My dissertation consists of three essays that study the college premium in China and how it has been affected by China’s higher education expansion. In the first essay, I utilize the high education expansion as exogenous source to estimate the college premium. The rapidly changing access to college provides a rare opportunity to estimate a local treatment effect (LATE) of college education on earnings by utilizing the drastic increase in college admission rate in 1999. I also utilize the yearly admission rate as an instrumental variable for the endogenous college education. Using China Household Income Project 2013, the two IV estimates of college premium are 75.7 and 57.5 log points respectively. The second essay examines the trends of the college earnings premium by age groups from 1995 to 2013 in China. Specifically, based on China Household Income Projects, the college premium for the younger group (age 25-34) stagnated, while the college premium for the older group (age 45-54) increased substantially. I attribute the stagnation for the younger group to the fast-growing relative supply of younger college workers due to China’s higher education expansion. Holding the age cohort and survey year constant, a one unit increase in log relative size of college workers leads to 10.3 log points decrease in college premium. The third essay further explores the channel through which the cohort size affects the college premium. Using Blinder-Oaxaca decomposition, I find that, for all survey years and age groups, the differential of the higher-skilled occupations share between college and non-college educated workers only explains a small part of college premium, 10%-30%. The part due to the higher-skilled occupational premium is negligible. Over 70% of the college premium is contributed by the college premium among the workers with lower-skilled occupations.

This dissertation consists of two chapters. In the first chapter, I explain changes in the life-cycle earnings profile for different birth cohorts. The second chapter assesses the quantitative importance of federal aid for college education in explaining college premium. In the first chapter, I document the life-cycle earnings profile for the 25-year- old college- and high school-educated white men in 1940, 1950, 1960 and 1970. I find that later cohorts have flatter average life-cycle earnings profile. Using a version of the Ben-Porath model, I propose an explanation based on the composition effect. In my model, all individuals have a high school diploma and are differentiated by their ability. They must decide whether to work or go to a four-year college. There is a threshold ability above which individuals choose to attend college and below which they work. All cohorts face the same ability distribution and an exogenous sequence of wage rate per unit of human capital that grows at a constant rate. A higher initial level of wage rate increases college attainment implying that the average ability is lower for both college- and high school-educated individuals. From the Ben- Porath model, lower ability individuals have less steep increment in their earnings. This implies that the average college (and high school) life-cycle earnings profile for the 1970 cohort will be flatter than that of the 1940 cohort. My model is able to quantitatively explain 67 and 35 percent of the flattening in the average life-cycle earnings profile for college and high school-educated individuals, respectively. Since the late 1970s, there has been a strong increase in the college premium. While most papers focus on skill-biased technical change, the second chapter explores the role of federal aid as a possible source of inequality. I build a model where all individuals have a high-school diploma but are heterogeneous with respect to their innate abilities and initial human capital. They decide whether to attend college to accumulate more human capital before working, or to start working right away. The production function for human capital in college requires two inputs: human capital and goods. In this context, two mechanisms are key for the behavior of the college premium. First, federal aid makes it easier to afford the goods input in the human capital technology. This induces college students to accumulate more human capital and consequently, they have higher earnings. Second, as more individuals attend college due to rising income, the composition of college graduates changes: more low- ability individuals attend college, implying a decrease in average college earnings. A calibrated version of the model accounts fully for the rise in the college premium. Federal aid alone accounts for about 70 percent of the rise.

Among better-educated employed workers, the fraction of full-time full-year (FTFY) workers is quite high and stable over time in the U.S. Among those with low education levels, however, this fraction is much lower and considerably more volatile. These observations suggest that the composition of unobserved skills is subject to sharp movements within low-educated employed workers, while the scale of these movements is potentially much smaller within high-educated ones. The standard college premium framework accounts for the observed shifts between education categories, but it cannot account for unobserved compositional changes within education categories. This thesis uses Heckman&#039 / s two-step estimator on repeated Current Population Survey cross sections to calculate a relative supply series that corrects for unobserved compositional shifts due to selection in and out of the FTFY status. We find that the well-documented deceleration in the growth rate of relative supply of college-equivalent workers after mid-1980s becomes even more pronounced once we correct for selectivity. This casts further doubt on the relevance of the plain skill-biased technical change hypothesis. We conclude that what happens to the within-group skill composition for low-educated groups is critical for fully understanding the trends in the relative supply of college workers in the United States.

Among better-educated employed workers, the fraction of full-time full-year (FTFY) workers is quite high and stable over time in the U.S. Among those with low education levels, however, this fraction is much lower and considerably more volatile. These observations suggest that the composition of unobserved skills is subject to sharp movements within low-educated employed workers, while the scale of these movements is potentially much smaller within high-educated ones. The standard college premium framework accounts for the observed shifts between education categories, but it cannot account for unobserved compositional changes within education categories. This thesis uses Heckman&#039 / s two-step estimator on repeated Current Population Survey cross sections to calculate a relative supply series that corrects for unobserved compositional shifts due to selection in and out of the FTFY status. We find that the well-documented deceleration in the growth rate of relative supply of college-equivalent workers after mid-1980s becomes even more pronounced once we correct for selectivity. This casts further doubt on the relevance of the plain skill-biased technical change hypothesis. We conclude that what happens to the within-group skill composition for low-educated groups is critical for fully understanding the trends in the relative supply of college workers in the United States.

This dissertation consists of three chapters, which focus jointly on the effects of education policy on the functioning of labor markets.De-industrialization and technological progress have changed job markets fundamentally. The most fundamental change is that the concept of a worker as a unit of production relatively insensitive to inherent characteristics has been overthrown. Service sectors that have taken over from manufacturing as the engines of economic activity rely primarily on human capital for autonomous production. This is especially true for internationally tradable services. Their stark development was rendered possible by the informationcommunication revolution. Skills and talent, as well as their allocation to suitable tasks matter for production, now more than ever. We argue in this dissertation that the ability of education policy to facilitate optimal task allocation plays a role in maximizing aggregate production and in influencing education earnings premia, as well as employment volumes in various sectors of activity. / Doctorat en Sciences économiques et de gestion / info:eu-repo/semantics/nonPublished