Is there Complete , Partial or No Recovery from Childhood Malnutrition ? - Empirical

The long term consequences of malnutrition during childhood are well established in the literature. Yet, little is known about the extent to which individuals are able to recover from some of the deficits in health outcomes caused by childhood malnutrition. In order to address this question, I construct a panel data for children between 3-59 months in 1993 and follow them through 1997 and 2000 waves of the Indonesian Family Life Survey. I estimate a dynamic conditional child health demand function to capture the association between malnutrition during childhood and future health status. Variants of the IV/GMM estimation strategy are used in this paper to deal with the endogeneity problems encountered in estimating a dynamic health demand function. I find that an upward bias in the estimated coefficient on the one period lagged health status is eliminated as I move from an OLS estimation strategy to a first-difference GMM strategy. The coefficient on lagged health status reduces from 0.53 using OLS to 0.26 following a firstdifference GMM strategy. A coefficient of 0.26 on the one period lagged health status indicates that malnutrition during childhood does not manifest into poor health status in the future i.e. there exists partial “catch-up effects”. I show that the first-difference GMM strategy adopted in this paper addresses biases arising from both omitted variables and measurement error in data. In addition, the GMM strategy adopted here also relies on much weaker stochastic assumptions as compared to earlier work in the literature. I also find some evidence to support that catch-up effects differ by age i.e. younger children are likely to experience higher catch-up effects as compared to their older cohorts.