INTRODUCTION

The model of the American political economy described in chapter 8 has several empirical implications. These are tested in this chapter using United States data for the period 1915 to 1988. We examine the behavior of the annual growth rate of the gross national product and of vote shares in presidential elections and in elections for the House of Representatives.

Throughout this volume we have stressed the interdependence of politics and economics. A basic theme is that partisan politics introduces uncertainty in both the polity and the economy; both the voters and the economic agents need to hedge. Specifically, uncertainty in presidential elections leads economic agents to hedge by expecting an inflation rate equal to an average of the inflation policies that would be followed by the two parties. Similarly, voters adjust their on-year congressional votes so that there is not a decisive tilt towards either party. At midterm, the identity of the party of the president for the next two years is known; thus the voters can complete their balancing of the president.

In other words, expectations about future policies tie the political and economic systems together. Therefore we undertake a simultaneous estimation of four equations: one for GNP growth; one for the vote share in presidential elections; one for House vote shares in on-years; and another for House vote shares in midterms.

Our most general formulation of the GNP growth equation incorporates the two basic determinants of growth highlighted in the previous two chapters: the partisan macroeconomic policies of the two parties and the competence of different administrations.