Full text of Business Review (Federal Reserve Bank of Philadelphia) : November/December 1981

View original document

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

Federal Reserve Bank of Philadelphia
IS S N 0007-70 11

18
91
NOVEMBERDECEMBER

EDITOR’S NOTE: LOOKING AT TAX POLICY
The national debate over tax policy reached a seeming climax earlier this year with the
passage of a new tax package for individuals and businesses. The aim of the law was to
increase incentives to work, save, and invest—to get the economy off its back and
growing again. Far from ending the tax debate, however, this legislation appears rather to
have spurred it on.

NOVEMBER/DECEMBER 1981

Federal Reserve Bank of Philadelphia
100 North S ixth Street
(on Independence Mall)
Philadelphia, Pennsylvania 19106

The BUSINESS REVIEW is published by
the Department of Research every other
month. It is edited by John J. Mulhern with
the assistance of Michael McNamara.
Artwork is directed by Ronald B. Williams.
The REVIEW is available without charge.
Please send subscription orders and
changes of address to the Department of
Research at the above address or telephone
(215) 574-6428. Editorial communications
also should be sent to the Department of
Research or telephone (215) 574-6426.
Requests for additional copies should be sent
to the Department of Public Services.
*

The Federal Reserve Bank of Philadelphia
is part of the Federal Reserve System—a

System which includes twelve regional banks
located around the nation as well as the
Board of Governors in Washington. The
Federal Reserve System was established by
Congress in 1913 primarily to manage the
nation’s monetary affairs. Supporting func
tions include clearing checks, providing coin
and currency to the banking system, acting
as banker for the Federal government, super
vising commercial banks, and enforcing
consumer credit protection laws. In keeping
with the Federal Reserve Act, the System is
an agency of the Congress, independent
administratively of the Executive Branch,
and insulated from partisan political pres
sures. The Federal Reserve is self supporting
and regularly makes payments to the United
States Treasury from its operating surpluses.

The present issue of the BU SIN ESS REVIEW contains two contributions to the ongoing
discussion. Stephen Meyer and Robert Rossana focus on the tax legislation and ask
whether it will have its desired effect on individuals. Their answer is that it won’t—that
since most people will not be paying a smaller tax in years to come despite reductions in
rates, the hoped-for incentive gains will not materialize. Only a much larger reduction in
tax rates, they suggest, would alter the incentives to any effect. Ira Kaminow points out
that beyond the issue of tax reduction lies the question of tax efficiency: how can the tax
burden be distributed so as to induce the smallest distortions in economic behavior? He
suggests that the social costs of our inefficient tax system are enormous, and he discusses
several proposals that could yield large efficiency gains.—J.J.M .

Did the Tax Cut
Really Cut Taxes?
By Stephen A. M eyer and Robert J. Rossana*

In response to slowing productivity growth
and a public outcry in favor of tax cuts,
President Reagan proposed and Congress
adopted a package of tax cuts for individuals
and businesses. For individuals the major
element of the tax package is a twenty-five
percent cut in personal income tax rates,
spread over three years.1

The President’s tax proposals were sub
jected to a lively debate, both in the legislature
and in the press. Much of the argument
focused on the question of how people would
respond to so large a cut in income taxes.
Would they save the extra take-home pay or
would they spend it? Would people respond
to higher after-tax wages by working harder
and longer? Would owners of small busi
nesses seek to expand and undertake more
investment as their personal tax rates were
cut?
Our analysis suggests that the three-year
cut in personal income tax rates will have
little effect on people’s behavior, because
few taxpayers will face lower tax rates in
1983 than they did in 1980. We construct
estimates of the new tax rates that households
in the U .S. will face over the next few years.
Our estimates indicate that the tax rate on
any given real income will not fall from 1980
to 1983. Inflation will continue to push
people into higher tax brackets (higher dollar

‘ Stephen A. Meyer is Senior Economist in the Money
and Macroeconomics section of the Research Depart
ment. He received his Ph.D. from Yale University.
Robert J. Rossana is Assistant Professor of Economics
at the Pennsylvania State University. He was previously
Senior Economist at the Philadelphia Fed.
■^Even though the President and Congress refer to a
“twenty-five percent cut” in personal income tax rates,
the new law actually provides a twenty-three percent
cut. On October 1,1981 tax rates were cut 5%. On July 1,
1982 tax rates will fall by a further 10% from their levels
on June 30, 1982. Then on July 1,1 9 8 3 tax rates will be
cut by 10% more, from their levels on June 30, 1983.
Overall this is equivalent to a 23% cut in tax rates from
their levels in mid-1981.

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

income) as fast as the tax rate on any given
dollar income falls. When social security
taxes are added to personal income taxes, we
find that most families will actually wind up
facing higher Federal tax rates in 1983 than
they did in 1980, if their dollar incomes keep
pace with inflation.
The twenty-five percent cut in personal
income tax rates over three years, as proposed
by President Reagan and enacted by Con
gress, is more properly described as an
attempt to offset built-in tax increases caused
by inflation than as a real cut in tax rates.
Unless inflation from 1981 through 1983
turns out to be much lower than the Adminis
tration has projected, Federal income tax
rates will be slightly higher (for the same real
income) in 1983 than now. Of course Federal
income tax rates will be substantially lower
in 1983 than they would have been without
the tax package adopted earlier this year.
Because people with the same real income
will face roughly the same tax rates as last
year, however, it is unlikely that the personal
income tax package adopted in 1981 will
induce people to change their real economic
behavior from what it is now.

leads to bracket creep ; as incomes rise just
enough to offset the effects of inflation,
taxes due on those incomes rise still faster.
For example, husbands and wives who filed
joint returns in 1980 and who earned a
$30,000 taxable income paid $6,238 in
Federal income taxes for that year. Now
suppose that all prices rise 10 percent. If
taxable income also rises by 10 percent
(enough to preserve its purchasing power),
these households would pay $7,348 in Federal
income taxes—an increase in tax payments
of nearly 18 percent. Bracket creep means
that real tax payments rise when inflation
occurs. This raises the real receipts of the
government just as if Congress had passed
legislation to raise taxes. These unlegislated
tax increases are a major factor accounting
for recent declines in real after-tax incomes.
. . . As The Social Security Wage Base
Rises. The decline in after-tax incomes has
been reinforced by changes in the social
security program. This system, designed to
provide a part of the retirement income of
older Americans, is financed by contributions
(taxes) levied upon both firms and workers.
Workers who participate in financing this
program are then entitled to its benefits
during their retirement years. With the pro
gram heading for apparent insolvency,
Congress has been forced to raise both the
tax rate and the wage base upon which taxes
are levied (see Table 1). The wage base is the
maximum amount of each worker’s wage
and salary income that is subject to social
security tax.
These changes have the effect of raising
the tax burden not only because the tax rate
has risen, but also because the wage base has
risen so sharply between 1978 and 1981.
Many people had wage income higher than
the old wage base, so they paid no social
security tax on part of the income they
earned during the year. With the increase in
the wage base, many of these individuals
now find themselves paying social security
tax on their entire wages; as a result they
now pay a larger fraction of their total

