The Effects of Cash Transfers on Social Preferences

2021-04-12

2021-09-30

Participants’ children in each of the three groups (treatment, spillover, and pure control) will be surveyed and will complete several economic games. The surveys will measure self-reported social behavior, with modules measuring specifically trust, envy, depression, social norms, social ties, perceptions about sharing and redistribution, locus of control, and IQ. The games will be standard economic games that have been modified to be appropriate for children. In the games, the children will get limited information on the other player, namely their age and gender. In some rounds of the games, children will also know the other player’s family household wealth relative to their own family’s wealth. By comparing the behaviors of the children across the three groups, we are able to assess how cash transfers to parents affect children’s prosocial behavior. And by comparing the behaviors within-child across different kinds of partner identities, we are able to assess how prosocial behavior depends on the current distribution of wealth.
In addition, we will resurvey adults, asking them questions about asset ownership, household consumption, and individual well-being. These questions will allow us to assess the persistence of the cash transfers impact on these outcomes for each of the children’s households.

We describe the analysis plan for a lab-in-the-field study that will examine the effects of unconditional cash transfers on children’s social preferences. The study will be conducted with children living in Nakuru County, Kenya whose parents participated in a previously implemented RCT that provided cash transfers to poor households. Participating households either received a cash transfer, did not receive a transfer but lived in the same village as those who did, or did not receive a transfer and lived in a village in which nobody received a transfer. Participants will be surveyed and will complete several standard economic games measuring their social preferences. For these tasks, they will be paired with individuals of varying relative wealth to assess their social preferences and how social preferences vary with the identity of the other player.

Randomization was completed by the previous study for the Unconditional Cash Transfer treatment (Haushofer,Johannes, Robert Mudida, and Jeremy P Shapiro, “The Comparative Impact of Cash Transfers and a Psychotherapy Program on Psychological and Economic Well-being,” NBER Working Papers, 2020.). Randomization for the lab experiments (e.g. characteristics of the other player) are randomized by a tablet.

The treatment from Haushofer, Mudida, and Shapiro (2020) was randomized at the village level. We randomize for each game a child plays the characteristics of the other player.

No

120 villages

We aim to visit 4,424 children within our target age range of 6 years old to 17 years old in 1,804 households. This will correspond to 906 children in the treatment households, 1,446 children in the spillover households, and 2,072 children in the pure control households.

We aim to visit 4,424 children within our target age range of 6 years old to 17 years old in 1,804 households. This will correspond to 906 children in the treatment households, 1,446 children in the spillover households, and 2,072 children in the pure control households.

The Effects of Cash Transfers on Social Preferences