Sources of change in the federal government deficit, 1970-86.

Wakefield, Joseph C. ; Holloway, Thomas M.

Sources of Change in the Federal Government Deficit, 1970-86

AT the beginning of 1970, the Federal Government sector in the national income and product accounts (NIPA's) was in approximate fiscal balance. In 1986, based on the fiscal year 1986 budget, the Federal deficit is expected to be about $164 billion.1 The deficit's growth over this period is not uniform. Moreover, changes in receipts and in expenditures do not contribute uniformly, nor do the changes in the various categories of receipts within total receipts or the various categories of expenditures within total expenditures.

1. All measures of the budget are on a NIPA basis unless otherwise noted and are from NIPA table 3.2. All annual estimates are for calendar years. Estimates for 1985:1-1986:3 are based on the translation of the fiscal year 1986 unified budget into the Federal sector of the NIPA's; see Joseph C. Wakefield and Richard C. Ziemer, "Federal Fiscal Programs,' SURVEY OF CURRENT BUSINESS 65 (February 1985): 10-15. Estimates for 1986 are averages of the first three quarters of calendar 1986.

This article examines the sources of change in the Federal deficit. Because the deficit is simply receipts minus expenditures, the article discusses the sources of change by examining trends in receipts and expenditures, their composition, and their automatic responsiveness to the business cycle and to inflation. The first section of the article provides an overview of the sources of change in total receipts and expenditures, and the resulting changes in the deficit. The automatic response of the budget to the business cycle (hereafter referred to as automatic cyclical effects) and to inflation (hereafter referred to as automatic inflation effects) are discussed. The second section features changes in the composition of receipts and expenditures, by category, for three subperiods that roughly correspond to the Nixon-Ford administrations, the Carter administration, and the Reagan administration.

Overview

Table 1 shows Federal receipts, expenditures, and the deficit for 1970-86. The table shows that, on an annual basis, the budget is in deficit the entire period. The deficit declines in as many years as it increases, although, cumulatively, the increases are much larger than the declines. As a percentage of GNP, the deficit increases sharply in the mid-1970's, declines in the late 1970's, then increases sharply in the early 1980's before declining somewhat at the end of the period. In chart 4, the budget measures are plotted as percentages of GNP. Over the period, there is no clear upward or downward trend in the measures of receipts. There is, however, a very strong upward trend in expenditures. The lack of a clear trend in receipts combined with a sharp increase in expenditures leads to a sharp increase in the deficit as a percentage of GNP.

Table 2 and chart 5 show total changes in receipts, expenditures, and the deficit as well as changes due to automatic cyclical effects, automatic inflation effects, and legislation and other factors. The second panel of chart 5 shows the automatic cyclical effects. These effects reflect the automatic responsiveness of receipts and expenditures to the business cycle and are estimated by calculating what budget levels would be if the economy were operating on a hypothetical trend of GNP--middle-expansion trend GNP--rather than its actual path.2 These automatic effects on receipts are much larger than those on expenditures. As a result, the effects on the deficit of the business cycle mainly follow the pattern of the effects on receipts.

2. See Frank de Leeuw and Thomas M. Holloway, "The High-Employment Budget: Revised Estimates and Automatic Inflation Effects,' SURVEY 62 (April 1982): 21-33; Frank de Leeuw and Thomas M. Holloway, "Cyclical Adjustment of the Federal Budget and Federal Debt,' SURVEY 63 (December 1983): 25-40; and Thomas M. Holloway, "Cyclical Adjustment of the Federal Budget and Federal Debt: Detailed Methodology and Estimates,' Staff Paper No. 40, BEA, June 1984.

The estimates of the automatic cyclical effects reflect cyclical developments as shown in chart 6. The cyclical position of the economy is defined by the GNP gap--the gap between actual and trend GNP, relative to trend GNP. Changes in the cyclical position of the economy (that is, a declining or an expanding economy) are indicated by changes in the GNP gap, which underlie the estimates of changes due to automatic cyclical effects shown in the second panel of chart 5. The sharp declines in 1974-75, 1980, and 1982 shown in chart 6 contributed to automatic cyclical increases in the deficit shown in chart 5. The economy's recovery since 1982, as in 1972-73 and 1976-79, contributed significantly to reducing the deficit. However, a deceleration of the rate of increase, which is typical following the strong growth of the early stages of cyclical recoveries, contributes to smaller deficit reductions in 1985 and 1986 than in 1984.

