FEMA flood maps are not particularly glamorous or technologically exciting. They have done their work for many years and, provided that they are up to date, are an effective way of communicating a generalized level of flood risk. FEMA flood maps have been the primary, flood insurance underwriting tool for the National Flood Insurance Program (NFIP). That program is currently billions of dollars in debt to the U.S. Treasury due to premium subsidization.
Also, FEMA flood maps have been used extensively by local governments in their efforts to keep development out of the floodiest areas of the U.S. That has not worked out so well, either, as coastal and riverine developments in the most flood-prone areas have abounded over the past 30-years. As we saw with the Hurricane Harvey disaster in 2017, FEMA flood maps are far from perfect. Nearly a hundred thousand homeowners in the FEMA X Zone (500-year, or 0.2% annual risk) in Houston, who were told that they did not need or were not required to purchase flood insurance, had their homes severely damaged by Harvey’s heavy rainfall flooding.
In 2016, we at Coastal Risk Consulting observed that FEMA flood maps did not provide comprehensive enough flood risk assessments to allow individuals and businesses to make accurate “buy, sell, protect and insure” decisions at the property level. So, Coastal Risk identified a number of improvements that could be made to the FEMA flood maps, including downscaling risk mapping to the individual property level, and made them publicly available for purchase at www.floodscores.com for any property in the U.S.
FEMA flood maps don’t tell the whole story about your risk of flooding.
First, as was seen with Hurricane Harvey flooding in Houston, FEMA flood maps don’t include heavy rainfall flooding risks to your home or business. Properties in FEMA X Zones, which don’t require flood insurance for federally insured mortgages, may definitely need to be insured to protect them from heavy rainfall flooding. Coastal Risk models heavy rainfall flood risk for every property in the U.S.; FEMA flood maps do not.
FEMA flood maps also don’t include coastal tidal flooding risks to your home or business. Tidal flooding is a property damage “threat multiplier.” When hurricanes come ashore at high tide or even a King Tide, which often occurs in the fall on the Atlantic Coast of the U.S., properties with existing tide flooding are at much greater risk of damage and loss than those that don’t experience tidal flooding. FEMA flood maps do not take this type of flooding into account, now and into the future as sea levels rise. King Tides occurs during the height of hurricane season. Because King Tides are due to astronomy and not weather, scientists know precisely when they will occur.
FEMA flood maps also underestimate the height of hurricane storm surge, as compared with NOAA models. NOAA surge models typically show higher water heights than the FEMA Base Flood Elevations (BFEs), which are a key component of the FEMA flood maps. The higher the surge, the greater the economic damage and loss to properties, and the greater the risk of injury and death to those who don’t evacuate in advance of these storms.
The federal National Flood Insurance Program (NFIP) underwrites the overwhelming majority of residential flood insurance policies in the U.S. As of April 2018, more than 5 million NFIP policies were in force nationwide (4.8 million residential), representing slightly more than $1.28 trillion in coverage ($1.17 trillion residential). For decades, the NFIP has been homeowners’ only option for flood insurance, but over the past several years a small private market for residential flood insurance has emerged. Policymakers are increasingly interested in learning whether the expansion of this market could help meet the policy goals of increasing the number of homeowners with flood insurance or offering more affordable coverage.
Stakeholders—in congressional testimony, op-eds, reports and other forums—have offered diverging opinions as to the appetite of the private sector in writing more flood insurance, on the existing barriers to private coverage and on the implications for the NFIP. The present state of the market is unclear, particularly because there is no nationwide database on the companies writing residential flood insurance, coverages offered, policy terms, pricing and any differences between private and NFIP flood insurance. This makes it difficult to evaluate the market’s future evolution and relationship to the NFIP.
This report aims to fill these knowledge gaps and has two primary objectives:
to document the current state of the private, residential flood insurance market across the U.S.; and
to identify the main factors influencing the number and form of flood insurance policies offered by the private market.
To meet these objectives, we conducted in-depth, semi-structured interviews with 63 insurers, reinsurers, state brokers and other market participants. We also gathered and analyzed current private market data from a range of sources, including public documents, congressional testimony, news articles, state regulators and private firms.
