This paper presents asymptotic theory and Monte-Carlo simulations comparing maximum-likelihood bivariate probit and linear instrumental variables estimators of treatment effects in models with a binary endogenous treatment and binary outcome. The three main contributions of the paper are (a) clarifying the relationship between the Average Treatment Effect obtained in the bivariate probit model and the Local Average Treatment Effect estimated through linear IV; (b) comparing the mean-square error and the actual size and power of tests based on these estimators across a wide range of parameter values relative to the existing literature; and (c) assessing the performance of misspecification tests for bivariate probit models. The authors recommend two changes to common practices: bootstrapped confidence intervals for both estimators, and a score test to check goodness of fit for the bivariate probit model.
This paper estimates average and marginal returns to schooling in Indonesia using a non-parametric selection model estimated by local instrumental variables, and data from the Indonesia Family Life Survey. The analysis finds that the return to upper secondary schooling varies widely across individual: it can be as high as 50 percent per year of schooling for those very likely to enroll in upper secondary schooling, or as low as -10 percent for those very unlikely to do so. Returns to the marginal student (14 percent) are well below those for the average student attending upper secondary schooling (27 percent).
Data from three rounds of nationally representative health surveys in India are used to assess the impact of selective mortality on childrens anthropometrics. The nutritional status of the child population was simulated under the counterfactual scenario that all children who died in the first three years of life were alive at the time of measurement. The simulations demonstrate that the difference in anthropometrics due to selective mortality would be large only if there were very large differences in anthropometrics between the children who died and those who survived. Differences of this size are not substantiated by the research on the degree of association between mortality and malnutrition. The study shows that although mortality risk is higher among malnourished children, selective mortality has only a minor impact on the measured nutritional status of children or on that status distinguished by gender.