This article was originally published in the The Encyclopedia of Taxation and Tax Policy (1999 Urban Institute Press), edited by Joseph J. Cordes, Robert D. Ebel, and Jane G. Gravelle. The Encyclopedia is available in cloth and CD-ROM editions. Order online at www.uipress.org or call toll-free 1-877-847-7377.
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
This report is available in its entirety in the Portable Document Format (PDF).
Annual procedures whereby state governors submit budgets for legislative review and disposition.
While state governments in the United States are often noted for their diverse approaches to politics and policies, their budget processes are to a great extent very similar (Hutchison and James 1988).
In almost every state, proposed budgets are put together in the executive branch. Most state fiscal years run from July 1 to June 30, and governors begin the process in July or August by having their budget offices send out instructions for budget requests to the various departments and agencies. These instructions usually give each spending unit a rough idea of what they might expect in continued, expanded, or (rarely) reduced resources. By the fall, agency requests are received by the budget office and reviewed, often with an eye toward cutting expenditures seen as unnecessary. Hearings on the budget requests may be conducted within the executive branch and might include representatives of the governor, budget officials, and department heads, who may wish to appeal the decisions of the budget office. The governor then prepares a final budget proposal, which includes projected revenues and appropriations, and a budget message (Rosenthal 1990).
Governors customarily submit proposed budgets to their state legislatures in late January or early February. Most commonly, the budget is referred to one committee in each house, usually called the "appropriations" committee. This is very different from (and far more efficient than) budgeting at the federal level, where each house has a budget committee, 13 authorization committees, and an appropriations committee, divided into many subcommittees. After committee hearings, debates, markup sessions (where bills are gone over and changed line by line), and votes, the budget bill is sent to the floor of the whole chamber, where it may be amended and is ultimately voted on.
Budget bills usually originate in the lower house, mirroring the practice for revenue bills at the federal level, and following the philosophy that all revenue actions should be initiated in the chamber closest to the people in terms of number of constituents per representative and frequency of elections. A similar process then occurs in the Senate, with committee work preceding debate and final action on the Senate floor. In most cases, a budget that passes a state senate will differ in some manner from the budget that passes the house, and a conference committee will be appointed by legislative leaders to work out a compromise. That compromise must then be approved by both houses of the state legislature.
The budget bill is then sent to the governor. Governors may sign the bill, veto the entire bill (which almost never occurs), or in most states sign it with a line-item veto, allowing the governor to strike portions of the bill that he or she dislikes. Any vetoes then return to the state legislature for override attempts, usually requiring two-thirds of the state legislature. Some evidence suggests that while the item veto is thought of only as a tool of fiscal restraint, it is frequently used to alter policy (Gosling 1986). Item vetoes are rarely overridden.
Important variations in state budget processes do exist. In 21 states, the budget is a biennial one, passed every two years instead of annually. The argument for biennial budgeting is that it is more efficient, allowing the legislature to spend time on other important policy issues, time that would otherwise be consumed by the budget process. Evidence on the effect of biennial budgeting on state policy is unclear, to a great extent because there is no agreement on how to identify "good policy." In another variation, 15 states have joint budget committees, with representatives of both houses of the state legislature receiving the budget from the governor.
In a few states, including California, Rhode Island, and Nebraska, a super-majority of legislators is needed to pass a budget. The goal of super-majority approval is fiscal discipline, the argument being that it will be more difficult to get, say, two-thirds of a state legislature to agree to a budget with unnecessary spending.
Finally, variations exist in the veto powers of governors. Governors in 43 states have the power of line-item veto, meaning that they can veto portions of a budget passed by the legislature. But in 26 states, governors may also veto language in an appropriations bill, giving them more power (Council of State Governments 1996). In addition, provisions for override vary by state. Giving the governor the most power is an override that calls for two-thirds of elected legislators to override; less restrictive is the provision for override by two-thirds present at the time of the vote.
The roles of the main participants in the state budget process are fairly typical in all states. The governor may be viewed as budget balancer for two reasons. First, the governor is the one individual in the process who represents the entire state, and a balanced budget is a common denominator for most citizens, at least in principle. Second, submitting a budget in balance gives the governor power over the legislature in determining the ultimate shape of the budget. Facing a proposed budget in balance, a legislature desirous of tax cuts would have to find expenditure reductions to finance them, while a legislature that sought to increase spending over a governor's request would have to find the added revenues to finance it. It is usually easier to accept the broad parameters of the governor's budget.
