Methods: We use data from the American Dream Demonstration Experiment (ADD), a longitudinal, randomized IDA field experiment. Participants were individuals with household incomes below 150% of the federal poverty level, who were randomly assigned to either the treatment group (access to IDAs) or a control group (restricted from participation in any matched savings program). This study combines data from four time points: baseline, 18 months, 48 months (the end of the program), and 10 years after program enrollment. Using probit regression, we assess the impact of treatment assignment on the probability of enrollment and degree attainment in educational programs net of important covariates. We use a conservative intent-to-treat approach to ensure the highest reliability in our findings. Finally, we compare treatment effects on the educational outcomes between sub-groups of the sample defined by age, initial level of education, gender, race, and socio-economic status.
Results: Findings suggest that assignment to eligibility for an IDA may be associated with positive educational outcomes. People randomly assigned to IDA treatment were more likely, ten years after assignment, to have enrolled in an educational program than those assigned to the control group. There does not appear to be a significant impact of treatment on increase in level of educational credential.
Conclusions and implications: These unique data offer the opportunity to assess both the short- and long-term impact of the incentive for low-income adults to engage in further educational activities. These findings contribute to our understanding of the long-term impact IDAs have on the lives of participants and how assets can shape the opportunity structure faced by lower income populations in the United States. They also suggest potential intersections between policies focused on building assets and policies focused on expanding the educational attainment and human capital of lower-income people.
