1. How does the receipt of government disaster aid impact the demand for insurance?
While Individual Assistance (IA) grants provide important financial help to those in need after a
disaster, we find that it creates a significant moral hazard effect. Increasing the average IA
disaster grant in a ZIP code by $1,000 reduces the average individual demand for insurance in
that ZIP code by up to $6,000.
2. Does the impact depend on the size of the grant?
Yes. As theory would predict, the higher the grant, the more significant the effect. In fact, we
found that when the grant was on the high end of the distribution (top 75 percent quartile), then the
moral hazard effect could be up to three times larger. Interestingly, when the grant is on the
lower end (lower 25 percent quartile), individuals in that same ZIP code actually purchased more
insurance, probably because they found federal aid to be insufficient to cover their costs.
3. Do people cancel their insurance policy after they received disaster relief grants?
No. We find that free relief mostly has an impact on the quantity of insurance purchased, not the
decision to buy it. Government relief is typically associated with legal requirements to purchase
disaster insurance, and those requirements seem to be well enforced, as least for the years that
immediately follow the disaster.
4. Do all government relief programs have the same effect of creating additional risk taking?
No. This is another important finding. We looked at whether individuals change their insurance
purchase behavior after receiving a low‐interest disaster loan from the SBA and found no
systematic effect. The difference is that one program provides free grants while the other
provides liquidity to victims of disasters to repair or rebuild, but they then have to repay the loan
to the federal government over time with interest.
The question of the future of disaster risk financing has been raised several times after recent
disasters. Here we focus on federal government relief to individuals. A complementary question
is whether the Stafford Act, which guarantees that 75 percent of a state’s disaster losses (after a Presidential
disaster declaration) will be paid by federal taxpayers, also creates moral hazard. While likely, the
size of this effect is a matter of empirical analysis and has yet to be quantified.