Since Sandy made landfall as a post-tropical cyclone on October 29, 2012, in southern New Jersey, AIR has been analyzing the storm’s characteristics and impacts. Informed by the latest available information on surge height and extent from the USGS, surface wind speed observation data, and findings from AIR’s post-disaster survey teams,
AIR now estimates that insured losses from Sandy will be between USD 16 and USD 22 billion.
AIR estimates include wind and storm surge damage to onshore residential, commercial and industrial properties and their contents, automobiles, and time element coverage (additional living expenses for residential properties and business interruption for commercial properties).
Note that AIR clients can download a variety of resources from the Downloads page of the ALERT website, including a document that details the assumptions used in generating the loss estimates, as well as the results of sensitivity tests on the impact of various commercial flood coverage assumptions on insured losses. Also available are breakdowns of modeled losses by line of business, coverage, and peril (wind versus surge
The significant increase in estimated losses from AIR’s estimate issued on October 30, the day after Sandy’s landfall, is driven primarily by an increase in estimated losses from storm surge damage. This, in turn, is driven by a reassessment of the percentage of flood losses that will actually be paid, as well as an improved storm surge footprint run against high-resolution industry exposure information.
Issues of particular materiality to AIR’s revised estimates follow.
Commercial Flood Take-up
AIR’s default assumption—and the one used to generate insured loss estimates hours after Sandy’s landfall—is that 10% of commercial structures carry flood insurance. On the basis of a detailed review of insured exposures in the affected area and analysis of client data, AIR has concluded that a more appropriate assumption for this particular storm would be that 20% of the commercial storm surge losses would be covered. Therefore, the above range of losses reflects the assumption that 20% of the damage from Sandy’s storm surge to commercial and industrial properties will be covered. However, AIR acknowledges that there is still uncertainty in this assumption.
One hour before Sandy made landfall, the National Hurricane Center (NHC) issued a public advisory, which, in addition to updating the meteorological information and position of Sandy, stated that the storm had transitioned from a hurricane to a post-tropical cyclone. In the days following landfall, officials in New York, New Jersey, Connecticut, Delaware, Maryland, New York, Pennsylvania, Rhode Island and Washington, DC all ruled that insurers may not impose hurricane deductibles on homeowners policies in their jurisdictions.
In AIR’s initial loss estimates for Sandy posted on October 30, hurricane deductibles were applied. In the revised estimates, residential losses assume a flat $750 deductible.
Additional Sources of Non-Modeled Loss
Note that the range of losses above (USD 16 to 22 billion) does not include damage to infrastructure, such as occurred to New York’s subway system and highway tunnels, nor to marine craft (although clients can model marine craft exposures in their own portfolios in CLASIC/2).
Neither does the range of losses above include low-level losses in states far from the storm’s center, including Massachusetts, New Hampshire, and West Virginia. These losses on the periphery of the storm system were influenced by conditions not associated with the core of the storm system. AIR estimates that the inclusion of these losses would result in as much as an additional USD 500 million to USD 1 billion to the totals above.
When these non-modeled losses are considered, along with the continuing uncertainty with respect to commercial flood coverage, insured losses from Sandy could be higher than these estimates.