If [government] cannot raise its revenue in proportion to its expanse, it ought, at least, accommodate its expence to its revenue. (Smith [1776] 1976: 946) Pyramid-building, earthquakes, even wars may serve to increase wealth, if the education of our statesmen on the principles of the classical economics stands in the way of anything better. (Keynes, 1936: 129)
Public spending may be financed in one of three ways: by levying taxes on the private sector, by printing money, or by borrowing. In the United States and much of the industrialized world, the last of these tools was used sparingly prior to roughly 1970. Only wars and other national emergencies prompted governments to resort to the bond market; the public sector otherwise operated on a balanced-budget, pay-as-you-go basis. Furthermore, the issuance of public debt to finance wartime spending was simply an unavoidable expedient: "An immediate and great expence must be incurred in that moment of immediate danger, which will not wait for the gradual and slow returns of the new taxes. In this exigency government can have no other resource but in borrowing" (Smith [1776] 1976: 909). Once the hostilities had ended, the accumulated debt customarily was retired, often through the establishment of a ‘sinking fund’, consisting of revenues earmarked specifically for that purpose. In most times and places, the restoration of peace brought a return to public budget balance. The normative principle that, except for periods of ‘hard necessity’, government should live within its means was rarely questioned and widely practiced.
For reasons not yet well understood, that pattern was broken in the last third of the twentieth century (Anderson, 1986). Chronic budget deficits became the peacetime norm in the United States until 1998, when, owing to dramatic cuts in defense spending and to the robust economic expansion of the 1990s, the federal government’s receipts exceeded its outlays for the first time since 1969 (Alesina, 2000). The proximate cause of this period of persistent budget imbalance was massive growth in so-called entitlement programs, especially Medicare and Medicaid, established to help pay the health care bills of elderly and poor Americans. Such programs, which are open-ended in the sense of providing benefits to everyone who meets predetermined eligibility criteria, ensured that spending would rise continuously with increases in the populations of qualified recipients, even with no changes in benefit levels or eligibility requirements. And, indeed, if the future liabilities of these programs are added to those of the social security system, the ostensible budget ‘surpluses’ of the late twentieth and early twenty-first centuries quickly sink in a sea of red ink.
In fact, because governments generally do not follow the standard accounting practices accepted for use in the private sector, the magnitude of the public budget’s net balance (and even its algebraic sign) is a matter of considerable scholarly controversy. Whether or not the public budget is in surplus or deficit at any point in time — and by how much — depends on the treatment of items such as the financing of long-lived capital projects, the future obligations accrued by social insurance and other entitlement programs, the revenues and expenses of state-owned enterprises (and the disposition of the proceeds realized from the sale of such properties), the assets and liabilities of government loan programs, publicly owned lands and mineral rights, and many other public sector activities, both off-budget and on, having important fiscal consequences. Because accounting conventions differ widely across nations, as do the sizes and scopes of their public sectors, cross-country comparisons of fiscal stance are even more problematic (Blejer and Cheasty, 1991).
Whether one assesses the government’s budget conventionally as the simple difference between current revenues and current expenses, or adjusts it to include the present values of the public sector’s most significant future obligations, it seems clear that the last third of the twentieth century witnessed a unique period in the history of public finance. Why did budget deficits appear suddenly in the late 1960s and why did they persist for 30 years (or more)?
‘Ideas matter’ is one answer to that question. The norm of governmental fiscal responsibility rested for 160 years on the intellectual foundations laid by the classical economists, who for the most part viewed public debt as inimical to economic growth. With the exception of Thomas Malthus, the classicals thought that, by competing for scarce loanable funds and, hence, by diverting wealth from relatively productive private investment projects to relatively unproductive public spending programs, government borrowing impairs capital formation and, in so doing, makes a nation poorer than otherwise. That is true even if all of the public’s debt is ‘internal’ (i.e., held domestically). Adam Smith, for one, recognized that, while the government’s creditors have a general regard for the national welfare, insofar as their interest income is contingent on continued prosperity, the holders of the public debt do not have the same incentives for deploying resources to their highest valued uses as the owners of those resources, who are taxed to service the debt. An income transfer between these two groups therefore ‘must, in the long-run, occasion both the neglect of land, and the waste or removal of capital stock’ (Smith [1776] 1976: 928).
Smith’s reasoning explodes the hoary myth — hoary apparently even in 1776 — that the payment of interest on the public debt has no real economic consequences because "it is the right hand which pays the left" or because "we owe it to ourselves" (ibid.: 926). That "apology for the public debt", founded on the "sophistry of the mercantile system" that Smith exposed so ably, was also wrong as a factual matter then, as it is now, "the Dutch, as well as several foreign nations, having a considerable share of our publick funds" (ibid.: 927).
