In 2011, the Pew Center on the States published a study calling attention to the increasing inaccuracy of states’ revenue estimates. Looking at data from 1987 through 2009, one thing Pew shows seems obvious enough: During recessions, states tend to overestimate revenues and during periods of economic growth, states tend to underestimate revenues.
Revenue forecasts matter for planning and policy purposes. When revenues are underestimated, surpluses encourage government growth and swings in tax policy that might be ill-conceived. When revenues are overestimated, governments typically resort to financially damaging gimmicks, debt, and raids on reserves, just delaying the politically disruptive controversy that comes with prioritizing less spending. By not confronting reality, states can be led to impose otherwise unnecessary and ill-conceived tax increases.
Revenue forecasts could have been more accurate in many states during the last recession, as demonstrated by some states where forecasts were relatively accurate. Part of the problem is politics. There is a bias toward higher spending, and higher revenue forecasts allow budgets to be higher as well. During the last recession there was a systematic tendency on the part of revenue estimators to assume revenues would rise from one year to the next, even in the face of evidence that the recession would be long and deep. But unrealistically high revenue estimates, usually politically influenced, allowed for higher budgeted spending than what could have prevailed otherwise. This means institutional structures need to be put in place to counteract political incentives.
As discussed in this paper, rigorous statistical evidence indicates that wise budgeting and estimating policies will lead to more accurate revenue forecasts. The analysis relies on two statistical regression analyses of National Association of State Business Officials data regarding revenue estimate errors and the institutional structures that surround budgeting and revenue estimates across the states. In addition, academic studies of the subject were considered and found to generally agree with the results of the analysis conducted for this paper. Academic studies pointed to a few other institutional areas for reform as well.
Suggested institutional reforms that would likely result in more accurate revenue estimates and more realistic budgeting on a real-time basis include: 1) spending limits, 2) the adoption of Generally Accepted Accounting Practices in budget documents, 3) multiple revenue estimates to include at least one prepared by an official independent of the budgeting process, 4) an independent system for aggregating the independent estimates into a single estimate with the budget bound to the estimate, 5) increased frequency of publicly released estimates, 6) budgeting practices that strictly prioritize programs, and 7) a requirement that state spending be balanced to actual revenues.