This paper provides empirical evidence of advantageous selection in insurance markets. By using a novel insurance setting where moral hazard is not a concern, I am able to overcome an important obstacle in most studies of selection: the inability to distinguish moral hazard from selection. In the US market for area yield crop insurance, payouts are based on average county yields. Moreover, area yield insurance is only offered in counties where no farmer is large enough to affect the mean yield. I find that area yield insurance takeup is higher when average yields in the county are higher and show that this effect is not being driven by prices. This suggests that the net selection into these plans thus favors insurance providers, not buyers. One possible mechanism is that providers have better information about aggregate yields. Another is that the desirability of other, non-area yield, insurance options changes, a potentially important but previously overlooked driver of selection.
展开▼
