Social capital and savings behavior of the poor: evidence from the field

Abstract

Two thousand and fifty six Senegalese clients of a Microfinance Institution (MFI) participated in the experiment during four months. They were divided in three groups: a control group that did not receive any messages, and two treatment groups, one in which clients received generic messages that simply said they should save money, and another group in which clients received messages that referred to the savings behavior of individuals who live in their neighborhood. The goal of this study is to assess whether receiving additional information about other participants from the same area where the participants live (neighborhood social capital) affects savings behavior. The results of this study show no significant impact of generic messages on savings behavior. Social capital effectively encourages deposits, which also makes withdrawals more salient since the accounts are fully liquid, but only among female clients. Limited attention, psychological reactance, salience, and gender issues emerge as possible explanations.