
SUMMARY
We began the chapter by introducing the basic concepts and tools of game theory. We then used these tools to explore oligopoly when the number of firms is fixed. From the Cournot and Bertrand models, we learned one very basic lesson about oligopolistic interactions: there is a fundamental contradiction between what is privately rational (or profitmaximizing) and what is collectively rational. In any simple Nash equilibrium, firms have a clear incentive to collude. Yet in exploring the collusive model of oligopoly, we discovered that each party to a collusive agreement also has a private incentive to cheat by producing more than the agreedupon output or by charging less than the agreedupon price. The prisoners’ dilemma is the game that illustrates these conflicting forces most clearly. This contradiction arises in Cournot’s model of oligopoly, where firms independently choose quantities, and in Bertrand’s model, where firms independently choose prices. Evidence from experimental economics and from the history of real collusive agreements supports the view that oligopolists do recognize and act on these conflicting incentives at different times.
We saw, too, that Cournot’s model responds in an appealing way to changes in the number of firms. When there is only one firm, Cournot’s model is identical to the standard monopoly model. As the number of firms increases, aggregate output increases, and aggregate profit and price decrease. As the number of firms approaches infinity, Cournot’s model converges to the competitive equilibrium. Thus, Cournot’s model captures a wide range of market structures, from monopoly to perfect competition. Bertrand’s model has none of these appealing properties. In that model, price is equal to marginal cost whenever there are two or more firms.
In the second part of the chapter, we arrived at several additional insights as we explored the theory of oligopoly in the long run, when firms are allowed to enter a market. We began with the limitoutput model, in which established firms produce enough output (the limit output) that an entrant who takes the output of established firms as given will not enter. As we saw, the weakness of this model lies in the question of credibility: How believable is the threat by established firms to actually produce the limit output after entry occurs? In answering this question, we produced a theory of market structure based on Cournot’s model.
We closed the chapter by summing up the insights to be drawn from our foray into the fascinating but complicated world of oligopolistic decision making. It is fascinating because it is at heart a game of strategic manoeuvring: a game of positioning and reacting. Yet it is precisely the business of anticipating and responding to strategic moves of rivals (present and potential) that makes oligopoly so difficult to model.


 