Absence of conflict of interest.
Citation
Highlights
- The study’s objective was to examine the impact of the American Dream Demonstration (ADD) on retirement savings contributions.
- The study was a randomized controlled trial that assigned individuals into the treatment or control group. The primary data sources were a baseline and 10-year follow-up survey. Study authors used a statistical model to compare the outcomes between the treatment and control groups.
- The study found no significant differences between the treatment and control groups in the amount of retirement savings, the presence of retirement savings, or the sufficiency of retirement savings.
- This study receives a moderate evidence rating. This means we are somewhat confident that the estimated effects would be attributable to the American Dream Demonstration, but other factors might also have contributed. However, the study did not find any statistically significant effects.
Intervention Examined
American Dream Demonstration
Features of the Intervention
The American Dream Demonstration (ADD) began in the late 1990s with 14 philanthropic funded sites to provide individual development accounts (IDA) to encourage low-income adults to save money. The Community Action Program of Tulsa County in Tulsa, Oklahoma was one ADD site. The program in Tulsa provided participants with financial education, case management, and an IDA at the Bank of Oklahoma. The program targeted individuals who were employed and had a household income below 150% of the federal poverty level.
Features of the Study
The study used a randomized controlled trial to examine the impact of the ADD program on retirement savings. The study was conducted in Tulsa, Oklahoma and included 1,103 participants. Once participants completed the baseline interview, they were randomly assigned to either a treatment or control group. The treatment group received the intervention services: financial education, case management, and an IDA at the Bank of Oklahoma. Study authors were not explicit on the details of what services, if any, that the control group received. The study reports data on the fourth wave of participants who completed both a baseline and 10-year follow up. The analytic sample included 660 participants (311 in the treatment group and 349 in the control group) of whom were primarily white (55%), women (80%), had children under 18 living in the home (81%), and had a mean age of 35 at baseline and an age of 45 years at the 10-year follow up. The authors used statistical models to compare the outcomes between the treatment and control group participants.
Findings
Knowledge and skills for financial decision making
- The study found no significant differences between the treatment and control groups in the amount of retirement savings, the presence of retirement savings, the increase in retirement savings, or the sufficiency of retirement savings to replace 75% of income for one month, six months, or one year.
Considerations for Interpreting the Findings
The study authors state that of the 1,103 participants at baseline, only 855 participants completed the 10-year survey. The authors also excluded 39 participants from the original sample who were over retirement age at the 10-year follow-up and 156 participants who had missing data. This resulted in an analysis sample of 660 participants, a loss of 433 participants (40% overall attrition). Additionally, the authors do not identify how many participants were lost per group. With unknown differential attrition, the study was ineligible for a high evidence rating. However, the authors ensured that the groups were comparable at baseline.
Causal Evidence Rating
The quality of causal evidence presented in this report is moderate because it was a randomized controlled trial with unknown attrition, but the authors ensured baseline equivalence between the groups. This means we are somewhat confident that the estimated effects would be attributable to the American Dream Demonstration, but other factors might also have contributed. However, the study did not find any statistically significant effects.