State legislators can generally be characterized as spenders or tax-cutters, as they seek to please their constituents by bringing benefits back to their districts, or use the budget to reward special interests who may have provided the rationales for budget items, and perhaps gained access by means of campaign contributions. Department or agency heads are also generally spenders, since in the public sector size of budget is a measure of success, a substitute for profit in the private sector. Most individuals who head agencies also believe in their mission and are therefore likely to want to spend more, and have a clientele of their own to serve.
The governor has much more power in the state budgeting process than the president does in federal budgeting. First, governors usually submit a budget in balance, which as already noted makes it more difficult for the legislature to change. Second, governors have an informational advantage over state legislators, and the imbalance of staff resources at the state level tilts much more strongly toward the executive than it does at the federal level. Third, in many states the job of state legislator is a part-time one, or at least allows legislators to pursue other professional activities, giving the governor the advantages of time and focus. Fourth, the line-item veto gives a governor more power than the president has traditionally had, and is generally stronger than the version recently enacted by Congress, which allows a simple majority of members to specify a law as not subject to presidential line-item veto.
State budgeting for the most part continues to be "traditional" or "line-item" budgeting, with incrementalism the prevailing decision system employed. As a consequence, most of the emphasis of state budgeting is placed on analyzing the change in inputs from the previous year's budget (Rosenthal 1990: 136). While producing stability and making it easier to compromise, incrementalism does result in some programs outliving their usefulness, or at least growing beyond need. Comprehensive reform measures have for the most part been unsuccessful, including zero-based budgeting and planning, programming, budgeting systems (National Conference of State Legislatures 1995). Zero-based budgeting sought to have programs analyzed from top to bottom, with varying output levels estimated for various levels of resources. Planning, programming, budgeting systems sought to move from line-item budgeting to the presentation of budgets in program packages, with multiyear planning and performance measures. Both reforms failed as a result of calculation problems, as well as political and organizational resistance. Successful reforms have included better revenue estimating, often involving joint legislativeexecutive committees, and multiyear forecasting of both revenues and expenditures. Performance budgeting has increasingly found its way into state budgeting, especially on the executive side.
Budgeting at the state level is greatly affected by legal, political, economic, and demographic variables. Legally, all states except Vermont are either statutorily or constitutionally required to balance their budgets. Most states also have limits on how much they can borrow, usually expressed as a percentage of appropriations. Twenty-one states face constitutional or statutory tax and/or expenditure limitations (TELS), most a product of the tax revolt kicked off by California's Proposition 13 in 1978 (Advisory Commission on Intergovernmental Relations 1991). Taxes are often limited to a percentage of the tax base, or expenditure growth may be tied to inflation plus population growth. These TELS usually require a super-majority override by the legislature, or public override by referendum. Political variables include tax competitionthe fear that taxpayers will flee a state if its tax burden is much higher than that of a neighboring state.
Many exogenous variables have an important impact on state budgeting. Most important is the economy. A drop in economic performance reduces revenues while simultaneously causing increased expenditures for welfare and unemployment programs. Important demographic variables include the number of school-age children, as education spending is one of the two largest expenditure commitments of states; the number of youth in the 16-24 high-crime cohort, which has a great impact on the fastest-rising state expenditure, corrections; and the number of elderly, which affects state budgets most directly through increased Medicaid expenditures funneled to nursing homes. Finally, federal policies also have an impact on state budgets, through increases or decreases in federal aid, or through the imposition of federal mandates, especially those that are unfunded. While Congress recently passed a law limiting the use of unfunded mandates, it was not made retroactive.
In sum, because of balanced-budget requirements, limits on revenue increases for fear of taxpayer flight, and built-in expenditure pressures based on inflation and increasing populations served, state budgeting is revenue-constrained. Surges in revenue will allow states to embark on additional spending or to pursue tax cuts. But in the absence of such revenue windfalls, state budgeting is for the most part highly routinized, with incremental decisions made by actors constrained by their environment.
Advisory Commission on Intergovernmental Relations. Significant Features of Fiscal Federalism 1991. Vol. 1. Washington, D.C., February 1991.
Council of State Governments. Book of the States. Lexington, Ky., 1996.
Gosling, James J. "Wisconsin Item-Veto Lessons." Public Administration Review 46 (July-August 1986).
Hutchison, Tony, and Kathy James. Legislative Budget Procedures in the 50 States: A Guide to Appropriations and Budget Processes. Denver: National Conference of State Legislatures, September 1988.
National Conference of State Legislatures. Fundamentals of Sound State Budgeting Practices. Denver, May 1995.
Rosenthal, Alan. Governors and Legislatures: Contending Powers. Washington, D.C.: Congressional Quarterly Press, 1990.
This report is available in its entirety in the Portable Document Format (PDF).