Abstract:
This paper shows that, if the government smooths taxes, then the budget surplus should equal the present discounted value of expected changes in government expenditure. This implication of the tax-smoothing hypothesis imposes more stringent restrictions on the data than the more usual method of testing whether changes in tax rates follow a random walk. The tax-smoothing model is applied to the federal government budgets of Canada and the United States and, in each case, receives considerable empirical support. Copyright 1995 by Ohio State University Press.