In July 2012, Congress passed the Biggert-Waters Flood Insurance Reform Act of 2012 (Biggert-Waters), which increases the premiums charged by the National Flood Insurance Program (NFIP). This moves the program towards charging all policyholders actuarially-fair, unsubsidized premiums – a recommendation of the Domenici-Rivlin Debt Reduction Task Force.1 These reforms should eventually allow the NFIP to pay claims almost entirely out of its own revenues rather than continuing to rely heavily on borrowing from Treasury, as has been the case since 2005.
Many of those recently enacted higher premiums, however, currently face possible repeal in Congress because of the financial burden that they pose to policyholders. The political fallout was immediate when higher rates from Biggert-Waters went into effect in October 2013 for some policies that have historically been subsidized. The omnibus spending bill that passed on Thursday, January 16 effectively bars the NFIP from enforcing some of the higher rates, and the Senate is poised to take up legislation that would turn back the clock on more of them.
Even if a bill repealing most of the rate increases in Biggert-Waters is enacted, Congress should continue to insist on improvements to the NFIP rather than use this legislation as a reason to give up on making the NFIP fiscally sound.
The process of increasing NFIP premiums reveals important insights for policymakers interested in reforming civilian and military entitlements. Implementing entitlement changes that stick – particularly ones that have real “losers” and only the federal budget (i.e., the generic taxpayer) as a “winner” – is extremely difficult, and designing reforms with that difficulty in mind can minimize the impact on constituents and political blowback to policymakers.
We will be releasing another blog post detailing some recommendations for reforming the NFIP.
Approximately 5.5 million structures are insured by the NFIP
The NFIP, which is administered by the Federal Emergency Management Agency (FEMA), provides insurance for buildings and their contents in the event of a flood. At the end of Fiscal Year (FY) 2012, the NFIP held more than 5.5 million policies with a total insured value of $1.3 trillion. Those who live in Special Flood Hazard Areas (SFHAs) are required to hold flood insurance in order to be eligible for government-backed mortgages.2
Because flood damage tends to occur to many people in the same areas at the same time, proper risk-pooling for private insurance purposes is very challenging. For that reason, private flood insurance was almost nonexistent prior to the establishment of the NFIP in 1968. These days, the government takes the insurance risk and pays private companies to administer the program. The NFIP limits coverage on residential policies to $250,000 for the structure and $100,000 for contents and commercial policies to $500,000 each for the building and contents. Almost all policies are administered through “Write Your Own” (WYO) programs where private insurers issue, adjust, settle, pay, and defend flood policies. FEMA reimburses WYOs for claims plus expenses at rates that FEMA establishes.
Up until 2005, the NFIP occasionally had to borrow from the federal government to meet claims, but was able to pay Treasury back in a timely fashion.3 After the 2005 hurricane season, including the devastating Hurricane Katrina, the NFIP was forced to borrow $16.8 billion from Treasury to cover the unprecedented level of claims. When Hurricane Sandy made landfall in October 2012, NFIP was forced to borrow even more. As of July 2013, the NFIP owed Treasury approximately $24 billion in total and had no plan for repayment.
Actuarially fair premiums will move the NFIP towards self-sufficiency
When the NFIP started in 1968, FEMA had not completed the first set of Flood Insurance Risk Maps (FIRM) in order to accurately price insurance. Those buildings built before the relevant local map was completed or before 1975 – “pre-FIRM” or “subsidized” structures – were offered significant subsidies on flood insurance to ensure uptake.4 About 92 percent of those structures are single-unit primary residences. Though originally intended to be temporary, the subsidized rates persisted and, prior to passage of Biggert-Waters, accounted for approximately 20 percent—about 1.1 million—of the policies that the NFIP insures. These subsidies are effectively a transfer of wealth from American taxpayers to individuals who own structures that are at particularly high risk of flood damage.
Even with generous subsidies, pre-FIRM structures often have higher premiums than full-rate structures because they were built before modern building codes (e.g., elevation standards and design features) and are at high risk of flood damage. For example, in October 2011, subsidized, inland pre-FIRM structures at high risk of flooding had an average annual premium of about $1,200 while newer, unsubsidized structures in the same risk class faced an average premium of $500.
Biggert-Waters phases in actuarially-fair premiums
Just a few months before Hurricane Sandy, Biggert-Waters was passed as a section of the 2012 transportation reauthorization bill (“MAP-21”, P.L. 112-141). It reauthorized the NFIP through September 30, 2017 and removed subsidies for most pre-FIRM structures. Some of the other provisions required the creation of a repayment plan for current NFIP debt, set standards for revising outdated flood maps, created a reserve fund to pay future claims, and required the Government Accountability Office (GAO) and other government agencies to complete a variety of affordability and impact studies.5
Out of the 1.1 million subsidized policies at the time of Biggert-Waters’ enactment, the law immediately eliminated subsidies for 438,000 on October 1, 2013. The other phased-in reforms were then projected to reduce the number of subsidized policies from 715,000 to under 100,000 after 14 years.