The private residential flood insurance market in the U.S. is currently small relative to the NFIP. We estimate that private flood insurance accounts for roughly 3.5% to 4.5% of all primary residential flood policies currently purchased.
With the exception of Puerto Rico, more policies are written by surplus lines carriers than by admitted carriers subject to state rate and form regulations. This is unsurprising, because surplus lines firms tend to cover new or catastrophic risks for which consumers may have trouble finding coverage in the admitted market.
Roughly 20% of private residential flood policies (and 40% of admitted carrier policies) are in Puerto Rico; another roughly 20% are in Florida. No data are available to evaluate the size of the total private market in other states or at a substate level nationwide.
Private market growth to date has largely been driven by the interest of global reinsurers in covering more U.S. flood risk. In the admitted market, reinsurers are assuming most of the risk for primary insurers, often in excess of 90%. In the surplus lines market, Lloyd’s of London has played a major role, backing the majority of residential flood policies.
Among the small number of policies written by the private sector, we identified three broad policy types. The most prevalent is what we refer to as an “NFIP+” policy within the FEMA-mapped 100-year floodplain, where flood insurance is required for federally backed mortgages. NFIP+ policies have higher limits or broader coverages than NFIP policies. Most are stand-alone policies, although some are sold as endorsements to homeowners policies. A second type is a lower-coverage-limit policy issued as an endorsement in lower-risk areas. The third type, used by only a couple of firms, mimics the NFIP policy.
There does not exist data to ascertain how many homeowners previously uninsured against flood are purchasing private policies versus how many are switching from NFIP policies to private coverage. Insurers in the market believe their portfolios include both newly insureds and policyholders switching from the NFIP.
Because the NFIP will provide a policy to anyone in a participating community, private firms can operate only where they can price lower than the NFIP or provide broader or different coverages for which there is consumer demand. In a sense, then, the NFIP is a default benchmark for comparison with private flood insurance policies.
Companies have identified certain types of properties or risks where they believe they can profitably operate and compete with the NFIP. Those target areas of opportunity, however, vary across firms. For example, some are restricting themselves to areas that FEMA designates as having lower flood risk, and others are focusing on areas that FEMA designates as at higher flood risk.
The largest U.S. homeowners insurance companies have generally been hesitant to enter the flood market, although a few have begun to enter through subsidiaries. Their caution, we learned, stems from concern about being unable to adjust rating or policy coverages as they gain experience in writing flood because of state regulatory practices; concentration of risk in their portfolio; correlation of flood with existing wind exposure; satisfaction with the current arrangement; and concern about reputational risk should they need to raise premiums or scale back coverage as they explore the potential flood market.
More private capital is now willing to back private flood coverage in the U.S. Interviewees agreed that, as insurers’ familiarity with flood catastrophe models grows, as underwriting experience develops and as state regulatory structures evolve, the number of private flood policies in force could continue to grow, including among admitted carriers. As of this writing, there were multiple new rate filings in many states, suggesting a continued expansion of the market.
Whereas the NFIP is required to take all risks, private insurers are selective in their underwriting. All interviewees agreed that the private sector will never be able to write policies for certain properties or locations (e.g., repetitive loss properties or high-tide flooding areas) at a price homeowners would be willing to pay. Substantial public investment in risk reduction, combined with aggressive land-use management, they said, was essential for limiting future exposure and encouraging the private sector to move into those areas.
The private market participants we interviewed differed as to how much flood risk in the U.S., and storm surge risk in particular, they thought could be underwritten by the private sector. All agreed there would likely remain a large and important role for the NFIP to play, particularly in the near term.
Acceptance of private flood insurance by banks and financial institutions does not appear to be a major constraint on the market at present. With very few exceptions, private insurers have told us banks ultimately accept their products, though they may have some initial questions or concerns.
There is a need for expanded insurance agent education about flood risk and flood insurance products, both for the NFIP and private policies. Interviewees disagreed about whether the higher-than-market commissions paid by the NFIP were creating a disincentive for the private market.
Most interviewees saw limited demand for flood coverage today, whether offered by the NFIP or by a private provider, and said that consumers were price sensitive.
