We study road congestion as a mechanism design problem. In our basic model we analyze the allocation of a set of drivers among two roads, one of which may be congested. An additional driver on the congestible road imposes an externality on the other drivers by increasing their travel time. Each...

We consider a budget-constrained mechanism designer who selects an optimal set of projects to maximize her utility. Projects may differ in their value for the designer, and their cost is private information. In this allocation problem, the quantity of procured projects is endogenously determined...

This note establishes a revelation principle in terms of payoff for deterministic mechanisms under ex-post constraints: the maximal payoff implementable by a feasible deterministic mechanism can also be implemented by a feasible deterministic direct mechanism.

In mechanism design with (partially) verifiable information, the revelation principle holds if allocations are modelled as the Cartesian product of outcomes and verifiable information, giving rise to evidence-contingent mechanisms. Consequently, incentive constraints characterize the...

In many trade environments - such as online markets - buyers fully learn their valuation for goods only after contracting. I characterize the buyer-optimal ex-ante information in such environments. Employing a classical sequential screening framework, I find that buyers prefer to remain...

This paper shows that the possibility of collusion between an agent and a supervisor imposes no restrictions on the set of implementable social choice functions (SCF) and associated payoff vectors. Any SCF and any payoff profile that are implementable if the supervisor's information was public...

In mechanism design, Myerson regularity is often too weak for a quantitative analysis of performance. For instance, ratios between revenue and welfare, or sales probabilities may vanish at the boundary of Myerson regularity. This paper therefore explores the quantitative version of Myerson...

In school choice problems, the widely used manipulable Immediate Acceptance mechanism (IA) disadvantages unsophisticated applicants, but may ex-ante Pareto dominate any strategy-proof alternative. In these cases, it may be preferable to aid applicants within IA, rather than to abandon it. In a...

Numerous simple proofs of the celebrated Gibbard-Satterthwaite theorem (Gibbard, 1977, Satterthwaite, 1975) has been given in the literature. These are based on a number of different intuitions about the most fundamental reason for the result. In this paper we derive the Gibbard-Satterthwaite...

We study rotation programs within the standard implementation framework under complete information. A rotation program is a myopic stable set whose states are arranged circularly, and agents can effectively move only between two consecutive states. We provide characterizing conditions for the...