The third panel of chart 5 shows the automatic inflation effects. These effects reflect the automatic responsiveness of receipts and expenditures to price changes. These automatic effects increase both receipts and expenditures in every year of the period, but the increases in receipts are always larger than the increases in expenditures. Therefore, the automatic inflation effects contribute to reducing the deficit throughout the period.

The size of the automatic inflation effects reflects the course of inflation, shown in chart 6. Two periods of accelerating inflation, 1972-75 and 1976-81, result in sharp increases in both receipts and expenditures. Because receipts increase more rapidly, the automatic inflation effects contribute relatively large amounts to reducing the deficit. Two periods of decelerating inflation, 1975-76 and 1981-83, result in clearly smaller effects on receipts and expenditures. The sharp deceleration in 1981-83 lowers the contribution from automatic inflation effects to reducing the deficit from about $27 billion in 1981 to about $7 billion in 1983.

The bottom panel of chart 5 shows changes due to legislation and other factors. This source of change is the residual after netting changes due to automatic cyclical effects and automatic inflation effects from total changes. In what follows, this source of change will often be referred to as "legislation' for convenience, and mention will be made of legislation that brought about policy changes. However, other changes--such as real growth in the economy and demographic changes--are frequently as important, or more important, quantitatively. This source of change increases expenditures in every year, and receipts in all but 3 years. However, in all but 1 year, the increase in expenditures is larger than the increase in receipts.

Over the entire period, the automatic cyclical effects and automatic inflation effects move the budget toward surplus by about $25 billion and $230 billion, respectively. These two automatic effects, however, are exceeded by legislation, which moves the budget toward deficit by $406 billion.3

3. The estimates of the sources of change are made on a quarterly basis given the initial conditions of the quarter. The values in table 2 are derived by cumulating quarterly estimates. Because the initial conditions of a quarter depend on all earlier sources of change and not just the one of interest, cumulations of the sources must be interpreted cautiously. The cumulations are indicators of the effects over time, but are partly dependent on the history of the other sources.

The top panel of chart 5 shows that the budget moves toward surplus in 8 of the 16 years shown, but the magnitude of movement toward deficit in the other years is clearly overwhelming. Three moves toward deficit --in 1975, 1980, and 1982--were particularly sharp. All 3 years included cyclical troughs that accounted for large movements toward deficit in the automatic cyclical effects. However, the changes due to legislation were also important. In 1975, this source of change included a relatively large decline in receipts and a relatively large increase in expenditures; these changes resulted in a very large movement toward deficit. In that year, the automatic inflation effects resulting from rapid inflation only partly offset the other sources of change. In 1980, changes in expenditures due to legislation increased sharply relative to changes in receipts due to this source. Again the automatic inflation effects resulting from the rapid inflation helped offset the other sources of change, but still left a sharp increase in the deficit. Finally, in 1982, changes in expenditures due to legislation remained large, but changes in receipts due to this source declined. The result was another large movement toward deficit. The deceleration in inflation reduced the size of the offsetting automatic inflation effects and contributed to a very large increase in the deficit.

Changes in Components of Receipts and Expenditures

In this section, following a discussion of trends in the components of receipts and expenditures, the composition and sources of change in the components are examined for three subperiods. These subperiods approximately correspond to presidential administrations: Nixon-Ford, 1970:1-1977:1; Carter, 1977:1-1981:1; and Reagan, 1981:1-1986:3. The use of administrations to demarcate budget developments is somewhat arbitrary because, when taking office, an administration inherits many policy commitments from the previous administration. Nevertheless, they are often of interest, and are used in this presentation to subdivide the entire period.

