Dustmann, Christian Puhani, Patrick A. Schönberg, Uta
Year of Publication:
IZA Discussion Paper 7897
Despite its efficiency in tailoring education to the needs of students, a tracking system has the inherent problem of misallocating students to tracks because of incomplete information at the time of the tracking decision. This paper investigates the effects of attending a more advanced track in middle school on long-term education and labor market outcomes for Germany, a country with a very rigorous tracking system where the risk of misallocating students to tracks is, due to the early age at which tracking takes place, particularly high. Our research design exploits quasi-random shifts between tracks induced by date of birth, and identifies the long-term effects of early track attendance for a group of marginal students most at risk of misallocation. Remarkably, we find no evidence that for these students, attending a more advanced track leads to more favorable long-term outcomes. We attribute this result to the up- and downgrading of students between tracks after middle school when more information about their potential is available. Overall, our findings underscore that flexibilities built into a tracking system, which allow students to revise initial track choices at a later stage, effectively remedy even a prolonged exposure to a less advanced school environment.
tracking school quality peer effects regression discontinuity design