Introduction

One of the strongest and most consistent findings in development, health and labour economics is the positive relationship between schooling and child health. This empirical relationship has been confirmed in numerous studies across different time periods, countries and measures of child health. These studies generally treat education as a ‘black box’, however. What is measured is not what a person has learned in terms of skills, such as, for example, literacy and numeracy but rather what level or grade has been completed. Two main issues are involved here. First, the link between schooling and child health really goes from schooling to skills to productivity to child health. As the link between schooling and skills is more tenuous in developing countries, due often to poor school quality, it is imperative that this part of the process receives particular attention in empirical analyses in this context. Secondly, policies focusing on education rather than on skills might be misdirected. With multiple paths to achieving skills (including formal education and adult literacy programmes) and with limited public budgets, cost-effectiveness of programmes is essential.

In response to these issues, I suggest that literacy, numeracy and other skills be viewed as intermediate outputs in a production process where the main inputs are formal (child) schooling and non-formal (adult) literacy course attendance. Subsequently, literacy, numeracy and other skills enter as inputs in a production process to generate the final outputs of child health.