Revisiting the Link Between Electoral Competition and Policy Extremism in the U.S. Congress

Abstract

Prominent theories of American political parties imply that higher levels of competition cause lawmakers to be more responsive to the center of public opinion, but there is little empirical evidence to support this assertion. Furthermore, many studies have found that competition causes lawmakers to be less responsive to public opinion as a whole and more responsive to their own partisans. Parties and candidates pursue a "mobilization" rather than "moderation" strategy supposedly because competitive constituencies are inherently heterogeneous, which then creates a context for more ideological campaigns and a more partisan policymaking environment. An analysis of U.S. Senate roll call votes between 1989 and 2000 reveals that legislators from more marginal states are less responsive to the ideological center than legislators from safe states. Furthermore, there is no evidence of a relationship between constituency diversity and electoral competition or that greater partisanship is the result of greater diversity.

Incumbent positioning, ideological heterogeneity and mobilization in U.S. House elections

“…Thus, we should expect divergence to be greatest when seats are competitive. Gulati (2004) has found support for this proposition for incumbent senators but others have challenged this finding for candidates for the House of Representatives (Ansolabehere et al 2001;Burden 2004;Griffin 2006). These scholars have empirically documented that candidates become more moderate as partisan competition increases.…”

Are African Americans Effectively Represented in Congress?

“…To measure each MC's race, we use an indicator coded 0 for non-African American MCs and 1 for African American MCs. 11 Following a number of studies (Sullivan and Uslaner 1978;Ansolabehere, Snyder, and Stewart 2001;Gulati 2004;Griffin 2006), we measure district competitiveness for the 101st, 103rd, and 105th Congresses using the district's normalized two-party presidential vote share in the 1988, 1992, and 1996 elections, centered on the yearly mean and reflected so that higher values indicate more competitive districts. 12 We impute district competitiveness measures for the 102nd, 104th, and 106th Congresses using the 1988, 1992, and 1996 election data, respectively.…”

“…This measure of district competitiveness was preferred to congressional election outcomes for two reasons. First, election outcomes will be affected by how constituents were represented in the previous term and so should not be used to model representation in the current term (Gulati 2004;Griffin 2006). Second, the usefulness of congressional election margins to predict an incumbent's electoral security has declined substantially over time (Jacobson 1987).…”

Explaining legislators’ positions on outside income limits: Voting on honoraria ceilings in the U.S. Senate, 1981–1983

“…More specifically, marginal legislators will be less likely than other members to support proposals that their constituents oppose, while "safer" legislators will be less constrained to support unpopular proposals (Ansolabehere et al 2001;Bartlett 1979;Coates and Munger 1995;Kuklinski 1978;Griffin 2006;McRae 1952; though see Gulati 2004). Legislative pay raises are highly unpopular with voters (Bianco et al 1996;Clark 1996;Theriault 2005).…”