The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
ra©i1l\\
CG)
TI
JF1f
CC
D.
(Q)
May 8,1 981
Kemp-Rothand Saving
In his triumphal appearance before Congress
last week, President Reagan strongly reiterated his support for the Kemp-Roth Bill,
which would reduce Federal personalincome tax rates by ten percent in each of the
nextthree years. Thus, it's worthwhile
reviewing what the Administration hopes to
achieve through this controversial measure.
Under Kemp-Roth (according to this view),
workers would work more hours at a more
intense pace, because they would receive a
higher after-tax return for working. More
persons would enter the labor force, because
jobs would have higher after-tax rewards.
Unemployment would be lower, because
businesses would find more profit in hiring
and training unemployed workers. Government spending to help the disadvantaged
would contract, because the more prosperous economy wou Id reduce the need for
such spending. Business would invest more
in job-generating capital, because of lower
effective business-tax rates. Entrepreneurs
would work harder, innovate more, and take
greater risks, because such activities would
mean higher after-tax returns. The rich would
shift their wealth from unproductive tax
shelters to productive uses. Finally, savers
would save more, because they would
receive a higher after-tax return to saving.
Many Congressmen, businessmen and economists have rejected the Administration's
claims, however. They do not bel ieve that
workers wou Id work much more, that the
labor force would expand or unemployment
contract substantially, that the need for
government spending would fall much, that
the morale of entrepreneurs would improve
much, that the rich would desert their tax
shelters, or that savers wou Id save much
more. Indeed, one Congressman termed
these claims "hallucinatory," whfle another
Congressman asserted that he could not find
a shred of evidence to support any of these
claims.
Instead of analyzing all of these arguments in
detail, we would do well to concentrate on
the most controversial argument -that
Kemp-Roth would raise saving. Keynesian
theory, which has dominated the discussion
and formulation of macroeconomic policy
throughout most of the past generation, posits
that consumption is closely related to disposable (after-tax) income, so that households
spend a fraction of any additional disposable
income they receive on consumption goods.
Therefore, according to this theory, cutting
personal income-tax rates raises disposable
income and hence raises consumption. Private saving also rises because households
only consume a fraction of this increase in
disposable income, but government saving
(mi nus the government deficit) falls by the fu II
amount of the tax cut. Consequently, total
national saving must fall unless the tax cut
generates a massive boom-which
conceptually could bring in enormous amounts of
new tax revenue, reduce unemployment
compensation and other income-sensitive
transfer payments, and expand private
saving. However, in a pure Keynesian model,
it is theoretically impossible for the boom to
be so massive.
A few numbers might clarify this argument.
Suppose that Kemp-Roth reduces taxes by
$40 billion, and that households consume 90
cents of each additional dollar of disposable
income they receive. At the current level of
national income, government saving would
drop $40 billion, private saving would rise
only $4 billion (40 x (1-.9)), arid national
savingthus would drop by $36 billion. Therefore, in the absence of a major increase in
national income, Kemp-Roth would lower
saving.
In addition to raising disposable income,
however, Kemp-Roth would.raise the real
after-tax return to saving. The Administration
argues that the latter effect, which tends to
lower consumption, overshadows the former
JB)S\fr)\Jk<3J)
II
If i f
(Cli
CD)
Opinions
expressed in this newsletter do not
i'etlect the views of the rnanager'llent
of the Ferreral Reserve Ban k of
Francisco!
f1C)r of
i1o·arclcrf GO\/t:rrtC)fS eyf
FedE::raf
Reserve Systenl.
chart). The solid line shows the actual evolution of consumption between 1 954 and
1 967. The dotted line from 1 954 to 1963
shows the path of consumption implied by
the fitted consumption function, given the
actual path of disposable income. As the
reader can see, the dotted line is right on top
of the solid line during the 1 954-63 perioda period with no change in the Federal personal-income tax code.
effect, which tends to raise consumption. A
higher real after-tax return to saving would
lower consumption because it makes all of
the things for which households save-their
retirements, rainy days, and their children's
educations-more attractive relative to
current consumption.
Consider the case of a rich household that
wishes to save for its retirement 30 years
hence. If this household is in the 70-percent
tax bracket and makes a nominal before-tax
return of 15 percent, its nominal after-tax
return is 4.5 percent (15 x (1-.7)). If it expects
inflation to continue at the rate of 1 0 percent
a year, its real after-tax return is -5.5 percent
(4.5 - 10). Kemp-Roth would lower this
household's marginal tax rate to 50 percent,
thus raising its real after-tax return to -2.5
percent. Without Kemp-Roth, assets lose 5.5
percent of their value each year, so that sacrificing one unit of consumption now yields
.18 (= (1-.055)30) units of retirement consumption 30 years from now. But with KempRoth, that sacrifice yields .47 (= (1-.025)30)
units of retirement consumption. This effect
would be smaller, though still appreciable,
for all but those households in the lowest tax
brackets. Since Kemp-Roth would so markedly improve the trade-off between current
consumption and the future goals that saving
accomplishes, saving should rise sharply.
The Kennedy tax cut passed in February
1 964. One can clearly see that no sudden
surge in consumption occurred in 1964 or
thereafter. Indeed, it would be impossible to
pick out the year when the tax cut occurred
just from looking at the consumption series.
