FULL TEXT: CRS'S GRAVELLE TESTIFIES AT JEC COLLOQUIUM ON ECONOMY AND TAXES.

When considering the potential economic effects of tax policy and deficit spending, it is helpful to organize the discussion into three separate issues: aggregate short run effects on output, aggregate long-run effects on output, and criteria that might be used to evaluate individual policy changes aside from these aggregate effects. These criteria would include efficient allocation of a fixed amount of resources, distributional consequences, and administrative feasibility.

One of the complexities of discussing the first two issues is that there may be opposite effects in the short run than in the long run of various tax policies and deficit reduction policies. Other things equal, a typical deficit reduction will cause a contraction in the short run when assessed within standard aggregate demand models. That same deficit reduction will cause an expansion in output in the long run. One effect eventually shades into the other, and the short- run effect can also be altered by monetary policies.

The next section discusses the short-run effects of deficit reduction. It also discusses the effects of a deficit-neutral policy that finances tax incentives for savings or investment with spending cuts. The following section discusses long-run effects for the deficit reduction, tax changes that might potentially influence the private supply of labor and capital, and the issue of public investment. The final section briefly touches on the other criteria for evaluating tax and spending policies.

SHORT-RUN EFFECTS

Deficit Reduction

Mainstream economic analysis of the short-run effects of deficit reduction would predict that a decrease in the deficit reduces output and employment, because it reduces aggregate demand. This is the type of analysis that is embodied in the large-scale macroeconomic models, such as DRI/McGraw-Hill model, although it is difficult to determine how accurate the predictions of these models are.

A cut in government purchases of goods and services will directly reduce the government component of aggregate demand. An increase in taxes, or a reduction in transfers, will lower personal incomes and reduce the consumption component of aggregate demand. The contractions in demand will be somewhat magnified by a multiplier effect, since the reduction in demand will cause further reductions in income, further reductions in consumption, and so forth. ¹ A change in government spending or taxes of the same general magnitude would typically be predicted to cause roughly the same effects.

There are several cautionary observations that might be made about these effects.

First, if a dollar of tax increase partly reduces savings rather than consumption, a tax increase might have a smaller effect on aggregate demand than a spending cut. This effect is normally not very large, because only a small fraction of income is saved.

Secondly, short run effects arising from contractionary fiscal policy can, in theory, be offset by a relaxation of monetary policy. In practice, however, it may be difficult to determine how large this offset should be and the appropriate time pattern.

Thirdly, there is not a lot of consensus about the magnitude and timing of short run effects from contractionary fiscal policy. ² There is dispute in the economics profession about the relative power of fiscal and monetary policy which depends on a variety of behavioral responses. Also, in an open economy with flexible exchange rates and mobile international capital, fiscal contraction tends to have a smaller effect than it would in a closed economy. This occurs because the lower interest rates that occur as the government reduces its borrowing cause smaller net inflows of capital. Since foreign investors demand fewer dollars with which to make investments, and U.S. investors supply more dollars to purchase foreign currency to make investments abroad, the price of the dollar falls and U.S. goods become cheaper. This effect causes net exports, also a component of aggregate demand, to rise. These effects take some time, of course, and they depend on how responsive international capital flows are to interest rates.

Over time, any contractionary effect of a smaller deficit should fade. As it does, the additional capital freed up by the reduced deficit will add to the capital stock.

Tax Incentives for Savings/Investment Financed by Spending Cuts

In general, a policy that has no effect on the deficit would not be likely to have an aggregate economic effect. If, however, the tax incentives actually increased savings, the effects would be contractionary in the short run in this standard analysis. There are reasons to doubt, however, that such policies will have an effect on savings, as discussed further in the section below on long-term effects.

Tax incentives for investment (e.g. investment credits) financed by spending cuts could be neutral, contractionary, or expansionary depending on how powerful the effects are on investment spending. Even the more generous estimates of the effectiveness of investment incentives would not tend to predict investment effects larger than the revenue cost, so it is doubtful that such a policy would be expansionary. More pessimistic views of the efficacy of investment subsidies suggest that such policies could be contractionary if part of the investment subsidy is saved in the short run.

