关键词: 大气污染治理, 演化博弈, 稳定策略

Abstract: Air pollution is one of the most serious problems that endanger human survival, and air pollution prevention in our country at present also faces many difficulties. China's air pollution control mode is based on administrative division as the boundary of territorial governance. There is a contradiction between non-territorial and territorial governance of air pollution, so we need to carry out collaborative and cooperative governance, break the inter-provincial barriers of territorial governance, and form the central government as the core of the inter-provincial and multi-city air pollution cooperation governance alliance, in order to alleviate and solve the air pollution problem to the greatest extent.
In this paper, the evolutionary game method is adopted to analyze the stable strategies of air pollution control for multiple urban subjects in a certain inter-provincial region under four situations, namely, the spontaneous formation of cooperative governance alliance under unconstrained territorial governance, constrained territorial governance, and constrained territorial governance, and the cooperative governance alliance breaking territorial governance with the central government as the core. The city government with bounded rationality is taken as the decision-making subject, and an evolutionary game model of air pollution control with multiple game subjects is established. Each variable is represented by letters. Through the assumption of probability and derivation, using the income, cost and damage equalized under urban governance and non-governance as well as the rewards that the central government will give to each city, the expected income and average income under urban governance and non-governance of air pollution are calculated. According to the actual situation of different situations, the relationship between variables is determined, and the replication dynamic equation of urban governance of air pollution is constructed. We need to find the evolutionarily stable equilibrium point, which makes it reach the same probability of the city choosing any strategy in the region. Finally, under the different conditions of the four situations, the evolutionary stability strategy of all cities in the region is determined.
The results show that the high cost of controlling air pollution and the high income of not controlling air pollution make local governments choose to ignore the damage caused by air pollution. At the same time, there exists the phenomenon of “free riding”, which eventually leads to more and more serious air pollution in the region. In the case of constrained territorial governance, although provincial governments punish ungoverned cities, the punishment is not strong enough, and local governments often choose to bear the punishment in order to pursue economic development. At the same time, the punishment intensity of provincial governments is different. Cities in provinces with light punishment value choose to “free ride”, which leads to low governance efficiency. In the case of cooperative governance, alliance forms spontaneously under constrained territorial governance, and cities with high penalty value will absorb cities from provinces with low penalty value into the air pollution control alliance. Cities should enhance cooperation and promote innovation. However, it cannot solve the fundamental problem caused by the different penalty values in different provinces. With the central government as the core to form a cooperative governance alliance, the inter-provincial barriers can be broken, the restraint means of punishment and reward can be used, and the cost of air pollution governance can be reduced to promote air pollution governance. Therefore, the cooperative governance alliance with the central government as the core is the relatively optimal air pollution control scheme.Finally, in order to better study the specific situation of the evolutionary game of air pollution control under various circumstances, numerical simulation is carried out. The evolutionary game is simulated by assigning values to parameters, and sensitivity analysis is carried out in four cases by adjusting the size of parameters. The results of numerical simulation are analyzed and compared, and then the optimal scheme of air pollution control is worked out.
Therefore, this paper provides a certain reference for the design of air pollution control mechanism. In the case of territorial governance, the difference of fines between provinces is the root cause of “free riding” behavior and “high investment and low return”. It is relatively more scientific to break the territorial restriction, unify the punishment amount, and form the air pollution cooperation control alliance with the central government as the core. Under the leadership of the central government, relevant policies on air pollution control can be formulated more efficiently and mutual learning can be carried out, which can further reduce the cost of air pollution control.

Key words: air pollution control, evolutionary game, cooperation management