Does the early bird catch the worm? Instrumental variable estimates of educational effects of age of school entry in Germany

We estimate the effect of age of school entry on educational attainment using three different data sets for Germany, sampling pupils at the end of primary school, in the middle of secondary school and several years after secondary school. Results are obtained based on instrumental variable estimation exploiting the exogenous variation in month of birth. We find robust and significant positive effects on educational attainment for pupils who enter school at seven instead of six years of age: Test scores at the end of primary school increase by about 0.42 standard deviations and years of secondary schooling increase by almost half a year.