The partisan theory of economic cycles, fiscal policy and the congressional election

We develop and test a rational expectation model in which voters are heterogeneous and have endogenous preferences to examine the effect of electoral competition on macroeconomic policy, income distribution and voting behavior. In particular, we demonstrate how partisan monetary policy influences the choice of fiscal policy via economic cycles that affect the profile of pre-tax and transfer income distribution. Politics and the distribution of income are closely interrelated. Politics has affected economic growth and income distribution in the United States mainly through unanticipated monetary policy shifts following the outcome of presidential elections and through the flow of resources to transfer programs. At the same time, expected economic growth simultaneously affects voters\u27 tax preferences since it influences their relative valuation of expected costs and benefits of the fiscal program. When the cost of economic cycles fall mainly on the poor, we find that ex-post tax preferences of the middle class vary positively with the rate of economic growth. Under the institutional structural of Alesina and Rosenthal (1989a,b), our model predicts that the two-party vote share in congress is influenced by two factors; namely, moderating behavior and a shift in preferences effect. The former is the same as in Alesina and Rosenthal (1989a,b) while the latter results from an anticipated change in expected output growth in the upcoming congressional term. Our empirical study is divided into two parts. First, we find evidence supporting the partisan theory of monetary policy that the degree of policy surprise following the outcome of the presidential election helps to explain the degree of variation of output growth from trend in the second year of each administration. Second, test results also strongly support our version of the congressional election hypothesis