LONG-RUN EFFECTS ON ECONOMIC GROWTH

Deficit Reduction

In the long run, we turn from the short-run aggregate demand model to what is termed a neoclassical model. In such a model, there is no reason to expect any effect on involuntary unemployment since any fiscal contraction that initially occurs is transient. In the long run, deficit reduction should increase overall savings and investment and lead to a higher level of output. One must, of course, keep in mind that the short-run effects of policies on the deficit may differ from the long-run effects if the pattern of revenue effects changes over time. Thus, in assessing a particular policy, the first step is determining whether and to what degree permanent deficit reduction occurs.

Behavioral Effects and Public Capital Investments

Long run effects on output can also be influenced by the nature of the policies enacted, and any behavioral effects that might occur. We consider three of these issues in turn: labor supply, savings response, and public capital expenditures.

We turn first to the issue of labor supply. If the deficit reduction is financed via tax increases, it is possible that such increases will influence labor supply.

Economists have long recognized that the response of work effort to a change in after-tax wage rates can be either positive or negative, reflecting the opposing forces of income and substitution effects. When the wage rate falls, leisure becomes less costly in terms of forgone wages, and the individual might wish to increase leisure by reducing the amount of time worked. This is the substitution effect, and it is actually referring to the substitution of leisure for consumption. At the same time, the lower income makes individuals consume less of everything -- both consumption and leisure. Since the amount of leisure rises through one effect and falls with another, the direction of the effect is uncertain. Or, put another way, an individual with less after-tax income might try to earn more to make up for some of the loss.

Since theory does not provide a clear answer, one must turn to empirical evidence on the response of work effort to changes in taxes. Overall, the statistical evidence suggests that the labor supply response is probably small and perhaps even negative. For males, most studies find a negative labor supply elasticity that is small. That is, increases in the net wage result in smaller amounts of work. Studies on female response are mixed and many studies of married women report positive elasticities (i.e. a rise in net wage rates would increase work). ³

The finding of a small elasticity may, in part, reflect institutional constraints. Many salaried workers are employed for a standard work period or to do a standard job and cannot easily vary the amount of labor supplied.

It is also important to note that we do not have good information on all segments of the labor force, e.g. high-income wage earners. The claim is made that high-income workers may be more likely to hold jobs where they are able to vary work efforts, although this conclusion is not certain. ⁴ High-income wage earners may, however, receive nonmonetary rewards from their work and may be less responsive in general to financial considerations for that reason.

We now turn to the issue of savings. Suppose a deficit reduction is financed by a tax that falls on the return to savings. If private savings contracts as a result, the effect of the deficit reduction on overall capital and output will be dampened.

Although the conclusion may seem surprising, economic theory is ambiguous as to whether raising the rate of return will increase the savings rate. This theoretical result is, however, quite straightforward. Saving itself is not a commodity; rather the commodity is consumption at some future time which is being financed by saving. The rate of return affects the price of that future commodity. If the interest rate is ten percent, one has to forego consumption of slightly over 90 cents this year to consume a dollar's worth of goods and services next year, since that amount will grow to a dollar in a year. The "price" of future consumption is 90 cents. If the interest rate falls to five percent, one has to forgo slightly more current consumption, about 95 cents, to consume a dollar's worth of goods and services next year. Thus, at lower rates of return the price of future consumption (foregone current consumption) is higher, and this higher price will discourage future consumption. This price effect suggests that individuals will consume more in the future when the rate of return is higher, just as individuals tend to consume more of any commodity when the price falls. (This effect is referred to as a substitution effect).

Savings, however, is not consumption, but is expenditure on future consumption (price times quantity) Expenditure on future consumption -- savings -- can either rise or fall when the interest rate rises. If the individual were saving 95 cents at an interest rate of five percent, he can both consume more in the future and more in the present when the rate of return rises to ten percent. This ability to consume more in each period occurs because his income has gone up (due to the reduction in taxes on earnings from capital), and the effect is referred to as an income effect. He might save more; but he can also save less. For example, he could save only 93 cents and consume more in the future AND the present. Since, in any one period the majority of the budget tends to be devoted to current consumption, it is quite feasible for consumption in the future to rise relative to present consumption and nevertheless to have savings fall.