Trends

Table 3 and chart 7 show components of receipts and expenditures as percentages of GNP. Among receipts, personal tax and nontax receipts as a percentage of GNP are somewhat lower at the end of the period than at the beginning. The early part of the period is characterized by a downward drift in the percentage, followed by a steady increase during the middle part of the period, then a relatively sharp decline toward the end. The estimate of 8.8 percent for 1986 is below the entire-period average of 9.0 percent. Contributions for social insurance increase almost without interruption over the entire period and rapidly approach personal taxes as a percentage of GNP. The estimate of 7.3 percent for 1986 is well above the entire-period average of 6.4 percent. Corporate profits tax accruals decline gradually from the beginning of the period until 1982, before increasing somewhat. The estimate of 2.2 percent for 1986 is below the entire-period average of 2.7 percent. Indirect business tax and nontax accruals decline slightly from the beginning of the period until the end. However, the estimate of 1.4 percent for 1986 is about the same as the entireperiod average of 1.5 percent. Overall, total receipts as a percentage of GNP is slightly higher in 1986 than in 1970. The increased percentage is the result of the sharp increase in the percentage of contributions for social insurance that more than offsets the combined declines in the percentages of personal taxes, corporate profits taxes, and indirect business taxes.

Expenditure components also show both increases and declines as percentages of GNP. Transfer payments increase sharply from the beginning to the end of the period. They reach a peak in the early 1980's, then decline each year in the rest of the period. The estimate of 9.2 percent for 1986 is above the entire-period average of 8.8 percent, but is below every other year in the 1980's. The next largest category of expenditures, defense purchases, declines somewhat from the beginning to the end of the period. The decline is concentrated in the 1970's, reaches a low in 1978-79, and then begins a steady increase. The estimate of 6.5 percent for 1986 is well above the entire-period average of 5.7 percent. Net interest paid increases sharply from the beginning to the end of the period. During the 1970's, this category increases only slightly, but during the 1980's its growth accelerates noticeably. The estimate of 3.4 percent for 1986 is well above the entire-period average of 2.0 percent. Nondefense purchases and other expenditures (which consists of grantsin-aid to State and local governments, subsidies less current surplus of government enterprises, and wage accruals less disbursements) decline somewhat from the beginning to the end of the period. Overall, the sharp increase in expenditures as a percentage of GNP from 1970 to 1986 occurs as a result of sharp increases in the percentages of transfer payments and net interest paid that are partly offset by declines in the percentage of defense purchases--especially in the early 1970's--and in the percentages of nondefense purchases and other expenditures. However, in the late 1970's and in the 1980's, defense purchases contribute to, rather than offset, the increase in expenditures.

Detailed analysis

The discussion of the sources of change in total receipts, expenditures, and the deficit focused on separating them into automatic cyclical effects, automatic inflation effects, and changes due to legislation and other factors. Using receipts as an example:

(1) T = T(c) T(i) T(l)

where:

T = change in total receipts;

T(c) = change in receipts due to automatic cyclical effects;

T(i) = change in receipts due to automatic inflation effects;

T(l) = change in receipts due to legislation and other factors.

Table 4 shows the factoring of the deficit and the components of receipts and expenditures into these sources of change for the Nixon-Ford administrations, the Carter administration, and the Reagan administration. These estimates indicate that of the total rise in the deficit of about $156 billion, about 23 percent occurs during the 28 quarters of the Nixon-Ford administrations, about 6 percent during the 16 quarters of the Carter administration, and about 71 percent during the 23 quarters of the Reagan administration.

Only during the Nixon-Ford administrations did the economy end up at a lower cyclical position than at the start of the period.4 Consequently, changes due to automatic cyclical effects move the budget toward deficit in the Nixon-Ford administrations, but toward surplus in both the Carter and Reagan administrations. Changes due to the automatic inflation effects move the budget away from deficit in all three administrations, but by more in the Carter administration than any of the others. The size of the changes due to automatic inflation effects is noticeably smaller in the Reagan administration even though the economy (measured in current dollars) is much larger. Changes due to legislation move the budget toward deficit in all three administrations, with a marked acceleration during the Reagan administration.

4. The GNP gap, defined as actual minus trend GNP, divided by trend GNP, is 0.2 in 1970:1, -0.9 in 1977:1, -0.5 in 1981:1, and 1.7 in 1986:3.

