders.
Compensatory or Withholding Tax

The United Kingdom imposes an "Advance
Corporation Tax" (ACT) on qualifying distributions equal to the amount of corporate tax imputed
to shareholders (at a 25 percent grossed-up rate).
The corporation may apply ACT payments against
its regular tax liability (mainstream tax) subject to
the limitations described below. Because preference income generates no mainstream tax, ACT
effectively recaptures preferences at the corporate
level on the distribution of preference income,
thereby assuring that preference income ultimately
is taxed at shareholder rates.
The amount of ACT that may be applied
against mainstream tax is limited to an amount
that equals 25 percent of the corporation's taxable
income for the year. Excess ACT may be carried
back for up to 6 years and may be carried forward indefinitely. Alternatively, current year and
surplus ACT can be surrendered to a more than
50 percent-owned subsidiary. Because excess
ACT is not refundable, uncredited ACT represents an additional tax liability to the corporation
until the corporation earns sufficient additional
taxable income to ab~orb it. In practice, because
of the numerous tax preferences provided by U.K.
law, many corporations carry excess ACT credits
on their books. 32

Imputation Credit

When a corporation makes a "qualifying
distribution" (described below) to its shareholders,
the distribution carries with it an imputation
credit. The shareholder includes the amount of the
credit in his taxable income in addition to the
amount of the distribution and may use the credit
against his income tax liability. The amount of the

Dividends Defined, Bonus SluJres,
Share Repurchases

The U.K. system generally defines a qualifying distribution to include any non-liquidating
distribution of cash or property made by a corporation with respect to its shares, other than a
Liquidating
repayment of share capital.

182

Appendices

distributions are not treated as qualifying distributions, and thus neither the ACT nor the gross-up
and credit mechanism applies.

individuals are eligible for an annual capital gains
exclusion of £5,000, also indexed for inflation.

Treatment of Preference Income
Bonus issues are not qualifying distributions.
This rule prevents corporations from having to
pay ACT on bonus issues. However, cash distributions on bonus issues of redeemable shares
made within 1 years of their issuance generally
are qualifying distributions even if paid out of
share capital.

°

Share repurchases are generally treated as
qualifying distributions to the extent that the
amount paid exceeds share capital, and the corporation must pay ACT on the amount so treated.

Allocation of Credits to Dividends
Because the gross-up and credit mechanism
described above applies to each qualifying distribution at the assumed 25 percent rate, no allocation rules are necessary.

Tax Rates
The corporate rate, until recently, was 25
percent for income up to £100,000 and 35 percent
for income greater than £500,000. (The U.K.
system phases out the reduced corporate rate,
which resulted in a marginal rate of 37.5 percent
for corporate income between £100,000 and
£500,000.) On March 29, 1991, the Chancellor of
the Exchequer introduced a budget that (l) reduces the 35 percent corporate rate to 34 percent
retroactive for profits earned in financial year
1990, and to 33 percent for profits earned in
1991, and (2) raises the ceiling on the 25 percent
rate to £250,000.
The individual rate is 25 percent for income
up to £20,700 and 40 percent for income over this
level.
Capital gains are taxed at the same rate as
ordinary income. In calculating the amount of
gain on disposition of a capital asset, the basis in
the asset is indexed for inflation. In addition,

As discussed above, the ACT generally
prevents corporate preferences from being extended to shareholders (preference income is taxed at
shareholder rates when distributed). However,
crediting ACT against mainstream tax has the
effect of treating distributions as made fITst from
taxable income.
The U.K. system provides corporations with
a variety of tax preferences. The most significant
is accelerated capital allowances or "writing
down" allowances (equivalent to accelerated
depreciation or amortization). To provide investment incentives, accelerated cost recovery is
allowed for certain types of capital expenditures.
Generally, all investments in business machinery
and equipment are "pooled," i.e., treated as a
mass asset. In lieu of depreciation, taxpayers are
pennitted to recover 25 percent of the pool each
year, on a declining balance basis. Scientific
research expenditures and certain oil exploration
costs in the U.K. can be fully deducted in the
year incurred even if they create an asset. Capital
expenditures on industrial and commercial buildings in enterprise zones are deductible in full
when incurred. Additional preferences are available for mineral extraction operations, industrial
buildings, and patents and know-how.

Treatment of Domestic
Intercorporate Dividends
A U.K. corporation paying a qualifying
distribution to another U.K. corporation generally
must pay ACT on the distribution, but the recipient corporation is exempt from tax on the distribution. A U.K. corporation receiving a dividend
generally cannot claim a refund or credit of ACT
paid on that dividend. However, the recipient
corporation can redistribute a dividend that has
been subject to ACT (franked investment income)
without incurring further ACT, and its shareholders are entitled to a credit for the ACT paid

Appendices

183

by the original distributing company. The effect
of imposing ACT on intercorporate dividends is to
recapture preferences prior to distribution of
preference income out of corporate solution.
If a recipient corporation receives more
franked investment income than it distributes, it
can carry forward the excess franked investment
income indeftnitely. Alternatively, the recipient
corporation may claim a refund of ACT paid on
the excess franked investment income by offsetting the excess against any losses for the year. If,
in a subsequent year, payments by the corporation
of franked investment income exceed receipts of
franked investment income, any refund of ACT
received in the earlier year is recaptured.

Qualifying distributions between U.K. corporations are not subject to ACT if a group dividend
election has been made. Such an election may be
made with respect to dividends from a more than
50 percent owned subsidiary. If a group dividend
election is made, the distribution is not treated as
franked investment income and thus is subject to
ACT when redistributed.

Treatment of Foreign Source Income
U.K. corporations are taxed on their worldwide income, with relief from double taxation
provided through a foreign tax credit system.
U.K. corporations are allowed a credit for foreign
taxes paid subject to the following limits.33 First,
the foreign tax credit is allowed only against U.K.
tax payable on foreign source income from the
particular source with respect to which the foreign
tax was paid. Second, unused foreign tax credits
may not be carried forward or back.
Foreign tax credits cannot be used to satisfy
liability for ACT where qualifying distributions
are paid out of foreign source income. Thus, the
benefit of the foreign tax credit is washed out
with respect to distributed foreign source income.
The amount of ACT that may be applied
against mainstream tax imposed on foreign source
income effectively is the lesser of (I) the mainstream tax on foreign source income and (2) 25

percent of foreign source taxable income. The
effect is that foreign tax credits are allowed before
the ACT and ACT that is unused because of
foreign tax credits is carried back or forward.
This ordering rule favors taxpayers because
surplus ACT, unlike surplus foreign tax credits,
can be carried forward. 34

Treatment of Tax-Exempt
Shareholders
A tax-exempt shareholder is entitled to a
refund of the shareholder credit. The primary
entities exempt from tax on investment income are
charities, pension plans (called "exempt approved
schemes"), and building societies.

Treatment of Foreign Shareholders
The treatment of dividends paid by U.K.
corporations to foreign shareholders varies depending on whether they are entitled to treaty
beneftts. Except as provided by treaty, onl y
shareholders that are U.K. residents are entitled to
imputation credits on dividends received from
U.K. corporations. On the receipt of such dividends, a foreign shareholder not entitled to treaty
beneftts is treated as having income equal only to
the amount of the distribution (rather than the
distribution plus the imputation credit), the rate of
tax applicable is the same as for residents (25 or
40 percent for individuals), the foreign shareholder is treated as having paid tax at the 25 percent
rate on the distribution, and the foreign shareholder generally is not entitled to the imputation
credit.
Under tax treaties, foreign shareholders
generally are entitled to some or all of the imputation credits otherwise allowable to resident shareholders with respect to a dividend from a U.K.
corporation, and the rate of tax is reduced (the
amount of the reduction may vary depending on
whether the shareholder is a portfolio or nonportfolio investor). For example, for aU. S. shareholder owning less than 10 percent of the stock of
the distributing corporation, the U.S. treaty
entitles the shareholder to the full imputation
credit and reduces the tax to 15 percent of the

Appendices

amount of the dividend grossed up for the credit
(imposed as a withholding tax). For a U.S.
shareholder owning at least 10 percent, the shareholder is entitled to one-half of the imputation
credit and the rate of tax is reduced to 5 percent
of the dividend grossed up for the amount of the
credit allowed (also imposed as a withholding
tax).35

Streaming
The U.K. system contains several anti-streaming provisions. For example, tax-exempt shareholders purchasing at least 10 percent of a corporation are subject to tax at a 10 percent rate on
dividends made out of pre-acquisition earnings
(but may use attached credits to offset the tax).
Restrictions on entitlement to imputation credits
apply where there is an arrangement to channel
credits to shareholders of a close investment
holding company.
In addition, the United Kingdom has adopted
measures to prevent trafficking in excess ACT.

184

The principal limitation is triggered where, following a major change in share ownership (a
more than 50 percent increase by one or more
5 percent shareholders over a 3 year period),
there is a major change in nature or conduct of
the corporation's business or a considerable
revival of business that had been negligible prior
to the ownership change. In such a case, prechange surplus ACT cannot be used to offset postchange mainstream tax.

Treatment of Interest
Interest paid by U.K. corporations generally is
deductible if the indebtedness is incurred for
business purposes. Interest received by a resident
lender generally is includable in the lender's
income. Foreign lenders are taxed on U.K. source
interest at the same rate as residents, but this tax
may be reduced or eliminated under treaties. For
example, U.K. source interest received by a U.S.
resident is exempt from U.K. tax under the U.S.
treaty.

c:

ApPENDIX
EQUIVALENCE OF
DISTRIBUTION-RELATED INTEGRATION SYSTEMS
The dividend exclusion, imputation credit and
dividend deduction systems produce equivalent
results if corporate and shareholder tax rates are
the same, all shareholders are taxable, and no
corporate tax preferences exist. This appendix
illustrates that equivalence and shows how the
three systems diverge when each of these
assumptions is relaxed.

C.1

same amount split between a cash dividend and a
tax credit. Under either the dividend deduction or
the imputation credit system, the shareholder has
the same after-tax income and pays the same
amount of tax. Thus, the corporation's behavior
should be the same economically under both
systems. To achieve equivalence under the three
systems, in the example above, the corporation
must adjust its cash dividends to leave its shareholders in identical after-tax positions. This
assumption probably better reflects long-term than
short-term behavioral responses to the various
integration mechanisms.

EQUIVALENCE OF SYSTEMS
IF TAX RATES WERE EQUAL

Table C.I illustrates the equivalence of the
three different types of systems when individual
and corporate tax rates are equal (34 percent in
the example), all shareholders are subject to tax,
and no corporate tax preferences exist. For
simplicity, all examples assume that corporations
distribute all income when earned.

C.2

EFFECTS OF RATE
DIFFERENCES, PREFERENCE
INCOME, AND EXEMPT
SHAREHOLDERS

Rate Differences
It might appear counterintuitive that the
dividend deduction and imputation credit systems
lead to exactly the same result. Nevertheless,
from an economic perspective, the two systems
are equivalent under these assumptions. This
equivalence depends on the assumption that
shareholders are indifferent between receiving a
certain amount of money as a cash dividend or the

If corporate and shareholder tax rates differ,
the three systems no longer produce equivalent
results. A dividend exclusion system eliminates
whatever shareholder level tax would otherwise be
imposed. A dividend deduction system eliminates
the corporate level tax and retains the shareholder
level tax.

imputation credit
system can be structured to
tax distributed earnings at
either the corporate or individual rate. To tax distribut~ons at the individual rate, a
credit would be allowed to
shareholders for the full
amount of corporate tax paid
with respect to a distribution.
This credit would be allowable against tax on other
income, or, if there were no
such tax, fully refundable. To
tax distributions at the corporate rate, a credit would be
An

Table C.I
Equivalence of Distribution-Related Integration Systems
Classical
System

100
Corporate income
Distribution
66
Corporate tax
34
Shareholder credit
0
Cash received
66
66
Shareholder income
Shareholder taxi
22
56
Total tax paid
ITax due after credits, if any.

Dividend
Exclusion

Dividend
Deduction

Imputation
Credit

100

100
100
0
0
100
100
34
34

100

66

34
0
66
0
0

34

185

66

34
34
66

100
0
34

186

Appendices

allowed only for tax at the shareholder rate on the
sum of the cash distribution and the credit ($95.65
in the second to last column in the example
below).l
Table C. 2 assumes a shareholder rate of 31
percent and a corporate rate of 34 percent.

the policy choice is to extend preferences, where
corporate and shareholder rates are equal, the
system could determine the shareholder credit as
though the corporation had paid tax at the full rate
on all income, i.e., by grossing up the cash
distribution at the full corporate rate. 2 Passing
through preferences where there are rate differences is somewhat more difficult. 3

Preference Income
If some corporate income is not taxed, or is

taxed at a lower rate, the alternative systems also
do not produce equivalent results. Without modification of the sort described in Section 2.B, a
dividend exclusion would automatically extend
corporate tax preferences to shareholders, because
preference income would not be taxed (or would
be taxed at a lower rate) at the corporate level
and, with an exclusion for all dividends received,
would not be taxed at the shareholder level. A
dividend deduction system would not extend
preferences to shareholders because shareholders
would include dividends in income.
An imputation credit system can be designed
to achieve either result. If, as this Report recom-

mends, the policy choice is not to extend preferences to shareholders, a system can be designed
to limit the shareholder credit to the corporate tax
actually paid with respect to the distribution. If

To illustrate the effects of preferences, holding
tax rates equal, Table C.3 assumes that the corporate rate and the shareholder rate are both 34
percent.

Tax-Exempt and Foreign Shareholders
If certain shareholders are wholly or partially
exempt from U.S. tax, the alternative distribution-

related integrated systems do not produce equivalent results, even if corporate preferences are not
taken into account. A dividend exclusion system
replicates the current treatment of tax -exempt
shareholders, because corporate income is taxed
at the corporate level, and a tax-exempt shareholder would receive no additional benefit from a
shareholder level exclusion. 4 In contrast, a dividend deduction system produces an absolute
benefit to tax -exempt shareholders because corporations could reduce or eliminate the corporate
level tax that appli~s to income from equity

Table C.2
Effect of Rate Differences
Imputation Credit
Classical
System
Corporate income
Distribution
Corporate tax
Shareholder credit
Cash received
Shareholder income
Shareholder taxi
Tota! tax paid

100
66
34
0
66
66
20.46
54.46

Dividend
Exclusion

100
66
34
0
66
0
0
34

Dividend
Deduction

100
100
0
0
100
100
31
31

At Shareholder
Rate

100
66
34
29.65
66
95.65
0
34

At Corporate
Rate
100

66
34
34
66
100
0
31 2

ITax due after credits, if any.
"The shareholder would have an excess credit of $3 that would be refunded or could be used to
offset other tax liability.

Appendice

187

Table C.3
Effect of Preferences
Imputation Credit
Classical
System
Corporate income
Preference income
Taxable income
Distribution
Corporate tax
Shareholder credit
Cash received
Shareholder income
Shareholder tax?
Total tax paid

Dividend
Deduction

Dividend
Exclusion

100
40
60
79.6 1
20.4
0
79.6
79.6
27.06
47.46

100
40
60
79.6 1
20.4
0
79.6
0
0
20.4

Preferences
Passed Through

Preferences Not
Passed Through

100
40
60
79.6 1
20.4
41
79.6
120.6
0
20.4

100
40
0
100
0
0
100
100
34
34

100
40
60
79.6 1
20.4
20.4
79.6
100
13.6
34

IThis is the maximum amount the corporation can distribute after payment of the corporate level tax.
2Tax due after credits, if any.

supplied by tax-exempt shareholders by deducting
payments of dividends to tax-exempt shareholders.
A dividend deduction system also would maintain
the same benefit relative to taxable investors that
tax-exempt shareholders enjoy under current law.
An imputation credit system with full refundability would have the same effect as a dividend
deduction system. An imputation credit system
that does not permit credits to be refunded to tax-

exempt shareholders would have the same effec
as a dividend exclusion system.
Table C. 4 assumes that all shareholders an
fully exempt from tax and that the corporatior
pays tax on all of its income at a 34 percent rate.
The treatment of foreign shareholders unde]
each system is similar. A dividend deductior.
system would extend automatically the benefits oj

Table C.4
Effect of Tax-Exempt Shareholders
Effect of Tax-Exempt Shareholders
Imputation Credit
Classical
System
Corporate income
Distribution
Corporate tax
Shareholder credit
Cash received
Shareholder income
Shareholder tax2
Total tax paid

100
66
34
0
66
66
0
34

Dividend
Exclusion

Dividend
Deduction 1

100
66
34
0
66
0
0
34

100
100
0
0
100
100
0
0

Refundable

100
66
34
34
66
100
03
0

Not
Refundable

100
66
34
34
66
100
0
34

INo withholding on dividends. (A dividend deduction system with a nonrefundable "withholding" tax of 34 percent would duplicate the results under a dividend exclusion system or an
imputation credit system with nonrefundable credits.)
2Tax due after credits, if any.
Jyhe tax-exempt shareholder would receive a $34 refund.

Appendices

integration to foreign shareholders, because only
one level of tax (the current withholding tax on
dividends) would be collected on corporate income distributed to foreign shareholders. A
dividend exclusion system would automatically
deny the benefits of integration to foreign

188

shareholders (assuming, again, that the current
withholding tax remains in place). In contrast, an
imputation credit system would extend benefits to
foreign shareholders if the imputation credit is
refundable and would deny benefits if the credit is
not refundable to foreign shareholders.

189

Notes

NOTES
PART I
Chapter 1
1. If corporate income were not subject to tax until distributed to shareholders, retained eamings would be taxed under the
individual income tax system only when shareholders realize capital gains on the sale of stock. Shareholders could defer or
avoid individual income tax simply by retaining earnings in corporations. See Pechman (1987) and Warren (1981). While
this argument counsels against repeal of the corporate income tax, it does not apply to the integration proposals discussed
in this Report, none of which permit such indefinite deferral of tax on corporate income.

Some have suggested that a mark-to-market regime for corporate stock would remove the potential deferral associated
with investment in corporations and, thus, the need for the corporate tax. Under a mark-to-market regime, shareholders
would recognize each year the change in the value of the corporation, including corporate income. See Shakow (1986) and
Thuronyi (1983). While marking to market corporate stock could be considered a method of integrating the corporate and
shareholder tax systems, it also would tax shareholders on income that is unrealized at the corporate level. We do not explore
that approach in this Report, because abandoning the realization requirement goes well beyond the changes necessary to
achieve integration.
2. Tax Reform Act of 1986, P.L. 99-514, 100 Stat. 2085, Oct. 22, 1986.

3. General Utilities & Operating Co. v. Helvering, 296 U.S. 200 (1935).
4. This increase in welfare compares favorably to that estimated for the 1986 Tax Reform Act at the time of its adoption.
See Fullerton, Henderson, and Mackie (1987).
5. Appendix A contains a more detailed discussion of the taxation of corporations under current law.
6. Characterizing the corporate income tax as a double tax rests on the assumption that the corporate level tax reduces
corporate income available to shareholders. If the corporate tax does not reduce profits but instead increases prices charged
to consumers or lowers wages paid to workers, little or no additional tax may be paid on dividends. Section 13.G discusses
the incidence of the corporate tax. In addition, not all income earned by corporations is taxed when earned, and not all
shareholders are subject to taxation. Chapter 5 discusses tax preferences and Chapters 6 and 7 examine the issues of taxexempt and foreign investors.
7. The Omnibus Budget Reconciliation Act of 1990, Pub. L. 101-508, 104 Stat. 1388 (the 1990 Act) made three changes
in the individual income tax rate structure. First, the 1990 Act increased the top marginal tax rate for individuals to 31
percent from 28 percent. A number of other statutory provisions may affect statutory marginal rates. For example, the 1990
Act created an explicit phaseout of personal exemptions for taxpayers with adjusted gross income (AGI) above certain
thresholds. For a married couple filing jointly, for example, the deduction for personal exemptions phases out at a rate of
2 percent for each $2,500 of AGI above $150,000. The 1990 Act also enacted a rule disallowing a portion of itemized
deductions otherwise allowable to high-income taxpayers. Itemized deductions (other than medical, casualty and theft, and
investment interest deductions) are generally reduced by 3 percent of AGI in excess of $100,000, except that the disallowance
cannot exceed 80 percent of the affected itemized deductions.
8. Interest received by foreign lenders that are related to the borrower or by foreign banks on loans made in the ordinary
course of business, is, however, subject to withholding tax at 30 percent or a lower treaty rate.
9. In addition to the distortions created by the two-tier tax, distortions may result from the rules used to measure business
income. For example, the Code generally fails to correct for distortions in the tax base attributable to inflation or to the
requirement that a capital gain be realized before being subject to tax. These measurement problems affect both corporate
and unincorporated business income. Because the general reform of business income measurement rules is beyond the scope
of this Report, we take the existing system of income measurement rules as given.
10. See Harberger (1962 and 1966) and the subsequent studies cited in Chapter 13, note 1.

Notes

190

11. This simple example abstracts from other factors affecting the cost of capital, including: (i) differences between tax and
economic depreciation; (ii) differences in tax rates among investors; and (iii) inflation.
12. See Gravelle (1991). These calculations assume (i) a rate of inflation of 4 percent; (ii) an average holding period of 7
years; and (iii) that tv.'o-thirds of capital gains are deferred until death.
13. Data for the past few years (some of it preliminary) shows a reduction in the size of the corporate sector relative to the
noncorporate sector and the overall economy. Particularly since 1986, S corporations have accounted for an increased share
of corporate profits. Long-term comparisons of corporate activity '.vith general economic activity, however, present no clear
trend toward disincorporation. See Chapter 13.
14. See Gravelle (1991).
15. Inflation adds a complication here. Because the tax system taxes nominal rather than real returns, the deductibility of
interest expense under current law offers an even greater tax advantage to corporate debt financed investments (relative to
corporate equity financed or noncorporate investments) in the presence of inflation, since corporations typically deduct
nominal interest payments at a higher tax rate than the rate at which lenders are taxed on these payments. See Fullerton,
Gillette, and Mackie (1987) and Gertler and Hubbard (1990).
16. While both book-value and market-value measures are subject to criticism, market-value measures of debt burdens are
generally superior for measuring bankruptcy risks because they reflect inflation and other factors that influence the value of
alternative claims on the firm. See, e.g., Bernanke and Campbell (1988) and Warshawsky (1991).
During inflationary periods, book-value measures tend to overstate the burden of debt and to understate the value of a
finn's assets. The debt burden may be overstated because with inflation part of the interest rate reflects a return of principal,
not a real cost to the firm. As a result of inflation, new debt can be issued without increasing the effective debt burden of
the finn; some new debt would merely represent a rollover of the portion of the real principal that must be repaid, rather
than a net issuance of new debt. In addition, to the extent that inflation is higher than anticipated, the burden of a given
amount of debt falls because real income is transferred from bondholders to shareholders. Book-value ratios also understate
the value of the finn's assets because traditional accounting measures of asset values are based on the historical price of the
asset, not on its current market (replacement) price. In addition, because book-value debt to asset ratios do not reflect
changes in equity values, they may be misleading indicators of the true burden of debt, especially during periods (such as
the 19808) with large increases in stock prices,
While market-value measures of the firm's debt and equity reflect adjustments for inflation and for other changes in the
market value of the finn and its securities, they also may be criticized. First, market-value measures generally are estimated
rather than directly observed. One approach for estimating the market value of equity and debt, for example, is to capitalize
dividend and interest payments, respectively. The Federal Reserve market value ratio shown in Figure 1.5 is a more
sophisticated measure, but it also relies on estimates of equity and debt values. Second, market-value ratios are inaccurate
if stock market prices do not reflect fundamental values.
17. See, e. g., Shoven (1987) and Auerbach (1989). Share repurchases are discussed further in Chapters 8 and 13.
18. See the evidence in Shaven (1987) and Auerbach (1989).
19. Estimates are based on data for dividends and buybacks from the COMPUSTAT II database, Standard and Poor's
COMPUSTAT Services, Inc. Assuming the corporate AAA bond interest rate for all years, the figures represent the
maximum interest properly attributable to the increase in share repurchases because they assume that (1) repurchases were
financed completely by debt, and (2) the additional debt remains outstanding during the 1980s. The elimination of the capital
gains exclusion by the 1986 Act reduced the attraction for investors of share repurchases, since the gain component of the
distribution is no longer generally taxed at preferential rates. Share repurchases continued strong through 1989, but declined
in 1990.
20. Similarly, leveraged buyouts (LBOs), which replace substantial equity with debt, also may have contributed to the
IDcrease in corporate debt during the 1980s. The dollar value of completed mergers and acquisitions in the United States rose
at an annual rate of 14.3 percent between 1981 and 1989. The LBO share of this activity rose 8.6 percent in 1983 to 22.7
percent in 1986, but receded to 18.4 percent in 1989 (excluding RJR Nabisco), dropping sharply to 9.3 percent in 1990.
(Source: Mergers and Acquisitions, Almanac and Index, May-June 1985-1991). By the end of 1988, outstanding LBO debt
was estimated to be about 20 percent of the (book) value of outstanding corporate bonds or more than 9 percent of the (book)
value of total nonfinancial corporate debt (based on data from the Federal Reserve Board's Flow of Funds Accounts,

191

Notes

Financial Assets and Liabilities, Year End (1967-1990), hereinafter cited as Flow of Funds Accounts). See Gertler and
Hubbard (1990».
21. See, e.g., Warshawsky (1991).
22. See Friedman (1990) and Gertler and Hubbard (1990).
23. Potential nontax benefits of debt finance are discussed in Chapter 13. See also Jensen (1986) and Gertler and Hubbard
(1990).
24. See Chapter 13 and Gordon and Malkiel (1981).
25. The Congressional Research Service estimates that the shareholder level effective Federal income tax rate on dividends
is 32 percent, compared to 11 percent or less on capital gains attributable to retained earnings. See Gravelle (1991).
26. This assumption is controversial, since not all economic models of the effects of taxation on dividend payments maintain
that nontax benefits are associated with dividend payments. There are two leading explana~ions of why corporations continue
to pay dividends in spite of the greater investor level tax burden on dividends than on capital gains attributable to retained
earnings or share repurchases: the "traditional view" and the "new view." The "traditional view" asserts that dividends offer
nontax benefits to shareholders that offset their tax advantage. Accordingly, dividend taxes distort payout decisions and raise
the cost of capital. The "new view" assumes that dividend payments offer no nontax advantages to shareholders and that
corporations have no alternative to dividends for distributing funds to shareholders. Under this assumption, dividend taxes
reduce the value of the firm, but do not affect firms' dividend or investment decisions. This Report adopts the framework
suggested by the "traditional view. " The two approaches are discussed in more detail in Chapter 13.
27. These studies are discussed in Section 13.B.
28. The 1970 data in the text are from Shoven (1987). The 1989 and 1990 data are from Department of the Treasury
calculations based on tabulations of the Standard and Poor's COMPUSTAT Industrial and Research files.
29. The effect of taxation on savings is uncertain because changes in the after-tax rate of return have an ambiguous effect
on savings. A higher after-tax return makes future consumption cheaper than foregone present consumption. This substitution
effect encourages households to reduce present consumption and increase savings. However, a higher after-tax return also
allows a given level of future consumption to be reached with less savings today. This second effect, called the income effect,
reduces saving. Because the substitution effect of a rise in the after-tax return increases saving, while the income effect
reduces saving, the net effect of a rise in the after-tax return is an empirical question.
30. As noted in note 29, the net effect of changes in the after-tax rate of return on saving is difficult to determine because
it depends on opposing income and substitution effects. There is less theoretical uncertainty about the direction of the effect
of capital taxation on investment. The distinction between saving and investment is an important one in an analysis of
corporate taxation. In an economy without international trade and investment flows, national saving equals national
investment, and the average cost of capital summarizes tax incentives to save as well as to invest. International capital flows
break the equivalence of domestic saving and investment, however. In a world with perfect international capital mobility,
incentives for domestic investment would be governed by the pre-tax return needed to cover taxes and the worldwide
opportunity cost of funds. At the same time, domestic saving would depend on the after-tax return earned by savers from
investing at the worldwide rate of return. Hence, domestic investment depends on domestic corporate level taxes, while
domestic saving depends on domestic individual level taxes.
31. U.S. Department of the Treasury, Tax Reform for Fairness, Simplicity, and Growth (1984) (hereinafter cited as Treasury

D, Vol. 2, pp. 135-144 and The White House, The President's Tax Proposals to the Congress for Fairness, Growth, and
Simplicity (1985) (hereinafter cited as The President's 1985 Proposals), pp. 120-129. See also U.S. Department of the
Treasury, Blueprints for Basic Tax Reform (1977) (hereinafter cited as Blueprints).
32. See Appendix B.
33. See, e.g., McLure (1979).
34. So-called partial integration (referred to in this Report as distribution-related integration) has been viewed as a
compromise between the passthrough ideal and considerations of administrability. A conventional definition of full integration

Notes

192

is given in McLure (1979), p.3: "... income earned at the corporate level, whether distributed or not, would be attributed
to shareholders, as in a partnership. and taxed only at the rates applicable to the incomes of the various shareholders. "

35. Appendix C discusses the effect of rate relationships on integration proposals.
36. For general discussion of economic benefits of neutrality in the taxation of capital income, see Institute for Fiscal Studies
(1978) and Bradford (1986).
37. See Sections 2.D and 4.F.

38. This Report also does not generally address tax distortions created by inflation.
39. Under a corporate cash-flow tax, corporations would be taxed on the net cash flow from their business activities.
Corporate cash-flow taxes have generally been advanced as part of an overall restructuring of the tax system that would
replace the individual income tax with a consumption or cash-flow individual tax. See Institute for Fiscal Studies (1978),
Aaron and Galper (1985), and Bradford (1986). Recently, however, some economists have proposed cash-flow taxes on
businesses, while the current income tax rules would be maintained at the individual level. See, e.g., King (1987), Feldstein
(1989), and Hubbard (1989).
Under one corporate cash-flow tax proposal, a corporation would determine its tax base by subtracting from its receipts
from sales of goods or services its cost of purchasing real goods and services for production. No deductions for financing
investments would be allowed; that is, neither dividends nor interest payments would be deductible. Several significant
changes would be required to convert the current corporate income tax base to a cash-flow tax base, including replacing
depreciation deductions with a deduction for the cost of capital assets in the year of acquisition (expensing), and eliminating
corporate investment interest deductions. Other ways to define the base of a corporate cash-flow tax are discussed in Institute
for Fiscal Studies (1978) and King (1987).
Proponents of a cash-flow tax emphasize that, because the initial purchase of assets would be deductible, the system
would generate a zero marginal effective tax rate on investment. In effect, the tax system would not distort the cost of capital
investment decisions. Income generated in the corporate sector, however, would continue to bear a tax at the individual level.
In contrast, noncorporate business income would face no tax at the margin if it were taxed on a cash-flow basis. Hence, a
bias against investment in the corporate sector would still exist.
Because interest payments would not be deductible, the tax advantage that debt enjoys under the current system would
be eliminated, but a cash-flow tax would not achieve neutrality with respect to choice of finance. Rather, under the
reasonable assumption that the marginal individual tax rate on dividends exceeds the marginal effective accrual tax rate on
capital gains, retained earnings would have an advantage over either debt or new equity .as a source of corporate finance.

40. See generally Treasury I.

Part II
Introduction

1. While the prototypes discussed in this Part and in Part IV contain considerable technical detail, they do not provide a
comprehensive summary of technical changes that would be required. For example, the prototypes do not address the effect
of an int~gration system on groups of corporations filing consolidated returns. We concluded, consistent with the approach
to consohdated return matters under the current corporate tax system, that consolidated return issues are better addressed
after a basic integration approach is selected.
2. The distribution-related integration systems of several major U. S. trading partners are described in Appendix B.
Chapter 2
I. Peel (19~5) also proposes a dividend exclusion system. While Peel's proposed system resembles the dividend exclusion
prototype dIscussed here (e.g., in allowing shareholders to exclude dividends only to the extent of income that has been taxed
fully at the corporate level), there are significant differences. For example, Peel's proposed system would track taxable

193

Notes

income rather than taxes paid, would extend the benefits of integration to foreign shareholders by statute, and would treat
foreign taxes like U.S. taxes in determining the extent to which a corporation's income has borne tax.
From 1954 to 1986, the Code provided a very small exclusion for dividends received by individuals. Immediately
preceding repeal, IRC § 116 provided an exclusion of up to $100 of dividends received ($200 on a joint return).

2. Although a detailed treatment of the financial accounting consequences of adopting an integrated system is beyond the
scope of this Report, and the fmancial accounting authorities have never addressed the integration prototypes developed in
this Report, a few preliminary observations can be made. Because the dividend exclusion prototype generally retains the
current rate structure and rules for calculating corporate income subject to tax, adoption of the prototype should not
significantly change corporations' provision for income tax expense or the determination of taxes currently payable or payable
at a future date. Of course, the economic effects of moving to an integrated tax system, e.g., changes in corporations'
distribution policies and capital structures, would be reflected in fmancial statements.
3. This is similar to an imputation credit system that taxes corporate income at a 34 percent rate and allows shareholders
imputation credits at the individual shareholder rates.
4. An imputation credit system that demes refundability of imputation credits to tax-exempt shareholders achieves the same
results. See Section 11.E.
5. An imputation credit system that relies on a shareholder credit limitation rather than a compensatory tax reaches the same
result. See Section 11. B.
6. An imputation credit system reaches the same result if foreign taxes are not added to the shareholder credit account. See
Section I1.D.
7. In an imputation credit system, this result can be achieved by denying refundability of imputation credits to foreign
shareholders and continuing to impose withholding tax. See Section 11. E.
8. For simplicity, Table 2.1 (and the corresponding tables in Chapters 3, 4, and 11) refer to the tax imposed on a foreign
investor's noncorporate equity income as a withholding tax, lwN, although the method and rate of taxation actually vary
depending on the type of income. Very generally, a foreign investor is taxed on income from an equity investment in a
noncorporate business as if the foreign investor had earned directly the income earned by the business. A foreign investor
is generally subject to tax at rates applicable to U.S. persons on income that is "effectively connected" with a U.S. trade or
business. A partnership generally must withhold tax from a foreign partner's distributive share of effectively connected
income under IRC § 1446. A partnership also withholds tax on a foreign partner's distributive share of dividends, interest,
and other income to the extent required by IRe § 1441.
9. A compensatory tax is used in some foreign imputation credit systems, e.g., the Umted Kingdom, France, and Germany,
to ensure that corporate level preferences are not extended to shareholders. See Appendix B.
10. Because the prototype treats AMT as corporate taxes paid, it does not treat as taxes paid the portion of a later year's
regular taxes that are offset by the AMT credit allowed by IRC § 53.
Example. A corporation earns $100 of preference income. The corporation's regular tax is $0, and its AMT is $20.
The addition to the EDA is $38.82 «$20/.34)-$20). This is the amount of hypothetical income that would be left
for distribution if the corporation had earned taxable income of$58.82 and paid $20 of regular tax at the 34 percent
rate (58.82 X .34 = 20).
11. In mathematical terms, for each dollar of taxes paid, the corporation can add (1 It) - 1 to its EDA, where t is the
corporate tax rate. This formula also can be expressed as (l-t)/t.
The graduated rates set forth in IRe § I 1(b) for corporations with incomes of less than $75,000 would continue to be
available. Converting the entire amount of taxes paid at a 34 percent rate provides a simple rule and should not harm most
corporations, because the benefit of graduated rates begin to phase out for corporations with taxable incomes greater than
$100,000. It would, however, be possible to modify the EOA conversion formula to reflect graduated rates. One possibility
is to build the graduated rate structure into the EDA formula for corporations with taxable incomes of less than $100,000
by permitting conversion of the first $7,500 of taxes paid at the 15 percent rate (into $42,500 of ED A) and conversion of

Notes

194

the second $6,250 of taxes paid at the 25 percent rate (into $18,750 of EOA). These amounts would be reduced for
corporations in the phaseout range.
12.

Example. In year one, a corporation reports $100 of income and pays $34 of tax. The corporation's EOA
balance is $66, and it pays an excludable dividend of $66. In year two, the corporation incurs a net
operating loss of $50 and files a claim for refund of $ 17. Making that adjustment retroactive to year one
would require adjusting shareholders' incomes to reflect a taxable dividend of $33. Because this is
impractical, the prototype requires that the refund in the year of the adjustment be carried forward to be
applied against future corporate taxes.

