14 October 2013Re/insurance

A girl named Sandy

Superstorm Sandy surprised many in the US when it struck New York in October of 2012, resulting in the second costliest storm in US history. Eclipsed only by Hurricane Katrina seven years earlier in terms of economic loss, Sandy resulted in extensive damage to the US east coast—New York in particular—with insured losses reaching $30 billion according to Aon Benfield, $7.2 billion of which were paid by the federally-run National Flood Insurance Program (NFIP).

While the event had precedents—Hurricane Donna of 1960 and Hurricane Irene of August 2011, which came close to New York—the event was still regarded by many as unexpected. The storm surge component of the storm proved particularly costly, prompting those affected by Sandy to re-evaluate thinking regarding their exposures, understanding and re/ insurance buying behaviour as a result.

Bermuda:Re spoke with a host of players in the market about the major lessons to have emerged from Sandy and found the east coast preparing for another major storm in light of Sandy losses.

MTA: a capital market response

For the New York Metropolitan Transport Authority (MTA) the impact of Sandy was severe. As Nora Ostrovskaya, a senior manager of strategic initiatives at the MTA outlined, the MTA is responsible for not only the New York subway system, but also two railway networks and two bus companies that are together responsible for the transit of more than 8.5 million people each day. With many of these assets being subterranean, low-lying or in exposed locations, the impact of Sandy proved costly. Economic losses incurred by the MTA reached $5 billion, with insured losses totalling $1.07 billion, as the MTA coped with the inundation of vast tracts of tunnel and low-lying infrastructure. While lessons had been learned and applied from the close call of Irene the year before, said Ostrovskaya, the impact of the storm on low-lying areas was significant and—to a large extent—unavoidable for the MTA.

The MTA has long recognised its exposure to catastrophic flood damage and as a result has had a sizable all-risk property insurance policy in place, with a programme limit of $1.075 billion subject to a $25 million self-retention, explained Laureen Coyne, director of risk and insurance management at the MTA. However, its upper layer—the $275 million beyond $800 million—did not cover catastrophe risk, providing instead business interruption coverage, although with limits in certain exposed areas. The MTA’s insurance programme helped stem losses from Sandy—with the federal government stepping in to help with much of the remaining cost— but its support from the private market was sharply curtailed as a result of Sandy losses. As Ostrovskaya explained, in very rough terms pricing doubled and capacity halved for the MTA following the storm, forcing a rethink in the way it purchases its capacity. MetroCat would become the answer.

"Despite these headwinds MetroCat proved a successful issuance, with the bond providing the MTA with $200 million of protection from storm surge resulting from named storms."

Coyne said that the MTA was deeply concerned about the forthcoming April renewals and the effect that a contraction of capacity and a marked rise in pricing would have on its re/insurance programme. As a result, the MTA began to explore capital market alternatives, which eventually led to the issuance of MetroCat. Ostrovskaya explained that the MTA was under tremendous time pressure to complete the deal and provide complementary capacity to its existing re/insurance programme before April renewals. The MTA also faced considerable execution risk, she said, with the MTA expecting—but not being certain of—investor appetite, while the bond is also the first to cover the risk of storm surge through the capital markets.

Despite these headwinds MetroCat proved a successful issuance, with the bond providing the MTA with $200 million of protection from storm surge resulting from named storms. The bond delivers competitively priced coverage that would have been hard to find in the traditional market and complements a changed re/insurance programme at the MTA.

As Coyne detailed, post May 2013, the company now has a traditional reinsurance programme with limits of $500 million—including coverage for catastrophe perils and with no sublimits for low lying areas—alongside its MetroCat bond, which provides $200 million of coverage in excess of the $500 million for flood for the next three years. Ostrovskaya said that the MetroCat deal was a welcome addition to the MTA’s insurance buying toolkit, with the MTA now being in a position to add other layers through similar capital market deals. Coyne predicted that the traditional market would take a few years to return.

Endurance: surge, autos and a changing dynamic

The threat posed by storm surge—which, while understood by the industry, was not fully appreciated by consumers—the extent of auto losses, and the changing dynamics of the market postevent have been some of the key themes to have emerged from Superstorm Sandy.

That is the view of David Bigley, head of North America property catastrophe reinsurance at Endurance, who said that in many ways Sandy found consumers unaware of the impact that storm surge can have in the Northeast US. He said that Endurance had been aware of the exposures posed by Northeast storm surge for some time and had actively minimised commercial exposures which carry surge coverage, but others were evidently less well prepared.

