We derive testable implications of model in which first best allocations are not achieved because of a moral hazard problem with hidden saving. We show that in this environment agents typically achieve more insurance than that obtained under autarchy via saving, and that consumption allocation...