13. Payment of a refund when the EDA balance is exhausted would, in effect, refund corporate taxes that have already been
used to qualify distributions as excludable by shareholders; only by requiring a negative balance in the EDA could this be
compensated for in later years.
14. We rejected the alternative of permitting refunds and NOL carrybacks to create a negative EOA. If such an approach
were adopted, a negative EOA would be increased by subsequent payments of corporate tax. In addition, a corporation with
a negative EOA would be required to pay additional tax to increase its EDA to zero upon certain events, e.g., upon
liquidation.
15. While a 100 percent dividends received deduction could be extended to all corporate shareholders to defer completely
taxation of corporate preference income until it is distributed out of corporate solution, it would add approximately $400
million to the revenue cost of the dividend exclusion prototype. Because of the additional complexity that would arise from
a partial dividends received deduction under an imputation credit system, we make a different recommendation under that
system. See Section 11.B.
16. As under current law, hybrid instruments and derivative products (e.g., convertible debt and options may allow a taxexempt or foreign investor to capture the portfolio benefits of holding stock while avoiding corporate level tax.
17. One anti-streaming mechanism is inherent in the prototype. Because all dividends paid reduce any positive balance in
the EOA, a corporation cannot simultaneously pay excludable dividends on one class of stock and taxable dividends on
another. The imputation credit system, described in Chapter 11, allows greater flexibility in attaching shareholder level tax
credits to dividends and, as a result, demands additional anti-streaming restrictions.
Requiring dividends to reduce the EOA does not prevent all streaming, however. For example, excludable dividends
can be paid to taxable shareholders to the extent of the EOA and thereafter all taxable dividends can be paid to tax-exempt
shareholders. Further, complex corporate structures and corporate reorganization (either acquisitive or divisive) also might
be used to stream excludable dividends by isolating or shifting shareholders' interest in a corporation's EOA. If necessary,
anti-abuse rules can be formulated to prevent such arrangements.
18. IRC § 246 (which governs corporations' eligibility for the dividends received deduction) may provide a model for
developing related rules.
19. IRC § 1059 limits the ability of corporate shareholders to strip dividends by claiming the dividends received deduction
with respect to distributions more properly treated as a return of capital. It does so by requiring stock basis to be reduced
to the extent of the dividends received deduction with respect to extraordinary dividends paid within 2 years of an acquisition
of stock. The appropriate scope of an IRe § 1059-type basis adjustment will depend on the treatment of capital gains under
integration. See Chapter 8.
As discussed in the text under "Corporate Shareholders," an excludable dividend received by a corporate shareholder
increases the recipient's EOA. Consideration should be given to whether additional anti-streaming rules are necessary to
prevent streaming through the shifting of EDA balances among corporations.
20. Under IRC § 305(b)(2), a distribution (including a deemed distribution) by a corporation of its stock is treated as a
dividend if the distribution (or a series of distributions of which distribution is a part) has the result of (1) the receipt of
money or other property by some shareholders, and (2) an increase in the proportionate interests of other shareholders in
the assets or earnings and profits of the corporation. For example, assume a corporation issues two classes of common stock
in an attempt to stream excludable dividends to certain shareholders. The first class pays excludable dividends and is intended
to be held by taxable persons. The second class pays stock dividends (or receives an increased interest in the corporation's
assets) and is intended to be held by tax-exempt persons. In such a case, IRC § 305 would impute dividends on the second
class of stock and the corporation's EDA would be reduced accordingly.

195

Notes

Similarly, IRC § 305(c) authorizes the Department of the Treasury to issue regulations treating a wide variety of
transactions as constructive distributions to any shareholder whose proportionate interest in the corporation's assets or
earnings and profits is increased thereby. For example, IRC § 305(c) would prevent a corporation from issuing preferred
stock on which a redemption premium substitutes for dividends.
21.

Example. Corporation X is owned by a tax-exempt shareholder, and its only asset is a $100 EDA balance,
e.g., because it previously distributed preference income and retained only enough cash to pay the tax
liability when the preference subsequently turned around. Corporation Y is owned by taxable shareholders
and has substantial preference income and cash but a $0 EDA balance. Corporation Y acquires corporation
X in a tax-free merger described in IRC § 368(a)(a)(A), and subsequently uses X's EDA balance to
distribute $100 of Y's cash as excludable dividends. If Y's $0 EDA balance is attributable to deferral
preferences, it will ultimately owe tax when the preferences turn around. However, the acquisition of X's
EDA enables Y to defer tax on the preference income that otherwise would have resulted from V's current
distribution of dividends.

22. The American Law Institute, Reporter's Memorandum No.3, (1991), pp. 7-8, makes a similar recommendation in
discussing an integration proposal involving maintenance of a "taxes paid account" at the corporate level.
23. In the interim, the rules of IRC § 269 could be applied to prevent the most obvious tax-motivated acquisitions.
24. Similar issues arise under the shareholder allocation and imputation credit prototypes, but we do not discuss them
separately in Chapters 3 and 11. The dividend exclusion prototype taxes corporate equity income once at a 34 percent rate,
regardless of the tax rate of the shareholder. Thus, if an interest disallowance rule applied, it should apply regardless of
whether the dividends paid on the stock are excludable or taxable. While excludable dividends bear a superficial similarity
to tax-exempt interest under IRC § 103, one level of tax on the earnings used to pay the dividend has been collected.
Similarly, taxable dividends paid, for example out of preference income, to a taxable shareholder also bear one level of tax,
although at the shareholder's rate. Thus, if an interest disallowance rule were adopted, it would be inappropriate to apply
it only to the extent of excludable dividends. On balance, this Report does not recommend deVeloping rules to deal with the
potential rate arbitrage of equity holders borrowing from low rate or tax-exempt lenders for either excludable or taxable
dividends. See note 25.
25. As under current law, the general deductibility of interest permits significant rate arbitrage through the issuance of debt
by taxable issuers to tax-exempt and foreign lenders. the relative importance of the rate arbitrage potential of borrowing to
purchase corporate stock may be less in an integrated system that does not change the treatment of interest generally. In
contrast, CBIT generally eliminates businesses' ability to pay interest to tax-exempt and foreign lenders without the payment
of one level of tax. Thus, in CBIT, we found it appropriate to eliminate investor level rate arbitrage through borrowing as
well. Compare IRC § 246A.
26. No other country with an integrated system has adopted this approach, however.
27. If such treatment of foreign taxes were permitted, special rules would be required to ensure that appropriate amounts
are added to the EDA when foreign tax rates exceed the U.S. rate. If the foreign tax rate is less than the U.S. rate, foreign
taxes paid could be converted into the appropriate EDA balance by applying the formula set forth in Section 2.B.
Example 1. A corporation has $100 of foreign source income and pays $20 in foreign taxes. After applying the IRC
§ 904 limitation, the corporation would be entitled to credit all $20 of foreign taxes against its U.S. tax liability of $34.
The U.S. residual liability would be $14, which would convert into a $27 ($14/.34-$14) addition to the EDA. The $20
of foreign taxes paid would convert into a $39 ($201.34-$20) addition to the EDA. The total EDA would be $66, which
would enable the corporation to distribute its after-tax earnings of $66 as excludable dividends.
However, if foreign tax rates exceed U.S. tax rates, the foreign taxes cannot be converted into an EDA balance using
the formula set forth in Section 2.B. In that case, the foreign taxes must be converted using the higher foreign tax rate.
Example 2. A corporation has $100 of foreign source income and pays $40 in foreign taxes. After applying the IRC
§ 904 limitation, the corporation would be entitled to credit $34 against its U.S. tax liability of $34. The U.S. residual
liability would be $0. It would be inappropriate, however, to add $66 to the EDA, because the corporation has only $60
($100 income-$4O foreign taxes) of after-tax earnings to distribute. Adding $66 rather than $60 would permit the
distribution of $6 of U.S. source preference income without shareholder level tax. Thus, the amount to be added to the

Notes

196

EOA should be limited to $60, which can be accomplished by applying the EOA fonnula to actual foreign taxes paid
using the higher foreign rate ($401.4 - $40).
This approach would create some complexity at the corporate level, because it would require separate tracking of foreign
taxes paid and foreign tax rates. The alternative of tracing foreign income and adding to the EOA foreign income less foreign
taxes is likely to be at least as complex.

28. A low taxable income is not necessarily inconsistent with wealth. For example, a low-bracket individual may have large
amounts of income from tax-exempt sources, e.g., tax-exempt bond interest. Alternatively, a low-bracket individual who
is retired may have a small income but a large accumulation of wealth. That is, individuals may prefer to maintain a level
of consumption over their lifetime, and thus reduce consumption during high-income working years in order to be able to
maintain consumption during low-income retirement years. See, e.g., Ando and Modigliani (1963).
29. The credit fonnula is: Credit = (DIV/.66) x(.34-t), where DIV is the dividend and t is the shareholder's marginal rate.
This credit fonnula is designed to replicate the excess credit under an imputation credit system, i.e., the difference between
the imputation credit (.34 X (DIV 1.66» and the amount of shareholder tax due on the grossed up dividend at the shareholder
rate (tx(DIV/.66».
30. Alternatively, relief for low-bracket shareholders also might take the form of a deduction. The credit formula could be
converted into a deduction formula by dividing the credit by the shareholder tax rate: [(DIV1.66) X (.34 -t)]/t, where DlV
is the net dividend and t is the shareholder's marginal rate. Thus, a shareholder in the 15 percent bracket would be entitled
to a deduction of $127 «$66/.66)x.19/.15).
31. A corporation's EOA would be allocated among shareholders in proportion to the amount of other assets distributed to
them.
32. The policy underlying the reorganization provisions is that imposition of tax is inappropriate if a corporate reorganization
merely effects a readjustment of shareholders' continuing interests in corporate property under modified corporate forms.
This policy applies equally under the prototype, because it reflects ajudgment about when income should be recognized under
a realization-based tax system that does not require corporate assets or stock to be marked to market, not a judgment about
whether two levels of tax should be imposed on recognized corporate income.
33. Under current law, earnings and profits of the distributing corporation in a divisive reorganization that qualifies as a
reorganization under IRC § 368(a)( 1)(0) are divided between the distributing corporation aI,ld the controlled corporation based
on the relative fair market value of their assets.
34. Under current law, nonliquidating distributions to shareholders are treated as dividends to the extent paid out of the
corporation's post-February 28, 1913, accumulated earnings and profits or its earnings and profits for the current taxable
year. The earnings and profits rules may be viewed as serving two principal functions with respect to dividend taxation. First,
the earnings and profits rules may be seen as a mechanism to assure that corporate preferences are not extended when
preference income is distributed to shareholders. Second, the rules may be seen as a mechanism to distinguish whether a
distribution represents a distribution of income earned on the shareholder's investment or a return of that investment.
35. IRC § 30I(c).
36. See, e.g., Andrews (1956), Blum (1975), and American Bar Association (1986).
37. Earnings and profits also are relevant in contexts other than determining dividend taxation. Earnings and profits are
relevant, for example, in determining the extent to which gain on a disposition of IRC § 306 stock is recaptured as ordinary
income, whether certain corporate divisions qualify for tax-free treatment under IRC § 355, the amount of taxes paid by a
foreign corporation that under IRC § 902 are credited to its 10 percent corporate shareholder upon receipt of a dividend,
the amount of Subpart F income that must be currently included in income by a United States shareholder of a controlled
foreign corporation, whether an S corporation with substantial passive income is subject to entity level tax on such income
under IRC § 1375 or whether such income causes the termination of S corporation status under IRC § 1362(d)(3); the amount
of any basis adjustments in the stock of consolidated subsidiaries pursuant to the consolidated return regulations, and the
amount of the adjusted current earnings adjustment for AMT purposes. In some contexts, it is possible to eliminate references
to earnings and profits or to devise alternatives that are simpler. Nevertheless, in other contexts-especially in the rules
governing the taxation of foreign income--developing simple alternatives may prove more difficult. The benefit of

197

Notes

eliminating the earnings and profits rules for purposes of dividend taxation is considerably reduced if alternatives are not
found for the rules in other contexts.
38. Recently, the American Law Institute Reporter circulated a draft memorandum that would eliminate earnings and profits
as part of its distribution-related integration proposal. American Law Institute, Reporter's Memorandum No.3 (1991), p.
5.
39. Just as under current law, however, the connection between earnings and profits and the economics of shareholder
investment is severed, however, by sales of stock and other transactions or events increasing or decreasing shareholder basis
without adjusting earnings and profits. Preserving the connection would require earnings and profits accounts to be
maintained and adjusted on a per share basis. Thus, for example, a seller of stock in a corporation with retained earnings
would recognize dividend income to the extent of the earnings and profits attributable to such stock and the earnings and
profits account for the stock would be reduced to zero. This system would not be feasible for actively traded stock.
Accordingly, the earnings and profits rules may yield arbitrary and incorrect results from the shareholder's perspective. The
alternative rules are likely to be no more accurate in distinguishing between income distributions and returns of capital
because they also do not take into account changes at the shareholder level. Indeed, by eliminating earnings and profits as
a limitation on dividend taxation, the alternative rules would tend to increase the likelihood of imposing dividend taxation
on a distribution that economically is a return of shareholder investment.
40. For a discussion of the equivalence of deducting the cost of an investment and exempting investment income from tax,
see Graetz (1979), Warren (1975), Andrews (1974), and Brown (1948).

Chapter 3
1. If income is not taxed at the corporate level (because of tax preferences or foreign tax credits), there is no additional tax
burden on retained earnings, and therefore corporations will tend to retain preference income. Under the dividend exclusion
prototype, as well as under the current system, retained preference income is taxed at the shareholder level only when the
stock is sold. To the extent that retaining preference income increases the value of stock, it also increases the capital gain
realized on the sale. Thus, distribution-related integration treats retained corporate preference income more favorably than
distributed preference income.

2. Because the shareholder allocation prototype would generally continue to tax the corporation in the same manner as under
current law, it should not significantly change a corporation's fmancial statement provision for income tax expense, taxes
currently payable, and taxes payable at a future date. The prototype's denial of carrybacks for net operating losses and
removal of the corporate AMT will, however, be reflected in the reporting of corporate tax liability for financial accounting
purposes.
The denial of carryback treatment for net operating losses may increase the provision for income tax expense in certain
circumstances. For fmancial accounting purposes, when a operating loss can and will be carried back, the tax effects of such
carryback generally increase net income, or reduce the net loss, during the loss period. See Accounting Principles Board,
Opinion No. 11 (1967), paragraph 44 and Financial Accounting Standards Board, Statement No. 96 (1987), paragraph 52.
The tax effect of the NOL carryback (which is included in the determination of net income or loss) is based on income, or
loss, reported for financial accounting purposes rather than for tax purposes. The refund of taxes expected as a result of the
carryback is recorded as a current asset. Any difference between the tax loss and financial accounting loss carryback benefit
is recorded in the deferred tax account. The shareholder allocation prototype would preclude corporations from recognizing
the benefits of NOL carrybacks.
Because the shareholder allocation prototype eliminates the corporate AMT, it would reduce the provision for tax
expense in those limited situations in which a corporation would otherwise calculate a hypothetical AMT liability. For
fmancial accounting purposes additional tax expense is only provided with respect to the corporate AMT when the application
of the AMT rules to financial accounting income would result in a hypothetical AMT liability, i.e., to the extent AMT relates
to deferral items no additional tax expense is recorded for fmancial statement purposes. The corporate AMT also affects the
fmancial statement allocation of tax expense among taxes currently payable and taxes payable at a future date. Accordingly,
the shareholder allocation prototype also could affect these allocations.
3. Because both the dividend exclusion and shareholder allocation prototypes retain the corporate interest deduction, interest
paid to tax-exempt organizations and foreign investors generally escapes U.S. tax, while corporate equity income distributed
to such investors is subject to at least one level of U.S. tax. Achieving equal treatment of debt and equity under a shareholder
allocation system would require a corporation to allocate its taxable income to both bondholders and shareholders each year,

Notes

198

whether or not interest or dividends were paid. A bondholder, like a shareholder, would be entitled to a credit for the
corporate level tax on the income allocated, and the bondholder's basis would increase by the after-corporate tax amount of
income allocated. Tax-exempt and foreign bondholders would not be entitled to claim refunds of tax credits. Unlike current
law, which requires accrual-basis bondholders to include interest in income whether paid or not, a shareholder and
bondholder allocation system might limit bondholders' interest income to the amount of the corporation's earnings.
Such a system would require rules for allocating corporate earnings to classes of debt as well as stock. The allocation
rules in such a system should provide that earnings would be allocated first to interest payable or accrued on debt, and any
remaining income would then be allocated to equity. One method for allocating income to traditional debt instruments would
determine the maximum amount of income to be allocated to a given class of debt based on the current law rules for accrualbasis taxpayers (or for holders of bonds with original issue discount). Available earnings could then be allocated to each class
of debt according to its priority, i.e., first to senior debt, then to senior subordinated debt, and then to subordinated debt.
For example, assume that a corporation has $100 of earnings and three classes of debt. The first class of debt is bank debt,
senior to the other two classes. The second and third classes are of equal priority. The interest accruing on the bank debt
is $80; the interest accruing on the second class is $30; and the interest accruing on the third class is $10. Of the
corporation's $100 of earnings, $80 would be allocated to the bank debt. The remaining $20 would be allocated
proportionately between two classes of junior debt, so that $15 (or $20 multiplied by $30/$40) would be allocated to the
second class, and $5 (or $20 mUltiplied by $30/$40) would be allocated to the third class. No earnings would be allocated
to equity.
4. For a more detailed examination of problems involved in administering a widely held passtbrough entity, including
reporting issues, allocating items (such as built-in gain on contributed property) to members, and collection issues, see
Department of the Treasury, Widely Held Partnerships (1990). Proposals are pending in the Congress to modify the conduit
treatment of certain large partnerships. Under H.R. 2777 and S. 1394, !02d Congress, 2d Session (1991) the income of
partnerships with at least 250 partners would be consolidated at the partnership level, resulting in a reduction in the number
of separate items that would be reported to partners. Audit adjustments would result in a single, current year adjustment to
partnership income, rather than adjustments to the returns of prior year partners. Under these bills, the tax administration
of large partnerships would move toward an entity approach and away from the aggregate approach that dominates current
law partnership rules.
In 1966, Canada's Carter Commission recommended a modified shareholder allocation integration system, but Canada
did not adopt the recommendation. See Royal Commission on Taxation (1966). Similarly, the United States did not adopt
the Blueprints proposal for a shareholder allocation integration system. In 1971, the Federal Republic of Germany's Tax
Reform Commission rejected a shareholder allocation integration system because of administrative complexity. See
Gourevitch, (1977), pp. 48-54. In addition, other countries have implicitly rejected shareholder allocation integration by
adopting distribution-related integration systems, although most countries have passthrough entities that are taxed under a
shareholder allocation integration approach.
5. For ease of computation, the discussion and examples in this chapter use a 31 percent corporate tax rate. The shareholder
allocation prototype could retain the current 34 percent corporate tax rate but provide credits to shareholders at a 31 percent
rate if maintaining the credit rate differential were desirable or necessary. The revenue estimates set forth in Chapter 13
assume a 34 percent corporate rate. Maintaining the corporate tax rate at 34 percent would require an adjustment in the
amount of tax passed through to shareholders to allow shareholders a tax credit no greater than the maximum 31 percent
individual rate. For example, if a corporation reported $100 of taxable income and owed $34 of tax, only $31 of tax would
be passed through to shareholders. Retaining the rate differential would necessitate numerous calculations to transform
corporate level preferences into shareholder level preferences; for example, if a corporation also had a $10 low-income
housing credit, the shareholders should be entitled only to 31134 of the credit.
6. The additional economic income sheltered by the credit, absent an upward adjustment of the shareholder's basis, will be
taxed upon distribution by the corporation or sale of the shareholder's stock.

If the corporation had a $40 credit, shareholders would be allocated $31 of tax credits, and the $9 excess credit would
be carried forward at the corporate level to the extent permitted under the Code. As discussed above, a shareholder with tax
liability less than the amount of credit allocated to him could use excess credits against other income. As in the imputation
credit prototype discussed in Chapter 11, consideration might be given to providing a carryforward at the shareholder level
for unused credits. See Chapter 11, note 33.
7.

Example. A corporation earns $100 of taxable income and pays $31 of corporate tax. The corporation's shareholders
increase their basis in their stock by $69, the after-tax income of the corporation. This achieves the same result as
a partnership that earns $100 of taxable income and distributes $31 in cash to P~!oplll' the tax.

199

Notes

8. Because the shareholder allocation prototype treats distributions first as a nontaxable return of capital to the extent of
shareholder basis and second as capital gain to the extent of any excess over basis, the earnings and profits rules are not
needed. Compare note 14, below.
9. To mitigate somewhat the effect of eliminating loss carrybacks, consideration might be given to extending somewhat the
carryforward period, for example, from 15 to 18 years, so the total period in which corporate losses could be used would
not be reduced under shareholder allocation.
10. Corporations with more complicated capital structures may require more complicated allocation provisions. See
Section 3. F.
11. While noting that corporate level payment would facilitate payment of tax, Blueprints did not include such payment in
its model system. See Blueprints, pp. 73-74. Compare IRC § 1446, which requires withholding by partnerships on income
that is effectively connected with a trade or business in the United States and that is allocable to a foreign partner.
12. If passthrough of losses were permitted, corporate losses, like partnership and S corporation losses, could be used by
shareholders to the extent of share basis. Losses in excess of share basis might be carried forward at the shareholder level.
See IRC § 704(d).
13. One method for eliminating most preferences would require corporations to allocate AMTI, rather than taxable income,
to shareholders. Each corporation would thus impute to shareholders the full amount of both taxable and preference income
(at least to the extent preference items are included in AMTI), regardless of whether the corporation was subject to the AMT.
Example. Assume that a corporation has $100 of taxable income and $30 of tax-exempt interest as its only
preference item. The corporation would not be subject to the AMT, because the tentative AMT ($26) would not
exceed the regular corporate level tax ($31). Nevertheless, the corporation would allocate $120 of income among
its shareholders.
Under this approach, corporations would continue to pay corporate level tax as under current law, at either the regular
or AMT rate, whichever is applicable. Shareholders would be entitled to credit both corporate level tax and AMT but would
not be entitled to credit corporate tax to the extent it was offset in later years by the AMT ci·edit.
The following example illustrates this method. The example assumes a 31 percent corporate and shareholder rate and
a 20 percent corporate AMT rate.
Year I

Year 2

Year 3

$100

$100

$164

Corporate Level Tax Calculation
Corporate taxable income
Corporate preference income

200

0

o

AMTI

300

100

164

Tentative AMT

60

20

33

Regular tax

31

31

51

AMT

29

0

o

0

II

18

60

20

33

AMT credit
Net corporate tax payable
Shareholder Level Tax Calculation

$300

$100

$164

Shareholder tax

93

31

51

Credit for corporate taxes paid

60

20

33

Net shareholder tax payable

33

II

18

Shareholder income

In this case, a total of $175 of tax has been paid on $564 of economic income (a 31 percent rate).
This approach would effectively eliminate corporate level preferences, whether or not distributed, by taxing corporate
preference income currently at shareholder rates. A shareholder in the 31 percent bracket would generally be liable for
additional shareholder level tax to the extent that corporate AMTI exceeded corporate taxable income. Thus, corporate level
preferences essentially would be taxed the same as corporate level taxable income, unless the absence of a corporate level

Notes

200

tax were signjficant. For example, a tax-exempt shareholder would not owe additional shareholder level tax, with the
consequence that allocated preference income would be tax-exempt (except to the extent of the corporate AMT).
14. The following approach would tax preference income to shareholders only upon a distribution or a sale of stock.
Corporations would track taxes paid, which would include payments of regular tax and AMT, as well as any AMT credits
for AMT paid in prior years. An amount of deemed income equal to the amount of income that would give rise to the actual
amount of corporate tax paid if tax had been imposed at a 31 percent rate would be allocated among shareholders. Thus,
each $1 of regular tax or AMT would give rise to $3.23 of deemed income ($11.31). Shareholders would report the deemed
income and would be entitled to a credit for corporate taxes paid. Because this approach treats the amount of income that
would be allocated to shareholders as if it had been taxed at the maximum corporate rate, no shareholder would owe
additional tax on corporate level preferences currently and lower bracket shareholders could use excess credits to offset other
tax liability. Share basis would increase by the amount of deemed income reported to the shareholder, net of the credit for
taxes paid.
The following example compares the treatment of two corporations, only one of which, corporation B, is an AMT
taxpayer. It assumes a 31 percent corporate rate and shareholder rate and a 20 percent AMT rate.
Corporation A

Corporation B

$645
350
200
0
200

0

Corporate Level Tax Calculation
Corporate taxable income
Corporate preference income
Regular tax

AMT
Total corporate taxes paid

$1,000

0
200
200

Shareholder Level Tax Calculation
Shareholder income
Shareholder tax
Credit for corporate taxes paid
Net shareholder tax payable

$645
200
200
0

$645
200
200
0

Under this approach, corporations with significant preference income would pay tax at corporate AMT rates, but no
additional shareholder level tax would be imposed currently. Additional shareholder tax would be collected only when
preference income is distributed or shares are sold. Tax would be collected at that time because share basis is increased only
by the amount of the deemed income. Thus, if a corporation has income that is taxed at less than a 31 percent rate, the
shareholders' aggregate basis in their shares will be less than the corporation's aggregate earnings available for distribution.
When distributions exceed shareholder basis (or when shares are sold for amounts in excess of basis), additional shareholder
tax will be paid.
Example. A corporation has $100 of assets and a single shareholder with a stock basis of $100. During the year,
the corporation earns $200 of preference income and pays AMT of $40. The corporation allocates $129 ($40 +- .31)
of income and $40 of tax credit to the shareholder. The shareholder's basis increases to $189 ($100 original basis
plus ($129 -$40»). The corporation has $260 of assets available for distribution. Ifthe corporation distributes $260
to its shareholder, the shareholder will recognjze gain of $71, the amount of preference income not previously taxed
at 31 percent.
Under this approach, distributed preference income is generally taxed at capital gain rather than at ordinary income rates,
because distributions in excess of basis are treated as gains from the sale of stock. In contrast, under current law and under
distribution-related integration, only retained preference income (which increases share value) is taxed as capital gains, while
distributed preference income is taxed as ordinary income.
In contrast to the treatment of dividend distributions under current law, this method treats distributions first as a return
of capital, so preference income is not taxed until share basis is exhausted. This stacking order is not consistent with the
dividend exclusion or CBIT prototypes or the imputation credit prototype, described in Chapter 11, which require that
distributions in excess of fully-taxed income be treated as taxable distributions of preference income before they are treated
as returns of the shareholder's investment. It is possible, however, to conform the stacking order in the shareholder allocation
prototype to the stacking in those prototypes. To do so, a corporation would be required to maintain an accumulated earnings
and profits account (essentially under the rules of current law). Within the earnings and profits account, the corporation

201

Notes

would maintain a subaccount for fully-taxed earnings and profits (computed by tracking taxes paid, as in the EOA). See
Section 2.B. Distributions in excess of the fully-taxed earnings, up to the amount of earnings and profits, would be treated
as taxable dividends, rather than a return of the shareholder's investment.
15.

Example. Assume that a shareholder has a basis of $10 in stock of a corporation. If the corporation earns
$100 of taxable income and receives $50 of tax-exempt bond interest in year one, the corporation would
pay $31 in tax. The shareholder would include $100 in income and would be entitled to offset the $31
shareholder tax by the $31 credit for corporate level tax. The shareholder's basis would increase by $119
(the tax-exempt interest income plus the taxable income, reduced by the amount of taxes paid). Thus, the
corporation could distribute its net cash of $119 without giving rise to shareholder level tax. This basis
adjustment differs from the $150 adjustment that would be made in a partnership because of the $31 of tax
collected at the corporate level.

16.

Example. Assume that a shareholder invests $100 in stock of a corporation. The corporation invests the
$100 of contributed capital in an asset that costs $100. Assume that the corporation earns $100 and is
entitled to expense the asset in year one, rather than depreciating it over its economic life of three years.
The deferral preference will reduce the corporation's income subject to corporate level tax in year one to
$0. In years two and three, however, the preference turns around, because the corporation will have more
income than it would have if the asset had been depreciated over 3 years. Thus, the corporation's and the
shareholder's income in years two and three will be higher.

17.

Example. A corporation's only income is $100 of tax-exempt interest on bonds described in § 57(a)(5).
Thus, its taxable income is $0 and its AMTI is $100. The corporation pays $20 of AMT. Assume that an
individual taxpayer with a 31 percent marginal tax rate holds all the stock of the corporation and has no
other income. Disregarding AMT exemption amounts, the shareholder would include the $100 of corporate
AMTI in his own AMTI, and thus would owe individual AMT of $24. The shareholder could then credit
the $20 of corporate AMT against his own AMT liability, resulting in a net AMT liability of $4.

Ifthe shareholder had other income, e. g., $100 of wage income, the shareholder would pay $31 of regular tax and $17
of AMT ($200 AMTI X .24-$31). The $20 of corporate level AMT paid at the corporate level would be creditable to reduce
the total tax due from the shareholder to $28. The shareholder would have an AMT credit of $17 to use against future regular
tax liability but no corporate level AMT credit would be allowed.
18. Permitting shareholders to credit corporate AMT paid against their regular tax liability without including any amounts
in shareholder AMTI, in effect, would refund the corporate AMT to taxable shareholders.
Example. The facts are the same as in the example in the preceding footnote. The 31 percent bracket shareholder
also has $100 of wage income. If the AMT paid at the corporate level were creditable against regular tax, but no
AMTI were imputed to the shareholder, the shareholder would pay only $11 of regular tax.
19. One approach would continue to impose the corporate AMT without any current credit to shareholders for corporate
AMT paid. Shareholders would benefit from corporate AMT payments only when the corporation made the AMT credit
allowed by IRe § 53 to reduce a subsequent year's regular tax liability. The AMT credit would be passed through to
shareholders like other credits. This rule would, however, deny integration benefits to shareholders of corporations that are
chronic AMT taxpayers, because those corporations may never use their AMT credits. This system also would require
modifications to the shareholder basis rules to decrease share basis to reflect the payment' of a noncreditable, nondeductible

tax.
An alternative rule would impute to shareholders, in addition to the corporation's taxable income, an amount of income
based on the corporate AMT paid, and allow shareholders to credit the corporate AMT against their regular tax. The
additional income imputed would equal the amount of corporate AMT paid, grossed up at 31 percent.

Example. The facts are the same as in the examples in the preceding two footnotes. Instead of including corporate
AMTI in shareholder AMTI, the corporation would allocate $64.52 of additional income ($201.31) to its
shareholder. The shareholder would then credit the $20 of corporate AMT against his regular tax liability. Thus,
the shareholder's total taxable income would be $164.52; total tax liability would be $51; and the shareholder would
be allowed to credit the corporate AMT to reduce the tax due to $31.
This approach is similar to the method described in note 13. Unlike that method, however, this rule imputes a grossed-up
amount of income to shareholders only to the extent of corporate AMT paid. As a consequence, it produces erroneous basis

Notes

202

adjustments in the case of deferral preferences, because deferral preference gives rise to partial basis when AMT is paid and
subsequently gives rise to the full amount of basis when the preference turns around and generates regular taxable income.
The basis adjustments could be corrected by continuing to calculate basis adjustments based on grossed-up taxes paid (rather
than taxable income allocated to shareholders). Such alternative basis adjustments would require complex rules, complicated
information reporting, and would create basis adjustments the timing of which differ from the timing of income passed
through to shareholders.
20. S corporations allocate income items pro rata. An S corporation allocates to each share of stock exactly the same amount
of each item arising in a taxable year. This system is simple and administrable; it works well for S corporations because they
may not have more than one class of stock. IRC § 1361(c)(4) permits classes of stock in an S corporation to have different
voting rights, but other differences generally are prohibited. Thus, the system achieves simplicity by requiring all the stock
of an S corporation to possess similar economic rights. An integration proposal that limits all corporations to a single class
of stock, however, is neither feasible nor economically desirable. The variety of existing corporate capital structures
precludes serious consideration of such a system.
2 \. IRC § 704(b).
22. Treas. Reg. § 1.704-1.
23. The complex capital account maintenance rules contained in the regulations under IRC § 704(b) illustrate the variety of
issues that would have to be addressed. An alternative approach would look to IRC § 305 to impute income to a shareholder
whose proportionate interest in the corporation increases as does the holder of class B stock in the example in the text. We
do not explore the implications of such an approach.
24.

Example. Two shareholders form a corporation and contribute $100 each. One shareholder receives
preferred stock with a liquidation preference of $100 and a return of 10 percent. The other shareholder
receives common stock, which is entitled to the remaining income and assets. Assuming the corporation
makes no cash distributions, corporate income would be allocated as follows:

Allocations

Year-End Capital Accounts

Corporate
Income

Preferred

50

10

40

110

140

2

50

11

39

121

179

3

50

12

38

133

217

Year

Common

Preferred

Common

Under the terms of the preferred stock, the liquidation preference of the preferred stock increases each year as its capital
account increases. In year one, the preferred shareholder is treated as if it received $10 and purchased an additional $10
of preferred stock. As a consequence, the preferred shareholder is allocated $11 in year two (10 percent of $110 of
preferred stock). If the corporation is liquidated at the, end of year three, the corporation has total assets of $350 and
the preferred stock has a capital account (liquidation preference) of $133. The common stock would thus receive the
remainder of the assets, or $217.
As the text notes, capital accounts would be adjusted to reflect corporate losses. Assume that the corporation is not
liquidated until year four and there is a $100 loss in year fOUi, so the corporation's assets are reduced to $250. In that
case, no income would be allocated to either shareholder in year four, but the $100 loss would reduce the common
shareholder's capital account to $117. Upon liquidation at the end of year four, the preferred shareholder would receive
$133 and the common shareholders would receive $117.
25. The full integration proposal in Blueprints used an annual record date method and designated the shareholders on the
first day of the taxable year as the shareholders of record to avoid "trafficking" in shares of loss corporations at year end.
Blueprints, pp. 70-71, rejected a "last day" rule because, at year end, the market would have information indicating that the
corporation would incur a tax loss for the year, and shares could then be sold to high-bracket taxpayers to whom the loss
would be most useful. Because the shareholder allocation prototype does not permit the passthrough of losses to shareholders,
loss trafficking is not an issue. The quarterly record date approach also minimizes tax-motivated year-end trading to capture
credits for corporate taxes paid by limiting the benefit of year end ownership to one quarter of income and its proportional
share of tax.

203

Notes

26. It may be desirable to allow (or require) corporations to close their books under certain circumstances. For example,
a seller of a majority stock interest in a corporation may wish to ensure that income generated by activities after the sale will
not be allocated to her. Similarly, the government could have an interest in requiring closing of the books after extraordinary
corporate events to assure that net income and loss are allocated to the appropriate shareholders.
27. The effect of A's loss is to defer taxation of $10.35 of corporate income until the purchaser sells his stock. If A can fully
use the capital loss, A's loss offsets the tax on $10.35 of corporate income. The purchaser, however, has a basis of $144.85
($117.25 plus $27.60) in the stock of a corporation having assets with a value of $155.2,0. The purchaser thus has built-in
gain of $10.35 in his stock.
28. The Code provides that a partnership's taxable year closes with respect to a partner whose entire interest is sold. See
IRe § 706(c). If a partner's interest varies during a year, the Code simply provides the general rule that tax items are to
be allocated to take into account this variation. Specific rules are provided for a few items of cash basis taxpayers, such as
interest and taxes, which must be allocated on a per day basis throughout the taxable year. See IRC § 706(d)(2).
29. In that case, each prior quarter's income would be unaffected by subsequent events, and each future quarter's income
would be allocated to the purchaser.
30. We also rejected the alternative of allocating a corporation's income on a per share per day basis throughout the taxable
year. Although current law employs this system for S corporations, which must allocate income among stockholders on a
strict pro rata basis, including daily allocation of income where there has been an ownership change, we believe that this
system could not successfully be applied to large corporations with publicly held stock in which there is frequent trading.
Publicly traded partnerships are widely held, publicly traded entities that are required under current law to allocate certain
items among partners on a per day approach. However, these partnerships typically adopt conventions to minimize the
difficulties of tracking frequent changes in ownership, for example, by allocating each month's share of partnership income
to the partner holding the partnership unit on the first day of the month. Compared to publicly traded partnerships, publicly
held corporations have more shares of stock outstanding, and the stock is traded more frequently; for example, trading of
the most actively traded stock can exceed one million share per day. A per share per day approach would require tracking
of many millions of transfers during a year, and therefore a daily allocation method would be impractical for publicly traded
corporations.
31. The Blueprints system is one example. That system did not include a corporate level tax, taxed capital gains at ordinary
income rates, and permitted unlimited use of capital losses against ordinary income. See Blueprints, p. 77. Accordingly, the
Blueprints system permitted a shareholder of record who sold stock during the year to calculate gain or loss calculated by
reference to his basis at the beginning of the year, based on the observation that the allocation of current year income would
not affect the difference between the sale proceeds and his basis as of the date of sale. The corporate income or loss that
he would have to report as the shareholder of record would be exactly offset by a corresponding basis adjustment. See
Blueprints, pp. 71-72.
The results are somewhat different under the shareholder allocation system, which retains a corporate level tax. The
introduction of a corporate level tax means that allocations of taxable income increase share basis but do not create any
additional shareholder level tax liability (because the corporate tax rate is at least equal to the maximum shareholder rate).
For example, under the Blueprints system, an unexpected increase in allocable earnings of $100 would increase a selling
shareholder's taxable income by $100 but would increase basis (and reduce gain, or increase loss, on sale) by the same
amount. Ignoring differences in character (which may have significant consequences), the shareholder's total income would
be the same. Under the shareholder allocation prototype, however, an unexpected $100 increase in earnings would result
in an allocation of $100 of earnings and $31 of tax credits. If this increase occurred for a period prior to the period in which
the sale took place, e.g., an unexpected increase in earnings for the first quarter with respect to stock transferred in the
second quarter, the withholding credit will be available to the selling shareholder. The parties to the transfer would need to
estimate the potential for material changes in earnings on pricing the stock. Blueprints acknowledged that the addition of a
corporate level tax complicates calculation of gain on sale. Blueprints, p. 74.
The current treatment of capital gains and losses would complicate calculations under a record date system. A
shareholder who sold stock with a basis of $100 for $150 might not be indifferent between $50 of capital gain (if gain were
calculated at the time of sale) and $75 of ordinary income and a $25 capital loss (if calculation of gain were deferred and
the corporation earned $75 for the year).
32. Where corporate tax is imposed at a rate greater than or equal to the maximum individual rate, the government does not
suffer from delay in attributing income to the proper corporate entity. An upper tier corporation that held stock in a lower
tier corporation might be required to report its income from the lower tier corporation with a one year delay. Thus, if an

204

Notes

upper tier corporation purchased stock in another corporation during year one, the upper tier corporation would report no
income from the investment in year one. The upper tier corporation's share of the lower tier corporation's year-one income
would be reported in year two. The government's interest would not suffer, as the lower tier corporation's income would
have been subject to tax at the corporate rate in year one. The upper tier corporation and its shareholders would, however,
suffer a detriment to the extent that the corporate rate exceeds shareholder rates and shareholders would have been entitled
to use excess credits for corporate taxes paid. In that case, the upper tier corporation's shareholders have, in effect, made
an interest-free loan of the excess credits to the government.
Such a system could be restricted to situations where the upper and lower tier corporations have identical taxable years.

