and Tennessee allowed a more-limited tax on dividends and interest in lieu of a comprehensive income tax.
Individual income taxes have become an important source of revenue for most states that have adopted them, receiving more revenue from this source than any other. The combined tax collections for the states from individual income taxes, however, remain significantly smaller than those collected by the federal government from this source, predominantly because the tax rates applied by states are typically much lower. The structure of many state income taxes parallels and mirrors that of the federal income tax. Many states allow taxpayers to copy information directly from federal returns when computing state income tax liability. About 3,500 cities also levy individual income taxes, but in most cases, they limit coverage to payroll income, not income derived from all sources. Many of these cities are in Pennsylvania; about 900 cities are outside of Pennsylvania.
Forty-four states and the District of Columbia use a corporate income tax. Exceptions are Nevada, South Dakota, Texas, Washington, and Wyoming. Michigan applies a single business tax, which is a modified value-added tax. Complicating the application of state corporate income taxes is the operation of many corporations in more than one state, and sometimes more than one country. Though some income can be identified as originating in a particular state and therefore legitimately subject to its corporate income tax, the origin of other income is sometimes more difficult to determine. To facilitate the calculation of corporate income tax due it, each state with a corporate income tax has adopted its own income apportionment formula, to determine how much of the total profit earned by the multistate corporation it will tax. The most common approach is the single-weighted three-factor formula, with state shares of total corporate property, payroll, and sales used as approximations of corporate activity in the state. Some states do not use the formula, however, and weight the factors differently. Most states lack sufficient auditors to verify corporate tax returns and must accept the calculations done by corporations. Consequently, state corporate income tax avoidance is a common problem, and a local corporate income tax is totally infeasible.
MARCIA LYNN WHICKER
Conlan, Timothy J., Margaret T. Wrightson, and David R. Beam . 1990. Taxing Choices: The Politics of Tax Reform. Washington, DC: Brookings Institution.
Mikesell, John L. 1991. Fiscal Administration. Belmont, CA: Wadsworth.
Pechman, Joseph A. 1987. Federal Tax Policy. 5th ed. Washington, DC: Brookings Institution.
Pechman, Joseph A., ed. 1988. World Tax Reform: A Progress Report. Washington, DC: Brookings Institution.
Samuelson, Paul A., and William D. Nordhaus. 1989. Economics. 13th ed. New York: McGraw-Hill.
Stein, Herbert, ed., 1988. Tax Policy in the Twenty-First Century. New York: John Wiley and Sons.
Stiglitz, Joseph E. 1986. Economics of the Public Sector. New York: W. W. Norton and Co.
INCREMENTAL BUDGETING. The routines of budgeting that are marked by predictability and stability. Incremental budgeting is a description of both the outcomes of year-to-year changes in budgets and the method used to make the budget decisions. Year-to-year decisions, interpreted as outcomes of a budget process, are measured as percentage changes from the previous year. Therefore, the increment is the budget in year t minus the budget in year t -- 1, divided by the budget in year t -- 1. This provides the percentage change from one year to the next; the percentage change, when displayed graphically over several years, will show a pattern of stable growth. Increments are expected to be modest from year to year. The modest changes reflect new activities, new programs, and increases to existing programs.
These statistical properties of budgets that demonstrate incrementalism as an outcome were summarized over three decades ago by Aaron Wildavsky as follows, "The largest determining factor of the size and content of this year's budget is last year's budget" ( Wildavsky 1984, p. 13).
The specification of incremental budgeting as modest percentage changes from last year's budget is less obvious than it seems at first. One of the first questions to ask about increments is: What percentage increase qualifies as an increment? Is a 5 percent increase an increment? What about a 10 percent increase? Could a decrease in a budget from one year to the next be called an increment? Scholars debated this very point without reaching consensus because there is simply no objective way to define what constitutes a "modest" percentage increase over last year's budget.
Perhaps the best answer to these questions is that the size of the budgetary increment is not the important issue. Instead, it is the stability of the year-to-year changes that defines an incremental budget process. Using this definition, year-to-year changes in the one percent to three percent range may be incremental; so may year-to-year changes falling into a 5-to-9 percent range. Notice that it is not the size of the increment that is important but the predictability of the changes that defines a budget process as incremental.
The statistical description of incrementalism described previously has additional ambiguity. Stable budgetary patterns at one level of organization may mask considerable budgetary variation at a different level of organization. For example, a department with four bureaus may show budgetary stability when the budget for the entire department