Cato Policy Analysis No. 45 December 19, 1984

Policy Analysis

Deficits and Taxes:
Federal Budget and Fiscal Policy in the 1980's

by Thomas S. McCaleb

Thomas S. McCaleb is associate professor of economics at Florida State University and a former senior staff economist with the Council of Economic Advisers.

Executive Summary

The economic program proposed by the Reagan administration at its inception in 1981 was designed to reduce government spending and taxes relative to the economy's total output or gross national product (GNP). As a first step in implementing the program, the administration proposed a phased reduction in tax rates over the three years from 1981 to 1983. This was to be accompanied by reductions in the level of spending proposed by the Carter administration for fiscal year (FY) 1982. The tax reductions were in fact adopted by Congress, and a set of spending reductions was incorporated into the First Congressional Budget Resolution. The budget process for 1982 was never completed, however, and the 1981-82 recession intervened.

The net result of these efforts has been that tax rates are lower now than in 1980, but not lower than rates in 1979. The reductions in aggregate federal expenditures relative to GNP, however, have not materialized. Indeed, during the first three years of the Reagan administration, federal spending as a percentage of GNP increased to historically high peacetime levels. Because the decline in the rate of growth of tax revenues has not been matched by a decline in the growth of expenditures, the government's budget deficit in real terms has also reached unprecedented peacetime levels. The 1983 deficit was almost 6 percent of GNP. Projected deficits for 1985 and 1986 exceed 4 percent of GNP. These levels are of the same order of magnitude as those reached during the Great Depression of the 1930s. Without a reversal of the tax reductions or significant real spending cuts, the projected deficits will not fall below 3 percent of GNP until 1989.

The projected federal budget deficits have now become the central focus of economic policy. Furthermore, deficit mania has swept Wall Street, the national media, and even the groves of academe, where the National Bureau of Economic Research has launched a major project to investigate the impact of government budget deficits. The deficit issue poses three major questions: (1) Where did the deficit projections come from? (2) Where did the projected deficits come from? (3) Where will the deficit reductions come from?[1]

Where Did the Deficit Projections Come From?

The deficit projections are a residual item derived from separate projections of government outlays and government receipts. The budget document submitted each January includes outlay requests and forecasts of receipts for the next fiscal year, which begins the following October. In addition, since the inauguration in 1976 of the current budget process, the budget document contains outlay and receipts projections for four years beyond the next fiscal year, the so-called out-years. Thus the FY 1985 budget contains the administration's budget requests for FY 1985 and projections for FY 1986 through FY 1989.

These projections of expenditures and revenues depend on two major forces. First, of course, they depend on the program initiatives that the administration proposes. Second, they depend on the expected state of the economy over the five-year period. The expenditure and revenue estimates, and therefore the deficit projections, are determined by the economic assumptions that the administration adopts. [2] To determine where the deficit projections come from, then, one must know where the economic assumptions come from.

Formulation of these economic assumptions is a joint responsibility of the Council of Economic Advisers, the Treasury, and the Office of Management and Budget (OMB). Based on their own best judgment, the principals of these agencies--the chairman of the Council, the secretary of the Treasury, and the director of OMB--agree on a set of assumed values for the rate of growth of real GNP, the rate of inflation, the unemployment rate, and interest rates. Using these assumed values as inputs into a set of economic relationships, their staffs then gene-rate detailed projections of other relevant economic variables, such as personal income, corporate profits, and wages and salaries. The economic relationships are designed to ensure internal consistency, but the assumed values of the major variables are imposed on this set of relationships and are not derived from them.

The economic assumptions that underlie the budget projections are not true economic forecasts although they are widely proclaimed as the government's official forecast. They are not derived from any econometric model, unlike the forecasts produced in the private sector. The typical private-sector forecast makes certain assumptions about monetary and fiscal policy, variables that are subject to choice by government policymakers. Then, based on these assumptions about the government's policy actions, the private-sector forecast produces estimates of the major economic and financial variables from a model that attempts to capture the historical relationship between policy actions and the state of the economy. The government's budgetary process, by contrast, begins with an assumed set of values for the major variables, imposes them on a set of relationships, and then generates from that exercise a set of expenditure and revenue projections.

