Stephen Morris Publications

We study price discrimination in a market in which two firms engage in Bertrand competition. Some consumers are contested by both firms, and other consumers are “captive” to one of the firms. The market can be divided into segments, which have different relative shares of captive and contested consumers. It is shown that the revenue-maximizing segmentation involves dividing the market into “nested” markets, where exactly one firm may have captive consumers.

Consider a market with identical firms offering a homogeneous good. A consumer obtains price quotes from a subset of firms and buys from the firm offering the lowest price. The “price count” is the number of firms from which the consumer obtains a quote. For any given ex ante distribution of the price count, we obtain a tight upper bound (under first-order stochastic dominance) on the equilibrium distribution of sale prices. The bound holds across all models of firms’ common-prior higher-order beliefs about the price count, including the extreme cases of full information ( firms know the price count) and no information (firms only know the ex-ante distribution of the price count). A qualitative implication of our results is that a small ex ante probability that the price count is one can lead to a large increase in the expected price. The bound also applies in a wide class of models where the price count distribution is endogenized, including models of simultaneous and sequential consumer search.

Consider a market with many identical firms offering a homogeneous good. A consumer obtains price quotes from a subset of firms and buys from the firm offering the lowest price. The “price count” is the number of firms from which the consumer obtains a quote. For any given ex ante distribution of the price count, we obtain a tight upper bound (under first-order stochastic dominance) on the equilibrium distribution of sale prices. The bound holds across all models of firms’ common-prior higher-order beliefs about the price count, including the extreme cases of complete information ( firms know the price count exactly) and no information ( firms only know the ex ante distribution of the price count). A qualitative implication of our results is that even a small ex ante probability that the price count is one can lead to dramatic increases in the expected price. The bound also applies in a wide class of models where the price count distribution is endogenized, including models of simultaneous and sequential consumer search.

Consider a market with identical firms offering a homogeneous good. A consumer obtains price quotes from a subset of firms and buys from the firm offering the lowest price. The “price count” is the number of firms from which the consumer obtains a quote. For any given ex ante distribution of the price count, we derive a tight upper bound (under first-order stochastic dominance) on the equilibrium distribution of sales prices. The bound holds across all models of firms’ common-prior higher-order beliefs about the price count, including the extreme cases of full information (firms know the price count) and no information (firms only know the ex ante distribution of the price count). A qualitative implication of our results is that a small ex ante probability that the price count is equal to one can lead to a large increase in the expected price. The bound also applies in a large class of models where the price count distribution is endogenously determined, including models of simultaneous and sequential consumer search.

We consider demand function competition with a finite number of agents and private information. We show that any degree of market power can arise in the unique equilibrium under an information structure that is arbitrarily close to complete information. In particular, regardless of the number of agents and the correlation of payoff shocks, market power may be arbitrarily close to zero (so we obtain the competitive outcome) or arbitrarily large (so there is no trade in equilibrium). By contrast, price volatility is always less than the variance of the aggregate shock across all information structures.

We describe a methodology for making counterfactual predictions when the information held by strategic agents is a latent parameter. The analyst observes behavior which is rationalized by a Bayesian model in which agents maximize expected utility, given partial and differential information about payoff-relevant states of the world. A counterfactual prediction is desired about behavior in another strategic setting, under the hypothesis that the distribution of and agents’ information about the state are held fixed. When the data and the desired counterfactual prediction pertain to environments with finitely many states, players, and actions, there is a finite dimensional description of the sharp counterfactual prediction, even though the latent parameter, the type space, is infinite dimensional.

We describe a methodology for making counterfactual predictions in settings where the information held by strategic agents is unknown. The analyst observes behavior assumed to be rationalized by a Bayesian model, in which agents maximize expected utility, given partial and differential information about payoff-relevant states of the world. A counterfactual prediction is desired about behavior in another strategic setting, under the hypothesis that the distribution of the state and agents’ information about the state are held fixed. When the data and the desired counterfactual prediction pertain to environments with finitely many states, players, and actions, the counterfactual prediction is described by finitely many linear inequalities, even though the latent parameter, the information structure, is infinite dimensional.

We describe a methodology for making counterfactual predictions when the information held by strategic agents is a latent parameter. The analyst observes behavior which is rationalized by a Bayesian model, in which agents maximize expected utility, given partial and differential information about payoff-relevant states of the world, represented as an information structure. A counterfactual prediction is desired about behavior in another strategic setting, under the hypothesis that the distribution of the state and agents’ information about the state are held fixed. When the data and the desired counterfactual prediction pertain to environments with finitely many states, players, and actions, there is a finite dimensional description of the sharp counterfactual prediction, even though the latent parameter, the information structure, is infinite dimensional.

We describe a methodology for making counterfactual predictions when the information held by strategic agents is a latent parameter. The analyst observes behavior which is rationalized by a Bayesian model in which agents maximize expected utility given partial and differential information about payoff-relevant states of the world. A counterfactual prediction is desired about behavior in another strategic setting, under the hypothesis that the distribution of and agents’ information about the state are held fixed. When the data and the desired counterfactual prediction pertain to environments with finitely many states, players, and actions, there is a finite dimensional description of the sharp counterfactual prediction, even though the latent parameter, the type space, is infinite dimensional.

We describe a methodology for making counterfactual predictions in settings where the information held by strategic agents and the distribution of payoff-relevant states of the world are unknown. The analyst observes behavior assumed to be rationalized by a Bayesian model, in which agents maximize expected utility, given partial and differential information about the state. A counterfactual prediction is desired about behavior in another strategic setting, under the hypothesis that the distribution of the state and agents’ information about the state are held fixed. When the data and the desired counterfactual prediction pertain to environments with finitely many states, players, and actions, the counterfactual prediction is described by finitely many linear inequalities, even though the latent parameter, the information structure, is infinite dimensional.

We characterize revenue maximizing mechanisms in a common value environment where the value of the object is equal to the highest of bidders’ independent signals. If the object is optimally sold with probability one, then the optimal mechanism is simply a posted price, with the highest price such that every type of every bidder is willing to buy the object. A sufficient condition for the posted price to be optimal among all mechanisms is that there is at least one potential bidder who is omitted from the auction. If the object is optimally sold with probability less than one, then optimal mechanisms skew the allocation towards bidders with lower signals. This can be implemented via a modified Vickrey auction, where there is a random reserve price for just the high bidder. The resulting allocation induces a “winner’s blessing,” whereby the expected value conditional on winning is higher than the unconditional expectation. By contrast, standard auctions that allocate to the bidder with the highest signal (e.g., the first-price, second-price or English auctions) deliver lower revenue because of the winner’s curse generated by the allocation rule. Our qualitative results extend to more general common value environments where the winner’s curse is large.