Public in-Kind Relief and Private Self-Insurance

摘要

This paper provides a new angle on the question of crowding effects of publicpolicies. We examine how non-hypothetical self-insurance behavior by households responds tovariations in public investments in relief capabilities based on a large disaster preparednesssurvey (n = 19,071) conducted in Japan in 2012. In our specific setting which looks atemergency drinking water, (i) government provides in-kind, rather than cash, relief and (ii)the crowding effect observed is more apt to be total, rather than partial. In contrast to much ofthe literature studying crowding effects of cash relief, we find little evidence for crowding outin emergency drinking water, with an upper bound of 2% at the intensive margin. We alsoidentify important benefits of targeting in-kind relief to households with minors.