Essays in Microeconomic Theory: Information Design, Partnerships, and Matching

Bela Szabadi

doi:https://doi.org/10.21985/n2-ttnq-8y22

Work

Essays in Microeconomic Theory: Information Design, Partnerships, and Matching

Public Deposited

Download PDF

Download All Files (.zip)

This dissertation consists of three essays in Microeconomic Theory that study the interplay between mechanism and information design, provide insights into the design of efficient dispute resolution mechanisms for partnerships, and analyze the stability of fractional matchings in two-sided markets.', "In the first chapter, we study the optimal disclosure policy in a setting in which both a monopolistic seller and a single potential buyer of an indivisible good face product-related uncertainty. The buyer's valuation depends on his privately known type and the unknown attributes of the object. The seller designs a disclosure policy with a publicly observable outcome and interacts with the buyer only after the information is revealed. We identify the condition on the underlying preferences that guarantees the optimality of full disclosure: if buyer types are ranked uniformly across states by their willingness to pay, the seller prefers to disclose all information about her product. Whenever this condition is violated, there always exist type distributions for which full disclosure is suboptimal. We also study the optimal information disclosure policy in cases in which the released information might affect the ranking of the types. In environments with two types, the seller prefers partial disclosure that never reverses the ranking and fully reveals states in which the net gain from restricting output by selling only to the higher type is above a certain threshold. In settings with many types, the optimal disclosure policy depends on the distribution of types in a more complicated way. We establish structural properties that the groups of types that purchase the object at different posteriors must fulfill under the optimal policy.The second chapter, which is a joint work with Daniel Fershtman, studies the efficient resolution of partnership disputes where efficiency need not imply dissolution. The modeled endogeneity of dissolution is a departure from the partnership dissolution literature emanating from Cramton et al. (1987), which has effectively focused on circumstances in which dissolution is unavoidable; the analysis yields predictions that contrast sharply with their analogs in such environments. First, we show that unless a dispute is sufficiently severe, its efficient resolution is infeasible without external subsidy. Furthermore, we prove that the severity of a dispute has a non-monotonic effect on the cost of its efficient resolution. We also consider an alternative class of mechanisms and characterize the disputes for which a profit-oriented arbitrator benefits from a more conservative approach in opting for dissolution. The latter characterization has implications for the design of arbitration in environments in which the partners' decision to trigger a dispute is itself endogenous.", 'The third chapter describes a new stability concept for fractional matchings in two-sided markets in which agents can be matched with multiple partners over time. The new definition, called C-stability, is equivalent to the non-existence of unstable implementing time schedules and is less restrictive than strong stability proposed by Roth et al. (1993). We characterize this condition both geometrically and in terms of the underlying preferences. The geometric characterization reveals the relationship between the set of fractional matchings satisfying C-stability, the polytope of matchings, and the polytope of weakly stable fractional matchings. The preference-based characterization uses these results to provide necessary and sufficient conditions on the underlying preferences that a fractional matching must fulfill to satisfy C-stability. These findings also highlight the difference between strong stability and C-stability and identify the exact cases in which strong stability excludes matchings that can be implemented only by stable time schedules.

Last modified