If taxable years differ, the upper tier corporation would report the lower tier income in its taxable year in which the lower
tier corporation's taxable year ends. If two corporations own stock in each other, this system could result in a continuous
delay in proper attribution of the income. Under such a system, taxpayers would have an incentive to structure their
investments to minimize relationships that cause detrimental reporting delays. To the extent such arrangements are
impractical, however, a shareholder allocation system would treat intercorporate investments more harshly than direct
investment.
33. The pending tax simplification bills would adopt a similar approach for large partnerships. See The Tax Simplification
Act of 1991, H.R. 2777 and S. 1394, to2d Cong., 1st Sess. (June 26, 1991). See also U.S. Department of the Treasury,
Widely Held Partnerships (1990).
34. This problem is closely analogous to the problem of extending preferences to shareholders, discussed in Section 3.E.
35.

Example. A U.S. corporation's only income is a dividend from a foreign subsidiary. Under IRC § 902,
the corporation includes $100 in income and receives a credit for foreign taxes paid of $40. Under the
foreign tax credit limitation rules of IRC § 904, the corporation's foreign tax credit is limited to $31. The
corporation's sole shareholder is Shareholder A who has a marginal tax rate of IS.percent and wage income
of $100. Without foreign tax credit limitation rules at the shareholder level, Shareholder A will treat $31
as a credit for taxes paid and use the excess credit of $16 to offset all tax due on his wage income.

Section 11. D discusses the feasibility of using a shareholder level exclusion of foreign source income to avoid the
application of IRC § 904 at the shareholder level if foreign taxes were treated like U.S. taxes under the imputation credit
prototype.

Chapter 4
1. Although there are no existing models of this prototype, others have suggested a similar approach using a bondholder
credit. See, e.g., Steuerle (1989) (describing a "simplified integrated tax" that would be withheld by corporations at the
maximum individual or corporate rate); Seidman (1990) (describing an FDIC proposal to require corporations to withhold
34 percent of all their dividend and interest payments and require recipients to report the grossed-up amount of the
distributions and claim a credit for the tax withheld by the corporation); H.R. 4457, WIst Cong., 2d Sess. (1990) (introduced
by Congressman Vander Jagt, and proposing an approach similar to the FDIC proposal outlined by Mr. Seidman). For
proposals that resemble CBIT even more closely, see Jacobs (1987) (describing a 28 percent "single business tax" on capital
income that would be imposed by disallowing business interest deductions and excluding interest and dividends from
investors' taxable income); Bravenec (1989) (describing a "nontraditional approach to integration" that would deny
corporations interest deductions and exclude from income of investors dividends and interest received from corporations).
The financial accounting ramifications of CBIT are, in many respects, the most direct of all the integration prototypes.
The nondeductibility of interest expense would increase corporations' income tax burden, thereby increasing the provision
for income taxes and reducing earnings per share. Generally, we would expect an increase in the provision for income taxes
and a reduction in earnings per share for net borrowers. In the rare case of certain net lenders, the provision for income taxes
could be reduced and earnings per share could be increased. Because nondeductibility of interest expense would increase
taxes currently payable, CElT also would serve to increase the reported current liability for income taxes and the cash flow
requirements associated with this current liability. The recommended gradual phase-in of CBIT should allow for gradual
changes in capital structures and enhance the comparability of interperiod financial results.
A less obvious financial accounting effect of CBIT arises if a compensatory tax is imposed. The standards for accounting
for income taxes generally require corporations to recognize as income tax expense both the taxes currently payable and the
taxes that are payable during a future period but are, nonetheless, associated with earnings during the current period. See
Accounting Principles Board, Opinion No. 11 (1967), paragraph 34, and Financial Accounting Standards Board, Statement

205

Notes

No. 96 (1987), paragraph 7. Under these standards, income tax expense recognized by a CBIT entity would include the
potential compensatory tax liability that is associated with preference income which is earned and retained by the entity.
Thus, the compensatory tax could serve to further increase the provision for income taxes. The fmancial accounting for a
compensatory tax has never been formally considered, however, and it is conceivable that the financial accounting authorities
might permit corporations to disregard potential tax expense associated with future compensatory taxes provided the
corporation's earnings distribution policy suggests that the likelihood of a distribution of preference income is remote. The
Accounting Principles Board has adopted such a position with respect to the provision for taxes that may arise with respect
to distributed earnings of subsidiaries, e.g., foreign subsidiaries or subsidiaries that are not consolidated for tax purposes.
See Accounting Principles Board, Opinion No. 23 (1972), paragraphs 9-14.
2. See Chapter 10.
3. CBIT is related to, but not identical with, a bondholder credit system that taxes interest income at the debtholder level
through an imputation credit system. CBIT differs from a bondholder credit system where the borrower and lender have
different marginal tax rates. See Section II.H, which describes a bondholder credit system and discusses the differences
between that system and CBIT.
4. See Section 13.H. If gains on sale of CBIT equity and debt are not subject to tax, losses on such securities also would
not be allowed. Given the difficulty of the analysis of capital gains in the context of integration (see Chapter 8), we simply
note here that the CBIT prototype would be revenue neutral at a 31 percent rate with full exclusion of capital gains and losses
on sales of CBIT equity and debt at the investor level.
5. See Sections 4.F and 6.D.
6. Compare Sweden's flat rate tax on capital income, adopted in 1991 as part of a comprehensive tax reform package. See
Swedish Ministry of Finance (1991) and Lodin (1990). Under the new system, a flat tax rate of 30 percent applies to all
capital income received by individuals, including dividends, interest, and capital gains. Earned income is taxed separately,
at graduated marginal rates ranging from approximately 31 to 50 percent. Unlike CBIT, Sweden's flat rate tax on capital
is not an integration proposal. Sweden generally retains the classical system of corporate taxation, taxing corporate income
at a rate of 30 percent. The Swedish system provides a limited dividends paid deduction for new equity and a "tax
equalization reserve" that reduces the effective tax burden on retained earnings to approximately 23 percent. Swedish
Ministry of Finance (1991), p. 39.
7. A gradual phase-in also would provide an opportunity to evaluate the extent to which imposing one level of tax on interest
paid to tax-exempt and foreign investors might induce those investors to change the composition of their portfolios or the
level of their investment in U.S. business. Adjustments in the application of CBIT to these investors can be adopted to reduce
such effects if undesirable portfolio shifts or changes in capital flows occur. See Section 4.F. Partial steps toward a CBIT
regime that would narrow distinctions between debt and equity also are possible on a revenue neutral basis. See Section 6.D.

8. As recommended, the CBIT prototype can use a 31 percent rate--equal to the top individual marginal rat~rather than
a 34 percent rate without losing revenue relative to current law. See Section 13.H.
9. Carrybacks would not, however, be permitted if they would create a negative balance in the EDA.
10. Fully-taxed income is determined in the same manner as under the dividend exclusion prototype. See Sections 2.B and
4.D.
11. Several nations have expressed concern about their increasing inability to tax capital income, and some interest has been
shown in the adoption of a withholding tax of 10 to 15 percent on capital income, although concern over the potentially
adverse implications of the unilateral adoption of such a tax has precluded general acceptance of such a tax. In 1989, the
European Commission (EC) proposed a 15 percent withholding tax on savings bank and bond interest income earned by
residents of the EC. This proposal, which would not have affected Eurobonds or resid'!nts of countries outside the EC, was
not accepted, although an informal meeting of the Finance Ministers of the member countries supported a withholding tax
on capital income if such tax also were supported by the United States, Japan, and other countries. See TUITO (1989). EC
Tax Commissioner Madame Scrivener subsequently proposed a 10 percent tax on interest income, but this proposal also was
not generally accepted; see Goldsworth (1990). Since then, Madame Scrivener has continued to express the view that a
general withholding tax on interest income is the best solution to the problem of tax avoidance in a world of increased capital
mobility. See Nagle (1990) and Daily Tax Report (November 8, 1991).

Notes

206

12. This Report explores CBIT as an integration prototype directed to the taxation of equity and debt income generated by
businesses. The CBIT approach, however, might be extended to other types of interest income. Such an expansion of CBIT
might provide a means of taxing all interest income at a uniform rate. Economic efficiency suggests that taxing capiial income
at a uniform rate might improve welfare. ·While an expanded CBIT approach is beyond the scope of this Report, we note
that it raises difficult issues.
Home mortgage interest would be one important issue in considering an expanded CBIT regime. Under current law,
home mortgage interest generally is deductible by the payor and includable in the income of the recipient. While the basic
CBIT prototype retains the current law treatment, an expanded CBIT regime might subject home mortgage interest to CBIT.
Subjecting home mortgage interest to the CBIT rules would ensure that one level of tax is collected on home mortgage
interest. Under current law, home mortgage interest paid to tax-exempt or foreign investors (who may hold mortgage
passthrough certificates) escapes the U.S. tax base entirely. Depending upon the level of interest rates following adoption
of an expanded CBIT regime, the average homeowner with a mortgage might be better off with CBIT treatment than with
the deductibility of current law.
In addition, if all capital income were taxed at a single rate at the payor level, the distinction between interest and other
types of capital income that may have a significant interest component would become more important. "Identifying Disguised
Interest" in Section 4. G discusses the implications of CBIT for the current law distinction between true leases that are treated
as leases and financing leases that are treated as loans. That section reflects our judgment that, under the CBIT prototype,
no important changes in current rules for distinguishing between interest and other types of capital income are necessary.
In an expanded CBIT regime, however, the pressure on the line between interest and other capital income would be greater.
13. hterest and dividends received from a nonCBIT business would be included in the taxable incomes of individual and
business investors, and capital gains realized on the disposition of interests in nonCBIT businesses would be taxable without
regard to any change due to CBIT.
14. We anticipate that entities might move freely from CBIT to nonCBIT status based on annual gross receipts, i.e., a
business which had gross receipts of $75,000 in year I, $125,000 in year 2, and $75,000 in year 3 would report its income
under current law provisions in years 1 and 3 and file a CBIT return in year 2. CBIT tax paid in year 2 would allow payment
of tax-free distributions attributable to the taxed amounts in year 3 and later nonCBIT years. The impact of year to year
changes would cause some complexity and would cause a rate notch effect as an entity moves in and out of CBIT status.
An alternative would allow organizations that generally meet the gross receipts test to remain nonCBIT entities until they
have exceeded the floor for several years.
15. If the lower bound were higher, an aggregation rule probably would be required. The least complicated approach would
require individuals with more than a threshold amount, e.g., $100,000, on Schedules C and K of their Forms 1040 to pay
tax at a schedular rate of 31 percent on the excess. While this approach would inhibit multiplication of entities to avoid the
CBIT loss limitation, it would not be effective to prevent use of mUltiple entities to evade the CBIT interest deduction
disallowance rule. A refinement could require all nonCBIT small entities to report to their shareholders and partners their
deductions for business {nterest paid. (Individual proprietors would, of course, know this amount for Schedule C activities.)
Individuals could then be required to add these amounts to the income reported on Schedules C and K in computing the
schedular tax described above.
16. An alternative would adopt graduated CBIT rates to reduce the impact of CBIT on small businesses. Because the 31
percent CBIT rate eqlL1ls the top individual rate, this would have the effect of imposing CBIT at rates identical to those at
the individualleve!. The principal disadvantage of this approach is that it would require complex rules to combat multiple
use of the graduated rates by common owners. Compare IRC § 1551 (denying the benefit of graduated rates to corporations
under common contro\).
Another alternative that we rejected as unduly complex would subject all corporations and unincorporated businesses
to CBIT, but tax all income of owner-managers at their personal rates rather than at the CBIT rate. Once owner-managers
have been identified, the business would proceed to calculate its CBIT tax, excluding the share of profits and other income
attributable to the owner-managers (whether that income is called salary, bonuses, partnership income, dividends, or interest)
from the CBIT income of the business. The owner-managers then would include these amounts in their personal income when
they calculate their taxes. This alternative, however, would introduce a set of complexities that a receipts-based exception
avoids. One example would be the need to separate all interest income and expense items between their business and persJnai
cOP.Jponents. Some taxpayers will see this task as unnecessarily difficult, while others will see it as an opportunity for tax
planning. For example, a proprietorship operated out of the proprietor's home should bear a (nondeductible) portion of the
home mortgage interest expense. Additional rules would be needed to address these problems. Taxpayers would likely find
the rules to be complex, arbi trary, and unfair.

207

Notes

The criteria for being considered an owner-manager might be similar to the requirements for "material participation"
under the passive loss limitations of mc § 469. Another possible set of criteria would treat as owner-managers all individuals
who report net earnings from self-employment under IRC § 1402. (Net earnings from self-employment, as defined in IRC
§ 1402(a), would have to be modified for CBIT purposes by adding back all the capital income that is excluded from the
current self-employment tax. See, e.g., IRC § 1402(a)(I) and (2), which exclude most rents, dividends, and interest from
self-employment income.) A third possibility would follow the concept in IRC § 911(d)(2)(B), which identifies individuals
who are engaged in trades or businesses in which both personal services and capital are material income-producing factors.
That identification also was used to apply the maximum tax on earned income of former IRC § 1348, repealed in 1981, and
the IRS and the courts developed a considerable body of law on whether services and capital are material income-producing
factors in a given trade or business.
17. On the other hand, imposition of tax on distributed preference income (at either the corporate or shareholder level) may
be viewed as retaining, in small part, the current system's bias against dividend distributions. See Chapter 5.
18. See Section 2.B. To illustrate the functioning of such a mechanism under CBIT, assume that a corporation earned $100
of taxable income and $100 of preference income. The corporation would pay tax of $31 and would add $69 of fully-taxed
income to its EDA. The balance in the account would translate into $69 of excludable distributable income. Thus, if the
corporation distributed $75 during the year, $6 would be deemed made from preference income and would be includable
in the investor's taxable income.
19. Other solutions may be possible. For example, a compensatory tax could be imposed, but a tax credit like that described
in Section 4.F could be provided to tax-exempt and foreign investors. A compensatory tax would raise sufficient revenue
to allow a refund of up to 50 percent of entity level taxes paid to tax-exempt and foreign investors. We expect that such a
credit would significantly reduce the distortion in payout decisions the compensatory tax would create. As Section 11.B
discusses, the compensatory tax creates a real increase in the tax burden on distributed preference income because we do
not recommend refunding it to tax-exempt and foreign investors. If the compensatory tax were completely refundable to such
investors, the amount of tax collected from investments by those investors would remain the same, and one would expect
businesses and investors to adjust, in the long run, to what is merely a change in the collection mechanism without an
additional burden. A partial refund of entity level tax would mitigate the distortions created by a compensatory tax. See also
Section 6.D.
20. If a compensatory tax is adopted in CBIT, consideration could be given to allowing payments of compensatory tax to
be credited against subsequent regular tax liability. Such a rule would allow the most taxpayer-favorable stacking of taxable
income and preference income earned in different years. However, the existence of excess compensatory tax carry forwardslike excess ACT accounts in the U.K. system-may create "trafficking" concerns. See American Law Institute, Reporter's
Memorandum No.3 (1991).
Example. A corporation earns $100 of preference income in year 1 and distributes $69, incurring $31 of
compensatory tax. In year 2, the corporation earns $100 of taxable income and owes $31 of tax, which is offset by
the previous year's payment of compensatory tax. The corporation now has a zero EDA and will owe $31 of
compensatory tax when it distributes the second year's income.

If compensatory tax is not creditable against regular tax liability, the corporation would owe $31 of regular tax in year 2 but
would have a $31 EDA. This is the approach we generally follow in discussing a compensatory tax under CBIT.
21. The CBIT prototype uses the imputed interest and OlD rules to distinguish payments of interest from payments of
principal; similar rules may be required for preferred stock. See "Current Law Interest Deduction Limitations Under CBIT,"
in Section 4.G. These rules are necessary to ensure that payments representing a return of debt or preferred stock capital
do not reduce the EDA and are not subject to compensatory tax or investor level tax.
The role, if any, of the current earnings and profits rules requires reconsideration under CBIT. Although earnings and
profits could be computed under CBIT principles, i.e., without an interest deduction, it is unclear whether those rules would
be necessary or appropriate as an additional (or alternative) mechanism for identifying payments that represent a return of
equity or debt capital. The dividend exclusion prototype, which applies only to stock, retains the earnings and profits rules.
See Section 2.F.
22. The tax paid would result in an addition to the EDA and would ensure that the income would not be taxed again when
redistributed.

Notes

208

23. It may be desirable to provide a 100 percent deduction without regard to the degree of affiliation between the payor and
the recipient. Although the dividend exclusion prototype retains current law, that prototype applies only to equity. Under
CBIT, which applies to both debt and equity, there seems to be no reason to accord a larger deduction to a related creditor
than to a portfolio creditor, and maintaining parity between debt and equity requires the same treatment for shareholders.
24. Imposing a 31 percent tax on all individual income in excess of $100,000 reported on Schedules C and K of Form 1040
might be required to achieve these simplifications. See note 15 supra.
25. Historically, the corporate and individual minimum taxes were enacted in response to public perceptions that corporations
and individuals with substantial economic income were not paying any income tax. Although eBiT may result in some
taxpayers not writing checks to the IRS (because most of their income is excludable CBIT interest and dividends), individuals
do not in fact escape tax on interest and dividends paid by a CBIT entity, because the investors' income tax is prepaid at
the entity level and at the CBIT rate (which is equal to the top individual rate and exceeds the individual AMT rate).
26. Other countries with integrated systems of corporate taxation typically treat foreign source income in a similar fashion;
the domestic tax on foreign source income that is not initially collected because of foreign tax credits (or an exemption rule)
is collected, at the shareholder's tax rate, when the foreign source income is distributed to resident shareholders. Collection
of this tax is not considered inconsistent with income tax treaty obligations to grant relief for foreign taxes. If a compensatory
tax were imposed under CBIT, the domestic tax would be collected at the 31 percent CBIT rate, rather than the rate paid
by the shareholder on its other income.
27. See U.S. v. Goodyear Tire & Rubber Co., 493 U.S. 132 (1989).
28. Under this approach, the CBIT prototype collects U.S. tax currently on foreign source income of a branch used to pay
interest. We view this as the correct approach. Unlike other differences typically found between the U.S. and foreign
computations of the foreign source income base (e.g., depreciation or inventory), the treatment of interest under CBITwouid
be a major systemic difference. The decision not to permit a foreign tax credit against the portion of a branch's foreign
source income base used to pay interest can be analogized to placing such income in a separate limitation or "basket" under
IRC § 904( d). Since the foreign jurisdiction can be expected never to impose tax on this income, it is appropriate to prevent
the averaging of high foreign taxes imposed on other foreign source income against the "zero" rate of tax imposed on the
income used to pay interest.
We recommend that the foreign tax credit limitation be computed as the lesser of (1) .31 times foreign source income
computed with a deduction for interest expense allowable under foreign law and (2) actual U.S. tax liability. This approach
has a disadvantage in that dividend income received by aU. S. corporation from a foreign subsidiary will be included in the
foreign source income base without a reduction for interest expense allocable to the corporation's investment in that
subsidiary, i. e., because that interest expense will not be deductible for foreign tax purposes. The resulting inflation of the
limitation will permit the U. S. corporation to absorb excess foreign tax credits generated by non-dividend income.

An alternative approach would compute the foreign tax credit limitation by taking into account the interest expense that
would be deductible and allocable to foreign source income under current law rules. See IRC §§ 861 and 864. Under this
approach, the foreign tax credit limitation formula would be: .31 X (worldwide income) x (foreign source income/worldwide
income), where worldwide income is reduced by interest expense that would be deductible under current law and foreign
source income is reduced by interest expense that would be allocable to such income under current law. An obvious
disadvantage of this approach is that it would require the retention of current law provisions that determine the deductibility
and allocation of interest expense. On balance, the choice between these alternatives depends upon whether the complexities
associated with retention of current law interest rules are more or less acceptable than the potential averaging that would arise
from reliance on foreign law. See also Section 4.G.
29. Computation of the earnings of a foreign subsidiary without a deduction for interest might be considered appropriate on
the ground that such earnings are calculated under IRC § 902 in order to determine the U.S. tax liability of the U.S.
corporate shareholder (a CBIT entity), and not of the foreign subsidiary. In other words IRC § 902 deems the U.S. corporate
shareholder to have earned the earnings used to pay the dividends it receives from the foreign subsidiary and to have paid
the associated foreign tax. If this approach were adopted, an indirect credit could be granted for interest payments received
by a 10 percent U.S. corporate shareholder from a foreign subsidiary. Compare IRC § 904(d)(3)(C). A U.S. corporate
shareholder receiving both interest and dividends from a foreign subsidiary with no other creditor would then receive a full
indirect credit for foreign taxes paid by the subsidiary. This would permit the use of foreign tax credits to shelter the interest
income from U.S. tax, however, which, as discussed in the context of foreign branch income, we consider objectionable.
Moreover, in cases where a foreign subsidiary paid interest to a creditor other than a 10 percent U.S. corporate shareholder,
this approach would result in the stranding of foreign tax credits at the subsidiary level. Specifically, the computation of

209

Notes

foreign subsidiary earnings without an interest deduction would reduce the indirect credit available to aU. S. corporate
shareholder with respect to dividends received from the subsidiary, i.e., because those dividends would represent a reduced
proportion of a larger, hypothetical amount of subsidiary earnings. It would be impossible for the U. S. shareholder to obtain
a credit for the full amount of taxes paid with respect to income distributed as dividends because a portion of such taxes
would be deemed to have been paid on income paid out as interest to a third party creditor. This would be the case, even
though the foreign subsidiary was not actually taxed on income paid out as interest, by virtue of the availability of an interest
deduction for foreign tax purposes. To avoid this result, we have proposed that the earnings of a foreign subsidiary be
calculated for IRC § 902 purposes with an interest deduction based on the interest expense claimed under foreign law.
30. In the case of foreign operations conducted through a foreign partnership, this may raise an issue of comparability with
a foreign branch. This issue is discussed below at note 37.
31. Introduction of CBIT might induce some U.S. corporations to reorganize foreign branch operations as foreign
subsidiaries. The nondeductibility of interest under U.S., but not foreign, tax law would effectively reduce the foreign taxes
available to offset U. S. tax, thus providing greater incentives for operating in corporate form abroad in order to defer U. S.
taxation.
32. The branch profits tax also would be repealed because, in the absence of a dividend withholding tax, it would no longer
be needed to maintain parity between U.S. branches and U.S. subsidiaries of foreign corporations.
33. Significant exceptions to the portfolio interest exemption, i.e., interest paid to a foreign bank on a loan made in the
ordinary course of business and interest paid to related foreign persons, give the United States some leverage to obtain
withholding rate reductions in treaties negotiated under current law.
34. See IRC § 882(c) and Treas. Reg. § 1.882-5.
35. Note that the 30 percent withholding rate would perform a function here analogous to the 31 percent schedular tax
discussed in note 15. Reduction or elimination of the 30 percent tax by treaty might encourage the use of multiple small
business entities to avoid CBIT.
36. The term "nonCBIT debt" refers to debt issued by entities that are not subject to CBIT. NonCBIT debt includes Treasury
securities, home mortgages (and mortgage pass through certificates), debt issued by tax-exempt entities, and debt issued by
foreign governments and businesses, all taxable to U.S. persons. State and local government debt is nonCBIT debt also;
however, it would remain tax exempt to the extent provided in current law.
37. U.S. CBIT entities needing funds for foreign operations could borrow through foreign subsidiaries. Borrowing through
a foreign branch would not be desirable, however. Because a foreign branch would be a component of a CBIT entity, it
would not be permitted to deduct interest expense. Thus, the branch would probably fmd it advantageous to borrow in the
United States (where its ability to pay excludable interest could be expected to produce a lower interest rate) rather than
paying higher, nonCBIT interest rates that would be required to attract foreign lenders. An alternative would treat foreign
branches as if they were foreign subsidiaries for CBIT purposes. Interest paid by a foreign branch would then be deductible
by the branch and taxable to the lender. Rules similar to those of IRC § 861 (a)(l)(B)(i) (providing foreign sourcing for
interest paid by foreign branches of U.S. banks on bank deposits) could be applied to avoid the imposition of any applicable
CBIT on such interest paid to a foreign lender. This approach would raise numerous technical and compliance issues. For
similar reasons, borrowing through a foreign subsidiary would not be advantageous if borrowed funds were to be used in
the United States.
38. Alternatively, the credit could be fully refundable, without regard to the taxpayer's other tax liability. Making the credit
nonrefundable is, however, consistent with the decision in Chapters 3 and 11 not to permit refunds of excess imputation
credits to low-bracket shareholders and with the treatment of tax-exempt and foreign investors described in the text below.
Although interest and dividend income would not be taxable under CBIT, most low-bracket individuals who would invest
in CBIT entities should have sufficient tax liability on wages and nonCBIT income to use the CBIT investor credit.
39. See also American Law Institute, Reporter's Memorandum No.3 (1991).
40. See Chapter 6. Under a distribution-related integration system that denies refunds of imputation credits on corporate
dividends, tax-exempt investors would have an incentive to invest in debt rather than equity. By imposing a tax on investment
income, the taxation of debt and equity would be conformed, and tax-exempt entities would have an incentive to invest in
dividend-paying stock to use the excess imputation credits against the tax due on other income. This structure would

Notes

210

encourage tax-exempt entities to hold a mixture of debt and equity, since the excess credits associated with corporate
dividends could be used to offset the tax due on other kinds of investment income.
41. In theory, the policies which led Congress to enact IRC § 263A(f) would justify its retention for small business entities;
however, given the capitalization threshold for application of IRC § 263A(f)(I)(B) (assets costing more than $1 million or
having long life or production period), retention of its complexity may not be justified for the few situations in which it
would apply. In contrast, absent special rules to equate self-constructed and purchased assets, capitalization of interest for
CBIT entities could undercut the CBIT revenue base by converting some nondeductible interest into basis eligible for cost
recovery.
42. The rules of IRC § 265 would, however, be expanded to limit the deduction of expenses attributable to CBIT interest
and dividend income. See Section 4.1.
43. A similar expansion of IRC § 265(a)(4) to cover regulated investment companies and other conduits which hold stock
and debt of CBIT entities also will be required. See Section 4.H.
44. If A's lender were taxable, the disallowance of interest deductions to A would result in the collection of a double tax.
However, the potential for tax arbitrage described in the text led us to adopt the disallowance solution.
45. As discussed in Section 4.E, the prototype computes the foreign tax credit limitation by calculating a branch's foreign
source income taking into account the interest deduction allowed to the branch under foreign law. The alternative is to fe{{uire
allocation and apportionment of interest expense to the foreign source income as under current law. In that case, the
provisions listed in the text would continue to be relevant for purposes of determining the foreign tax credit limitation.
46. For example, if the seller enjoys a reduced rate on capital gains, compared to a zero rate on CBIT interest, this tension
will be reduced, but not eliminated. See also Chapter 8.
47. The Service's guidelines for ruling that a lessor is the owner of assets for tax purposes (and hence that the lessee's
payments are rents) include rules governing (1) the length of the lease compared to the useful life of the property, (2) the
residual value of the property at the end of the lease, (3) options to purchase or sell property at the end of the lease term,
and (4) the lessor's e{{uity investment in the property. See Rev. Proc. 75-21, 1975-1 C.B. 715. See also Rev. Proc. 75-28,
1975-1 C.B. 752, Rev. Proc. 76-30, 1976-2 C.B. 647, and Rev. Proc. 79-48, 1979-2 C.B. 529.
In theory, every leasing transaction has an interest component, because the lessee obtains current performance (the
possession of the property) but makes deferred payments. In that sense, a lease is economically similar to an installment sale
of the property. Compare Halperin (1986) (several different types of accelerated or deferred payments contain implicit loans);
Mundstock (1991) (economic equivalence of loans and leases). The degree of similarity between the two, however, depends
on several factors, including the term of the lease agreement and the rights retained by the lessor with respect to the property.
The tax law historically has respected a broad range of leases, and we do not think it necessary to change that treatment in
the move to CBIT, although it would be possible to consider CBIT treatment for certain rents and royalties.
48. That the courts' efforts in this area have led to inconsistent results is hardly surprising given the factual nature of each
inquiry into who is the true owner of property that is the subject of complex contractual arrangements between parties. No
case shows this inconsistency better than the Supreme Court's only examination of this area in the last 50 years, Frank Lyon
v. United States, 435 U.S. 561 (1978), rev'g 536 F.2d 746 (8th Cir. 1976), rev'g 75-2 USTC 1 9545 (E.D. Ark. 1975).
Based on all of the facts and circumstances, the trial court upheld the taxpayer's contention that it was the true owner of the
building. The Court of Appeals, however, analogizing the rights of a property owner to a bundle of sticks, agreed with the
government's argument that taxpayer "totes an empty bundle and that the term 'owner' for tax purposes cannot reasonably
be attached to the empty wrapping taxpayer has retained." 536 F.2d at 751. The Supreme Court then undertook its own
evaluation of the facts, and cited some two dozen facts to support its conclusion that the taxpayer was the tax owner of the
building. Statutory standards might help the courts to reach more consistent results.
49. See IRC §§ 483, 1274. IRC §7872 also should be retained in order to characterize properly the interest component of
certain below-market loans.
50. It may be possible to simplify the current OlD rules for CBIT debt, because neither the issuer nor the lender must
currently accrue deductions or income. Thus, it may be sufficient to adopt rules that correctly identifY the character of
payments. Compare IRC § 483. Similar rules may be needed to distinguish dividend payments from redemption payments
on preferred stock. See § 305(c). The treatment of capital gains under CBIT may, however, result in some retention of the
current timing rules. If capital gains on CBIT debt are taxed, it may be appropriate to provide debtholders with an increase

211

Notes

in basis (with a corresponding debit to the issuer's EO A) to ensure that accrued discount on CBIT debt is not taxed as capital
gains when the debt is sold. See Section 9.B.
51. Consideration might be given to providing Treasury with the option of issuing both taxable debt and tax-exempt debt.
52. See IRC § 103.
53. The exemption also may permit distributions to be taxed at a lower rate, if the beneficiary is in a lower tax bracket after
retirement.
54. "CBIT income" refers to dividends and interest on CBIT debt and equity (and, if capital gains on CBIT debt and equity
are exempt from tax under CBIT, capital gains on such assets). The two accounts would increase when the pension fund
receives contributions, nonCBIT income, or CBIT income, and would decrease when the pension fund makes distributions
to beneficiaries. If CBIT income were reinvested in nonCBIT assets, only the return on those assets would be added to the
nonCBIT income account. If no compensatory tax is adopted, CBIT income would include only excludable CBIT interest
and dividends.
Pension funds would, as under current law, also track nondeductible employee contributions, which are exempt from
tax when distributed.
The transition to the new regime should be straightforward. Pension funds would calculate the sum of all previous
contributions and investment earnings on the date of enactment of CBIT. Those earnings would go into the nonCBIT account,
and any future CBIT earnings would go into the CBIT account.
55. Special rules may be needed to limit the allocation of EOA balances to preferred stock upon liquidation. For example,
it may be inappropriate to allocate any EOA to preferred stock on which current, fully excludable dividends have been paid.
In that case, the liquidation proceeds simply represent a return of capital.
56. IRC § 732 prevents a step-up in basis, however, thereby preserving a potential tax whenever the distributee partner
disposes of the distributed asset.
57. Such exceptions might be patterned on existing IRC §§ 731-732 or prior IRC § 333, which was repealed in 1986.
58. See Treas. Reg. § 301.7701-2. In general, an organization that has associates and an objective to carry on business for
joint profit is classified as a corporation rather than a partnership if it has more corporate characteristics than noncorporate
characteristics. The corporate characteristics relevant to this determination are (1) continuity of life, (2) centralization of
management, (3) limited liability for debts, and (4) free transferability of interests.
59. IRC § 7704.
60. IRC § 851 et seq. A RIC also may retain and pay tax on long-term capital gains, in which case shareholders must include
such gains in their income and are credited with their share of corporate tax paid.
61. IRC § 856 et seq. REITs are allowed a dividends-paid deduction for distributions of both ordinary income and capital
gains income, but are not allowed to impute retained capital gain income to shareholders.
62. IRC § 860A et seq.
63. See IRC § 1381 et seq. which generally apply to cooperatives. See also IRC § 501(c)(12) (certain cooperative telephone
or electric companies); and IRe § 521 (farmers' cooperatives).
64. These changes also would apply to sole proprietorships not eligible for the small business exception.
65. IRC § 265(a)(4) should be expanded to cover CBIT investments of all three conduit entities. As discussed in the context
of rules for savings and loan associations under CBIT, policy makers could consider imposing a withholding tax of 31 percent
on distributions from RICs, REITs, and particularly REMICs to tax-exempt investors attributable to home mortgage interest
to prevent unfair competition between these entities and savings and loan associations.
66. The patronage dividend mechanism is sufficiently flexible that it should permit the cooperative to shift income attributable

Notes

212

to the disallowance of interest deductions to patrons. In effect, the cooperative could substitute a patronage deduction for
the interest deduction if patrons are generally in a tax bracket under 31 percent.
67. For example, consideration might be given to allowing banks to pay deductible (and includable) interest on a limited class
of deposits. The possibility of such an option for savings and loan associations is discussed in the text below.
68. Unlike the alternative approach, this rule would require a provision defining the institutions eligible for its special rule;
e.g., the special rule could apply to CBIT entities that earn at least 80 percent of their total income from interest and
dividends.
69. The potential problems could be exacerbated if losses arising from nonapplication of IRC § 265(a) to financial institution
operating expenses were allowed to generate net operating losses that could be used by other members of a consolidated
group.
70. S&Ls may well argue that such a provision is necessary to preserve parity with REMICs and other entities which we
recommend retain their conduit status. Since REMICs, for example, could market mortgage pass through instruments to taxexempt institutions without imposition of an entity level tax of 31 percent, REMICs would clearly have an advantage in
raising funds from the tax-exempt sector over S&Ls. As suggested earlier, an alternative solution to this result might be to
impose a 31 percent withholding tax on REMIC distributions to tax-exempt organizations or impose such a tax directly on
tax-exempt organizations receiving tax-exempt interest through a REMIC by treating such income as unrelated business
taxable income. Under current law, interest paid on REMIC regular interests is tax free to tax-exempt investors and, in
general, to foreign investors. A portion of the income on REMIC residual interests is subject to UBIT in the hands of taxexempt organizations and is subject to 30 percent withholding tax when distributions are made to foreigners.
71. Under current law, insurance companies generally include in gross income premiums and investment income and deduct
from gross income general business expenses and distributions to policyholders and beneficiaries. In addition, the companies
are allowed to deduct the net increase in the amount of insurance reserves during the taxable year. If reserves decrease, the
amount of the decrease is included in income. Over the life of any insurance policy, the net deduction for reserves is always
zero (since the reduction in reserves as claims or benefits are paid generates items of income that offset the earlier
deductions). Thus, the reserve deduction affects the timing of insurance company deductions for claims and benefits, but does
not increase the ultimate deductions to more than the amount of claims and benefits actually paid.
Tax reserves are calculated on a discounted basis to reflect the time value of money. The deduction for the net increase
in insurance reserves serves two purposes. First, it prevents that portion of premiums needed to fund future casualty or
benefit payments from being taxed. Second, it provides for a deduction equal to the expected investment return on reserve
funds. As a result of the combined deduction for reserves, claims and benefits, insurance companies are able to deduct
currently the present value of anticipated future payments, instead of deducting those payments when made. The difference
between the present value of future payments and nominal amount of those payments decreases over time, and each year a
deduction is allowed to the extent of the decrease during the taxable year.
Insurance companies also make dividend payments to policyholders. Policyholder dividends consist of various
components, one of which is an interest component. Dividends paid to policyholders are generally deductible from income
and, among other things, provide a mechanism for life insurance companies to adjust effectively the amount of the reserve
deduction for changes in the rate of investment return. Thus, the interest-like deduction available to insurance companies
under current law is spread among deductions for the change in reserves, for claims and benefits paid, and for policyholder
dividends paid. For a more complete discussion of the issues related to insurance company policyholder dividends, see U.S.
Department of the Treasury, Report to the Congress on Life Insurance Company Taxation (1989) and U.S. Department of
the Treasury, Report to Congress on Property and Casualty Insurance Taxation (1991).
72. CBIT would not alter current law rules which result in exclusion of much of the amount paid to policyholders in the form
of claims, benefits, or policy dividends. Under current law, virtually all death benefit distributions payable under life
insurance policies are fully excluded from gross income. Casualty claim payments are typically offset by loss (lRC § 165)
or rollover (lRC § 1033) deductions allowed to the recipient. However, some other insurance company distributions are
included in income. Business policyholders of casualty policies must generally include policyholder dividends in income,
because they generally may deduct the related premiums. Individuals receiving policyholder dividends from either P&C or
life policies or receiving policy surrender distributions from life policies generally are required to take those distributions
into income only to the extent that they exceed the total of previous premium payments less previous distributions. As a result
of these rules, very little of the investment income earned on cash value is included in taxable income at the individual level
under current law.

