This Oct. 22 article, Coastal homes lack flood insurance because of maps, by Kate Spinner in the Sarasota Herald Tribune presents a good starting point for discussing the design of the National Flood Insurance Program and some of the problems that arise from flood risk maps, the limited mandate for flood insurance, and the reliance on insurance companies and their agents to sell policies.

Coastal homes lack flood insurance because of maps

by Kate Spinner, Sarasota Herald Tribune

Despite the sweeping Gulf of Mexico view from Tony and Kathy Zumbano’s ground-level kitchen — and the fact that a hurricane could pummel it with crashing 10-foot waves — the federal government does not consider their home a high flood risk.

In fact, when they bought the Venice beachfront property, their real estate agent told them they were lucky not to need flood insurance.

The reason: The maps underlying the nation’s flood insurance program do not account for direct hurricane strikes, which can cause catastrophic flooding. Instead, they rely on the average risk of flooding over a long period.

As a result, tens of millions of people in coastal counties across the nation underestimate the likelihood that their property could be inundated — and the possibility that they could take a huge financial loss or worse — if they base their home insurance or evacuation decisions on the flood maps.

… People are essentially on their own to assess their storm surge risk, says Bill Read, director of the National Hurricane Center. Too many assume they will never be in danger of flooding because they misinterpret the federal flood maps.

…To protect lives, scientists give states computer programs that show how high the storm surge could rise under the worst-case scenarios.

But it is up to the states and counties to share the information with residents. Keeping the public informed is complicated because some communities do not want to bring attention to their vulnerability, given the potential impact on real estate and business.

“Who’s going to tell you you need flood insurance? Not your Realtor, not your city,” Read said at a conference earlier this year.

Congress created the National Flood Insurance Program in 1968 to reduce the federal cost of natural flooding disasters. It requires anyone in the 100-year flood plain — an area where the flood risk over a 30-year span is an extraordinary 1 in 4 — to buy flood insurance if they have a federally backed mortgage.

But few outside that defined scenario buy flood insurance, even though the chances of flooding are much greater than the likelihood of fire or many of the other perils covered by standard policies.

Most major floods and storm surges cause significant water damage outside the 100-year floodplain. That gap is demonstrated almost every time a major storm strikes, including during Hurricane Irene, which slammed North Carolina in August.

Irene’s flooding from surge and waves caused $2 billion in damage in coastal regions alone, only about 10 percent of which was insured, said Tom Larsen, vice president of Eqecat, a disaster-modeling firm for insurance companies.

First let me say that it is great that the Sarasota Herald-Tribune is focusing on this aspect of the flood insurance program. The national media, to the extent that they write about flood insurance at all, only write that the premiums should be raised and allege without proof that the availability of flood insurance encourages and subsidizes risky development.

But how can flood insurance be facilitating development if property owners are not buying it and cities and Realtors and insurance agents are telling homebuyers that they do not need it? No developers build homes near the water because flood insurance is available there. They try to build outside the mandatory purchase areas so they can tell prospective buyers that they do not need flood insurance. Developers and Realtors and city and county officials commonly dispute proposed flood map changes that would expand the zones in which flood insurance is mandatory.

Although this article is from Florida, the issue it raises really is more of a problem in other places. Florida has a lot of low, coastal property, so flood insurance is required in many parts of the state. More than 2 million NFIP policies are in force in Florida, 37% of the 5.5 million policies nationwide. Yet, in the history of the program, Floridians have received less than 10% of the amount of total claims payments. In the National Flood Insurance Program, Florida has not been subsidized by ratepayers and taxpayers from other states.

Unfortunately, some statements in the article are not true or are misleading, beginning with the title. The Flood Insurance Risk Maps are not responsible for the many coastal homes that do not have flood coverage. The maps establish the zones where flood insurance is mandatory for a federally-backed mortgage and designate the building elevations within those zones. The zones do not divide those who should buy flood insurance from those who should not. They divide those who must buy flood insurance from those who should consider buying it.

