Estimating Effective Subsidy Rates of Student Aid Programs
Stacey H. CHEN
Annals of Economics and Statistics,
Every year millions of high school students and their parents in the US are asked to fill out complicated financial aid application forms. However, few studies have estimated the responsiveness of government financial aid schemes to changes in financial needs of the students. This paper identifies the effective subsidy rate (ESR) of student aid, as defined by the coefficient of financial needs in the regression of financial aid. The ESR measures the proportion of subsidy of student aid under the assumption that all students who enroll in college apply for financial aid. Problems of self-selected applications are considerable because more than one-third of financial aid applicants are ineligible for any form of financial aid; and because about 10 percent of financially needy students do not apply. I cope with endogenous applications by estimating a Tobit-with-Selection model using maximum likelihood estimation methods. The results show that conventional OLS estimates considerably understate the ESR because a large portion of applicants are from high-income families with no financial needs and are ineligible for any form of aid. My finding suggests that it is important to include endogenous applications for financial aid in evaluations of student aid programs.