213

Notes

PARTID
Introduction

1. Under these conditions, any system of integration would result in the imposition of a single level of tax at a single tax
rate, regardless of whether corporate earnings were distributed or retained. For example, assume that a corporation earns
$100, and all corporate and individual income is taxed at a flat rate of 34 percent. ,Under the shareholder allocation
prototype, $100 of income would be imputed to the shareholder, who would pay $34 in tax. The tax due also would be $34
under any of the three distribution-related integration systems. In each system, the corporation would pay $34 of tax. Under
the dividend exclusion prototype the corporation could distribute its $66 of after-tax earnings tax-free to shareholders. Under
the imputation credit system discussed in Chapter 11, when earnings were distributed, the shareholder would have a $34
credit, which would exactly offset his tax liability. In a dividend deduction system, the corporation would have a $100
deduction that would offset its tax liability in the year of distribution, and the shareholder would pay tax of $34. Under
CBIT, the earnings would be subject to $34 of tax at the corporate level but would not be taxable upon distribution as interest
or dividends to investors.
2. The equivalency analysis set forth in the preceding note does not take into account the possible additional burden created
by taxing capital gains on corporate stock. See Chapter 8. Appendix C discusses the equivalence of distribution-related
integration systems.

Chapter 5
1. Although no agreement exists on the precise specification of the standard accounting rules, there is sufficient conformity
that most analysts are able to ascribe to an accepted list of preferential items. See, e. g., Budget of the United States
Government. Fiscal Year 1992, Ch. XI, "Tax Expenditures."
2. See IRC § 312. Because corporate shareholders generally claim a dividends received deduction for both regular tax (lRC
§ 243) and minimum tax (lRC § 56(g)(4)(c)(ii» purposes, preference income flows through to most corporate shareholders
under current law.
3. See McLure (1979), pp. 131-32, and Polito (1989), pp. 1036-37 (both arguing that corporate preferences should be passed
through to shareholders under a fully integrated tax system); and Kitchen (1987), p. 360 (defending the ability to pass
preferences through under Canada's integrated tax system).
4. Congress has at times indicated a willingness to discriminate between corporate and noncorporate preferences. For
example, IRC § 291 restricts the availability to corporations of certain preferences that are otherwise available to both
corporate and noncorporate entities alike. See also IRC § 56(b), which specifies several AMT adjustments that apply only
to taxpayers other than corporations, and IRC § 56(c) and (g), which specify adjustments that apply only to corporations.
5. See, e.g., the tax expenditure estimates presented in the Budget of the U.S. Government, cited in note 1. Although the
approximately $50 billion annual corporate tax expenditures noted in the 1992 Budget overstates the magnitude of revenue
cost (primarily because behavioral adjustments are not considered in the tax expenditure estimates) this figure serves to
illustrate the significant revenUe impact that would result from extending preferences to shareholders.
6. As discussed in Chapter 13, a complete analysis of the economic effects of the integration prototypes should include an
examination of the efficiency cost of the revenue offsets.
7. See Avi-Yonah (1990), pp. 199-202.
8. See Section 2.B. The same is true of an imputation credit system of distribution-related integration. Under such a system,
extending preferences to shareholders can result in shareholders receiving tax credits that exceed the corporate level taxes
paid. This occurs if the integration rules implicitly (and incorrectly) assume that the corporation has paid taxes on preferencerelated income, and if the corporation tax rate exceeds the individual tax rate. For example, such errors would occur if a
shareholder imputation credit method required that a shareholder compute his credit as a fixed percentage of dividends
received (if the percentage is based on the statutory rate of tax), gross up the dividend by the amount of the credit, apply
his tax rate to the grossed-up dividend, and apply the credit to the resulting tax liability. This procedure would extend
preferences to shareholders whenever the corporate and personal tax rates are equal, but it would provide greater subsidy
for preferences if the corporate tax rate exceeds the shareholder tax. rate.

Notes

214

9. If it were desired to extend some (but not all) preferences to shareholders, a distribution-related integration system could
be structured to accomplish this result. Preferences in the form of tax credits could be passed through simply by treating such
credits the same as taxes actually paid. The relative ease of passing such credits through in an integrated system should
encourage policy makers so to structure any tax preferences that it desired to pass through to shareholders. Exemption
preferences also could be passed through, but, in an imputation credit system, that would require additional accounts at the
corporate level and separate treatment at the shareholder level. Deferral preferences create the most substantial mechanical
problems if passed through to shareholders. See also Section 3.E.
10. A compensatory tax ensures that full corporate level tax has been paid on distributed income by assessing a "toll charge"
on the corporation with respect to each distribution of preference income. Section 11.B and Appendix C examine different
types of compensatory tax systems. To determine the amount of the toll charge, corporations would maintain an account of
corporate tax paid or of fully-taxed income to determine the amounts of fully-taxed and of preference income. A "stacking"
rule could then be applied to determine the extent to which distributed earnings were made from the corporation's fully-taXed
or preference income. The stacking rule most favorable to taxpayers is to treat corporate distributions as paid first from fullytaxed income and then from preference income. Thus, if the corporation has sufficient fully-taxed income to apply to
distributions, the corporation and its shareholders will suffer no adverse consequences from a decision not to extend
preferences to shareholders. Chapter 11 contains a discussion of stacking alternatives and their economic effects. The
principal alternative is a pro rata stacking rule, which would treat distributions as containing a proportionate share of the
corporation's retained preference income.

If the compensatory tax rate is set equal to the corporate tax rate, the effect is to recapture corporate tax preferences.
In that case, if a corporation distributes only fully-taxed income (determined under stacking rules), no additional tax liability
results. For distributions in excess of fully-taxed income, each dollar of tax-exempt preference income is subject to the full
corporate tax rate, and the full amount of tax paid is available as a shareholder credit. If the shareholder credit is fully
refundable, the tax system collects no additional net taxes from a compensatory tax. If the credit is not fully refundable, then
the tax system collects an additional tax on preferences distributed to shareholders who have insufficient tax liability to absorb
the credit or who are tax-exempt.
If the compensatory tax is set at a rate below the corporate tax rate, distributions in excess of fully-taxed income result
in additional corporate level tax liability on preference income, but at less than a dollar-for-dollar rate. This achieves a result
somewhat analogous to the current alternative minimum tax, because distributed preference income bears tax at a rate lower
than the corporate tax rate. Setting the compensatory tax at a rate lower than the corporate tax rate differs from an alternative
minimum tax: the compensatory tax is triggered only on distributions, while the current alternative minimum tax applies
regardless of whether funds are retained or distributed.
A third alternative sets the compensatory tax rate equal to the shareholder rate rather than the corporate rate. This
approach, adopted in the U. K. imputation system, effectively taxes the corporation at the shareholder rate on distributed
preference income and allows shareholders a credit at the same rate. For shareholders who pay tax at that shareholder rate,
the compensatory tax acts as a withholding tax on funds distributed to shareholders. If the shareholder credit is not refundable
and cannot be carried forward, the compensatory tax creates an additional tax burden on distributed preference income for
shareholders whose tax rate is less than the statutory rate. Only refundability of tax credits will eliminate such consequences
for tax-exempt shareholders.
Section 11.B examines the treatment of preference mcome distributed to tax-exempt shareholders under both a
compensatory tax and a credit limitation approach.
11. See Section 11. B for a discussion of the different methods for limiting the shareholder credit to corporate level tax
actually paid. This method requires the corporation to maintain an account of corporate taxes paid. In a dividend exclusion
system, the amount of taxes paid is converted into a corresponding amount of fully-taxed income. The account would be
increased by corporate tax paid and the amount of credits from dividends received from. other corporations and decreased
by the amount of credits attached to distributions made to shareholders (or the fully-taxed income equivalents). As with the
compensatory tax, a stacking rule is necessary to determine the extent to which distributions are made out of fully-taxed
income. Shareholder credits with respect to distributions would thus be allowed only to the extent the corporation's account
was sufficient to fund the credits. Distributions considered made out of preference income would not carry imputation credits
and, thus, would be subject to tax at the shareholder tax rate, as under present law.
12. See Section 11.B.
13. See Section 2. B. If integration were extended to retained earnings through a dividend reinvestment mechanism, a decision
by restricting
the dividend
not to extend corporate level tax preferences to shareholders could readily be implemented
/'
._--_.
-.

215

Notes

reinvestment option to fully-taxed retained earnings. This could be accomplished by limiting the dividend reinvestment option
to the balance in the corporation's EDA, in the dividend exclusion and CBIT prototypes, or the SCA, in the imputation credit
prototype. See Chapter 9.
14. See Section 4.D.
15. See Section 3.E.

Chapter 6
1. In some cases, the Code also permits deductibility of donors' contributions as charitable contributions (lRC § 170), while
contributions to pension funds are generally deductible as business expenses (lRC § 404).
2. This is true only when individuals' tax rates are constant over their working life and in retirement. If tax rates during
retirement are lower, current law treatment of pension savings is even more valuable.
3. Income from an exempt organization's investments in a publicly traded partnership is subject to UBIT, regardless of
whether the partnership's business is unrelated to the entity's exempt purpose.
4. As Chapter 5 notes, most preference items confer tax deferral rather than complete exemption. Corporate income sheltered
from tax by a deferral preference can be distributed to a tax-exempt shareholder without shareholder level tax, preserving
the value of tax deferral until the preference "turns around" and additional tax is imposed at the corporate level. Corporate
preference income distributed as interest to tax-exempt debtholders receives even more favorable treatment: not only is the
income exempt from tax at both the corporate and shareholder level, but the interest deduction may be available to offset
otherwise taxable income. This benefit is not available for all preference income. IRC § 265, for example, disallows
deductions for interest and other expenses attributable to tax-exempt bond interest.
5. In 1989, tax-exempt entities were allocated $1.6 billion in income from partnerships, or approximately 2 percent of the
total amount allocable to all partners. Of the tax-exempts' share, an estimated $260 million was trade or business income
that could have been subject to UBIT. The remainder consisted of rents, royalties, interest, dividends, and other forms of
income not subject to UBIT.
6. Depending on the integration system adopted, there could still be an advantage in distributing corporate preference income
to tax-exempt shareholders. For example, under a shareholder credit limitation system, preference income would be exempt
from tax at the corporate level and would be exempt from tax at the investor level if distributed to a tax-exempt shareholder.
Retained preference income, realized in the form of capital gains on stock, also would be exempt from tax in the hands of
a tax-exempt shareholder. A compensatory tax, discussed in Section 11.B, would impose a corporate level tax on distributed
preference income, but would not change the treatment of retained preference income.
7. A dividend exclusion system would not provide equivalent treatment of debt and equity held by tax-exempt investors unless
interest also were nondeductible at the corporate level and excludable by the recipient. This regime is CBIT; see Chapter 4.
8. See Sections I1.E and 12.A, respectively. A dividend deduction system without withholding would equalize the treatment
of debt and equity investments by tax-exempt investors. Corporations would be able to deduct dividends paid, as they now
deduct interest, and neither type of income would be taxable to the tax-exempt investor. This result could be changed by
denying the deduction (or the benefit of the zero rate) for dividends paid to such tax-exempt shareholders, but such an
approach would require corporations to track the identities and tax status of shareholders. Coupling a nonrefundable
"withholding" tax with a dividend deduction could achieve results similar to a nonrefundable credit under an imputation credit
method of integration.
9. The United Kingdom refunds the imputation credit to tax-exempt investors. However, while the U.K. 's imputation credit
is fully refundable to all domestic shareholders, including tax-exempt shareholders, the U. K. has a partial distribution-related
integration system, so earnings distributed to a tax-exempt shareholder still bear a tax equal to the excess of the corporate
rate over the credit rate. See Appendix B. Tax-exempt organizations own approximately 40 percent of the outstanding stock
of U.K. companies.
10. An effort to provide tax-free treatment for corporate income allocated to tax-exempt or tax-favored investors under CBIT
would raise major problems. For income distributed in the form of interest and dividends, the relative advantage of such
investors could be maintained by providing refunds of corporate tax paid with respect to funds distributed. For undistributed

Notes

216

income, however, eliminating the corporate level tax would require allocating undistributed income to the shareholdersexactly the type of administrative complexity that occurs under a shareholder allocation system and that the CBIT approach
to integration seeks to avoid.
11. A dividend deduction proposal passed by the House of Representatives in 1985 would have made a portion of dividends
received by certain tax-exempt organizations subject to UBIT. See H.R. 3838, 99th Cong., 1st Session, § 311 (1985) and
H. Rept. No. 426, 99th Cong., 1st Sess. (1985), p.240.
12. For example, under an imputation credit system of distribution-related integration, providing full shareholder imputation
credits on dividend income to tax-exempt investors would allow them to invest in a mix of equity and debt so the credits
could be used to offset the tax on other investment income. This approach is similar to Australia's system for tax-exempt
investors, adopted shortly after enactment of a shareholder credit limitation integration system. Allowing the credit to offset
other investment income also discourages streaming of franked dividends to taxable investors and unfranked dividends to taxexempt investors.
For example, assume a tax-exempt entity earns $100, of which $25 is dividend income and $75 is interest income.
Assume, in addition, that the dividend carries an imputation credit for corporate tax paid at a 31 percent rate and that the
tax-exempt entity is subject to tax on all investment income at a 12 percent rate. The net dividend of $25 would be treated
as a gross dividend of $36.23, with a tax credit of $11.23. The tax-exempt entity would have a tax liability (before credits)
of $13.35 (.12XII1.23), which would be offset in part by the $11.23 credit. The net tax due would be $2.12.
13. If credits are nonrefundable, the revenue neutral rates are as follows: 8.4 percent for shareholder allocation, 7.6 percent
for the imputation credit prototype, 7.2 percent for CBIT with no taxation of capital gains, and 6.1 percent for CBIT with
current law capital gains taxation.

Chapter 7
I. Unlike many other countries, the United States also taxes the worldwide income of all U.S. citizens and U.S. corporations,
whether or not they are residents of the United States.
2. Some or all of the U.S. shareholders of a foreign corporation may, however, be subject to current U.S. tax on all or a
portion of the corporation's income if it earns income which is either passive, e. g., interest, dividends, royalties, and similar
income or particularly mobile or holds assets that produce such income. See, e.g., IRC §§ 951, 1293.
3. Thus, for example, if a foreign subsidiary of a U.S. company earns $100 abroad, pays $40 in foreign corporate level
taxes, and remits $27 in dividends to its U.S. parent ($30, net of a $3 withholding tax imposed by the foreign country), the
parent must report $50 in foreign source dividend income ($27 plus $3 plus 50 percent of $40), and can claim a credit
(subject to the appropriate limitations) for direct foreign taxes of $3 and indirect foreign taxes of $20.
4. Merely acquiring U. S. stock and debt securities does not constitute aU. S. trade or business.
5. See, e.g., Bergsten, Horst, and Moran (1978) and Caves (1983). In the public economics literature, studies by Musgrave
(1969), Horst (1980), and Giovannini (1989) have attempted to compare the relative efficiency of capital export and capital
import neutrality under various stylized assumptions. See also the overview in JCS-6-91 (1991).
6. See "Savings and Investment" in Section I.B.
7. Setting tax rates independently implies that countries take policies of their trading partners as given, and misestimate
effects of their own policies. See, e.g., Gordon (1983). In particular, analyses of capital export neutrality often assume that
foreign countries' tax rates are independent of the resident country's tax rates. The source country may, of course, take into
account that most investment from abroad originates from countries that grant a worldwide credit for foreign taxes paid. The
source country may, therefore, be able to increase taxes on foreign investment without reducing capital inflows because
foreign governments, not investors, would absorb the tax. In effect, a policy of capital export neutrality may lead to a
transfer from the resident country's treasury to that of the source country.
8. The foreign tax credit tends to promote capital export neutrality, because it eliminates an investor's U.S. tax liability to
the extent of foreign taxes paid, but requires the investor to pay a residual U.S. tax if the U.S. tax rate is higher than the
foreign tax rate. In this situation, the investor is neutral between domestic and foreign investment, because the investor bears

217

Notes

the same tax burden in either case. For additional discussion, see Hines and Hubbard.(1990) and JCS-6-91 (1991). As
explained in the text, however, the foreign tax credit does not always have this effect.
9. The indirect credit thus provides equal treatment for foreign direct investment by U.S. corporations, whether through a
foreign subsidiary or a foreign branch operation.
10. This conclusion turns on accepting, as we do in Chapter 13, the traditional view of dividends. See Section 13.B. For
additional discussion of these issues, see Hines and Hubbard (1990) and the studies cited therein.
11. The statutory exemption for portfolio interest reflects the difficulty of taxing highly mobile debt capital. The exemption
for capital gains represents an incentive to foreign persons to invest in U. S. capital markets and a concession to the
administrative difficulties of determining gain and collecting tax where the income is not physically paid from U.S. sources.
12. Treaties also suggest another explanation for the nondiscrimination rule--to protect the bargain agreed to by the parties.
Treaties limit withholding rates but generally do not impose direct limitations on a source country's right to tax business
profits. This creates some risk that the source country may alter the bargain, without directly affecting withholding rates,
by changing the way that business profits are taxed to foreign investors. The nondiscrimination rule indirectly prevents this
by requiring that changes in the taxation of business profits burden domestic and foreign capital equally.
13. The shareholder allocation prototype treats foreign taxes by statute like U.S. taxes, but we do not recommend adoption
of that prototype.
14. The following examples illustrate the tension between a policy of avoiding additional taxation of foreign source profits
and a policy of collecting one level of U.S. tax on profits from all sources. Assume that a U.S. individual owns 100 percent
of a domestic corporation that in turns owns 100 percent of a foreign corporation. The V.S. corporate rate is 34 percent,
the individual rate is 31 percent, and the United States has adopted a dividend exclusion system. The foreign corporation
earns $100 of foreign profits in the relevant taxable year and pays foreign taxes of $25. The foreign subsidiary later
distributes the after-tax income to its domestic parent, which distributes the dividend (net of any U.S. tax) to its sole
individual shareholder. If the domestic parent is required to include $100 of profits in income for the taxable year of the
distribution but is given a tax credit of $25 against its U. S. tax liability, and the individual is allowed to exclude the dividend
altogether, then the aggregate level of tax of the foreign profits will be no greater than if the profits were from domestic
sources. No additional taxation will exist. Compared to current law, exempting the dividend in the hands of the individual
shareholder will significantly reduce the United States' portion of the aggregate tax burden borne by the foreign profits. The
United States' portion of the total tax paid will only be $9 out of $34, or 26 percent of the total, compared to the United
States' portion under current law: $29 out of $54, or 54 percent of the total.

If, in contrast, the tax regime provides a credit for the $25 of foreign taxes paid by the subsidiary to the domestic parent
but requires the individual shareholder to pay tax upon the appropriate portion of the subsequent distribution by the parent
under the dividend exclusion prototype, then the foreign profits will bear an additional amount of tax relative to a similar
amount of domestic profits. The domestic corporation will owe $9 of additional tax upon receipt of the distribution from the
foreign shareholder, and the individual shareholder will owe a tax of $15 upon the subsequent distribution of a grossed-up
dividend of $49. The foreign profits will have been subject to aggregate foreign and U.S. taxation of $49, in comparison
with aggregate taxation of $34 for similar profits from domestic sources. Under this approach, the United States' portion
of the total taxes paid for such income will be $24 out of $49, or 49 percent of the total. However, the total tax burden on
the earnings decreases to $49 from current law's $54, because there is only one level of U.S. residual tax.
15. This problem would be even more severe if shareholder credits in a shareholder allocation or imputation credit system
were actually refundable, rather than simply available to offset tax liability on other income.
16. See Sections 2.C and 11.D.
17. See Section 4.D.
18. See Section 3.1.
19. For domestic corporations owned by foreign shareholders, the first level of tax is the normal domestic corporate tax and
the second level is the 30 percent withholding tax on dividends. For a U.S. branch of a foreign corporation, the first level
is the corporate tax on the branch's U.S. business income and the second level is the branch profits tax under IRC § 884(a).

Notes

218

20. Other countries with integrated tax systems, as a rule, have not extended benefits of integration to U.S. shareholders
except as a result of tax treaties. However, the U.S. treaties with the U.K., Germany, and France extend some benefits of
integration to U. S. shareholders in certain cases. On the other hand, Australia generally extends the benefits of integration
to foreign shareholders by statute. See Appendix B.
21. The following example illustrates the problem in the context of an imputation credit system that refunds imputation
credits to foreign shareholders. The issues would be the same in a dividend exclusion system that refunded corporate tax to
foreign shareholders. Assume, for example, that two domestic corporations each earn an annual pre-tax profit of $100.
Corporation A has one shareholder, a U.S. resident individual. Corporation B also has one shareholder, a nonresident alien
individual who resides in a country that has a tax treaty with the United States. The tax treaty limits the U.S. dividend
withholding rate to 15 percent for portfolio investors (including the shareholder of corporation B) and contains a standard
prohibition against discrimination based on capital ownership. Assume also a 34 percent corporate tax rate, a 31 percent
individual tax rate and that corporate taxes are credited to shareholders at the 31 percent individual rate.

Jf neither corporation distributes earnings, each pays a tax of $34 on its $100 profit. No discrimination exists between
the two corporations, and the withholding rules are not implicated. If, instead, each corporation distributes one-half of profits,
the domestic shareholder receives a cash distribution of $33, an imputation credit of $14.83, and a grossed-up dividend, i.e.,
including credit of $47.83. See Section 11.B. The domestic shareholder will have a tax liability with respect to the gross
distribution of $14.83, which will be exactly offset by the imputation credit. Thus, for corporation A both distributed and
retained earnings are taxed at a 34 percent rate.
There is a significantly different result for corporation B. The foreign shareholder receives a cash dividend of $33. If
he also receives an imputation credit of $14.83, his gross dividend will be $47.83. The withholding tax on this distribution
will be $7.17, entitling him to a refund of $7.66. In this case, undistributed profits are taxed at 34 percent, but distributed
profits are taxed at 18.7 percent ($50 of pre-tax income that bears $17-$7.66 of tax).
22. In the past, countries with nonintegrated tax systems, including the United States, have responded that this argument is
highly stylized, that it ignores the economic reality that profits distributed to foreign shareholders bear a higher level of tax
than profits distributed to domestic shareholders, and that such an integration regime is discriminatory. As noted in the text,
this response has generally been rejected by countries with integrated systems, although the United States has had some
success in negotiating partial integration benefits for its shareholders.
23. See Section 2.A.
24. This would not be true in an integration system that imposed both a nonrefundable compensatory tax and a withholding
tax on dividends. A nonrefundable compensatory tax combined with a withholding tax would subject distributed preference
income to two levels of tax, rather than the one level of tax imposed under current law. (Note that, if a compensatory tax
were adopted in CBIT, the current withholding tax on dividends would be repealed.) See Section 4.E.
25. See Section 3.I.
26. See Section 6.D, which describes such an approach for tax-exempt entities. Such an approach would minimize portfolio
shifts by foreign shareholders and would provide an opportunity for achieving greater parity between debt and equity
investments in U.S. corporations by foreign investors.

Chapter 8
1. Presumably, if shareholders were not taxed on gains, they would not be allowed losses on stock sales.
2. As described in Section I3.B, we accept the traditional view of dividends, under which the value of $1 of retained
earnings is $1 as long as the managements of corporations maximize firm value. Under the new view, also described in that
section, distributions to shareholders in the form of dividends are unavoidable. For a dividend paying corporation in this
view, an incremental dollar of retained earnings raises share value by less than $1.
3. The value of stock in a corporation that has retained earnings may include the value to a prospective purchaser of the
resulting capital loss that will be realized when the stock is resold after the earnings are distributed, although the value of
this loss to a purchaser depends on the purchaser's marginal tax rate and ability to use capital losses, and the amount of time
the purchaser expects to elapse before the earnings are distributed and it dispose of the stock.

219

Notes

Assume, for example, that a dividend exclusion system is adopted and that the corporate and shareholder tax rates both
are 34 percent. A corporation earns $100 of fully-taxed income in year one and pays $34 in tax, so it has retained earnings
of $66 and an EDA balance of $66. How much should a prospective purchaser pay for all the stock? The answer is that the
purchase price of the stock will vary between $66 and $100, depending on the tax attributes of the purchaser and the expected
timing of the distribution of the $66 of retained earnings and the purchaser's resale of the stock.
The after-tax value of the retained earnings to any purchaser is $66. In addition, if the corporation distributes all of its
earnings, the shareholder will realize a capital loss upon disposition equal to the amount paid for the stock. (The amount
realized on the disposition would be zero, assuming the corporation has no assets after the distribution.) In theory, the value
of the capital loss may be as great as $34 (and thus, a purchaser would be willing to pay $100) if: (1) the distribution of the
earnings and the disposition of the stock are expected to occur very shortly after the purchase of the stock, (2) the purchaser
expects to have sufficient capital gains against which to use the capital loss, (3) the purchaser expects to face a 34 percent
marginal tax rate, and (4) the distribution does not reduce the basis of the shares.
The value of the capital loss may be much less. The value of the capital loss will be less if the shareholder does not
dispose of the stock immediately, cannot use the capital loss immediately, or is subject to tax at a marginal rate of less than
34 percent. If, for example, the capital loss is worth zero, the purchaser would pay only $66 for the stock.
4. Depending on marginal tax rates, the tax system may collect as little as no tax or as much as two full levels of tax on
corporate earnings. If the corporate tax rate does not exceed the individual rate, the tax system may collect virtually no tax
on corporate earnings if, for example, a seller of stock is tax-exempt and a purchaser is taxable. In that case, the seller will
not pay tax on capital gains attributable to fully-taxed retained earnings, but, after the earnings are distributed, the taxable
purchaser can sell his stock and realize a capital loss. That loss may be valuable enough to offset tax collected on the
earnings at the corporate level. On the other hand, the tax system may collect two full levels of tax if, for example, a seller
of stock is taxable and a purchaser is tax-exempt. In that case, the initial shareholder's capital gain is taxed in full, but the
offsetting capital loss creates no tax benefit to the purchaser. Current law in some cases limits the availability of a capital
loss following a distribution. See, e.g., IRC § 1059 (basis reduction for extraordinary dividends).
5. The analysis in the text oversimplifies this issue to illustrate the general point. The analysis can be complicated if
preferences are subsequently distributed or if the preference is a deferral or tax credit rather than an exclusion of income.
6. This could be accomplished by increasing inside basis in a manner similar to the treatment of electing partnerships under
IRC § 754 and electing purchasers of corporate stock under IRC § 338. Applying such a rule to small acquisitions of stock
(particularly where there is frequent public trading) would be administratively impossible; however, using a dividend
reinvestment plan could provide some relief. See Chapter 9.
7. Halperin and Steuerle (1988) indicate that total capital gains in the economy over time are approximately equal to gains
attributable to inflation plus retained earnings. Their research indicates that the real gains in value in one sector, e.g., land
in the 1970s, tend to be offset by real losses in another sector, e.g., corporate stock in the 1970s. According to Halperin
and Steuerle, from 1948 to 1985 the total change in economywide net worth equals the sum of (1) average net investment
of 12.3 percent of net national product (NNP), (2) average inflationary gains in value of 16.1 percent of NNP, and (3)
average real gains in value of -2.6 percent of NNP. See also Steuerle (1991). If total capital gains are attributable only to
inflationary increase in asset values and retained earnings, the case for reduced taxation of nominal capital gains on corporate
stock is much stronger.
8. See IRC §§ 705 and 1367. Treas. Reg. § 1.1502-32 provides a comprehensive set of basis adjustments for C corporations
that are members of a consolidated group.
9. In cases where expected increases in future earnings that are reflected in the price of equity never materialize, an equity
holder may realize a gain that never creates a corresponding amount of income to be taxed under CBIT at the entity level.
In that case, however, the purchaser of the equity interest will realize a corresponding loss, and disallowing both the gain
and the loss achieves a roughly accurate solution.
Example. A purchases Corp. X stock for $100, when Corp. X is expected to earn $1,000 per year. One year later,
Corp. X announces a new product line that is expected to increase its earnings to $1,500 per year. A sells his stock
to B for $150. Six months later, one of Corp. X's competitors introduces a superior product. Corp. X's expected
future earnings decline to $1,000 per year. B then sells his stock for $100.
Without taking into account the time value of money, the marginal tax rates of the two investors, or capital loss limitations,
A's $50 gain is offset by B's $50 loss.

Notes

220

10. A complete exemption also may create an incentive to restructure transactions. For example, because investor level gains
on a sale of stock would be exempt but entity level gains on a sale of assets would not, there would be a considerable
incentive to structure acquisitions of corporations with appreciated assets as stock sales rather than asset sales. This is similar
to the bias that exists under current law, under which sales of stock result in only one level of tax, while sales of assets,
which typically either are preceded by a liquidating distribution of assets or followed by a liquidating distribution of sales
proceeds, generally result in two levels of tax.
11. Proposals made in other contexts, e.g., a mandatory IRC § 338 election, might be considered. Current law permits
certain purchasers of 80 percent or more of a corporation's stock to elect to treat a stock purchase as an asset purchase. A
mandatory IRC § 338 election, adapted for CBIT, would require recognition of gain at the entity level if a certain percentage
of the equity of a CBIT entity changes hands. A mandatory IRC § 338 election may be more palatable in an integrated
system than under current law, because any gain realized would be subject to only one level of tax. Gain would be taxed
solely at the entity level, and no additional investor level tax would be due.
Another possible approach would tax capital gains realized on the sale by a CBIT entity of its equity interest in another
CBIT entity, e.g., a corporation'S sale of the stock of a subsidiary. For the reasons discussed above, taxing capital gains on
CBIT equity realized by a CBIT entity would tend to impose a second level of tax on earnings. Taxing entity level capital
gains on CBIT equity also would create disparities between equity investments held directly by individuals and those held
through other entities, e.g., affiliated groups of corporations. On the other hand, extending the exemption for capital gains
on CBIT equity would multiply the potential for deferral of entity level tax. Without special rules limiting tax-free
contributions of assets to subsidiaries or partnerships, CBIT entities would be able to structure some sales of assets as sales
of CBIT equity.
12. Auerbach (1990) discusses alternative means of retrospective capital gains taxation that approximate accrual-equivalent
capital gains taxation.
13. The text focuses on the different sources of capital gains for traditional forms of equity and debt. The sources of capital
gains for hybrid instruments may reflect both equity-type and debt-type gains. For example, fixed rate, nonconvertible,
cumulative preferred stock of a creditworthy company may react to interest rate changes in much the same way as debt.
14. The credit quality of debt may change because of changes in the underlying value of the firm. For example, debt issued
by a manufacturing firm might rise in value because the demand for the firm's product rises unexpectedly, thereby increasing
the likelihood that the firm will payoff the debt in a timely manner. In essence, the debt is more valuable because the finn
has become more valuable. The rise in value represents a capital gain to the debtholder. Such a gain is analogous to the gain
an equity holder would realize from the same event, and the deferral concerns are the same.
15. An unexpected fall in the market interest rate, for example, could generate a capital gain to the holder of long-term, fixed
rate, noncallable debt. The value of the debt would rise until the debt's interest payments would provide a new investor with
a return equal to the market interest rate.
Example. A noncallable perpetuity is a debt instrument that never matures. If the interest rate at issuance is 10
percent, a $100 perpetuity would pay $10 of interest per year. If the market rate of interest drops unexpectedly to
5 percent, the value of the perpetuity would double to $200, so its $10 annual interest payment would represent a
5 percent rate of return on the value of the debt. If the debt holder sold the perpetuity, he would realize a capital
gain equal to the $100 increase in value.
The effect of changes in interest rates is less pronounced for short-term bonds because there is a shorter period over
which off-market interest payments will be received and because the present value of the prepayment of principal is a more
significant component of price. For example, if the bond in the example above were scheduled to mature in one year, an
unexpected drop in interest rates would cause the bond to increase in value only to $104.76 ($110/1.05), rather than to $200
as with the perpetuity. However, a change in market interest rates creates an equal and offsetting gain or loss to the
borrower. A decline in the market interest rate increases the amount the borrower must pay to eliminate his debt. If the
borrower repurchased the debt in the example for $200, he would recognize the loss in the form of a $100 deduction. See
Treas. Reg. § I. 163-4(c). If market interest rates increased, the borrower could repurchase his debt for less than its issue
price and would realize income from the cancellation of indebtedness. See Treas. Reg. § 1.61-12(c).
Interest rate changes also can affect the value of equity. For example, an increase in interest rates may decrease the value
of common stock to the extent that stock price reflects the discounted present value of future cash flows on the stock because
the higher interest rate also will decrease the discounted present value of future cash flows from corporate assets. An increase

221

Notes

in interest rates also may create an offsetting increase in the value of common stock if a corporation has outstanding low-rate
noncallable debt.
16. Thus, if CBIT included a compensatory tax and gains on CBIT equity were exempt, considerations of simplicity may
support exempting gains and denying losses on CBIT debt (to both borrowers and lenders) as well. Although gains and losses
on debt that are attributable to changes in interest rates represent real accretions to wealth (or real reductions in wealth) to
borrowers and lenders, distinguishing between gains and losses on debt arising from changes in the value of the firm and
those arising from changes in interest rates would be virtually impossible. Further, a change in interest rates creates no net
gain in the tax system, because the lender's gain or loss is offset by the borrower's loss or gain. To the extent that debt
holders and equity holders face the same tax rate and would pursue the same realization strategy, the Treasury would collect
the same tax revenue if such gains and losses were included in taxable income as it would if such gains and losses were
ignored. This conclusion is weakened if differences in tax rates and differences in the timing of realization are taken into
account. Excluding all gains and losses on debt could create a net loss of tax revenue to the system in some cases, e.g., if
interest rates increase and the lender is tax-exempt and the borrower is taxable. Strengthening the case for exempting such
gains and losses is the observation that they are most important for long-term, fixed rate debt with call restrictions. Longterm, fixed rate debt has become less important in recent years. For nonflllancial corporations, the ratio of long-term debt
(corporate bonds, mortgages, and tax-exempt bond) to total credit market debt has fallen from 71.6 percent in 1962 to 56.4
percent in 1990. See Flow of Funds Accounts (1991). To the extent that even long-term debt has more flexible interest rate
adjustment than in the past, long-term fixed rate debt is even less important than the above calculation would suggest.
17. See IRC § 302. A redemption also may qualify for sale treatment if it terminates a shareholder's interest in the
corporation or is made to a noncorporate shareholder in a partial liquidation.
18. The analysis in the text generally applies to individual shareholders. Corporate shareholders, which are entitled to a
dividends received deduction (DRD), may favor dividends over share repurchases even under current law. A corporation
entitled to a 100 percent DRD would always prefer a dividend, which would be entirely tax-free and would preserve share
basis to offset later gains. A corporation entitled to a 70 or 80 percent DRD might prefer dividends in some cases.
The problems raised by share repurchases under the classical system are discussed at length in the American Law
Institute (1989), which recommends adopting "a minimum tax on distributions" of 28 percent (the maximum rate applicable
to individual taxpayers at the time) on the gross amount of any nondividend distribution to ensure that the second level of
tax is collected. See Section 12.C.
19. Thus, a shareholder with a basis of $150 in his stock would pay the same amount of tax on a $200 distribution and a
$200 payment in full redemption of his stock. In each case, the $200 payment would result in $50 of capital gain.
The rules determining stock basis should be reexamined under shareholder allocation. Although each share of stock has
traditionally been viewed as having a separate basis, an aggregate basis approach may be more suitable under shareholder
allocation, as under the partnership rules. For example, if aggregation is not permitted and a shareholder holds both low basis
shares and high basis shares, a pro rata distribution might result in recognition of gain on the low basis shares, while an
equivalent amount paid in full redemption of only a portion of the stock might be tax-free because the shareholder could
choose to surrender only high basis shares.
20. A DRIP would reduce the bias against share repurchases out of taxable income. DRIPs are discussed in Chapter 9.
21. Some have contended that the best approach would recharacterize a share repurchase as a pro rata dividend, followed
by sales of shares among shareholders to reflect the fact that, after a share repurchase, some shareholders have cash and
others have an increased proportionate interest in the corporation. All shareholders wo'uld pay tax on ordinary dividend
income and would add that amount to share basis. Selling shareholders would recognize gain or loss measured by the
difference between the amount realized on the sale and their basis in the shares. See Chi rei stein (1969). Abandoning the
realization requirement to tax nontendering shareholders would create additional complexity and administrative difficulties.
Indeed, since integration reduces the tax incentives for share repurchases over dividends in comparison to current law, a
change in that policy does not seem appropriate or necessary. Moreover, allocating the EDA balance among all shareholders
would require income allocations as complex as those required in the shareholder allocation prototype. See Chapter 3.
22. Attempting to treat third-party sales of shares as dividends that would be excludable to the extent of the issuing
corporation's EDA balance would entail information reporting (by brokers to the issuing corporations and by issuing
corporations to selling shareholders and the IRS) to an unprecedented degree. Such a system would be highly impractical
and undesirable.

