The structural relationship between nutrition, cognitive and non-cognitive skills: evidence from four developing countries
Año | : | 2013 |
---|---|---|
Autor/es | : | Alan Sánchez |
Área/s | : | Educación y aprendizajes |
[2013] SANCHEZ, Alan. The structural relationship between nutrition, cognitive and non-cognitive skills: evidence from four developing countries. Oxford: Young Lives, 2013. Working paper, 111.
Both cognitive and non-cognitive skills are rewarded in the labour market. However there is little evidence about how these abilities are formed in the context of developing countries. I study the way in which cognitive and non-cognitive skills are simultaneously acquired in the transition from childhood to adolescence using longitudinal data from four countries: Peru, India, Vietnam and Ethiopia. I estimate a linearized version of the technology of skills formation, linking inputs observed at 7 to 8 years to outputs observed at 11 to 12 and 14 to 15 years. I find evidence of selfproductivity mainly for cognitive skills and cross-productivity for both types of skills. I then extend the technology of skills formation to account for the role of nutritional status in the acquisition of skills. A child’s height-for-age Z-score is used as a proxy of nutritional status. To deal with the endogeneity of nutrition, this variable is instrumented using a set of selected household shocks that are country-specific. Height-for-age is found to be a relevant input in the skills formation model, having a direct as well as an indirect effect on skills accumulation. To obtain estimates of the long-term impact of nutritional investments during the early childhood period on later abilities, I use evidence gathered from a second model that links early height-for-age to cognitive ability at 7 to 8 years. Linking results from both models, I find that an increase of 1 standard deviation in early height-for-age tends to increase cognitive skills during adolescence by 6%, 9%, 17% and 7% in Peru, India, Vietnam and Ethiopia, respectively. It also increases non-cognitive skills by 2% and 4% in India and Vietnam, respectively. Finally, I test the notion of complementarity and find that the rate of return of cognitive skills is considerably lower for children that were stunted in mid childhood, whereas the rate of return of non-cognitive skills remains unchanged.