The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
JB)@iffiJk\ II
CCV
§@iffi IP1fSiiffiCC
II
CD')
June 12, 1981
High-Employment Budget
A government working group has recently
revised and updated a major tool for fiscal
analysis-the high-employment budget
concept. This important fiscal tool may be
due for a revival in popularity. In 1962,the
President's Economic Report used the
full-employment budget concept in support
of the Kennedy tax-cut proposal as a means of
stimulating the economy. In 1973, again, the
President's Economic Report noted that the
balance in the Federal budget at full employment "was the best single guide to budget
policy," since it represented the desired rate
of growth of the national economy.
The high-employment budget provides a
benchmark for inter-year comparisons of the
impact of the Federal budget upon the
economy-specifically,
the impact when the
economy is operating at a level of high (or
full) employment. Casual observers might
bel ieve that fiscal pol icy becomes more
sti mu lative when one year has a larger budget
deficit than another year. This is not
necessari Iy true, however, because the
economy cou Id be operati ng at different
levels of capacity utilization in the two years.
More often than not, the economy operates
below estimated capacity-except in such
years as 1955, 1966-68,and early 1973.
Valid comparisons can be made, however,
when two years are compared with the use of
a single yardstick-i.e., the high-employment budget.
Economists accomplish this by estimating the
level of Gross National Product at which the
economy would be operating with
reasonably full utilization of resources. This
measure, potential GNP, is the maximum
level of output which can be sustained with
presently available resources. With potential
GN P estimates in hand, analysts can then
complete the exercise by estimating the
levels of Federal budget receipts and expenditures at high employment.
Estimatingpotential GNP
Potential GN P, in constant-dollar terms, may
be calculated as the product of four
quantities: (1) working-age population; (2)
ratio of labor force to population; (3) ratio of
employment to labor force; and (4) ratio of
constant-dollar GN P to employment.
While the working-age population is more or
less fixed in the short run, each of the three
ratios varies over the business cycle. Analysts
thus can obtain high-employment values of
each by estimating the gap between the
actual unemployment rate and the rate
consistent with high employment. So, in
effect, there are two gaps involved in the
estimation of the high-employment budget:
the GN P gap (the shortfall of actual output
from potential output) and the unemployment gap (the difference between actual and
"high-employment" unemployment).
Economists calculate the G N P gap from the
unemployment gap via the three ratios.
Budget receipts and expenditures move with
these gaps; receipts tend to increase, and
expenditures tend to decline, as the
unemployment gap closes. But there is an
added complication -the unemployment
rate corresponding to full employment has
not been constant over time. For example,
this rate was estimated at 4.0 percent of the
labor force in 1955, but at 5.1 percent in
1979. This increase over time reflects the
sharp increase in the number of younger
persons in the labor force - persons who
exh i bit higher -than-average u nem ployment
rates.
Estimatinghigh-employmentreceipts
To obtain high-employment budget receipts,
analysts apply tax elasticities for different
types of income to the tax base (that is,
income), thus calculatingthetax receipts that
would be generated at high employment.
(Elasticity represents the ratio of changes in
tax receipts to changes in the tax base.)
Opinions expressed in this newsletter do not
necf".;jsarily reflect the
of the management
of lhe Federal Reserve Bank of San Francisco.
{Jr 01 the Bc)<:ud of (; OVPfTlors
of the Federal
Reserve Svstem.
one-fourth of total Federal spending. Most
other expenditure categories are insensitive
to cyclical fluctuations, which means that
theiractual expenditures are equivalent to
their high-employment expenditures.
Elasticities differ by different types of income,
and also differ over time with changes in the
business cycle.
The individual income tax shows the highest
and most stable elasticity, with a value
ranging from 1 .30 to 1 .47 during the 1 955-79
period as a whole. For the 1 975-79 segment
of that period, the elasticity figure averaged
1.42. Actually, this represents a weighted
elasticity, since the calculation involves
personal income from a variety of sources.
