Revenue-Stabilizing Tax Rates over the Business Cycle: Implications for States
Enomoto, Carl E., Erickson, Christopher A., Ghosh, Soumendra N., Quarterly Journal of Business and Economics
The timing of tax collections over the business cycle is an important issue for policy makers at the state level because state authorities often face political and legal constraints that limit fiscal flexibility. For example, 39 states have constitutions that include some kind of balanced budget provision. Although tax collections not always are required to cover all expenditures, these provisions often limit deficit financing to capital projects that are preapproved by popular vote. Even when state constitutions do not proscribe deficits, political circumstances often do. Large budget surpluses are also politically undesirable. Large surpluses imply tax rates and associated deadweight losses that are greater than necessary. Political opponents can point to surpluses as evidence of poor management.
Political and constitutional balanced budget requirements can limit the ability of states to finance government operations over the business cycle. Because sales tax revenues are generated by taxing economic activity, they tend to be pro-cyclical. Even small declines in tax revenues during periods of slow or negative growth can force state governments to curtail or eliminate programs, disrupting service flows. Unusually large levels of collections during business upturns may be difficult to justify when tax rates could have been reduced. In fact, variable cash flows are just as much a problem for governments as they are for firms. A reduction in state government cash flow variability can be viewed as one way of reducing risk.
Despite the importance of business cycles to state government finances, only limited work has been done in this area.(1) The purpose of this paper is to help fill this gap in the literature. A model of private and government sector behavior is formulated and used to derive a time path for tax rates that minimizes variation in state revenue about a predetermined level. Although the methodology is general, this research concentrates on gross receipts (sales) tax rates because gross receipts typically are the most important source of government revenue at the state level.
Changing tax rates each year to balance the budget may be impractical. Tax rate changes often are associated with intense lobbying efforts. Constituent pressure on legislators, especially not to raise taxes, may be intense. It is unlikely that legislators are willing to enact rate changes frequently. Therefore, the optimal tax path is derived not only for the case when the budget is balanced annually, but also for when the budget is balanced over two, three, four, and five years. The five year case is particularly interesting because this corresponds to balancing the budget over a typical business cycle.
Although the main purpose of this paper is to derive a revenue-stabilzing tax time path, the results also shed light on the effects of the business cycle on state economies. In particular, the model can be applied to any state to show how changes in Gross National Product (GNP) and tax rates affect personal income and taxable gross receipts. It is found that balancing the budget annually increases the variability of state personal income, causing the state to suffer a more severe business cycle.
The concept of optimal tax rates has been discussed widely in the literature. (See, for example, Diamond and Mirrlees, 1977a, 1977b; Mirrlees, 1976; Sandmo, 1976; Sadka, 1977; Garfinkel, Moreland, and Sadka, 1986.) The methodology used typically has been to choose taxes to maximize a social welfare function subject to a production and government revenue constraint. Such an approach is subject to criticism to the extent that a social welfare function is not well defined. Apart from these theoretical considerations, optimal tax theory has been difficult to apply because data necessary about preferences are not available.
Another trend in the tax policy literature has been to use simulation models to determine the effect of alternate tax policies (Salathe, et al. …