You are here: Home Projects B: Product market regulation B04: Decentralized markets

B04: Decentralized markets

The aim of this project is to analyze the workings of decentralized markets both theoretically and empirically, and to draw lessons on the effectiveness of market design by an intermediary or a regulatory authority. We study price determination and contracting arrangements. In particular, we extend existing search theory to account for learning about market conditions. We also extend the existing theory of relational contracting to multilateral settings. We plan to utilize data from online marketplaces to test the theory and to evaluate and suggest changes to market design and regulatory policies.


Project members (B04)


Policy outreach (B04)


Publications (B04)


Discussion papers (B04)



Decentralized markets are ubiquitous: Housing, labor, online markets, over-the-counter asset markets. Search models are central tools in economics for the analysis of decentralized markets. Uncertainty and learning about market conditions as well as informal contracting arrangements are critical characteristics of decentralized markets. Inequality of opportunity and inefficient allocations may arise due to differential access to markets, differential contracting possibilities or differential information.

Policy relevance

New market designs, especially concerning informal contracting and search with learning, raise questions about product and financial market regulation: Does uncertainty about market conditions imply market failures? Is there a need for transparency-increasing regulatory interventions to enhance information about market conditions? Theoretical and empirical evaluation of the market design employed by (online) intermediaries: Do intermediaries choose transparent or opaque markets? Is there a need for consumer protection policies?

Project plan

Work Package 1 - Theory of search with learning

  • Develop tractable models of individual search with learning; utilize tools from the literature on experimentation with multi-armed bandits.
  • Study the magnitude of distortions induced by incomplete information about market conditions, in terms of the distance of prices from market clearing, the extent of price dispersion, and the reduction of liquidity.
  • What determines the magnitude of distortions? Search process, bargaining protocol, divisibility, availability of monetary transfers
  • Analyze the incentives for a private market designer (intermediary) to choose a transparent or opaque market structure.
  • Analyze the influence of uncertainty about market conditions and of learning on entry and sorting into markets, and especially on sorting by prior expertise and information.

Work Package 2 - Relational contracting in one-to-many settings

  • Use the theory of relational contracts to analyze interactions in anonymous goods markets.
  • Extend the theory to trading between one (or several) long-lived and many short-lived agents.
  • Study conditions under which decentralized relational contracting can solve classical efficiency problems related to moral hazard and adverse selection.
  • Model traders' incentives for voluntary and unbiased evaluation of trades.
  • Key: Reputational concerns through “ratings” designed by the platform.

Work Package 3 - Testing the theory

  • Use cooperation with researchers at firms to access data and possibly run field experiments in online markets (eBay, Airbnb, Thumbtack).
  • Test for influences of the organization of search results by platforms.
  • Test for searchers' limits in information processing.
  • Test for searchers' or bidders' learning.
  • Test for agents' incentives to build reputation and the effect of reputation on trading outcomes.
Document Actions