The effect of budget rules on fiscal policy.

• Author: Poterba, James M.
• Position: Research Summaries

Theories of what affects the size and persistence of government budget deficits have been advanced for centuries, surely since the beginning of government debt finance. Two recent developments in fiscal policy have generated renewed interest in the political economy of deficit determination, and in the impact of budget rules on fiscal policy outcomes. First, substantial, sustained, peacetime U.S. federal budget deficits have risen in the early 1980s. On the basis of national income and product accounts, federal deficits averaged 0.8 percent of gross domestic product (GDP) during 1960-79, compared with 3.3 percent of GDP in 1980-94. Second, there is substantial international dispersion in deficit experiences. Countries that appear to be similar in terms of standards of living, demographic composition, and other observable attributes may differ substantially in their ratio of deficit-to-GDP. This time-series and cross-national variation in deficits has led both researchers and policymakers to seek to understand the factors that contribute to deficit experiences.

As a reaction to chronic U.S. federal budget deficits, there has been debate about the structure of budget rules and whether they should be changed in an effort to alter budget outcomes. The 1986 Gramm--Rudman--Hollings antideficit bill and the 1990 Budget Enforcement Act are examples of recent federal legislation that has altered the budget process. The frequently discussed Balanced Budget Amendment, which would place explicit limits on fiscal policy outcomes, is a more extreme example of such an institutional reform. These actual and proposed reforms alter the way in which the participants in the budget process interact, and the rules under which they are expected to reach agreement. The maintained assumption underlying such proposals is that budget process rules matter for fiscal policy outcomes, and that changing these rules can affect fiscal deficits.

While a substantial body of largely theoretical research in political science supports the notion that legislative rules have real effects on policy outcomes, there is limited empirical evidence to support this view. Moreover, much of the empirical research in public economics that has considered the determinants of taxes and government spending proceeds as if budget institutions are simply veils, through which voters and elected officials see, and which have no impact on ultimate policy outcomes. For example, studies of spending determination in the "median voter" tradition usually maintain the hypothesis of institutional irrelevance by omitting these variables from analysis.

During the last several years, I have carried out a series of empirical research projects designed to analyze the impact of budget institutions on fiscal policy outcomes. These studies have taken various forms, but all have relied on cross-state variation in budget institutions in the U.S. states. My work has touched on a variety of topics, including the effect of balanced budget rules on state deficits, the impact of capital budgeting institutions on the level and mix...