We estimate the effect of social capital on the ability of households to insure consumption after unexpected negative shocks. Many theoretical models argue that strong ties to extended family members and to one’s community help protect families when an adult becomes ill or disabled. Using household-level longitudinal data on Indonesian families, we test whether consumption declines less after a negative health shock for those with many ties to their community and for those in a community with dense ties. We take advantage of a particularly rich set of measures of social capital including measures of civic participation; the existence of traditions of mutual cooperation; ethnic and linguistic diversity; and long-term relationships in the community. We also examine the role of a large and prosperous extended family. We find little support for the hypothesis that social capital is the capital of the poor.