Health insurance lowers the out-of-pocket price of healthcare, and it is well-established that this leads to higher utilization of care. This type of “moral hazard” is typically viewed as a social cost of insurance. Within a standard model, we show that there are two important ways in which the consumer’s ability to change her behavior in response to insurance can play a central role in the ability of insurance to protect her from risk. These are (i) by allowing optimal exploitation of real income gained in bad states, and (ii) by enabling more resources to be shifted from good states to bad states than otherwise could be. We provide a theoretical characterization of these cases and quantify their importance empirically. Under standard parameterizations of demand for healthcare and health insurance, estimates in the literature imply that moral hazard accounts for as much as half of the total value of risk protection derived from insurance. Preventing consumers from changing their behavior in response to insurance would lower healthcare spending, but also result in a major loss of risk protection, on-net reducing social and consumer welfare in the population we study. Our results suggest that under-utilization of healthcare may thus be an equally important threat to welfare as over-utilization.