Abstract: Air pollution has significant cross-regional characteristics, and cooperation between local governments is the only way to control air pollution. Under the situation of cooperation, a social dilemma arises , which is free-riding in controlling air pollution. The game of air pollution treatment between local governments is long-term and dynamic, and the evolutionary game model of multi-player two strategy is established. The evolutionary stable strategy of local governments without restriction mechanism and under restriction mechanism and the related influencing factors are studied respectively. Considering the pollution industries transferred between governments, this paper establishes the evolutionary game model considering pollution compensation mechanism that government shall compensate for the receiving party for the transfer of polluting industries ,and studies the condition of optimal compensation-governance strategy turning into evolutionary stable equilibrium under unconstrained and constrained mechanism, separately. The researches show that cooperative governance strategy will not be affected by the externality between local governments. The cooperative governance strategy will be significantly affected by the cost of implementing regulation, the emission reduction during implemented regulation and the emission increase during unimplemented regulation, the rewards of central government to implemented regulation government and the punishment of central government to unimplemented regulation government. It can effectively promote the evolution of local government strategies to participate in cooperative governance by increasing the weight coefficient of environmental quality index in achievement assessment system. The optimal compensation-governance strategy can't arrive at ESS under local government independent choice. Under the supervision of central government, the optimal compensation-governance strategy can arrive at ESS when economic punishment is within a certain range.

Key words: cooperative governance, evolutionary game, local government, evolutionary stable strategy

摘要： 大气污染具有跨区域性,府际间合作治理是大气污染防治的必由之路,合作治理中存在“搭治污便车”的困境。由于府际间大气污染治理博弈的具有长期性和动态性,由此建立了多人两策略的演化博弈模型。分别研究,无约束机制和有约束机制两种情况下地方政府演化稳定策略及其影响因素。考虑到政府间污染产业转移,建立了污染产业转移政府对承接方政府治污补偿机制下的演化博弈模型,分别研究无约束和有约束两种情形下(补偿,治理)策略成为演化稳定均衡状态的条件。研究表明:府际间的合作治理策略选择不会受到外部效应的影响,大气污染治理成本、治理时污染物减排量、不治理时污染物排放增加量、中央政府对治理的奖励和对不治理的惩罚力度是影响地方政府决策的重要因素。将大气质量作为政绩考核的重要指标,提高其权重系数,可有效促使地方政府策略向参与合作治理演化。无约束下自主选择时,最优策略(补偿,治理)不会成为稳定均衡策略,引入中央政府约束后,当经济惩罚在一定范围内,(补偿,治理)成为稳定均衡策略。

关键词: 合作治理, 演化博弈, 地方政府, 演化稳定策略