Using the recent China Family Panel Studies, we identify the subjects studied by college (2–3 years) graduates and university (4–5 years) graduates. For the university graduates, we can further distinguish universities by the tier of selectivity (i.e., Key and Ordinary Universities). We take advantage of the rich information on the respondent's school cohort, hukou status at age 12, and the mother's age and education to estimate university applicants' simultaneous choice of subject and tier of prestige of higher education institutions (HEIs). Using the doubly robust Inverse Probability Weighted Regression Adjustment method to account for selection – on observables – into subjects and tiers, our treatment effect estimates suggest that pooled OLS and random-effect models substantially underestimate the effect of attending universities that are more prestigious for graduates of both genders in law, economics, and management (LEM). We also demonstrate that the recent massive expansion of the higher education sector resulted in reduced returns to HE for all graduates, except for graduates who studied LEM or Other non-STEM (sciences, technology, engineering and math/medicine) subjects at the most prestigious universities. The results are robust to treating subjects as predetermined for the selection into HEIs by tiers of prestige.