Unemployment benefits represent the most
important of the seven cyclically-responsive
categories. The other categories include a
variety of transfer payments, such as
social-security benefits and food stamps. On
the basis of 1 979 data, analysts estimate that a
decrease of one percentage point in the
unemployment rate would result in a
decrease of $2.4 billion in regular
unemployment benefits, and a decline of
about $0.9 billion for the other six
expenditure categories. The sensitivity of
response varies substantially among this
group; a decline of one-percentage point in
the unemployment rate induces a
7.7 -percent decline in food-stamp spending,
but a 25.8-percent decline in regular
unemployment benefits.
The tax elasticity is smaller and less stable for
corporate income than for individual
income, because profits are more sensitive
than personal income to the business cycle,
and because the law allows for offset of
corporate losses. For the entire 1 955-79
period, the elasticity of corporate income
taxes averaged 0.80. The elasticity of
social-security taxes has increased over time
to a figure of 0.90 in 1 979, reflecting a rise in
both the tax rate and the covered wage base.
Meanwhile, the elasticity of unemploymentinsurance taxes has remained almost
constant since 1 972, at a figure of 0.68,
reflecting an earnings base much lower than
the social-security wage base.
Use of high-employment budget
Economists use the high-employment budget
to measure shifts in fiscal policy from one
year to another, thus removing the effect of
changes in the level of economic activity on
the budget. For example, in 1 978 the budget
showed an actual deficit of $27.7 billion buta
high-employment deficit of $1 3.4 billion; in
1 979 the budget showed an actual deficit of
$1 1 .4 billion but a high-employment surplus
of $5.4 billion. The high-employment budget
thus moved from -0.6 percent of potential
GN P in 1 978 to +0.2 percent of potential
GN Pin 1 979 -from expansionary to slightly
restrictive. This shift took place because of the
automatic response of receipts-which
would have risen-and expenditureswhich would have fallen-if the economy
had been operating at high employment. It
took place because of automatic changes
responding to changes in growth in output
and employment, as the utilization rate in
the economy slipped from 98.4 percent in
1 978 to 98.0 percent of potential GN P in
1 979, rather than as a result of discretionary
policy changes.
Various sources of Treasury receipts thus
show wide variations in response to changes
in the level of economic activity. The average
response in 1 979 varied from 1 .46 for the
personal-income tax to 0.68 for the unemployment-insurance tax, with most other
taxes showing elasticities of about 0.80. The
greater elasticity of the personal-income tax
reflects, of course, the progressive nature of
the rate structure compounded by inflation.
Estimating high-employment expenditures
To obtain high-employment expenditures,
analysts make adjustments to show what
Federal spending would be at a high level of
employment. The procedure is essentially the
same as that used in obtaining highemployment receipts, but the estimated
adjustments are limited to seven expenditure
categories which account for about
2
Budget Position
High
Employmenl
Budget
{
r-. - - - - - - - - - - - - -
Surplus
Oehcll
:-..
ci'
;;-,:§"
,§'.
High Employment
Rate of Resource
Utilization
• Actual Rate 01 Resource Use
- --- --- --- --- --- --- --- --- -- --- --- --- --- --- --- --- -- --- --- --- --- --- --tically as income rises with prices - the so
ca lied "bracket drift" - whereas budget
expenditures tend to lag with respectto rising
prices.
Consider the case of a tax cut unaccompanied by other changes in the Federal
budget - i.e., a discretionary policy change
rather than an automatic change. At the
existing rate of GN P utilization, this would
result in a (stimulative) increase in the highemployment budget deficit, shifting the
budget line as noted in the chart. The actual
budget deficit would also increase, implying
greater Treasury demands upon credit
markets. Or, consider a decrease in Federal
spending concentrated in highly cyclical
expenditure categories. Besides being
restrictive in their impact on the economy,
these cuts may also force changes in the
estimation of high-employment expenditure
ad justments. As they become less sensitive,
such adjustments will probably become
smaller. As a result, the high-employment
budget might show a smaller deficit as
automatic adjustments become less
responsive to cyclical changes.
