Risk Sharing within the Extended Family: Evidence from the Indonesia Family Life Survey

In recent years, there has been an increase in the availability of longitudinal household surveys in developing countries that allow the researchers to construct linkages between households with familial ties. This type of data allows researchers to look at whether some household outcomes are influenced by decisions made at the extended-family level. This article utilizes data from two waves of the Indonesia Family Life Survey (1997 and 2000) to construct a panel of extended families in order to focus on one particular example of such outcome: whether consumption risk sharing occurs among households within the extended family. The findings show that after controlling for the extended family fixed effects, changes in the household's own income still matter for changes in its consumption, rejecting full consumption risk insurance. However, the coefficients on household own income are small, indicating significant degrees of risk sharing. Results from a set of reduced-form estimations suggest that household consumption is affected by characteristics of other households in the extended family. This article contributes to the literature on household surveys by shedding light on how one could define a household in a panel analysis that uses data from longitudinal household surveys in which split-off households are tracked.