TAXPAYERS FEEL THE PINCH
Many Americans know that their incomes
have risen in the past few years, but after
paying taxes they feel unable to purchase as
much as they could previously. This obser
vation is generally correct. The interaction
of inflation with our progressive income tax
code, and rising social security taxes, have
combined to reduce after-tax real incomes.
Bracket Creep Raises Taxes . . . By social
consensus the U .S. constructed an income
tax code which requires those with greater
ability to pay to shoulder a larger burden in
financing the activities of government. As
taxable income rises so too does the marginal
income tax rate—the extra tax incurred on
each extra dollar of taxable income. How
ever, the tax system does not recognize the
difference between nominal and real (infla
tion-adjusted) income. This blind spot then

FEDERAL RESERVE BANK OF PHILADELPHIA

TABLE 1
SOCIAL SECURITY TAX RATES AND EARNINGS LIMITS
Year

Tax Rate (°?b)

Wage Earnings Maximum

1978

6.05

17,700

1979

6.13

22,900

1980

6.13

25,900

1981

6.65

29,700

1982

6.70

32,700

1983

6.70

35,700

N.B.

This data reflects increases in tax rates and wage income limits scheduled under current
law.

TABLE 2
FEDERAL INCOME AND SOCIAL SECURITY TAXES
AS A PERCENTAGE OF ADJUSTED GROSS INCOME (AGI)
AGI
(1978$)

1978

1979

(%)
1980

1981

13,000

13.8

13.9

14.8

17.4

15,000

15.2

15.3

16.4

18.9

17,000

16.7

20.5

20.4

19,000

17.6

17.9

19.0

21.6

22,500

18.8

19.4

20.9

22.9

27,500

20.7

21.2

22.7

24.6

40,000

25.7

26.3

28.2

30.1

N.B.
These Figures illustrate the rise in taxes paid by presenting average tax rates for a family
of four with one wage earner. We assume that almost all income is wage income, up to the level of
the social security earnings base. For simplicity we present tax rates for families who take the
standard deduction. Average tax rates for other families have moved similarly.
1981 average tax rates are shown before the tax cut. 1981 income taxes will be cut only iy4%,
(more at very high income levels), so 1981 average tax rates will be only slightly lower than shown
here.

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

earnings in social security taxes. If we
combine the effects of bracket creep and
higher social security taxes, most of us have
indeed experienced an increase in our real
tax payments during the past few years.
Table 2 shows why taxpayers are now up
in arms. The fraction of income taken by the
Federal government has risen substantially
since 1978. A family of four earning a
constant real income of $13,000 in 1978
dollars found that the Federal government’s
tax bite rose by 26 percent, in real terms,
during the past four years. For a $40,000 real
income the tax bite rose 17 percent. The
results are similar for other income levels.

Such is the stuff of which tax rebellions are
made.
TAXES AND LABOR SUPPLY:
MARGINAL TAX RATES
While most of the public outcry over taxes
has focused on the share of income paid in
taxes, economists have looked at marginal
tax rates. (The marginal tax rate is the
fraction of one additional dollar of income
that would be taxed away.) Economists have
focused on the marginal tax rate because it is
the marginal tax rate that affects incentives
to work, to save, and to invest (See: Tax
Rates And Incentives).

TAX RATES AND INCENTIVES
The debate about the economic effects of this year’s tax cuts involved, among other issues, the
likely impact of such policies upon incentives to work. Proponents of “supply-side economics”
argue that cutting marginal tax rates will increase incentives to work, thus raising the labor supply
available to firms. How does a tax cut do this?
One incentive that strongly affects people's willingness to work is hourly take-home pay. Cutting
marginal tax rates increases the take-home pay which one can earn by working additional hours. So
cutting marginal tax rates increases the real quantity of goods and services that an extra hour of
work will buy. That is, giving up an hour of leisure time (and working instead) allows a worker to
obtain more goods and services, compared to the amount that she would obtain by sacrificing an
hour of leisure when there is a higher marginal tax rate. When marginal tax rates are cut, some
workers respond to the opportunity to get more consumption than before by working (or becoming
willing to work) extra hours. And other people, who were not working, choose to enter the labor
force to try to take advantage of the increased after-tax wages.
But a tax cut can increase hourly take-home pay in two ways, which have very different effects on
the incentive to work extra hours. If only the marginal tax rate is cut (leaving unchanged the amount
of taxes a worker pays on his initial income), then the incentive to work extra hours is strong. One
can take advantage of a cut in the marginal tax rate only by working extra hours; working the same
hours as before leaves one’s after-tax income unchanged. So no one has an incentive to work less.
Some workers would be willing to put in more hours, and some the same number of hours, but the
total labor supply would rise.
However, if average tax rates are cut (so that taxes due on a worker’s initial income fall), but
marginal tax rates are left unchanged, then total labor supply would fall. Cutting the average tax
rate means that a worker’s spendable income rises if she works the same number of hours as
initially. She can actually work slightly fewer hours (have more leisure time) and still end up with a
somewhat higher after-tax income than before taxes were cut. Not surprisingly, some people
choose to work less, and enjoy more leisure activities, when only the average tax rate is cut. So the
total labor supply would decline.
These two offsetting influences on labor supply suggest that Congress should be careful about
how it cuts taxes, if the objective of a tax cut is to induce people to work more. Giving each taxpayer
a tax cut by allowing him to calculate his income tax on today’s forms, and then subtract $500 from
the taxes due, would lower the average tax rate without affecting the marginal tax rate. This would

