讲座题目：Why Play Tough? Divide-and-Rule and Commitment Problem （分而治之与承诺问题）
陈学政，现任四川大学经济学院副研究员，2013年毕业于英国华威大学（University of Warwick），获得经济学博士学位，研究领域为政治经济学、实验与行为经济学。陈学政博士已在《经济研究》等期刊发表论文多篇。
Acemoglu, Robinson and Verdier (2004) develop a theoretical model to explain how the strategy of divide-and-rule is implemented by a kleptocratic ruler maximize her personal payoff. Their model is built on a crucial assumption that the ruler is committed to punish whichever group who initiates the challenge, and it is credible to both groups. In this paper, we introduce a model of divide-and-rule without this assumption, and examine how this will affect the implementation of the divide-and-rule strategy. We elaborate the analysis on the subgame optimal strategy for the ruler after her power is challenged off the equilibrium path, and find that it is not necessarily subgame optimal for the ruler to punish the group who initiates the challenge. The introduction of the above assumption may change the optimal strategy for the ruler along the equilibrium path, entitling the ruler more flexibility in implementing the strategy of divide-and-rule that brings her greatest payoff. The analytical results in this paper help explain why the rulers in real world may have the incentive to build a reputation of being a strong ruler by playing tough on the opposition groups.