Equilibria in a stochastic resource management game under imperfect information

The paper considers equilibrium solutions in a stochastic resource management game. The countries choose frequently between cooperative and noncooperative (retaliation) modes of play. The choices between cooperative and noncooperative subgames are based on imperfect estimates of the fishing efforts (fleet sizes) of the co-harvesting country has been too intensive during the previous period, i.e., the estimated fishing effort exceeds a trigger value, then either a noncooperative retaliation period follows or cooperation is continued. After a noncooperative period the countries begin to cooperate once again. It is shown that equilibria in which cooperative and noncooperative periods alternate can occur in dynamic games. Different types of equilibria for the game exist; the countries may be able to sustain efficient cooperation all the time, they may oscillate between cooperation and noncooperation, or they may fail to cooperate at all. Conditions under which an equilibrium coincides with a Pareo optimal solution are studied.