FEDERAL RESERVE BANK OF PHILADELPHIA

reduce labor supply. On the other hand, cutting tax rates applicable to each income bracket and at
the same time abolishing the personal exemption (now equal to $1,000) could lower marginal tax
rates without substantially affecting the average tax rate on a worker’s initial income. Doing this
would provide a strong incentive to work additional hours, so labor supply would rise.
The tax cut adopted in 1981 actually provides for an across-the-board cut in personal income tax
rates. If rates had been cut enough to offset the effects of bracket creep and higher social security
taxes, then both marginal and average tax rates would fall. How would this have affected labor
supply? The evidence suggests that an across-the-board cut in income tax rates would generate a
small increase in people’s willingness to work.* If this extra labor supply were put to use by
employers, then real GNP would rise.
Should the marginal tax rates that are relevant for workers’ labor supply decisions include the
employee’s share of social security taxes? It could be argued that higher social security taxes won’t
reduce incentives to work, because those taxes buy higher benefits when a worker eventually
retires. In reality, however, the social security benefits that any individual stands to receive in the
future are not closely related to the social security taxes that she pays today. Rather, future benefits
are determined by what Congress chooses to enact at that time. Today’s social security taxes pay for
today’s benefits. Because there is no direct link between social security taxes paid today and the
future level of benefits, today’s social security taxes affect labor supply decisions in the same ways
as today’s income taxes. Thus the wage rate which is relevant for the decision about whether or not
to work an additional hour is the after-tax wage, net of both income and social security taxes.
The impact of taxes on potential GNP is more properly measured by looking at effective marginal
tax rates rather than statutory rates. Effective tax rates measure the extra taxes paid, given the tax
preferred status of many types of spending or saving, associated with an extra dollar of income from
whatever source. These numbers are generally unavailable, but some estimates have recently been
constructed which are close to this concept.** These effective rates are generally lower than those
which we report, reflecting the fact that additional income is often put into tax shelters which lower
effective tax rates. Nonetheless, inspection of these rates over time suggests that there is little
reason to expect effective rates to behave differently from our estimates of statutory rates.
*See Footnote 2 to main text.
**See John J. Seater, “Marginal Federal Personal and Corporate Income Tax Rates in the U .S ., 1909-1975,”
Journal of Monetary Economics (forthcoming).

Cutting marginal tax rates raises the after
tax wage earned by working additional
hours. Proponents of “supply-side” policies
argue that workers will respond to higher
after-tax wages by working more hours,
which will result in greater output of goods
and services in the U .S. economy. Although
economists do not know just how large this
increase in labor supply would be, available
studies do indicate some response to changes
in tax rates.2

On the other hand, raising marginal tax
rates reduces the incentive to work additional
hours. Marginal tax rates have risen even
faster than average tax rates over the past
four years. The columns for 1978 through
1981 in Table 3 tell the story. A striking
example is given by the case of a family of
four earning a constant real income of $19,000
per year (in 1978 $). When that family earned
$19,000 in 1978 it faced a marginal Federal
income tax rate of 25 percent applied to
every extra dollar of taxable income. Their
total marginal rate (including social security
taxes) was the same, as their wage income
was likely to be well above the social security
maximum for that year. Now let’s see how
they fare in 1981. With the same family size

^See A. Protopapadakis, “Supply-Side Economics:
What Chance for Success?” Business Review, Federal
Reserve Bank of Philadelphia (July/August, 1981) for a
discussion of empirical estimates of labor supply re
sponses to changes in tax rates.

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

TABLE 3
MARGINAL TAX RATES
WITHOUT TAX CUT
AGI
(1978$)

1978
1979
1980
1981
1982
1983
Fed. Total Fed. Total Fed. Total Fed. Total Fed. Total Fed. Total

13000

.22

.28

.21

.27

.21

.27

.24

.31

.24

.31

.24

.31

15000

.22

.28

.24

.30

.24

.30

.24

.31

.28

.35

.28

.35

17000

.25

.31

.24

.30

.28

.34

.28

.35

.28

.35

.32

.39

19000

.25

.28

.34

.28

.34

.32

.39

.32

.39

.32

.39

22500

.28

.32

.39

.37

.44

.37

.44

27500

.36

.37

.43

40000

.45

.43

.49

.54

N.B.

Fed. = Marginal rate from Federal tax code.
Total = Sum of Federal marginal rate and social security rate.
Data apply to joint return of four person household using standard
deduction. Tax rates are rounded to the nearest percent.

and constant real income, that family faces a
marginal income tax rate of 32 percent. If an
extra dollar of income comes from wages
(rather than interest or dividends), the
government taxes away almost 39 percent—
an increase of 14 percentage points above
their 1978 total marginal tax rate!3 Marginal
tax rates rose from 1978 to 1981 for other
income classes, but by less. Economists
worry that increases in marginal tax rates
have reduced work effort and investment in
new capital equipment, thereby contributing

to lagging productivity growth in the United
States.
If the new tax package is intended to
increase the number of hours people wish to
work, it must actually reduce marginal tax
rates on wage income. Does President
Reagan’s tax package really reduce marginal
tax rates? We know enough about the details
of the tax package adopted this year to
construct some good estimates of the mar
ginal tax rates that workers wil face in the
next few years.
We constructed estimates of marginal tax
rates that households would have faced had
there been no tax cuts, as well as estimates of
tax rates that will result from the tax pack
age adopted in 1981. (See the APPENDIX for
complete details on the method of calcu
lation.) While it is obvious that marginal
rates will indeed be lower than they would

3Keep in mind that the social security rates are based
on wage income which is typically below AGI, so that
exactly which income classes are below the social
security wage base is not known with precision. Also,
Table 1 embodies tax rate and wage base limits which
are currently scheduled, but which may change in
future legislation.

FEDERAL RESERVE BANK OF PHILADELPHIA

otherwise have been, the vast majority of
U .S. households will still face higher mar
ginal tax rates than they faced in 1980 or in
1978, which is the last year for which detailed
tax and income data are available.
Marginal Tax Rates Were Scheduled to
Rise . . . Bracket creep and increases in
social security taxes will continue between
1981 and 1983. So if no personal income tax
cuts had been enacted by the Congress,
marginal tax rates would have risen sub
stantially. This is shown in the columns for
1980 through 1983 in Table 3. Families at all
income levels would have found themselves
facing higher marginal tax rates.
A family with adjusted gross income (AGI)
in 1980 equal to $22,500 in 1978 dollars
($26,990 nominal income in 1980) would
have faced both higher income tax rates and
higher social security tax rates in 1983 if its
dollar income grew at the inflation rate.
Though no better off in real terms, that
family’s marginal income tax rate would
have risen from 32 percent in 1980 to 37
percent in 1983. The social security tax on an
extra dollar of wage income would have
risen from zero to 6.7 percent, as both the
social security tax rate and the wage base
rose. So the total marginal tax rate on an
extra dollar of wage income would have
risen from 32 percent in 1980 to 43.7 percent
in 1983, if the Congress had not passed a tax
cut.
Of course Congress did enact a cut in
personal income taxes. What will happen to
marginal tax rates under the tax program
proposed by the President and adopted by
Congress?
. . . And Will Rise Even With the 25Percent Cut. The Treasury Department has
issued tax tables which embody the tax cuts
proposed by the Reagan Administration and
adopted by Congress. Using these we can
construct a set of (new) marginal rates and
compare these to previous results. Our
analysis suggests that when these new rates
are compared to those in 1980, most people
will find that they really haven’t received

