Federal Trade Commission

The ability of incumbent firms to maintain their market power by deterring entry has been a topic of considerable interest for some time. With the advent of game theory, economists have recognized that entry deterrence strategies must be credible. Imposing this credibility requirement on an incumbent's strategy in a price or quantity setting game with the entrant has had a profound effect on the analysis of several types of entrydeterrence strategies, such as limit pricing (e.g., Milgrom and Roberts 1982) or capacityexpansion (e.g., Dixit 1980). Typically, however, these analyses have not considered thecredibility of the incumbent's decision to remain in the market or to exit. They have assumed parameter values such that exit is never optimal (the no exit threat is always credible) or that the incumbent can commit not to exit the market. In industries with large fixed, but not sunk, costs, however, exit can be the profit maximizing strategy when faced with an entrant who may be much more efficient or have a far superior product. American Airlines, for example, decided to abandon its San Jose hub and many routeswithin California shortly after Southwest Airlines entered the San Jose market (San Jose Mercury News 1993; Washington Post 1993). In fact, as the above quote indicates, the credibility of the no exit threat could become increasingly important as rapidly advancing technology provides opportunities for entrants to topple dominant firms in many industries. Thus, it is more important than ever to analyze the effect of the exit option on the ability of incumbents to deter entry.

Agents play an important role in many search markets. In the real estate market, for example, agents improve the efficiency of the search process by helping buyers and sellers locate trading partners, identify situations where there are gains from trade, and take care of the technical details of a transaction. Agents may also have an important effect on the flow of information between buyers and sellers, depending on whom the agents work for and the nature of their legal responsibilities. For instance, if an agent helps a buyer to find trading partners, it is likely that he will learn a great deal about the buyer's preferences. Unless the agent is bound by a duty of confidentiality, he may reveal what he knows to sellers, who would presumably benefit from access to this information, since they could use it when formulating their bargaining strategy. It may be in the agent's interest to strengthen the seller's hand if, as is common in the real estatemarket, his compensation is proportional to the sales price that the buyer and seller negotiate. In this book we investigate how the role that agents play in a search market affects welfare. Our principal question is whether agents should be able to transmit information about buyers'willingness-to-pay to sellers, assuming that society's goal is to maximize the discounted expected gains from trade that the market generates. To this end, we analyze the equilibrium that arises in a search market under alternative assumptions about the agents' role. For concreteness, we develop our results using a model of the real estate market, but our conclusions are applicable to any search market in which sellers wish to sell and buyers wish to buy one unit of an indivisible good, buyers are privately informed about their valuations, and sellers' valuations are commonly known