Association for Public Policy Analysis & Management

Low-income students attend college at lower rates than high-income students, and many students who do attend college default on their federal student loans (Blagg, 2018; Scott-Clayton, 2018). Promise programs, which often provide significant college subsidies, could both induce additional students to attend college and affect students’ debt levels. In this paper, we use synthetic control methods to examine how a variety of Promise programs affect college enrollment, college graduation, and student debt. This is one of the first studies to examine the effects of Promise programs by program characteristics, as well as one of the first to study the effects of Promise programs on student finances after college.

We use data from the College Board, the National Student Clearinghouse (for college enrollment and graduation), and TransUnion credit reports (for students’ credit scores, student debt, credit card debt, auto debt, and home debt). We consider students to be treated if they attend high school in any district that the W.E. Upjohn Institute for Employment Research has identified as having a Promise program (Miller-Adams, Hershbein & Timmeney, 2017), and construct the counterfactual using synthetic control methods (Abadie, Diamond & Hainmueller, 2010). We include students in the high school cohorts of 2004 – 2017 (for college enrollment), 2004 – 2012 (for college graduation), and 2004 – 2010 (for financial outcomes). We estimate program heterogeneity by college type included in the program (e.g. two versus four year), program intensity (i.e. subsidy amount), and student characteristics (e.g. predicted income).