complete relief from bracket creep, let alone
the effects of rising social security taxes.
Two sets of estimates are provided—one
for a household using the standard deduction
and one for a family which itemizes deduc
tions (see Table 4).
Our results show that for families of four
using the standard deduction, no family with
an Adjusted Gross Income in the range from
$13,000 to $40,000 (in 1978 $) will face a
lower personal income tax rate or combined
marginal tax rate in 1983 than in 1980. This is
true even after taxes are cut in the way
suggested by the Reagan administration. For
those in the lowest income class ($13,000
1978 dollars), the Reagan program will offset
bracket creep and the total marginal rate will
be nearly unchanged as well. As incomes
rise, the gap between 1980 and 1983 rates
will widen, with those in $22,500 class
facing a total marginal rate that will be
roughly one-quarter higher in 1983 than it
was in 1980. Even ignoring social security
taxes the Reagan program will not quite
offset bracket creep; all but two income
groups will face marginal Federal income
tax rates in 1983 which exceed 1980 rates.
The comparison is even more dramatic
when we look at 1978 and 1983 marginal tax
rates for households which claim the standard
deduction. The total marginal tax rate will
rise substantially for all but the lowest income
level, and even for that group the marginal
tax rate will rise somewhat. (Compare the
1978 column in Table 3 with the 1983 column
in Table 4.)
For those who itemize, the result is much
the same. Comparing 1980 with 1983 mar
ginal tax rates, the lowest income group will
see a slight decline in its marginal personal
income tax rate. All other groups will ex
perience flat or rising marginal tax rates. The
same is true when we add in social security
taxes. Indeed, those with 1980 incomes of
$22,500 (in 1978 $) will find themselves
facing a total marginal tax rate which is more
than one-third higher in 1983 than it was in
1980.
9

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

TABLE 4
MARGINAL TAX RATES AFTER REAGAN TAX CUT
Household Of Four Filing Jointly
(Using Standard Deduction)
1982
1980
1981
Fed. Total
Fed. Total
Fed. Total

AGI
(1978$)

1983
Fed. Total

13000

.21

.27

.22

.29

.22

.29

.22

.29

15000

.24

.30

.22

.29

.25

.32

.25

.32

17000

.28

.34

.25

.32

.25

.32

.28

.35

19000

.28

.34

.28

.35

.28

.34

.28

.35

22500

.32

.28

.35

.33

.40

.33

.40

27500

.37

.39

40000

.39

.44

.49

Household of Four Filing Jointly
(Itemizing Deductions)
1980
1981
1982
Fed. Total
Fed. Total
Fed. Total

AGI
(1978$)

1983
Fed. Total

13000

.18

.24

.16

.23

.16

.23

.16

.23

15000

.18

.24

.19

.26

.19

.26

.19

.26

17000

.21

.27

.19

.26

.22

.29

.22

.29

19000

.21

.27

.22

.29

.22

.29

.25

.32

22500

.24

.25

.32

.25

.32

.28

.35

27500

.32

.28

.33

40000

.43

.39

.44

N.B.

Tax rates are rounded to the nearest percent.

We see the same results when we compare
1983 tax rates with 1978 rates.4 Marginal tax

rates will rise for all families who itemize
deductions, except those in the lowest
income class.
Our conclusions about marginal tax rates
from 1981 through 1983 depend upon pro
jections of inflation, as explained in the
APPENDIX. If inflation turns out to be less

4In 1978, for those itemizing deductions, marginal
tax rates with totals in parentheses are: $13000-. 19 (.25),
$15000-.19 (.25), $17000-.22 (.28), $19000-.22 (.22),
$22500-.25 (.25), $27500-.28 (.28), $40000-.39 (.39).

FEDERAL RESERVE BANK OF PHILADELPHIA

than we have projected, then people will not
be pushed into higher tax brackets as rapidly
as we have calculated, so marginal tax rates
for families with constant real income [before
tax) actually might fall from 1981 to 1983. On
the other hand, if inflation were to continue
at its current rate, then marginal tax rates
would rise more than shown here. Our calcu
lations are based on virtually the same total
inflation during 1981 through 1983 as that
projected by the Reagan administration in its
fiscal 1982 budget proposals.
We have not estimated marginal tax rates
for those families with really high incomes,
because the tax code provisions which apply
to those families are so complicated. Various
tax shelters tend to reduce the true tax rates
faced by high income families. On the other
hand, interest and dividend income earned
by these same people has been taxed at
higher rates than wage income, which tends
to raise their marginal tax rates. The new tax
legislation adopted in 1981 affects very high
income families in two ways. First, the top
tax rate on interest and dividend income is
reduced from 70 percent to 50 percent. Second,
the top tax rate on wage and salary income is
not cut; it remains at 50 percent. So those
who have very high incomes will definitely
face a lower marginal tax rate on their
interest and dividend income in 1983 than
they did in 1980. As with the rest of us,
however, high income families will find no
cut in the marginal tax rate on their w age
income.

The 1981 tax cuts certainly do cut rates
from levels that would otherwise be achieved,
but not by enough to lower marginal tax
rates from current levels. Most taxpayers
will find that the trend of rising taxes will
continue. Except for one income group,
every other household studied here is going
to face a higher total marginal tax rate in
1983 than it did in 1978 and 1980.5 Insofar as
rising taxes are reducing productivity growth,
they still will be, although to a lesser extent.
We find that the Reagan tax cuts can only be
viewed as an imperfect attempt to offset
bracket creep.
Perhaps the most important aspect of the
tax package adopted in 1981 is the decision
to index the tax code beginning in 1985.6 If
done properly, indexing can prevent bracket
creep and thus automatically prevent de
clines in labor supply and potential GNP
caused by rising marginal tax rates. Although
Congress adopted the indexing provision
with little debate, it is clearly one of the most
significant changes in the personal tax code
in recent memory.
Did this year’s tax cut really cut taxes? Tax
rates will be lower than they would other
wise have been. Tax rates on a constant real
income will be higher in 1983 than they were
in 1980, however. Bracket creep and higher
social security taxes will more than offset
the 25-percent reduction in income tax rates.
Most families will find themselves facing
higher marginal tax rates in 1983 than they
did in 1980.

THE BOTTOM LINE
A program of cutting marginal tax rates
could have a substantial impact upon pro
ductivity growth and potential output, if it
succeeded in stimulating labor supply in
response to higher after-tax real wages. To
do this, and thus achieve some of the ob
jectives set out by its advocates, the Reagan
program would have to cut tax rates to offset
bracket creep and social security tax hikes,
and then some. This year’s tax cuts are not
big enough to do so.

5The share of personal income going to Federal taxes
will also rise slightly through 1983. The share of GNP
going to Federal taxes will decline slightly, however,
because corporate taxes are cut by the 1981 tax law.
®Each year from 1985 on. ’income tax brackets are to
be adjusted by the percentage increase in prices that
occurred during the year ending the previous September
30.