222

Notes

23. Rules similar to those in IRC § 302 would be retained. Because corporations, for example, may have an incentive to
use redemptions of tax-exempt shareholders' stock in a dividend exclusion system, it might become necessary to reduce EDA
balances in proportion to shares redeemed.
24.

Example. A corporation owns an asset worth $100 and its sole shareholder has a basis of $100 in her stock.
The value of the asset declines to $60, and the shareholder sells her stock for $60, realizing a $40 capital
loss. If the corporation then sells the asset for $60, it too will realize a capital loss.

A shareholder level loss that mirrors an unused net operating loss at the corporate level is similar to a shareholder level
loss attributable to unrealized depreciation.
Example. The facts are the same as in the preceding example, except that the corporation sells the asset before the
shareholder sells her stock. The corporation has no taxable income (and no EDA balance), so that the $40 loss
represents an NOL carryforward available to offset future income. The shareholder sells her stock for $60 and
realizes a $40 capital loss.
25. Under current law, capital losses of individuals are allowed only to the extent of capital gains plus $3,000 of ordinary
income. See IRC § 1211(b). It would be possible to allow capital losses on corporate stock only to offset capital gains on
corporate stock (plus $3,000 of ordinary income) and generally match loss and gain duplication to reduce loss duplication.
See also IRC §§ 269 and 382-84; Treas. Reg. §§ 1.1502-21 and -22.

Chapter 9
1. A system of basis adjustments for retained earnings is inherent in the shareholder allocation prototype. See Chapter 3.
A DRIP also may be appropriate in the imputation credit prototype described in Chapter 11. Section 11.1 discusses special
considerations in adopting a DRIP in the imputation credit prototype. A DRIP would be unnecessary under CBIT if gains
and losses are not taxed to investors, because basis in such investments would be irrelevant.
2. This would not be true in the case of a dividend deduction system, discussed in Chapter 12. Under such a system, deemed
dividends would be taxable to shareholders but would give rise to a corporate level deduction. Thus, at minimum, a DRIP
in a dividend deduction system would require shareholder consent, as under current law. While we do not address the issue
further, we question whether a DRIP should be allowed in a dividend deduction system. Rate arbitrage might occur if a
corporation and its shareholders can elect a current corporate level deduction in return for a shareholder level tax.
3. For example, under the dividend exclusion prototype, a shareholder must meet a 45 day holding period in order to exclude
dividends received. See Section 2.B.
4. For example, dividend stripping generally results in basis reduction under current law, and the same rules may be
appropriate in the context of a DRIP. Basis allocation rules also might be used.
Example. The facts are the same as in Example 1, except that the fair market value of X shares at the time of the
DRIP distribution is $10 per share. Under these circumstances, the basis of both Lot A and Lot B shares will exceed
fair market value under either allocation method. In these circumstances, basis sufficient to bring the basis of all
shares up to fair market value should be so allocated. The balance should be allocated to all shares, pro rata.
5. The EDA would continue to be available to pay excludable dividends (or interest, in CBIT) on any class of stock (or debt,
in CBIT). In theory, it would be possible to maintain a separate EDA, as well as a deemed dividend account, for each class
of stock. However, such an approach would require unacceptably complex allocations of the EDA among classes of stock,
similar to ihe allocations of corporate income required under the shareholder allocation prototype. See Chapter 3.
6. We rejected three alternative rules. First, the stacking rule could treat cash distributions first as a return of capital to the
extent of previous deemed dividends. The rule recommended in the text is more favorable than this rule for any corporation
with a remaining EDA balance, because shareholders would generally prefer excludable dividends to basis reduction. Second,
the stacking rule could follow current law and treat cash distributions as a return of capital only after a corporation's earning
and profits are exhausted. Deemed dividends that had been declared would reduce earnings and profits by the amount of the
deemed dividend and cash distributions would be tax free to the shareholder to the extent treated as payments out of the
remaining EDA. This rule would be consistent with the current treatment of corporate dividends and with the notion that
shareholders recover capital only after recovering earnings. Under this rule, however, a corporation that had used the DRIP
to eliminate its EDA balance but had additional earnings and profits attributable to retained preference income would be

223

Notes

required to pay taxable dividends before it could treat distributions as a return of capital. While corporate shareholders
entitled to a DRD might prefer taxable dividends to basis reduction, we believe that the rule in the text is more favorable
to taxpayers in most cases. Finally, cash distributions might be treated entirely as dividends and no earnings and profits
account or account of deemed dividends would be kept. The advantage of the third alternative is that corporations would not
need to keep an account of deemed dividends. This approach, however, may discourage use of DRIPs.
7. We would not permit DRIPs for debt in CHIT because interest is generally paid in cash as it accrues. As Section 4.G
discusses, CBIT would generally retain OlD or imputed interest rules to distinguish payments of interest from payments of
principal. CBIT would not, however, retain the current rules governing the timing of imputed interest income.
This approach raises the question of how accrual, e.g., zero-coupon, and payment-in-kind bonds would be treated.
Consideration should be given to adopting rules that would prevent accrued discount (which, like interest, is not taxable to
a debtholder when received) from being taxed as capital gain if the debt instrument is sold before the discount is paid. One
approach would be to maintain the current OlD timing rules. Accrued discount would increase a debtholder's basis (but
would not be includable in income) and would decrease the issuer's EDA (but would not be deductible). Similar issues are
presented by discount preferred stock. See IRC § 305(c).

8. Mechanically, a mandatory DRIP would operate like the elective DRIP, except that a corporation would be required to
reduce its EDA to zero at the end of each year through deemed or actual distributions. A mandatory DRIP might cause
restrictions on the forms of equity eligible for DRIP distributions to be more desirable.

Chapter 10
1. Auerbach (1990) presents an overview of issues relating to gains and losses during the transition to integration.

2. As indicated in Chapter 13, we believe the best empirical evidence supports the traditional view of dividends, which holds
that the existing two-tier corporate tax has not been fully capitalized into share values. Accordingly, we believe that
integration may create some transition gains to owners of corporate stock but that such gains will not be as great as those
anticipated by advocates of the new view.
3. The second and third transition concerns described in the text are sometimes referred to as carryover problems.
4. See Graetz (1977).
5. See Section 2.B and Section 4.D, respectively.
6. The stacking order rules for distributions from the EDA (see Sections 2.B and 4.D) may prolong the deferral of the tax
on the retained earnings, however.
7. The American Law Institute Reporter's Study Draft (1989) on corporate tax reform contains a deduction for dividends
paid that would apply only to new equity. The proposals avoid the complexity of tracking new and old equity instruments
by limiting the deduction to the product of a specified rate and capital contributed after the date of enactment of the
proposals, less extraordinary distributions. American Law Institute (1989). See Section 12.C.
8. The current rules governing the conversion of a C corporation, i.e., a corporation taxed under the classical system, to
one of the various passthrough entities suggest the difficulties and complexities that would be involved in attempting to isolate
old equity from new equity. These rules, which include the rules that apply to C corporations that convert to a partnership,
an S corporation, or a RIC or REIT are concerned in varying degrees with preventing corporate income attributable to
preconversion years from escaping the two-tier tax. None provides a particularly satisfying approach to dealing with the
transition to an integrated corporate system.
For example, an approach modeled on the existing rules for taxing C corporations that convert to partnerships would
treat the corporation as though it had distributed all its assets to its shareholders in a liquidating distribution in with built-in
gain or loss with respect to the assets is realized at the corporate level and built-in gain or loss with respect to the stock is
realized at the shareholder level. The shareholders would then be treated as recontributing the assets to the corporation. This
mark-to-market approach would tax all the built-in gain or loss with respect to assets at the corporate level and all the built-in
gain or loss with respect to stock at the investor level. (Alternatively, an approach modeled on the existing rules for taxing
C corporations that convert to passthrough status as a RIC or REIT would confine the mark-to-market approach to the
corporate level, with shareholders taldng a carryover basis in their stock. See Notice 88-19, 1988-1 C.B. 486.) Although

Notes

224

the mark -to-market approach would eliminate any long-range transition effects from the change to an integrated corporate
system. the substantial and immediate tax cost, together with the administrative burden that would ensue from the need to
value all corporate assets, makes this approach unacceptable.
A transitional approach also could be modeled on the existing rules for taxing C corporations that convert to S
corporation status. Current law does not treat the conversion as a taxable event. However, S corporation shareholders are
taxable on distributions from earnings and profits accumulated in C corporation years to the extent the S corporation's
distributions exceed its cumulative taxable income. IRC § 1368. In addition, IRC § 1374 provides that if the S corporation
recognizes gain on an asset held at the time of the conversion within a 10 year period following the conversion, the gain is
subject to a corporate level tax. The total amount of gain subject to corporate level tax cannot exceed the net built-in gain
inherent in the corporation's assets at the time of the conversion. IRC § 1374(c)(2). Certain items of income and deduction
that are attributable to periods before the conversion but have not yet been recognized are taken into account in computing
the corporation's built-in gain. IRC § 137 4( d)(5). This approach avoids the immediate tax cost associated with the partnership
conversion model but does not avoid the valuation problem. It is administratively more burdensome than the partnership
conversion model because the corporation has to make valuations on an asset-by-asset basis and monitor assets held at the
time of the conversion (as well as income and deduction items attributable to pre-conversion periods) for a 10 year period.
In addition, this approach distributes the tax burden of the transition to integration in an unequal manner because it allows
those corporations with wasting assets or assets on which gain can be deferred beyond the end of the 10 year period to escape
corporate level tax on the gain.
9. The choice between limiting integration to newly contributed equity and extending it to all equity reflects assumptions
about the extent that investor level taxes affect corporate dividend decisions and share prices. If dividend payments are
unavoidable and shareholders do not place an intrinsic value on dividends relative to retained earnings, the classical system
does not create any bias against dividend distributions, and investor level taxes on dividends are already capitalized into share
values. This is the new view of dividend distributions. See Section 13.B. If that view is correct, then applying integration
to dividends from accumulated as well as newly contributed equity would not encourage dividends and would confer a
transition gain to holders of existing equity, the price of which would increase. As discussed in Chapter 13, however, we
reject the new view. Accordingly, we believe that extending integration to existing equity, partiCUlarly under a phase-in,
would not confer unacceptable transition gains, and that retaining the classical system for existing equity would maintain the
tax bias against dividends for such equity.
10. The Department of the Treasury recommended a phase-in approach in its 1984 proposal to provide relief from the double
taxation of corporate income. That proposal generally would have allowed corporations a deduction equal to 50 percent of
dividends paid to their shareholders and also would have reduced the corporate dividends received deduction from 75 percent
to 50 percent. The proposed 6 year phase-in rule would have allowed a 25 percent dividends paid deduction in the first year
that would have increased by 5 percentage points in each of the next 5 calendar years. Similarly, the dividends received
deduction would have been 75 percent in the first year, with a 5 percentage points decrease in the deduction for each of the
next five calendar years. See Treasury I, Vol. 2, pp. 136-137, 140.
II. The imputation credit prototype described in Chapter 11 could be phased in. The imputation credit prototype contemplates
additions to the SCA and associated shareholder level credits by reference to the maximum tax rate applicable to
shareholders, currently a 31 percent rate. Where the corporate tax rate is less than the maximum shareholder rate, it would
be appropriate to base shareholder credit account and imputation credit amounts on the lower corporate tax rate. This level
of integration might be phased in two alternative ways. First, a phase-in rate might be set as a percentage of the maximum
shareholder rate to accomplish a smooth phase-in of integration. For example, a 5 year phase-in could base the shareholder
credit account additions and allowable shareholder credits on a rate equal to 20 percent of 31 percent (6.20 percent) in the
first year, 40 percent of 31 percent (12.40 percent) in the second year and so on. Alternatively, the imputation credit
prototype might be phased in by linking imputation credits to a shareholder tax rate less than the maximum individual rate.
For example, SCAs and imputation credits might be based on the 15 percent individual rate for a several years before moving
to the 31 percent rate. If only partial distribution-related integration were contemplated, this system could be used
indefinitely. Such a system would be similar to the United Kingdom's imputation system. See Appendix B.
12. See generally Graetz (1977).
13. Most corporate debt may be called without premium after a period of time, typically 5 to 7 years. Debt instruments
typically permit the debt to be called earlier upon payment of a redemption premium. A CBIT phase-in is likely to
signi ficantly mitigate the increase in the cost of borrowing because corporations would be able to call their debt in substantial
part before the disallowance of the interest deduction is fully phased in.
14. See Section 4.G.

225

Notes

15. If an accrual method taxpayer accrues but does not pay interest before the CBIT phase-in begins, then pays the previously
accrued interest in a CBIT transition year, this approach assures that either holder level tax (in the form of the portion of
dividends and interest includable in the income of shareholders or debtholders) or compensatory tax is paid on such interest.
16.

The formula for transition years' additions to the EDA would be:

Additions to EDA = p [U.S. tax paid for taxable year -u.s. tax paid for taxable year]
.31
+ p(dividends and interest received from CBIT entities) + p(allowable interest deduction)

where p is the transition percentage.
17. As Section 4.D discusses, payments of interest and dividends reduce the EDA in the order in which they are made. These
examples assume, for purposes of illustration, that interest payments are made first and thus reduce the EDA first.

PART IV

Introduction
1. Australia, Denmark, Finland, France, Germany, Ireland, Italy, New Zealand, and the.United Kingdom have all adopted
imputation credit systems. See Appendix B for a discussion of certain of these countries' systems.
2. Differences among dividend exclusion, dividend deduction and imputation credit systems of integration are due to
differences in tax rates applicable to different shareholders or types of income. See Appendix C.

Chapter 11
1. Individual shareholders subject to rates less than 31 percent would be allowed to use the credits against tax on other
income. See Section II.E.
2. The grossed-up dividend is the cash dividend received by the shareholder divided by one minus the maximum individual
tax rate (cash dividend/I-.3I).
3. Additional restrictions on the amount of the credit would be imposed to prevent streaming of credits to taxable
shareholders, and consideration could be given to requiring corporations to frank dividends with credits at the full 31 percent
rate as long as there is a balance in the SCA. See Section 11.F.
4. See also note 48, below.
5. A compensatory tax may take either of two forms. First, it might apply only to distributions of earnings that have not been
taxed at the full corporate rate. This requires a corporation to determine the amount of corporate tax deemed to have been
paid with respect to each distribution and to pay additional tax to the extent that earnings used to make the distribution have
not been subject to tax at the full corporate rate. The French and German systems follow this model. See Appendix B.
Alternatively, the compensatory tax might be imposed on all distributions, regardless of the amount of corporate tax
previously paid, with the compensatory tax allowed as a credit against regular corporate tax. Under such an "advance tax"
system, a corporation is not required to determine explicitly the amount of tax deemed paid on a particular distribution. In
an advance tax system in which the shareholder credit is computed using a corporate tax rate of 34 percent, a corporation
is required to pay a compensatory tax on all dividends equal to 51.5 percent of the dividend (.34/.66). The corporation would
be entitled to credit this tax against its regular corporate tax liability. Shareholders would be entitled to a credit equal to 51.5
percent of the amount of any cash distribution, and the credit would be included in income together with the cash distribution.
The 51.5 percent rate applied to net cash dividends is used in lieu of applying the 34 percent corporate rate to a grossed up
amount; 51.5 percent of a $66 cash dividend ($34) equals 34 percent of $100, the $66 cash dividend grossed up at the 34
percent rate ($66/.66). A corporation's ability to credit the compensatory tax against its regular corporate tax liability means
that the compensatory tax results in additional tax liability only to the extent that distributions exceed the amount of fullytaxed earnings between the two regimes. The United Kingdom's Advance Corporation Tax (ACT) system represents an
example of the second type of compensatory tax.

Notes

226

The principal substantive difference is that the advance tax system implicitly treats distributions as made first out of fullytaxed income, while a compensatory tax can, in theory, be combined with any stacking rule. In practice, most existing
compensatory tax systems, such as those in France and Germany stack distributions first against fully-taxed income. While
they differ mechanically, the two alternatives have similar economic impact on corporations subject to the compensatory tax.

6. If a compensatory tax is set at the corporate tax rate and is refundable to shareholders so it acts solely as a withholding
tax, all distributed income is taxed only once, at shareholder rates. Although the tax is collected at the corporate level, rather
than at the shareholder level, no net separate corporate level tax is imposed. The compensatory tax, however, serves to
ensure payment of the shareholder level tax as preference or shielded foreign source income is distributed. The refund of
imputation credits associated with distributions means that the net amount of tax borne by the distribution will be determined
solely by the shareholder's tax rate and taxable or tax-exempt status.

7. Tin:ing preferences, as well as exclusion preferences, would increase the corporate level cost of dividends in a
compensatory tax system, A compensatory tax requires current payment of tax on distributed preference income, thus
removing the tax deferral created by timing preferences. Consider a firm with $100 in economic income in year one and
$100 worth of timing preferences. Suppose further that in year two its economic income is zero (but tax is due on the $100
deferred from the year before) and that the firm distributes all of its income in year one. With a compensatory tax, the firm
has to pay $34 in year one; there is no mainstream tax to which the credit can be applied. Therefore, it carries over the $34
credit to year two, so that in year m'o its tax liability is zero. In contrast, under a credit limitation system, no tax is paid
in year one, but $34 is paid in year two. Thus, if the firm's economic income is distributed as it is earned, the present value
of timing preferences to the firm under the credit limitation scheme is greater than under the compensatory tax scheme. On
the other hand, taxable shareholders would receive credits in year one in a compensatory tax regime that they would not
receive in a credit limitation system. The overall effect, therefore, would depend on the relationship of the compensatory
tax rate to that of the shareholders.
8. The imputation credit prototype, like the dividend exclusion prototype, is not expected to change significantly corporations'
provision for income tax expense or the determination of taxes currently payable or payable at a future date for financial
accounting purposes. Note 1 in Chapter 4 discusses the possible effect of a compensatory tax on corporate financial reporting.

9. Mechanically, one can determine which distributions are made out of fully-taxed income either by tracing taxable and
preference income or by tracking taxes paid. A tracing-of-income approach requires the corporation to maintain different
accounts for earnings and profits that have been taxed at different rates, including different accounts for income earned in
different years, if tax rates have changed from year to year. We consistently recommend tracking taxes paid rather than
tracing taxable income. See Section 2.B, Section 4.D, and Section 12.A. Tracking taxable income is significantly more
complicated than tracking taxes paid and does not seem to offer any offsetting advantages. Australia's imputation credit
system tracks taxes paid. The French and German imputation credit systems illustrate the complexity of tracking income.
See Appendix B.
10. The following example compares three alternative stacking rules. The example assumes that the corporation pays tax at
either 34 percent (nonpreference income) or 0 percent (preference income) and that corporate taxes paid are credited at the
3 1 percent shareholder rate.
Alternative Stacking Rules
Stack
Stack
Preferences
Preferences
Pro Rata
Last
First
Stacking
Economic Income
100
100
100
Preference Income
10
10
10
Taxable Income
90
90
90
Tax (@34%)
30.6
30.6
30.6
Preference Income Available for Distribution
10
10
10
Nonpreference Income Available for Distribution
59.4
59.4
59.4
Cash Distribution
50
50
50
Tax Deemed Paid on Distribution
17.97
22.46
20.22
The "stack preferences last" approach treats each dollar distributed as coming first from nonpreference income. The $50
distributed is less than the amount of nonpreference income available for distribution, thus, the distribution is deemed to be
entirely nonpreference income. The "stack preferences first" approach treats each dollar distributed as coming first from
preference income (taxed at zero percent) and then from nonpreference income. Thus, the first $10 distributed is deemed

227

Notes

to have borne no tax. The pro rata stacking approach treats each dollar as from preference and nonpreference income in the
slime proportion as the corporation's after-tax preference and nonpreference income. The pro rata approach thus treats each
dollar distributed in the example as having borne tax at an effective rate of 30.6 percent (90/loox34%)+(1O/lOOxO%).
The indirect foreign tax credit allowed under IRC § 902 to certain U.S. corporate shareholders uses a pro rata stacking rule
to determine the amount of foreign taxes associated with distributions from foreign corporations to related U.S. corporations.
11. The ACT in effect stacks distributions first against fully-taxed income. For example, assume that the corporate rate is
33 percent and the credit rate is 25 percent, and that a corporation earns $100 of fully-taxed income and $100 of preference
income in a year. If the corporation distributes $100, it will pay ACT of $33.33 (.25x$1001.7S).1t will owe mainstream
tax for the year of $33 and will be permitted to credit $25 of ACT against the mainstream tax. Thus, its tax liability for the
year will be $8. The effect is the same as if the corporation had first paid $33 of mainstream tax and then paid a $133.33
grossed-up distribution, deemed to be composed of $100 of fully-taxed income and $33.33 of preference income.
Compensatory tax of $8.33 (.25 x$33.33) would be due on the distribution. In both cases, the total tax paid is $41.33.
In contrast, the French and German systems explicitly adopt stacking rules that stack preferences last. The German
system uses an "available net equity" account to track taxable and preference income. Available net equity is divided into
separate "EK" baskets, consisting of income taxed at various rates. The balances in EK 50, EK 36 and EK 0 represent
income taxed at the statutory retained earnings rate, the statutory distribution rate and at a zero rate, respectively. However,
the corporation's income may actually be subject to rates other than those for which corresponding EK categories exist. The
German system converts each category of income subject to tax at some other rate into equivalent amounts of EK 36 and
either EK 50 or EK 0, as appropriate.
The following equation converts pre-tax income subject to tax at some non-EK rate into equivalent amounts of pre-tax
income subject to tax at the distribution rate (36 percent) and either the statutory rate (50 percent) or the zero rate: .36X+(.5
or 0) X (Y - X) = T, where Y equals the total amount of pre-tax income (known), X equals pre-tax income subject to the
distribution rate, (Y - X) equals pre-tax income subject to either the statutory rate or the zero rate, and T equals the amount
of tax paid with respect to Y (known). Because X and (Y -X) must be positive, the effective tax rate, T/Y, determines
whether the equation must contain the statutory rate or the zero rate (and whether the residual amount of income is converted
into EK SO or EK 0). The following equations convert the pre-tax amounts, X and (Y - X), into their after-tax EK amounts:
EK36 = (l-.36)xX
EK SO (lfT/Y > .36) = (l-.SO)X(Y-X)
EK 0 (lfT/Y < .36) = Y-X
French corporations are required to segregate fully-taxed income from income potentially subject to the compensatory
tax or precompte mobilier for tax accounting purposes. In general, dividends eligible for the imputation credit or avoir fiscal

are deemed to be distributed first out of current fully-taxed income, and then out of fully-taxed retained income of each of
the immediately preceding S years. Once fully-taxed income for this S year period has been exhausted, a corporation may
choose to allocate a dividend distribution to (1) dividends received from foreign subsidiaries, (2) the long-term capital gains
reserve, or (3) other miscellaneous preference income, in any order. France thus allows stacking of dividends last against
preference income.
Appendix B discusses these systems in more detail.
12. The formula set forth in the text is based on the formula used to determine the EOA in the dividend exclusion and CBIT
prototypes. Multiplying the EOA formula by (1/.69-1) converts after-tax income at the 34 percent corporate rate into
imputation credits at the 31 percent maximum shareholder rate.
13. If the 34 percent corporate rate were the credit rate, the credit in the example in the text would equal $17 and the 31
percent shareholder would have an excess credit of $2.17 to offset other tax liability.
14. This is the method used, for example, by New Zealand. See Appendix B, Section B.S.
15. In general, the treatment of the adjustment as a current year item should extend only to determining the SCA balance.
Interest on deficiencies or overpayments should be calculated as under present law. Under a compensatory tax, if liability
is adjusted upward, the corporation would either be allowed to use accumulated excess compensatory tax to satisfy the
liability or, if there is no excess, would be required to pay additional tax. If a corporation's prior year tax liability is adjusted
downward, it would either increase the balance in its excess compensatory tax account, or to the extent it did not use the
prior year tax liability to avoid compensatory tax on distributions, it would receive a refund. The corporation would not
receive a refund of the corporate tax payment where it has been used to avoid compensatory tax because this corporate tax

Notes

228

payment has been claimed as a credit by shareholders. If a refund were allowed, shareholders would have been able to claim
a credit for taxes that the corporation, after allowance of the refund, did not actually pay.
16. The contrary approach, which would treat audit adjustments as an adjustment to the SCA in the taxable year to which
the adjustment relates, is complicated and burdensome. Under that approach, a corporation that receives a refund of corporate
tax paid may have reported to shareholders credits in excess of its adjusted balance in the SCA. An unanticipated reduction
would occur in the SCA for the year in question, which the corporation would have to satisfy by reducing its remaining SeA
in that year, or, if there were no remaining SCA, by paying tax equal to the deficit SCA balance (together, possibly, with
imposition of penalty or interest).
17. Allowing a loss to be carried back to obtain a refund of some or all of the taxes used to frank a dividend may be
appropriate in theory, particularly if the corporation's shareholders are the same at the time of the dividend and the loss, but
would be difficult to implement in practice. For purposes of determining shareholder level consequences, the franked
dividend could be recharacterized retroactively as a return of capital or a distribution of preference income, depending upon
whether the corporation had sufficient retained preferences income at the time of the dividend. If the distribution constituted
a return of capital, no shareholder level tax would be due, but basis in the stock would be reduced by the amount of the
distribution (which would not be grossed up for the credit). If the distribution were paid out of preference income, the
amount of shareholder level tax would be computed only on the amount of the distribution (which also would not be grossed
up for the credit). Requiring retroactive adjustments in shareholders' basis or tax liability would be impractical to administer,
however, especially if shares of a corporation are widely held.
The argument that tax refunds should be limited to the SCA balance is weakened somewhat because, under the credit
limitation system without full refundability, amounts withdrawn from the SCA to frank past dividends may not actually have
been used by shareholders. Shareholders cannot obtain refunds of imputation credits, and thus tax-exempt, foreign and some
low-bracket shareholders may not enjoy the benefit of some imputation credits. In contrast, in a system with full refundability
of imputation credits, all SCA amounts used to frank dividends would be fully used by shareholders. While there is thus
some theoretical justification for allowing refunds in excess of the SCA to the extent that the imputation credits were not fully
used, it would be impractical to trace the use of the imputation credits by shareholders.
18. Current law contains limitations on the ability of taxpayers to accelerate the recognition of losses or to increase the
amount of loss recognized for tax purposes to an amount exceeding the loss incurred economically. Such limitations include
limitations on the deductibility of investment interest, passive activity losses, and amounts in excess of the amount the
taxpayer has at risk with respect to an activity. Under present law, these limitations either do not apply to C corporations
or apply only to C corporations that are personal service corporations or closely held corporations (essentially defined as
corporations more than 50 percent of the stock of which is held by or for five or fewer individuals).
By eliminating or reducing substantially the tax disadvantages of incorporation, distribution-related integration may
encourage the use of corporations to avoid these rules. Because distribution-related integration removes the double tax on
distributed corporate earnings, taxpayers may view corporations as attractive vehicles for engaging in activities designed to
accelerate or increase tax losses. For example, individuals might use passive activity losses by contributing a loss-producing
passive activity and an income-producing active business to the same corporation. The deferral benefit achieved by this
structure would continue until the earnings sheltered by the preference were distributed. Distributed income would be fully
taxable to taxable shareholders, although it would be tax-exempt in the hands of exempt shareholders. In addition, the income
generated when the preference reverses would be subject to only one level of tax. Thus, it may be appropriate to extend some
of or all the loss limitation rules described above to C corporations if, after distribution-related integration is adopted,
experience shows that taxpayers are using C corporations to avoid those rules.
19. A dividends received exclusion (DRE) would be as effective as a DRD in preventing multiple taxation of corporate
dividends. The two could, however, produce different technical effects increases where Code limits or classifies taxpayers
based on receipts or income. For example, dividends are taken into account under IRC § 448(b)(3), which limits the
availability of cash method accounting for certain taxpayers with annual gross receipts in excess of $5 million. See Treas.
Reg. § 1.448-1 T(f)(2)(iv). By contrast, dividends are excluded under IRC § 263A(b)(2)(B), which limits capitalization of
cost requirements for certain taxpayers whose annual gross receipts do not exceed $10 million. See Treas. Reg. § 1.263A1T(d)(2)(iv)(B). Regardless of the general approach, however, special adjustments may be provided wherever appropriate.
See, e. g., IRC § 170(b)(2)(B) (corporate charitable deductions are limited to 10 percent of taxable income determined without
regard to the DRD). During any period of transition to integration, the current law DRD could be increased in stages from
70 percent to 100 percent as the percentage of integration increases. During periods when there is less than 100 percent
integration, a 100 percent DRE would be inappropriate and also would require appropriate phase-in.

229

Notes

20. If all dividends were either fully unfranked or completely franked, it would be relatively easy to retain the current 70
or 80 percent DRD. The mechanics would be similar to those discussed in Section 2.B in the context of the dividend
exclusion system. Partially franked dividends would create signi ficant complexity, however. To determine its DRD a
corporation eligible for only a 70 or 80 percent DRD would have to separate a partially franked dividend into a fully franked
portion and a completely un franked portion.
Example. A corporation that has a zero SCA balance owns 5 percent (by vote and value) of the stock of a second
corporation and has no other assets. The second corporation pays a cash dividend of $166, which carries an
imputation credit of $29.65.
The recipient corporation must convert the partially franked dividend into fully franked and un franked components.
A $29.65 imputation credit would fully frank a cash dividend of $66. Thus, the un franked dividend is $100 ($166$66). After taking into account the 70 percent DRD, the corporation must pay tax of $10.20 on $30 of income.
Using the formula in Section 11.B, the corporation would add $38.55 ($29.65 for the credits received on the franked
portion plus $8.90 with resped to the $10.20 of tax paid on the un franked portion) to its SCA. If the corporation
then distributed all its remaining cash to shareholders, it would distribute $155.80 of cash ($166 -$10.20) and attach
an imputation credit of $38.55. Assuming a 31 percent shareholder rate, shareholders would pay tax, after claiming
imputation credits, of $21.70 «$194.35 gross dividendX.31)-$38.55). This represents shareholder tax at the 31
percent rate on the remaining $70 of preference income not taxed in the hands of either corporation.
21. The alternative would tax the recipient corporation on the dividend and permit the tax to be offset by any imputation
credit attached to the dividend. The imputation credit and any additional corporate taxes paid on the dividend would increase
the recipient's SCA. This alternative rule would eliminate tax preferences upon the initial distribution of preference income,
whether the distribution was made to a corporate or an individual shareholder.
22. A compensatory tax system might suggest a different result. Once the decision is made to tax distributed preference
income to the distributing corporation, the rationale for extending preferences while the distributed income is in corporate
solution may not be compelling. See Section 4.D. As noted in the text, however, some countries with compensatory tax
systems (notably the United Kingdom) forgo the compensatory tax for certain intercorporate dividends.
23. See H. Rept. No. 426, 99th Cong., 1st Sess. (1985), p. 302; S. Rept. No. 313, 99th Cong., 2nd Sess. (1986), p. 515.
24. If, unlike the prototype recommended here, the SeA were based on tracing taxable income, difficulties with respect to
the AMT would arise in determining the amount of tax that has been paid with respect to a particular distribution by a
corporation that has paid AMT. However, under the tracking-tax-paid approach, adding minimum taxes to the SCA can be
done directly. As indicated in note 26, the amount added to the SCA would be adjusted to reflect the maximum 31 percent
rate at the shareholder level. Indeed, the need to allow imputation credits with respect to corporate AMT is an important
reason for preferring the tracking-of-taxes-paid approach to a tracing-of-taxable-income approach under the credit limitation
system.
25. The corporate AMT also seems appropriate under a compensatory tax. While a compensatory tax would prevent the
passthrough of preferences to shareholders, it would not ensure that corporations pay some level of tax on retained income.
Imputation credits attached to a dividend represent tax prepaid at the corporate level and thus should be allowed for
purposes of the individual AMT.
Example. A shareholder with a 31 percent marginal rate has $100 of AMT preference income, a $100 gross
dividend, and a $31 imputation credit. Her AMTI is thus $200. She should owe only $17 in AMT ($48 of tax less
the $31 imputation credit). Mechanically, this can be accomplished by computing her regular tax for AMT purposes
as zero ($31 of tax less $31 imputation credit), but allowing the full imputation credit in computing tentative
minimum tax. Thus, her tentative minimum tax is $17 ($48-$31) and her AMT is $17 ($17-0).
Similarly, we recommend that excludable dividends not be viewed as preference income for individual AMT purposes
under the dividend exclusion and CBIT prototypes. See Section 2.E and Section 4.D.
26. Although the AMT rate is 20 percent, compared with the maximum shareholder rate of 31 percent, corporate AMT
payments are not added dollar-far-dollar to the SCA but instead, like regular tax, are reduced to reflect the difference
between the corporate and shareholder rates. This rule is necessary because corporate AMT payments give rise to an equal
AMT credit that offsets regular corporate tax at the 34 percent rate.

Notes

230

Example. A corporation invests $100 in an asset that will produce $100 per year for 2 years. As a deferral
preference, the corporation is entitled to expense the asset in the first year.
Year

2

AMT

Regular tax
before credit

credit

Tax
due

Cummulative
SCA

20

nJa

nJa

20

17.44

nJa

34

20

14

29.65

Cash
flow

Taxable
Income

AMT

100

0

lOO

100

At the end of year two, the corporation has an SCA of $29.65 and $66 of retained earnings. The corporation
distributes $66 to shareholders, and no additional tax is due. If the AMT were instead added to the SCA dollar-fordollar, the corporation would have an SCA of $32.21 and excess credits of $2.56.

