EC 2025
INFORMS Workshop on Market Design

      Location: Stanford University (SIEPR 130)

The 2025 INFORMS Workshop on Market Design will be held on the morning of July 10th, 2025 in conjunction with the EC 2025 conference. The workshop will bring together researchers and practitioners that work on market design, and will represent a broad range of perspectives from more theoretical to more applied and empirical work. As with previous iterations, the workshop is sponsored and organized by the INFORMS Section on Auctions and Market Design, i.e., the workshop is a successor of

• INFORMS Workshop on Market Design 2024

• INFORMS Workshop on Market Design 2023

• INFORMS Workshop on Market Design 2022

• INFORMS Workshop on Market Design 2021

• INFORMS Workshop on Market Design 2019

• INFORMS Workshop on Mathematical Optimization in Market Design 2018

Market design is a field of applied and theoretical research that sits comfortably on the intersection of computer science, economics, and operations research. In recent decades, the theory and applications of market design have blossomed. In this workshop, we will focus on a set of promising, new applications of market design. We are particularly interested in exploring research areas which are just recently making their impact on practice, including empirical work on evaluating mechanisms and approaches in applications. Topics include but are not limited to:

• - Machine learning and generative AI in market design
• - Mathematical optimization and pricing in markets
• - Iterative multi-object auctions
• - Procurement auctions
• - Energy markets
• - Matching with constraints and complex preferences

Schedule (July 10, 2025 )

• 9:00 - 9:05 — Vahideh Manshadi / Itai Ashlagi: Opening Remarks
• 9:05 - 9:40 — Yang Cai (Yale University): Simple vs. Optimal under Endogenous Information
  We study multi-item auctions where the seller jointly designs the selling mechanism and the information structure for the buyer to learn his values. Unlike Hart and Nisan (2019), who show that simple mechanisms perform poorly under exogenous information, we find that when the seller controls information, even uniform pricing guarantees at least ½ of the optimal revenue. Item pricing performs even better, achieving strictly more than half of the optimum in general. We further identify sufficient conditions under which item pricing is revenue-optimal. Our results highlight the power of information design in making simple item pricing mechanisms competitive in multi-dimensional settings. Joint work with Yingkai Li and Jinzhao Wu.
• 9:40 - 10:15 — Irene Lo (Stanford University): Optimization Meets Participation: Iterative School Zone Generation with LLMs
  Designing zones for school choice systems is a complex problem that requires eliciting complex preferences and balancing multiple stakeholder objectives. To address this challenge, we propose a stakeholder-in-the-loop framework that iterates between using optimization to generate zone boundaries for given preferences, and allowing stakeholders to participate and learn their preferences by reacting to zones. To facilitate stakeholder participation, we use LLMs to translate between natural language preferences and optimization constraints. To enable real-time use of our framework, we develop faster computational approaches for the multi-school zoning problem using both optimization and sampling-based methods. Our framework is being used to support zone design in the San Francisco Unified School District.
• 10:15 - 10:45 — Coffee Break
• 10:45 - 11:20 — Eric Richert (University of Chicago): Indirect Inference Techniques as an Equilibrium Computation Method
  This paper proposes a computational method to solve for equilibria in Bayesian games and incomplete models, addressing challenges associated with approximation, stability, and computational costs in existing methods. It solves for the set of action (bid) distributions that are consistent with a primitive (value) distribution in a known or estimated set of primitive distributions. The method uses a mapping from actions to primitives, which can be computed using readily available empirical techniques. A set of simulations verifies the performance and provides guidance for implementation. The method allows for the solution of the multi-unit auction game with step-function bids that characterize, for example, electricity and Treasury auctions. I apply the method to solve for equilibria in a counterfactual uniform price auction for the Turkish Treasury (Hortacsu McAdams (2010)) and to evaluate the merger effects in that setting.
• 11:25 - 12:00 — Mike Ostrovsky (Stanford GSB): Effective and Equitable Congestion Pricing: New York City and Beyond
  In this paper, we argue that the New York City congestion pricing scheme that was launched on January 5, 2025 has a major shortcoming: it has a much more severe impact on the drivers of personal vehicles than on the passengers of taxis and ride-hailing vehicles or on the clients of delivery services. In addition to being inequitable, this shortcoming also makes the congestion pricing scheme relatively ineffective at solving the traffic congestion problem inside the Central Business District, due to the fact that the drivers of personal vehicles constitute a minority of traffic there. We provide empirical evidence from the launch of the current plan, and propose a simple modification to the scheme that addresses this shortcoming. Joint work with Frank Yang.

Organizers

• Christina Aperjis (Power Auctions)

• Martino Banchio (Bocconi & Google Research)

• Shoshana Vasserman (Stanford GSB & Google Research)