View original document

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

November 15. 1991

eCONOMIC
Federal Reserve Bank of Cleveland

Generational Accounts: A New
Approach to Fiscal Policy Evaluation
by Alan Auerbach, Jagadeesh Gokhale, and Laurence J. KotlikofF

De"espite recent attempts to impose discipline on the federal budget-making process, federal budget deficits have continued
to escalate over the past several years. The
prospect of still larger deficits over the next
few years has stimulated discussion about
how to bring them under control. Some
have suggested reforms that would enable
the federal government to achieve balanced
budgets, not necessarily in every year in
the future, but at least on average over a
number of years.1
On the face of it. such recommendations
seem to be sensible. On January 30. however, the government released 1991 projections for not one. but as many as four
different types of deficits, each based on
a different definition of items included in
the calculation.' The question then
arises: Which of these deficits should be
targeted for elimination? This paradoxical
set of events suggests a need to reexamine the concerns prompted by large
budget shortfalls.
Two components underlie the generally
accepted wisdom about why budget deficits are important. First, as a matter of intergenerational equity, it is important to
know how much of the burden of paying
for government expenditure is being
shifted onto future generations. Many
people mistakenly think that the reported
annual budget deficit measures the extent
of such a redistribution resulting from
current policy. Second, a shifting of the
payments burden onto future generations
is likely to adversely affect these generations' incentives to work, save, and

invest. Thus, large redistributions of
wealth away from future generations
are likely to be detrimental to future
economic growth.
Correct measurement of the intergenerational redistributive impact of any fiscal
policy is, quite obviously, a precondition
for evaluating the policy's implications
for future economic growth. This Economic Commentary surveys the reasons
for doubting that deficits, as conventionally defined, are adequate measures of
the effect of fiscal policy on intergenerational redistribution. It also suggests
that evaluation on this score would be
improved by looking at generational
accounts — a system designed to reflect
consistently the intergenerational redistribution implications of current fiscal
policy and of future policy changes.
• The Deficit — an Inadequate
Measure of Fiscal Policy
Dissatisfaction with reported deficits has
led to suggestions for making various corrections tojhe deficit and to public debt
numbers. These include adjustments for
inflation, for market values of government assets, for economic growth, and
for state and local budget surpluses.3
However, the problem with using the
deficit as an indicator of policy runs
deeper than any of these modifications
can fathom. It can be shown that deficit
numbers, no matter how they are measured, reflect nothing more than the rules
of accounting used in their computation.

The federal government often adopts
fiscal policies whose long-term implications for intergenerational wealth
redistribution are not correctly reflected in published budget-deficit
numbers. Correct measurement of
such policy-induced redistribution is
important in order to gauge the likely
impact of policy changes on future
economic growth. Generational
accounting provides a new method
for investigating this issue. An application of this method for the United
States reveals that today's fiscal policies involve a substantial imbalance
in the treatment of current and future
generations.

I

and that such numbers are devoid of
economic content.
The starkest example of the dependence
of the size of the deficit on accounting
conventions is provided by the Social
Security system. At its inception, the
system was unfunded: The elderly generations then alive received old-age
benefits that were financed from Social
Security contributions of the thenworking-age generations. Implicit in
this scheme was a promise of future
benefits for the current contributors.
The contributions, however, were
called "taxes." and the old-age benefits
to be paid in the future were labeled
"transfers." If the words describing
these transactions had. instead, been
"borrowing" and "repayment of principal and interest." respectively, the
sizes of the recorded government deficits as far back as the 1960s and 1970s
would have been many times larger
than the S200 billion deficits reported
during the late 1980s.4
The installation of an unfunded Social
Security system is just one example of
how the government can engage in redistributive fiscal policies whose long-term
implications are not correctly reflected
in the reported deficit. An equal revenue shift from sales/excise taxation to
income taxation, for example, would
redistribute wealth from younger and
future generations to older generations.
This is because the young pay relatively more through income taxes, while
the old pay relatively more in sales/
excise (consumption) taxes. A revenueneutral elimination of investment incentives would also redistribute wealth
from younger and future generations to
older generations."
Hence, by combining policies that are reflected in the deficit with others that are
not. by arbitrarily labeling government
receipts and payments, and by changing
accounting conventions and inclusion
rules, the government can essentially report any number as the current deficit
while following the same underlying fiscal policy. Deficit numbers can mislead
the public into thinking that government
fiscal policy is profligate when it may

