The benefits and costs of alternative strategies to improve educational outcomes

Few empirical relationships have been investigated more frequently than that between years of schooling and earnings. Hundreds of studies using a wide variety of datasets from developed countries, spanning many decades, and employing alternative specifications to correct for various potential sources of bias, have consistently found positive private returns per year of schooling.¹ Returns are frequently equal to or above long-run average market returns to other investments. Estimated returns to schooling in developing countries have been comparable in magnitude to returns found in developed countries. Table 4.1 presents ordinary least squares (OLS) estimates of returns from a standard Mincerian earnings function applied to sixty-three household datasets from forty-two developing countries. The results are presented separately for males and females and for urban and rural residents. These datasets were selected because the variable definitions could be harmonized across countries and because separate returns could be estimated for men and women and for urban and rural residents.² The same model was estimated for all countries so that the variation is not due to specification choice. Several interesting results are apparent. First, private returns, estimated as the percentage increase in annual earnings obtained from an additional year of schooling, are almost universally positive. In only one case for women, four cases for men, three for urban residents and two for rural residents did education fail to raise earnings.