Using new and unique panel data, we investigate the role of long-term health and childhood malnutrition in schooling outcomes for children in rural India, many of whom lack basic numeracy and literacy skills. Using data on students’ performance on mathematics and Hindi tests, we examine the role of the endogeneity of health caused by omitted variables bias and measurement error and correct for these problems using a household fixed effects estimator on a sub-sample of siblings observed in the data. We also present several extensions and robustness checks using instrumental variables and alternative estimators. We find evidence of a positive causal effect of long-term health measured as height-for-age z-score (HAZ) on test scores, and the results are consistent across several different specifications. The results imply that improving childhood nutrition will have benefits that extend beyond health into education.
CSAE (2010) CSAE WPS/2010-14, Centre for the Study of African Economies, Oxford, UK, 59 pp.