The government's strategy in making assumptions about such major economic variables as real GNP is to focus on the overall trend, not to attempt predictions of year-to-year fluctuations. In some years, for example, actual growth will exceed the assumed value, and in other years actual growth will fall short of it. Forecasting the year-to-year fluctuations for the purpose of shaping the budget and guiding the evolution of government programs is considered unnecessary. Small or even moderate variations of actual outcomes from assumed values have little significance for budgetary planning.

The value of the deficit projections as indicators of forthcoming events depends on the closeness of the economic assumptions to actual outcomes. The record since 1975 provides a test of the predictive accuracy of the government's annual economic assumptions (see Table 1).[3] Consider first the assumptions about real GNP growth one year ahead, an estimate made in December or early January for the calendar year that begins that same January. On average this one-year-ahead assumption has deviated from actual growth by 2.8 percentage points. The assumption for real GNP growth published in January 1983 for calendar year 1983, for example, was 1.4 percent, whereas the actual real GNP growth in 1983 was 3.3 percent, an error of 2.9 percentage points. For 1984 the budget process assumes real GNP growth of 5.3 percent. Even if this assumption were equal to or better than the average of one-year-ahead assumptions from 1976 through 1983, it would still mean that real growth for the year could fall anywhere between 2.5 percent and 8.1 percent.

The average error of assumptions made two years ahead has been 2.5 percentage points. Applying this to the 9.6 percent real growth assumed between 1983 and 1985 suggests that the economy's actual growth could be anywhere between 7.1 percent and 12.1 percent over this period. For five years ahead, the

Table 1
Accuracy of Economic Assumptions
Total Percentage Increase of Real GNP over Period
  1 Year Ahead(a) 2 Years Ahead(b) 5 Years Ahead(c)
Average error of assumption 2.8 2.5 9.1
Administration's January 1984 assumption 5.3 9.6 23.3
Range with error 2.5-8.1 7.1-12.1 14.2-32.4

Source: Calculated by the author from data in Budget of the United States Government, various years.

(a)Assumptions for calendar years 1975 through 1983 published at beginning of each year from 1975 through 1983. (b)Assumptions for calendar years 1976 through 1983 published at beginning of each year from 1975 through 1982. (c)Assumptions for calendar years 1979 through 1983 published at beginning of each year from 1975 through 1979.

average error has been 9.1 percentage points. Therefore, based on the government's assumption of 23.3 percent real growth from 1983 through 1988, actual growth could fall anywhere between 14.2 percent and 32.4 percent. And these ranges presume that the assumptions published in early 1984 will be at least as close to actual outcomes as the average of the assumptions made between 1976 and 1983.

The economic assumptions have also exhibited substantial volatility from year to year. In other words the assumed value for some economic variable in a particular future year may be quite different in one year's budget as compared with the previous year's budget. This volatility is illustrated in Table 2, which shows the values of real GNP growth, inflation, and the unemployment rate for 1983 and 1984 as assumed in the FY 1984 budget process and in each of the three prior years. In January 1980, for example, the administration assumed that

Table 2
Volatility of Assumptions
Assumptions for 1983 and 1984
Yeats in Which Assumptions Adopted(a)s Real Growth (b) Inflation(b) Unemployment Rate(c)
  1983 1984 1983 1984 1983 1984
1980 5.0 4.9 7.4 6.8 5.9 5.1
1981 3.5 3.7 8.7 7.7 7.1 6.7
1982 5.2 4.9 5.1 4.7 7.9 7.1
1983 1.4 3.9 4.9 4.6 10.7 9.9
(1983 actual) (3.3) --- (3.2) --- (9.6) ---
1984 --- 5.3 --- 4.4 --- 7.8

(a) Published in January of each year.
(b) Percentage change from previous year.
(c) Average for entire year.

1983 would be a year of relatively high growth and virtually full employment but with high inflation. One year later, the administration's expected rate of growth for 1983 was reduced by more than 1 percentage point, and both the unemployment rate and the inflation rate were at least 1 percentage point higher. The assumptions changed in the opposite direction again in January 1982, with yet another reversal in January 1983. Furthermore, actual data have revealed that all of the values assumed early in 1983 were too pessimistic.

