Racial Disparities in Student Debt: Evidence from a National Investigation of Low- and Moderate-Income Households

Background/Purpose

Student debt remains central to ongoing research and policy on higher education. Recent evidence indicates that minority and lower-income students incur disproportionate levels of student debt compared to their white and upper-income peers. Furthermore, the accumulation of student debt may more negatively impact minority and lower-income students, contributing to higher dropout rates, loan default, and diminished post-college returns.

Despite some research on this topic, little is known about the extent of racial gaps in student debt among exclusively low-income populations.

This study employs propensity score matching and two-part modeling to test whether low- and moderate-income (LMI) minority students, particularly black students, are at heightened risk of utilizing education loans. We also examine whether loan utilization among LMI black students heightens this population's risk of accumulating student debt in excess of their LMI white peers.

In light of recent state- and federal-level reductions to student aid programs most beneficial to these students, this study's findings are on-par with evidence-based policy considerations. Additionally, our methodological approach advances our field’s scientific inquiry in the context of social inequality.

Methods

Data and sample. This study uses a national sample of LMI households (N = 17,684) from the Household Financial Survey (HFS) of the Refund to Savings initiative.

Measures and analysis. Outcome measures of student debt include a dichotomous indicator of education debt acquisition and a continuous measure of education debt amount. The main predictor variable is a categorical measure of race/ethnicity. Additional educational variables include enrollment status (enrolled or not) and educational attainment (less than high school tio graduate/professional degree). Other covariates include demographics such as gender, age, nativity status, gross income, number of assets, and credit card or other liabilities.

We perform analyses in two stages to probe robustness of results and disentangle confounding socioeconomic risk factors. First, we estimate a two-part model to account for a large number of zero observations in the dependent variable (student debt amount). Second, we implement a propensity score matching estimator to match respondents based upon observed demographic and financial characteristics.

Results. Results of the two-part model indicate that black students are two times more likely to have student debt and accumulate $7,175 more than white students. When degree status is considered, we find that these significant disparities persist. Among equally educated students with a college degree, black graduates are predicted to incur $6,975 more in debt than white graduates. Propensity score analyses extend this finding, indicating that LMI black students accumulate $6,482 more in student debt.

Conclusions/Implications

Using robust methods, this study provides some of the first evidence which suggests that LMI black students accumulate significantly more debt than do white counterparts, even after accounting for race- and debt-related confounders. While college access has increased, the notion that a college degree comes at a higher price for LMI minority students raises questions about higher education as a vehicle of guaranteed social and economic mobility and the possible consequences of student debt for LMI minorities. We discuss evidence-based policies to continue the enhancement of minority access to college, while promoting equitable outcomes after graduation.