We estimate the long-run cointegrating relationship between schooling and income inequality using annual data from 48 contiguous U.S. states between 1946 and 2000. Our study contributes to the literature in several ways in terms of data and empirical methodology. First, we take into account integration and cointegration properties of the data and estimate the cointegrating relationship between schooling and income inequality using Fully Modified OLS (FMOLS) following Pedroni (2000, 2001). Second we investigate the direction of the causality. Using several different measures of income inequality, we find that an increase in the average level of schooling decreases income inequality in the long-run and (Granger) causality runs from schooling to income inequality, not the other way around.