Association for Public Policy Analysis & Management

Many countries expanded their systems of higher education in the past decades (including U.S., UK, Russia, China, Italy, etc.). Effects of such expansions on college access and persistence, education quality, and graduates’ labor market outcomes vary, partly depending on type and scale of the expansions, and how such expansions were financed. This study analyzes one of the most radical and large-scale higher education expansion programs in modern history, which began in 1999 in China. Number of available spots in higher education institutions for high school graduates increased by 50 percent in the first year of the expansion, and increased by another 40 and 20 percent in the following two years respectively.

Methodologically, China’s substantially expanded access to higher education institutions over a short period of time provides variation in college admission that is arguably unrelated to pre-existing factors like students’ ability, motivation, and family background. I merge survey data with province-level administrative and archival data, and adopt a difference-in-difference (DiD) approach (using exogenous variation in both students’ birth cohorts and the intensity of expansion across provinces) to estimate the causal effects of the expansion on college access, educational attainment and graduates’ labor market outcomes, such as earnings, employment stability and occupation.

Moreover, I conduct subgroup analysis to explore heterogeneous effects of the expansion on students with different levels of high school achievement and family background. Results show that students who have mediocre high school achievement but higher family socio-economic status are more likely to benefit from the expansion. I also find that older cohorts who were not directly affected by the expansion but were still enrolled in college when an increasing number of new students entered college are adversely affected by the expansion, mainly because the expansion crowed out recourses per capita and decreased quality of instruction and teaching.

Apart from heterogeneous and spillover effects for certain populations, in general, younger cohorts who would have attended college after the expansion are more likely to attend and complete college. However, students are slightly less likely to attend national flagship universities, but more likely to attend community colleges and less selective 4-year universities. My DiD estimators imply that the expansion decreases earnings premium for non-selective university and community college graduates, but increases earnings for selective 4-year university graduates. Results suggest that deteriorating education quality (measured by spending per capita, faculty qualifications, teacher-student ratio, etc.) and decreasing ability of marginal college graduates are two major channels through which these changes happen.