really be prudent, and vice versa Because conventionally reported deficits
are arbitrary, we need to seek bcttsr
measures of fiscal policy.
• The Need for an Alternative
Measure of Fiscal Policy
The annually reported deficit is a number reflecting the current net cash flow
of the government. The government,
however, often engages in fiscal policies that have long-term consequences.
Several of its policies, such as the introduction and modification of an unfunded Social Security system, changes
in the relative importance of consumption and income taxation, and the introduction and subsequent elimination of
investment incentives, have significant
long-term effects on the resources of
private individuals.
Consider, for example, the 1983 Social
Security Amendment. This amendment
mandated higher payroll taxes now and
lower benefits in the future in order to
reduce the burden on future working
generations of supporting the baby
boomers who will begin to retire early
in the next century. The Social Security
system is currently generating large
surpluses which, when added to the
rest of the government's budget, reduce
the magnitude of the reported federal
budget deficit today. In 1983. however,
when the policy was changed, the
reported deficit was unaffected.
The long-term implications of a given
policy will typically vary for individuals
belonging to different generations. In general, fiscal policies redefine how much
each generation will pay for government
spending now and in the future. In short,
these policies are really generational policies, and they need to be recognized as
such. Thus, in order to evaluate current
fiscal policy, one must examine its generational stance — how it redistributes
the burden of paying for government
spending among different currently living
and future generations. This, in fact, is
the concern underlying the word deficit.
• What Are Generational Accounts?
risci! policies "r-jl: :r. zr. ~pr?pr:p.!:?of resources for the government's oper-

ation. The burden of financing government spending will not. in general, fall
given fiscal policy, members of each
generation make payments to and receive payments from the government at
different points during their lifetimes.
A change in any component of fiscal
policy implies a change in the amounts
and timing of these payments. Generational accounts (GAs) track the distribution of the net-payments burden
among different existing and future
generations that is implied by the current set of fiscal policies.
The GA for any given generation measures the present value of the net payments
that each member of that generation is
projected to make, on average, under current policy over his or her remaining lifetime. Thus, a policy-induced change in
GAs will indicate which generations lose,
and which gain, as a result of the change
in policy. The changes in the GAs will reflect current as well as projected future
losses and gains arising from the policy
change. In evaluations of how any such
modification affects the intergenerational
distribution of paying for government
consumption, it is these policy-induced
changes in the GAs of different current
and future generations, rather than the
reported deficit, that should be the focus
of attention.
• How Are Generational
Accounts Computed?
The entire current and future consumption
spending by the government must be
paid for by at least one of three sources:
1) the government's current net wealth,
2) resources taken from generations currently alive, and 3) resources taken from
generations as yet unborn. Computing
GAs. then, involves the following steps:
First, obtain the present value of the government's current and future projected
spending levels that are implicit under
current spending policies. Second, estimate the government's current net wealth.
Third, compute the present value of the
projected net payments that each existing
generation is expected to make under current policy over its remaining lifetime.

