Matching Theory and College Admissions


Yeon-Koo Che
Kelvin J. Lancaster Professor of Economic Theory


The research develops new methods for analyzing two-sided matching, whether it is organized via a decentralized market or a central clearinghouse. The projects focus on college admissions as a possible application, although the new theory is general and will apply to many other contexts. The first project develops a canonical model of a decentralized college admissions game, analyzes the strategic yield management decisions made by colleges and universities, and evaluates the resulting equilibrium outcomes. The second project uses data from Australia, where college admissions are centralized in each state. The data analysis determines the effect of specific features of this centralized process on the resulting assignment of students to universities. The third project is again economic theory; the team seeks to develop new and more general conditions that guarantee existence of stable matching in an asymptotic sense; as applied to college admissions this would mean a match that is stable as the economy grows in the number of students and in the number of college places.

Broader impacts of this project include new insights into the strengths and weaknesses of different ways of encouraging college access through changes in admissions processes.