A Game-Theoretic Analysis of China's Village Elections.

From the perspective of good governance, a competitive election is considered better than a non-competitive election. Competition will bring uncertainty of an election result. Each candidate, who assumes that they will have a chance to win, will outline their policy positions in the campaign to sway voters and win the election. Thus, candidates will be held accountable to voters for their policy positions. In other words, competition of an election will give each candidate an incentive to campaign and candidates will be held accountable to the electorate through campaigning. Throughout my field research in China's rural areas, I found the roles and functions of elections were very different from what I imagined. My game-theoretic analysis explores the effects of competitive elections on their outcomes. This is a significant question because previous studies tend to assume that competitive elections have a positive effect on elected cadres' accountability; in short, they assume that competitive elections are better than non-competitive ones. I challenge this assumption. My attempt is supported by many of the empirical findings from China's village elections, which suggest (1) competitive elections are often corrupt; (2) vote-buying candidates are often elected by competitive and corrupt elections; and (3) candidates elected by non-competitive elections are often not corrupt and provide public goods. My comparative-static analysis shows logic of how, given the institutional setting of village elections and unique characteristics of the village community, a vote-buying candidate is more likely to win if an election is competitive, and a good cadre is elected by a non-competitive election. The size of a village, the unit of elections, is very small and the laws regulating vote-buying are poorly enforced through they formally stipulate severe punishment. Therefore, incorporating these conditions, the model's comparative-static analysis implies that competitive elections tend to be corrupt and non-corrupt elections tend to be non-competitive. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]