From “I thought homeowners insurance covered that,” to “I’m in a low risk flood zone, so I don’t need flood insurance,” flood insurance agents have heard all the myths about why homeowners don’t need flood insurance. Unfortunately, these myths often undercut the true dangers of flooding and leave home and business owners across the country woefully underprepared if a flood event does occur.
In 2017 alone, the National Oceanic and Atmospheric Administration reported 16 separate disasters in the U.S., each with damages exceeding $1 billion, which generated total record losses in excess of $306 billion. A large percentage of these damages were caused by flooding, including damages associated with Hurricane Harvey in Texas and Louisiana, and record high water levels in Missouri, Arkansas and Illinois. While this devastation has certainly brought the conversation about flooding and flood damage back to the forefront, it also may have relayed a subtle, worrisome message to others: Flooding events only affect certain regions at certain times of the year, and 2017 was an anomaly.
The reality is quite the opposite.
When it comes to flooding, these disasters can happen any time of the year, anywhere across the country. In fact, 25% of all flood damage in the U.S. occurs in what are classified as low- to moderate-risk flood zones by the National Flood Insurance Program (NFIP), and destructive flood events have occurred in 98% of counties across the country. This is especially troubling when, as the NFIP estimates, just one inch of water intrusion can cause more than $20,000 in damages, and the average NFIP claim is around $43,000. In addition, one-third of FEMA disaster assistance goes to properties in these low- to moderate-risk zones, but it’s rarely enough to cover the damages.
The data from these costly flood events reveals a story of loss and hardship that needs to be more widely understood, and now—with the bevy of new private flood insurance options in place—is the perfect time for agents and organizations to research the facts and widely broadcast the message to the home and business owners they protect.
What Agents Need to Know
Flooding is not restricted to heavy tropical storms. It takes multiple forms that can affect other regions in the U.S., including rapid rainfall or structural failure leading to flash floods, spring snowmelt, changing weather patterns, clogged rainwater systems and new building development. Flooding can even be triggered by drought and wildfires. Why? Because wildfires and droughts alter soil conditions, leading to reduced absorption and increased runoff during any heavy rain that follows.
Simply put, flooding is the costliest and most common natural disaster in the U.S., especially considering that it can strike any time of year. Ultimately, a single flood event can destroy a property and wreak irreparable damage on the finances of uninsured home and business owners. Yet, despite these dangers, only 12% of homeowners have flood insurance. For those in low-risk areas or those without mortgage loans not required by their lenders to purchase NFIP policies, this is a huge problem. Consider the following:
A March 2017 survey by InsuranceQuotes found that 56% of respondents mistakenly believed that a standard homeowners policy covers flood damage.
Most business insurance policies exclude flood damage.
FEMA flood maps may be outdated, and flood risks are changing rapidly. Some research suggests that current maps vastly underestimate those in the high-risk, 1-in-100-year floodplain, with one study suggesting that 40 million Americans, instead of the current estimate of 13 million, are at high risk.
Hurricane Harvey hit low-risk flood areas, and fewer than 20% of homes damaged in Hurricane Harvey were flood-insured. In total, just 15% of all the 1.8 million homes in Harris County (Houston) had flood insurance, including only 28% of the homes in high-risk areas.
NFIP policies in low-risk, preferred risk areas are as low as $500 a year, while a flood claim averages $43,000.
Collectively, these statistics highlight the importance of flood insurance for all Americans, no matter where they live. As agents, it’s our job to understand the potential risks in our communities and educate clients—and potential clients—on their flood risk to advise them in the best possible way.
What Are the Solutions?
For agents and others in the insurance industry, spreading the flood risk message requires perseverance and a keen understanding of the different insurance products that are available. While it’s important that owners and tenants know their FEMA-assigned flood risk, it’s also key that they understand the shifting nature of flood risk: that the dangers are changing, and every property is at risk at any time of year. Additionally, while government-backed NFIP policies are a critical way for many Americans to secure coverage, they are not the only option. Some homes and businesses may need excess coverage to cover up to replacement cost, and still other eligible properties may benefit from the array of coverage options available with private insurance products instead of an NFIP plan.
