Mechanism Design with Inspection

This mini-course focuses on a central technique in mechanism design with inspection. Rather than surveying a broad list of papers, the course develops a unified approach: representing a mechanism through its interim allocation and inspection probabilities, formulating the principal’s problem as a linear program in these reduced forms, and using duality and complementary slackness to uncover the structure of the optimal mechanism.

Each session is organized around one main paper, with related work providing additional context.

Prerequisites

Participants should be familiar with standard one-shot mechanism design with transfers, including incentive compatibility and the revelation principle. Basic knowledge of linear programming, duality, and complementary slackness is also assumed.

Session 1 — The BDL Benchmark via LP Duality

This session derives the benchmark model of optimal allocation with costly verification using linear programming and duality.

Main references:

• Ben-Porath, E., Dekel, E., and Lipman, B. L. (2014), “Optimal Allocation with Costly Verification,” American Economic Review, 104(12).

• Vohra, R. V. (2012), “Optimization and Mechanism Design,” Mathematical Programming, 134(1), 283–303.

Session 2 — Dynamic Inspection and the Secretary/Prophet Connection

This session studies dynamic mechanisms with verification and explores their connection to secretary problems and prophet inequalities.

Main reference:

• Epitropou, M. and Vohra, R. V. (2019), “Dynamic Mechanisms with Verification,” Symposium on Algorithmic Game Theory (SAGT).

Session 3 — Noisy Inspection and Noisy Self-Information

The final session considers environments in which inspection is imperfect and agents have noisy information about their own values. The discussion focuses on which parts of the reduced-form approach continue to apply.

Main references:

• Weitzman, M. L. (1976), “Optimal Search for the Best Alternative,” Econometrica.

• Khalfan, Xia, and Vohra (2026), working paper.