We provide evidence on the long-run impact of vouchers for private secondary schools, evidence collected twenty years after students applied for the vouchers. Prior to the voucher lottery, students applied to either an academic or vocational secondary school, an important mediating factor in the vouchers’ impacts. We find strong tertiary education and labor market effects for those students who applied to vocational schools with almost no impact on those who applied to academic schools. The labor market gains for vocational students are strongest at the top of the distribution and null at the bottom of the distribution. We find additional long-run impacts on consumption, and teen-age fertility. The expected net present value of benefits to participants and to taxpayers was large and positive implying that the program was welfare improving unless net externalities were large and negative
