Assessing and Managing Hurricane and Flood Risks through Economic Incentives and Insurance Contracts

Principal Investigator: 
Other Researchers: 
Erwann Michel-Kerjan
Performance Period: 
October 2009 to September 2010
Commercialization Status: 
N/A
Abstract: 
1. New era of catastrophes   Today, trillions of dollars of assets are exposed to hurricane risks in coastal states from Texas to Maine. Florida constitutes the nation’s area must vulnerable to hurricane risks reflecting its growth in population, from 2.8 million in 1950 to 18.7 million in 2010, an increase of almost 600 percent. As of December 2007, insured exposure on the coast of Florida was $2.4 trillion. New York also has coastal exposure of approximately $2.4, and there is nearly $900 billion of exposure on the coast of Texas.   2. Focus of this study   As Florida is subject to the most severe losses from the hurricane risk with accompanying water damage from flooding and storm surge, this study focuses on this state.  We will examine the roles that risk assessment, building codes and behavioral biases play in individuals’ and businesses’ decisions as to whether to invest in loss reducing (mitigation) measures, and then suggest new approaches to managing the risk.  We will also briefly summarize the implications for other hurricane-prone states, three of which were studied in detail for hurricane risk (New York, Texas and South Carolina).   3. Role of risk assessment   While catastrophes are often characterized as low-probability, high-consequence events, the data suggest that they are occurring with a much higher frequency than in the past. Using outputs from catastrophe models provided by Risk Management Solutions, we determined aggregate residential losses associated with a series of hurricane scenarios in Florida and three other states (New York, South Carolina and Texas).  This analysis reveals a 15 percent annual probability of an insured loss of at least $10 billion in Florida and a 5 percent annual probability that insured losses in that state will exceed $25 billion.  The insured losses from hurricanes in the postal zones within Miami-Dade County were compared with forty-six counties in the northern part of Florida that had the same expected annual insured losses as Miami-Dade County ($900 million) but a different variance. The standard deviation of losses was $1.4 billion higher in Miami-Dade County ($4.2 billion) than it was for the forty-six northern counties combined ($2.8 billion). This implies that there is a much greater likelihood that residents in Miami-Dade will experience a hurricane with catastrophic consequences than their counterparts in the northern part of the state.   4. Impacts of building codes on risk reduction   The adoption of building codes significantly reduces damage from hurricanes.  Based on a sample of over 5,600 homeowners affected by Hurricane Charley in Florida in 2004, residences built under the wind-resistant standards that were enforced in 1996 had a claim frequency that was 60 percent lower than those that were built pre-1996. If all homes in Florida were assumed to meet the current building codes, the reduction in damage from hurricanes with return periods of 100 years and 500 years ranged from 61 percent to 52 percent. For a 100-year hurricane, mitigation would reduce the potential losses by 44 percent in South Carolina, 39 percent in New York, and 34 percent for Texas.       5. Water damage from hurricanes and inland flooding   Hurricanes also have the potential to create severe water damage from flooding and storm surge, both on the coast and inland. This was the case in Florida where over 80 percent of the state’s flood claims in 2004 and 2005 (two record-high years) were caused by storm surge. The amount of flood insurance sold through FEMA’s National Flood Insurance Program (NFIP) has increased significantly in recent years: as of December 2008, the NFIP covered $1.2 trillion in assets nationwide. But the flood insurance market is highly concentrated, with Florida having more than 40 percent of the entire number of NFIP policies-in-force. Around 70 percent of these policies are located in five states—Florida, Texas, Louisiana, California, and New Jersey. Mitigation is effective in dealing with the hurricane risk.  An analysis of 45,000 flood claims paid by the NFIP in Florida between 2000 and 2006 reveals that flood risk mitigation can significantly reduce the level of claims. For example, elevating the house reduces flood claims by an average of 77 percent across the state during this period.   6. Lack of interest in mitigation  Empirical data reveals that many residents in hazard prone areas do not invest in cost-effective risk reduction measures voluntarily. They are thus exposed to much greater losses from future disasters than they would experience had they invested in these measures prior to a disaster. Given that many homeowners are underinsured or not insured at all, the general taxpayer is likely to bear some of the costs of disaster relief.   7. Lack of adequate flood insurance protection Empirical data also reveals that many residents in hazard prone areas do not have flood insurance coverage even though coverage in flood-prone areas is required as a condition for a federally insured mortgage. Based on an analysis of the portfolio of the National Flood Insurance Program (NFIP) nationwide from 2001 to 2009, we show that on average new flood insurance policies lapse after only 3 years. This finding is robust over time, across states and levels of risk. Some individuals move elsewhere, some underestimate the risk or are told they are safe, and others think the government will bail them out in case of a disaster loss, so why purchase insurance.   8. Proposed Innovation: Long-term insurance contracts   We propose concrete solutions to overcome the behavioral biases impeding residents’ investment in mitigation measures and purchase of proper flood insurance. These include the following measures: (1) moving from the traditional one-year insurance contract to long-term insurance coupled with long-term home improvement loans; FEMA can play a critical role here by having the NFIP start offering long-term flood insurance, (2) having sellers or buyers of new or existing homes obtain a seal of approval from a recognized inspector that the structure meets or exceeds building code standards, (3) having communities encourage residents to pursue mitigation measures by providing them with tax incentives, and (4) developing zoning ordinances that better communicate risk.   9. Future Research   The concluding section suggests a set of guiding principles for developing long-term flood insurance, as well as the impact that climate change is likely to have on the pricing of these contracts. These areas are ones we have discussed with FEMA/DHS, Congress, climate scientists and other interested parties in the past few months.