Econometrica, Vol. 74, No. 4 (July, 2006), 885–928 ADMISSION, TUITION, AND FINANCIAL AID POLICIES IN THE MARKET FOR HIGHER EDUCATION B Y DENNIS EPPLE,RICHARD ROMANO, AND HOLGER SIEG 1 We present an equilibrium model of the market for higher education. Our model simultaneously predicts student selection into institutions of higher education, finan- cial aid, educational expenditures, and educational outcomes. We show that the model gives rise to a strict hierarchy of colleges that differ by the educational quality provided to the students. We also develop a new estimation procedure that exploits the observed variation in prices within colleges. Identification is based on variation in endowments and technology. It does not rely on observed variation in potentially endogenous char- acteristics of colleges such as peer quality measures and expenditures. We estimate the structural parameters using data collected by the National Center for Education Statis- tics and aggregate data from Peterson’s and the National Science Foundation. KEYWORDS: Higher education, peer effects, college competition, nonlinear pricing, equilibrium analysis, estimation. 1. INTRODUCTION OVER THE PAST SEVERAL YEARS, research has investigated normative and pos- itive consequences of competition in primary, secondary, and higher educa- tion, and the likely effects of policy changes, including vouchers, public school choice, and changes in education financing. 2 Some of this research has relied on general equilibrium models. Given the absence of large scale policy exper- iments, these models have been a primary tool for evaluating the impact of a variety of education reform measures. To date, the predictions of these models 1 We would like to thank a co-editor and two anonymous referees for very helpful comments. We would also like to thank Joseph Altonji, Charles de Bartolome, Pat Bayer, Steve Berry, Richard Blundell, Martin Browning, Steven Coate, Steven Durlauf, David Figlio, James Heckman, Carolyn Levine, Dean Littlefield, Charles Manski, Erin Mansur, Robert Moffitt, Tom Nechyba, Ariel Pakes, Steve Stern, Chris Taber, Miguel Urquiola, Michael Waldman, and seminar partici- pants at Brown University, the University of California in Berkeley, University College London, the University of Colorado, Cornell University, CREST, the University of Kentucky, the Univer- sity in Louvain, Northwestern University, the Norwegian School of Economics in Bergen, Ohio State University, the University of Oslo, Stanford University, the University of Toronto, the Uni- versity of Virginia, the Brookings Conference on Social Interactions, the SITE Workshop on Structural Estimation, the Triangle Applied Micro Conference, the CAM Workshop on Charac- teristics Models, the ERC conference in Chicago, the SED in New York, the NASM at UCLA, and the NBER Public Economics Meeting. We would also like to thank Gary Gates for research assistance, the National Center for Education Statistics and Peterson’s for providing us with data used in this paper. Financial support for this research is provided by the National Science Foun- dation, the MacArthur Foundation, and the Alfred P. Sloan Foundation. 2 Recent theoretical studies include Benabou (1996), Caucutt (2002), de Bartolome (1990), Epple and Romano (1998, 2003), Fernandez and Rogerson (1998), Manski (1991), Nechyba (2000), and Rothschild and White (1995). Some recent empirical studies include Bergstrom, Rubinfeld, and Shapiro (1982), Carlton, Bamberger, and Epstein (1995), Downes and Greenstein (1996), Ferreyra (2003), Fuller, Manski, and Wise (1982), Hoxby (2000), and Rouse (1998). 885