Markets are systems that allocate scarce resources to individuals, coordinate aggregate behavior, and determine societal outcomes. Research on the design of markets has impacted the FCC spectrum auctions, the National Residency Matching Program, and school choice programs, among others. The goal of this project is to bring similar fruit to bear on so-called developing markets that primarily serve poor and disadvantaged participants. The project will use two distinct markets, each with their own theoretical and practical challenges, as examples of this research agenda. The first is an agricultural market in Uganda. The goal of this market is to find an efficient set of trades among subsistence farmers and buyers. The second is an insurance market in Thailand. The goal of this market is to build a system of lending and borrowing for villagers. In addition to improving the lives of those who are impacted by developing markets, found all over the world, the project will provide opportunities for graduate and undergraduate students to participate in the research, as well as a rich interdisciplinary interaction between theoretical computer scientists and development economists.

Developing markets present unique challenges to a market designer, and present an opportunity to advance the state-of-the-art in market design. The primary challenge is an informational one: due to technological constraints, the designer has significantly less knowledge of and access to the underlying economic environment than is typically assumed. Populations in these markets often have urgent need for access to capital, while also being highly transitory. Both these factors push markets to clear rapidly. This, however, sacrifices market thickness and so degrades the quality of market outcomes. This project considers techniques to address this, including allocation schemes that incorporate stochastic assumptions about market participants. A secondary challenge concerns the need for simplicity and transparency. Given that, such populations are understandably suspicious of outside interventions, and lack access to technologies that help them optimize their strategies within a market, the mechanisms in this project will be designed to be easy to comprehend and easy to play, including restricting the design space to a subclass of mechanisms that closely approximate the optimal one.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Project Start
Project End
Budget Start
2018-10-01
Budget End
2021-09-30
Support Year
Fiscal Year
2018
Total Cost
$100,000
Indirect Cost
Name
Harvard University
Department
Type
DUNS #
City
Cambridge
State
MA
Country
United States
Zip Code
02138