Changes in the ratio of the budget components to GNP reflect their growth relative to growth in the economy. (Here, as in the rest of the article, the terms of equations that are expressed as ratios are shown as percentages in the tables.) Changes in the ratio, using receipts as an example, can be expressed as:

(2) (T/Y)= T/Y-(T-1/Y-1)g

where:

T =total receipts;

Y =GNP in current dollars;

g =relative growth in current-dollar GNP ((Y - Y-1)/Y).5

5. The change in the ratio of a budget component (total receipts here, for example) to GNP can be factored as follows:

(T/Y) =T/Y-T-1/Y-1

=T-1 T/Y-T-1/Y-1

= T/Y T-1Y-1/YY-1-T-1Y/Y-1Y

= T/Y T-1(Y-1-Y)/YY-1

The result is that shown in equation (2).

The equation shows that the direction of change in the receipts-to-GNP ratio depends on whether the growth in receipts relative to current-period GNP exceeds a measure of relative GNP growth times the lagged receipts-to-GNP ratio. Substituting equation (1) into equation (2) gives:

(3) (T/Y)= T(c)/Y T(i)/Y T(l)/Y-(T-1/Y-1)g

Changes in the budget components (receipts here), which comprise the numerators of three of the terms in equation (3), are from table 4. Table 5 shows all of the terms in equation (3) for the deficit and the components of receipts and expenditures. The budget developments factors ( T(c)/Y, T(i)/Y, and T(l)/Y in equation (3)) roughly measure the effects of a change in a component of the budget on the change in the component-to-GNP ratio. The GNP growth factor--

(T-1/Y-1)g, in equation (3)

--measures the effects of changes in GNP on the change in the budget component-to-GNP ratio.

Over the entire period, the deficit-to-GNP ratio increases 3.5 percentage points, the result of a 0.1-percentage-point decline in the receipts-to-GNP ratio and a 3.4-percentage-point increase in the expenditures-to-GNP ratio. All of the increase in the deficit-to-GNP ratio occurs during the Nixon-Ford and Reagan administrations. In the Nixon-Ford administrations, the deficit-to-GNP ratio increases 1.9 percentage points, all the result of an increase in the expenditures-to-GNP ratio. In the Carter administration, the deficit-to-GNP ratio declines 0.4 percentage points; a 1.5-percentage-point increase in the receipts-to-GNP ratio more than offsets a 1.1-percentage-point increase in the expenditures-to-GNP ratio. Finally, in the Reagan administration the deficit-to-GNP ratio increases 2.0 percentage points, the result of a 1.6-percentage-point decline in the receipts-to-GNP ratio and a 0.4-percentage-point increase in the expenditures-to-GNP ratio.

The Nixon-Ford administrations.-- An increase in the deficit-to-GNP ratio in the Nixon-Ford administrations was the result of an unchanged receipts-to-GNP ratio and an increase in the expenditures-to-GNP ratio. The approximately unchanged receipts-to-GNP ratio occurred because the sum of about equal declines in the ratios of personal taxes and indirect business taxes was almost exactly offset by an increase in the ratio of contributions for social insurance. The ratio of corporate profits taxes was unchanged.

As noted in the discussion of equation (3), the contribution of the three budget developments factors relative to the GNP growth factor accounts for the changes in these ratios. Of the budget developments factors, the one associated with the automatic inflation effects was the largest for most categories.6 For personal taxes, this factor approximately offsets the GNP growth factor (which, from equation (3) carries a negative sign), but the budget developments factors associated with the automatic cyclical effects and with legislation declined and thus contributed to a decline in the ratio. For indirect business taxes, the budget developments factor associated with the automatic cyclical effects declined slightly; the two others increased, but by amounts too small to offset the GNP growth rate factor. The increase in the contributions for social insurance ratio resulted from a relatively large increase in the budget developments factor associated with legislation. Among the most important legislated actions contributing to the increase in the ratio were increases in the maximum taxable wage for Social Security and two increases in the payroll tax rate.

6. The effect of inflation on the ratio of a budget component to GNP depends on the elasticity of the component with respect to inflation-induced changes in GNP. If the elasticity is greater than one, the ratio will increase; if it is less than one, the ratio will decline; if it is equal to one, the ratio will not change. Equation (2) rewritten as:

(T/Y)= T-T-1 Y/Y-1/Y

illustrates this.