When individuals are saving for some fixed target (such as college expenses), then a higher rate of return will result in less savings, since a smaller amount of savings growing at a higher rate of return will achieve the savings target.

In sum, economic theory cannot establish the direction of the effect of a tax reduction on savings. Empirical evidence must be used to assess the direction as well as the size of any effect on savings.

Attempts to study the effects of rate of return on savings have examined these variables over time. These studies have largely found no evidence of a strong savings effect or, in most cases, no clear evidence of any savings response. ⁵ Most of these studies did not take into account data from the eighties. The case that higher rates of return induce higher savings would be even weaker when the experience of the 1980s is taken into account. During this era of high interest rates and low effective tax rates (due to tax benefits adopted in 1981), private savings rates actually fell.

There were a number of studies that considered the effects of individual retirement accounts. The economic theory discussed above would suggest that, because of dollar limits on the amount that can be contributed, these provisions would be less likely to increase private savings than a general reduction in tax burden through, for example, rate reductions. ⁶ For individuals who would have saved at or above the dollar limit without IRAs, there is no substitution effect.

The empirical studies of IRAs were different from the standard savings studies because most of them involved cross section data -- looking at the behavior of different individuals within a short time period. While some studies have claimed to find a strong savings response, there are some fundamental problems with the underlying methodology of those studies. Others have found no evidence of such a response. ⁷ Although one must be cautious about interpreting evidence from a single episode in time, one might also note that the introduction of universal IRAs did not coincide with an overall increase in private savings.

Note that in an open economy with mobile capital, changes in savings will not necessarily translate dollar for dollar into changes in domestic investment. Moreover, it is possible to increase domestic investment by providing tax benefits for investment, even if savings are not changed. However, when investment is financed by foreign suppliers of capital, the returns to that investment are not available to domestic individuals. It is a change in national savings that will yield increases in output that are available to increase the domestic standard of living.

While it is difficult to obtain clear evidence on the effects of taxation on labor supply and savings, on the whole the evidence does not appear to suggest a strong response, or even necessarily a negative effect. These observations also suggest that financing tax incentives for savings with spending reductions would be unlikely to increase long-run output very much, and could even decrease it.

A final point to make about government policies and long-run output effects is that shifting from spending on consumption to public investment will increase output in the long run, assuming that such investment is in projects that have a social value. The concept of investment can be extended beyond capital investment (such as roads) to investment in research and development and in human capital (e.g. education). For some of these investments, however, it is difficult to determine the return.

It is also important to note that even if there were a labor supply contraction as a result of higher taxes, economists generally would not count the loss of output as a reduction in welfare. Rather, the cost is the difference in the valuation of lost consumption relative to the increase in leisure. Similarly, we would not value the increase in output due to increased savings as coming without cost, since consumption must be foregone to obtain such an increase. The value of such increased output is the difference between the value of current and future consumption.

OTHER CRITERIA FOR EVALUATING POLICIES

There are criteria that might be used to evaluate the effects of particular tax and spending policies other than the aggregate effects on the capital and labor supply. These issues will be mentioned briefly.

First, policies that allocate a given amount of capital more efficiently would improve long-run economic welfare. Currently, for example, tax burdens are much higher on some types of capital (e.g. corporate equity) than on others (e.g. corporate debt). There may be efficiency gains due to a reallocation of capital from policies that even up tax rates on different types of investments. Many economists, for example, might argue for integration of corporate and individual income taxes.

Similarly, policies that subsidize activities with positive spillovers or penalize those activities with negative spillovers may improve overall welfare without changing the overall amount of resources used. An example of an activity with a likely positive spillover is investment in research and development, which may yield social benefits that cannot be captured by the firm. For that reason, policies that allow patent protection, joint research ventures, direct spending or subsidies, or tax benefits for research and development may result in a more efficient economy. (There are also problems with some of these policies. For example, patent production and joint ventures may foster monopoly profits, and for direct spending and subsidies it may be difficult to target those investments with the highest social rates of return.)

