The optimal timing of school tracking

We develop a simple model which determines the optimal timing of school tracking as the outcome of the trade off between the advantages of specialization, which call for early tracking, and the costs of early selection, which lead to later tracking. We calibrate the model for Germany and study how relative demand shifts toward more general skills and changes in the (exogenous) rate of technical progress affect the optimal tracking time as well as the efficient allocation of students to general and vocational tracks.