The increased expenditures-to-GNP ratio occurred because increases in the transfer payments and other expenditures ratios were considerably larger than declines in the defense purchases ratio. The nondefense purchases and net interest paid ratios increased only slightly.

Of the budget developments factors, those associated with legislation were the largest for most expenditures categories. For transfer payments, the budget developments factors associated with automatic inflation effects and, especially, legislation more than offset the GNP growth factor. The automatic inflation effects were mainly from the indexation of transfer payments programs, including programs indexed prior to 1970 (such as civilian and military retirement) and those first indexed between 1970 and 1976 (such as social security benefits, food stamps, and supplemental security income). 7 Among the most important legislated actions contributing to the increase in the ratio were statutory increases in social security benefits, the initiation of new programs (such as supplemental security income and the earned income credit), and the enactment of special temporary programs (such as the Federal supplemental benefits and supplemental unemployment assistance programs during the 1974-75 recession.)

7. See Thomas M. Holloway, "Sources of Change in Federal Government Transfer Payments to Persons, 1970-81,' SURVEY 62 (October 1982): 25-32. Other categories of transfer payments, such as medicare, are not directly indexed, but are sensitive to the rate of inflation, and their responses are included in the measures of automatic inflation effects.

For the other expenditures category, the budget developments factor associated with legislation was most important. The largest increases came from new or expanded expenditures for a large number of categories of grants-in-aid to State and local governments: Revenue sharing was initiated, and grants for public assistance (such as aid to families with dependent children and medicaid), child nutrition programs, education programs, and Environmental Protection Agency programs all grew during the subperiod.

Partly offsetting the increases in these categories was the decline in the defense purchases ratio. Budget developments factors associated with automatic inflation effects and legislation contributed to relatively small increases in the ratio, but not nearly large enough to offset the GNP growth factor.

The Carter administration.--In the Carter administration, the deficit-to-GNP ratio fell as the result of a larger increase in the receipts-to-GNP ratio than in the expenditures-to-GNP ratio. The increase in the receipts ratio occurred because increases in the ratios of personal taxes, indirect business taxes, and contributions for social insurance--increases of approximately the same size--offset a decline in the corporate profits taxes ratio. For personal taxes, the budget developments factor associated with automatic inflation effects alone more than offset the GNP growth factor and was an important source of increase in the ratio. For indirect business taxes and contributions for social insurance, the budget developments factor associated with legislation was particularly important. The enactment of the windfall profit tax on domestic producers of crude oil accounted for much of the increase in the indirect business taxes ratio. Increases in payroll (and self-employed) tax rates accounted for some of the increase in the contributions for social insurance ratio. The decline in the corporate profits taxes ratio was partly due to a cut in the top corporate profits tax rate and to the decline in the corporate profits share of GNP--after removing the effects of the business cycle--during the Carter administration.

The increased expenditures-to-GNP ratio occurred mainly as a result of the increase in the net interest paid ratio. The ratios of defense purchases, nondefense purchases, and transfer payments increased slightly, but were partly offset by a decline in the ratio of the other expenditures category. Indexing and the acceleration of inflation were reflected by the budget developments factor associated with the automatic inflation effects for transfer payments. This budget developments factor was larger during the Carter administration than during the Nixon-Ford administrations; however, the same accelerating inflation also contributed to a larger offsetting value of the GNP growth factor for transfer payments. The result was a relatively small increase in the transfer payments ratio during the Carter administration. The relatively large increase in the net interest paid ratio was primarily due to the budget developments factor associated with legislation. The increase in this factor reflected the effects of rapidly rising real interest rates, persistent cyclically adjusted deficits, and accumulated cyclically adjusted debt.

The Reagan administration.--The deficit-to-GNP ratio increases sharply in the Reagan administration as a result of a decline in the receipts-to-GNP ratio and an increase in the expenditures-to-GNP ratio. The decline in the receipts ratio occurs because declines in the ratios of personal taxes, corporate profits taxes, and indirect business taxes to GNP offset the relatively small increase in the contributions for social insurance ratio. The increase in contributions is partly the result of several increases in the payroll (and self-employment) tax rates.