27. Mechanically, the limitation on additions to the SCA allows distributions by the U. S. corporation out of earnings
attributable to dividends from the foreign corporation to be treated in the same manner as distributions out of earnings
attributable to preference income from U.S. sources.
28. IRC § 901.
29. Section 2.C discusses a shareholder level exclusion of foreign source income.
30. Continuing to tax income distributed to shareholders but preserving the benefit of preferences for tax-exempt shareholders
under a compensatory tax system would require making imputation credits attributable to the compensatory tax fully
refundable to tax-exempt shareholders. If policymakers were to choose to tax preference and foreign income as well as
nonpreference income received by tax-exempt shareholders, a compensatory tax should be adopted with nonrefundability of
credits to tax-exempt shareholders. This result cannot be accomplished under a credit limitation system without a
compensatory tax. Such a compensatory tax system might be limited to preference income, but this would require separate
tracking of foreign source income, which could continue to be paid free of U.S. tax to tax-exempt entities. Alternatively,
if, contrary to the recommendations here, one chooses to tax neither preference nor nonpreference income distributed to taxexempt shareholders, credits should be made refundable to tax-exempt shareholders; a system of refundable credits could
be provided with either a compensatory tax or a credit limitation system. Refundability, however, would cause significant
revenue loss.
31. See also Section 6. D for a discussion of an alternative approach under an integrated system that could be designed to
maintain the overall level of tax revenues collected on corporate capital supplied by tax-exempt entities and achieve greater
neutrality between the tax burden on their debt and equity capital.
32. Assume, for example, that a U.S. corporation with 1,000 shares outstanding of a single class of stock and an SCA
balance of $2,000 makes a distribution of $10 per share and designates $2 per share as the applicable imputation credit with
respect to each share. One hundred of the corporation's shares are owned by a foreign person subject to U.S. withholding
tax at a rate of 15 percent under an applicable tax treaty. The foreign shareholder will be subject to U. S. withholding tax
of $150 on the distribution of $1 ,000 (100 shares x $ 10 distribution X 15 percent withholding tax). The corporation will reduce
its seA by $2,000, although the foreign shareholder cannot offset the imputation credit against the U.S. withholding tax.
33. Consideration might be given to allowing a shareholder to carryforward unused imputation credits for some period of
time, such as 5 years. Such a carryforward would add complexity, but should serve to enable virtually all shareholders
subject to original tax rates below 31 percent and those currently in a tax-loss position to use any excess credits.
34. If imputation credits were fully refundable to all taxpayers, corporations and their shareholders would have no tax
incentive to develop strategies for directing the credit to particular taxpayers. Because fully refundable credits would be
equally valuable to all taxpayers, taxpayers would be indifferent to the form of a distribution, e.g., a $69 dividend carrying
a $31 credit versus a $100 dividend carrying $0 credit or $100 of interest or other income such as rent or wages. However,
in accord with the recommendations of Chapters 6 and 7, this prototype does not permit refunds of credits to tax-exempt
or foreign shareholders. Credits thus would be available only to offset tax liability the taxpayer would otherwise owe on the
dividends or other income. As a result, certain taxpayers, e.g., tax-exempt and foreign shareholders, would not be indifferent
between receiving a dividend carrying a credit and a higher cash dividend distribution because to them the credit would not
be the C(]uivalent of cash

231

Notes

If the alternative tax on investment income, described in Section 6.D, were adopted, imputation credits would be used
by tax-exempt entities to reduce or eliminate that tax and the incentives for streaming would be reduced.
35. One difference is that the imputation credit prototype allows low-bracket shareholders to use excess credits to offset tax
on other income.
36. New Zealand requires a corporation generally to frank all dividends paid during a year to the same extent even if the
dividends relate to different classes of stock. A corporation may change its franking ratio during a year only if an officer
of the corporation declares that the change is not "part of an arrangement to obtain a tax advantage" and the corporation
notifies the tax authorities of the change.
Australia has adopted several rules to prevent a corporation from underfranking a dividend. These rules require the
corporation (1) to take into account all dividends that are paid on the same day, that have been declared but not yet paid,
or that the corporation is committed to pay later in the same year (a "committed future dividend"), e.g., dividends on
preferred stock, in allocating franking credits to a given dividend, (2) to frank a dividend that was a committed future
dividend at the time of payment of an earlier dividend at least to the same extent as the earlier dividend, and (3) to frank
a dividend at least to the same extent as any other dividend paid on the same day. These rules, however, do not prevent a
corporation from franking an earlier dividend at one rate and franking a later dividend at a lower rate if the corporation is
not committed to pay the later dividend or the later dividend is paid in the next year.
Additional anti-abuse rules might be adopted as necessary. See Appendix B for a discussion of anti-streaming rules
adopted by certain of our trading partners.
37. The implementation of distribution-related integration may require certain adjustments to the treatment of qualifying
reorganizations to reflect the shareholder credit system. One issue is whether the current law treatment of "boot" (money
or property other than stock or securities in a corporate party to the reorganization) is appropriate under distribution-related
integration. Under current law, a shareholder receiving boot in a reorganization recognizes gain equal to the lesser of the
gain realized and the amount of boot received. If the receipt of boot has the effect of a dividend, gain recognized is taxed
as a dividend to the extent of the shareholder's ratable share of the corporation's earnings and profits. Dividend equivalency
is tested by treating a target shareholder as receiving only stock of the acquiring corporation and the acquiring corporation
as then redeeming an amount of the shareholder's steck equal to the amount of boot received.
The current treatment of boot raises problems under distribution-related integration because of the rule that limits the
amount of boot that is taxable to the amount of the recipient's realized gain. Under distribution-related integration, this would
allow the distribution of preference income to high-basis shareholders without shareholder level tax. It also would allow the
distribution of fully-taxed income to high-basis shareholders without a reduction in the seA, so amounts in the SeA
subsequently could be used to frank distributions of preference income. This is similar to the issue created by share
repurchases. If policy makers adopt special rules for share repurchases, similar rules may be appropriate for boot. See
Chapter 8.
38. Assume, for example, that a corporation has two active businesses, each generating a mix of taxable and preference
income. If the corporation could isolate each of the businesses in a separate corporation but leave the entire SeA balance
in one corporation, shares of the corporation without any SCA balance could be distributed to tax-exempt shareholders, and
shares of the corporation with the seA balance could be retained by taxable shareholders.
39. In April 1990, Representative Vander Jagt introduced legislation that essentially adopts this approach. H.R. 4457, Wist
Cong., 2d. Sess. (1990). The Vander Jagt bill would allow a tax credit to a shareholder or bondholder equal to the "gross-up
amount" included in the holder's income. A recipient of a cash dividend or interest payment from a C corporation would
include the gross-up amount, as well the cash received, in income. However, the amount of the credit would be limited to
a portion of the taxpayer's tax that equals the ratio of his interest and dividend income to his total income. A corporation
would be required to attach credits to a payment of interest or dividends representing the same proportion of the corporation's
post-1989 taxes as the ratio of the amount of the net dividend or interest payments bears to post-1990 undistributed earnings
and profits. No deduction would be allowed for interest or original discount paid or accrued by a C corporation. See also
note 1 in Chapter 4.
The AU Reporter's recent integration memoranda also adopt such an approach. See American Law Institute, Reporter's
Memorandum No.3 (1991).

Notes

232

40. A bondholder credit system could be adopted either while retaining the current deduction for interest paid by corporations
or in a system denying deductions for either interest or dividends at the corporate level. Retaining the deductibility of interest
would require imposing a withholding tax on interest payments and allowing recipients a credit for such withholding. The
following example shows the calculation of the imputation credit with and without an interest deduction.
Example. For simplicity, this example assumes that the corporate rate is 31 percent. A corporation earns $100 of
taxable income and agrees to pay $50 of after-tax interest. If no interest deduction is allowed, the corporation would
pay tax of $31 and would add $31 to its taxes paid account. The taxes paid account would represent available
imputation credits for both interest and dividends. The corporation could attach an imputation credit of up to $22.46
to the interest payment. The $8.54 remaining in its taxes paid account would fully frank its remaining after-tax
earnings of $19.

If an interest deduction is allowed but a withholding tax on interest is imposed, the corporation would have to pay
gross interest of $72.46. Net of the 31 percent withholding tax ($22.46), the interest payment would be $50. Taking
into account the $72.46 interest deduction, the corporation would have taxable income of $27.54 and would owe
tax of $8.54. Thus, the total tax paid would be $31 ($22.46 +$8.54). The corporation's SCA balance, which would
be available only to frank dividend payments, would be sufficient to frank a dividend of its remaining after-tax
earnings of $19.
41. Therefore, CBIT might be viewed, to some extent, as substituting taxation of the payor for taxation of the recipients.
To illustrate the concept of substitute taxation, assume a manufacturer borrows $100 for one year and agrees to pay $10 of
interest to the lender. Assume both the manufacturer and the lender have a 31 percent marginal tax rate. The manufacturer
plans to use the $100 to produce a product that will provide a return sufficient to pay $110 to the lender at the end of the
year. At the end of the year, the manufacturer sells the product for $115. Under current law, the manufacturer's taxable
income is derived by deducting from its $115 of gross sales $100 for wages, materials, and other costs of producing the
product, and $10 for interest expense. The manufacturer would be liable for tax of $1.55 ($5x.31), and would use the
remaining $113.45 ($115 -$1.55) to repay the $100 principal on the loan and the $10 interest, leaving an after-tax return
of $3.45. The lender would pay $3.10 of tax on its interest income ($lOx.31) and would receive an after-tax return of
$6.90.
Under CBIT, the lender need only be paid $6.90 in interest. The manufacturer's taxable income would be determined
by deducting from gross sales the $100 for wages, materials, and other production costs. Thus, the manufacturer would have
taxable income of $15 ($115 - $1 (0) and would pay $4.65 of tax ($15 X .31). The manufacturer would then use the $110.35
in after-tax gross receipts ($115 -$4.65) to pay $100 in principal on the loan and $6.90 in interest to the lender. The lender
would not include the $6.90 of interest it received in its taxable income, because the tax on that income was by the
manufacturer. The manufacturer's after-tax return would be $3.45 ($110.35-$106.90), and the lender's after-tax return
would be $6.90. Compared to current law, the manufacturer's $4.65 CBIT liability can be viewed as including the same
$1.55 of income tax on the manufacturer, and an additional tax of $3.10 on the lender's interest income; CBIT substitutes
an additional $3.10 of tax on the borrower for the income tax that current law would impose on the lender.
42. The fact that the imputation credit system taxes income at the shareholder's or lender's rate creates other differences
between the two models. For example, no small business exception would be needed. The bondholder credit system, like
an imputation credit system, also provides greater flexibility to change policy recommendations in the future. For example,
relief could be provided to tax-exempt and foreign investors simply by permitting full or partial refunds of imputation credits.
Compare Section 4.F. As with the imputation credit system, however, this flexibility is earned at the cost of substantial
complexity.
43. It may be appropriate to retain the withholding tax for un franked dividends and interest payments. The issue is the same
as the treatment of taxable dividends and interest payments if no compensatory tax is imposed under CBIT. See Section 4.E.
44.

Example. A corporation earns $100 of taxable income, pays tax of $34, and adds $29.65 to its SCA. See
Section 11. B for a discussion of how the SCA balance is calculated. The corporation could elect to pay
deemed dividends of up to $66 «$29.651.31-$29.65) = $66). If the corporation declared a deemed
dividend of $66, shareholders would include $95.65 in income and would be entitled to imputation credits
of $29.65. Share basis would increase by $66.

45. Excess credits could be used to offset other tax liability, but would not be refundable, as with imputation credits attached
to a cash dividend.
46. See Section 9.A for a discussion of the allocation of basis among shares.

233

Notes

47. The prototype also adopts a holding period requirement and extends certain other rules of current law. See Section lI.F.
Those rules would apply to deemed dividends as well as to cash dividends.
48. The rule described in the text would not prevent a corporation from adopting a dividend policy under which it pays
unfranked cash dividends. It would, however, prevent a corporation from both paying partially franked or un franked
dividends and using the elective DRIP. Neither of the two common reasons that might lead a corporation to pay partially
franked or unfranked dividends arise in circumstances in which a DRIP would be useful. First, a corporation might want
to distribute cash but have an insufficient SCA balance to frank all dividends fully. In that case, however, the SCA balance
will be completely exhausted by the cash distributions, and the corporation will neither need nor be able to use the DRIP.
Second, the corporation might want to retain an SCA balance to frank future distributions. If the corporation intends to retain
an SCA balance for future use, however, it would not use the DRIP to reduce its SCA balance.

Chapter 12
1. See Treasury I, Vol 2, pp. 136-37,140; and The President's 1985 Proposals, pp. 122-26. A partial or full deduction for
dividends paid is often expressed in terms of a split rate system, in which distributed earnings face a lower tax rate than
retained earnings. With a full dividend deduction, a split rate system results in a zero corporate tax rate for distributed
earnings. With partial dividend deductibility, the effective rate of deduction is (tc - tJ/tc , where tc and td are, respectively,
the tax rate on retained earnings (the corporate rate) and distributed earnings.

2. Although a dividend deduction could avoid extending integration benefits to tax-exempt and foreign shareholders by
imposing non-refundable, corporate level withholding, such a system replicates the imputation credit discussed in Chapter 11.
For example, the imputation credit prototype could be duplicated by withholding at a 34 percent rate and allowing credits
at a 31 percent rate. The two systems may have different nontax consequences. See American Law Institute, Reporter's
Memorandum No.1 (1990), pp. 45-47.
3. See Section 13.H.
4. Compare Institute for Fiscal Studies (1991) and the Reporter's Study Draft proposals discussed in Sections 12.B and 12.C,
which avoid this problem by imputing a deduction on equity capital rather than tracking actual dividend payments.
5. See Section 2.B. This account would restrict the dividends paid deduction to the amount of income that otherwise would
have been taxed fully at the corporate level. For example, if a corporation paid tax of $34 under current law it should
beallowed a dividend deduction of up to $100--the pre-tax earnings, not the after-tax amount of $66 added to the EDA. The
difference occurs because the dividend deduction system operates on a pre-tax basis whereas the dividend exclusion system
operates on an after-tax basis. Presumably, the corporate AMT be retained and the interaction between dividend deductions
available for regular tax purposes and for AMT purposes would have to be addressed.
6. The following examples illustrate how such results would occur, absent a limitation mechanism similar to the EDA.
Example 1. A corporation earns $100 of tax-exempt bond interest income in one year. The corporation has no
additional earnings in the next year and distributes the $100 of exempt income it earned in the first year. The
corporation has a dividend deduction of $100, creating a net operating loss that can'be carried forward to shelter
$100 of future retained taxable income from tax.
Example 2. A corporation earns $100 of foreign source income and pays foreign taxes of $34 in one year. After
the foreign tax credit, it pays no U. S. tax. In the second year, the corporation has no additional earnings but
distributes $66. The corporation has a dividend deduction of $66, which creates a $66 net operating loss that can
be carried forward to shelter $66 of future taxable earnings.
7. An alternative approach, suggested in The President's 1985 Proposals, would require the distributing corporation to report
to shareholders the portion of the dividend deducted. The deducted portion would be fully taxable to the corporate
shareholder. The nondeducted portion would be eligible for a 100 percent dividends received deduction. Thus a corporate
shareholder would be entitled to a 100 percent dividend received deduction with respect to dividends received in excess of
the distributing corporation's previously taxed earnings. This approach would preserve preferences until distributed out of
corporate solution.
8. See Chapter 9, note 2.

234

Notes

9. See Institute for Fiscal Studies (1991) and the description in Gammie (1991).
10. While the proposal would reduce tax-induced distortions in corporate financing decisions, if capital gains from retained
earnings were to receive very favorable tax treatment at the investor level the IFS proposal would tend to encourage
retention.
11. Shareholders funds are defined as:
(1)
(2)
(3)
(4)
(5)
(6)

shareholders' funds for the previous period, plus
any new equity contributed, plus
the AFCE allowance for the previous period, plus
the taxable profits for the previous period, less
the tax paid on those profits, less
dividends and distributions to shareholders and capital repaid.

A new corporation would have shareholders' funds for the initial period equal to the value of the equity capital contributed
by shareholders. Additional rules would be needed to determine an existing corporation's shareholders' funds on the date
of introduction of AFCE.
12. The following example illustrates the difference between intercorporate equity and debt investments under the proposal.
If Corporation A uses $100 raised from new equity to buy shares in Corporation B, shareholders' funds are $0 for A and
$100 for B. If, on the other hand, A raised $70 from equity and $30 from debt to buy shares in B, A would have
shareholders' funds of -$30. The negative AFCE allowance would reduce the interest deductible on the $30 of debt against
A's profits.
13. See American Law Institute, Reporter's Study Draft (1989).
14. According to the Reporter's Study Draft new equity capital includes "all amounts paid in for stock or as shareholder
contributions to capital after the date of tbis proposal." The critical distinction is between "accumulated" and "contributed"
equity. Earnings on new "contributed" capital become "accumulated" capital, do not increase the QCC, and, therefore, do
not qualify for a dividend deduction. The intent is to treat contributed equity capital in a manner consistent with new
borrowing. That is, if the allowable rate for deduction were 7 percent, an increase in contributed equity of $1 million would
generate $70,000 in dividend deductions. Earnings on the $1 million invested would not qualify for a dividend deduction.
15. An important difference between the IFS and Reporter's Study Draft proposals is that the former grants dividend relief
to both accumulated and new equity, wbile the latter grants relief only to new equity. The Reporter's Study Draft
distinguishes between accumulated and contributed equity. An allowable dividend deduction is computed as the product of
new contributed equity and the allowable rate.
16. As a consequence, low-bracket investors would be subject to a lower tax burden on dividends than on nondividend
distributions.
17. The four Reporter's Study Draft proposals include coordinating rules to ensure that any particular transaction is subject
to no more than one of these rules. For example, the MID is imposed only to the extent that a distribution does not trigger
interest disallowance or a reduction in the capital base for the dividends paid deduction. The MID also does not apply to
the purchase of stock as a portfolio investment. A distribution does not trigger interest disallowance to the extent that it
reduces the capital base for the dividends paid deduction.
18. See Chapter 10 and Section 13.B.

PART V
Chapter 13
1. See, e.g., Shoven and Whalley (1972), Shoven (1976), Ballard, Fullerton, Shoven, and Whalley (1985), and Fullerton,
Henderson, and Mackie (1987).
2. See Gravelle and Kotlikoff (1989).

235

Notes

3. Whether these distortions in fact create significant efficiency costs depends on the response of business enterprises to the
tax bias against incorporation. Gordon and MacKie-Mason (1991), analyzing data on individual business enterprises, find
that changes in organizational form (between C and S corporations, and between S corporations, partnerships, and
proprietorships) are sensitive to changes in tax rates and other tax policy incentives.
4. For example, some potential investments that benefit from corporate organization on. account of liquidity of corporate
securities or access to capital markets will not be undertaken even if they earn more (before taxes) than comparable
investments in the noncorporate sector. Publicly traded partnerships, including master limited partnerships with units traded
on organized exchanges, can have the liquidity of publicly traded corporations without the corporate taxes if they limit their
investments to certain types of activities, principally real estate and natural resources. REITs, REMICs, and ruCs avoid a
second level of tax provided they satisfy certain restrictions on assets and business activities. Alternatively, businesses may
elect S corporation status. This allows them to retain some of the benefits of incorporation, but at the expense of confonning
to certain restrictions. For example, S corporations have limitations on the number of investors they can have and the type
of stock they can issue. See IRC § 1361 (b).
5. In addition to corporate domestic income as a percentage of net national product, mentioned earlier, Figure 13.2 shows
gross domestic product of all corporations and nonfmancial corporations, relative to gross domestic product; and gross
domestic product of nonfmancial corporations relative to GNP, from 1950 to 1990.
6. Compare the declines in 1989 and 1990 in corporate profits relative to net national product (Figure 13.1) and in total
income in the corporate sector relative to net national product, gross domestic product and gross national product (Figure
13.2) with the stability in income of proprietorships and partnerships relative to net national product (Figure 13.1).
7. S corporation income here is measured consistent with pre-1987 figures.
8. In the Midsession Review of the Budget (1990), estimated corporate receipts were decreased by approximately $7.5 billion
to reflect revisions of the 1986 Act's effect on corporate income taxes and the greater than anticipated use of Subchapter S
filings by corporations.
9. A bias would remain, however, if business tax preferences and losses that reduce the effective tax rate on noncorporate
income did not pass through corporations to their shareholders.
10. A common rule of thumb is that the accrual-equivalent tax rate on capital gains is about one-fourth the statutory rate.
See Poterba, "Tax policy and corporate saving" (1987) and the references therein. This adjustment captures reductions
attributable to deferral and to the fact that the basis of inherited property is stepped up to fair market value (eliminating the
tax on capital gains accrued before the holder's death).
11. For example, in the late 1970s, marginal tax rates on individuals were as high as 70 percent for unearned income, while
the top marginal rate on corporate income was 46 percent and there was a 60 percent exclusion for long-term capital gains.
This created an incentive in some cases to shift income into corporations, because the combination of the corporate tax rate
and the effective capital gains rate was lower than the individual tax rate on the same amount of income. See Feldstein and
Slemrod (1978). This was particularly likely to be true for corporations with income low enough to take advantage of the
graduated corporate rate structure.
12. In comparing corporate and noncorporate investments, however, the degree of bias may be reduced by the existence of
accelerated depreciation allowances. The relative importance of those allowances depends upon the marginal business level
tax rate facing the corporate or noncorporate enterprise. In the case of the debt-equity choice, the focus is on a corporation
contemplating the best method to finance that portion of net investment that is not being funded by the government through
a policy of accelerated writeoffs. The existence of accelerated allowances is immaterial to that choice.
13. In certain special cases, however, debt may not enjoy a tax advantage over equity. Consider, for example, a corporation
whose tax liability is determined under the AMT. That corporation faces a 20 percent corporate income tax rate. Thus, if
the accrual-equivalent capital gains rate were sufficiently low relative to the shareholder tax rate on interest income, equity
might be the tax preferred form of financing for the minimum tax corporation.
Because statutory corporate tax rates are graduated, a corporation with taxable income under $75,000 also would face
a relative low (15 to 25 percent) corporate tax rate. For such a corporation, equity is less tax-disadvantaged than for
Corporations with larger profits that face the 34 percent statutory tax rate. In addition, a corporation with a substantial net
operating loss can be thought of as having a low corporate tax rate and, therefore, as deriving little benefit from debt as
opposed to equity financing.

Notes

236

14. The idea that debt can improve managerial incentives is at the core of Jensen's (1986) "free cash flow" theory, a
prominent explanation of the increase in debt financing. Jensen contends that managers, if given the leeway, will take
advantage of the inability of suppliers of funds to ascertain whether the firm is investing efficiently. Managers may squander
cash flow by investing for their own benefit in projects with negative present value. An arrangement in which outside lenders
hold debt and managers hold the residual claims minimizes this misuse of cash flow. Higher productivity (and, hence,
shareholder profitability) could result from better managerial incentives. Some studies providing empirical evidence in support
of this proposition are reviewed in Bernanke (1989).
This theory is subject to challenge, however. While debt fmancing is one way to mitigate the problem Jensen describes,
it may not be the best option. If the objective is to make managers bear more residual risk, other means could be used
(including incentive-based management compensation or reform of the oversight role, which in principle is exercised by
boards of directors). Tax considerations have likely played a role. If taxes have contributed to increased debt, then high debt
levels may not be the most efficient way to operate the firm.
15. This is true to the extent that debt is costly to renegotiate. See Gertler and Hubbard (1990). The idea is that managers
should be made residual claimants only on the component of profits they can influence: the firm specific component. For
example, managers should not be punished if the business does poorly during a recession but no worse on average than its
competitors.
16. See Warshawsky (1991).
17. Looking at changes in debt to asset ratios in the "upper tail" (the ninetieth percentile corporations) reveals that some
firms are close to having negative net worth on a market-value basis.
18. See Bernanke and Campbell (1988), Bernanke, Campbell, and Whited (1990), and Warshawsky (1991).
19. The empirical evidence on the effect of taxes on corporate borrowing decisions is mixed. Studies by Ang and Peterson
(1986), Long and Malitz (1985), Bradley, Jarrell, and Kim (1984), and Marsh (1982), for example, fail to find plausible
or significant tax effects. Other studies, in contrast, find significant relationships between tax policy variables and corporate
borrowing. See, e.g., Auerbach (1985), Bartholdy, Fisher, and Mintz (1985), MacKie-Mason (1990), and Masulis (1983).
At least two studies have directly estimated the responsiveness of corporate debt financing to changes in the tax advantage
of debt. Nadeau (1988) estimates that a 1 percent increase in the tax advantage of debt relative to equity will cause a 0.2
percent increase in the fraction of external funds obtained by issuing debt. Rangazas and Abdullah (1987) estimate that a 1
percent increase in the tax advantage of debt relative to equity will cause a 0.12 percent increase in the debt to value ratio
in the short run, and a 0.4 percent increase in the debt to value ratio in the long run.
20. This argument is made formally in Gertler and Hubbard (1991).
21. Some financial economists have maintained that tax parameters are irrelevant for dividend payout decisions, arguing
that share prices of dividend paying firms are set by investors who face equivalent (typically zero) tax burdens on dividends
and capital gains. See, e.g., Miller and Scholes (1978).
22. See, e.g., Bhattacharya (1979) and Miller and Rock (1985).
23. The new view (sometimes described as the "tax capitalization" or "trapped equity" approach) is developed in King
(1977), Auerbach (1979), and Bradford (1981). See also the survey in Poterba and Summers (1985).
24. A temporary change in the dividend tax rate would change both dividend payments and investment incentives because
of intertemporal substitution.
25. Again, investment incentives are only affected by transitory changes in investor level dividend tax rates.
26. Under the new view, other tax factors such as the corporate tax rate and capital cost recovery allowances affect the
corporation's dividend distributions and the investment policy. To understand why, under the new view, permanent dividend
taxes do not affect investment incentives, one must recognize that this view assumes that retained earnings provide the funds
for marginal corporate equity financed investment. Consider, for example, a corporation that wants to invest $1 of capital
by retaining an additional dollar of earnings. To retain the dollar, the corporation must reduce dividends by $1. At a 20
percent marginal individual income tax rate, the $1 of dividends foregone represents $0.80 net of the personal level tax on
dividends, so $0.80 represents the cost of the investment in terms of dividends foregone. In the following period, suppose
the investment earns a 6.4 percent pre-tax return, leaving $0.043 to distribute to the shareholders after paying corporate tax

237

Notes

at a 34 percent marginal corporate income tax rate (0.043 = 0.064 X (1-0.34». Upon distribution, the shareholder receives
a net dividend of $0.034, after paying the 20 percent tax on the dividend distribution (0.034 = 0.043 x(I-0.20».
In determining investment incentives, however, it is the return to the shareholder relative to the cost of the investment
that is relevant. In our example, the investment costs the shareholder only $0.80 in terms of foregone dividends, since that
is how much she would have had to invest if the $1 had been distributed to her rather than reinvested within the corporation.
Consequently, the rate of return relevant for determining whether the investment should be undertaken is 3.4 percent divided
by 80 percent (4.3 percent), the pre~dividend tax return. Because the cost of the investment is always reduced by the dividend
tax in exactly the same proportion that the return from the investment is reduced by the dividend tax, the dividend tax does
not affect investment decisions under the new view.
The new view does assume, however, that share appreciation on investments fmanced by retained earnings is subject
to capital gains tax. The effective accrual tax rate on capital gains does affect investment incentives, even under the new
view. To see why, assume that the effective accrual tax rate on capital gains is 6 percent. When the corporation retains a
dollar, the investor owes capital gains tax of $0.06*q, where q gives the share appreciation caused by $1 of retained
earnings. We assume that the firm pays dividends, so that q must equal 0.851 (0.851 = (1-0.2)/(1-0.06» to insure that
shareholders are just indifferent between dividends and retained earnings. Thus, the shareholder pays capital gains tax of
$0.051, thereby sacrificing a total of $0.851 in after-tax income to make the investment of one dollar. In the next year, the
investment pays a dividend of $0.043, of which the investor keeps $0.034 after paying taxes at a 20 percent rate. To measure
the investor's after-tax rate of return, we must adjust for the fact that only $0.851 was sacrificed rather than $1. As a result,
the investor earns a 4 percent rate of return (0.04 = 0.034/0.851) after taxes. Note, however, that since the investment yields
4.3 percent before investor level taxes, the investor level tax rate is simply the 6 percent effective tax rate on capital gains
(0.04 = 0.043*(1-0.06». Thus, the capital gains tax, but not the dividend tax, reduces the investor's incentives under the
new view.
27. Under the new view, managers are assumed to maximize shareholder value, and corporations can be described as
"immature" (with desired investment spending exceeding internal funds) or "mature" (with internal funds exceeding desired
investment spending). Immature firms use their available internal funds from retained earnings, then seek more costly
external finance. They would never pay dividends and issue new shares at the same time. Investors in mature firms must
be indifferent at the margin between receiving a dollar in dividends or receiving a capital gain on the reinvested dollar. If
the value of an additional dollar of investment in the firms is denoted by q, the investor must be indifferent between receiving
a dividend of $ I-valued at I-m, where m is the investor level tax on dividends-and a capital gain of q dollars-valued
at q(l-z), where z is the investor level accrual-equivalent tax rate on capital gains. Hence, I-m=q(l-z), so that q=(lm)/(I- z) < 1. Under certain assumptions, q is related to the ratio of the market value of the firm to the replacement cost
of the firm's capital stock. Hence, the dividend tax is capitalized in share values (i. e., decreasing m would increase q and
the value of the firm).
28. Under the traditional view, dividends offer nontax benefits to shareholders, so that tax-disfavored dividends are not a
cheaper source of funds for the firm than external finance. Using the notation of the previous note, q = 1, and investor level
dividend taxes are not capitalized in share values.
29. See Poterba, "Tax policy and corporate saving" (1987). The Tax Reform Act of 1986 is assumed in the analyses
discussed in this chapter to have increased the payout ratio from the 0.61 value reported by Poterba to 0.73 under current
law.
30. Statistical analysis is difficult because it is often difficult to isolate changes in tax rates on income from dividends that
occur independently of changes in tax rates on nondividend income (which would affect the required return on corporate
equity, share values, profits, and dividends in equilibrium).
31. Brittain (1966) analyzes data on U.S. corporations from 1920 through 1960. For the corporate sector as a whole, he finds
that in the short run (first year) a 1 percent increase in dividend tax rates would reduce the dividend payout ratio by 0.18
to 0.42 percent. As corporations gradually adjust to the new tax system, they respond more fully, and in the long run the
behavioral responses are larger, ranging from 0.61 to 1.02 percent. Brittain concludes that the dividend tax rate explains
dividend payout better than any of a variety of measures of the tax penalty on dividends relative to capital gains.
Feldstein (1970) examines the dividend payment behavior of British firms from 1953 through 1964, and finds that payout
decisions were sensitive to the tax penalty on dividends relative to capital gains. Feldstein finds that in the short run (first
year) a 1 percent increase in the tax penalty on dividends relative to capital gains (measured as the opportunity cost of
retentions in terms of foregone dividends) will reduce the dividend payout ratio by between 0.27 percent and 0.68 percent.
In the long run, Feldstein's estimates are close to 1.0.

Notes

238

King (1971, 1972) examines data on British corporations from 1949 through 1967. He finds behavioral responses that
are lower than Feldstein's by about one-half. However, Feldstein (1972) countered that King's estimates are biased downward
because of data problems, and maintains that the true response is closer to his own original estimates than to King's
estimates.
Poterba and Summers (1985) also examine data on British firms, using information through 1983. They find that
dividends are very sensitive to the tax penalty variable. They estimate that a 1 percent increase in dividend tax rates would
reduce dividend payout rates by 0.18 to 0.54 percent in the short run and by 1.03 to 2.6 percent in the long run.
Poterba, "Tax policy and corporate saving" (1987) provides estimates based on data for the United States for the period
1948 through 1986. Poterba estimates short-term responses in the dividend payout ratio with respect to the dividend tax
penalty ranging from 0.61 to 0.78 percent. In the long run, Poterba's elasticities range from 1.56 to 4.00 percent.
Another type of evidence comes from studies of changes in asset prices in response to taxes. Such studies attempt to
test whether investor level dividend taxes are capitalized in share prices. Poterba and Summers (1985) studied the reaction
of prices of British stocks to the announcement in 1970 that an integrated tax system would replace the double taxation of
dividends. They found no significant increase in stock prices, suggesting that dividend taxes were not capitalized into share
values.
32. This estimated sensitivity, in principle, could reflect investors' perceptions that dividend tax changes are temporary. Even
in the new view, a temporary decrease in dividend tax rates would increase dividend payout. Poterba and Summers (1985)
argue, however, that empirical evidence is consistent with an effect on payout of "permanent" dividend tax changes.

33. See Shoven (1987) and Poterba, "Tax policy and corporate saving" (1987).
34. The calculations follow Poterba (1987), and are based on tabulations of the COMPUSTAT Industrial and Research files.
35. In different contexts, see Lintner (1956), Easterbrook (1984), Jensen (1986), Gertler and Hubbard (1991).
36. See the discussion in Fazzari, Hubbard, and Petersen (1988) and Hubbard (1990).
37. Empirical evidence in support of the proposition that capital income taxes affect investment is more conclusive than for
the case of saving. Modern theoretical models of business fixed investment build on early work by Jorgenson (1963), which
demonstrated a link between capital spending and the cost of capital, which in tum depends in part on tax rates. Initial
empirical evidence by Hall and Jorgenson (1967) bolstered this view. Criticism of the Hall-Jorgenson approach by Eisner
and Nadiri (1968) and Eisner (1969) (see also later work by Chirinko and Eisner, 1983) centered on the Hall-Jorgenson
approach of combining output and cost of capital effects in a single term. In this work by Eisner and others, the cost of
capital effect in isolation was small. A significant effect of taxes on investment spending has been demonstrated in recent
models using a range of underlying theoretical approaches. See, for example, Summers (1981), Feldstein (1982), Feldstein
and Jun (1987), Fazzari, Hubbard, and Petersen (1988), and Auerbach and Hassett (1990, 1991).
38. See Shaven and Whalley (1984) for a discussion of computable general equilibrium models.
39. The assumptions underlying the models were made to conform to each other whenever possible. Common assumptions
include inflation rates (3.5 percent), asset holding periods (seven years), share of capital gains excluded from tax through
step up in basis at death (two-thirds), historical dividend-payout ratios (two-thirds of the real return), and historical debt
shares (40 percent for corporations, 34 percent for noncorporate enterprises, and 38 percent for owner occupied housing).
Each model generally characterizes the production technologies in a particular industry in a similar way, and where possible
the models assume consistent behavioral responses of dividend-payout ratios and debt to equity ratios to changes in taxes.
Only Federal taxes on capital income are taken into account in measuring investment incentives.
40. By taxing distributions out of tax-favored or foreign-taxed income, a compen3atory tax can significantly offset the
efficiency gains otherwise resulting from integration. In particular, had a compensatory tax been incorporated into the eBIr
prototype (rather than the investor level tax actually recommended), the decision to retain, rather than distribute, current
earnings would be as distorted by tax considerations as under current law.
41. The analysis of corporate borrowing in the model is based on Nadeau (1988). He estimates an elasticity of the fraction
of total external financing in the form of debt to the difference between the real rate of return required on equity and the
real interest rate of o. 224. The representation of corporate borrowing in the model is consistent with an elasticity of the debt

239

Notes

to asset ratio with respect to the tax advantage of debt of 0.3. Nadeau measures the tax advantage of debt as 1-[(1-t,)(1tJ/(I-tJ], where tdis the tax rate on debtholders, tc is the corporate tax rate, and fe is the effective tax rate on the real return
to equity (including the benefit from the preferential treatment of capital gains). Rangazas and Abdullah (1987) have
estimated that this elasticity is about 0.4 in the long run, somewhat larger than the behavioral response assumed in the model
used in this Report.
42. The gain to shareholders from a dollar distributed as a dividend relative to an additional dollar of retained earnings is
given by (l-m)/(l-z), where m is the tax rate on dividends and z is the accrual-equivalent tax rate on capital gains. The
model assumes an elasticity of the dividend payout ratio with respect to this measure of relative after-tax values of
approximately unity. This estimate is conservative. For example, Poterba (1987) estimated the long-run elasticity to be in
the range from 1.6 to 4.0, while Feldstein (1970) estimated long-run elasticities ranging from 0.85 to 1.33.
43. In all calculations, noncorporate business is assumed to be financed using 34 percent debt, and owner-occupied housing
using 38 percent debt. These calculations are based on information from Balance Sheets for the U.S. Economy, Board of
Governors of the Federal Reserve System, various issues.
44. In fact, because nominal interest payments are deductible, the effective marginal tax rate on debt-financed investments
is negative in these calculations.
45. These calculations assume that retentions are never distributed. Thus, they may overstate the difference between the
taxation of dividends and retentions. This assumption is probably appropriate for the calculations below, however, since
incentive effects in these calculations are based on a marginal expansion of the capital, stock. Hetained earnings used to
fmance such an expansion would be retained indefmitely.
46. In the scaled-tax-rate calculations, and compared to current law, all prototypes reduce slightly the overall average cost
of capital for the economy, and encourage additional savings and investment. The small reduction in the overall average cost
of capital is caused by the reduction in the premium that corporate investments must earn to compensate investors for taxinduced corporate financial distortions. The direct tax cost of investment has, by assumption, remained fixed at its current
law level. Since CBIT reduces fmancial distortions most significantly, it generates the largest reduction in the overall average
cost of capital. This effect is not the focus of the present analysis, however.
47. The incidence of the corporate income tax is discussed in detail in Section I3.G.
48. Mackie (1991) describes the technical details of the model outlined in this section. The model is based upon Fullerton
and Henderson (1989).
49. See, e.g., Gordon and Malkiel (1981), Fullerton and Gordon (1983), and Gertler and Hubbard (1990).
50. Even though in the scaled-tax-rate calculations the integration prototypes may leave constant the effective tax rate on
investment, they still might encourage capital formation by reducing tax-induced distortions in corporate fmancial policy.
Although small in an absolute sense, this effect may be large relative to the other gains brought on by the integration
prototypes. Nonetheless, the static, single period calculations reported in the tables do not incorporate such an effect.
51. We use a modified version of the Mutual Production Model introduced by Gravelle and Kotlikoff (1989).
52. Corporate financial behavior in the MPM is based on CES functional forms with an elasticity of dividend payout ratio
with respect to the tax penalty on dividends relative to capital gains equal to -3.0, and an dasticity of the leverage ratio with
respect to the tax advantage of debt relative to equity equal to 0.3. Thus, the fmancial behavior in the MPM is consistent
with, but not identical to, that assumed in the augmented Harberger model described earlier. For technical details of the
MPM, see Gravelle (1991).
53. As statutory tax rates rise to make the distribution-related prototype revenue neutral, the tax advantage of debt relative
to equity also rises because the higher tax rates increase (1) the value of deducting nominal interest, and (2) the tax rate on
purely inflationary capital gains. At the set of tax rates needed for revenue neutrality, these two effects, combined with a
relatively large distortion in dividend policy, are sufficient to counteract the effect of the dividend exclusion or credit. As
a result, relative to current law the tax benefit to debt rises, and corporations actually increase slightly their use of debt.
54. The portfolio allocation model is described in Galper, Lucke, and Toder (1988).

