Free beta access. 50 free AI tokens each week. Optional top-ups available. See what is included
Opening Word document…

Fiscal policies and taxation implemented by the government have always been considered a key part of the neoclassical economic model.

Kotlikoff 84 - Kotlikoff, Laurence J. Taxation and savings-A neoclassical perspective. No. w1302. National Bureau of Economic Research, 1984.

ABSTRACT This paper discusses recent neoclassical analyses of taxation and savings. Contrary to the popular view that fiscal policy has highly ambiguous impacts on savings, neoclassical models admit a host of policies with clear and potentially quite powerful affects on the accumulation of wealth. The paper considers four fundamental types of fiscal policies and compares their quantitative effect on savings. The essential elements of these policies involve inter— and intragenerational redistribution, marginal and intra— marginal taxation, and the level of government consumption. Conventional accounting measures of "taxes", spending, and "deficits" provide, at best, little guide to changes in underlying fiscal instruments and, at worst, precisely opposite indicators of the direction of such changes. Indeed, the continued use of and concern with conventional fiscal measures is symptomatic of widespread fiscal illusion. These points are developed within the context of certainty models. The paper also considers the role of fiscal policy in both mitigating and exacerbating economic risks facing the private sector. Since precaution is a major motivation for saving, governments can greatly influence wealth accumulation either by using fiscal policy to pool private risks or by making fiscal policy itself highly uncertain. Laurence J. Kotlikoff National Bureau of Economic Research 1050 Massachusetts Avenue Cambridge, NA 02138 I. Introduction The ability of governments to alter an econon"s rate of wealth accumulation is a question that has intrigued economists since the inception of the discipline. Recently the issue has received considerable attention from researchers puzzled by historically low U.S. saving rates and dramatic differences in saving rates across industrialized countries. While no consensus on the appropriate model of saving behavior has been reached, investigation of tax policy within a variety of neoclassical models suggests that governments can significantly influence the accrual of national wealth. This assessment of the current state of knowledge about neoclassical savings behavior and government policy differs markedly from a view, still accepted in many circles, that tax policy within the standard economic paradigm has highly ambiguous impacts on wealth accumulation. The ambiguity is associated with allegedly opposing income and substitution effects arising from changes in tax rates, in particular, capital income tax rates. While this allegation is true for a subset of tax changes, neoclassical models admit a host of policies with clear and potentially quite powerful effects on national savings. "Structural" tax changes provide an important example of such effective policies. They are defined here as changes in either the tax base or the progressivity of the rate structure used to generate a given time path of receipts while holding the time path of government consumption fixed. Holding constant the time path of receipts provides a prescription for setting tax rates through time. A key feature of these policies is that they "compensate" the private sector for imposing a new tax by removing an old one. Since these policies involve no changes in the timing and level of the government's direct absorption of resources, they leave unchanged the private sector's intertemporal consumption and leisure possibility frontier.1 If it so chooses the private sector can consume as much and work as little in the new tax regime as it does in the old. Such a response clearly leaves the private sector with the same collective resources to finance its own as well as the government's unaltered time path of consumption. The private sector will, however, generally respond to new tax regimes by choosing a different point on its budget frontier. The movement along this intertemporal frontier and its associated changes through time in savings and labor supply are determined by unambiguous substitution effects and, potentially, income effects. Although there is no aggregate change in income for the private sector, since its possibility frontier has not changed, structural tax changes typically redistribute resources across and within generations. However, for any structural tax policies the net income effects of this redistribution reinforce the substitution effects in raising or lowering savings. The distinction between compensated and uncompensated policy changes is clarified by considering the government's intertemporal budget constraint. Excluding the possibility that government debt can grow indefinitely at a faster rate than that of the economy, this constraint requires governments to equate the present value of receipts, inclusive of base money creation, to the present value of expenditures, inclusive of interest and principal payments on net government debt. While restricting the set of policies, including structural tax policies, that can be used over time to alter wealth formation, the government's long term budget is consistent with a wide range of short and medium term policies. In particular, the government can permit debt to grow fora long time at a faster rate than the economy, although indefinite use of this policy is not feasible.2 In addition to Structural Tax Policy, the government's intertemporal budget constraint suggests three other classes of policies for organizing a discussionof taxation and saving. These are: Tax—financed Changes in Government Consumption—— changes in the time path of government consumption that are contemporaneously financed by adjustments in tax rates of a particular tax base. Intergenerational Tax Policy—— reduced taxation of particular cohorts financed by increased taxation of other cohorts, holding constant the tax base and the time path of government consumption. In contrast to structural tax policy, this rule for setting tax rates involves changes in the time path of total annual tax receipts since total annual tax receipts need not equal annual government consumption. Intragenerational Tax Policy—— reduced taxation of particular members of a cohort contemporaneously financed by increased taxation of other members of that cohort. In contrast to broad structural changes in either the tax base or tax progressivity which are not cohort— _14 specific and, consequently, redistribute intergenerationally, these cohort—specific policies involve only intragenerational redistribution. Analysis of these four classes of policies provides insight into the economic impacts of the essentially unlimited broader combination of these and other policies. This survey is limited to the savings effects of fiscal policy in "full employment" neoclassical models, assuming, in most of the