Bigley said that Sandy has helped to sharpen the debate around the risk posed by both wind and water, with surge damage encouraging a refocusing on the water component of Atlantic storms.Bigley said that one of the most surprising lessons to have emerged from Sandy was the extent of auto losses. He said that the storm broke all historical precedents, with only hurricane Katrina resulting in more sizable auto losses. Allstate, for example, incurred $1.08 billion of losses during Sandy, 40 percent of which was auto loss, with 78 percent of that located in New York. Details from PCS published in December 2012 put the number of damaged vehicle claims after Sandy at 230,000. As Bigley explained, either people had more than one vehicle or they had reason to leave their cars behind—either way, the auto component was substantial.

Sandy also raised the prospect of change within the private market and at the NFIP, where reform is likely to be a more protracted affair. Bigley said that Sandy had created interest in alternative capital deals such as the MTA’s MetroCat—which explicitly protects against storm surge—and had encouraged movement among traditional players in the Northeast. He said that he had seen increased competition along the coast as those players that don’t have significant Northeast exposure look for market dislocation and increased rates, while some that experienced large claims from Sandy are seeing their concerns reaffirmed and are moving away.

It seems that a fluid reinsurance landscape exists post-Sandy. Sandy has also encouraged reinsurers to consider their exposures more closely, with many carving out surge and flood from wider wind policies. With surge being the most damaging component of Sandy, many players are increasingly cautious about such exposures.

Sandy is also encouraging a much-needed rethink regarding the long-term viability of the NFIP, said Bigley. The programme is $26 billion in debt following Sandy, a figure that is simply unsustainable in the long run. Making the leap towards the commercial provision of flood coverage is, however, a significant challenge and raises issues of affordability. Homeowners in the Northeast are currently highly subsidised and it is unlikely that an increase in premiums paid would be welcomed. Bigley said that a stepped solution was probably the most palatable, but added that the industry would need to be satisfied that rate adequacy could be achieved, particularly following events such as Sandy.

ABIR: Sandy providing impetus for flood programme reform

Supporters of the Biggert-Waters Flood Insurance Reform Act 2012 received added impetus following Sandy, with skirmishes in Congress over rate-adequate pricing, increased risk transfer into the private sector and the sustainability of the existing programme being shaped by the storm.

That is the view of Brad Kading, president and executive director of the Association of Bermuda Insurers and Reinsurers (ABIR), who said that while there has been wrangling over the implementation of Biggert-Waters, the general consensus—helped in part by Sandy—is that planned reforms need to press ahead.

The NFIP extends 5.6 million policies providing more than $1.2 trillion of coverage, according to statistics from the Federal Emergency Management Agency (FEMA), but its position has become increasingly untenable. Kading said that post-Sandy the NFIP was obliged to raise the borrowing authority by a further $9 billion in order to pay claims. The NFIP is now around $26 billion in debt, with Sandy helping to focus people’s minds on future debt obligations associated with the programme, said Kading.

This has in turn raised three key talking points. First among these is the need for rate-adequate pricing. A stepped approach has been established under Biggert-Waters—a 25 percent annual increase until risk-based pricing is achieved—but political opposition has sought to derail the reform. Kading said Republicans in Congress are however fighting to defend the changes. He said the Sandy experience was such that they are digging in against any roll-back of rate increases.

Kading added that despite the political and public noise around reform, those in Sandy-affected areas such as New York and New Jersey have lent their support to the bill. Governors Chris Christie and Michael Bloomberg evidently recognise the value of the bill and the need for reform, he said.

Second, Sandy has also helped to sharpen minds regarding the role private reinsurance can play in supporting the US flood programme. Kading said that private involvement was considered positively by those who wrote the legislation, but added that the FEMA was taking a rather more conservative view of private sector involvement. FEMA will await the results of research into reforms to the flood programme, said Kading, which will begin in September. Final results will be published in a year, but intermediary findings will be made public in the interim. Kading explained that the case could be made for a private reinsurance element right now, but said that FEMA is awaiting the cost:benefit analysis from the report.

Finally, Sandy and the mounting debt bill have encouraged a rethink regarding the viability of the existing flood programme. Suggestions pre- Sandy that there would be federal forgiveness of the debt evaporated following the storm, with decision-makers in Congress unwilling to grant such a move, said Kading. This will necessarily force change.