A P P E N D IX . . .
ii

. . . C O N S T R U C T IN G M A R G IN A L
IN C O M E T A X R A T E S
For the purpose of constructing marginal personal income tax rates, we require detailed
information on taxes paid, deductions, and adjusted gross income (AGI) for U.S. households. The
last year for which such data are available is 1978. This information is provided by the Internal
Revenue Service in a publication entitled “Individual Income Tax Returns, 1978 Statistics of
Income,” Publication 79 (3-81).
Choosing seven AGI classes where husbands and wives filed joint returns, we define the AGI of
the typical households in each group to be the mid-point of the AGI range for that class. For
example, AGI is assumed to be $13000 in 1978 for households in the $12000-$14000 AGI class found
in the “Statistics of Income.”
Exemptions claimed per return averaged 3.7, so for simplicity we assumed that each household
claims four exemptions.
For a household of four that does not itemize deductions, Taxable Income (TI) is computed using
the formula
(1)

AGI - 4 • ($/EXEMPTION) = TI

Given TI, we can refer to the tax table to obtain the relevant marginal, statutory tax rate. Dollars per
exemption were $750 in 1978 and $1000 in 1979 and beyond.
For those who itemize, we can use the “Statistics of Income” to find deductions per itemized
return (D) in 1978. We arrive at TI using the formula
(2)

AGI - (D-ZB) - 4 • ($/EXEMPTION) = TI

where ZB is the zero bracket amount (the amount of taxable income at the zero percent rate). For
joint returns, zero bracket amounts are $3200 in 1978 and $3400 for 1979 onward.
Finally, to compute the marginal tax rates for years subsequent to 1978, we raised 1978 AGI and D
at the inflation rate (actual or forecasted), applied formula (1) or (2) and then took the resulting TI
figures into the tax tables to find the reported marginal tax rates. Tax rates under the Reagan
program were obtained from tax tables provided by the Treasury Department. To measure inflation
we used the actual or projected rate of growth of the implicit deflator of personal consumption
expenditures (a good measure of inflation). Inflation projections for 1981 to 1983 were obtained
from recent economic forecasts made by Data Resources, Inc. Projected inflation rates are 8.8
percent in 1981, 8.7 percent in 1982, and 7.8 percent in 1983. In any given year these rates differ
from the Administration’s forecast, but over the whole 3 years our assumptions about inflation are
virtually identical with those the Reagan Administration used in its fiscal 1982 budget proposals.
We have ignored state and local income taxes in making these estimates. The reported tax rates
are those for Federal taxes alone. Including state and local taxes would raise the total marginal tax
rates faced by workers. Unless state and local income tax rates were to fall between 1980 and 1983,
however, our conclusions about the changes in marginal tax rates would be unaffected by including
state and local income taxes.

FEDERAL RESERVE BANK OF PHILADELPHIA

The Philadelphia Fed’s Department of Research occasionally publishes research papers
written by staff economists. These papers deal with local, national, and international
economics and finance. Most of them are intended for professional researchers and
therefore are relatively technical.
For a complete list of RESEARCH PAPERS currently available, write to RESEARCH
PAPERS, Department of Research, Federal Reserve Bank of Philadelphia, 100 North
Sixth Street, Philadelphia, Pennsylvania 19106.

FEDERAL RESERVE BANK OF PHILADELPHIA

The Merits
of Efficient Taxation
By Ira P. Kaminow*

resources lost by the private sector to the
government; the excess burden is the loss to
all of society because resources are wasted
and misused in an attempt to avoid taxes.
Tax incentives to encourage labor and in
vestment are efficient only to the extent that
they lead to the kinds of work-effort and
capital that produce the goods we want,
when we want them, and at minimum cost.
Otherwise, these incentives contribute to the
excess burdens of wasted and misused
resources.
One recent study concluded, for instance,
that though half of all saving in America
escapes taxation (or is very lightly taxed),
these tax breaks are so ill-conceived on
efficiency grounds that they have done
nothing to make society better off on the
whole. In contrast, estimates of the potential
gains from a truly efficient tax system range
into the hundreds of billions of dollars.
Unfortunately, many of the proposals recent
ly under discussion in the tax-reform debate
failed to give due weight to efficiency matters.

The old foundations of American tax policy
have become suspect. Taxes that once
promised to finance a great society are rou
tinely criticized for destroying incentives to
work and save. And interest in the distri
bution of tax burdens between rich and poor
has, by all appearances, diminished con
siderably in the past several years. For the
moment at least, incentives have replaced
equity as the key concern of the tax-reform
debate. But many economists believe the
incentives issue is overemphasized and
would prefer to focus more attention on tax
efficiency.
An efficient tax system permits society to
meet private and public demands without
incurring unnecessary costs—costs which
economists label the excess burden of taxa
tion. The direct burdens of a tax are the
*Ira P. Kaminow, formerly Vice President and
Economic Advisor at the Federal Reserve Bank of
Philadelphia, is Director of Economic Studies for The
Government Research Corporation, Washington, D.C.

NOVEMBER/DECEMBER 1981

BUSINESS REVIEW

BURDENS AND EXCESS BURDENS

that society is a net loser. In economists’
jargon, the lower the excess burden, the
more efficient the tax system and the
economy.
Tax efficiency is an old concept that in
volves careful distinctions between tax
revenues and tax structures—that is, between
how much we collect in taxes and how we
collect taxes. One useful way to clarify this
concept is to compare two kinds of taxes—a
lump-sum tax and a tax on income. Unlike
an income tax, a lump-sum tax is a flat
assessment levied without regard to any
measure of economic activity or well-being.
Since a lump-sum tax is not tied to people’s
economic behavior, it imposes no burden
beyond the revenue they forgo; there are no
high tax rates to discourage work or savings
or otherwise impose excess burdens. The
direct burden of a lump-sum tax, however,
could be extraordinarily heavy and unfair.
An equally distributed assessment would
mean a tax bite of roughly $12,000 for a
family of four, regardless of income or wealth,
assuming today’s government revenue levels.
So most people consider a lump-sum scheme
impractical on equity grounds. But, practical
or not, the likely effects of such a tax are
highly instructive.
Most of us, if slapped with a flat $12,000
tax bill, would try to increase earnings
through overtime work, a second family
income, or some other means, in order to
cushion the tax burden. Contrary to some of
today’s fashionable rhetoric, high taxes
encourage work and other productive efforts.
And this is as it should be. When people
demand more from government, whether
because of external events such as war or
because of changing perceptions of govern
ment’s role in society, they should meet at
least some of the bigger tax burden by
increasing production. The rest will be met
by cutting back on private consumption. The
combination of additional effort and less
private consumption corresponds to the
higher total burden of increased government
spending.

On average, each man, woman, and child
in the nation pays Uncle Sam about $3,000 in
taxes each year—$3,000 that cannot be spent
for rent, food, movies, and bicycles. This is
the direct burden, per individual, of the
$700-billion Federal tax take. It is part of the
finance necessary to pay for what govern
ment buys. And while taxes can be reduced
if government substitutes other kinds of
finance (borrowing or printing money], the
burden of government can only be reduced if
government spending itself is reduced. A
missile does not take less steel, nor a public
hospital fewer bricks, if taxes are lower.
Shifting government finance among taxes,
borrowing, and printing-press money may
shift the burdens of government, but it can
not reduce these burdens.
The situation is quite different in the case
of the so-called excess burden of taxation.
To raise $700 billion a year in revenues, we
have developed a complex tax structure that
can distort incentives so that what is socially
productive often leads to private loss and
what is socially inefficient can be privately
profitable. This reflects the excess burden of
taxation—the burden over and above the
necessary shift of resources to the govern
ment to finance expenditures. Excess burden
includes, for example, work that isn’t done
because after-tax wage rates are so low, and
savings that people forgo in the face of a
negligible after-tax return.
The direct burden of taxes may or may not
be worthwhile.1 It all depends on how wisely
the government uses the resources. But the
excess burden is a dead-weight loss to the
economy at large because it reflects a loss to
some with no offsetting gain to others, so
■
’■Spending, taxation, debt, and inflation are highly
interrelated, and the costs and benefits of government
can be determined only after the size and mix of the four
have been analyzed as a unit. This article focuses on the
burdens of taxes and devotes little analysis to the costs
of inflation and government debt or the benefits of
government spending.

FEDERAL RESERVE BANK OF PHILADELPHIA

Higher tax burdens should also encourage
high levels of savings in a lump-sum tax
world. We save, for example, to buy homes
and cars, to go on vacation, and to fit our
children with braces. And we also could save
to pay taxes. If lump-sum assessments are
expected to come due during retirement, we
have to put something aside today to prepare
for them. Expectations of higher future tax
burdens, just like expectations of increases
in other future expenses, should trigger
higher private saving.
Imagine now that the hypothetical lump
sum tax is replaced by an income tax (while
government spending levels remain the
same). In moving from a lump-sum to an
income tax, incentives to work and save will
be less, since wages and interest payments
will be lower on an after-tax basis. Not only
do we lose, for private use, the resources
transferred to the government (just as we do
under the lump-sum tax), but production is
lost because of the disincentives. The value
of this lost output is part of the measure of
the excess burden of the income tax. But the
income tax is in no way unique in generating
an excess burden. Every tax levied on a use
ful economic activity imposes an excess
burden on society by discouraging that activ
ity.2
An efficient tax system will try to keep
these distortions to a minimum. But, contrary
to what some popular discussions seem to
imply, designing such a structure does not
involve recreating patterns of work and
savings as though government spending
didn’t exist. Rather the task is to approxi
mate—with due account to fairness and
other social objectives—private behavior
that would exist under a lump-sum tax. In
other words, the tax system should be de
signed to minimize the impact of tax dis
tortions (not of taxes) on private economic
decisions. This idea has some rather power

ful and diverse implications for evaluating
the efficiency of our current tax system and
of popular tax-cut proposals.
Consider, for example, the debate sur
rounding the likely success of supply-side
tax cuts. These cuts are aimed at reducing
tax rates in the hope of encouraging ad
ditional work and saving. Opponents of the
cuts are quick to point out that most statis
tical studies conclude that high taxes have
had little or no impact on work or production.
Cuts in taxes, they argue, cannot restore
what high rates have not taken away.
But there is another way to interpret this
evidence. W e’ve just argued that high taxes
should increase work effort, production, and
savings if they are to efficiently meet large
demands for government services. So evi
dence that our income tax has had little
impact on saving and labor supply need not
mean that the tax is innocuous. To the
contrary, it points directly to the excess
burden of the tax. When the tax burden
grows into the hundreds of billions, we
should be working and saving more so that
we can pay the tax without unduly sacrificing
our private standards of living. That we have
not suggests that the overall burden of the
income tax rates has entirely offset the
tendency to work and save more when our
tax obligations increase. This implies that
the overall burden of the income tax on our
economic well-being must be quite high.
EXCESS BURDENS AND
SOME POPULAR TAX PROPOSALS
America’s search for a more incentiveoriented tax structure has come down to a
number of variations on two major themes:
reduce the high marginal tax rates that have
weakened incentives to work and to invest,
and make additional tax cuts designed
specifically to encourage business invest
ment and private saving. Unfortunately,
these proposals and their variants are too
often evaluated only in terms of their likely
impacts on work, saving, and investment.
This focus can be extremely misleading.

2Taxes on harmful activities provide an excess gain.
The use of taxes to discourage undesirable activities is
another long-standing issue in tax theory.

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

tax bill. Under the Kemp-Roth scheme, the
average married man’s tax bill would fall to
$833; but because his tax rate also would fall
(by 30 percent], disincentives to work would
drop and his excess burden would be only
$128. If fully implemented, the Kemp-Roth
proposal would yield less revenue than last
year’s tax law. Therefore, for any given level
of government spending, the Kemp-Roth
proposal would require a larger government
debt or more inflationary printing-press
money than would the 1980 law. But both
government debt and inflation can impose
their own costs and excess burdens on society.
So a full comparison of Kemp-Roth with the
1980 tax code would require some measure
of these additional costs. Unfortunately,
Hausman could not easily measure them,
and as he points out, this makes it difficult to
make the comparison.
Hausman was able to compare the full
excess burdens under last year’s law and
under Friedman-type alternatives. He
designed variants of Friedman’s proposal
which would match tax revenues under the
1980 tax code dollar for dollar. Under one
such variant, the first $4,000 of income
would be exempt from any tax. All income
over $4,000 would be taxed at a rate of 20.7
percent. According to Hausman, this plan
would be superior to 1980 law since it would
raise the same revenue but would cut the
excess burden of labor supply disincentives
in half. Hausman also claims that the tax
would be at least as progressive as 1980 law,
at least for incomes up to $25,000 or so.
While marginal tax rates under this scheme
are not progressive, average tax rates are,
because the first $4,000 of income is tax free
(see A FLAT TA X RATE . . .).
The case that Friedman, Hausman, and
others make for the low-top-marginal-taxrate plan is remarkably persuasive. With
little or no loss of revenue, the plan sharply
reduces excess burden while keeping the
distribution of the direct tax burden roughly
unchanged. In short, the plan seems to offer
substantial gains in effic ien c y at little cost in

