July 2013
Program on Education Policy and Governance, Harvard University

Providing the first experimental estimate of the long-term impacts of an offer of a private-school voucher to low-income families, we link data from a privately sponsored voucher initiative, which awarded the scholarships by lottery, to college enrollment information maintained by the National Student Clearinghouse, which we are able to do for 99 percent ofparticipants. Overall, we find no significant effects on college enrollment of the offer of avoucher to attend private school. However, we find evidence of large, significant impacts for African-American students and smaller but statistically insignificant impacts for Hispanic students.