The classical analysis of budget deficits and public debt was swept away by the Keynesian revolution. Writing at a time when the global economy seemed to be in freefall and the private market economy seemed incapable of self-correction, John Maynard Keynes (1936) argued forcefully that full employment would be restored only if governments intervened aggressively, using their fiscal policy tools to offset the calamitous and apparently permanent decline in private investment spending that began in 1929. Keynes prescribed increases in government spending, on useful public works if possible, but on pyramids if need be, in order to inject purchasing power into the economy and put people back to work. Every new dollar spent by the public sector would increase national income many times over through the operation of a ‘multiplier effect’ as consumption spending by the initial recipients passed into the hands of merchants, who in turn spent a portion of their now higher incomes, enriching others who increased their own expenditures, and so on and so on. If, in addition, the increase in government spending was financed by debt (what Keynes called ‘loan expenditure’), the holders of the bonds would experience a ‘wealth effect’ that would generate further increases in private consumption and investment spending. Using its considerable resources to augment aggregate demand, the public sector is thus able, in the Keynesian system, to jump-start a stagnant economy, setting it on the return path to full-employment equilibrium.
In imparting intellectual respectability to deficit spending, Keynes destroyed the norm of public budget balance (Buchanan and Wagner, 1977). Indeed, the relationship between public revenues and expenditures became an unimportant byproduct of government’s Keynesian responsibility of actively countering the peaks and troughs of the business cycle so as to maintain the economy at full employment. These ideas were taken to their logical extreme by Keynes’s disciple, Abba Lerner (1943), who rejected totally the classical orthodoxy in favor of a doctrine of ‘functional finance’, which judges fiscal policy, not by its impact on budget balance, but by its impact on the economy. Lerner went to great lengths in attempting to dispel the ‘fairy tales of terrible consequences’ from undertaking a prolonged program of deficit spending, if that was what was needed to deal with chronic economic stagnation. He clung tenaciously to the view that, because "we owe it ourselves" (or "to our children or grandchildren and to nobody else"), internal debt’s possible adverse effects are largely "imaginary". Government can therefore borrow freely, even for the purpose of paying interest on its outstanding debt, without imposing any untoward burden on the economy.
Public choice scholars entered the fray at a time when post-Keynesian macroeconomists were in the process of reexamining the effects of public debt and concluding that government borrowing does in fact impair private capital formation as the classicals had taught. Analyzing the public debt problem from the perspective of the individual economic actors who, as citizens of a democratic polity, collectively must choose methods of financing the expenditures of government, James Buchanan ([1958] 1999: 26-37, [1964] 1982) observed that the decision to borrow involves a tradeoff between present and future taxes. The fact that the national debt must be serviced and retired in the future implies a future tax liability. Accordingly, it is future taxpayers who shoulder the burden when government borrows to finance current spending.
The debate became livelier following Robert Barro’s (1974) reformulation of what has since erroneously been called the doctrine of ‘Ricardian equivalence’ (O’Driscoll, 1977). Barro’s theoretical model starts with the assumption that the members of each generation care about the welfare of the next (more precisely, that the utility attained by one generation depends partly on its own consumption and partly on the utility attainable by that generation’s immediate descendants). If, in addition, there exists a "chain of operative intergenerational transfers" (private bequests) that connects the current generation to future generations, individuals will behave as if they live forever. Under these assumptions, the issuance or retirement of public debt has no differential impact (relative to the tax alternative) on personal wealth, on aggregate demand, or on capital formation because current taxpayers will alter their bequests to offset the implied change in future tax liabilities. Bequests will be increased to compensate future generations fully for the heavier tax burden otherwise imposed on them by increases in public indebtedness and, when debt is retired, bequests will be lowered by the full amount of the reduction in the future tax burden. Changes in future tax liabilities, in other words, are fully capitalized in inter-generational wealth transfers, thereby neutralizing completely the effects of changes in the government’s budget balance. Debt and taxes are equivalent tools of public finance.
Buchanan (1976) replied that such equivalence is illusory because rational taxpayers will predictably respond to an increase in the public debt, which implies a corresponding increase in future tax liabilities, by shifting income from the future to the present. The attempt on the part of current taxpayers to lower their future tax bills by reallocating their incomes intertemporally means that individuals will save less under deficit finance than they would under the revenue-equivalent current period tax, thereby impairing capital formation. It also means that, at prevailing tax rates, future tax collections will not be adequate for meeting debt service and amortization obligations.
Buchanan’s broader point was that, even if future tax obligations are fully anticipated, taxpayers are placed in a prisoners’ dilemma situation with respect to public debt issues. In particular, future tax liabilities in the Barro model are contingent because each individual is required to make spending plans for himself (and for his immediate descendents) under the assumption that everyone else will likewise plan to discharge his pro rata share of the community’s deferred tax liabilities. However, if any one taxpayer fails to do so, the other members of the community will find their future tax bills to be larger than expected even though they themselves acted responsibly. Moreover, if one taxpayer has incentive opportunistically to shift some (or all) of his future tax burden to others, everyone does.
