Drop-out from Individual Development Accounts: Prediction and Prevention

Individual Development Accounts (IDAs) are a new policy instrument designed to help the poor save and accumulate assets. IDAs provide matches for savings used for home purchase, post-secondary education, or microenterprise. IDAs cannot help participants, however, if they drop out. What determines drop-out, and what can be done to help participants to stay in? Three findings emerge from an analysis of IDAs in the American Dream Demonstration. First, drop-out depends more on transaction costs and previous debt than on income. Second, program design—and match rates in particular—affect drop-out risk. Third, drop-out can be predicted with some accuracy, so IDA programs could use statistical targeting to identify candidates for special preventive attention before they drop out. Acknowledgments The authors are grateful for help from Constance Dunham and Timothy Bates and from Margaret Clancy, Lissa Johnson, Jami Curley, Michal Grinstein-Weiss, Min Zhan, Sondra Beverly, and Suzanne Fragale. We are also thankful for cooperation from the Corporation for Enterprise Development and from the host organizations in the American Dream Demonstration. Some financial support for this research came from the Division of Asset Building and Community Development of the Ford Foundation.