School Competition and Product Differentiation∗

This paper provides the first combined theoretical and empirical analysis of the impact of private school entry on private schools’ strategic choices of quality, addressing the concern that school competition can harm poorer students. I develop a model where schools select their quality when (1) the match between a school and a student affects test scores, and (2) wealthier students’ enrollment decisions are more responsive to quality. The model demonstrates that private school entry can increase test scores for wealthy students and decrease them for poor students in the same school. Using Pakistani data, I find that entry increases within-school test score inequality by 0.1 SD. Structural estimates of schools’ equilibrium choices show that schools choose match-specific qualities that strongly favor wealthy students, and this intensifies with entry. Moving a poor student from the average private school to one that caters to her increases her test score by 0.2 SD. ∗I am deeply indebted to my advisors, Chris Avery, Roland Fryer, Asim Khwaja, and Nathan Nunn for their guidance and encouragement. I am also grateful to Victor Aguirregabiria, Nava Ashraf, Oriana Bandiera, David Baqaee, James Feigenbaum, Christopher Jencks, Larry Katz, Robert McMillan, Ariel Pakes, Doug Staiger, Bryce Millett Steinberg and numerous seminar and conference participants for helpful comments. I gratefully acknowledge the support of the NSF Graduate Research Fellowship and the Harvard Inequality and Social Policy Fellowship. †University of Toronto and CIFAR, contact: [email protected].