Even when taxes have no apparent impact
on work, saving, or investment, they can still
impose substantial excess burdens. And
work, saving, and investment that are used
inefficiently or that produce goods no one
wants are not very helpful.
Personal Tax Rate Cuts. In 1978, Con
gressman Jack Kemp and Senator William
Roth proposed an across-the-board cut in
personal income tax rates of 10 percent a
year for three consecutive years. Since then,
the idea of a cut in tax rates has moved
steadily toward the center stage of American
politics. In July, the Congress passed Presi
dent Reagan’s program, which reduces tax
rates 25 percent over the next three years.
Some students of tax theory favor a still
bolder approach to tax-rate reductions. The
idea is to lower the top tax rate dramatically,
and it’s most closely identified with Nobel
laureate Milton Friedman, who proposes
limiting the top rate to a mere 25 percent.3
Friedman predicts that slashing the top rate
would ease the worst of the income tax’s
excess burden with little or no revenue loss.
He points out that in 1977 only about 13
percent of all Federal revenue came from
rates above the 25-percent mark. And with
the top rate cut by almost two-thirds from the
present 70 percent, the incentives to use
revenue-draining tax shelters would decline,
perhaps bringing the government even more
revenues.
A recent study by Jerry A. Hausman
attempted to compare the excess burden
under the 1980 tax system with that under
the Kemp-Roth scheme and under variants
of the Friedman proposal.4 Some of the
estimates are quite interesting. The average
married man in Hausman’s sample now pays
a tax of $1,077. In addition, as a result of taxinduced disincentives to work, he bears an
excess burden of $235 or 21.8 percent of his
^Newsweek, August 18, 1980.
4Jerry A. Hausman, “Income and Payroll Tax Policy
and Labor Supply,” National Bureau of Economic
Research, Working Paper No. 610, December 1980.

FEDERAL RESERVE BANK OF PHILADELPHIA

A FLAT TAX RATE
CAN YIELD A PROGRESSIVE TAX SYSTEM
Average Tax Rate

Gross Income
Current
Law

Alternate
Proposal*

$ 4,000

.119

8,000

.104

.147

16,000

.155

.173

24,000

.172

.188

’ First $4,000 of income tax free, 20.7-percent rate on all income above $4,000.
SOURCE: Hausman, “Income and Payroll Tax Policy and Labor Supply.”

differential between two similar activities is
large, people will shift out of the useful but
heavily taxed activity and into the activity
subject to little or no tax. The more people
try to beat taxes by finding loopholes, the
more will private activity be distorted per
dollar of revenue raised.
What does this have to do with the effi
ciency of taxing income from savings? Each
of us is faced with the choice among leisure,
current consumption, and future consumption
(see THE REAL CHOICES overleaf). An
efficient tax system will tax these three
activities in such a way as to discourage
switching among them to beat the tax man.
But if wages are taxed, choices will be
distorted away from consumption and toward
more leisure. People can avoid the wage tax
by not working. The efficiency of a savings
tax depends on whether the wage tax, in
addition to reducing the total amount of
consumption, also distorts lifetime con
sumption patterns. If the wage tax does not
distort consumption patterns, but merely

terms of equity. It is a proposal that should
be taken far more seriously in the future, as
we try to make further progress toward the
ideal tax system.
Savings Incentives. Among the most hotly
debated questions of tax policy is how to
treat savings. Should income from savings
be taxed like other income? Perhaps, as
many have suggested, we should eliminate
the tax on saving altogether and tax only
consumption expenditures. Or maybe savings
should be subsidized. The proper tax treat
ment of savings is one of the most crucial,
but difficult, areas to address from an effi
ciency perspective. The way savings are
treated can magnify or reduce the excess
burden associated with other taxes—in
particular, a tax on wages. Unfortunately,
though, a key piece of information necessary
for deciding whether to tax savings or even
subsidize them has yet to be uncovered.
A straightforward rule for efficient taxation
is easy to state: similar activities should be
taxed at similar rates. When the tax-rate

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

THE REAL CHOICES
The focus on the trade-off between consumption and saving can be quite misleading and can
understate the true distortions of taxation. Consider, for example, the case of the identical clones
Mervyn and Marvin.
Mervyn lives in the most fictitious of all countries. It has neither inflation nor taxes. Mervyn has
been earning $2,000 a month since he started working at 25 and will go on doing so until he retires at
age 65. Each month, Mervyn spends $1,670 and puts $330 in the bank at 3-percent interest.
Marvin also lives in a fictitious country. It has no inflation and no wage tax, but all interest
income is taxed at 50 percent. Like Mervyn, Marvin will work from age 25 to age 65; he earns
$2,000, spending $1,670 and saving $330 at 3 percent (before taxes) every month.
Judging by their consumption and saving decisions, Mervyn and Marvin behave identically and
the tax on interest income has had no impact on Marvin’s choices. But look again. As the example
works out, untaxed Mervyn will have enough in the bank to go on spending $1,670 a month until he
is 85. Marvin, who has been taxed on his interest income, retires with far less in the bank. Poor
Marvin will be able to spend only about $850 a month if he lives to 85.
When Marvin chose to keep saving unchanged despite the tax, he chose, in a more meaningful
sense, to bear his entire tax burden during his retirement. If he had wanted to spread the tax burden a
little more evenly between his youth and old age, he might have increased his savings to $400 a
month. This would allow retirement spending of $1,264 a month to age 85. Even if the tax
encourages Marvin to save more, it still can reduce retirement consumption relative to consumption
during the working years. The reason is that part of Marvin’s saving will go to pay taxes.
The often discussed trade-off between work and leisure misses the point as well. Leisure (the
consumption of time) is an end, work is merely a means to acquire and consume goods sold in the
marketplace. The real choice is between the two ends: the consumption of time and the
consumption of goods. If a wage tax has no effect on the individual’s choice between work and
leisure, the entire tax burden falls on the consumption of goods (because lower after-tax income will
allow the purchase of fewer goods). To distribute the tax burden more evenly between the
consumption of time and the consumption of goods, labor supply must increase, and leisure time
must fall.
Analyzing the impact of taxes on economic well-being, therefore, requires unraveling taxation’s
effect on the quantity and mix of consumption. Today’s tax problems should be examined in the
context of the trade-off among three basic goods: leisure, current consumption, and future
consumption. Work, saving, and investment are of only indirect concern.

trillion. 5
Suppose, however, that the wage tax not
only reduces consumption spending but also
redistributes it over time. Perhaps the tax
causes us to cut back on future consumption
more than current consumption. Then, the
wage tax reinforces the distortions of a tax
on savings income: both taxes encourage
current consumption and discourage saving

reduces all consumption expenditures,
present and future, by the same percentage,
income from saving should not be taxed. The
addition of a tax on saving would do nothing
but increase excess burden by discouraging
savings (hence future consumption) and
encouraging current consumption. Another
distortion would be added to that created by
the wage tax. Lawrence Summers has
estimated the excess burden of a tax on
savings (assuming that a wage tax has this
proportionate effect on present and future
consumption) at 10 percent of each year’s
GNP or an astronomical total cost of $20

^Lawrence Summers, “Taxation and Capital A c
cumulation in a Life Cycle Growth Model," American
Economic Review, forthcoming.

