Institutions for fiscal stability.

Alesina, Alberto

The seventies and most of the eighties have been a period of fiscal profligacy in many countries around the world. Several OECD countries have accumulated debt/GDP ratios which are unprecedented, except for the aftermath of major wars. Low public savings have been at the root of Latin America's "lost decade," the eighties. Currently, the goal of achieving and maintaining fiscal stability is the main macroeconomic issue in many parts of the world.

This evolution of fiscal policy raises many intriguing questions: Why have many but not all countries abandoned fiscal discipline? What explains the very large cross-country variance in fiscal stance? Why did large and persistent deficits appear in the mid-seventies and not before? What explains the likelihood of success of fiscal adjustments? Why did certain countries make swift and very successful fiscal consolidations, while others are still struggling?

The answers to all of these questions cannot rely purely on economic factors since economically similar countries exhibit very large differences in fiscal performance. In a series of recent papers, I have addressed these questions by considering politico-institutional explanations.

In a paper(1) coauthored with Roberto Perotti, I identify two critical institutional variables as important determinants of the fiscal policy stance: the degree of government fragmentation, and the nature of budget institutions. In terms of the former variable, we argue that coalition governments are more likely to delay the adoption of stabilization policies, because of inter-coalition struggles leading to legislative deadlocks. Thus, after the oil shocks of the seventies, countries ruled by fragmented coalitions reacted more slowly and less decisively, letting budget deficits accumulate. In a second paper,(2) we show that when coalition governments actually attempt to stabilize the budget, they often fail because they do not have the political strength to deal with structural budget cuts in social spending and government wages.

While this argument about government fragmentation is relatively well understood, the issue of budget institutions is more complex and multifaceted. In the last few years a vast research program to which I have contributed has investigated how different procedures influence fiscal outcomes, both in the OECD group of countries, and in a sample of Latin American countries and the American states. This research effort leads to the conclusion that budgetary institutions do matter as a determinant of fiscal outcomes, and therefore different choices about fiscal institutions may lead to a higher or lower propensity to run excessive deficits.

In a third paper,(3) Perotti and I identify several theoretical and empirical issues which are central for the discussion of how institutions affect fiscal outcomes. We define budget institutions as all of the rules and regulations according to which budgets are drafted, approved, and implemented. We focus mainly on three issues: balanced budget rules; procedural and voting rules; and transparency.

Balanced Budget Rules

Well-known economic arguments suggest that balanced budget rules are not optimal, because they do not allow deficits to fluctuate over the cycle and in the event of major and temporarily high spending needs. However, since for many political reasons politicians may have incentives to run excessive deficits, balanced budget rules may serve the purpose of correcting a politically induced distortion in fiscal policy. However, an unpleasant consequence of balanced budget rules is that they generate incentives for circumventing them. In so doing, policymakers engage in tactics of creative accounting which make the budget less transparent, creating additional obstacles to an effective control on fiscal discipline.

For all of these reasons, balanced budget rules at the national level may be counterproductive. Instead, fiscal discipline can be enhanced by appropriate procedural rules (discussed later) which do not require a numerical target on the budget balance. The same argument, however, may not apply to subnational levels of government, such as American states. Local and state governments may need less flexibility because their budget are less cyclically sensitive. Several authors have investigated the effects of restrictions on budget deficits in American states: this literature concludes that more stringent balanced budget rules lead to more fiscal discipline. My own contribution to the literature includes a paper with Tamim Bayoumi of the IMF.(4) In this article, we show that more stringent fiscal rules enforce fiscal discipline without any apparent negative effect on state output volatility.

Procedural Rules

One can identify three phases in the budget process: the formulation of a budget proposal within the executive; the approval of the budget in the legislature; the implementation of the budget within the bureaucracy. In my research I have focused almost exclusively on the first two points.

Voting procedures can be classified on a hierarchical-collegial dimension. Hierarchical procedures are those that, for instance, attribute strong prerogatives to the Prime Minister (or Treasury Minister) to overrule spending ministers in intragovernmental negotiations. Also, hierarchical institutions limit the latitude of legislative amendments on the budget. For instance, in some cases the legislature can change the proposed budget without affecting the balance. Even more stringent rules require the legislature not to increase either the level of spending or the deficit. Thus, in the latter case, the legislature can only change the budget allocation between spending programs. Collegial institutions have the opposite features: they emphasize "consensus" at every stage of the process, by enhancing the prerogative of all spending ministers in the government, the prerogative of the legislature vis-a-vis the government, and generally by upholding the right of the minority in every stage of the process.

One can identify a tradeoff between the two types of institutions. Hierarchical institutions are more likely to enforce fiscal restraint, to avoid large and persistent deficits, and to promote swift fiscal adjustments when needed. On the other hand, the same institutions are less respectful of the prerogative of the minority not in the government, and therefore are more likely to generate budgets tilted in favor of the governmental coalition. Collegial institutions have the opposite features.