Notes

240

55. Households hold debt and corporate equity, directly and indirectly, through certain pension holdings. The household
allocations of debt and corporate equity in Table 13.9 reflect direct holdings. Pension holdings of debt and corporate equity
are shown separately.
56. Household wealth includes small net holdings of foreign equity. As a result, total wealth slightly exceeds the value of
total physical capital, so shares can differ between the top and middle panels of Table 13.9.
57. Though not shown, the PA model also simulates changes in portfolio shares across income groups. The shareholder
allocation, imputation credit, and CBIT prototypes shift stock ownership from high-income to low-income groups; the
dividend exclusion prototype shifts stock holdings to higher-income groups. In all cases, the shifts are quantitatively small.
Larger cross-household shifts in taxable debt accompany the prototypes, especially CBIT. Broadly speaking, all of the
prototypes reduce the share of total debt held by low-income groups, while raising the share held by middle- and high-income
groups.
58. Note that this can result simply because existing businesses in the noncorporate sector decide to incorporate. It does not
necessarily imply a change in ownership of assets.
59. Both the augmented Harberger model (AH) and MPM simulations suggest that each integration prototype would improve
economic welfare. The models also suggest possible gains at both real and financial margins. Nonetheless, there are
substantial differences between these two models' results. Perhaps most noticeably, the MPM produces much larger shifts
in physical capital and in economic welfare than does the AH model. There are some key differences in the models'
predictions about corporate financial policy, real capital shifts, and welfare changes, as described below.
Changes in corporate financial policy. For a given prototype and fmancing mechanism, the two models predict very
similar changes in the corporate dividend payout ratio. In the lump-sum calculations, furthermore, the two models predict
fairly similar changes in the corporate leverage ratio. In contrast, with the scaled-tax-rate replacement mechanism, the two
models predict somewhat different changes in the corporate leverage ratio, especially under the two distribution-related
prototypes. Such differences can be traced to the fact that the two models (1) start with somewhat different statutory rates,
(2) use slightly different behavioral responses in estimating corporate financial behavior, and (iii) have different equal-taxyield requirements.
Changes in capital allocation. The MPM generally produces larger shifts in physical capital than does the AH model.
This difference reflects in part the MPMs greater scope for substitutability between corporate and noncorporate resources.
The greater substitutability stems from two sources: (1) a much larger implied substitution elasticity between corporate and
noncorporate business in each industry; and (2) a corporate-noncorporate choice in the provision of rental housing that is
not considered in the AH model.
Changes in welfare from improved consumption. The MPM predicts larger gains from imprOVed consumption choices.
This difference is due principally to the MPM's greater shifts in capital (and other resources) discussed above. The greater
substitution between the corporate and noncorporate form in the MPM means that, because investors are quite sensitive to
tax differences, current law does more to distort the allocation of real resources in that model than in the AH model.
Consequently, relieving the tax distortion produces a larger gain in the MPM than in the AH model.
Changes in welfare from corporate financial policy. The MPM generally produces larger changes in welfare from
changes in corporate debt and dividend policy. Some differences between the models' welfare results reflect di fferences in
the predicted changes in the leverage and dividend payout ratios, as discussed above. In addition, for each prototype the
MPM has a larger fraction of the economy's stock of capital allocated to the corporate sector under current law than does
the AH model. Thus, the same per unit fmancial distortion would produce a larger absolute (i.e., dollar) loss in the MPM
than in the AH model.
60. Our gains also are on the same order of magnitude as those estimated for the 1986 Act. See, e.g., Fullerton, Henderson,
and Mackie (1987).
61. See Harberger (1966), Shoven (1976), and Fullerton, et al. (1981).
62. See Fullerton (1984).
63. See Fullerton and Henderson (1989).

241

Notes

64. Others also have emphasized the role of debt ftnance and capital gains taxes in reducing the size of the corporate tax
wedge, and so reducing the efftciency cost of the corporate tax system. See, e.g., Gordon and Malkiel (1981) and Stiglitz
(1973).
65. The important differences are three. First, in this Report, only Federal income taxes distort investment decisions, while
in Fullerton and Henderson, state and local income and property taxes also act to distort investment decisions. (All other
things constant, this would tend to make the welfare gains from integration in Fullerton and Henderson larger than those in
this Report.) Second, Fullerton and Henderson's calculations are based on the new view of dividend taxes while this Report
uses the traditional view. (All else constant, this would tend to make the welfare gains from integration in Fullerton and
Henderson smaller than those in this Report.) Finally, in this Report the model has been augmented to account for taxinduced ftnancial distortions. (This would tend to make the welfare gains from integration, even those due to real resource
allocation alone, smaller in Fullerton and Henderson than those in this Report.)
66. Fullerton and Gordon (1983), for example, estimate that eliminating the tax incentive for corporate debt would generate
gains equivalent to about 0.8 percent of consumption, while Gordon and Malkiel (1981) estimate that it would generate gains
of about 0.4 percent consumption.
67. Neither Gravelle (1989) nor Fullerton, Henderson and Mackie (1987) considered the welfare costs of distortions of
corporate financial decisions.
68. Harberger (1977 and 1980) argues that evidence on rates of return on capital is consistent with capital mobility. On the
other hand, Feldstein and Horioka (1980) found that domestic saving and investment rates moved too closely together in the
1960s and 1970s to be consistent with capital mobility. Feldstein and Horioka reasoned that if capital were perfectly mobile
internationally, national savings rates should be independent of national investment rates. Capital would flow to wherever
it received the highest return, and so returns would be equalized globally. Therefore, if saving increased in a country, rather
than reducing interest rates below the global interest rate and thereby increasing investment at home, the additional saving
would flow abroad. However, examining data from OECD countries, they found that, over long periods, national saving
and investment rates were highly correlated. In a regression of national investment rates on national saving rates, the
estimated coefficient on saving was statistically significant and close to unity. They interpreted this to mean there was very
little international capital mobility, so that a one dollar increment to national saving produced almost a one dollar increment
to national investment.
Since Feldstein and Horioka, there has been a series of papers examining the saving-investment relationship in time series
and cross-section studies, generally with the intent of overturning their result. The result has, however, until recently, held
up remarkably well for data from many countries over a long period. Recently, however, studies by Feldstein and Bacchetta
(1989) and Frankel (1990) indicate that the close correlation between saving and investment may have broken down during
the 1980s. Using data from the OECD countries, Feldstein and Bacchetta found that the coefficient on saving in a savinginvestment regression is markedly lower for the 1980-1986 period than for prior years. Frankel used a long time series of
U.S. data and found that the relationship between saving and investment held up well before 1980, but for the 1980-1987
period the estimated coefftcient on saving is relatively small and statistically insignificant.
Several authors have pointed out that national savings and investment rates are both endogenous variables. Hence if there
are exogenous variables that are correlated with both saving and investment, one could find a significant correlation between
the two even in the presence of perfect capital mobility. See, e.g., Obstfeld (1986), Summers (1986), and Frankel (1986).
Feldstein and Bacchetta (1989) rejected most of these explanations.
More recently, researchers have studied impacts of domestic capital market imperfections on capital flows. For example,
Gertler and Rogoff (1990) present a model in which capital is perfectly mobile internationally, but capital market
imperfections can lead domestic saving to be correlated with domestic physical investment. In their model, there is a domestic
sector consisting of risky projects. There also is an international market for a riskless asset which yields a world rate of
return. Foreigners can invest funds directly in the risky domestic projects, but because of asymmetric information they do
not know how much of their funds are actually used in the project and how much reinvested in the international capital
market. The probability of the project's success depends on how much money is actually invested in it. There is
underinvestment of foreign funds in the risky domestic sector, but foreign investment increases with increased domestic
investment in the risky sector. If saving increases, thereby increasing investment of domestic funds in the risky sector,
foreigners will be willing to contribute more funds too. This may cause saving and investment to be correlated. While this
model is stylized, it does point out that international mobility of capital in one market (for low-risk assets) need not imply
that returns are equated internationally in markets for risky assets.

Notes

242

69. Most of the empirical evidence pertains to debt secunhes. When looking at secuntles (as opposed to saving and
investment rates), the appropriate test is whether returns are equalized across national boundaries. To implement this test,
one needs to define (and measure) the relevant returns that should be equalized. This is not always easy.
Mishkin, "Are Real Interest Rates Equal Across Countries" (1984), Mishkin, "The Real Interest Rate" (1984), and Mark
(1985) found evidence against real interest parity. In a less direct test, Barro and Sala-i-Martin (1990) estimated a system
of country real interest rate and investment equations derived from a macroeconomic model. They found some evidence that
global factors, e.g., global stock returns, are more important in determining a country's real interest rate than country
specific factors. Of course, real interest parity may not hold even in the presence of perfect capital mobility if there is an
expected change in the real exchange rate or an exchange rate risk premium. A test for capital mobility that allows for the
existence of expected changes in the exchange rate or exchange rate risk premia is whether covered interest parity holds.
Frankel and MacArthur (1988) and Frankel (1990) present evidence that covered interest differentials have narrowed over
time, and that they are currently small for major industrial countries.
The covered interest differential measures only the extent of institutional barriers and market imperfections that impede
capital flows. It does not measure the substitutability of domestic and foreign assets in investors' portfolios. The uncovered
interest differential is a better indicator of capital mobility capturing asset substitutability. The difference between the
uncovered interest differential and the covered interest differential is the exchange rate' risk premium, the size of which
provides a measure of the substitutability of assets across currencies. Froot and Frankel (1989), Giovannini and Jorion
(1987), and others have rejected uncovered interest parity, suggesting the presence of a risk premium. Frankel (1990)
presents some evidence that much of these differences is accounted for by expected changes in real exchange rates rather
than exchange rate risk premia.
To summarize, there appears to be substantial integration in asset markets for short-term debt. Of course, even if there
is a high degree of capital mobility in these markets, imperfect substitution between these markets and other asset markets
(for equity or long-term debt) could still be consistent with weak overall integration of capital markets.
Tests of equity market integration in the capital asset pricing model have generally rejected international integration. See,
e.g., Stehle (1977) and Jorion and Schwartz (1986). This may be due in part to the sample period (which does not include
much of the 1980s). French and Poterba (1991) stress informational problems as an explanation for imperfect international
diversification in equity markets.
70. See Mutti and Grubert (1985) for details.
71. The model assumes not only that debt capital is more internationally mobile than equity capital, but also that debt is more
important in cross-holdings of assets. In the model's calibration, 66 percent of foreign holdings of U. S. assets are in the form
of debt, while 60 percent of U.S. holdings abroad are in the form of debt.
72. This is true even for shareholders that are tax-exempt institutions. Taxes borne by pension and life insurance funds reduce
the incomes of their beneficiaries, and taxes falling on charitable and educational institutions reduce the services they can
provide.
73. See, e.g., Harberger (1962), Shoven and Whalley (1972), Shoven (1976), Pechman (1987), and Gravelle and Kotlikoff
( 1989).
74. See Harberger (1962).
75. See Ebrill and Hartman (1982) and Gravelle and Kotlikoff (1989).
76. See, e. g., Stiglitz (1973). The risk of bankruptcy may constrain the use of debt to finance the marginal investment, and
that risk plays an independent role in the effect of the corporate tax. See, e.g., Gordon and Malkiel (1981).
77. See, e.g., Harberger (1983), Mutti and Grubert (1985), and Pechman (1987).
78. See Young (1988), Murthy (1989), and Gravelle (1991).
79. This possibility seems likely for the United States since the corporate tax is not a residence-based tax. American
multinationals pay taxes on repatriated income to the United States in excess of foreign taxes paid. The U.S. corporate tax,
in fact, is both residence-based and source-based, since taxes on earnings retained and reinvested abroad can be deferred.

243

Notes

80. Other assumptions have sometimes been used by other analysts. While Pechman (1987) allocated the corporate income
tax to all capital income, Pechman and Okner (1974) and Pechman (1985) used five different assumptions to allocate the
corporate income tax: (1) to dividends, (2) to property income in general, (3) half to dividends and half to property income
in general, (4) one-half to dividends, one-fourth to consumption, and one-fourth to employee compensation, and (5) half to
property income in general and half to consumption. In its original (1987) study of tax burdens and in the (1988) update,
the Congressional Budget Office allocated the corporate tax burden in two ways: (I) entirely to capital income and (2) half
to capital income and half to labor income. The Joint Committee on Taxation has not attempted to allocate the burden of
corporate income tax to individuals.
The assumptions correspond to those conventionally employed in contemporary analyses of the distributional implications
of tax changes. Early analyses by the Department of the Treasury in the 1930s and 1940s allocated the burden of the
corporate income tax by income class on the basis of dividends or stockholdings. More recently, Department of the Treasury
analyses of the distribution by income class of federal income taxes have consistently allocated the burden of the corporate
tax to owners of capital. In Blueprints, the corporate income tax was allocated on the basis of total capital income. Similarly,
in constructing Family Economic Income, the Department of the Treasury has allocated the corporate tax to families on the
basis of their total capital income.
81. The tax rates reflect the burden of the corporate tax borne by foreign investors and tax-exempt institutions, other than
pensions, through their ownership of V.S. capital. The portion of the corporate tax falling on assets owned by pension funds
is allocated to the individuals with rights to the pension reserves.
82. Family economic income is constructed by adding to adjusted gross income: unreported and underreported income; IRA
and Keogh deductions; nontaxable transfer payments such as Social Security and AFDC; employer-provided fringe benefits;
inside buildup on pensions, life insurance, and IRA and Keogh accounts; tax-exempt interest; and imputed rent on owneroccupied housing. Capital gains are computed on an accrual basis, adjusted for inflation to the extent reliable data allow.
Inflationary losses of lenders are subtracted and gains of borrowers are added. The economic incomes of all members of a
family unit are summed to produce the family economic income used in the distributional analysis.
83. The rate of inflation is assumed to be 3.5 percent per annum.
84. The revenue estimates have assumed an average excludability rate of 56 percent, implying that 56 percent of the
distributions of corporations will be excluded from income have tax credits attached that can be used by the recipient of the
distribution to offset taxes. This rate consists of a base rate of 51 percent and an additional 5 percent representing carryovers
of excess amounts in Earnings Distribution Accounts from prior years to exclude dividends in the current year.
The low average excludability rate is accounted for by the fact that many corporations that distributed income to
shareholders have paid no (or little) tax on that income. That is, much of the income distributed represents preference or
foreign source income not taxed at the corporate level. Moreover, many corporations whose income is taxed more fully have
low dividend payout ratios. The assumed excludability rate of 56 percent is based on Department of the Treasury
calculations.
85. The EDA is calculated as taxes after credits mUltiplied by (1-tc)/tc' where tc is the corporate tax rate, to gross up the
amount of income available to pay excludable dividends. For example, for income of $100 and taxes paid of $34, $66 is
available to pay dividends. The EDA also is $66 [(0.66/0.34) x$34].
86. Thus, individuals cannot exclude dividends from foreign source income except to the extent that V.S. tax is paid.

APPENDICES
Appendix A
1. Treas. Reg. § 301. 7701-2(a)(I). Two characteristics, associates and an objective to carry on business and divide the
profits, are common to partnerships and corporations and are therefore not material in distinguishing between partnerships
and corporations.
2. IRC § 7704.

3. IRe § 851 et seq.

Notes

244

4. IRC § 856 et seq.
5. IRC § 860A et seq.
6. Exceptions include: (1) interest on purported debt that is properly viewed as equity (see, e.g., IRC § 163(e)(5», (2)
interest on debt used to finance certain tax-favored income (see, e.g., § 265(a)(2», and (3) interest that must be capitalized
because the debt relates to the production of future income (see, e.g., IRC § 263A(f».
7. The Code treats a distribution as a dividend to the extent of current and accumulated earnings and profits of the
distributing corporation. Distributions that exceed earnings and profits are treated as a tax-free return of basis to the extent
of the shareholder's basis in the stock. To the extent that the distributions exceed basis, they are generally treated as capital
gains. IRC § 301(c).
8. Capital gains of individuals are subject to a maximum tax rate of 28 percent. IRC § 1(h).
9. A domestic corporation also is entitled to a dividends received deduction (in the percentage specified in IRC § 243) for
the U.S. source portion of dividends received from a foreign corporation that is at least 10 percent owned by the U.S.
corporation. The deduction is 100 percent for a wholly owned subsidiary whose income is all effectively connected with a
U.S. trade or business. IRC § 245.
10. IRC § 385(b).
II. The data reflect corporate taxes at both the central government and local levels. Comparisons of corporate tax receipts
for central governments only would be misleading because some countries have much greater corporate taxation at the local
level than others. Organisation for Economic Co-operation and Development (1991), Table 13, p. 78.

Appendix B
1. We believe that the descriptions that follow are complete as of December 1991. They are based in part on secondary
sources. We are grateful to those government officials, academics, and practitioners who gave us their comments.
2. The amount of the imputation credit is [F/(1- .39)] X .39, where F equals the amount of the distribution from the franking
account.

n

3. The amount added to the franking account each year is (61139 X + D, where T is the total Australian tax paid by the
corporation in the relevant period and D is the amount of franked dividends received from other resident corporations that
period.
4. For example, an individual shareholder owns a share with a paid-up value of AU$1.00 and a market value of AU$2.50.
The shareholder's basis in the share is AU$2.00. The corporation buys the share (and has taxable income sufficient to frank
fully all dividends paid that year). If the buyback is off-market, then the difference between AU$2.50 (amount paid) and
AU$l.OO (paid up value) is a dividend (AU$1.50). That part of the purchase price not treated as a dividend (the paid up
value of AU$l.OO) is consideration received in the sale. Thus, the shareholder also has a capital loss of AU$l.OO (AU$l.OO
paid up value minus AU$2.00 basis). If the buyback is instead on-market, the total purchase price (AU$2.50) is consideration
in the sale, and the shareholder has a capital gain of AU$O.50 (AU$2.50 minus AU$2.00 basis). The corporation, however,
must debit its franking account by AU$I.50, the amount that would have been a dividend if the purchase were off-market.
5. The required franking amount equals: CD X [RFS/(TD+CFD+SDD)], where CD is the current dividend and RFS is the
franking surplus. RFS is reduced by any unpaid dividends with an earlier reckoning day. (The reckoning day is normally
the day that the dividend is paid, but sometimes dividends that are part of the same distribution are not paid on the same day.
In that case the reckoning day is the day that the first of those dividends is paid.) TD is the total amount of dividends paid
or to be paid on the same class of shares and under the same resolution as the current dividend. CFD is the amount of the
committed future dividends (not in TD) at the beginning of the reckoning day for the current dividend. SDD (same day
dividends) have the same reckoning day but are paid or to be paid under a different resolution or under the same resolution
on a different class of shares.
6. Thus, the corporation pays a franking deficit tax equal to the franking deficit grossed-up at the corporate rate and then
multiplied by that rate: [FD/(l- .39)] X .39, where FD equals the amount of the franking deficit.

245

Notes

7. Implementation of an accompanying foreign investment fund regime recently was postponed to July 1, 1992. This regime
is similar in purpose, though not in details, to the U.S. PFIC rules of IRC §§ 1291-1297.
8. For example, if a shareholder receives a taxable dividend of $100, he includes $125 in income and receives a Federal tax
credit of $16.75. Assuming the provincial rate is 50 percent of the Federal liability, the $16.75 Federal credit reduces
provincial tax liability by $8.38 ($16.7512). The total tax saved as a result of the credit is $25.13.
9. The following table illustrates the Canadian system with respect to the business income of a Canadian corporation. (This
analysis does not deal with the investment income of a Canadian private corporation, which is subject to a somewhat different
regime.) The table assumes, for purposes of the provincial tax, that the dividend paying corporation is both resident in, and
doing business in, Ontario, and that the individual Canadian shareholder also is resident in Ontario. Three cases are shown:
a normal Canadian corporation, subject to a 28 percent Federal tax plus a 3 percent surtax and a 15.5 percent Ontario tax;
a Canadian manufacturing company, subject to a 23 percent Federal tax plus a 3 percent surtax and a 14.5 percent Ontario
tax; and a small business corporation subject to a 12 percent Federal tax on its business income (not exceeding $200,000
per year) plus a 3 percent surtax and a 10 percent Ontario tax. The shareholder is assumed to be subject to Federal income
tax at the top rate of 29 percent (before credit) plus a 5 percent surtax, and an Ontario tax equal to 53 percent of the Federal
tax (after shareholder credit). For simplicity, these rates do not reflect the Federal and provincial surtax on high-income
individuals.
Normal
Corporation

Manufacturing
Corporation

Small Business
Corporation

100.00

100.00

100.00

28.00

23.00

12.00

0.84

0.84

0.84

Ontario tax

15.50

14.50

10.00

Total Federal and provincial tax

44.34

38.34

22.84

Maximum distribution to shareholder

55.66

61.66

77.16

25 percent gross-up

13.92

15.42

19.29

Taxable income of shareholder

69.58

77.08

96.45

Federal pre-shareholder credit income tax

20.18

22.35

22.97

9.28

10.28

12.86

10.90

12.07

15.11

Federal surtax (5 %)

0.55

0.60

0.76

Ontario tax
(53 % of pre-surtax, post-credit, Federal tax)

5.78

6.40

8.01

Total Federal and provincial shareholder tax

17.23

19.07

23.88

Total value of credit to shareholder
(Federal credit plus .53 % of Federal credit)

14.20

15.73

19.68

102.0%

102.0%

102.0%

32.0%

41.0%

86.2%

Net income of Canadian corporation
Federal tax
Federal surtax (3 %)

Dividend received credit (67 % of gross-up)
Federal tax after shareholder credit

Value of credit as a percentage of gross-up
Credit as a percentage of Federal and provincial
corporate tax

10. These amounts are indexed for inflation.
11. Assume, for example, that a regular corporation earns $25 of preference income and $100 of taxable income. Assume,
in addition, that a regular corporation is subject to Federal tax at a net rate of 28 percent (i.e., after the provincial abatement)
and that a shareholder is subject to Federal tax at a rate of 29 percent (both assumptions disregard surtaxes). Taking into
account only Federal tax, the corporation pays $28 of tax. When net income of $97 is distributed, the shareholder includes
$121.25 in income ($97 X 125 percent), has tax liability of $35.16 and is entitled to a credit of $16.25, reducing shareholder
tax to $18.91. The total Federal tax burden on $125 of economic income is thus $46.91 ($28+$18.91), or 47 percent. Thus,
the income has been taxed at a rate greater than either the shareholder or the corporate rate. If, on the other hand, the
corporation had earned $125 of preference income and $100 of taxable income, the total Federal tax burden on $225 of
economic income would be $46.91, or 21 percent.

Notes

246

12. Special rules apply with respect to dividends on redeemable preference shares.
13. When the avoir fiscal was enacted in 1965, the French corporate tax rate on distributed (and retained) profits was 50
percent. The 50 percent avoir fiscal percentage was chosen in order to provide shareholders with a partial imputation credit
equal to 50 percent of the taxes actually paid by a corporation on distributed profits. When the corporate tax rate was reduced
to 42 percent in 1988, however, the avoir fiscal percentage also was not reduced to preserve the 50 percent relationship
between the avoir fiscal and actual corporate tax payments. Instead, the avoir fiscal percentage was maintained at 50 percent
as a means of introducing a greater degree of integration. As a result, the avoir fiscal represented a greater percentage (69
percent) of actual corporate tax payments on distributed profits. With the further reduction of the tax rate on distributed
profits to 34 percent for 1992, the avoir fiscal will represent almost the entire amount of corporate level tax paid on
distributed profits.
14. Net operating losses generally may be carried forward for 5 years, although net operating losses attributable to
depreciation may be carried forward indefinitely. If a net operating loss fully offsets taxable income in a carryover year, a
dividend distribution out of carryover year income will incur the precompte mobilier. A corporation may elect, however,
to spread a net operating loss carryover over the 5 year carryover period in order to leave some fully-taxed income in each
year of the carryover period from which to make dividend distributions.
Alternatively, a corporation may elect to carry back over a 3 year period a tax credit calculated by applying to the
amount of the loss the standard corporate tax rate in effect at the end of the loss year. The tax credit may be used to offset
income tax liability on undistributed fully-taxed profits realized during the 3 year carryback period. Any excess credit
remaining thereafter is refunded.
Net operating losses cannot be carried back to offset any portion of the prior years' income for which tax liability was
satisfied using avoir fiscal or other tax credits.
IS. Rather than separating income into fully-taxed and untaxed baskets, France effectively relies on the ability of French
corporations to avoid the precompte mobilier out of retained earnings with respect to income taxed at rates less than 34
percent. For example, assume that a corporation has Fl000 of gross income, FSOO of which is taxable at 34 percent and
FSOO of which effectively is taxable at 19 percent, e. g., a dividend from a foreign corporation resident in a treaty country
paid to a French nonparent corporation that is subject to a IS percent foreign withholding tax. If the corporation distributes
its entire after-tax income of F735, this amount will be subdivided into two parts: a dividend of F330, which has borne
regular corporate tax, and a dividend of F40S, which has not borne corporate tax. The precompte mobilier will be imposed
on F40S at a rale of 50 percent, reSUlting in an additional tax liability of F202.50. Thus, the total tax liability of the
corporation will be F467.50, and the corporation will be required to pay the additional F202.50 liability out of retained
earnings.
As a practical matter, a corporation wishing to distribute tax-sheltered income will reduce the amount of its dividend
so it can pay its precompte mobilier liability out of current after-tax income. In the above example, the corporation would
pay a dividend of F600, equal to F330 (income that has borne regular 34 percent corporate tax) plus F270 (income that is
subject to a precompte mobilier of 50 percent). The corporation's total tax liability would be F400, equal to F265 regular
corporate tax plus F135 precompte mobilier.
16. The participation exemption results in an effective tax rate of (1) 2.55 percent on the gross amount of a dividend
(including the amount of the avoir fiscal) received from a 10 percent-owned French subsidiary, and (2) 1.70 percent on the
gross amount of a dividend (including the amount of a credit for foreign withholding tax) received from a 10 percent-owned
subsidiary in a treaty country.
17. In some circumstances, a French company may elect to be taxed on all foreign branch income. In such cases, the
precompte mobilier is not imposed upon distribution of the foreign branch income.
18. The purpose of the special rules is to avoid an effective tax surcharge that arose under pre-1990 law. Dividends received
by a French holding company from a foreign subsidiary are exempt from French income tax in the hands of the holding
company by virtue of the participation exemption. Prior to 1990, however, the foreign source dividend income was subject
to the precompte mobilier upon redistribution by the holding company. Payment of the precompte mobilier by the holding
company entitled the recipient to claim an avoir fiscal credit with respect to the redistribution. If the recipient was a French
10 percent shareholder of the holding company, however, the participation exemption would exempt the income again in the
hands of the 10 percent shareholder. Thus, the avoir fiscal was not needed to offset income tax liability of the 10 percent
shareholder with respect to the dividend income. Under pre-1990 law, moreover, the avoir fiscal could not be used to offset
income tax liability of the 10 percent shareholder with respect to other types of income. Pre-1990 law did permit the 10

247

Notes

percent shareholder to use the avoir fiscal to offset any precompte mobilier liability that it might incur upon a subsequent
distribution of preference income; if the 10 percent shareholder did not have sufficient preference income however, all or
a portion of the avoir fiscal (which had been "paid for" by the French holding company) was lost.
19. The amount of the excess tax equals the amount distributed out ofEK 50 (or EK 56), grossed-up to its pre-tax equivalent,
and then mUltiplied by the difference between 50 percent (or 56 percent) and 36 percent (the distribution rate). Accordingly,
if D equals distributions out of EK 50 (or EK S6), the corporation receives a refund of D/.50 x .14 (or D/.44 x .20). For
example, if a corporation earns DM100 and pays tax of DMSO, it will have DMSO in its EK SO account. If it then
redistributes DMSO out of EK SO, the corporation will receive it refund equal to DM14 (DMSO/.SO x .14).
20. The following table illustrates the application of the German split rate and imputation credit systems. For simplicity, the

table ignores any surtaxes.
Income before taxes
Tentative corporate tax
Decrease in corporate tax on full distribution
Amount available for distribution
Withholding tax (2S percent)
Shareholder includes in income
Cash dividend
Withholding tax credit
Imputation credit
Shareholder tax liability (S3 percent rate)
Shareholder credit
Withholding tax credit
Imputation credit

DMlOO.OO
DMSO.OO
DMI4.00
DM64.00
DMI6.00
DMlOO.OO
DM48.00
DMI6.00
DM36.00
DM100.00
DMS3.00
DMS2.00
DM16.00
DM36.00
DMS2.00

Net amount due

DMl.OO

21. The following equation converts pre-tax income subject to tax at some non-EK rate into equivalent amounts of pre-tax
income subject to tax at the distribution rate (36 percent) and either the statutory rate (SO percent) or the zero rate:
.36X+(.S or O)x(Y-X) = T, where Y equals the total amount of pre-tax income (known) subject to some non-EK rate,
X equals pre-tax income subject to the distribution rate, (Y - X) equals pre-tax income subject to either the statutory rate or
zero rate, and T equals the amount of tax paid with respect to Y (known). Because X and (Y - X) must be positive, the
effective tax rate, T/Y, determines whether the equation must contain the statutory rate or zero rate (and whether the residual
amount of income is ultimately converted into EK SO or EK 0).

The following equations convert the pre-tax amounts, X and (Y - X), into their after-tax EK amounts:
EK 36 = (1-.36)XX
EK SO (ifT/Y > .36) = (l-.SO)x(Y-X)
EK 0 (if T/Y < .36) = Y-X
22. Specifically, the calculation converts the DM100 into DM71.4 of income subject to the 36 percent distribution rate (.36+
.SX(DMlOO-X) = DM40) and the remainder, DM28.6, into income subject to the SO percent statutory rate (DM100DM71.4 = DM28.6). This translates into available net equity of DM4S.7 in the EK 36 category (.64 xDM71.4) and
DM14.3 in the EK SO category (.SOxDM28.6).
23. Specifically, the calculation converts the DMlOO into DM69.4 of income subject to the 36 percent distribution rate (.36X
= DM2S) and the remainder, DM30.6, into income subject to the zero rate (DMlOO-DM69.4 = DM30.6». This translates
into available net equity of DM44.4 in the EK 36 category (.64xDM69.4) and DM30.6 in the EK 0 category (DMlOODM69.4).
24. The rules for carrybacks and carry forwards of net operating losses are designed to prevent the refund of an amount of
tax that, by virtue of the imputation credit, has already been used to offset shareholder taxes. In summary, when a German
Corporation suffers a net operating loss for a year, it first enters the full amount of the loss as a negative adjustment to its

Notes

248

EK 02 account. The corporation may then carry back the loss for two years and (to the extent the loss is not absorbed in
these years) may carry forward the loss indefinitely.
With respect to carrybacks, the loss may be deducted in the earlier year, and generate a refund, only to the extent of
taxable income in that year less the sum of (1) any distributions in that year and (2) the distribution tax (36 percent) on such
distributions. In effect, a carryback deduction is only allowed against taxable income if the tax. on such income has not
already been returned to shareholders by way of credit.

If the NOL is not absorbed through carrybacks, it is carried forward and deducted in later years. As the loss is deducted
(and is thereby automatically reflected in the EK 50 account), it is credited against the original negative adjustment to the
EK 02 account.
25. All German enterprises (including foreign corporations with permanent establishments in Germany) also are subject to
the municipal "trade tax." This tax has both income tax and capital tax components. The basic trade tax rates are set by the
Federal Government, but the local governments (which collect the tax) have considerable discretion to increase them. The
income tax component is typically 15 percent. The trade tax is deductible in computing a corporation's normal tax liability.
The trade tax is not taken into account in the examples in this summary.
26. Tax is always withheld on dividends at the statutory 25 percent rate at the time of distribution (except as noted below).
Shareholders entitled to reduced withholding under a treaty must apply the German tax authorities for a refund of the excess
withholding. This rule applies even to publicly traded shares.
Some treaties contain an anti-avoidance rule designed to discourage corporations from distributing profits to nonresident
shareholders who reinvest these profits in the same corporation in order to gain the benefit of the lower distribution rate on
what are, in effect, retained profits. Such distributions are subject to a higher withholding tax than normal distributions. (The
1954 U.S.-Germany treaty had such a provision, but it was unilaterally waived by Germany in 1981.)
27. The following example illustrates the treatment of foreign source income and foreign stockholders. Assume a German
corporation has two foreign branches, the first in a treaty country (Country 1) and the second in a nontreaty country (Country
2). The corporation has DM 100 of German profits, DM 100 of Country 1 profits, and DMlOO of Country 2 profits (all pretax). The German profits are taxed at the statutory rate of 50 percent. The Country 1 profits are taxed in Country 1 at a rate
of 25 percent and are exempt in Germany (under the Business Profits and Double Taxation articles of the treaty). The
Country 2 profits are taxed in Country 2 at a rate of 30 percent and are subject to tax in Germany, but the German tax is
reduced by a foreign tax credit. During the next year (when the corporation has no profits anywhere), all of the prior year
profits are distributed to a foreign shareholder (who enjoys no treaty benefits).
The German profits of DM100 produced equity of DM50 in the EK 50 account. When these profits are distributed, the
corporation receives a refund of DM 14, also distributed to the foreign shareholder. The distribution of DM64 is subject to
25 percent withholding of DMI6. The foreign shareholder receives no imputation credit with respect to this distribution.
The Country 1 profits of DMlOO produced equity of DM75 in the EK 01 account. When this equity is distributed, it
is subject to the 36 percent distribution tax (DM27), but the tax is credited and refunded to the foreign shareholder. The
entire distribution (DM75-DM27+DM27) is subject to 25 percent withholding (DM18.75).
The Country 2 profits of DM 100 were reduced by DM30 of Country 2 tax, and then by an additional DM20 of German
tax (at the statutory rate of 50 percent after the foreign tax credit). In allocating this income to EK accounts, the corporation
is considered to have paid tax. of DM20 on profits of DM70 (an overall tax rate of 28.6 percent). Specifically, the
corporation is treated as having paid a 36 percent tax on DM55.6 and a 0 percent tax on DMI4.4. This produced equity of
DM35.6 in EK 36 (55.6-(55.6 x36%» and DMI4.4 in EK 01. When the profits are distributed, the distribution out ofEK
36 is not subject to any further tax. and produces a refunded credit of 36/64, or DM20. The distribution out of EK 01 is
subject to the 36 percent distribution tax, which is refunded. This results in a distribution, including refunds, of DM70
(DM35.6+ DM20+ DMI4.4- DM5.2+ DM5.2 = DM70). The total distribution is subject to statutory withholding of 25
percent (DM 17.5).
The treatment of pre-1977 profits is illustrated by the following. Assume the corporation in the above example had only
DM 100 of German profits, which were earned in 1976 and were subject to a tax of 56 percent at that time. The net profits
of DM44 were placed in EK 03 in 1977, when the integration system was implemented. When these profits are distributed
to a foreign shareholder in 1990, they are subject to a distribution tax of 36 percent (DM15.8), which is credited and
refunded to the shareholder, producing a total distribution of DM44. This total distribution is subject to statutory withholding
of25 percent (DMIl).

249

Notes

28. The following example illustrates the mechanics of New Zealand's credit system. A corporation earns income of
NZ$I00, of which NZ$60 is taxable, and the tax is NZ$19.80 (at a 33 percent rate). The corporation distributes the
remaining NZ$81.20 to its shareholders. The payment of tax of NZ$19.80 gives rise to a credit to the ICA in the same
amount. The maximum amount of credits that can be allocated to the distribution is NZ$33.99 (NZ$60 x .33/(1- .33».
However, the corporation only allocates a credit of NZ$19.80 to the distribution to avoid having a negative ICA and
incurring penalties. Not taking into account the refundable resident withholding tax, the shareholder would include NZ$loo
in income (the cash distribution plus the attached credits), and have tax liability of NZ$33 and a credit of NZ$19.80. As a
result, the shareholder has additional tax liability of NZ$13.20.
29. Until March 31, 1991, the CFC regime applied only to a transitional list oflow-tax countries (the "black list" countries).
As of April 1, 1991, the new regime applies in full to a CFC resident in any country other than Australia, the United States,
the United Kingdom, Japan, France, Germany or Canada (the "grey list" countries). The CFC rules apply to investors in
a CFC resident in a grey list country only if the CFC has taken advantage of overseas "specified tax preferences," and only
if New Zealand tax exceeds the foreign tax that would be payable if the item were not a preference under that foreign
country's tax laws. To date, there is only one scheduled specified tax preference, namely, any exemption from income tax
for income derived from a business carried on outside the country.
30. The shareholder continuity rules do not apply to any corporation whose shares are listed on the New Zealand Stock
Exchange.
31. The amount of the imputation credit
distributions.