TABLE 1

THE COMPOSITION OF MALE AND FEMALE GENERATIONAL ACCOUNTSPRESENT VALUES OF RECEIPTS AND PAYMENTS
(THOUSANDS OF DOLLARS)
Receipts

Pavments
Labor
Social
Generation's
Net
Income Security
Age in 1989 Payment Taxes
Taxes

Males:
0
10
20
30
40
50
60
70

80
Future
generations

73.7
116.8
169.1
194.5
176.2
114.1
18.9
-42.7
-35.6

Capital

Sei-

Prop-

Excise
Taxes

Income
Taxes

gnorage

erty
Taxes

1.6
2.6

24.8
40.8
61.9
69.6
60.9
42.1
20.2
4.0
0.6

26.5
43.6
66.2
74.4
65.1
45.0
21.5
4.3
0.6

22.9
29.8
33.9
34.2
30.5
24.4
17.9
11.9
7.5

9.5

0.0

15.6
24.8
38.4
49.8
52.1
44.1
29.3
17.2

0.1
0.1
0.1
0.0
0.0
0.0
0.0
0.0

14.0
23.3
34.8
35.1
29.7
20.4
9.3
2.0
0.0

14.9
24.9
37.2
37.5
31.7
21.8
9.9

20.2
27.2
32.2
33.1
30.1
24.2
17.4
11.5
7.2

3.5
5.9
9.3
14.9
21.4
25.0
23.4
17.3
8.8

0.0
0.1

4.1
6.1
7.6
7.1
6.1
4.5
3.0

Medicare

Unem-

Welfare
b

C

ployment rood
General Insurance Stamps

OASDI"

HI

4.5
6.7
9.5
14.3
21.9
37.1
62.6
61.9
36.9

1.1
1.9
2.9
4.6
7.4
12.4
22.0
29.6
24.4

0.3
0.5
0.7
0.8
0.5
0.3
0.1
0.0
0.0

1.5
2.5
3.9
6.1
9.8
16.3
27.8
36.8
29.9

2.3
3.8

7.8
7.8

5.2
3.5

9.2
8.5
7.8
7.3
7.2
6.5

AFDC

4.4
4.6

5.3
5.4

5.3
5.4

5.6
4.9

3.0

1.0
1.6
2.4
2.3
1.8
1.1
0.3
0.0
0.0

0.3
0.5
0.8
0.9
0.7
0.5
0.3
C.2
0.1

0.4
0.7
1.1
1.0
0.7
0.4
0.2
0.0
0.0

1.3

89.5

Females:

0
10
20
30
40
50
60
70
80
Future
generations

36.4
60.4
85.5
90.9
78.2
41.0
-22.5
-60.2
-50.8

2.2
0.0

0.0
0.0
0.0
0.0
0.0
0.0
0.0

2.1
3.5
5.4
7.4
8.6
8.9
8.2
6.9
5.1

5.0
7.5
10.9
15.7
21.9
34.0
55.1
56.5
37.4

1.7
0.6
0.0
0.0
0.0

4.5

3.3
2.4
1.4
0.7
0.4

0.3
0.2

44.2

a. Old Age Survivors and Disability Insurance.
b. Health Insurance.
c. Aid to Families with Dependent Children.
SOURCE: Alan Auerbach. Jagadeesh Gokhale. and Laurence Kotlikoff. "Generational Accounts: A Meaningful Alternative to Deficit Accounting." in David Bradford,
ed.. TJ.\ Policy and the Economy, National Bureau of Economic Research and MIT Press. No. 5. 1991.

Once the present value of government
spending and the present value of two
of the sources for financing it are
known, the present value of the third
component — the amount that needs to
be taken from future generations—can
be obtained as a residual. We do not
know how this payments burden will
actually be distributed among future
generations. For the purpose of illustration, however, it can be assumed to be
distributed equally except for an adjustment for economic growth. Future
generation-specific population projections can then be used to obtain the per
capita net-payment burden on future
generations.

