Reforming the National Flood Insurance Program


Editor’s Note: This is the second post in a two-part series about reforming the National Flood Insurance Program. The first post, also authored by Logan Strother, was titled Insuring Risk.”

Major flooding events have highlighted many problems with the National Flood Insurance Program (NFIP) as a policy and as an institution. As a matter of review, NFIP was created in 1968 to address the (perceived) problem of under-insurance in flood-prone areas. Since that time, numerous flooding events have tested NFIP’s utility. First, in 1972 Hurricane Agnes made it abundantly clear that the majority of people eligible for NFIP coverage had not purchased flood insurance. In an attempt to remedy this problem, Congress passed the Flood Disaster Protection Act of 1973, which required federally insured mortgage lenders to require flood insurance on all properties in Special Flood Hazard Areas (SFHA). Similarly, twenty years later the National Flood Insurance Reform Act of 1994 sought to increase sanctions on lenders due to continued low levels of participation in the NFIP program which were made evident by the Midwest Floods of 1993. Indeed, as of 2010, 67.73% of all NFIP policies in force come from five states (Florida, Texas, Louisiana, California, and New Jersey) (Michel-Kerjan 2010, 171). Even so, in the state of Florida, which alone accounts for 38% of all NFIP policies, three-fourths of all properties are uninsured. Louisiana is little better, reporting only 26% insurance penetration. That is to say, market penetration remains minimal, even after two rounds of reforms aimed at increasing participation. Additionally, the Flood of ’93 highlighted the problem of repetitive- and severe-repetitive-loss properties (RLPs and SRLPs, discussed below). Finally, Hurricane Katrina drove home the reality of NFIPs many shortcomings on a scale much grander than the previous disasters, including low market penetration, RLPs, and tax-subsidized insurance price-breaks. Within a month of Hurricane Katrina experts were calling for reforms to NFIP, citing its many failures evidenced by the unfolding disaster on the Gulf Coast.

Numerous major reforms to NFIP were suggested by flood policy experts in and out of government after years of investigation spurred by the failures made evident by Katrina. The most common call for reform was to eliminate altogether subsidized premiums for flood insurance, and thus have individuals and businesses actually pay for their own risk (this move would end government-subsidized risk taking). That is, in many cases the government offers a significantly discounted insurance premium to property owners in high-risk flood zones in an effort to keep insurance “affordable.” Eliminating subsidies would require property owners to pay for the risk they face.

Another important suggested reform concerned repetitive loss properties: suggestions ranged from special rate brackets for such properties to insurance pay-out caps to government buy-backs. A repetitive loss property (RLP) is defined as an insured property that experiences two or more losses of more than $1,000 in any given 10-year period. A severe repetitive loss property (SRLP) is a property which was incurred at least four losses of more than $5,000, or the cumulative amount of all claims exceeds $20,000, or which has incurred two separate claims whose cumulative value exceeds the market value of the building. The logic underlying these proposals is to ensure that individuals who rebuild in high risk areas pay for their own risk.

The final reform I’ll focus on here has to do with the Flood Insurance Rate Maps (FIRMs). FIRMs are constructed by FEMA, the USGS, and occasionally the Army Corps of Engineers. These documents reflect data collected through topographical and hydrological surveys of the United States, which is used to construct maps describing the 100-year and 500-year floodplains. These maps are then used to determine the actuarial risk of a flood disaster for a given property, and thus to establish the premium rate for properties located in risk zones. Unfortunately, many of these FIRMs are significantly outdated. Out of date FIRMs mean that many property owners are paying less (though some are paying more) than they should be, simply because the maps no longer reflect reality. That is, in some cases that floodplain has changed – gotten larger or smaller, in some places the value of the property has changed – due to development, dilapidation, etc., and in some cases floods themselves are bigger or smaller due to a multitude of factors. Updating and correcting FIRMs would thus give property owners (and prospective buyers) an accurate picture of the risk they face.

On July 6th, 2013, President Obama signed the Biggert-Waters Act (Public Law No. 112-141) into law, enacting the above reforms (among others). Biggert-Waters promised to slowly phase out subsidies. Non-primary residences (vacation homes, condos, businesses, etc.) would see a 25-percent-per-year increase in subsidies, so that after four years premiums will reflect actuarial risk. Similarly, all subsidies will be phased out for all repetitive- and severe-repetitive-loss properties. Additionally, properties newly purchased, or new policies purchased on properties after July 6, 2013, as well as policy lapses after October 4, 2013, will move straight to full-actuarial premiums. Starting in 2014, all other subsidized rates (presumably including primary residences) will be phased out, as FIRMs are updated. Finally, Biggert-Waters requires that FEMA update all FIRMs nationwide, and requires that FEMA not allow the rate maps to fall so terribly out of date again. Importantly, Biggert-Waters provides substantial funding to carry out the updating and maintenance of FIRMs moving forward.

