UC Berkeley

This dissertation examines low-income college attendance and financial aid. The first chapter is an overview of the field and reviews relevant literature. The second chapter examines the growing income gradient in education. Using data from the National Education Longitudinal Study of 1988, I observe an income gradient in academic achievement in 8th grade, and demonstrate that it grows over the course of the students' education. This paper sheds light on this phenomenon by examining five critical steps on the way to a degree (middle school achievement, academic preparation in high school, applying to college, enrolling in a four-year college, and graduating) and the junctions between them. For each junction, I decompose the overall income difference in the probability of success and find poor students are disadvantaged in three ways: they start off behind; when they successfully complete a step, they are less likely to proceed to the next step; and missing a step hurts their future chances more than it hurts those of higher-income students. I find the last factor explains a surprisingly large amount of the income gradient. Wealthy students do, in fact, falter on their way to college but they are more likely to graduate from college anyway. Another key result is, conditional on being academically prepared and applying to college, students across the income distribution enroll in four-year colleges at essentially the same rate. Thus, while income plays a large role in college enrollment, it is due almost entirely to differences in academic qualifications and application behavior. This suggests policies attempting to increase low-income college enrollment should focus earlier.

The third chapter examines early commitment financial aid programs, a relatively new category of financial aid that typically enrolls low-income students while still in middle school, guaranteeing them financial assistance for college in exchange for fulfilling certain requirements. By removing uncertainty about college affordability for young students, these programs aim to increase high school preparation and ultimately college attendance. In this chapter, I examine the impact of Indiana's Twenty-first Century Scholars Program, the first public state-wide early commitment program, on high school retention, college enrollment, and Pell Grant receipt. Using a unique dataset with information from the State Student Assistance Commission of Indiana and the U.S. Department of Education's Office of Postsecondary Education, I find positive effects of the early commitment program on all three outcomes. A 10 percentage point increase in program enrollment can be expected to increase high school retention by 2.1 to 2.7 percentage points and Pell Grant receipt by 0.9 to 3.0 percentage points. I find that the program also increased four-year college enrollment by 2.2 to 2.4 percentage points. The results suggest that most scholarships are awarded to students who would not have attended college in the absence of the program. Moreover, the program's effects on college enrollment are considerably larger than the expected effects of grant aid alone. These results suggest that the incentive structure of early commitment programs can increase educational attainment among lower-income students.