Any property owner can buy flood insurance and any owner whose home is in a storm surge evacuation area should know to buy flood coverage. If you live within a few blocks of the ocean or some other large body of water and your neighborhood will be urged to evacuate when a hurricane approaches, there really is no excuse for not having flood insurance. Homes that are not in mandatory purchase areas get very cheap rates. It does not help that local government officials, home builders, real estate agents, lenders, and insurance agents often tell coastal home buyers that they do not need flood insurance if it is not required.  The fact that many homes that are in evacuation zones are not covered by flood insurance exposes a failure of insurance companies and their agents of their responsibility as NFIP contractors.

Let me explain how the program is supposed to work and why it doesn’t work well in some ways.

The flood insurance program, as with any government program, was created through political compromise. Before NFIP, no flood insurance was available, so flood victims depended on federal disaster assistance to recover from flood losses. The primary purposes of NFIP are to replace disaster assistance with flood insurance funded by premiums paid by the owners of flood-prone properties and to smuggle in some federal flood plain management rules as a condition of the insurance program.

Paranoia about federal government authority is endemic, so even while setting up a federal insurance program Congress limited the federal authority over local land use and zoning. Local governments can choose to participate in NFIP and must adopt certain flood plain management rules if they do so, but the process allows them plenty of input into the mapping that sets the zones, rates, and building requirements.

Almost every aspect of the flood insurance and the flood plain management rules is based on mapping the 1% chance flood risk, aka the 100-year flood. I am not sure how the 100-year flood became the standard for the program, but it is probably a reasonable level of risk to expect people to insure against. It is not a magic number or anything, but the insurance industry and its economists, actuaries, underwriters, and modelers have created a myth that they can accurately estimate the 1-percent chance losses for various risks.

Determining the 100-year flood risk is not even close to an exact science. After a local government opts to participate in NFIP, FEMA contractors use a combination of the historical record and modeling to propose flood maps estimating the 100-year flood elevations on a topographical map of the city or county. 

A lengthy comment period then begins during which local government officials, business leaders, and residents argue to change the lines on the maps to take some locations out of the mandatory purchase areas or to move some locations from a V-zone to an A-zone. V-zones are the areas estimated to be subject to the waves and forces of hurricane storm surge in the 100-year flood.  A-zones are subject to more conventional flooding from flowing or standing water without significant waves or tidal surge. Rates are higher within a V-zone because of the destructive force of the waves. Building requirements also are more extensive and more expensive within a V-zone. So one of the most contentious issues with any coastal flood map is locaton of the line dividing the V-zones and the A-zones.

A general rule used by FEMA is to divide V-zones from A-zones at the 3-foot wave line, but that is a very debatable and inexact standard. In most storms, the surge is nothing like a tsunami wave. The on-shore winds and the forward movement of a hurricane cause the water to pile up for hours in advance of landfall. The water level keeps getting higher and higher and then at landfall there is a faster, higher surge on top of the already high water. The storm surge has force because of the volume of water, not because of the heights of the waves. The wave-heights are more a function of the wind speed and the terrain than of the height of the water. So the 3-foot wave line is questionable as a means of defining the storm surge risk, but it is a standard that works in the favor of local governments and developers who try to limit the reach of V zones by hiring their own engineers to challenge the estimated wave heights.

Every city, town, or county has a separate mapping project for its jurisdiction, without coordination for a shared watershed, so it is common for the 100-year flood elevations to be different on adjacent properties separated by a city, county, or state boundary line.  This lack of watershed coordination between communities also creates problems with flood plain management and mitigation efforts.

All of the above is to explain that what we have are politically feasible flood maps, with the federal interest compromised by the local government interest, with the local officials commonly acting as surrogates for influential residents, developers and business interests who do not want the flood insurance mandate and do not want the building standards or premium rates that go with high risk. In every coastal community where I have compared the evacuation maps with the flood maps, there are areas identified as flood-prone on the evacuation maps that are not within the 100-year flood zones on the flood maps. Homes in those areas are not required to purchase flood insurance and not required to comply with flood-resistant building standards.  