• Generational Accounts as of 1989
A breakdown of the net payments of current generations according to types of
receipts and payments is given in table 1.
Several features are notable: First, the
GAs show a significant life-cycle profile
of the distribution of remaining lifetime
net-payment burdens on current generations. Young and middle-aged generations make positive net payments to the
government in present value, while older
generations receive, on net, in present
value, mainly because of large Social
Security and Medicare benefits.
Second, middle-aged generations pay
more in present value than do the very
young, because the former are closer to
their high-earning and high-taxpaying
years. Third, older female generations

begin receiving, on net, earlier than
males because of Social Security survivor benefits and the higher mortality of
their generally older spouses. These
generations also receive, on net. more
than males because of their relatively
lower income and payroll tax liabilities
over their remaining life spans.
Particularly striking is the comparison between the GA for current (1989) male
newboms (573,700) and that for future
generations ($89,500). The growthadjusted differential between these two
numbers is 20.5 percent.8"9 In other
words, if the treatment of current generations is maintained as under current
policy, and if future projected per capita
government spending remains the same,
each member of every fjrjrr j r r

TABLE 2

ADDITIONAL PRESENT VA'LUE OF NET PAYMENTS
NEEDED TO EQUALIZE GENERATIONAL BURDENS
.THOUSANDS OF DOLLARS)

Males Aged:

0
10
20
30
40
50
60
70
80
Future
generations

Income
Tax

Payroll

fax

Sales/
Excise Tax

Capital
Income Tax

1.8
3.0
4.5
5.4
5.2
4.2
2.6
1.2
0.5

2.1
3.4
5.0
5.5
4.7
3.2
1.4
0.3
0.0

2.3
3.0
3.3
3.3
2.9
2.3
1.7
1.1
0.6

1.4
2.3
3.5
5.1
6.1
5.9
4.6
2.7
1.4

-13.5

-13.1

-12.9

-13.9

1.0
1.6
2.2
2.5
2.5
2.1
1.3
0.7
0.2

1.2
2.0
2.8
2.7
2.3
1.5
0.6
0.2
0.0

2.1
2.7
3.1
3.2
2.9
2.3
1.6
1.1
0.6

0.5
0.9
1.3
2.0
2.8
3.0
2.5
1.7
0.6

-6.6

-6.3

-5.5

-7.0

Females Aged:
0
10
20
30
40
50
60
70
80
Future
generations

SOURCE: Alan Auerbach. Jagadeesh Gokhale. and Laurence Kotlikoff. "Generational Accounts: A New
Approach for Understanding the Effects of Fiscal Policy on Saving." Scandinavian Economic Review,
forthcoming 1992.

will bear a 20.5 percent larger burden
than that imposed on current newborns.
This indicates that current fiscal policy
involves a substantial generational imbalance.
A generationally balanced fiscal policy, if
kept in place, will result in the balance being preserved through time. Under such a
policy, every new generation will pay an
amount that leaves the same growthadjusted, net-payment burden on subsequent generations. On the other hand, a
generationally unbalanced policy that requires current generations to pay too little
would require future generations to pay
more. If kept in place, each new generation would pay too little, and the relative
payment burden between current newboms and future generations would
worsen over time.

What if the government did nothing about
correcting the generational imbalance implied by current policy — say, for 10 or
20 years? A recomputation of GAs by extending current policy treatment to generations projected to be bom between
1990 and 1999 indicates that the growthadjusted differential in the GAs of newboms in 1999 and 2000 will increase to
35 percent. If generations bom before
2009 are treated as they would be under
current policy, the differential between
the GAs of those born in 2009 and 2010
will rise to 57 percent. A postponement
of policy directed at equalizing burdens
on current and future generations will
gradually exacerbate the generational imbalance in fiscal policy.
• Correcting the
Generational Imbalance
What can be done to bring the generational accounts of newborns in 1989 and

