This dissertation features party cohesion as a factor influencing how voters hold ruling elites accountable. It consists of three essays that dissect the impacts of party cohesion on democratic accountability in the United States. In a nutshell, this dissertation finds that when party cohesion increases, the focus of politics is shifted toward partisan rivalries, and hence the ruling elites may not be held accountable to the public as they pursue partisan goals. The first chapter is an essay titled Income Inequality and Electoral Costs of Party Loyalty in the United States, 1976-2008. The United States features a political system where party loyalty has been traditionally seen as electorally costly to individual legislators. However, if being loyal to the party is costly, why did legislators in the U.S. increase their party loyalty to a high level during the past decades? I argue that the electoral cost of party loyalty could be reduced, and I suggest that such a reduction of electoral cost was resulted from the increases in income inequality. The evidence from the U.S. House elections between 1976 and 2008 indicates that party loyalty becomes less costly when income inequality increases. This finding may not only complement the existing theories of party government, but also shatter the conventional wisdom on the role of the party in personal-vote systems. The second essay is titled Clarity of Responsibility and Clarity of Party Line: The Impacts of Party Cohesion on Economic Voting in the United States, 1980-2008. Powell and Whitten (1993) maintain that governing party cohesion is a stable feature of political systems that enhances clarity of responsibility and economic voting. However, party cohesion is not stable, and there are ebbs and flows in party cohesion. Further, research finds that at the micro level, voters respond to partisan cues and become more partisan when the elites from each party stand with their in-party members and that economic conditions matter less when there are more partisan voters. With data from the American National Election Studies (ANES) between 1980 and 2008, this essay demonstrates that party cohesion is a double-edged sword. As party cohesion increases from a low to mid level, it may strengthen economic voting by enhancing clarity of responsibility. When it is at a high level and party line becomes clear, it may hinder economic voting. Finally, the essay, Resisting Being Held Accountable: The Impact of Party Cohesion on Concentration of Votes in the United States, 1992-2012, is the third chapter. Party cohesion has been seen as a constraint on particularistic exchanges between politicians and voters. However, using the concentration of votes as an indicator of particularistic exchanges between parties and voters, I find that party cohesion has a positive impact on the concentration of votes in presidential elections between 1992 and 2012. This finding suggests that a smaller range of the electorate is served by the incumbent party when the party is more cohesive. Further, I demonstrate that the concentration of votes received by an incumbent party positively affects the vote share of the party. Hence, a higher level of party cohesion does not improve accountability. Instead, it contributes to the ability of a party to resist being held accountable to the interest of the public. The three essays show how accountability may vary with the changes in party cohesion. The evidence presented suggests that party cohesion may not improve accountability as what political scientists have wished for. Rather, high levels of party cohesion may hinder voters from holding ruling elites accountable for the public interest.
Read
