A Simple Mechanism-Design Approach to Informed-Principal Problems
Last Updated: October 06, 2026
PaperAbstractA principal who privately knows her type and designs a mechanism creates a signalling problem. Her choice of mechanism can reveal her information, often generating equilibrium multiplicity depending on beliefs. Informed-principal problems are therefore often avoided in applied work. I show that, with independent one-dimensional private values and quasilinear utility (allowing nonlinear preferences over allocations), the problem reduces to standard optimal mechanism design. Using BIC--DIC equivalence, I show that every principal type attains her full-information payoff, with no signalling analysis and no equilibrium construction. The solution is a Myerson strong solution and pins down the principal’s payoff in every perfect Bayesian equilibrium. It is also robust to beliefs, on and off the equilibrium path. Truthful reporting is a dominant strategy for agents and participation is optimal ex post, meaning no assumption about how agents interpret the principal’s choice of mechanism is needed.
Commitment via Third-Party Contracts in Bilateral Trade: A Three-Way Equivalence
Paper [Supplement] AbstractA buyer with private value makes a take-it-or-leave-it offer to a seller with private cost. Which trading outcomes are implementable if the seller can sign observable and binding contracts ex ante with a third party that specify transfers as a function of the price posted and whether trade occurs? We establish a three-way equivalence: contract-implementable outcomes coincide with those achievable if the seller commits ex ante to an observable cost-dependent price-acceptance strategy, which are outcomes implementable with direct bilateral trading mechanisms subject to ex post monotonicity of the allocation in the seller's cost, and buyer’s interim incentive and participation constraints. We show royalty schemes can implement ex post efficiency, but also turn buyer monopsony into seller monopoly.
Works in Progress
Shaping the Stock: V-Shaped Pricing with Random Discount Windows