The reliability of expenditure, revenue, and deficit projections also depends on policy actions taken during the intervening years. Each year, the administration prepares two sets of projections, one based on continuation of current program levels and the other based on the assumption that all of the administration's policy initiatives will be enacted into law. The result is a pair of hypothetical deficit projections. Of course, expenditure and tax policies do not remain unchanged during the intervening years; nor are the administration's proposals all adopted exactly as offered. Therefore the reliability of deficit projections three, four, or five years ahead is further reduced.

The average error of one-year-ahead deficit projections over the period 1975 through 1983 was $11.3 billion, or 20.2 percent of the actual deficit (see Table 3). For two-years- ahead projections, the average error was $38.6 billion, or 49.9 percent of the actual value; and for five-years-ahead projections, the average error was $98.2 billion, or 123.7 percent. In January 1983 the Reagan administration projected a budget deficit for FY 1983 of 207.7 billion. The actual deficit turned out to be $195.4 billion, so the administration had overestimated the deficit by more than 6 percent. And this one-year-ahead projection was published in January 1983, with one-quarter of the fiscal year already elapsed and with preliminary information on that quarter already in hand.

Table 3
Accuracy of Deficit Forecasts
  1 Year Ahead(a) 2 Years Ahead(b) 5 Years Ahead(c)
  $ Billion Percent of Actual $ Billion Percent of Actual $ Billion Percent of Actual
Average error of projection 11.3 20.2 38.6 49.9 98.2 123.7
Administration's January 1984
183.7 --- 180.4 ---- 152.0 ---
Range with error adjustment 172.4-195.0 --- 141.8-219.0 --- 53.8-250.2 ---

Source: Calculated by the author from data in Budget of the United States Government, various years.

(a)Assumptions for calendar years 1975 through 1983 published at beginning of each year from 1975 through 1983.
(b)Assumptions for calendar years 1976 through 1983 published at beginning of each year from 1975 through 1982.
(c)Assumptions for calendar years 1979 through 1983 published at beginning of each year from 1975 through 1979.

For 1984 the Reagan administration projected a deficit of $183.7 billion. Based on the average error rate from the 1975- 83 period, an acceptable range of values for 1984 would have been $172.4 billion to $195.0 billion. The actual deficit turned out to be $175.3 billion, at the lower end of the error range, so that the January projection overestimated the actual deficit by almost 5 percent. For 1985 the projected deficit is $180.4 billion, but an acceptable range based on the average error rate of the 1976-83 period would be $141.8 billion to $219.0 billion; the lower figure is well below 3.4 percent of the assumed value of GNP in 1985. For 1988, five years ahead, the range of values with respect to the administration's projection, derived from the average error rate of the 1979-83 projections, is $53.8 billion to $250.2 billion; the lower figure is barely 1 percent of assumed 1988 GNP.

Like the economic assumptions from which they are derived, these deficit projections are highly volatile. The change in the administration's economic assumptions between January 1983 and January 1984 reduced the projected deficit for 1984 by almost 15 percent, the projected 1985 deficit by over 16 percent, and the projected 1986 deficit by almost 15 percent without any major policy changes. One final point clearly demonstrates the political rather than economic nature of these long-term deficit projections: The five-years-ahead projections for FY 1979 through FY 1982 all showed surpluses in the government's budget, with the projected surpluses exceeding $40 billion for both 1980 and 1982.

Where Did the Projected Deficits Come From?

The short answer to any question concerning the causes of the actual deficits over the past 25 years is that government expenditures have grown faster than government revenues. A more revealing answer, however, emerges from a look at the relative size of the budget and at its components. As Table 4 shows, government expenditures since 1960 have risen faster than the economy's total output. Federal outlays in 1960 were 18.5 percent of GNP; by 1980 they had grown to 22.4 percent. In 1988 federal expenditures will still be over 22 percent of GNP even if all of the Reagan administration's proposals are adopted. At the same time, federal receipts as a percentage of GNP rose from 18.6 in 1960 to 20.1 in 1980. In 1988 receipts will be 19.3 percent of GNP, provided that all the tax changes enacted by Congress in 1981 and 1982 are left intact and provided that indexing of the individual income tax is implemented as scheduled in 1985.

