Roberto Burguet’s research while affiliated with University of Central Florida and other places

We study targeting of advertising based on information about the consumer's likely ranking of products. Firms send ads to more promising consumers. Thus, targeting increases the expected willingness to pay for the product the consumer learns about but also reduces the expected differentiation of these products. The two effects push search, and so prices, in opposite direction. The first effect is more important for lower search costs. The second is stronger the larger the number of products. The improved selection under targeting increases firms' profitability and so advertising effort.

We propose a new solution concept for TU cooperative games in characteristic function form, the SCOOP, which builds on the symmetric Nash Bargaining Solution (NBS) by adding a consistency requirement for negotiations inside every coalition. The SCOOP specifies the probability of success and the payoffs to each coalition. The players share the surplus of a coalition according to the NBS. The disagreement payoffs are computed as the expectation of payoffs in other coalitions, using a common probability distribution that is derived from the prior distribution. The predicted outcome can be probabilistic or deterministic, but only an efficient coalition can succeed with probability one. We discuss the necessary and sufficient conditions for an efficient solution. In either case, the SCOOP always exists, is generically unique for superadditive games, and is easy to compute. Moreover, in the spirit of the Nash program, we propose a non-cooperative protocol whose stationary equilibrium identifies the SCOOP as the limit equilibrium outcome.

We analyze personalized pricing by a monopsonist facing a finite number of ex ante identical, capacity constrained suppliers with privately known costs. When the distribution of costs is sufficiently smooth and regular, the buyer chooses to make the same offer to all suppliers, leading to a posted price. When demand is sufficiently concave (convex) this price is lower (higher) than the classical monopsony price. In the limit as the seller capacities tend to zero, we obtain the classical monopsony price. Therefore, our model provides a decentralized micro-foundation for monopsony.

We present a novel microstructure for the market for athletes. Clubs simultaneously target bids at the players, in (Nash) equilibrium internalizing whether—depending on the other clubs' bids—a player not hired would play for the competition. When talent is either scarce or has low outside options, we support—and generalize to heterogeneous players—the Coasian results of Rottenberg (1956) and Fort and Quirk (1995): talent allocation is efficient and independent of initial “ownership” and revenue sharing arrangements. We also characterize equilibria when talent is abundant (or has a high outside option). The analysis uses a nonspecific club objective with an endogenously derived trade‐off between pecuniary and nonpecuniary benefits.(JEL J4, L1, L2)

We model oligopolistic firms, producing substitutes, who compete for inputs from capacity constrained suppliers in a decentralized market. Compared to a price-taking input market, the incentive to foreclose downstream competitors leads to higher input prices and to a higher aggregate amount of input acquired. This novel feature mitigates the output reducing effect of downstream market power and may even restore efficiency in the unique (input) market clearing equilibrium. Other equilibria, where firms coordinate on which suppliers to target, result in excess supply (involuntary unemployment, if input is labor) and even higher input prices. Our insights generalize to alternative vertical structures.

I investigate the design of optimal procurement mechanisms in the presence of corruption. After contracting with the sponsor, the contractor may bribe the inspector to misrepresent quality. The mechanism affects whether bribery occurs. I discuss the cases of both fixed and variable (with the size of misrepresentation) bribes, and also uncertainty about the bribe amount. In all cases, the optimal contract curtails quality for low efficiency contractors but also for the most efficient contractors. I also present models of bribe negotiations whose reduced form coincide with the model analyzed in the paper, and discuss implementation and the effect of competition.

We propose a new model of simultaneous price competition, where firms offer personalized prices to consumers, who then independently decide which offer to accept, if any. Even with decreasing returns to scale, this decentralized market mechanism has a unique equilibrium, which is independent of any exogenously imposed rule for rationing or demand sharing. In equilibrium, the firms behave as if they were price takers, leading to the competitive outcome (but positive profits). Given the unique result for the short-run competition, we are able to investigate the firms’ ex ante capital investment decisions. While there is underinvestment in the long-run equilibrium, the overall outcome is more competitive than one-shot Cournot competition.