An example of an activity with a negative spillover is one that produces pollution, so that regulatory restrictions or tax penalties might be appropriate.

It is desirable from an efficiency standpoint to divert resources from private to public uses for goods that are not supplied or are undersupplied in private markets (e.g. defense, highways, human capital investment). The challenge to policy makers is to ascertain the appropriate amount and types of these expenditures.

Tax and spending policies may also be evaluated according to their distributional effects. Economists can try to describe these effects; the desirability of any given degree of income redistribution, however, is not a subject easily addressed by economic analysis.

Finally, tax and spending programs may be evaluated on the grounds of feasibility of administration. Some provisions that would seem desirable on other grounds might be so costly to administer and comply with that they may not be beneficial.

² An economic theory has also been advanced that suggests that deficits do not matter because individuals will recognize the additional liabilities undertaken by the government and offset these effects on themselves and their descendants by altering their personal saving to compensate. This theory is called the "Ricardian equivalence" hypothesis and could apply to both short-run and long- run effects. Many, perhaps most, economists, do not find this theory plausible because they doubt that most individuals are as rational, as far-sighted, and as well informed as the theory suggests. Also individuals may not take into account the effects on all of their descendants (and some individuals do not have descendants). In addition, the evidence of the 1980s seems to contradict this theory, since savings did not rise to offset the increased deficits.

³ See Charles L. Ballard, et al., A General Equilibrium Model for Tax Policy Evaluation. National Bureau of Economic Research. Chicago: University of Chicago Press, 1985, pp. 136-137 for a summary of much of this work. The labor supply response for married women is particularly difficult to estimate because many married women do not participate in the labor force and there is no way to observe their wage directly. Indeed, a recent study suggests that errors in statistical techniques may be responsible for the large responses found in many previous studies, and that labor supply responses for married women may be more in line with male supply responses. See Thomas A. Mroz: The Sensitivity of an Empirical Model of Married Women's Hours of Work to Economic and Statistical Assumptions, Econometrica, Vol. 55, No.,4, June 1987.

⁴ This claim, though frequently made, is not entirely clear. High-income individuals have a higher fraction of income from self- employment and partnership operations. On the other hand, the availability of paid overtime may be more likely in blue-collar jobs.

⁵ See Michael Boskin, Taxation, Savings, and the Rate of Interest, Journal of Political Economy, v. 86, January, 1978, pp. s3- s27; Barry Bosworth, Tax Incentives and Economic Growth, Washington D.C.: Brookings Institution, 1984; A. Lans Bovenberg, Tax Policy and National Savings in the United States: A Survey, National Tax Journal, v. 42, June, 1989, pp. 123-138; Irwin Friend and Joel Hasbrouck, Saving and After Tax Rates of Return, The Review of Economics and Statistics, v. 65, November, 1983, pp. 537-543; E. Philip Howry and Saul H. Hymans, The Measurement and Determination of Loanable Funds Savings, Brookings Papers on Economic Activity, No. 3, 1978, pp. 655-705; John Makin and Kenneth A. Couch, Savings, Pension Contributions, and the Real Interest Rate, The Review of Economics and Statistics, v. 71, August, 1989, pp. 401-407.

⁶ Some economists have argued that individuals will be influenced by the attraction of the up-front deduction of the front- loaded IRA, or that IRAs will increase savings because of a "mental accounts" notion. This latter notion speculates that individuals do not have adequate self-discipline to save and that the setting aside of savings in a separate account with a penalty for withdrawal will cause them to save more. This theory seems less persuasive for the high-income individuals who tended to use IRAs. Note that conventional theory does suggest that a front-loaded (or deductible IRA) should produce initial savings relative to a back-loaded, non- deductible IRA in the amount of the up-front tax savings, assuming that individuals recognize the larger tax liabilities in the former when funds are withdrawn. There should be no difference in overall national savings rate for a deficit financed plan, however, since the increased private savings offset the higher budgetary costs.

⁷ See Jane G. Gravelle, Do Individual Retirement Accounts Increase Savings? Journal of Economic Perspectives, v. 5, Spring, 1991, pp. 13-148, for a review.