2013 Theses Doctoral

Essays on the Economics of Education

Simpson, Steven Troy

Post-secondary education is becoming increasingly more common for students around the world. As quantity of education increases, it becomes less of a distinguishing factor to be simply a college graduate. For those who want to stand out, the quality aspects of education become more salient. Moreover, as this expansion happens in the number of colleges and college students, it becomes less common for governments to generously fund the college education of a lucky few. In addition, the cost to colleges to provide an education is also increasing. Taken together, simply as a measure of cost-comparison, choosing between colleges based on the potential quality-for-money is also an important reason for college quality's increasing salience. College quality matters, and this dissertation endeavors to show how and to what extent. The following three separate chapters estimate the returns to different forms of college quality. There has been an extensive literature that shows, in general, that more schooling is better. These chapters seek to shift the margin of analysis from the extensive margin of quantity to the intensive margin of quality. Thus, I ask the question: is better schooling better or, to put it another way, how much better is better schooling? In the first chapter, I estimate the returns to college quality, operationalized mainly through peer quality, using a regression discontinuity design and exploiting the two separate rounds (early and regular) of college admissions in Taiwan. In the second chapter, focusing on college prestige, I again use a regression discontinuity design to estimate the returns to scoring just above (vs. just below) the admissions cutoff for the lowest-ranked national college. The theory of action is that national colleges are uniformly more desirable than private colleges (excluding a few elite private colleges), if for no other reason than that their tuitions are subsidized by the government and thus much lower for the individual. The final chapter looks at a set of 11 colleges that had already been meeting the minimum requirements for being labeled a university (an important distinction in Taiwan's system), but for bureacratic reasons had not been allowed to change their label/rank until a policy change in 1997. Treating this policy change as a natural experiment, I use a difference-in-differences framework to show that cohorts entering these newly upgraded 11 universities earn statistically significantly more than cohorts entering prior to the change at the same colleges. A consistent picture emerges out of these three papers: college quality matters on several dimensions. These chapters are set apart from other papers in the literature by the causal interpretation given to both choice of college AND choice of college major. My estimates show that those who attend higher quality colleges, within the same college major, end up earning between one-tenth to one-fifth of a standard deviation more in their first year of employment after graduating. Peer quality, college prestige, and college reputation all appear to provide a return. But choice of college major appears to be one of the most important dimensions through which college quality operates, with the science-track college majors receiving most of those returns to quality.