Mouseover the graphic to view the impact of Biggert-Waters before and after the proposed Senate bill
Actuarially-fair premiums face repeal because they are extremely costly to owners
Several provisions of Biggert-Waters face possible repeal over the next few weeks. Some criticism has focused on the lack of adequate data to implement the law; for instance, FEMA has insufficient data to determine which residences are primary and which are not. It also does not have elevation data – an essential determinant of full-risk rates – on 97 percent of pre-FIRM properties nor a plan for obtaining this information. These are legitimate issues that must be addressed. If rates for policies are based on inaccurate or outdated information, financial shortfalls could result totally apart from any intentional subsidies.
But the primary reason that many lawmakers are seeking repeal on behalf of their constituents is the impact of large premium increases on property owners. The subsidized premium amounts do not account for elevation (one of the main determinants of flood risk) and, on average, represent only about 40-45 percent of the full flood-risk rates, according to GAO. Unsubsidized premiums would generally be significantly higher, but the increase will vary widely depending on the elevation assessments.11 Many property owners will struggle to pay unsubsidized premiums and might not be able to remain in their homes or continue their businesses. Should they want to sell, they may find their properties unmarketable. Individuals who hold federally backed mortgages (most mortgage holders) are required to hold flood insurance—so going without is not an option.
Although FEMA estimates that most subsidized premiums will approximately double to reach the full risk rate, some buildings could see much larger increases. In one extreme example, the owner of a pre-FIRM home in Sutter County, CA at an elevation of 40 feet faces an increase of over 12 times the current premium. The structure, which is valued at $150,000 and has contents valued at $60,000, would have a full-risk annual premium of $5,800. Thus, the new monthly premium of $483 would be higher than the current annual premium of $459.
The increases affect individuals in every state. California, Florida, Louisiana, New Jersey, New York, and Texas each have more than 40,000 subsidized policies. Both Democrats and Republicans have constituents who were affected when subsidy reductions (i.e., premium increases) went into effect on October 1, 2013. The backlash was immediate. The omnibus spending bill that is currently pending before Congress denies funding for implementation of the Biggert-Waters provision that will eventually begin to phase out subsidies to buildings remapped into higher-risk areas.
Moreover, a bipartisan bill to delay more of the increases has 30 cosponsors in the Senate and will be voted on soon, and a related bill with 178 cosponsors has been introduced in the House. The Senate bill would effectively repeal many of the Biggert-Waters rate increases that have already started by delaying them for more than four years, when the NFIP must be reauthorized.
The most relevant provisions of the Senate bill would restore subsidies to policyholders for most structures that already had the subsidies removed entirely—policies that lapsed or that were purchased after the enactment of Biggert-Waters—and would prevent map updates from affecting premiums for anyone, making permanent the omnibus’ denial of implementation funding.12 The bill does not address those policyholders who are facing 25-percent annual increases in premiums.
In short, many of the provisions of Biggert-Waters are effectively being repealed by delaying them until after the NFIP will have to be reauthorized in October 2017. To be precise, the Senate bill would delay the higher premiums for two years from the date of enactment for FEMA to finish the still unfinished affordability study originally mandated to be completed nine months after the enactment of Biggert-Waters, another 18 months for FEMA to draft an affordability framework that might include targeted assistance to individual policyholders, another six months for Congress to pass the affordability framework under an accelerated process, and then further delay them until implementation of the affordability framework is completed.13 The Congressional Budget Office (CBO) estimates that this will take until at least 2018.
CBO estimates that NFIP expenditures will continue to be greater than revenue, and further finds that the Senate bill would exacerbate this problem by decreasing NFIP premium collections by $2.1 billion over the 2014 to 2024 period, relative to current law under Biggert-Waters. Nevertheless, technical issues force CBO to conclude that the bill would have zero net impact on the deficit in the ten-year window. We believe that a score of zero is misleading and that a realistic estimate of the impact of the Senate bill is $2.1 billion in increased deficits over the 2014-2024 period.14
Real losers and no clear winner increases likelihood of repeal
The NFIP is only a small program and the $24 billion that the NFIP owes Treasury is minor relative to the federal budget of more than $3.5 trillion in FY 2014. The attempt to repeal most of the Biggert-Waters reforms so soon after they were enacted, however, should be a lesson for policymakers trying to reform entitlements in the future.