1990 into balance? Obviously, either
more revenue will have to be extracted
from generations currently living, or
government spending will have to be
reduced. Several avenues are available
for achieving this kind of fiscal balance.
Current generations' net payments can
be raised through higher income, consumption, or payroll taxation or through
a combination of these with reductions
in government spending.
Table 2 shows the effects on the netpayment burdens of different generations
when various tax rates are increased. In
each case, average tax rates are raised so
that the growth-adjusted GA for future
generations is equalized to that of 1989
newboms. The average income tax rate,
for example, would have to be increased
from 14.5 to 15.3 percent. As a result, the
present value of a 30-year-old "s net payment would rise by S5.400 for a male
and by S2.50O for a female. However, the
present-value gains to future male and
female generations would be SI3.500
and S6.600. respectively.
Reducing government consumption
alone, on the other hand, would leave the
net-payment burden on current generations unchanged, but would nevertheless
reduce the amount that future generations
would be required to pay. Calculations indicate that the generational imbalance
would be eliminated if government
spending were permanently reduced by
3.3 percent or by S37 billion annually.
It should be noted that each method of
correcting the generational imbalance
would result in a different profile of annual revenue increases through time.
For example, calculations show that
the 1990 increase in revenue from raising payroll taxes in order to eliminate
the imbalance would have been S37 billion, while that from raising capital income taxation would have been S33 billion. Hence, although all the policies
equalize net-payment burdens on current and future newboms. each will
generate a different stream of recorded
deficits through time.

•

Cnnclniinn

The government's fiscal policies can
have significant long-term effects on the
resources available to members of different generations. Such policies do not
affect all generations equally, and their
implications for cross-generational redistribution are not reflected in the current
deficit numbers. Policy evaluation would
be better served by looking at generational accounts, a system that reveals the differential impact of policy changes on the
resources of members of different current
and future generations.

•

Miller. "Playing by the Rules," Federal Reserve Bank of Minneapolis. Annual Report.
1990.
2. See Congressional Budget Office, The

Alan Auerhach is a professor of economics at
the University of Pennsylvania and an associate of the National Bureau of Economic
Research. Jagadeesh Gokhale is an

Economic and Budget Outlook: Fiscal Years

economist at the Federal Reserve Bank of

1992-1996. January 1991. Arelatedarticle is

Cleveland, and Laurence J. Kottikqffis a

by Howard Gleckman, "How Do We Meas-

professor of economics at Boston University

ure the Deficit? Let Us Count the Ways,"

and an associate cfthe National Bureau of

Business Week. February 18. 1991, p. 32.

Economic Research. The authors wish to

3. See Robert Eisner and Paul J. Pieper, "A

thank William Gavin and Mark Sniderman

New View of the Federal Debt and Budget

for helpful comments. Most of the results

Deficits," American Economic Review, vol.

reported here appear in two previous articles

74, no. 1 (March 1984), pp. 11-29.

by the authors (see the source lines in table 1

4.

Computations for the United States indicate that current policy (as of 1989) involves a substantial generational imbalance. If the treatment of all currently
living generations continues unchanged,
future generations will have to bear, on
average, a 20.5 percent larger lifetime netpayment burden as compared to the burden on current newboms. Undertaking
corrective policy changes soon is imperative if this large generational imbalance is
to be prevented from becoming worse.

Footnotes

1. See. for example. V.V. Chari and Preston

For a more detailed exposition of this

argument, see Laurence Kotlikoff. "Deficit
Delusion," The Public Interest, Summer

and table 2). Another resource is Kotlikoff.
Generational Accounts: Knowing Who Pays,
and When. For What We Spend. New York:
The Free Press, forthcoming

1992.

1986, pp. 5 3 - 6 5 .
5. See Martin Feldstein, "Social Security, In-

The views stated herein are those of the

duced Retirement, and Aggregate Capital Ac-

authors and not necessarily those of the

cumulation," Journal of Political Economy.

Federal Reserve Bank of Cleveland or of the

vol. 82. no. 5 (September/October 1974), pp.