We should not pretend that the engineers, actuaries, modelers, and flood plain managers can accurately figure a 100-year flood elevation for a particular location. Flood risk is ridiculously complicated. Within the watershed of a river, bayou, or any tributary stream, the flood risk changes constantly because of development upstream, downstream, across the stream, and because of erosion, deforestation, and other land use changes. Almost any place with  significant flood risk is affected by levees, dams, reservoirs, spillways, floodways, seawalls, floodwalls, or other projects or structures that attempt to control the flood risk. Those projects prevent frequent flooding, but they can create a false sense of security in the protected areas. If flood control structures fail or if a flood exceeds their design specifications, the result is often catastrophic.

Even if it were possible to accurately map the probability and severity to determine the 100-year flood elevation, there is a simple math anomaly that guarantees that wherever a major hurricane hits, the flood maps will have underestimated the storm surge. Consider the more than 1,000 miles of Gulf of Mexico coastline from Corpus Christi, Texas to Sarasota, Florida. Every year, there probably is about a 10% chance that a major hurricane will hit somewhere in there with a storm surge of 15 feet or higher. Yet for any specific community within that area, the probability of a major hurricane landfall is probably less than 1 percent. So almost every flood map from Texas to Florida is based the local 100-year surge of  around 10 to 12 feet, but about once a decade, some place on the Gulf Coast gets a storm surge of 15 to 20 feet or higher. The only map exceptions are the Mississippi Coast flood maps maps. Those maps expected a 12-foot 100-year surge before Katrina but the new post-Katrina maps estimate storm surge of 20-feet or higher as the 100-year event. So, what we have is a  system guaranteed to underestimate flood risk before a major hurricane and then require higher building elevations and stronger building requirements to the devastated communities trying to recover. The inefficiency of this system is evident at the Mississippi-Alabama state line where the 100-year flood estimate is six feet higher on the Mississippi side than on the Alabama side. Alabama will not get comparable maps until after it has been devastated by a direct hit from a major hurricane.

This disconnect between the evacuation maps and the flood maps is not just a Gulf Coast concern. The flood maps up the Atlantic in South Carolina, Virginia, Long Island, Cape Cod, and in-between are not designed for a direct hit from a hurricane. There are parts of Virginia Beach where the 100-year flood elevation is only 7 feet above sea level, but the water has risen that much from tropical storms that passed by without coming within 50 miles of the city. 

The problem is not whether the maps are accurately estimating the 100-year flood. The problem is that the 100-year flood is a stupid standard for designing a flood insurance program and building requirements and flood plain management rules in any place that is at risk of hurricane storm surge. As stated in the article from the Herald Tribune, lots of houses that would be flooded by a direct hit from a hurricane are not required to buy flood insurance. But, in my opinion, the more important mistake is that homes were built on slabs 12 feet above sea level. If they have a 15-foot surge, they will suffer substantial uninsured losses. If they have a 20-foot surge, their homes will be destroyed and some people probably may drown. I am not so concerned about raising their premiums. We need to be raising the houses, or at least requiring new homes to be built a few feet off the ground.

The simplest way to fix the flood maps is to sync them with the evacuation maps and then reset the premium structure and the building and flood mitigation standards to the new maps. If the home in the article above is estimated to be outside the 100-year flood zone but inside the area that would be flooded by storm surge from a direct hit by a major hurricane, then the owners should be required to buy flood insurance but the rates should be set to reflect the low probability. New homes in that area should be required to be built a few feet off the ground and the premiums discounted to reflect that reduced risk.

We probably are stuck with the system for estimating the probability and severity for setting premiums, but we should be able to use modern technology to update the maps to reflect development and other changes in the floodplain.

Unfortunately, any of these changes would require Congress to override the opposition of local governments and local developers to stricter floodplain management and building standards and give the federal government the authority and resources to establish and enforce those stronger standards, and that is not going to happen any time soon.

I am going to cut this post off here and continue the discussion of various issues with levees and other flood controls, and then finally get to my point about the failure of insurance companies as contractors to operate NFIP.