discussion, well functioning capital markets. The emphasis on "full employment" neoclassical models may be justified by the secular, rather than cyclical nature of the subject. In addition, the very significant new body of literature based on such models provides more than enough material for a single survey. Finally, much of the popular discussion concerning the behavior of savings and related variables over the cycle invokes these models, although not always validly. Section II provides background information for the subsequent discussion; it describes long—term trends in U.S. saving rates and domestic investment rates. Striking differences in postwar saving behavior among industrialized nations are also documented. This historical and international evidence provides a healthy perspective on the issue of taxation and saving. Few would argue that long—term trends in U.S. saving rates and the current low rate of U.S. saving relative to those of our principal trading partners are due solely, or even primarily, to differences in fiscal environments. Government policy is a significant determinant of national savings, but it is only one of several important forces at play. Other critical factors that jointly influence savings are intertemporal preferences, demographics, existing technology and the rate of technical change, patterns of human capital formation, the extent of market adianta—family lending, and the degree of market and intra—family risk—pooling. Each of the four stylized policies is considered sequentially in Sections 14through T. Their examination raises a number of traditional topics including the impact of government debt on national savings, the extent to which government consumption substitutes for private consumption, the distinction between investment and savings incentives, the savings impact of progressive taxation, the relationship between human and physical capital accumulation the degree to which borrowing constraints alter the analysis, and the welfare implications of alternative tax structures. The discussion in these sections, which takes up a large portion of this article, ignores issues of uncertainty. Many important and interesting insights can be obtained by working with oxides of certainty. The introduction of uncertainty, in some cases, obscures these insights. In addition, many of the results from certainty models carry over to models with uncertainty. In general, certainty models provide a benchmark for examining current policies in more realistic settings in which tastes, technologies, and indeed future government behavior is uncertain. Analysis of the four classes of policies as constrained by the government’s long term budget makes it clear that the fundamentals of fiscal policy are not aggregate "taxes", "spending", and "deficits", rather they are changes in marginal incentives, intra—marginal redistribution across and within generations, and direct government consumption. Section III examines government and private intertemporal budget constraints for purposes of precisely defining the policies considered in Sections IV through VII. The main objective of Section III is, however, to identify the government's underlying policy instruments and to caution the reader that official accounting definitions of taxes, expenditures, and debt provide little guide to the underlying fiscal structure. On the contrary, our proclivity to discuss such issues as savings and government policy in terms of this conventional nomenclature is symptomatic of widespread fiscal illusion. To drive home this point, Section III demonstrates how identical fiscal policies can be conducted with the government reporting essentially any level of taxes, spending, and deficits that it desires. Describing fiscal policy in terms of its effects on marginal incentives, intramarginal transfers, and direct absorption of resources requires changes in vocabulary and accounting, but a new fiscal language is necessary to discuss consistently the savings impact of our plethora of fiscal policies, all of which ultimately involve these basic elements. A key feature of the new fiscal language should be descriptions of the lifetime budget constraints of a small set of representative households as well as descriptions of the fundamental fiscal instruments that affect these constraints. Household budget constraints depend only on net marginal prices and net intramarginal endowments. Since the calculation of these prices and endowments are free of accounting conventions, a set of representative budget constraints would provide a coherent basis for analyzing the changes in and the consequences of a large variety of fiscal policies. Section VIII addresses the impact of taxation on savings in uncertain environments. A series of contributions (Evsey Domar and Richard Musgrave, l941; James Tobin, 1957; Jan Mossin, 1968; Peter Diamond, 1977; James Mirrlees, 19714; Diamond and Mirrlees, 1978; Martin Bailey, 1978; Steven Shavell and —7— Laurence Weiss, 1979; Hal Varian, 1980; Roger Gordon, 1981; Alan Auerbach, 1981;Jerenv Bulow and Lawrence Summers, 1982; Jonathan Eaton and Harvey Rosen, 1980;Robert Merton, 1981) expressly or by implication suggest that a variety of tax/transfer schemes constitute implicit insurance markets. To the extent that private provision of insurance is both available at the margin and is as efficient as government insurance, the government's pooling of risks through the tax/expenditure system will have few or no implications either for risk—pooling for national saving. At the opposite extreme one could imagine that, in the absence of government insurance, particular risks would be pooled neither in formal private markets nor in informal family settings. In this case the effect of government taxation on savings is analyzed by simply considering the impact of the availability of each particular type of insurance on national savings. Section VIII describes the potential insurance properties of a number of fiscal policies including capital and labor income taxation, and Social Security's unfunded tax/benefit provisions. Unfortunately, there are few studies that directly compare wealth accumulation with and without particular forms of insurance. Section VIII, as a consequence, draws inferences about taxation, insurance, and savings that are rather speculative in nature. These types of comparisons certainly represent a promising area for future research. The government as a cause of uncertainty, rather than a source of insurance, is an alternative possibility explored briefly in Section VIII. In principle the private sector could self—insure against random government redistribution ang current generations. Capricious redistribution between current and future generations is also potentially insurable assuming intergenerational altruism as in Robert Barro (19Th). Without such private insurance, the impact on savings of government imposed uncertainty is again determined by comparing economies with and without the particular risks generated by the government.

Sign in to recut, update, or save this card.