Remittances and child labor in Pakistan : a tale of complementarities

We investigate the impact of remittance income on the household decision to send a child to work. Using data from a Multiple Indicators Cluster Survey conducted in the Punjab province of Pakistan in 2014, we isolate the causal impact of remittance income by employing statistical matching to construct counterfactuals that allow us to compare the occurrence in child labor in comparable households that differ solely in their access to remittances. We find that remittances have in general failed to mitigate household reliance on child labor in Punjab. However, the impact depends critically on whether remittances originate from within Pakistan or outside, the age and gender of the child, and on the nature of employment. Specifically, internal remittances increase the labor force participation of the youngest children in the 5-11 age group, with girls being more likely to work in household production and boys being additionally more likely to work as wage labor in nonhazardous occupations. By contrast, international remittances impact the oldest children in the 14-17 age group. While girls in this age group participate more in household and nonhazardous market production, boys are additionally more likely to participate in hazardous activity.