Floods remain the most common and destructive natural disaster in the United States, according to the National Association of Insurance Commissioners’ Center of Insurance Policy and Research. All 50 states have experienced flooding or flash floods in the last five years. Some reports suggest that as much as 90% of all natural disasters involve flooding. Flooding is serious business, and should be taken into account no matter where you live or where your business is located.
The Federal Emergency Management Agency (FEMA) developed the National Flood Insurance Program (NFIP) following the 1968 passage of the National Flood Insurance Act by Congress. The Act was passed in recognition of the lack of readily available flood coverage in the private insurance market for many of the most exposed homeowners and business owners. A series of widespread flood events along the Mississippi River Valley in the early to mid-1960s drove most of the private flood insurers out of the market, the frequency and severity of the losses made flood seem uninsurable in the private sector. Enter the NFIP.
The NFIP
As with many national social engineering projects, the NFIP addressed an immediate need but dramatically underestimated the long-term fiscal impact of the program on the American taxpayer. The NFIP allows insureds who would otherwise not have access to any flood insurance to purchase at least a stop-loss policy. Over time, though, the soft white underbelly of the NFIP has come to light — the limits offered are inadequate for the exposures in many areas, and the rates charged are nowhere near actuarially sound, meaning that the NFIP is perpetually in debt to the U.S. Treasury as losses outstrip premiums collected.
The US General Accounting Office puts the NFIP debt at $20.5 billion (you read that right, billion with a “B”) as of February 2018. It is worth noting that the $20.5 billion number is after Congress forgave $16 billion in outstanding debt (some going back to 2005) as part of the Additional Supplemental Appropriations for Disaster Relief Requirements Act of October 2017. Without that forgiveness, the NFIP would be $36.5 billion underwater, if you will.
In recent years the NFIP has been purchasing reinsurance from the private sector, to help offset some of their losses. In 2019 they report purchasing $1.32 billion in reinsurance from some 28 reinsurers, as well as $800 million of additional funds from CAT bonds. This is a step in the right direction but is inadequate to meet the losses of NFIP’s recent bad loss years: 2005 – $17.7 billion; 2012 – $9.5 billion; 2017 – $8.7 billion.
The House of Representatives passed a bill in 2017 that would extend the NFIP through September 30, 2022, but the Senate has yet to pass a similar bill, so no formal reauthorization has even made it to conference committee, let alone the President’s desk to be signed into law. The NFIP narrowly avoided another sunset on May 31, 2019 (nothing new, the NFIP was extended 17 times between 2008 and 2012, and actually lapsed four times in 2010 – 2011). The flood program has been extended through September.
Time will tell whether that expiration date will be met with more continuing resolutions or meaningful and thoughtful improvements to the legislation. These continuing resolutions keep the NFIP in business, but time and again fail to address the elephant in the room: Repetitive loss properties and properties in the highest hazard areas must be charged a premium more commensurate with the exposure for the program to ever be self-sustaining.
Flood losses
According to FEMA, as of September 30, 2017, Florida led the pack with 1,729,505 NFIP policies in force. Texas had 598,815, and Louisiana had 494,830.
For the fiscal year from October 1, 2015, through September 30, 2016, two of these three states also received the most claim payments from NFIP policies: Louisiana – $804,890,000, Texas- $696,029,000 — Florida came in at $17,024,000. The state with the most NFIP losses in that fiscal year has a population roughly equivalent to Los Angeles and accounts for 1.4% of the total U.S. population. Exacerbating the problem are discounted and grandfathered rates (when FEMA updates the flood hazard maps, current policyholders are permitted to keep the rate from the old flood zone determination if it is lower than the new rate), meaning that even when FEMA takes positive steps towards identifying the highest hazard properties, the policyholders in those areas do not always have to pay higher rates. A 2017 Congressional Budget Office report estimates 85% of zone V (the zone with the highest potential for flood losses) policies received the benefit of either discounted or grandfathered rates, or both.
Murky waters
The problem is stark and easy to see — those at the greatest risk for flood losses each year are concentrated in relatively small geographic areas (with apologies to my many friends in Texas, relative to the entire U.S., even Texas is relatively small). The vulnerable populations in both Florida and Louisiana, not to mention the other Gulf states, live in economically depressed regions, or in the case of Florida, boast large populations of seniors on fixed incomes. Charging actuarially sound flood insurance rates in these areas would likely make the coverage unaffordable for many who desperately need it. Unfortunately, the FEMA / NFIP program of subsidizing the rates for the most vulnerable properties by surcharging the least vulnerable properties doesn’t solve the problem. Adverse selection is hard at work in flood insurance, those who feel that their hazard is low often do not purchase the coverage, so the rate surcharges collected are inadequate to offset the subsidies given, and losses outstrip premium collected frequently.
Unlike the marketplace of the late 1960s, in the early 21st century there is appetite for flood insurance in the private sector. Carriers and reinsurers are developing programs for both homeowners and commercial flood insurance, and have been deploying them. Robert Muir-Wood, chief research officer for RMS, a catastrophe modeling company, estimates that as much as 4.5% of all residential flood insurance policies in the US are in the private market. Many private insurers are looking to private flood insurance as a tool to help balance and diversify their portfolios, while others see private flood insurance as a logical extension of the wind/hurricane coverage that they already offer — recent storm events such as Katrina in 2005, Sandy in 2012, Harvey in 2016, and Florence in 2018 resulted in greater losses from flooding than from wind.
Congress has taken some positive steps in the area of flood insurance in recent years. The Biggert-Waters Act of 2012 set some ground rules for private flood insurance and government-backed mortgages: If the private flood policy provides coverage equivalent to or greater than the NFIP, then Fannie-Mae and Freddie-Mac backed mortgages will accept the private flood insurance as meeting the requirement for purchase in high-hazard flood zones.
These guidelines have been updated a couple of times since to clarify expectations. Now it is time for a comprehensive overhaul of the NFIP —we need to sort out which elements of the program are working as designed, and which are falling short. An updated and revised NFIP would make the private flood market even more appealing to the insurers who are sitting on the edge of the pool debating dipping a toe into the water.
The NFIP’s future
There will always be a place for the NFIP, some properties and regions will defy any attempt at actuarially sound rating — meaning that private insurers are unlikely to compete for the business. Private insurers have a need to operate profitably, after all. There are, however, private insurers seeking the high-risk, high reward of writing flood business in high hazard zones, just as there are those more interested in lower hazard flood zones. The CBO reports that in the period from 1978–2006, properties in flood zone X accounted for 45% of the net adjusted loss that the NFIP experienced (zone A accounted for 54.5%, zone V 0.5%). If the private flood market increases its market share over time, it almost has to improve the results for the NFIP — the CBO report found no flood zones in which the NFIP was writing business without experiencing a net adjusted loss. Just as a rising tide floats all boats, a waning flood should have the opposite effect — spreading the exposure across many insurers (public and private) should alleviate much of the pain.
Michael Brown (michael@goldenbear.com) is vice president and property department manager at Golden Bear Insurance Company. The opinions expressed here are the author’s own.
Related: