The NREGS is an ambitious public works program intended to provide a basic safety net to the rural poor in India. Institutionalised in 2005, it is currently the largest public works program in the world. Despite its scale and the political importance attached to the program, its success in targeting and its impact on participating households has not been well studied. This paper attempts to study two aspects of the program's functioning using data from the Young Lives longitudinal Study conducted in Andhra Pradesh. Firstly it looks at the targeting of the program and the characteristics of those who self select into it. We find that poorer and lower caste households are more likely to register as are those affected by drought. We also find that having more than 5 influential relatives increases the probability of registration by 10.3 percentage points. We next attempt to estimate the impact of program participation on the children in participating households, looking specifically at anthropometric scores as indicators of health outcomes, and the incidence of child labour. While there seems to be a positive correlation between program participation and health outcomes, this does not remain robust across specifications. On the other hand we find that program registration reduces the probability of a boy entering child labour by 13.4% points and program take up reduces it for girls by 8.19% points.

We find that the targeting efficiency of the program seems to be largely effective and it seems to offer a viable security net for households with variable employment opportunities. It also seems to have an important effect on children, further strengthening the program's significance.