Abstract: The decision-making in key areas of multinational enterprises has become a thorny bottleneck distressing many countries. Based on the improved competition function and enterprise dependence (indirect rate of return), this paper constructs a sequential game model for multinational enterprises’ decision-making in key areas from the perspective of co-opetition. Four Nash equilibrium paths under the two scenarios of transnational enterprises as “forerunner” and “destroyer” are obtained by using the reverse induction method. We analyze the impacts of host government’s “stranglehold”, “legal war” and “list-making” strategies. Finally, the influences of decision execution on the equilibrium paths are given by numerical analysis. We find that, the game action order only changes the decision threshold of an indirect rate of the return on the four equilibrium paths; the decision threshold of indirect rate of the return has an inverse U relationship with the relative technological competitive advantage in key areas of host enterprises; when facing counterattack response, “stranglehold” strategy could effectively reduce the probability of multinational enterprises attacking highly dependent host enterprises. But when facing “non-counterattack” response, “legal war” strategy plays the most significant role; when facing “legal war” and “list-making” responses, host enterprises are more tolerant of the implementation of the “attack” strategy of highly dependent multinational enterprises.

Key words: enterprise decision, sequential game, co-opetition, technological competition, government intervention