That’s why it’s critical for agents to analyze the true needs of our clients and regularly communicate any changes or updates that may be necessary.
The NFIP provides coverage limits up to $250,000 for the structure of a home, and up to $100,000 for personal possessions. For business owners, coverage limits are up to $500,000 for a structure and $500,000 for contents. Depending on the value of their structure and belongings, some home and business owners may need excess coverage, provided by private insurers to supplement their NFIP policies and provide additional coverage options. For example, NFIP policies do not cover additional living expenses—like the rental of a hotel room if a family is displaced from their home—or coverage for basements and pools. Private flood insurance policies, however, may provide coverage for these things and more, to better serve consumers.
Homeowners and renters aren’t the only ones who can be helped by private flood insurance options. Often, much of the focus during catastrophic flood events is on homes and families, but the effect on businesses is just as devastating. Buildings and inventories have been destroyed, computer equipment wiped out and workers left with nowhere to go—and no job left to do. Often, the cost of recovery from a flood is too much for a business to manage, with 40% of small businesses never re-opening their doors after a disaster.
The fact is, whether a tenant or building owner, an uninsured company can lose everything in an instant when the water rises. The message agents deliver to business owners needs to mirror the one delivered to homeowners: The risks of flooding are high, and the cost of inaction could be financially unbearable. Agents need to make a special effort to educate business owners on the topic. After all, a company that is willing to invest in anti-theft measures, IT protection and liability insurance should recognize the real financial dangers that floods can deliver.
At the end of the day, private flood insurance options will only be embraced if agents and their clients are fully educated about the risk of flood, the limits of NFIP and all the changing options available. The bottom line is: Flood insurance options are expanding, and we as agents need to understand the full array of flood coverage possibilities available to ensure that those we serve have the opportunity to choose the coverage that will best protect their homes, their families and their businesses.
The U.S. West Coast is regularly confronted with the risk of conflagration if a wildfire is sparked. An estimated 3.6 million residential properties in California are situated within wildland-urban interface (WUI) areas, with more than one million of those residences highly exposed to wildfire events, according to a 2010 federal study.
However, the Tubbs wildfire – which rolled up with other California wildfires to result in some of the largest global reinsurance recoveries during 2017 – spread into the Coffey Park neighborhood that was situated outside WUI areas. This disaster amplifies the vulnerability to wildfire to even urban environments across the state and highlights the importance of continually mitigating wildfire exposure to protect people, homes and businesses. But what can be learned from this wildfire to enhance disaster planning and communications?
Mitigation approaches range from using generational lessons garnered through prior wildfires, as melded with science and technological gains, to adherence to public policy at the state, county and community levels. These strategies can unravel if absent of preventative measures taken by individuals at home or work to, for instance, expand defensible space at the property location and surrounding environment, combined with installing more effective fire retardant materials (i.e., roofing, siding, fencing, decks, etc.). All these measures are underpinned by the research and resources available through agencies or organizations such as CALFIRE, FEMA, IBHS and Firewise.
While the existing preventive measures did save lives and mitigate damage within Sonoma County, conditions and the speed of a widely spreading event combined to overwhelm emergency services and even many fortified structures that were eventually swept away by a wildfire which often flowed faster than water running downhill.
To elaborate upon such conditions, the humidity level was extremely low, westerly winds (Diablo) were fierce and swirling once engaged with wildfire, and embers the size of footballs were carried up to more than a mile away. The wind-driven conflagration ignited and leapt over a major highway into Coffey Park, an urban community situated on a flat setting and woven with tightly aligned residences often connected by fences, common shrubs and a canopy of overhanging trees. Enabled by such a common urban setting, the wildfire ignited into Coffey Park like a fast-burning fuse – the scope and speed of which can be seen in Berkeley Fire Department’s video as posted on the KTVU website. Evacuation efforts were the highest priority for first responders who also established a perimeter, aided by wider firebreaks on multiple sides, to eventually contain the wildfire flaring within Coffey Park. But the destructive wake of the wildfire ravaged Coffey Park, leaving behind nearly two dozen victims and the ruin of roughly 1,300 properties.
