The “social insurance” hypothesis posits that individuals join cooperative groups and share resources in order to reduce environmental risk. Despite its significance for explaining cooperative groups' formation, in small‐scale and in developing societies, the hypothesis has been subjected to little experimental testing. The present research is designed to examine the relative weight of the motivation for social insurance compared with other psychological motivations for sharing risk. We conducted two studies to test the tendency to share risk under different risk conditions and for groups of different sizes. A third experiment extends the risk‐sharing research to situations involving losses instead of gains. The findings of the first two studies lend strong support to the risk‐sharing hypothesis in the gain domain. For the loss domain, the results of the third experiment demonstrate an intriguing shift from strong reluctance to join groups under lower risk, to ubiquitous readiness to join groups under higher risk. We discuss these results in light of prospect theory and decisions from experience. Copyright © 2014 John Wiley & Sons, Ltd.

No abstract