Theses Doctoral

When Do Party Leaders Democratize? Analyzing Three Reforms of Voter Registration and Candidate Selection

Shoji, Kaori

Three independent studies drawing on the cases from different spaces and times comprise this research project, but they share a common theme: how do expansive reforms that open up paths to political participation take place? The first paper takes up the case of the motor voter reform, which allows people to register to vote at driver's license offices. The reform was widely legislated by U.S. states before the passage of the National Voter Registration Act in 1993. The paper investigates the factors that helped promote the reform at the state level by breaking down the reforms along two dimensions: the voter registration location and the implementation method. Motor voter legislation could either stand alone or be accompanied by agency-based registration (ABR), which includes registration at social service public agencies that primarily serve the poor. A reform could be implemented in an active or passive way. While ABR and active implementation had the potential to mobilize previously alienated socioeconomic groups, motor voter reform itself and passive implementation were expected to have a partisan-neutral and limited impact, respectively. Using data collected from the archived materials of the leading advocacy organization of the reform, Human SERVE, I test the following three general hypotheses statistically: 1) the Democratic Party is interested in mobilizing the poor, 2) electoral competition enhances mobilization efforts by parties, and 3) liberal political culture promotes inclusive electoral institutions. All three hypotheses find some support in the empirical analysis. The second paper focuses on a candidate selection method reform in contemporary Japan. Throughout the first decade of the twenty first century, the (then) opposition Democratic Party of Japan (DPJ) used kôbo, an open-recruitment candidate selection method, which was purported to open up the party nomination to non-traditional outsider aspirants. The DPJ's action presented a puzzle: searching for low-electability amateur candidates instead of traditional quality candidates seemed paradoxical for a party preparing to take over power. The paper reveals that using kôbo was a transitional strategy for a young party building itself under the mixed-member majoritarian system. I argue that recruiting "fresh faces" was not what really motivated the use of kôbo, by showing how kôbo increasingly produced insider candidates over time. The third paper investigates the development of direct primary in nineteenth century Pennsylvania. The historical origins of the U.S. primaries have mostly been discussed in terms of statewide legislations around the Progressive Era, which made the primaries mandatory for the two major parties. This paper focuses instead on the voluntary adoption phase that took place under the party by-laws, paying special attention to the case of Pennsylvania after 1842. I argue that the party elites of county organizations initiated the introduction of the primaries in order to prevent defection and to preserve party unity. As the vote share of a party increased, the party nomination became more valuable, and more people competed for nomination. More disgruntled nomination losers would run as independents, hurting the electoral prospects of a given party in the general election. For party leaders, whose overwhelming concern was the maintenance of party unity, the direct primary system offered a solution by presenting the primary winner as a focal candidate to the party voters. The primaries made it harder for losers to defect later, with the transparent features of their procedures. Thus, the stronger the party, the more likely it was to adopt the direct primary. The paper tests this hypothesis empirically with an original data set built from hundreds of archived local newspapers. To my knowledge, this is the first study on nineteenth century county-level party activities to use comprehensive data covering most counties from a single state. The findings have broader implications as to how party competition affects the choice of candidate selection methods, and the role which competition among elites plays in the democratization of the intraparty decision-making mechanism.

Geographic Areas


  • thumnail for Shoji_columbia_0054D_11442.pdf Shoji_columbia_0054D_11442.pdf application/pdf 3.04 MB Download File

More About This Work

Academic Units
Political Science
Thesis Advisors
Shapiro, Robert Yale
Ph.D., Columbia University
Published Here
May 30, 2013