For some time economists have acknowledged that reported budgetary data do not necessarily reflect actual economic activity. Agreement has not been reached, however, on how budget figures should be adjusted to reflect such activity accurately. In this working paper, Resident Scholar Neil H. Buchanan examines 13 alternative measures of the budget deficit in order to determine which, if any, are theoretically or statistically sounder than existing measures. He evaluates them in terms of their theoretical value, that is, their ability to capture benefits (such as higher levels of employment (subject to NAIRU constraints), higher rates of growth, and higher levels of private investment), and costs (higher rates of inflation and lower levels of private investment, consumption, and net exports) to the economy. He also tests whether the new definitions provide empirically more robust estimates than existing measures. Different measures can provide somewhat different gauges of the total macroeconomic effects of deficit spending. For example, since spending by any level of government has macroeconomic effects, it can be argued that all spending, not only that of the federal government, should be included in the derivation of the "government deficit." Some assert that the so-called inflation tax (the effects of inflation on changes in interest rate payments) should be subtracted from the deficit since inflation results in a decrease in the real amount of debt outstanding. An argument also can be made that changes in spending and revenues resulting from a change in the business cycle should be smoothed over the course of the business cycle. Another adjustment would separate government expenditures for capital items from expenditures for consumption items because capital items yield long-term returns. In addition, however, government's long-term liabilities in the form of, for example, unfunded loan and pension guarantees, should be accounted for in the definition of a capital budget.