Greater risk transfer into the private sector is perhaps the most obvious answer—if not a solution—to the mounting debt issue. As Kading explained, private involvement would reduce the likelihood of future debt issuance, even if it does not solve the problem of paying off the existing debt. Events such as Sandy have certainly helped to focus minds and it seems likely that a more significant private component to US flood insurance provision is in the offing. The results of the research report into flood insurance reform will undoubtedly be awaited with interest.

CoreLogic: storm surge a significant threat

Coming a little over a year after Irene, Sandy dramatically illustrated the destructive nature of storm surge, with the storm causing approximately $50 billion of economic damage. According to CoreLogic the storm exhibited “the worst possible trifecta of characteristics: an extremely large diameter, strong winds and high tide at landfall, which generated massive surge that inundated the coast from New Jersey to Connecticut”. While not unexpected in re/insurance circles, the storm did catch US businesses and homeowners by surprise, with the water component of damage proving particularly significant.

CoreLogic’s 2013 Storm Surge Report detailed the damage caused by Sandy, but also outlined just how exposed other east coast US cities are. “It is imperative to understand that future storms have the potential to affect other coastal areas just as dramatically, or possibly return to the New York/New Jersey area as larger events capable of even greater damage,” explained CoreLogic. Dr Tom Jeffery, senior hazard scientist, CoreLogic Spatial Solutions, added that “the fact that strong storms hit this area in consecutive years doesn’t mean they will continue to do so, but it is important to realise that the north east is going to experience more hurricanes and tropical storms in the future, and to plan and prepare appropriately”.

Jeffery added that Sandy was by no means the worst storm that the north east could face. Rather than provoking fear however, such knowledge should lead to “rational planning and decision-making in order to reduce the risk to life and property” from storm surge. He said that whereas the media’s attention had been very much focused on wind damage in the past, there is now an increasing interest in the threat posed by storm surge.

“This is primarily due to a better understanding of just how incredibly damaging a large volume of water that is forced onshore can be. When you realise that one cubic yard of water weighs approximately one ton—and Sandy was pushing a 13+ foot surge, then it becomes apparent just how damaging such events can be,” Jeffery explained.

CoreLogic warned that consumers have limited awareness regarding the threat posed by storm surge, adding that “the impact of Sandy clearly established just how many homes and coastal communities are vulnerable in an area that is not often associated with hurricane activity”. The report stated that homes outside the FEMA flood zones may also be vulnerable to storm surge flooding, but that appetite to buy coverage was limited, even in the face of Sandy losses.

Storm surge risk is not limited to the north east. As CoreLogic’s report outlined, the threat posed by storm surge is significant right along the Atlantic and Gulf of Mexico seaboard. New York remains the US metropolitan area with most significant exposure to storm surge, according to CoreLogic, but it is apparent that a host of US cities will need to consider the threat more closely. According to data from CoreLogic, “of the more than $1.1 trillion in residential property exposed to storm surge along the Gulf and Atlantic Coasts, more than $658 billion of that risk is concentrated in the top 10 metro areas”. Events such as Sandy are far from being one-offs.

Insurance Information Institute: Rates react to Congress and history of loss

The insurance industry was not surprised by Sandy, having expected and modelled for even more severe north east losses. And while rates in the north east rose following the storm, they were the product of impetus established by the Biggert-Waters Act and efforts to introduce risk-adequate pricing for flood risk, rather than as a result of Sandy.

That is the view of Dr Robert Hartwig, president of the Insurance Information Institute, who said that the impact of Sandy was rather more limited than the public perception. Hartwig said that it was well known in re/insurance circles just how vulnerable New Jersey and New York are to storm surge. As a result, there was no surprise or market dislocation following Sandy.

The storm did, however, provide a wake-up call to homeowners and businesses without flood insurance. They had been playing Russian roulette with their homes or businesses, said Hartwig, with tens of thousands having lost that bet. He said that there was a spike in interest in flood insurance following Sandy, but that Biggert-Waters was already determining the course of conversations regarding flood insurance provision and pricing.

Hartwig said that there was a remapping of flood zones and changes to subsidies underway as a result of the Act. These have in turn led to higher rates. Many ascribed the changes to Sandy, but upward momentum already existed before the storm as the NFIP introduced stepped changes to establish risk-adequate pricing, said Hartwig.

