Job Market Paper

Social interactions among low-income households and female labour force participation

Abstract

I use a unique dataset on detailed social networks at neighbourhood level to empirically analyse how social interactions among low-income women affect their labour market behaviour. Specifically, I study neighbourhoods, as they constitute a natural form of social interactions, to quantify the causal social effect of female neighbours’ behaviour on individuals’ decisions regarding participating in the labour force. To do so, I follow the literature on discrete choice with social interactions and extend a standard model of female labour force participation by including in the utility function the choices of other women within the neighbourhood. In order to identify social effects, I implement a novel strategy proposed in the latest literature on social interactions. In particular, I exploit the geographical location of households to identify neighbours of neighbours who are not closest neighbours, and their characteristics are used to construct instruments for the behaviour of closest neighbours. The results indicate the presence of significant and positive endogenous social effects on women’s decision to participate in the labour force. The estimated endogenous social effect in the preferred specification is positive and statistically significant and suggests that an increase of one percentage point in the female labour force participation rate of closest neighbours increases the individual probability of participation in the labour force by 10.5 percent.