Vocational training and general education are the two predominant forms of secondary schooling around the world. Most studies that compare the e?ect of vocational and general education on labor market outcomes in the cross-section su?er from selection bias since less able students are more likely to enroll in vocational programs. This paper exploits a 1973 educational reform in Romania that shifted a large proportion of students from vocational training to general education in order to avoid the bias caused by non-random selection. Using data from the 1992 Census and the 1995-2000 LSMS, we analyze the e?ect of this policy in the context of a transition economy that experienced a decline in manufacturing and a reallocation of labor to new jobs. We find that cohorts a?ected by the policy were significantly less likely to work in manual or craft-related occupations but showed no di?erences in unemployment, non employment, family income and wages than their counterparts who were not a?ected by the policy. We therefore conclude that the cross-sectional di?erences in labor market outcomes between graduates of vocational and general schools are largely a consequence of selection.