Chapter Summary

The characteristic feature of oligopolistic markets is interdependence among firms. In the Cournot model, each firm takes the quantities produced by its rivals as given; in the Bertrand model, in contrast, each firm takes its rivals' prices as given. Although the behavioural orientation of firms sounds very much the same in these two cases, the results are strikingly different. The Cournot model yields a slightly lower price and a slightly higher quantity than we would see if the firms colluded to achieve the monopoly outcome. In contrast, the Bertrand model leads to essentially the same outcome we saw under perfect competition.

A slightly more sophisticated form of interdependence among firms is assumed in the Stackelberg model, in which one firm plays a leadership role and its rivals merely follow. This model is similar in structure to the Cournot model, except that where the Cournot firms took one another's quantities as given, the Stackelberg leader strategically manipulated the quantity decisions of its rivals.

The interdependences among oligopolistic firms are often successfully analyzed using the mathematical theory of games. The four basic elements of any game are the players, the set of possible strategies, the payoff matrix, and the decision rule. A Nash equilibrium occurs when each player's strategy is optimal given the other player's choice of strategy. A strategy is called dominant if it is optimal no matter what strategy the other player chooses.

The incentives facing firms who attempt to collude are similar to the ones facing participants in the prisoner's dilemma. The difficulty in holding cartels together is that the dominant strategy for each member is to cheat on the agreement. Repeated interactions between a very small number of firms can support collusive behaviour under circumstances in which strategies like tit-for-tat are effective.

Incumbent firms may sometimes act strategically to deter potential rivals from entering their markets. Often this involves incurring higher costs than would otherwise be necessary.

The basic idea of the theory of contestable markets is that when the cost of entry and exit is very low, the mere threat of entry can be sufficient to produce an allocation similar to the one we see under perfect competition. Critics of this theory have stressed that there are almost always nontrivial sunk costs associated with entry and exit, and that even small sunk costs leave considerable room for strategic entry deterrence.

Monopolistic competition is defined by two simple conditions: (1) the existence of numerous firms each producing a product that is a close, but imperfect, substitute for the products of other firms; and (2) free entry and exit of firms. In the spatial model of monopolistic competition, customers have particular locations or product characteristics they most prefer. The result is that firms tend to compete most intensively for the business of products most similar to their own.

A central feature of the spatial model of monopolistic competition is the tradeoff between the desire for lower cost, on the one hand, and greater variety or locational convenience, on the other. The optimum degree of product diversity depends on several factors. Greater diversity is expected with greater population density and higher transportation costs (where, in the general case, "transportation costs" measure willingness to pay for desired product features). Optimal product diversity is negatively related to the start-up costs of adding new product characteristics or locations. The market metes out a certain rough justice in that the costs of additional variety tend to be borne most heavily by those to whom variety is most important.

The appendix to the chapter discusses entry and entry deterrence in more detail, and addresses the question of who pays for variety.