Catch-up Effects in Health Outcomes – Linear and Quantile Regression Estimates from Four Countries

Subha Mani, Fordham University
Jere Behrman, University of Pennsylvania
Andreas Georgiadis, University of Oxford

The objective of this paper is to capture the association between nutritional status at young ages and subsequent health. To do this, we estimate a dynamic linear panel data model using data from the three waves of the Young Lives Study. We find that the catch-up coefficient in the linear dynamic panel data model varies between 0 to 0.32, where, Ethiopia and India exhibit perfect catch-up and Peru and Vietnam exhibit partial catch-up in height-for-age z scores. To allow for the catch-up coefficient to vary along the entire distribution of child anthropometric outcomes, we also estimate a dynamic quantile regression instrument variable estimator. We find that the null of homogenous catch-up effects along the entire distribution of anthropometrics can be easily rejected.