SOEPpapers on Multidisciplinary Panel Data Research 598
We analyze the effect of education on wages using German Socio-Economic Panel data and regional variation in mandatory years of schooling and the supply of schools. This allows us to estimate more than one local average treatment effect and heterogeneous effects for different groups of compliers. Our results are in line with previous studies that do not find an effect of compulsory schooling on wages in Germany. We go beyond these studies and test a potential reason for it, namely that basic skills are learned earlier in Germany and additional years of schooling are not effective anymore. This is done by also estimating the effect of education on cognitive skills. The results suggest that education after the eighth year does not seem to have a causal effect on cognitive skills in Germany. This is consistent with the explanation for zero effects of schooling on earnings.