Limited insurance within the household: evidence from a field experiment in Kenya

In many developing countries, unexpected income shocks are common, formal insurance is absent, and informal inter-household risk-sharing networks are unable to provide full insurance. An important question is therefore whether risk sharing within the household is effective. I conducted a field experiment in Western Kenya in which 142 married couples were followed for approximately 8 weeks. Every week, each individual had a 50% chance of receiving an income shock equivalent to a few days' income. Since these shocks are, by definition, small relative to lifetime income, they should not affect intra-household bargaining power and should only affect a Pareto efficient household through the pooled budget constraint. However, I find that men increase their private consumption when they receive the shock but not when their wives do, a rejection of efficiency. I present evidence that such behavior is not specific to the experiment - both husbands and wives spend more on themselves in weeks in which their labor income is higher. The results suggest that insurance is limited even within the households in this sample.