Abstract:
Using 182 estimates from 140 studies in 55 countries, this paper compares ordinary least squares (OLS) and instrumental variables (IV) estimates of the private returns to schooling. IV returns average 9.7 percent–38 percent higher than OLS–and exceed OLS in nearly 80 percent of cases, with the largest gaps in developing countries. These patterns align with theories of diminishing marginal returns, scarcity rents, and attenuation from measurement error. While IV methods mitigate bias, instrument validity and external validity concerns persist. Evidence consistently shows substantial causal returns, particularly for disadvantaged populations, underscoring the need for rigorous research.