Without delay, and while functioning within areas situationally under control of authorities at the outset, the insurance industry moved swiftly and visibly to respond to honor obligations; immediately providing additional living expense funding, including for evacuation, and, when possible, settling property claims. However, the rebuilding campaign and P&C claims settlement process for the Tubbs wildfire, in an area with an extremely low vacancy rate prior to the disaster, will take many years to complete.
From the public policy perspective, governmental entities and citizens will revisit evacuation planning, especially in light of officials deciding to issue an e-mail alert and not a widely dispersed emergency cell phone alert due to concern that such a widespread alarm would hamper emergency efforts. However, when the wildfire expanded beyond worst expectations, the unintended consequence of such a decision led to a frenzied situation in Coffey Park, with citizens making a critical difference by racing from door to door to awaken neighbors, while staff at facilities in Santa Rosa resorted to using personal vehicles to clear patients from a hospital and residents from a senior care center.
A “back to the future” siren system, as a basic supplemental measure, might merit consideration.
The Public Utilities Commission may need to revisit how electrical power is supplied to mountainous areas where, during violent winds, swaying electrical lines or fallen transformers / poles often result in arcing wires that spark fires that easily spread.
With saving lives, most understandably, serving as the highest priority for emergency services, more insurers may consider contracting with private firefighting operations that have a superior ability to traverse steeply sloped and winding roads to reduce chances of property damage. These basic actions could involve, for example, clearing vegetation and combustible fencing with chain saws, accessing water sources with pumps, covering vents, clearing gutters and applying fire retardant foam spray to structures.
Underwriting teams commonly access software tools to address wildfire exposure as part of the selection and pricing processes, as well as review output generated through portfolio models to monitor aggregate exposure. However, while steadily improving and evolving through lessons learnt, these tools offer imperfect guidance. For wildfire models, what is expected is burning along the Wildland Urban Interface with minor encroachment into other areas, with such expectations being met during the past 30-plus years [model era] subject to the notable exceptions of the Oakland Hills  and Coffey Park  situations. Insurers, in view of the recent dimension of an urban tragedy striking Coffey Park, may choose to (re)explore “model miss” options that could influence spread of risk and reinsurance design strategies..
The industry will benefit through supporting community awareness campaigns to inspire customers to be vigilant about wildfire exposure; especially as another Coffey Park-type event could potentially recur in California or even happen elsewhere under similar conditions.
Following the series of US catastrophes, Aon visited the sites of Houston for Hurricane Harvey, Puerto Rico for Hurricane Maria, Florida for Irma and Northern California for the wildfires. The team surveyed the damage and assessed how each event evolved to affect both people and properties with the goal of enhancing catastrophe models and identifying lessons for the future. I was joined by Dan Dick, Steve Bowen, Steve Jakubowski, Jeff Jones and Weston Vosburgh.
Hurricanes Harvey, Irma and Maria laid bare fundamental inadequacies of the current flood insurance program in the U.S. Too few homeowners had flood insurance in place. Federal Emergency Management Agency (FEMA) flood maps were inadequate to encompass actual flood risks and, even more importantly, outreach programs by FEMA and the National Flood Insurance Program (NFIP) were inadequate to properly communicate risks to the market.
The current hurricane season revealed an astounding lack of resiliency in the U.S. on many levels: (1) lack of insurance coverages; (2) inadequate mapping of flood risks; (3) failure to properly educate homeowners about their actual flood risks; and, (4) gross under-investment in resilient infrastructure by all levels of U.S. governments.
As Congress prepares to slash budgets and cut taxes, the fact that Hurricane Irma easily topped City of Miami’s sea walls with only Category 1 strength winds and brought a four-foot river of water down Brickell Avenue, the center of Miami’s Financial District, shows how much work needs to be done to achieve resiliency in just one U.S. city.
In its rush to reduce taxes, Congress is ignoring the U.S. infrastructure deficit, which is estimated to be in the multiple trillions of dollars by the American Society of Civil Engineers. The slide deck that I prepared for the Future of Flood Summit discusses new, cost-effective tools for flood risk modeling, flood risk report production and flood risk communication. A central theme is: “Can the insurance industry do a better job of helping insureds and societies cope with the increasing risks they are facing, as our climate changes and sea levels rise?”