Legislation like Biggert-Waters faces tough prospects because it has clearly identifiable losers but only helps general taxpayers via a relatively small reduction in federal budget expenditures. With no defined group having a vested interest in keeping the legislation in place, policymakers are likely to face intense pressures from their constituents, and have little incentive to resist reverting back to more generous benefits.
In the particular case of flood insurance, the partial repeal of Biggert-Waters would be a large blow to the fiscal solvency of the program. Without increasing premiums on currently subsidized structures – as Biggert-Waters set in motion – the program will be unable to serve its intended purpose without continually needing taxpayer-funded bailouts. Furthermore, continuing to subsidize people to live and work in buildings with high risk of flood damage is a poor use of public resources, especially at a time when those resources are being constrained by federal budget spending caps and sequestration.
Some of the reforms in Biggert-Waters may have been too forceful and too fast, but we hope that any partial rollback of that legislation will be only an interim step towards sensible reform of the NFIP that includes comprehensive map updates and means-testing of any remaining subsidies, because the pending bill is far from a long-term solution.
In the coming days, the Bipartisan Policy Center will release another blog elaborating on recommendations to reform the program.
Alex Gold contributed to this post.
1 Actuarially sound or full flood-risk rates are the rates required to make the collection of premiums equal to expected future claims.
2 SFHAs are defined as areas that experience at least a 1 percent chance of flooding each year.
3 The NFIP is supposed to be a self-sufficient financial entity and therefore actually borrows any needed funds from Treasury up to a statutory limit and has to pay those loans back with interest. The NFIP’s unusual financial structure, however, made the program unlikely to be self-sufficient. We will address these issues more in a blog post on recommendations for reforming NFIP next week.
4 Buildings built either before the local FIRM was completed or before 1975 are eligible for subsidies. A small portion of the subsidized buildings fall into one of several other categories, including those with partially completed or insufficient levees or flood-control structures, provided sufficient progress has been made towards finishing those projects.
5 Much of this work has not been completed, including the creation of a repayment plan.
6 Mitigation assistance may include elevation, relocation, or flood proofing of structures or state purchase and demolition of properties that are at high risk of flooding. Such assistance may be offered after a major disaster or if the structure in question is a repetitive loss property. (See footnote 8 about repetitive loss properties.)
7 For non-primary homes, the transition started on January 1, 2013. It is not clear whether or how this provision was implemented, as FEMA has stated that its data on primary and non-primary residences may be out of date.
8 A repetitive loss property is one that has incurred serious flood damage at least twice – the average cost of which exceeded 25 percent of the value of the structure. Biggert-Waters introduced the concept of a severe repetitive loss property (SRLP). A single-family SRLP has one to four residences with four or more claims that each exceed $5,000 and cumulative claims that exceed $200,000, or two or more claims that cumulatively exceed the value of the property. Multi-family SRLP has not yet been defined.
9 FEMA currently only has data on whether structures are residential or non-residential. This means that FEMA does not have the data to distinguish non-business structures that will keep their subsidies – like schools or religious buildings – from the businesses that will lose subsidies.
10 To be precise, properties will lose their subsidies if buildings are damaged and the cost of repair exceeds 50 percent of the fair market value of the property or if the owner improves the building such that the value of improvements exceeds 30 percent of the pre-improvement fair market value of the property.
11 FEMA’s current plans rely on some policyholders to voluntarily complete these assessments. These assessments are unlikely to be completed because they are expensive to carry out and policyholders know that their completion may well lead to higher premiums.
12 For properties that experienced a lapse in coverage, premium increases would be delayed only if the property was not required to have coverage at the time of the lapse.
13 Under the Senate bill, if the affordability framework were to fail in Congress, the delay would end and the higher rates would be in effect according to Biggert-Waters.
14 The NFIP’s borrowing authority is set in statute. Therefore, CBO cannot assume that spending will occur that exceeds this borrowing authority, even if such spending is likely to happen. Under Biggert-Waters, CBO projected that the NFIP would reach its current borrowing limit in 2022 and not be able to spend any more funds thereafter. The Senate bill would decrease premium payments to the NFIP earlier in the scoring window. Therefore, under the Senate bill, the NFIP would be on track to hit its borrowing limit earlier – in 2020, rather than 2022 – and would thus cease further payments in 2020 (absent an increase in the borrowing limit). Because the increased outlays in early years are offset by outlays that would no longer be permitted (under the present borrowing limit) in later years, projected outlays are simply shifted earlier in the ten-year window, resulting in a CBO score with zero net impact on the deficit.