Table 4
Federal Expenditures and Receipts
as a Percentage of GNP
  1960 1970 1980 1983 1988(b)
Total expenditures(a) 18.5 20.2 22.4 24.1 22.3
   National defense 9.1 8.1 5.3 6.4 7.5
   Net interest 1.4 1.5 2.0 2.7 2.6
   Non-defense, non-interest 8.1 10.6 15.1 15.0 12.3
      Social security 2.2 3.1 4.6 5.2 4.7
      Health (including Medicare
and Medicaid)
0.2 1.3 2.1 2.5 2.6
      Food and Nutrition 0.0 0.1 0.5 0.5 0.4
      Education and Training 0.2 0.9 1.2 0.8 0.5
      All other 5.5 5.2 6.7(c) 6.0 4.1
Total receipts 18.6 19.9 20.1 18.2 19.3
   Individual income tax 8.2 9.3 9.5 8.7 8.8
   Payroll (social security) taxes 2.9 4.6 6.1 6.3 7.2
   Corporation income tax 4.3 3.4 2.5 1.1 2.1
   All other 3.2 2.6 2.0 2.0 1.3

(a) Percentages may not add exactly owing to rounding error.
(b) Assumes adoption of all proposals in Reagan administration's FY 1985 budget.
(c) Federal employee retirement and unemployment insurance benefits, and other income security programs account for 1 percentage point of the 1.2 percentage point increase since 1960.

The composition of federal expenditures has also changed dramatically. In 1960 defense spending accounted for 9.1 percent of GNP; in 1970, during the Vietnam war, defense spending had fallen to 8.1 percent of GNP; and by 1980 defense spending absorbed only 5.3 percent of GNP. The Reagan administration's proposals for increased defense spending would raise this percentage to 7.5 by 1988, still less than the levels of 1960 and 1970. Thus, if non-defense spending had not increased, the decline in defense spending between 1960 and 1980 would have reduced the share of GNP devoted to federal government programs by more than 3 percentage points. Furthermore, federal spending for transportation, commerce, housing, agriculture, and natural resources combined has been almost constant as a share of GNP since 1960

Net interest outlays rose from 1.4 percent of GNP in 1960 to 2.0 percent in 1980. Because of the high interest rates of the early 1980s, interest expense will continue to increase for several years into the future. The average maturity of the government's debt is just under four years, so the rise in interest expense could end by the late 1980s. If the projected deficits materialize, however, total interest expense will continue to remain at relatively high levels because of the increasing volume of outstanding debt.

The most rapid expenditure growth, both in dollar terms and as a share of GNP, has occurred in the so-called entitlement programs and in similar income-transfer programs. These include social security, Medicare and Medicaid, food stamps and other nutrition programs, and subsidies to individuals for education and other forms of training or "human capital investment." As a percentage of GNP, social security more than doubled between 1960 and 1980. Medicare and Medicaid, which did not exist in 1960, absorbed more than 2 percent of GNP in 1980. Food and nutrition assistance rose from a negligible amount in 1960 to 0.5 percent of the economy's total output in 1980. Federal subsidies for education and training rose from 0.2 percent of GNP to 1.2 percent between 1960 and 1980. Under the administration's current budget proposals, most of these entitlement and income-transfer programs--which may be referred to simply as "social programs"--will continue to grow during the 1980s, and none will return to the spending levels of 1960.

Where do the deficits come from? Between 1980 and 1983, expenditures relative to GNP rose by 1.7 percent and revenues fell by 1.9 percent. Among expenditure categories, defense rose by 1.1 percent while net interest increased by 0.7 percent. Thus, the growth in the deficit between 1980 and 1983 could be accounted for entirely by the increases in defense spending and interest outlays and the reduction in taxes. Comparing the deficit in 1980 with the projection for 1988 shows a deficit growth of 0.7 percent relative to GNP, accounted for entirely by lower taxes.