IS

[(D/(1-. 25)] x .25

=

D/3 where D equals the amount of net qualifying

32. The following example illustrates the mechanics of the imputation credit and ACT. The example assumes: (1) a corporate
tax rate of 33 percent, (2) a basic personal rate of 25 percent, (3) that all shareholders are taxed at a marginal rate of 25
percent, and (4) that the corporation distributes to shareholders all after-tax (including ACT) earnings.

A.
B.

C.

D.
E.

F.
G.

H.
I.
J.
K.
L.
M.
N.
O.

P.

Corporate income before preferences
Preference deductions or exclusions (e.g., accelerated cost recovery
in excess of book depreciation)
Corporate taxable income (A - B)
Corporate tax (.33 xC)
Cash distributions to shareholders «A-F-I) or (A-[(.33-.25)/
1.33)
ACT (Ex.25/(l-.25»
Limit on use of ACT (.25 xC)
ACT applied against mainstream corporate tax (lesser of F and G)
Net mainstream tax (D-H)
Total tax paid by corporation (F +I)
Retained earnings (A - E - 1)
Surplus ACT credit available for carryback or carryforward (F - H)
Shareholder income (E + F)
Shareholder tax (.25 xM)
Shareholder tax net of imputation credit (N - F)
Total corporate and shareholder tax paid (J + 0)

£100.00
40.00
60.00
19.50
71.40
23.80
15.00
15.00
4.80
28.60
0.00
8.80
95.20
23.80
0.00
28.60

If the shareholder in the example were instead a tax-exempt entity, the shareholder would be eligible for a refund of the

entire imputation credit of £23.80. Accordingly, the total tax paid by the corporation and the shareholder would be £4.80,
the net mainstream tax paid by the corporation.
33. An indirect foreign tax credit is allowed with respect to taxes paid by a foreign corporation to aU. K. corporation that
owns at least 10 percent of the foreign corporation. A similar credit is allowed if the foreign corporation is a controlled
foreign corporation the income of which is taxed currently to a U.K. shareholder.
34. Assume that a corporation earns £100 (of which £70 is U.K. source and £30 is foreign source income) and pays foreign
tax of £9 on the foreign source income (at a 30 percent rate). The corporation's mainstream tax is £24, of which £23.10 is
attributable to U.K. income (.33 x £70) and £0.90 is attributable to to foreign source income ((.33 x £30) -£9). The
corporation distributes £60 and pays ACT of £20. Under the general limit described in Section B.6.h, the corporation may
apply the ACT of £20 against its mainstream tax on U. K. -source income only to the extent of 25 percent of £70, or £ 17.50.
The corporation also may apply ACT against the £0.90 of U. K. mainstream tax payable on the foreign source income (the

Notes

250

lesser of the mainstream tax payable and 25 percent of £30 of foreign source income). Thus, the corporation offsets £18.40
(£17 .50+ £0.90) of ACT against its mainstream tax liability of £24 and therefore must make an additional payment of £5.60.
The corporation's total U.K. tax liability is £25.60.
35. The following example illustrates the difference in treatment of shareholders in countries with such treaties and
shareholders in countries without such treaties.
Example. A corporation distributes a total of £300, consisting of £75 to each of the following: Shareholder A, a
national of a nontreaty country, Shareholder B, aU. S. national owning less than 10 percent of the stock,
Shareholder C, a U.K. resident, and Shareholder D, a U.S. national owning at least 10 percent of the stock.
Shareholders A and C are subject to tax in the United Kingdom at a marginal rate of 40 percent, Shareholder B is
subject only to the 15 percent withholding tax, and Shareholder D is subject only to the 5 percent withholding tax.
The corporation pays ACT of £100 (£300 x .25/(1- .25), or £25 on each distribution.
Shareholder A is treated as receiving a distribution of only the £75 actually paid to him and is liable for tax of £30
(.4Ox£75). Shareholder A is treated as having paid tax of £18.75 (.25x£75) due to the ACT paid by the
corporation. Thus, Shareholder A must pay an additional £12.25.
Shareholder B is treated as receiving a distribution of £100 and is liable for tax of £15 (.15X£100). Shareholder
B is treated as having paid £25 (ACT paid on the distribution), and thus is entitled to a refund of £10.
Shareholder C also is treated as receiving a distribution of £100 and is liable for tax of £40. Shareholder C is
treated as having paid £25, and thus must pay an additional £15.
Shareholder D is treated as receiving a distribution of £87.50 (£75 actually distributed plus one-half of the ACT)
and is liable for tax of £4.38 (.05 x£87.50). Shareholder D is treated as having paid £12.50 (one-half of the ACT),
and thus is entitled to a refund of £8.13.

Appendix C
1. In that case, the credit would not only be nonrefundable but also would not be allowed to offset tax on other income of
shareholders subject to tax at less than the maximum rate. The imputation credit prototype, described in Chapter 11, is a
hybrid of these two approaches. It allows credits at the maximum shareholder rate but permits low-bracket shareholders to
use excess credits against other tax liability.
2. The second to last column of the example that follows in the text illustrates that this approach will pass through
preferences if the corporate and shareholder rates are identical.
3. Corporate tax credits could be passed through by treating credits as equivalent to corporate taxes paid. Corporate
preferences that are exclusions from income could be passed through to shareholders by a separate accounting at the
corporate level and exclusion at the shareholder level. Passing through deferral preferences, however, would be more difficult
because some account would have to be taken of their reversal over time. The corporate AMT, for example, has a credit
for AMT taxes against future regular income taxes. Alternatively, asset basis might be adjusted. Either of these approaches
would be complicated at the shareholder level. See McLure (1979), pp. 95-99.
4 .. There may be an indirect benefit to tax-exempt shareholders if a dividend exclusion system results in increased stock
pnces.

GLOSSARY
AFCE: Allowance for Corporate Equity. See Section 12.B.
ACT: Advance Corporation Tax (United Kingdom). See Appendix B, Section B.6.
ALI: American Law Institute.
AMT: Alternative minimum tax. See Appendix A, Section A.I.
AMTI: Alternative minimum taxable income. See Appendix A, Section A.I.
Capital export neutrality: The principle that investors should pay equivalent taxes on capital income,
regardless of the country in which the income is earned. See Section 7.B.
Capital import neutrality: The principle that all investments within a country should face the same tax
burden, regardless of whether they are owned by a domestic or a foreign investor. See Section 7.B.
CBIT: Comprehensive Business Income Tax. See Chapter 4.
C corporation: A corporation taxed under the classical system as set forth in Subchapter C of the
Internal Revenue Code. See Appendix A, Section A.I.
CGE model: Computable general equilibrium model. See Section I3.C.
Classical system: The two-tier corporate tax system, which taxes earnings on equity capital at both the
corporate and shareh'Jlder level.
Code: The Internal Revenue Code of 1986, as amended.
DRD: Dividends received deduction. See Appendix A, Section A.I.
DRIP: Dividend reinvestment plan. See Chapter 9.
EDA: Excludable Distributions Account. See Sections 2.B and 4.B.
EK: Eigencapital (equity capital) (Germany). See Appendix B, Section B.4.
FEI: Family Economic Income. See Section 13. G.
GDP: Gross domestic product. The value of fmal goods and services produced by factors of production
in the United States.
GNP: Gross national product. The value of fmal goods and services produced by U. S. owned factors
of production, including factors that are actually used overseas.
ICA: Imputation Credit Account (New Zealand). See Appendix B, Section B.S.
251

Glossary

252

Inbound investment: Investment by foreign persons in the United States. See Section 7.A.

IRS: Internal Revenue Service.
MPM: Mutual production model. See Section 13.F.

MTD: Minimum tax on distributions. See Section 12.C.
NNP: Net national product. GNP minus capital consumption (depreciation).
NOL: Net operating loss. See Appendix A, Section A.I.
OEeD: Organisation for Economic Co-operation and Development

OlD: Original issue discount. The OlD rules govern the accrual of discount on debt. Discount is
economically equivalent to interest.
Outbound investment: Investment by U.S. persons in foreign countries. See Section 7.A.
PA model: Portfolio allocation model. See Section 13.F.

REIT: Real estate investment trust. See Appendix A, Section A.I.
REMIC: Real estate mortgage investment conduit. See Appendix A, Section A.I.
RIC: Regulated investment company. See Appendix A, Section A.I.
R&D: Research and development.
S corporation: A corporation which bears no corporate tax and whose shareholders are taxed under the
passthrough regime set forth in Subchapter S of the Internal Revenue Code. See Appendix A,
Section A.I.

SCA: Shareholder Credit Account. See Section II.B.
S&L: Savings and loan association.
Subchapter C: The portion of the Internal Revenue Code that governs the taxation of corporations
under the classical system. See Appendix A, Section A.I.

UBIT: Unrelated business income tax. A tax-exempt entity is subject to UBIT on income derived from
a business unrelated to the entity's exempt purpose and on certain passive income to the extent it
is financed with debt.

BIBLIOGRAPHY
Aaron, Henry J., "A new view of property tax incidence," American Economic Review, Vol. 64 (May 1974). p. 212.
Aaron, Henry J. and Harvey Galper, Assessing Tax Reform. Washington: The Brookings Institution (1985).
Accounting Principles Board, Opinion No. 11, Accounting for Income Taxes (1967).
Accounting Principles Board, Opinion No. 23, Accounting for Income Taxes -- Special Areas (1972).
Agrawal, Anup and Gershon N. Mandelker, "Managerial incentives and corporate investment and financing decisions," The
Journal of Finance, Vol. 42 (September 1987). p. 823.
Ambarish, Rarnasastry, Kose John, and Joseph Williams, "Efficient signalling with dividends and investments," The Journal
of Finance, Vol. 42 (June 1987). p. 321.
American Bar Association, Section on Taxation, "Banking and savings institutions," Tax Lawyer, Vol. 37 (1984). p. 795.
American Bar Association, Section on Taxation, "Banking and savings institutions," Tax LaH-,),er, Vol. 38 (1985). p. 819.
American Bar Association, Section on Taxation, Earnings and Profits Work Group, "Elimination of 'earnings and profits'
from the Internal Revenue Code," Tax LaH-'}'er, Vol. 39 (1986). p. 285.
American Bar Association, Section on Taxation, and New York State Bar Association, Tax Section, Corporate Tax Reform:
A Report of the Invitational Conference on Subchapter C (1988).
American Law Institute, Federal Income Tax Project, Reporter's Study Draft, Subchapter C (Supplemental Study).
Philadelphia (1989). The Memorandum was written by Professor William D. Andrews of the Harvard Law School.
American Law Institute, Federal Income Tax Project, Integration of the Individual and Corporate Income Taxes, Reporter's
Memorandum No.1. Philadelphia (1990). The Memorandum was written by Professor Alvin C. Warren of the Harvard
Law School.
American Law Institute, Federal Income Tax Project, Integration of the Individual and Corporate Income Taxes, Reporter's
Memorandum No.2. Philadelphia (January 15, 1991). The Memorandum was written by Professor Alvin C. Warren
of the Harvard Law School.
American Law Institute, Federal Income Tax Project, Integration of the Individual and Corporate Income Taxes, Reporter's
Memorandum No.3. Philadelphia (draft of April 7,1991). The Reporter is Professor Alvin C. Warren of the Harvard
Law School.
Ando, Albert and Franco Modigliani, 'The life cycle' hypothesis of saving: aggregate implications and tests," American
Economic Review, Vol. 53 (March 1963). p. 55.
Andrews, William D., "Out of its earnings and profits: some reflections on the taxation of dividends," Harvard Law Review,
Vol. 69 (1956). p. 1403.
Andrews, William D., "Tax neutrality between equity capital and debt," Wayne Law Review, Vol. 30 (1984). p. 1057.
Andrews, William D., "A consumption-type or cash flow personal income tax," Harvard Law Review, Vol. 87 (1974). p.
1113.
Ang, James and David Peterson, "Optimal debt versus debt capacity: a disequilibrium model of corporate behavior,"
Research in Finance, Vol. 6, A. Chen, editor. Greenwich, CT: JAI Press (1986).
Auerbach, Alan 1., "Wealth maximization and the cost of capital," Quarterly Journal of Economics, Vol. 93 (August 1979).
p.433.
Auerbach, Alan 1., "Tax integration and the new view of the corporate tax: a 1980 perspective," Proceedings of the 74th
Annual Conference. Columbus: National Tax Association and Tax Institute of America (1981).
Auerbach, Alan J., "Corporate taxation in the United States," Brookings Papers on Economic Activity, Vol. 2 (1983). p.
451.
Auerbach, Alan J., "Taxation, corporate financial policy, and the cost of capital," Journal of Economic Literature, Vol. 21
(September 1983). p. 905.
Auerbach, Alan J., "Real determinants of corporate leverage," Corporate Capital Structures in the United States, Benjamin
Friedman, editor. Chicago: The University of Chicago Press (1985).
253

Bibliography

254

Auerbach, Alan J., "Tax policy and corporate borrowing," Are the Distinctions Between Debt and Equity Disappearing]
Richard Kopcke and Eric Rosengren, editors. Boston: Federal Reserve Bank of Boston (1989).
Auerbach, Alan J., "Debt, equity, and the taxation of corporate cash flows," Taxes, Debt and Corporate Restructuring, John
B. Shoven and Joel Waldfogel, editors. Washington: The Brookings Institution (1990).
Auerbach, Alan J. and Kevin Hassett, "Tax policy and business fixed investment in the United States," unpublished paper
(1990).
Auerbach, Alan J. and Kevin Hassett, "Recent U.S. behavior and the Tax Reform Act of 1986: a disaggregate view,"
Workmg Paper No. 3626, Cambridge: National Bureau of Economic Research (February 1991).
Auerbach, Alan J. and James M. Poterba, "Why have corporate revenues declined?" Tax Policy and the Economy, Vol. I.
Lawrence H. Summers, editor. Cambridge: MIT Press (1987).
Auerbach, Alan and David Reishus, "Taxes and the merger decision, " Knights, Raiders, and Targets: The Impact of Hostile
Takeovers, John Coffee, Louis Lowenstein, and Susan Rose-Ackerman, editors. New York: Oxford University Press
(1988).
Ault, Hugh J., "International issues in corporate tax integration, " Law and Policy in International Business, Vol. 10 (1978).
p. 461.
Avi-Yonah, Reuven S., "The treatment of corporate preference items under an integrated tax system: a comparative
analysis," Tax La}),yer, Vol. 44 (1990). p. 195.
Bagwell, Laurie Simon and John B. Shoven, "Cash distributions to shareholders," The Journal of Economic Perspectives,
Vol. 3 (Summer 1989). p. 129.
Ballard, Charles L., Don Fullerton, John B. Shoven, and John Whalley, A General Equilibrium Model for Tax Policy
Evaluation. Chicago: University of Chicago Press (1985).
Ballard, Charles L., John B. Shoven, and John Whalley, "The total welfare cost of the United States tax system: a general
equilibrium approach," National Tax Journal, Vol. 38 (June 1988). p. 125.
Barro, Robert and Xavier Sala-i-Martin, "World real interest rates," NBER Macroeconomics Annual 1990. Cambridge: MIT
Press (1990). p. 15.
Bartholdy, 1., G. Fisher and Jack Mintz, "Some theory of taxation and financial policy with application to Canadian
corporate data." Paper presented at the Econometric Society Fifth World Congress, Cambridge (1985).
Batten, Dallas S. and Mack Ott, "The President's proposed corporate tax reforms: a move toward tax neutrality," Federal
Reserve Bank of Sf. Louis Review, Vol. 67 (August/September 1985). p. 5.
Benge, Matt and Tim Robinson, How to Integrate Company and Shareholder Taxation: Why Full Imputation is the Best
Answer. Wellington: Victoria University Press for the Institute of Policy Studies (1986).
Bergsten, C. Fred, Thomas Horst, and Theodore H. Moran, American Multinationals and American Interests. Washington:
The Brookings Institution (1978).
Bernanke, Ben S., "Is there too much corporate debt?" Federal Reserve Bank of Philadelphia Quarterly Review (SeptemberOctober 1989). p. 3.
Bernanke, Ben S. and John Y. Campbell, "Is there a corporate debt crisis?" Brookings Papers on Economic Activity 1
(1988). p. 83.
Bemanke, Ben S., John Y. Campbell, and Toni M. Whited, "U.S. corporate leverage: developments in 1987 and 1988,"
Brookings Papers on Economic Activity 1 (1990). p. 255.
Bhattachaya, Sudipto, "Imperfect infonnation, dividend policy, and the 'bird in the hand' fallacy," Bell Journal of
Economics, Vol. 10 (Spring 1979). p. 259.
Bird, Richard M., "International aspects of integration," National Tax Journal, Vol. 28 (1975). p. 302.
Bittker, Boris 1., "A 'comprehensive tax base' as a goal of income tax refonn," Harvard Law Review, Vol. 80 (1967). p.
925.
Blair, Margaret and Robert Litan, "Corporate leverage and leveraged buyouts in the eighties," Taxes, Debt, and Corporate
Restructuring, John B. Shoven and Joel Waldfogel, editors. Washington: The Brookings Institution (1990).

255

Bibliography

Blazenko, George W., "Managerial preference, asymmetric information, and financial structure," The Journal of Finance,
Vol. 42 (September 1987). p. 839.
Blinder, Alan and Angus Deaton, "The time series consumption function revisited," Brookings Papers on Economic Activity
2 (1985). p. 465
Blum, Walter J., "The earnings and profits limitation on dividend income: a reappraisal," Taxes, Vol. 53 (1975). p. 68.
Boskin, Michael J., "Taxation, saving and the rate of interest," Journal of Political Economy, Vol. 86 (April 1978). p. S3.
Bosworth, Barry P., Tax Incentives and Economic Growth. Washington: The Brookings Institution (1984).
Bradford, David F., "The incidence and allocation effects of a tax on corporate distributions, " Journal of Public Economics
Vol. 15 (February 1981). p. 1.
Bradford, David F., Untangling the Income Tax. Cambridge: Harvard University Press (1986).
Bradford, William D., "The issue decision of manager-owners under information asymmetry," The Journal of Finance, Vol.
42 (December 1987). p. 1245.
Bradley, Michael, Gregg Jarrell, and E. Han Kim, "On the existence of an optimal capital structure: theory and evidence,"
The Journal of Finance, Vol. 39 (July 1984). p. 857.
Bravenec, Lorence L., "A nontraditional approach to corporate integration," Tax Notes (March 13, 1989). p. 1381.
Break, George F., "Integrating corporate and personal income taxes: the Carter Commission proposals," Law and
Comemporary Problems, Vol. 34 (1969). p. 726.
Break, George F., "Integration of corporate and personal income taxes," National Tax Journal, Vol. 22 (1969). p. 39.
Break, George F., "Corporate tax integration: radical revisions or common sense?" Federal Tax Reform Myths and Realities,
Michael J. Boskin, editor. San Francisco: Institute for Contemporary Studies (1978).
Break, George F. and Joseph A. Pechman, "Relationship between the corporation and individual income taxes," National
Tax Journal Vol. 28 (1975). p. 341.
Break, George F. and Joseph A. Pechman, Federal Tax Reform: the Impossible Dream? Washington: The Brookings
Institution (1975).
Breeden Douglas, "An intertemporal capital asset pricing model with stochastic consumption and investment opportunities,"
Journal of Financial Economics, Vol. 7 (1979). p. 265.
Brittain, John, Corporate Dividend Policy. Washington: The Brookings Institution (1966). p. 74.
Brown, E. Cary, "Business-income taxation and investment incentives," Income, Employment, and Public Policy: Essays
in Honor of Alvin H. Hanson, New York: W. W. Norton & Company, Inc. (1948).
Bulow, Jeremy I., Lawrence H. Sununers, and Victoria P. Summers, "Distinguishing debt from equity in the junk bond
era," Debt, Taxes, and Corporate Restructuring, John B. Shoven and Joel Waldfogel, editors. Washington: The
Brookings Institution (1990).
Campbell, John Y. and Richard Clarida, "The term structure of Euromarket interest rates: an empirical investigation,"
Journal of Monetary Economics, Vol. 19 (1987). p. 25.
Canellos, Peter C., "Corporate tax integration: by design or by default," Tax Notes (June 8, 1987). p. 999.
Caprio, Gerard and David Howard, "Domestic saving, current accounts, and international capital mobility," International
Finance Discussion Papers No. 244. Washington: Federal Reserve Board (1984).
Caves, Richard E., Multinational Enterprises and Economic Analysis. Cambridge: Cambridge University Press (1983).
Chirelstein, Marvin A., "Optional redemptions and optional dividends: taxing the repurchase of common shares," Yale Law
Journal, Vol. 78 (1969). p. 739.
Chirinko, Robert S., "Business investment and tax policy: a perspective on existing models and empirical results," National
Tax Journal, Vol. 39 (June 1986). p. 137.
Chirinko, Robert S., "The ineffectiveness of effective tax rates on business investment: A critique of Feldstein's FisherSchultz lecture," Journal of Public Economics, Vol. 32 (April 1987). p. 369.

Bibliography

256

Chirinko, Robert S. and Robert Eisner, "Tax policy and investment in major U.S. macroeconomic econometric models,'
Journal of Public Economics, Vol. 20 (March 1983). p. 139.
Clark, Robert C., "The morphogenesis of Subchapter C: an essay in statutory evolution and rdorm." Yale Law Journal, Vol.
87 (1977). p.90.
Cohen, Edwin S., "Possible solutions to practical problems in integration of the corporate and shareholders income tax, '
National Tax Journal, Vol. 28 (1975). p. 359.
Cohen, Edwin S., Alvin C. Warren, and William D. Andrews, "The meaning of changes within the framework of
Subchapter C and the impact on proposals for integration of the corporate and individual income tax," San Diego Lal\'
Review, Vol. 22 (1985). p. 317.
Congressional Budget Office, The Changing Distribution of Federal Taxes: 1975-1990 (October 1987).
Congressional Budget Office, The Changing Distribution of Federal Taxes: A Closer Look at 1980 (July 1988).
Consultative Committee on Full Imputation and International Tax Refonn (The Valabh Committee), Full Imputation, an
independent report to the New Zealand Government. Wellington (April 1988).
Consultative Committee on Full Imputation and International Tax Refonn (The Valabh Committee), International Tax
Reform: Full Imputation, an independent report to the New Zealand Government. Wellington (July 1988).
Consultative Committee on the Taxation of Income from Capital (The Valabh Committee), The Taxation of Distributions
from Companies, an independent report to the New Zealand Government. Wellington (July 1991).
Cooper, R., R. Krever, and R. Vann, Income Taxation. Sydney: The Law Book Company Limited (1989).
Cutler, David M. and Lawrence Summers, "The costs of conflict resolution and financial distress: evidence from the TexacoPenzoil litigation." Rand Journal of Economics, Vol. 19 (Summer 1988). p. 157.

Daily Tax Report, "EC Commissioner Scrivener vows to fight proposed environment taxes," (November 8, 1991). p. G-2.
Dooley, Michael, Jeffrey Frankel, and Donald J. Mathieson, "International capital mobility: what do saving-investment
correlations tell us?" International Monetary Fund Staff Papers, Vol. 34 (1987). p. 503.
Douglas, R., Consultative Document on Full Imputation. Wellington: New Zealand Ministry of Finance (1987).
Easterbrook, Frank H., "Two agency-cost explanations of dividends," American Economic Review, Vol. 74, (September
1984). p. 650.
Ehnll, Liam P. and David G. Hartman, "On the incidence and excess burden of the corporation income tax," Public
Final/ce, Vol. 37 (1982). p. 48.
Eisner, Robert, "Tax policy and investment behavior: comment," American Economic Review, Vol. 59 (June 1969). p. 379.
Eisner, Robert and M. Nadiri, "Investment behavior and neoclassical theory," Review of Economics and Statistics, Vol. 50
(August 1968). p. 369.
Engel, Charles and Kenneth Kletzer, "Saving and investment in an open economy with non-traded goods," Working Paper
no. 2141, Cambridge: National Bureau of Economic Research (1987).
Fazzari, Steven, R. Glenn Huhbard, and Bruce Petersen, "Financing constraints and corporate investment," Brookings Papers
Oil Economic Activil), 1, (1988). p. 141.
Federal Reserve Board, Flow of Funds Accounts, Financial Assets and Liabilities, Year End, Board of Governors of the
Federal Reserve System (various issues).
Feenherg, Daniel R. and Jonathan Skinner, "Sources of IRA savings," Tax Policy and the Economy, Vol. 3, Lawrence H.
Summers, editor. Cambridge: MIT Press (1988).
Feldstein, Martin, "Corporate taxation and dividend behaviour," Review of Economic Studies, Vol. 37 (February 1970). p.
57.
Feldstein, Martin, "Corporate taxation and dividend behaviour: a reply and extension," Review of Economic Studies, Vol.
39 (April 1972). p. 235.
Feldstein, Martin, "The incidence of the social security payroll tax: comment," American Economic Review, Vol. 62
(September 1972). p. 735.

257

Bibliography

Feldstein, Martin, "The income tax and charitable contributions: part II - the impact on religious, educational, and other
organizations," National Tax Journal, Vol. 28 (June 1975). p. 209.
Feldstein, Martin, "Inflation, tax rules, and investment: some econometric evidence," Econometrica, Vol. 50 (July 1982).
p. 825.
Feldstein, Martin, "Domestic saving and international capital movements in the long run and the short run," European
Economic Review, Vol. 21 (1983). p. 139.
Feldstein, Martin, "Imputing corporate tax liabilities to individual taxpayers," Working Paper No. 2349. Cambridge:
National Bureau of Economic Research (1987).
Feldstein, Martin, "Testimony on tax policy aspects of mergers and acquisitions," House of Representatives, Committee on
Ways and Means, Serial No. 101-10, Tax Policy Aspects of Mergers and Acquisitions, Part I (January 31,1989). p.
192.
Feldstein, Martin and Phillipe Bacchetta, "National saving and international investment," Working Paper No. 3164.
Cambridge: National Bureau of Economic Research (1989).
Feldstein, Martin, Louis Dicks-Mireaux, and James M. Poterba, "The effective tax rate and the pre-tax rate of return,"
Journal of Public Economics, Vol. 21 (July 1983). p. 129.
Feldstein, Martin and Daniel Frisch, "Corporate tax integration: the estimated effects on capital accumulation and tax
distribution of two integration proposals," National Tax Journal, Vol. 30 (1977). p. 37.
Feldstein, Martin and Charles Horioka, "Domestic saving and international capital flows,"
90 (June 1980). p. 314.

The Economic Journal, Vol.

Feldstein, Martin and Joosung Jun, "The effects of tax rules on nonresidential fixed investment: some preliminary evidence
from the 1980s," The Effects of Taxation on Capital Accumulation, Martin Feldstein, editor. Chicago: University of
Chicago Press (1987).
Feldstein, Martin and Joel Slemrod, "Inflation and the excess taxation of capital gains on corporate stock," National Tax
Journal, Vol. 31 (1978). p. 107.
Feldstein Martin, Joel Slemrod, and Shlomo Yitzhaki, "The effects of taxation on the selling of corporate stock and the
realization of capital gains: reply," The Quarterly Journal of Economics, Vol. 99 (February 1984). p. Ill.
Fieleke, Norman S., "National saving and international investment," Saving and Government Policy, Conference Series No.
25. Boston: Federal Reserve Bank of Boston (1982).
Financial Accounting Standards Board, Statement of Financial Accounting Standards No. 96, Accounting for Income Taxes
(1987).
First Boston Corporation, New York, High Yield Handbook (various issues).
Frankel, Jeffrey, "International capital mobility and crowding-out in the U.S. economy: imperfect integration of financial
markets or of goods markets?" How Open is the U.S. Economy?, R. Hafer editor. Lexington: Lexington Books (1986).
p. 33.
Frankel, Jeffrey, "Quantifying international capital mobility in the 1980s," Current Issues in International Trade and
International Finance, Dilip Das, editor. Oxford: Oxford University Press. (forthcoming).
Frankel, Jeffrey and A. MacArthur, "Political vs. currency premia in international real interest differentials: a study of
forward rates for 24 countries," European Economic Review, Vol. 32 (1988). p. 1083.
Frankel, Jeffrey and Kenneth Froot, "Using survey data to test standard propositions regarding exchange rate expectations,"
American Economic Review, Vol. 77 (1987). p. 133.
Freemen, Louis S., "Some Early Strategies for the methodical disincorporation of America after the Tax Reform Act of
1986: grafting partnerships onto C corporations, running amok with the master limited partnership concept, and generally
endeavoring to defeat the intention of the draftsmen of the repeal of General Utilities," Taxes, Vol. 64 (1986). p. 962.
Friedman, Benjamin J., "Views on the likelihood of financial crisis," Reducing the Risk of Financial Crisis, Martin
Feldstein, editor. Chicago: University of Chicago Press (1990).
French, Kenneth R. and James M. Poterba, "Investor diversification and international equity markets, n Working Paper No.
3609. Cambridge: National Bureau of Economic Research (January 1991).

Bibliography

258

Frenkel, J. and R. Levich, "Transaction costs and interest arbitrage: tranquil versus turbulent periods," Joumal of Political
Ecollomy, Vol. 85 (1977). p. 1209.
Froot, Kenneth and Jeffrey Frankel, "Forward discount bias: is it an exchange risk premium?" Quarterly Journal of
Ecollomics, Vol. 104 (1989). p. 139.
Fullerton, Don, "Which effective tax rate?" National Tax Joumal, Vol. 37 (March 1984). p. 23.
Fullerton, Don, Robert Gillette, and James Mackie, "Investment incentives under the Tax Reform Act of 1986,"
Compendium of Tax Research 1987. Washington: U.S. Govt. Print. Off. (1987).
Fullerton, Don and Roger H. Gordon, "A reexamination of tax distortions in general equilibrium models," Behavioral
Simulation Methods in Tax Policy Analysis, Martin Feldstein, editor. Chicago: University of Chicago Press (1983).
Fullerton, Don, Yolanda Henderson, and James Mackie, "Investment allocation and growth under the Tax Reform Act of
1986," Compendium of Tax Research 1987. Washington: U.S. Govt. Print. Off. (1987).
Fullerton, Don and Yolanda Henderson, "A dissagregate equilibrium model of tax distortions among assets, sectors and
industries," International Economic Review, Vol. 30 (May 1989). p. 391.
Fullerton, Don, John B. Shoven, and John Whalley, "Replacing the U.S. income with a progressive consumption tax,"
Journal of Puhlic Economics, Vol. 20 (1983). p. 3.
Fullerton, Don, A. Thomas King, John B. Shoven, and John Whalley, "Corporate tax integration in the United States: a
general equilibrium approach," The American Economic Review, Vol. 71 (September 1981). p. 677.
Gaffrey, Dennis J. and James E. Wheeler, "The double taxation of corporate source income: reality or illusion?" Tax
Advisor, Vol. 8 (1977). p. 523.
Gale, William G. and John Karl Scholz, "IRAs and household saving," unpublished paper (1990).
Galper, Harvey, Robert Lucke, and Eric Toder, "A general equilibrium analysis of tax reform," Uneasy Compromise:
Problems of a Hybrid Income-Consumption Tax, Henry J. Aaron, Harvey Galper and Joseph A. Pechman, editors.
Washington: The Brookings Institution (1988).
Gammie, Malcolm, "Corporate tax harmonisation: an 'ACE' proposal," IBFD European Taxation, Vol. 12 (1991). p. 545.
Gertler, Mark and R. Glenn Hubbard, "Taxation, corporate capital structure, and financial distress," Tax Policy and the
Economy 4, Lawrence Summers, editor. Cambridge: MIT Press (1990).
Gertler, Mark and R. Glenn Hubbard, "Corporate financial policy, taxation, and macroeconomic risk." Unpublished paper
(1991).
Gertler, Mark and Kenneth Rogoff, "North-south lending and endogenous domestic capital market inefficiencies," Journal
of Monetary Economics, Vol. 26 (1990). p. 245.
Gilbert, Richard 1. and David M. Newbery, "Entry, acquisition, and the value of shark repellent," Working Paper 8888,
University of California, Berkeley, (August 1988).
Gilson, Ronald J., Myron S. Scholes, and Mark A. Wolfson, "Taxation and the dynamics of corporate control: the uncertain
case for tax motivated acquisitions," Working Paper No. 24, Stanford Law School (January 1986).
Giovannini, Alberto, "Capital taxation: national tax systems versus the European capital market," Economic Policy, Vol.
4 (Oct. 1989). p. 345.
Giovannini, Alberto and Philippe Jorion, "Interest rates and risk premia in the stock market and the foreign exchange
market," Journal of International Money and Finance, Vol. 6 (1987). p. 107.
Gordon, Roger H., "Uncertainty and the analysis of corporate tax distortions, " Proceedings of the 74th Annual Conference.
Columbus: National Tax Association-Tax Institute of America (1981).
Gordon, Roger H., "An optimal taxation approach to fiscal federalism," Quarterly Journal of Economics, Vol. 98 (Nov.
1983). p. 567.
Gordon, Roger H., "Taxation of investment and savings in world economy: the certainty case," American Economic Review,
Vol. 76 (Dec. 1986). p. 1086.
Gordon, Roger H. and Jeffrey MacKie-Mason, "Effects of the tax reform act of 1986 on corporate financial policy and
organizational form," Do Taxes Matter? 1. Slemrod, editor. Cambridge: MIT Press (1990).

259

Bibliography

Gordon, Roger H. and Jeffrey K. MacKie-Mason, "Taxes and the choice of organizational form, " Working Paper No. 3781.
Cambridge: National Bureau of Economic Research (July 1991).
Gordon, Roger H. and Burton Malkiel, "Corporation finance," How Taxes Affect Economic Behavior, Henry J. Aaron and
Joseph A. Pechman, editors. Washington: The Brookings Institution (1981).
Gourevitch, Harry G., "Corporate tax integration: the European experience," Tax Lawyer, Vol. 31 (1977). p. 65.
Graetz, Michael J., "Legal transitions: the case of retroactivity in income tax revision," University of Pennsylvania Law
Review, Vol. 47 (1977). p. 126.
Graetz, Michael J., "Implementing a progressive consumption tax," Harvard Law Review, Vol. 92 (1979). p. 1575.
Gravelle, Jane G., Effective Tax Rates in the Administration and Ways and Means Tax Proposal: Updated Tables,
Congressional Research Service Report for Congress No. 85-1006E, Congressional Research Service, Washington, D.C.
(1985).
Gravelle, Jane G., Effective Corporate Tax Rates in the Major Revision Plans: A Comparison of the House, Senate, and
Conference Committee Versions, Congressional Research Service Report for Congress No. 85-1099E, Congressional
Research Service, Washington, D.C. (1986).
Gravelle, Jane G., "Differential taxation of capital income: another look at the 1986 Tax Reform Act," National Tax Journal,
Vol. 47(2) (December 1989). pp. 44l.
Gravelle, Jane G., Corporate Tax Integration: Issues and Options, Congressional Research Service, Washington, D.C.
(1991).
Gravelle, Jane G. and Laurence Kotlikoff, "The incidence and efficiency costs of corporate taxation when corporate and
noncorporate firms produce the same good," Journal of Political Economy, Vol. 97 (Aug. 1989). p. 749.
Griffith, Thomas D., "Integration of the corporate and personal income taxes and the ALI proposal," Santa Clara Law
Review, Vol. 23 (1983). p. 715.
Goldsworth, John G., "Status report on harmonization of direct taxation," Tax Notes lmernational (February 14,1990). p.
15.
Hall, Robert E., "Intertemporal substitution in consumption," Journal of Political Economy, Vol. 96 (April 1988). p. 339.
Hall, Robert E. and Dale W. Jorgenson, "Tax policy and investment behavior," American Economic Review, Vol. 57 (June
1967). p. 391.
Hall, Robert E. and Dale W. Jorgenson, "Application of the theory of optimal capital accumulation," Tax Incentives and
Capital Spending, Gary Fromm, editor. Washington: The Brookings Institution (1971).
Halperin, Daniell., "Interest in disguise: taxing the time value of money," Yale Law Journal, Vol. 95 (1986). p. 506.
Halperin, Daniel l. and C. Eugene Steuerle, "Indexing the tax system for inflation," Uneasy Compromise: Problems of a
Hybrid Income-Consumption Tax, Harvey Galper and Joseph A. Pechman, editors. Washington: The Brookings
Institution (1988).
Hammer, Richard M., "The taxation of income from corporate shareholders: review of present systems in Canada, France,
Germany, Japan and the U.K.," National Tax Journal, Vol. 28 (1975). p. 315.
Hansen, Lars and Robert Hodrick, "Risk averse specUlation in the forward foreign ex