The largest declines are in the personal taxes and indirect business taxes ratios. For personal taxes, the budget developments factors associated with the automatic inflation effects and with legislation are particularly important. The deceleration of inflation contributes to a noticeably smaller increase in personal taxes associated with automatic inflation effects than the comparable increase during the Carter administration. The effect is also smaller because indexing of the individual income tax beginning in 1985 eliminates inflation-induced bracket creep for a small part of the period. The decline due to the budget developments factor associated with legislation reflects the large tax rate reductions and other statutory tax code changes made under the Economic Recovery Tax Act of 1981.

For indirect business taxes, declining oil prices cause the windfall profit tax to decline significantly from its peak levels in early 1981. These sharp declines are partly offset by legislated increases in other types of indirect business taxes, such as cigarette and gasoline excise taxes. These influences on indirect business taxes contribute to a small decline in total budget developments factors. The result is a decline in the indirect business taxes ratio.

The increased expenditures-to-GNP ratio is the result of increases in the defense purchases and net interest paid ratios. The ratios of nondefense purchases, transfer payments, and other expenditures all decline. For both defense purchases and net interest paid, it is the budget developments factor associated with legislation that accounts for most of the increase in the ratios. Legislated increases in defense purchases underlie the increase in the defense purchases ratio. The same factors that were present during the Carter administration underlie the increase in the net interest paid ratio. The declines in the ratios for the remaining categories of expenditures can also be traced to the budget developments factor associated with legislation. In all of these categories, the contribution of legislation is much smaller (negative in one case) than during the Nixon-Ford or Carter administrations. Partly as a result, the total of the budget developments factors for these categories is less than the GNP growth factor for each category; consequently, the ratios decline.

Summary.--The increase in the deficit relative to the size of the economy --that is, the deficit-to-GNP ratio--occurs as a result of a variety of factors. Clearly, legislation is a very important source of its increase and of changes in the composition of receipts and expenditures.

In the Nixon-Ford administrations, defense spending declined in importance relative to transfer payments and grants, both of which increased sharply. New and expanded transfer payments and grant programs resulted in a large increase in total expenditures. Taxes were not increased enough to finance the increase, so the deficit-to-GNP ratio was up sharply.

In the Carter administration, both the receipts and expenditures-to-GNP ratios increased, but the receipts ratio increased more and reduced the deficit-to-GNP ratio. High and accelerating inflation during the Carter administration was a very important factor in the larger increase in receipts; legislated increases in indirect business taxes (windfall profit tax) and contributions for social insurance (mainly tax rate increases) also contributed. On the expenditures side, net interest paid began to increase in importance realtive to other categories of expenditures as a result of a growing stock of debt and high interest rates.

In the Reagan administration, the deficit-to-GNP ratio increases sharply as a result of a decline in the receipts-to-GNP ratio and an increase in the expenditures-to-GNP ratio. The decline in the receipts ratio largely reflects legislated tax cuts combined with a smaller boost to receipts than in the earlier administrations from the automatic effects of inflation (due to the deceleration of inflation). On the expenditures side, transfer payments and grants decline in importance relative to defense purchases and net interest paid, both of which increase sharply. Defense purchases increase as a result of legislative commitments. Net interest paid increases as persistent and large deficits enlarge the stock of outstanding Federal debt and interest rates remain high.

Table: 1.--Federal Government Receipts, Expenditures, and Surplus or Deficit(-)

Table: 2.--Sources of Change in Federal Government Receipts, Expenditures, and Surplus or Deficit(-)

Table: 3.--Components of Federal Government Receipts and Expenditures as a Percentage of GNP

Table: 4.--Sources of Change in Components of Federal Government Receipts and Expenditures, and in the Surplus or Deficit(-) for Three Subperiods

Table: 5.--Sources of Change in the Ratios of Components of Federal Receipts and Expenditures, and the Surplus or Deficit (-) to GNP for Three Subperiods

Photo: CHART 4 Federal Government Receipts, Expenditures, and Deficit as a Percentage of GNP

Photo: CHART 5 Sources of Change in Federal Government Receipts, Expenditures, and Surplus or Deficit (-)

Photo: CHART 6 GNP Gap and Inflation Rate

Photo: CHART 7 Receipts and Expenditures as a Percentage of GNP