The National Flood Insurance Program: Financial Soundness and Affordability

The National Flood Insurance Program (NFIP) was established in 1968 to provide insurance that had proved difficult for the private sector to provide at affordable rates and to promote floodplain management. Authorization for the program, which is administered by the Federal Emergency Management Agency (FEMA), expires on September 30, 2017. Lawmakers have sought information from CBO about the NFIP’s financial soundness and its affordability for policyholders.

CBO analyzed roughly 5 million policies in effect on August 31, 2016, which approximate the policies currently in place. The agency assessed the program’s financial soundness by comparing expected annual costs and premiums; expected costs included estimates of expected claims projected using commercially available models that simulate large numbers of potential flooding events along with their probability. To assess the NFIP’s affordability for policyholders, CBO compared premiums with household income.

CBO’s estimate of expected claims accounts for low-probability, high-cost events, such as Hurricane Harvey, which first made landfall in Texas one week before this report was published. As a result, the estimate is probably greater than actual costs would be in a typical year, although lower than costs could be in the aftermath of a catastrophic storm.

How Do Expected Costs Compare With Premiums?

The difference between the program’s expected costs and premiums depends on which costs are considered. CBO estimated that overall, considering all expenditures and premium income, the program had an expected one-year shortfall of $1.4 billion. An alternative measure is the actuarial shortfall, which amounted to $0.7 billion, according to CBO’s estimate. The actuarial shortfall compares premium income with the subset of costs associated with paying claims for existing policies and writing and servicing those policies; it excludes $0.7 billion for mapping floodplains, mitigating flood risk, and making interest payments on debt accumulated from previous claims.

What Accounts for the Estimated Program Shortfall?

The $1.4 billion estimated shortfall has two main sources:

• CBO’s estimate of expected claims exceeds FEMA’s estimate by about $1.0 billion. Because FEMA’s estimate is its basis for premium setting, the difference between the two estimates causes premiums to fall $1.0 billion short of CBO’s estimate of expected claims.
• The cost of providing discounted rates for certain policies is about $0.3 billion more than the receipts from surcharges created to help cover the costs of those discounts. The discounts are mainly for properties built before flood insurance rate maps (FIRMs) were developed. They are intended to prevent households from facing significant new costs that could impose hardship and cause some homeowners to forgo coverage.

The costs of activities other than writing, servicing, and paying claims for existing policies include the costs of mapping, mitigation assistance, and debt service—about $0.7 billion altogether. Offsetting those additional costs were collections of two charges paid by policyholders—a federal policy fee and a reserve fund assessment. (CBO’s estimate of the actuarial shortfall excludes the costs of those activities but includes the premium income from those fees, along with all other sources of income.) On net, those costs and charges did not contribute to or reduce the program shortfall. They may not balance in the future, however. Moreover, although the reserve fund assessment can be accessed at any time and can be used to cover nonclaim costs, it was created to build a fund to pay claims in years when costs are particularly high.

How Do Coastal and Inland Counties Contribute to the Program Shortfall?

The difference between expected costs and premiums varies from one region to another. In particular, the overall shortfall of $1.4 billion is attributable largely to premiums’ falling short of expected costs in coastal counties, which constitute roughly 10 percent of all counties with NFIP policies but account for three-quarters of all NFIP policies nationwide. Although some coastal counties generated surpluses of premiums over expected costs and some inland counties had shortfalls, the net shortfall measured over all coastal counties is $1.5 billion, whereas the net surplus measured over all inland counties is $200 million. A contributor to the coastal counties’ shortfall is the fact that premiums on policies for most homes in those areas do not cover the expected cost of wave damage from storm surges.

How Might Expected Costs and Premiums Differ in the Future?

The estimates of expected costs and premiums in this analysis approximate those for the NFIP currently, but those costs and premiums will change over time. Some changes, such as increases in costs because of rising sea levels or increases in premium receipts attributable to the phaseout of discounted policies, would occur over years or decades. Other changes could occur more quickly. For example, CBO expects that the NFIP’s interest payments to the Treasury (for borrowing to cover past claims) will increase because of a projected rise in interest rates.

How Does This Analysis Compare With CBO’s Baseline Projections?

CBO makes regular 10-year projections of annual expenditures and receipts for the NFIP assuming that the program will continue to operate as specified in current law. CBO uses such baseline projections as a basis for estimating the budgetary effects of proposed legislation that would affect the program. In CBO’s most recent baseline, for the 2018–2027 period, the NFIP’s receipts are projected to fall short of expected claims and other costs by about $1 billion. That amount is likely to change as a result of Hurricane Harvey.

The baseline projections differ from the estimates in this report in two important ways. First, although this report reflects a snapshot of the annual expected costs and premiums associated with the policies in effect on August 31, 2016, CBO’s baseline considers factors affecting costs and premiums that are likely to change over time, such as the share of policies that are discounted and their effects on the program’s income and expenditures. Second, CBO’s baseline projections are based on FEMA’s estimates of expected claims and not on the estimates used for this analysis, which were derived from commercial models. (CBO is exploring the applicability of those models for future baseline estimates.)

What Does a Flood Insurance Policy Cost?

The median annual premium for residential coverage under an NFIP policy in effect on August 31, 2016, was $520. Those premiums varied significantly, however. The central two-thirds’ range of the payments for such policies was $420 to $1,330. (That central two-thirds is often used to represent the most likely part of a range.) Premiums tend to be lower for primary residences than for others, in part because the annual surcharge of $250 for nonprimary residences and nonresidential properties is higher than the surcharge of $25 per year for primary-residence policies.

CBO’s analysis of premiums compared with household income (with both approximated by the median values observed in each census tract) suggests that premiums for primary single-family homes generally amount to less than 1 percent of household income, although the percentage could be significantly higher for individual households. (CBO lacked the data to compare premiums with income for each household with an NFIP policy.) The agency also found that, on average, NFIP policyholders tend to live in census tracts in which median income is somewhat higher than median income averaged across all tracts.

How Might Alternative Policies Affect Three Competing NFIP Objectives?

This report outlines 12 policy approaches that would generally involve tradeoffs between varying objectives. Although some could be difficult to implement, each could help achieve at least one of three goals for the NFIP:

• Improve solvency by increasing premium income from policyholders in general, reducing the use of discounted rates, or increasing the share of costs borne by certain categories of policyholders or by taxpayers generally;
• Better align premiums with risks by reducing the use of subsidies, including discounted rates and cross-subsidies (in which some policyholders are charged rates that are higher than their expected claims so that other policyholders can pay rates that are lower than their expected claims), or by adjusting premiums to better reflect underlying risk factors; or
• Keep costs low for some policyholders (perhaps while raising them for others) by targeting subsidies to low-income policyholders, shifting costs to taxpayers, or adjusting premiums to reflect the value of insured properties.