By itself, this would tend to suggest that the
high-employment budget has an inherent
bias towards restrictiveness. However, that
conclusion ignores inflation indexation,
which affects certain cyclically-sensitive
expenditure categories in a major way. Since
these categories are indexed on the basis of
the consumer price index, which tends to
overstate the actual rate of inflation,
inflation's impact on receipts can be offset
somewhat. Moreover, receipts are more
responsive than expenditures in periods of
real growth, so that the high-employment
budget may be pushed towards a surplus
without an actual discretionary shift towards
restrictiveness.
Even with these limitations, however, the
high-employment budget is a useful measure
for indicating the basic stance offiscal policy.
It gives analysts the ability to separate
changes in budget conditions that are due to
shifts in economic activity from those
changes that are due to policy shifts. And in
addition, it gives them the ability to measure
the effects' of automatic-stabi Iization
measures on the economy.
A simultaneous cut in taxes and expenditures
wou Id tend to reduce the automaticity of the
Federal budget in responding to cyclical
changes. Lower tax rates wou Id tend to
reduce the decline in revenues as general
business activity falls off. At the same time,
cuts in cyclically-sensitive expenditures
would tend to limit their further increase.
Thus, incomewouldtendtofall
morequickly
in a recession, necessitating greater fiscal
discretionary actions.
Herbert Runyon
Limitationsof estimates
In recent years, economists have shown less
interest in the high-employment budget
approach because of its failure to take
adequate account of inflation. High-employment budget estimates assume that there
is no "price gap" corresponding to the GN P
gap and the unerilploymentgap -an unrealistic assumption in the 1 970's and 1 980's.
The expression of high-employment budget
levels as a percentage of potential GN P
improves the usefulness of the highemployment surplus as a tool, but does not
completely solve the problem introduced by
inflation. With a progressive income tax,
inflation increases budget receipts automa-
Elasticitiesof VariousTaxes
With Respectto Income
1979
Individual income tax
Corporate profits tax
Indirect business taxes
Social security tax
Unemployment insurance tax
3
1 .46
0.78
0.80
0.90
0.6'8
UOlSU!4SI?M.4I?ln uoSaJO• I?pl?AaN o41?PI
•
•
!!I?MI?H• I?!
UJOJ!lI?J I?UOZ!JV• l?>fsl?lV
•
(DJ
BANKIN G DATA-TWELFTH FEDERAL
RESERVE
DISTRICT
(Dollaramounts millions)
in
SelectedAssetsand Liabilities
Large Commercial Banks
Loans
(gross,
adjusted) investments*
and
Loans
(gross,
adjusted) total#
Commercial industrial
and
Realestate
Loans individuals
to
Securities
loans
U.s. Treasury
securities*
Othersecurities*
DemanddeJX)sits total#
Demanddeposits adjusted
Savings
deJX)sits total
Time deJX)sits total#
Individuals,
part.& corp.
(Large
negotiable
CD's)
Weekly Averages
of Daily Figures
Amount
Outstanding
5/27/81
149,245
127,211
37,532
52,271
22,913
1,617
6,448
15,586
40,693
27,437
29,981
80,418
70,870
31,727
Weekended
5/27/81
Change
from
Change
yearago
from
Dollar
Percent
5/20/81
8.6
11,838
838
9.8
780
11,354
11.0
3,734
279
11.8
5,510
40
- 4.6
- 1,113
31
104.4
826
104
3.5
217
54
1.8
4
271
3.7
882
1,558
- 7.2
- 2,129
41
12.0
3,219
152
23.9
389
15,514
27.9
15,445
364
8,897
39.0
144
Comparable
Weekended
year-ago
period
5/20/81
Member Bank ReservePosition
Excess
Reserves )/Deficiency - )
(+
(
Borrowings
Net freereserves )/Netborrowed( )
(+
-
n.a.
148
n.a.
n.a.
132
n.a.
51
44
7
* Excludes
tradingaccountsecurities.
# Includesitemsnot shownseparately.
Editorial comments may be addressed the editor (William Burke) or to the author . ... Freecopiesof this
to
and other FederalReserve
publications can be obtained by calling or writing the Public Information Section,
Federal ReserveBankof SanFrancisco,P.O. Box 7702, SanFrancisco94120. Phone(415) 544-2184.
@FedFRASER