The Pell and the Poor: A Regression-Discontinuity Analysis of On-Time College Enrollment

Abstract

In this study, I capitalize on a natural experiment to examine the effect of federal Pell Grant eligibility on college enrollment for students who graduated from high school in spring 2004 and who completed the Free Application for Federal Student Aid (FAFSA). Data from the Education Longitudinal Study of 2002 are used to array students on the continuous Expected Family Contribution (EFC) variable, attained from the FAFSA, and divide them into two categories—those with an EFC at or below the federally designated amount for Pell eligibility, and those above. This regression-discontinuity approach allows me to measure whether students who vary regarding their Pell Grant eligibility, but who have essentially identical family incomes, differ in their college enrollment rates. I find no effect. While confirming prior studies examining the impact of the Pell Grant on college enrollment for traditional students, this finding remains inconsistent with findings on other need-based grants. This result suggests that the minimum Pell Grant amount, currently less than 10% of an average college’s tuition and fees, may be insufficient.