One related important issue concerns the order of voting. In some cases the budget procedures imply that the legislature first has to approve a balance, often in the context of a macroeconomic scenario for the coming fiscal year. Then, in later votes the allocation among different programs is decided. The alternative procedure implies that the balance of the budget is the residual of a series of votes on specific programs. The Budget Act of 1974 in the United States implied, among other things, a switch from the latter system to the former. Intuitively one would think that the system where the balance is voted first should promote more fiscal restraint. Indeed, this is what the cross-country empirical evidence seems to suggest. However, the theoretical underpinnings for this result are not very strong, if one assumes rational and forward looking behavior of legislators.

Transparency of the Budget

The budgets of modern economies are very complex, but sometimes they are more complex than they need to be. This complexity, partly unavoidable, partly artificial, makes it possible to hide the real status of public finances, in particular the current and future burden for the taxpayers of various spending decisions. Politicians have incentives to hide taxes, emphasize the benefits of spending programs, and hide government liabilities, equivalent to future taxes, by using various forms of creative accounting procedures. The more complex is a budget document, the easier it is to confuse the public.

The importance of lack of transparency cannot be overemphasized. A variety of tricks are used to strategically influence the information/beliefs of the taxpayers-voters: 1) Overestimation of the expected growth in the economy, so as to overestimate tax revenues, and to underestimate the level of interest rates so as to underestimate outlays. At the end of the fiscal year, the "unexpected" deficits can be attributed to "bad luck." 2) Over optimistic forecasts of the budget effects of various policies. 3) Strategic use of what is kept in and out of the budget, often with a creative use of the budget of various public organizations. 4) Strategic use of multi-year budgets, to the effect that difficult policies are permanently postponed to year two or three of a multi-year program and always delayed.

Issues of transparency and creative accounting are, in fact, at the forefront of the fiscal debate in Europe. The discussion about which countries can join the European Monetary Union has paid much attention to how "real" or "creative" are the fiscal adjustments in many European countries that are reaching the required deficit target. Several observers have noted how Germany, France, and especially Italy in recent years have used various ingenious methods to make their deficits appear as low as possible.

In summary, this discussion suggests that "hierarchical-transparent" procedures should be associated with more fiscal discipline. Thus, difference in procedures can contribute to explaining the cross-country differences in fiscal policy stance that are documented here.

Empirical work on this issue shows the difficulty of measuring institutions. Work by yon Hagen and his associates focused on European countries and concluded that fiscal institutions do matter in the expected direction. In my own work with coauthors, I have studied Latin American countries from this point of view.5 Using the answers from a survey distributed to the budget directors of all Latin American countries, and the text of the budget laws, we constructed a comprehensive index which summarizes several characteristics of fiscal procedures, along the "hierarchical transparent" to the "collegial non-transparent" dimensions. We then discussed the relationship between the index and various components of it, and the level and evolution of budget deficits in this region. We also examined cases of changes in procedures, namely whether one can detect a difference in the fiscal position of a country before and after a reform of its fiscal procedures. Our results confirm that budget procedures do matter. After controlling for several economic determinants of budget deficits, our index of procedures was still significantly correlated with budget deficits in the expected direction. A particularly important feature of such procedures is the one that requires a vote on the size of the deficit ex ante, in the context of the approval of the macroeconomic plan for the year, before the legislative discussion on the composition and allocation of the budget even begins.

Finally, evidence drawn from American States, European countries, and Latin American countries all points in the same direction: different budget procedures influence fiscal outcomes. Two critical issues then follow. What determines institutional choice? Why do different countries or states choose different fiscal institutions and, therefore, what determines institutional change? In other words, the research can be moved one step backward by looking at the determinants of institutions. The second issue is normative: this research can shed light on how to design institutions which contribute to maintaining fiscal stability and limit the extent of politically induced distortions.

1 A. Alesina and R. Perotti, "The Political Economy of Budget Deficits," NBER Working Paper No. 463 7, February 1994.

2 A. Alesina and R. Perotti, "Fiscal Expansions and Fiscal Adjustments in OECD Countries," NBER Working Paper No. 5214, August 1995.

3 A. Alesina and R. Perotti, "Budget Deficits and Budget Institutions" NBER Working Paper No. 5556, May 1996. Forthcoming in a conference volume edited by James M. Poterba and Jurgen yon Hagen and published by University of Chicago Press for NBER.

4 A. Alesina and T. Bayoumi, "The Costs and Benefits of Fiscal Rules: Evidence from U.S. States," NBER Working Paper No. 5614, June 1996.

5 A. Alesina, R. Hausmann, R. Hommes, and E. Stein, "Budget Institutions and Fiscal Performance in Latin America," NBER Working Paper No. 5586, May 1996.

Alberto Alesina is a Research Associate in the NBER's Program on Monetary Economics and a professor at Harvard University. His profile appears later in this issue.