The dotted line labeled A shows how consumption would have evolved ifthe 1954-63
consumption function had continued to hold
in the 1 964-67 period. Here we can see that
the previous consumption function no longer
tracks actual consumption after the tax cut.
Households consumed much less and saved
much more than Keynesian theory would
have predicted. The difference is probably
due to the higher after-tax return to saving
produced by the tax cut.
The dotted line labeled B shows how consumption would have evolved, had there
been no tax cut in 1964. It is obtained by
predicting what disposable income would
have been without the tax cut, and then usi ng
the fitted consumption function to calculate
what consumption would have been if disposable income had followed that predicted
path. This shows that the tax cut actually
loweredconsumption. Since households
saved more than 1 00 percent of the tax cut,
national saving would have risen even if
national income had not risen. But no one
disputes that the Kennedy tax cut raised national income. Therefore, the tax cut raised
national saving by a substantial amount.
Which provides a better answer-the
Administration view or the Keynesian view of
consumption and saving behavior? To answer that question, we shou Id look at what
happened after the last major reduction in
Federal personal-tax rates, the 1 964-65 Kennedy tax cut. Since Kemp-Roth would take
almost exactly the same form, the exercise
should prove instructive.
An estimated consumption-income relationship (Keynes' "consumption function")
based on pre-1 964 data should fit the data
very tightly and, more importantly, should
predict what happened to consumption and
saving after the tax cut. The relationship does
indeed fit well in the 1 954-63 period (see
Our evidence impl ies that Keynesian
consumption functions are not stable when
tax rates change-specifically that the
2
as when it is low. For this reason, the change
in saving that Kemp-Roth would produce is
likely to be in the same direction and comparable in magn itude to that produced by the
Kennedy tax cut.
Kennedy tax cut lowered consumption and
raised saving. Kemp-Roth, which would essentially replicate this tax cut, therefore
should also lower consumption and raise
saving. Of course, this kind of prediction may
be somewhat risky to make, because the U.S.
economy has changed a great deal since
1964, and may therefore respond differently
to Kemp-Roth than to the Kennedy tax cut.
For example, many of the Administration's
. critics have argued that households would
not save the tax cut because the inflation rate
is now about ten times higher than it was in
1964. Even though this high inflation rate
does depress the real after-tax return to saving
and hence the level of saving, a reduction in
personal income tax rates affects the tradeoff
between current and future consumption in
roughly the same way when inflation is high
The author concludes that, despite all the
changes in the economy since 1964, the best
available evidence supports the Administration's position that Kemp-Roth would raise
saving. The critics who assert that there is not
a shred of evidence to support this claim just
have not looked for it.
PaulEvans
(The author, Assistant Professor of Economics
at Stanford University, is the Visiting Scholar
this semester at the Federal Reserve Bank of
San Francisco.)
$ Billions
(1972$)
700
Consumption
Function
650
600
550
500
450
400
-
Actual consumption
.Predicted consumption
350
o
T
I
1954
I
1956
I
I
I
1958
I
I
1960
3
I
I
1962
I
I
1964
I
I
1966
55"1 :> .L5!:11:i
U018U!4Sl?M.4l?ln • uo8aJo • l?pl?AaN • 04l?PI
Hl?Ml?H • l?!UJOJ!Il?) • l?UOZpv • l?>jSl?IV
em
(G)}{(\illW?
'meJ 'O:lSpueJ:Iues
Z:S'L:
'ON llWH:Jd
OIVd
: J9Vl SOd's'n
ll VW SSV1J lSHI:I
JJ.
BANKING DATA-TWELF TH FEDERALRESERVE
DISTRICT
(Dollar amounts in millions)
SelectedAssetsand Liabilities
large Commercial Banks
Loans(gross,adjusted)and investments*
Loans(gross,adjusted)- total#
Commercial and industrial
Realestate
Loansto individuals
Securitiesloans
Treasurysecurities*
Other securities*
Demand deposits - total#
Demand deposits - adjusted
Savingsdeposits - total
Time deposits - total#
Individuals, part. &;corp.
(LargenegotiableCD's)
Weekly Averages
of Daily Figures
Member Bank ReservePosition
ExcessReserVes(+ )/Deficiency (-)
Borrowings
Net free reserves(+ )/Net borrowed(- )
Amount
Outstanding
4/22/81
Change
from
4/15/81
147,091
124,749
36,736
51,781
22,822
1,526
6,596
15,746
41,924
29,928
30,994
76,385
67,520
29,657
325
280
108
107
35
171
15
60
-3,757
-1,680
- 474
984
888
771
Weekended
4/22/81
n.a.
228.0
n.a.
Changefrom
year ago
Dollar
Percent
7,932
7,425
2,077
5,406
- 1,719
801
31
480
- 1,624
- 2,553
4,604
12,089
12,123
6,858
Weekended
4/15/81
n.a.
40.0
n.a.
5.7
6.3
6.0
11.7
- 7.0
110.5
0.5
3.1
- 3.7
- 7.9
17.4
18.8
21.9
30.1
Comparable.
year-agoperiod
82
148
66
* Excludestrading account securities.
# Includes items not shown separately.
Editorial comments may be addressedto the editor (William Burke) or to the author. , , . Freecopiesof this
andother FederalReServepublications can beobtained by calling or writing the Public Information Section,
FederalReserveBank of SanFrandsco, P.O, Box 7702, SanFrancisco94120. Phone(415) 544-2184.
@FedFRASER