The role of fiscal and political institutions in limiting the size of state government.

• Author: Krol, Robert
• Position: Report

In many states, tax and expenditure limits constrain government spending. All but one state have adopted balanced-budget rules. Some governors have the power to veto individual budget items (the so-called line-item veto). This article reviews the evidence linking fiscal and political institutions to state taxation, spending, and debt.

It appears that properly designed fiscal and political institutions are effective in containing the growth of state government. Constitutional tax and expenditure limits have been more successful than spending constraints established legislatively. Balanced-budget rules that prohibit deficit carryover to the following fiscal year are superior to rifles that allow deficit carryover.

Researchers have identified two other relationships. First, there is evidence that fiscal rules reduce state borrowing costs. Second, the citizen initiative process has played a role in controlling spending: states with a citizen initiative process spend less.

State Spending and Fiscal Controls

In the United States, state government spending has grown rapidly, nearly doubling since 1995. Population and prices have grown as well, but at a much slower rate. The result has been a significant expansion in real per capita state government expenditures. Although the finances of state governments have improved considerably in the last few years, overspending in the 1990s was a major contributor to the fiscal problems that plagued state governments in 2001-03.

Conventional views of government do not offer a justification for fiscal controls. For example, Downs (1957) argues that elected officials provide public services consistent with the preferences of the median voter. From this perspective, politicians act to maximize the net benefits of the median voter and there is no need to limit government. Similarly, Tiebout (1956) contends that businesses and individuals affect the size of government by voting with their feet, leaving jurisdictions with levels of spending they consider too high. In this view, competition between states brings about the right level of spending.

Alternatively, the special-interest view of government (Stigler 1971, Peltzman 1976, Becker 1983) and the Leviathan view (Niskanen 1975, Brennan and Buchanan 1979) predict that the influence of lobbying groups and the behavior of self-interested bureaucrats result in levels of government spending that exceed what the average voter desires. In contrast to conventional views of government, special-interest and Leviathan views place far more emphasis on the role public institutions and rules play in facilitating (and limiting) the growth of government.

Certainly state governments have a role to play in the provision of public goods and services, such as police, courts, and infrastructure. They also play a role in transfer programs. However, there is increasing concern over the growth in state spending. Economists and others worry that excessive spending results in higher taxes that, in turn, stifle economic activity (Barro and Sala-i-Martin 1995 and Engin and Skinner 1996). In the political arena, this concern has been manifested in the various attempts to establish fiscal and political institutions citizens can use to control the growth of state government.

Fiscal Institutions

Tax and Expenditure Limitations

Tax and expenditure limitations (TELs) are rules that attempt to constrain the growth of a state's revenues or expenditures. Most TELs limit the increase in expenditures to the growth of state personal income or the growth in state population plus inflation. Poulson (2005) reports that 30 states have some form of a TEL limitation. The majority (18) are constitutional, while the remaining (12) are statutory. Statutory TELs tend to be weaker and easier for the legislature to modify or avoid.

The early studies by Abrams and Dougan (1986), Cox and Lowery (1990), and Bails (1990) found TELs to be ineffective in controlling the growth of state government expenditures. However, those studies examined a cross section of states at one point in time. With such limited data, it is difficult to control for all the factors that influence spending.

More recent studies look at states over time. With far more observations, they are able to control for observable and unobservable (using fixed-effects estimators) factors that influence spending. These studies do, in fact, find that TELs reduce state government spending. In examining the effect of TELs, it is important to isolate a political response to a change in voters' attitudes (which may lead to the imposition of a TEL) from the direct effect of a TEL itself. Using an instrumental variables approach, Reuben (1996) finds that the presence of a TEL reduces state spending by about 1.8 percent (partially offset by an increase in local spending). Bails and Tieslau (2000) confirm Reuben's results. Their estimates indicate real state and local expenditures per capita to be 841 less, on average, in states with TELs.

Poterba (1994) examines the influence of state TELs on the budget adjustment process following an economic downturn. Looking at the 1991-92 recession, Poterba finds that budgets in states with TELs tend to adjust faster. When revenues fall, TEL states are less likely to increase taxes to balance their books. For a 81.00 budget deficit increase, TEL states increase taxes by 80.47 while non-TEL states increase taxes by 81.03. Poterba finds no difference in spending reduction behavior between TEL and non-TEL states; TEL states are more likely to run a deficit than to increase taxes in response to a negative budget shock. Poterba's work suggests that the most effective TELs are constitutional, written by voters, and those that limit spending rather than revenues.

Poterba and Rueben (1999) examine the impact of TELs on general obligation state debt yields. While state general economic conditions and the level of outstanding debt influence the yield, fiscal institutions have an independent influence. Poterba and Rueben examine data from 40 states with significant borrowing during the 1973-97 period. They find that where fiscal institutions are in place to limit expenditures, borrowing costs are an average of four basis points lower. In contrast, limitations that constrain revenue growth result in borrowing costs that are 17.5 basis points higher, on average. Clearly, financial markets react to statutory and constitutional fiscal constraints. With revenue limitations, lenders may fear that state governments would not be able to raise funds needed to service outstanding debt. Expenditure limitations have the opposite effect. They control the growth of general spending, making it easier to service outstanding debt.

To summarize, recent research indicates that TELs can slow the growth of government. TELs linked to expenditure growth also reduce borrowing costs on public debt.

Line-Item Veto

Governors exert considerable influence over state budgets. Thirty-three governors have line-item veto power. With this power, a governor can eliminate specific items (lines) in the budget without rejecting the entire budget. The line-item reduction veto provides the governor with even greater budget policy flexibility. If a governor thinks a program has some merit but considers it overfunded, he or she can reduce the level of funding without canceling the entire program. Eleven governors have item-reduction veto power. When legislators appropriate funds for programs that benefit special interests in their district, these appropriations may differ significantly from what the typical statewide voter would want. The line-item veto gives the governor the legal power to align program allocations with statewide priorities.

Much like the literature on TELs, the early studies on the line-item veto found little impact on spending (Rowley, Shughart, and Tollison 1986; Nice 1988; Ahn and Evers 1991). These studies had similar methodological problems. The focus was on cross-sectional analysis; researchers were unable to control for other...