Genetic algorithms and stochastic forest property rights enforcement

This paper analyses the impact of corruption in a timber industry by comparing stochastic and deterministic approximations of a 'log poaching-enforcement' difference game. In the paper it is argued that corruption leads to certainty about the effectiveness of enforcement efforts, a situation best represented by a deterministic game. By adding uncertainty, therefore, some light is shed on the impact of corruption. The model is hard to solve analytically, so results are demonstrated by developing heuristic genetic algorithms. The results show that increasing uncertainty induces a government to spend more on enforcement and an illegal logger to harvest less.