The Returns to Early Education in Germany

This paper looks at the role of pre-primary education in Germany using the German SocioEconomic Panel data set to determine if attending an early education program increases the probability of attending a higher level secondary school at age fourteen. In order to disentangle the effect of the early education program from family specific characteristics, I employ both family fixed effects estimation and quasi-experimental analysis. Initial Ordinary Least Squares (OLS) estimates on my data set show small and positive returns to early education on secondary schooling outcomes. The results of the family fixed effects estimation, however, show that once selection into education has been controlled for, the impact of attending an early education program is small, negative, and insignificant. I test for the presence of selection based on family characteristics and confirm that the OLS estimates are inconsistent since selection based on family characteristics is occurring. My quasi-experimental analysis, based upon a federal policy change in 1996, affirms the negative impact of early education on later schooling outcomes for the child induced into early education by the policy. The lack of positive, significant results, as often predicted in this context, may in part be due to unobserved early education program heterogeneity or intra-household dynamics.