There are issues other than problematic policy instruments that still need to be addressed. NFIP has run a cumulative budget deficit every year since 1973 – that is, its cumulative income, including premiums, has been less than its operating expenses and insurance claims. These findings are somewhat complicated, as King (2009) argues that policy premiums and fees covered all claims payments until 2005. Taken together, I think the best way to understand King’s and Michel-Kerjan’s findings (cited above) is that the cumulative budgetary shortfalls prior to Hurricane Katrina are due to administrative costs (i.e. the discounted rates were sufficient to cover claims prior to 2005, but were insufficient to cover the costs of administering the program). This clearly suggests that there are long-term problems with artificially low premiums for the long-run solvency of the program. Its cumulative budget shortfall hovered around four billion dollars for most of the program’s existence, until Hurricane Katrina, when it jumped to over twenty billion in running-deficit. While this is in part related to the problem of subsidized premiums outlined in more detail above, it is also due in large part to the underlying cause of the subsidies, and a stated goal of the National Flood Insurance Program, which is to provide “affordable” flood insurance to consumers. The absence of “affordable” insurance is not in itself a bad thing, however.

As noted in part one of this series, prohibitively expensive insurance disincentivizes building and investing in high-risk, flood-prone areas (recall specifically the discussion of the “moral hazard“). By making insurance “affordable” in these high risk areas, NFIP has actually contributed to the concentration of people and capital in flood-prone areas. And, as noted before, there is now a substantial population of people who benefit from the status quo, and who will likely resist any changes to those policy structures from which they benefit. Moreover, even if some minor changes (such as elimination of subsidies) succeed, there is still the fact that large populations and in places, whole communities, now reside in places that they simply should not. In the event of a major catastrophe, such as Hurricane Katrina or the Floods of 1993, income from premiums will be insufficient to cover the losses in these areas developed due to the existence of NFIP. All of this is to say that “affordable” insurance is not an appropriate goal for flood disaster mitigation. Emphasis should be on improving floodplain management, developing resilient structure, moving high concentrations of people and capital out of flood-prone areas, and teaching adaptive practices in those areas that can be used when there is no flood (i.e. farmland, etc.). To be clear, I am not advocating wholesale evacuation of all alluvial plains, but rather smarter policy emphasizing wise use, and operating on market principles (that is, free of market distortions such as subsidies). Unfortunately, it will likely prove very difficult, and very costly, to undo half a century development spurred by bad policy.

Unfortunately – but unsurprisingly – key members of the House and Senate agreed on October 9th, 2013 to delay many of the reforms to NFIP put in motion by Biggert-Waters. Maxine Waters (D-CA), ranking member of the House Financial Services Committee and one of the chief architects of Biggert-Waters, said that the delay is intended to assure that the “changes are implemented affordably” (Insurance Journal online). Lawmakers including Waters, Mary Landrieu (D-LA), Johnny Isakson (R-GA), Robert Menendez (D-NJ), Jeff Merkley (D-OR), Thad Cochran (R-MS), and Heidi Heitkamp (D-ND) have argued that the changes to NFIP brought on by Biggert-Waters has caused “harm” and “heartache” to Americans. They note that premiums are rising, which is hurting family budgets, and importantly, they claim that the changes are affecting home sales and property values. Thus, the delay measure postpones the promised (and necessary) phase-out of subsidies and reinstates the “grandfather clause.” In other words, this delay measure effectively un-does the vast-majority of the good that Biggert-Waters promised to do. In short, NFIP needs major reform – but those reforms are likely not even possible until we begin to address the real underlying causes of our present flood problems, and stop merely treating the symptoms.

King, Rawle O. 2009. “National Flood Insurance Program: Background, Challenges, and Financial Status.” Congressional Research Service Report R40650.

Michel-Kerjan, Erwann O. 2010. “Catastrophe Economics: The National Flood Insurance Program.” Journal of Economic Perspectives 24(4): 165-186.