Comparison of 1983 with 1970, however, tells a different story. In 1983 revenues were down by 1.7 percent and expenditures were up by 3.9 percent. Most of the increase in expenditures was accounted for by growth in the social programs. The increase in interest outlays was more than offset by a 1.7 percent decline in defense spending. Although the magnitudes are smaller, the same story emerges from a comparison of the 1988 projections with 1970. Finally, if 1960 is taken as the benchmark, the importance of declining taxes recedes further, and the contribution to the current and projected deficits made by increased spending on social programs is enhanced.

The growth in social programs since 1960 can be attributed to several factors. First, the number of individuals in the eligible population increased. The basic legislation governing many of these programs does not appropriate a fixed amount of money for a particular purpose. Instead, rules are established that define which people are eligible for benefits and the nature and amount of the benefits that each eligible person will receive. Sufficient funds must then be appropriated to pay for these benefits. Thus, even if the criteria for eligibility remains unchanged, outlays may expand if the number of individuals satisfying the eligibility tests increases. This is precisely what has happened, as society experienced substantial growth in both its young and its old populations.

Furthermore, in the early 1970s, Congress enacted legislation liberalizing the eligibility criteria for several social programs, thereby adding many more individuals to the recipient rolls. These programs are no longer directed, if they ever were, only at the poor and disadvantaged. In 1980, for example, total federal health services outlays were $63.8 billion. Only $19 billion--less than 30 percent of this total--went to individuals officially classified as poor. Over $14 billion was spent on non-poor and non-aged adults and children. Almost half of the total was spent for the care of the non-poor elderly population.[4] The incidence of poverty among the elderly is slightly less than its incidence among the population at large, and yet virtually every individual over 65 is eligible for some benefits from social security, with the largest benefits in dollar terms often going to those with the highest incomes.[5] In addition, some 20 percent of total federal spending on education provides assistance to students enrolled in institutions of higher learning; few of these students come from families that would qualify as poor or disadvantaged.[6] In the last 20 years the federal government has spent more and more in an attempt to secure virtually the entire population against all sorts of risks and adversities.

Finally, not only has the number of individuals receiving government assistance increased, but the real value of benefits per recipient has grown as well.[7] Between 1960 and 1970 the average annual benefit paid to each recipient by social security and Medicare combined rose from $853.44 to $1,277.73 (measured in constant 1967 dollars), an increase of 49.7 percent after allowing for inflation. During the same period, real per capita disposable personal income grew by only 35.3 percent. Between 1970 and 1980, social security benefits per recipient grew (after adjusting for inflation) by almost 40 percent, compared with a growth in real personal income of only 22.4 percent. Over the entire 20-year period from 1960 to 1980, the growth in the real value of the average benefit paid to social security recipients exceeded the growth in real disposable personal income by more than 40 percent.

The average food stamp beneficiary received $109.20 per year in 1970 and $166.94 a year in 1980 (measured in constant 1967 dollars). In other words, the benefit of a typical food stamp recipient rose, after adjusting for increases in the cost of living, by 52.9 percent, while real disposable personal income increased by less than half as much. Thus, not only have the social programs been taking care of increasing numbers of Americans, but they have been taking care of them better and better as well.

Where Will the Deficit Reductions Come From?

There is widespread agreement that deficits in excess of about 2 percent of GNP that persist for four to six years will do serious damage to the economy. Indeed, the adverse consequences are claimed to be so great that deficit reduction at all costs must be a paramount objective of federal budget and fiscal policy. What are these adverse consequences?[8] According to one view, if the Federal Reserve pursues a policy of restricting monetary growth in the face of rising demands for loanable funds, real interest rates will remain at their current high levels or rise even further.

The contention is that high interest rates caused by the deficits can severely damage the economy in three ways. First, high real interest rates cause too little investment, insufficient capital formation, and a continuation of the economy's recent experience of low rates of productivity and real income growth. Second, high real interest rates result in an overvalued dollar. Exports are discouraged and imports are encouraged, thereby reducing the demand for domestic output and increasing unemployment. Furthermore, while the trade balance is in deficit, the capital account is in surplus, meaning that foreigners are investing more in U.S. assets and that the United States is on its way to becoming a net borrower in contrast to its traditional role as a net lender. Third, the high interest rates may abort the current recovery before full employment is attained.

