This paper investigates the relationship between economic development and the average Level of education as well as the degree of inequality in the distribution of education, respectively. Approaching this question in a dynamic panel over 60 years and 143 countries with a system GMM estimator reveals strong support for the inclusion of an interaction term between the education Gini coeffcient and average years of schooling, indicating the existence of nonlinear effects. We contribute to the literature in providing strong evidence that more schooling is good for economic growth - irrespective of its distribution - but that the coeffcient is variable and substantially declining in inequality. On the other hand, inequality is positively related to economic growth for low average levels of education, whereas highly educated countries exhibit a statistically insignificant negative relationship between inequality and economic growth. From this it follows that at least a slight increase in the degree of inequality is necessary in order to haul initially poor and low educated economies out of the poverty trap. However, as economies become educated, the effect of educational inequality mainly works indirectly. Accordingly, countries that show greater educational inequality experience lower macro economic returns to education than more equal economies, on average. (authors' abstract) / Series: Department of Economics Working Paper Series

This paper investigates the relationship between the level and the distribution of education and economic development. We contribute to the literature by introducing an interaction term between the education Gini coefficient and average years of schooling. In a dynamic panel over 55 years and 134 countries we provide, on the one hand, strong evidence that more schooling is good for growth, but the coefficient is variable and substantially declining in the degree of inequality. The aggregate benefit to education thus depends on a country's position in the education distribution. On the other hand, we find a slight transitional increase in education inequality to be beneficial at a very low average level of schooling, but detrimental for growth at a relatively high average level. Allowing for the macroeconomic return to education to be heterogeneous with respect to the degree of inequality is therefore paramount in understanding the relationship between education and development. (authors' abstract)