Detracking Swedish compulsory schools: any losers, any winners?

In this article, I use within-school variation to estimate the effect of tracking. I exploit the fact not only that different tracking policies were practiced simultaneously in Swedish compulsory schools but also that tracking policies changed overtime within schools. I estimate not only if being in a tracked math environment had any effect on the probability of graduating from high school but also if tracking status had any impact on the math grade in high school. The results show that there are no significant average effects of tracking. However, there are effects in the lower part of the grade distribution. Students with a low-educated family background are more likely to fail math at high school if they have attended a compulsory school that practiced tracking compared to similar students in a non-tracked environment.