It is useful, though, to consider these arguments more carefully. The evidence to support the proposition that higher real federal deficits cause higher real interest rates is, to say the least, weak. There have been some 23 empirical studies of the deficit-interest rate linkage in the last 15 years.[9] A positive relation was reported in 10 of these, a negative relation in 2, and mixed results or no relation in the remainder. Only 3 of the studies, however, produced results that were unambiguously statistically significant. Furthermore, the argument that investment and capital formation have fallen during the past 10 to 15 years is based on a restrictive view that measures only investment in business fixed plant and equipment net of depreciation, ignoring altogether investment in human capital. In assessing the causes of the low productivity growth of the 1970s, this view also neglects the impact of demographic changes. Finally, the argument that investment and economic growth during the 1970s has been too low is predicated on the implicit assumption that the rate of growth achieved in the 1960s and the division of total output between consumption and investment during that decade were just about optimal. The argument thus reflects a very low implicit social rate of discount on future consumption relative to current consumption, and a relatively high implicit weight on the welfare of future generations relative to the current generation.

The presumption that the dollar is overvalued requires the substitution of a third-party judgment about relative currency values in place of the market's judgment. This argument also neglects factors other than interest rates that are currently affecting currency values. Better real investment opportunities in the United States, higher default risk in many countries that have traditionally been recipients of U.S. investment funds, and greater political instability abroad have attracted both U.S. and foreign capital back to the United States. Being a net borrower is not unambiguously evil, a point that was also lost on the seventeenth-century mercantilists; furthermore, someone has to be a net borrower if there are to be any net lenders. Finally, a trade deficit does not create aggregate unemployment but only shifts employment among industries. Indeed, with strong and rising aggregate demand, the trade deficit represents an outlet for inflationary pressures that may otherwise build up during the current recovery.

As recent U.S. experience shows, high interest rates have also not impeded recovery. Interest rates are the price of current consumption in terms of future consumption. Higher real interest rates signify a relatively stronger preference for current consumption and therefore discourage the provision for future consumption that investment and capital formation represent. Higher real interest rates do not affect the aggregate demand for output, however, but only alter its composition away from future consumption for which current investment is required and in favor of present consumption. Resources move from the production of capital goods to the production of consumer goods, but the high interest rates do not reduce total production or employment.

Of course, the Federal Reserve may respond to high federal budget deficits by increasing the rate of growth in the money supply. While this action would mitigate any pressure on real interest rates that emanates from the deficit, it would be inflationary in the long run. An anticipated inflation reduces the efficiency with which economic resources are utilized. An unanticipated inflation has unfortunate consequences for the distribution of income and wealth. Inflation, however, need not discourage investment or capital formation. Investment, capital formation, and growth depend on real interest rates, while inflation's effect is on the money rate, not the real rate, of interest. The problem here is not so much to restrain the deficit as it is to restrain the inappropriate behavior of the Federal Reserve.

The Impact of Tax Increases

Despite all these objections, however, let us suppose that large federal budget deficits do cause high real interest rates, insufficient investment and capital formation, and low rates of growth. What is to be done? The popular solution in Congress is to increase the rate of growth of federal tax receipts. Legislation intended to increase federal revenues has been proposed for three consecutive years. Even before final action was completed on the Deficit Reduction Act of 1984 (sometimes called "DEFRA," or "Son of TEFRA"), the congressional leadership announced plans for additional and larger tax increases in 1985 and beyond. Indeed, the 1984 revenue increases have been popularly referred to in Congress as only a down payment on the projected deficits.

Increasing taxes, so the argument goes, would increase the proportion of income that is saved, reduce interest rates, and bring about an increased rate of investment and capital formation. Of course, higher taxes distort consumption and production choices and prevent the economy from operating efficiently. Nevertheless, the harmful effects of higher taxes on economic efficiency are allegedly outweighed by the permanent damage that will result from the persistence of high deficits.