Commercial market conditions provided still further momentum. US property rates have been rising for some time, said Hartwig, unhindered by the influx of capital that is finding its way into the reinsurance space and reflecting experiences from a string of natural catastrophe losses.

Meanwhile, Biggert-Waters presents an opportunity for private insurers. As Hartwig outlined, many consumers wouldn’t need much convincing to opt for private insurance. With the NFIP’s performance much maligned, and incremental rate rises being introduced as a result of the Act, consumers will likely consider their options more closely.

Once rate changes introduced by Biggert-Waters reach an actuarially sound basis there will be greater opportunities for private insurers, said Hartwig. He warned that states may move to apply rate suppression as they do in states such as Florida, but if these efforts could be resisted that there could be significant opportunities for both insurers and reinsurers. Given assurances that risk-adequate pricing can be charged and will be sustained, developments under Biggert-Waters could be a major win for the industry and offer consumers more options, he said.

About the only cloud on the horizon is the prospect of fewer consumers buying flood insurance due to rising rates. Hartwig said that this was an issue the NFIP and industry would have to confront, but it is evident US consumers have been subsidised for too long. Events such as Sandy should help to hammer home this point and the need for them to purchase coverage.

Aon Benfield: a muted impact on reinsurers

Despite the economic impact of Sandy, its implications for reinsurers were decidedly limited, said Bryon Ehrhart, chief strategy officer for Aon Benfield and chairman of Aon Benfield Analytics and Aon Investment Banking Group. As he explained, substantial portions of the flood losses were either covered by government through programmes such as the NFIP or were simply uninsured. Those losses that did make it into the commercial market were largely retained by insurers—with losses proving insufficient to exceed their reinsurance retentions or risk tolerances.

Coverage has responded to Sandy losses, said Ehrhart, with programmes re-priced and restricted to reflect recent experience.Changes have largely been limited to the insurance space, with reinsurers having been generally unaffected by Sandy losses. The substantial flow of alternative capital into the reinsurance market has at the same time served to stymie any marked change in reinsurance pricing, said Ehrhart. Changes post-Sandy have been driven more by new capital flows than losses associated with the event, he said.

Deals such as MetroCat have certainly helped to change thinking around buying behaviour for those with sizable exposures significant enough to find a home in the capital markets.

RMS: increased granularity needed as exposures deepen

Re/insurers will need to closely consider the granularity of their data following Sandy, particularly their assumptions around their exposures to storm surge. That is the view of Claire Souch, vice president, model solutions at RMS, who said that following Sandy, re/insurers need to ensure that they have accurate information concerning risks insured. She said that the storm had demonstrated the need for location-specific data, with flood losses varying considerably even on the same street. As she explained, underwriting storm surge is all about data quality.

"The storm had demonstrated the need for location-specific data, with flood loses varying considerably even on the same street."

The need for increased granularity is all the more pressing considering the fact that Sandy was by no means a freak event or worst-case scenario, said Souch. As she explained, rising surface temperatures associated with climate change and the increased frequency of associated Cape Verde storms will likely increase the threat posed to the US east coast. Sandy was one such Cape Verde storm said Souch, which while not particularly severe, grew to around the size of Mongolia as it headed north, pushing a huge wall of slow-moving water ahead of it. The history of the storm out at sea is what built up the storm surge, she explained, with such events increasingly likely if sea surface temperatures continue to rise.

Even taking an agnostic view on climate change, she warned that the chances of another Sandy affecting an east coast city in the coming few years was significant. Singling out Baltimore, Tampa, Biloxi and Naples, Florida, as other cities with considerable exposure to storm surge, she said the industry needs to take a proactive approach to the threat posed.

Worsening matters is the urbanisation of coastal areas, said Souch, with increasing insured values evident right along the US east coast. Global trends are following a similar pattern, she said, with Asia likely to be particularly affected by storm surge as urban and industrial areas spring up in coastal locations to take advantage of power, cooling and transport.

In response, companies such as RMS have sought to get their arms around the complexities of storm surge. As Souch explained, until a few years ago modelling firms simply did not have the enormous computer power needed to understand storm surge. Understanding exposures is an extremely computer-intensive exercise, she said, requiring enormous amounts of power and data in order to establish an informed view of likely exposure. Only following considerable work and thanks to advances in computer power has RMS been able to build its own model view of storm surge, helping the industry better understand its exposures. With the risk likely to remain—and in all likelihood increase—re/insurers will need to pay close attention to data in order to better understand the threat posed by storm surge.