Under a pay-as-you-go system of public finance in which current spending is financed by current taxes, a steady stream of public revenue is more or less assured because fluctuations in individuals’ incomes tend to cancel out. No such offsets occur with loan-financed expenditure, however, because all individuals must accumulate sufficient funds to pay their shares of the community’s future tax bill. Under circumstances where it is costly to monitor the spending plans of fellow taxpayers and, moreover, where individuals can shift their future tax liabilities to others by acting irresponsibly from the community’s perspective, tax-financed expenditure will be preferred to debt-financed expenditure. Once again, ‘equivalence’ does not hold as a theoretical proposition. Nor does the weight of the evidence seem to support it (Evans, 1993; Stanley, 1998).
In any case, the historical record of the past 30 years, a period distinguished in much of the West by persistent government spending in excess of current revenues and ever-growing public debt, seems inconsistent with the neo-Keynesian orthodoxy, which calls for budget balance over the business cycle. One explanation for the theory’s failure to fit the facts is that the political institutions governing fiscal policy choices and the political actors who formulate and implement the chosen policies are absent from the analysis. Neo-Keynesians assume that government is exogenous to the economy and portray its fiscal policymakers as being guided by some version of the public’s interest, selflessly pursuing broad social objectives such as economic stabilization or ‘tax smoothing’ (i.e., acting to prevent volatile changes in tax rates over the business cycle). Fiscal policy thus responds mechanistically, predictably, and impartially to given economic conditions.
The neo-Keynesian gap between theory and political reality is filled by public choice, which brings government within the ambit of the macro economy. Public choice is essentially an exercise in modeling the behavior of self-interested political agents in a given institutional setting and, as such, offers a rich set of testable hypotheses about why democratic political processes might produce a bias toward budget deficits, a bias reinforced by the electorally foreshortened time horizons of politicians and by the ‘fiscal illusion’ of rationally ignorant voters, who underestimate their future tax liabilities. The literature proceeds on several levels.
At one level there is a consideration of the base-line incentives for politicians to prefer spending to taxes, and so to be driven by their self-interest to a policy of deficits. Deficits involve easy choices and budget balance hard choices for politicians. By deferring tax obligations to the future, loan-financed expenditures afford politicians the opportunity to shift the responsibility of paying for current government spending programs to voter-taxpayers, some yet unborn, living beyond the end of the incumbents’ electoral time horizons. A balanced-budget increase in public spending financed by higher current taxes, on the other hand, threatens incumbents with an immediate loss of political support. Hence, there is a natural tendency, grounded in the vote motive, for democratic governments to run deficits.
At a deeper level the various institutional features of representative democracies can influence the incentives of political agents to run deficits. Here, we encounter issues of dynamic policy consistency and the durability of public spending as aspects of political behavior that can tip the balance in favor of deficit finance (Persson and Svensson, 1989; Alesina and Tabellini, 1990). For example, a fiscally conservative administration may run deficits to tie the hands of a successor liberal administration. By forcing a larger fraction of future tax revenues to be used for servicing the public debt, a roadblock is placed in the way of the next regime’s plans for launching major new spending initiatives. Alternatively, a liberal government may systematically underestimate the future spending requirements of a policy proposal to get the camel’s nose under the tent, thereafter depending on the program’s beneficiaries or the fallacy of sunk costs to sustain a steady stream of funding.
The voters themselves play decisive roles in public choice analyses of fiscal policy choice. Simple majority voting rules afford opportunities for taxpayers who benefit disproportionately from public spending programs to shift the burden of financing those benefits to others (Browning, 1975; Tabellini and Alesina, 1990). Because voters in the present period cannot make binding contracts with voters in the next period, incentives likewise exist for supporting second-best policy options in order to avoid even worse outcomes in the future (Glaszer, 1989; Crain and Oakley, 1995). Because the configuration of costs and benefits facing individuals differs when choices are made collectively than when they are made privately, the rational behavior of self-interested voters may cause public spending to grow more rapidly than otherwise, requiring corresponding increases in the amounts that government borrows and taxes.
Bringing public choice principles to bear also suggests that popular ideas like term limits affect the fiscal behavior of politicians. The behavior of term-limited governors (‘lame ducks’) differs markedly during their last terms in office compared with earlier terms, for example. Taxes and public expenditures tend to be higher when the chief executive cannot run for reelection and, because of this, the time series of fiscal variables (taxes, spending, and debt) tends to be more volatile in states that impose gubernatorial terms limits than states that do not (Crain and Tollison, 1993; Besley and Case, 1995).
Emphasizing that institutions matter, public choice analyses of fiscal choices have shown that constitutional limits on taxing and spending can effectively constrain deficit finance and the growth of government (Porterba, 1997). Certain features of legislative organization and of public budgetary processes and procedures have likewise been found to be significant in maintaining fiscal discipline (Crain and Miller, 1990; Crain and Muris, 1995; Crain and Crain, 1998). Such rules and institutions are as important at a time of ostensible budget surpluses as they may have been during the era of high deficits. In the period since 1950, new spending has absorbed 73 cents of every surplus US budget dollar, on the average, whereas 21 cents was used to reduce the public debt and only a nickel was returned to the taxpayers (Vedder and Gallaway, 1998). The public choice lesson is that discussions of the appropriate size and scope of government cannot be separated from discussions of the appropriate mix of debt and taxes used to finance it.