FEDERAL RESERVE BANK OF PHILADELPHIA

once to the investor as dividends), but only
the corporation pays a tax on undistributed
profits. Also, because of the investment tax
credit on business equipment, the tax rate on
investment in equipment is only about 60
percent of the rate on investment in inven
tories and factories. Finally, taxes on contri
butions to employee retirement programs
are deferred, while taxes on savings account
deposits typically are not.
Several statistical studies show that the
wide variation in tax rates on similar activi
ties is indeed a major problem in the tax
treatment of savings and investment. Don
Fullerton, J. B. Shoven, and John Walley
conclude that though half of U .S. savings
income is untaxed, the excess burden from
disparate rates on different kinds of invest
ment and saving wipes out any efficien cy
gains.7 They estimate that taxing all personal
savings at the same rates and eliminating the
corporate income tax (to prevent double
taxation of corporate profits) would reduce
the national excess burden by about $200
billion. Many of the varied proposals recently
under discussion to increase savings and
investment by reducing taxes on this kind of
saving or accelerating depreciation on that
kind of investment would, at best, provide
only minimal reductions in excess burdens;
they could cause excess burdens to rise.
Charles Becker and Don Fullerton studied a
number of these proposals.8 Among their
conclusions is this statement:

for the future. Efficient taxation would, under
this condition, actually call for a negative tax
(that is, a subsidy) on saving for future
consumption and thereby offset the distortion
of the wage tax on consumption patterns.
The final possibility is that the wage tax
discourages present consumption propor
tionately more than it discourages future
consumption.6 Under this circumstance,
efficient taxation implies that a tax on saving
will discourage saving and restore more of a
balance between present and future con
sumption spending.
The proper tax treatment of savings from
an efficiency perspective clearly depends
very much on how taxes on wages affect
consumption patterns. Unfortunately, we
know very little about this issue. Until further
research unravels this mystery, it’s difficult
to say whether savings should be taxed,
exempted from tax, or subsidized.
While there is room for debate on the issue
of whether savings should be taxed, there is
no question but that our current tax treatment
of savings and investment violates the effi
ciency rule of similar taxes for similar activi
ties. Differences in taxes applying to savings
and investment abound, generating high
excess burdens as people take advantage of
loopholes to avoid taxes. Some examples:
If an individual buys a home for personal
use, no taxes are paid on the implicit rental
income; if that same individual sells the
home to a corporation and rents it back, the
business pays taxes on the rental income.
Again, distributed corporate profits are taxed
twice (once to the corporation as profits and

The plan most successful in terms
of generating new savings and
capital formation, is among the
least sucessful in terms of (effi

^Perhaps, we try to avoid the wage tax during our
working years by producing more at home and less on
the job—more home-cooked meals, more family enter
tainment and at-home education, less high-priced con
venience foods, movies, and outside piano lessons.
Avoid the tax man by avoiding the middle man. With
many current demands being met at home, relatively
more income is available to be saved. Present con
sumption expenditures fall relatively more than future
expenditures.

7Don Fullerton, J. B. Shoven, and John Walley,
“Dynamic General Equilibrium Impacts of Replacing
the U.S. Income Tax with a Progressive Consumption
T ax,” National Bureau of Economic Research, Con
ference Paper No. 55, October 1980.
^Charles Becker and Don Fullerton, “Income Tax
Incentives To Promote Saving,” National Bureau of
Economic Research, Working Paper No. 487.

BUSINESS REVIEW

NOVEMBER/DECEMBER 1981

leave it alone. But there are some clear cut
recommendations for designing a tax system
that fall out.
First, very high personal tax rates are
highly inefficient. And the top rates can be
lowered substantially with no loss in revenue.
Equity can be preserved by exempting low
income entirely from the tax. A persuasive
case can be made that neither fairness nor a
requirement for revenue demands a top per
sonal tax rate above 20 or 25 percent.
Second, while we may not know whether
to tax saving and investment, we do have
clear guidelines on how to tax them (if they
are taxed at all). Here the rule is simple: close
tax loopholes by taxing similar activities at
similar rates. As a rule of thumb, all invest
ment—whether in private homes, corporate
factories, or business machines—should be
taxed at the same rate. Otherwise, there will
be switches from efficient to inefficient
activities. Encouraging only some kinds of
investments or saving may not increase them
in the aggregate, and even if it does, it may
do nothing to reduce the apparently very
high excess burden in the U .S. tax system.

ciency) gains measures. The simu
lations serve to emphasize . . . that
increased capital is only valuable if
used properly.
Indeed, tax breaks for certain investments
have become so great under the 1981 tax law
that it may soon be profitable to buy a
machine just for the tax credits it produces,
even if it is left idle.
CONCLUSION
A tax system must consider equity if it is to
succeed; one that values only efficiency is
doomed to failure. But a tax system that
ignores efficiency can be quite costly to
society. The most popular objectives of tax
reform—incentives for work, saving, and
investment—often fail to hit the issue of
efficiency head-on. At best, they are mere
proxies for the true objectives of efficiency—
reduction of excess burdens. At worst, they
will lead to further inefficiencies.
Economic analysis and research leave
many questions about efficient taxation
unanswered. Most notable is the issue of
whether to tax or subsidize saving, or just

BIG GOVERNMENT,

a pamphlet written by
Lawrence C. Murdoch, Jr., Vice President at the Philadelphia
Fed, traces the growth of government in the United States
and puts recent calls for reducing the size of govern
ment into perspective. Copies are available without
charge from the Department of Public Services,
Federal Reserve Bank of Philadelphia,
100 North Sixth Street, Philadelphia, PA
19106.

Federal Reserve Bank of Philadelphia

FEDERAL RESERVE BANK OF PHILADELPHIA
BUSINESS REVIEW CONTENTS 1981
JANUARY/FEB RUARY

JULY/AUGUST

Laurence S. Seidman, “A Personal Consumption
Tax: Can It Break the Capital Formation Dead
lock?”
Robert J. Rossana, “Structuring Corporate Taxes
for a More Productive Economy”

John J. Mulhern, “The Defense Sector: A Source
of Strength for Philadelphia’s Economy”
John M. L. Gruenstein and Sally Guerra, “Can
Services Sustain a Regional Economy?”
SEPTEMBER/OCTOBER

MARCH/APRIL

Richard W. Lang, “Managing the Money Stock: A
Time of Transition”
N icholas Carlozzi, “Regulating the Eurocurrency
Market: What Are the Prospects?”

Brian Horrigan, “Indexation: A Reasonable
Response to Inflation”
Timothy Hannan, “Who Controls What in the
U.S. Economy?”
NOVEMBER/DECEMBER

MAY/JUNE

Donald J. Mullineaux, “Efficient Markets, Inter
est Rates, and Monetary Policy”
Aris Protopapadakis, “Supply-Side Economics:
What Chance for Success?”

100 North Sixth Street
Philadelphia, PA 19106

Stephen A. M eyer and Robert J. Rossana, “Did the
Tax Cut Really Cut Taxes?”
Ira P. Kaminow, “The Merits of Efficient Taxation”
Contents 1981

Full text of Business Review (Federal Reserve Bank of Philadelphia) : November/December 1981

FRASER