Experimenting With the Balancing Hypothesis

Abstract

The reasons why divided government is on the rise in the United States remain unclear. Of the explanations offered, Fiorina's (1992) balancing hypothesis - the idea that voters intentionally cast their ballots in a way that would increase the prospects of split party control - has drawn the most attention. This study gathered empirical evidence to test the hypothesis; its focus was not on whether citizens want divided government, but rather on whether they collectively act in a way consistent with balancing. In September 2000, during the national election campaign, a sample of undergraduates responded to one of five versions of a newspaper article (similar to actual articles about the campaign) that varied with respect to reported polling data on the competitiveness of the congressional and presidential races. The results cast doubt on the merits of the balancing hypothesis.