The truth, however, is different: just as high real interest rates discourage investment and deter productivity improvements, so do higher taxes. While the substitution of higher taxes for the projected deficits may lower real before-tax interest rates, the higher taxes also bring lower real after-tax returns to saving and investment. The lower real interest rates reduce the cost of capital, but the higher taxes reduce the net-of-tax returns that can be obtained from employing that capital. Furthermore, higher taxes and lower real interest rates reduce the reward for saving so that the supply of loanable funds falls, at least partly offsetting the lower government demand for funds. The final result may well be less saving and investment, with only a modest reduction in real interest rates.

Moreover, higher taxes discourage work effort, which in turn reduces the supply of labor to the market and leads to greater consumption of leisure and increased nonmarket production. While the saving-to-income ratio at each level of income is increased when income is raised, the level of income decreases. As a result, the supply of saving may fall even though the rate of saving rises.

No matter what happens to the level of investment when higher taxes are substituted for the projected deficits, the efficiency of investment declines. Higher taxes create incentives to divert funds away from more-productive investment activities that generate taxable income toward less-productive investment activities that receive favorable tax treatment. Less efficiency in the allocation of investment, even when accompanied by greater aggregate investment, is not conducive to significant gains in productivity and economic growth.[10]

Many advocates of higher taxes recognize these disincentive effects on saving, investment, productivity, and growth. They have proposed, however, that the increase in taxes be coupled with comprehensive tax reform. The strategy is to shift the tax base from income to consumption. They contend that a consumption tax could raise the additional revenues to reduce the deficit while exempting the returns of saving and investment from taxation. Thus investment would be favored while consumption would be discouraged. There is little support for this contention, however. While the rate of tax on saving and investment is lower with a consumption tax, the incentive to invest is blunted because of the reduced level of consumption. The ultimate purpose of investment, after all, is the future production of goods and services for consumption. Even the strongest proponents of consumption taxation emphasize its role in promoting efficiency in the allocation of investment funds among alternative types of capital and its administrative simplicity, not its effects on the overall level of investment and capital formation.[11]

Of course, the income tax could be replaced by a consumption tax without any increase in revenues. If consumption taxation does have a positive effect on saving and investment, the gains from shifting the tax base toward consumption would be even greater if the aggregate level of taxation were not simultaneously increased. A more effective strategy for increasing investment, productivity, and growth through tax reform, therefore, would be to move toward a consumption tax with no increase in the amount of revenue raised. The choice between consumption and income as a tax base is totally unrelated to the deficit issue.

Low productivity growth, insufficient investment and capital formation, and a slower rate of economic growth were not caused by too little taxation. Rising productivity, greater investment and capital formation, and a higher rate of economic growth cannot be achieved by increasing the level of taxation. Higher taxes may reduce the projected deficits; they may or may not lower interest rates; but they will not provide the stimulus to investment, productivity, and growth that is allegedly stifled by the current and impending deficits. Whether or not deficits of the projected magnitude actually have the detrimental economic effects that have been widely attributed to them is uncertain. There is no evidence, however, to support the contention that higher taxes will resolve the economic problems that are now being blamed on the deficits.

The Politics of Deficits

The real danger of large federal deficits is political, not economic. Deficit financing hides the true costs of federal expenditure programs by shifting those costs into the future. Political entrepreneurs and the special interests they serve are thus able to obtain current benefits at the expense of future taxpayers. Because those who benefit do not pay the full costs, programs are adopted that are privately profitable but socially inefficient. The availability of deficit financing encourages more and more government activity as government becomes ever larger and ever more intrusive. Every perceived problem calls for a governmental solution because the costs of the solution will be paid only by future generations.

Fundamentally, the problem is the size of government, not the method of financing government activity. The deficit is important only because it encourages more government activity. Raising taxes may reduce the deficit, but higher taxes will not reduce the size or the pervasiveness of government. Indeed, higher taxes serve only to confirm previous expansions of government activity. When seen in this light, the only deficit reduction strategy that makes sense is expenditure reduction.[12]

1984 The Cato Institute
Please send comments to webmaster