Board of Governors of the Federal Reserve

905-26: Alan J. Auerbach and Laurence J. Kot-

System.

likoff. Dynamic Fiscal Policy. New York:
Cambridge University Press: and Lawrence H.
Summers. "Capital Taxation and Accumulation in a Life Cycle Growth Model." American
Economic Review, vol. 71, no. 4 (September
1981). pp. 5 3 3 - 4 4 .
6. This section is a very brief summary of
the computational methodology described in
Alan Auerbach. Jagadeesh Gokhale, and
Laurence Kotlikoff. "Generational Accounts:
A Meaningful Alternative to Deficit Accounting." in David Bradford, ed.. Tax Policy and
the Economy. National Bureau of Economic
Research and MIT Press. No. 5. 1991.
7. In the computations reported below, the
projected future spending and payment
amounts were adjusted for economic growth
at 0.75 percent, and the present-value calculations were based on a 6 percent interest rate.
8. The absolute magnitudes of the GAs are
sensitive to alternative labeling of government receipts and payments. One should
therefore look only at the policy-induced
changes in the GAs for current and future
generations and at the percentage differential
between the GAs for current newborns and
future generations. These are invariant to
changes in the way payments and receipts
are labeled.
9. The ratio between male and female net
payments for future generations is assumed
to be the same as that for current newborns.

Fr-nnnmlr- Cnmrrrrtarv i<: a semimonthb' periodical "'swished by 'he FeHeml Reserve Bank of Cleveland Copies of the titles listed
below are available through the Public Affairs and Bank Relations Department, 1-800-543-3489.
Forecast Accuracy and
Monetary Policy
Michael F. Bryan and William T. Gavin
January 1, 1991

Defending Zero Inflation:
All For Naught
W. Lee Hoskins
April 1, 1991

The Sectoral and Regional Effects of
Oil Shocks: Who's over a Barrel?
Paul W. Bauer and Susan M. Byme
January 15,1991

Public Subsidies for Private Purposes
William P. Osterberg
April 15,1991

1991 Outlook: Mild Recession,
Mild Recovery
John J. Erceg
February 1, 1991
The Effect of War
Expenditures on U.S. Output
Charles Carlstrom, Jagadeesh Gokhale,
and Sharon Parrott
February 15, 1991
Deregulation, Money,
and the Economy
John B. Carlson
March 1, 1991
Is This Really a "White-Collar
Recession"?
Randall W. Eberts
and Erica L. Groshen
March 15, 1991

Address Correction Requested:
Please send corrected mailing label to
the above address.

|
j
j

Material may be reprinted provided that
the source is credited. Please send copies
of reprinted materials to the editor.

Federal Funds Rate Volatility
Diana Dumitru and EJ. Stevens
August 15, 1991

Price Stability and Regional Diversity
W. Lee Hoskins
May 1, 1991

Increasing National Saving:
Are IRAs the Answer?
David Altig and Katherine A. Samolyk
September 1, 1991

The RTC and the Escalating Costs
of the Thrift Insurance Mess
Christopher J. Pike
and James B. Thomson
May 15, 1991

Financial Fragility and
Regional Economic Growth
Katherine A. Samolyk
and Rebecca A. Wetmore
September 15, 1991

Realignment in the
U.S. Motor Vehicle Industry
Michael F. Bryan and John B. Martin
June 1, 1991

Exchange-Market Intervention
and U.S. Monetary Policy
Owen F. Humpage
October 1, 1991

Understanding the Recent
Behavior of M2
John B. Carlson and Sharon E. Parrott
June 15, 1991

Some Observations on
Central Bank Accountability
W. Lee Hoskins
October 15, 1991

Price Stability and
World Economic Recovery
W. Lee Hoskins
July 1, 1991

Why U.S. Managers Might Be
More Short-run Oriented Than
the Japanese
Gerald H. Anderson
November 1, 1991

Do Excess Reserves
Reveal Credit Crunches?
Joseph G. Haubrich
July 15, 1991

Federal Reserve Bank of Cleveland
Research Department
P.O. Box 6387
Cleveland, OH 44101

An Anemic Recovery?
John J. Erceg and Lydia K. Leovic
August 1, 1991

BULK RATE
U.S. Postage Paid
Cleveland, OH
Permit No. 385