Association for Public Policy Analysis & Management

Student loan cohort default rates are among a small number of metrics that are used in the system of accountability that determines institutional eligibility in the Federal Student Aid program.  However, we know relatively little about what drives variation in this measure both across institutions and across time.  In this paper I use a constructed data set to carry out a regression analysis that measures the extent to which variation in observed cohort default rates can be explained by fluctuations in macroeconomic conditions and borrower characteristics.  Covariates that capture information about students and the macro economy will be used to generate institution level adjusted default rates that allow for meaningful comparisons across institutions.  The constructed data set matches historical records on institution cohort default rates with institution characteristics from the Integrated Postsecondary Education Data System (IPEDS) and economic indicators including GDP growth and unemployment. This analysis sheds light on the important question of whether cohort default rates are an appropriate tool for measuring institution quality and provides insights for designing a better system.