Causes and Consequences of Student-College Mismatch

Our objective is to understand the observed patterns of student-college sorting and earnings premia associated with college quality in the United States. Higher quality colleges have higher graduation rates and their graduates earn more. Yet, a large fraction of high scoring students enroll in two-year schools and low quality four-year schools – this “undermatch” phenomenon is more pronounced for low income students. To understand these patterns, we develop a model with heterogeneous students and colleges that differ in human capital production technology and financial costs. We quantify our model using NLSY97 student-level and college transcript data, as well as quasi-experimental evidence on the impact of financial and information interventions. We find that college technology effects are at least as important as selection effects (∼ 50 − 75% vs. ∼ 25 − 50%) at explaining the observed college quality premia. Our results highlight the importance of access constraints in explaining the “undermatch” for low income high ability students.