| Abstract: |
This dissertation focuses on three aspects of higher education policy that are pertinent to both deans and policymakers. The first essay contributes to the debate about the relationship between tuition fees and demand for higher education. The literature indicates significant publication selection against positive estimates, suggesting that many researchers use the sign of the estimated effect as a specification test to show that education is unlikely a Giffen good. We found evidence of systematic dependencies between the estimated effects and data, methodological, and publication characteristics. The mean effect beyond publication bias is close to zero. The second essay shows how easily firms can substitute workers with different educational backgrounds. We evaluate the elasticity of substitution between skilled and unskilled workers, which is a key parameter in the analysis of wage inequality. We show that the empirical literature is consistent with both publication and attenuation bias in the estimated inverse elasticities. The publication bias-corrected estimates remain close to zero. The result is consistent with attenuation bias in the literature and implies an elasticity of 4 after correction for both biases. The third essay shows how the real implementation of the Russian Unified State Exam (USE) reform for high school graduates affected the returns to university education. The findings suggest that the reform has positively impacted education returns and the effect is statistically significant. The impact is more prominent in Moscow and St. Petersburg, where most elite schools are located. This could be due to the increased mobility of talented individuals from small cities, towns, and rural areas to bigger cities, where salaries are higher. |
