Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper aims to analyze levels of corruption under different types of party list proprotional representation (PR) in a cross-national setting. Employing more robust statistical methods and Monte Carlo simulations, it demonstrates that Chang and Golden's (2007) hypothesis concerning the positive association of open-list PR with political corruption after controlling for district magnitude is not robust to an extended dataset including the mixed-member systems. Despite weak statistical evidence in the pooled dataset, this paper shows that closed-list PR combined with plurality/LR-Hare formula can induce more corrupt behavior in presidential systems with weak party organizations, such as South Korea and Ukraine.