Split-Sample Instrumental Variables Estimates of the Return to Schooling

Two-stage least squares is biased in the same direction as ordinary least squares even in very large samples. The authors propose a split-sample instrumental variables estimator that is not biased toward ordinary least squares. Split-sample instrumental variables uses one-half of a sample to estimate parameters of the first-stage equation. Estimated first-stage parameters are then used to construct fitted values and second-stage parameter estimates in the other half sample. Split-sample instrumental variables is biased toward zero but this bias can be corrected. The authors use split-sample estimators to reexamine instrumental variables and two-stage least squares estimates of the returns to schooling.