Chairman Johnson, Ranking Member Crapo, and Members of the Committee, thank you for the opportunity to testify on the financial condition of the Federal Housing Administration. I am a professor at the University of Maryland's School of Public Policy and a faculty affiliate of the Center for Financial Policy at the Robert H. Smith School of Business at the University of Maryland. I am also a visiting scholar at the American Enterprise Institute and a senior fellow with the Milken Institute's Center for Financial Markets. I was previously Assistant Secretary for Economic Policy at the Treasury Department from December 2006 to January 2009.
Reforms are urgently needed to ensure that the Federal Housing Administration (FHA) plays its important role in helping to expand access to mortgage financing for low- and moderate income families who have the financial wherewithal to become homeowners. The fiscal condition of the FHA has deteriorated considerably in recent years, to the point at which the FHA nearly required a bailout last year and might well be required to draw on the Treasury this year to ensure its continued solvency. The FHA continues to have an outsized share of the housing market, especially for purchase loans, and thus displaces private sector activity, while providing backing for some houses worth more than $700,000—a level at odds with its mission. Moreover, the FHA underprices its insurance according to the appropriate measures reported by the Congressional Budget Office (CBO), meaning that taxpayers are not fully compensated for the housing risk they are taking on through FHA's guarantees.
"Reforms are urgently needed to ensure that the Federal Housing Administration (FHA) plays its important role in helping to expand access to mortgage financing for low- and moderate income families who have the financial wherewithal to become homeowners." Over the past several years as its financial outlook has worsened, we have all learned the hard way that the FHA grew too large and took on too much risk. Indeed, the Government Accountability Office (GAO) earlier this month added the FHA to its high-risk list, reflecting the need for actions to "restore FHA's financial soundness and define its future role."1 The financial fallout of this risk is encapsulated in the 2012 independent actuarial review indicating that the economic value of the FHA's Mutual Mortgage Insurance Fund (MMIF) is negative $16.3 billion. This is a continuation of the deterioration of the MMIF's economic value from positive $4.7 billion in 2010 and positive $2.6 billion in 2011, and comes despite useful actions taken by the FHA to improve its risk management and lender enforcement.
The first step in solving a problem is to acknowledge that one exists. This was not always the case with this administration, as can be seen in the dismissive response of the Department of Housing and Urban Development in 2011 to warnings from outside analysts such as Wharton Professor Joe Gyourko that the FHA was underestimating its risks on single-family mortgage guarantees, and that the capital position of the FHA needed to be improved to deal with these risks. 2 In this regard, it is laudable that the 2012 actuarial review adopts technical recommendations from the GAO, the HUD inspector general, and others to better model the risks and potential losses facing the FHA. These changes, along with a more realistic outlook for economic variables such as home prices, account for a good deal of the worsening of the FHA economic value. This revised outlook is not welcome news, but it is best to understand the risks clearly so that the FHA can take the actions necessary to be in a position to fulfill its mission.
The mission of the FHA is not flawed. Indeed, it is admirable that 78 percent of FHA home purchase loans in fiscal year 2012 went to first-time homebuyers, and that the FHA helped support around half of the home purchase loans made to African American and Hispanic borrowers. The key is for steps to be taken so that the FHA can continue to fulfill its mission in as effective a fashion as possible while protecting taxpayers.
More needs to be done to safeguard the financial stability of the FHA and thus to ensure that it carries out its mission; to protect taxpayers against even greater losses and the possibility of a future bailout; and to boost overall U.S. economic growth by ensuring that the private sector and not the government plays the leading role in allocating capital. I focus below on policy measures that would achieve these goals while better targeting government resources that are deployed in the form of support for homeownership through the FHA to families who most need assistance to become homeowners.
To be sure, the unwelcome financial condition of the FHA reflects the effects of the collapse of the housing bubble, ensuing recession, and subpar growth of jobs and incomes over the past four years. The combination of these developments led to elevated losses on FHA-guaranteed mortgages originated from 2000 to 2009, and especially on loans made starting in 2007 when the shutdown of subprime lending led riskier borrowers to migrate toward FHA-backed loans. The effects of these loan guarantees are still felt, as indicated in the most recent figures for the National Delinquency Survey released by the Mortgage Bankers Association that show a slight increase in delinquencies for FHA-backed loans at the end of 2012 when delinquency rates for other types of loans continued to decline.
These various factors affecting FHA-backed loans provide an explanation of the FHA's situation but not an excuse. Indeed, the negative value of the MMIF came about because the underwriting standards, insurance pricing, and other practices of the FHA have taxpayers provide an underpriced guarantee with 100 percent coverage of the mortgages taken out by risky borrowers, many with scanty downpayments and thus little protection against home price declines.
Some of the practices which brought about the FHA's problems have changed, with an end to seller-funded downpayments, some changes to practices for home equity conversion mortgages (HECM; so-called reverse mortgages), and increased actions to address problematic originators and underperforming servicers. But more needs to be done to protect taxpayers, to target FHA activities more effectively, and to ensure that the costs, risks, and benefits of FHA activities are transparent , accurately accounted for in government books, and understood by policymakers and the general public. In doing so, the FHA should return to its traditional share of about 10 to 15 percent of the housing finance market so that government does not take on an inappropriately high level of risk and distort the housing market.
"With insurance premiums that appear still to underprice risk, a continued large footprint for the FHA compounds the solvency risks rather than addresses them." In considering reforms, it is important both to address the present solvency concern and to ensure that the FHA is focused on its core mission while avoiding risks that pose a threat to its future solvency. Maintaining an oversized FHA is not the best approach to addressing solvency either today or into the future. With insurance premiums that appear still to underprice risk, a continued large footprint for the FHA compounds the solvency risks rather than addresses them. Moreover, maintaining an outsized role for the FHA means increased distortions in the broad housing market as the government seeks to artificially boost demand for housing—an approach that in the past led to considerable suffering for borrowers who prematurely attempted to take on the financial responsibilities of homeownership. It is intrinsically pro-consumer to strengthen underwriting standards to ensure that borrowers are capable of staying in their homes. The FHA Emergency Fiscal Solvency Act considered by the Committee in 2012 is a useful starting point for reform but is not enough. Additional measures are needed to ensure the solvency of the FHA, reduce its market share, and improve its efficiency, effectiveness, and ability to manage risk.
Measures that should be taken as part of FHA reform include:
1. Improve the pricing of FHA insurance. As was included in the FHA Solvency Act, it would be useful to increase both the minimum and maximum annual premiums on FHA loans and to utilize the scope for pricing insurance in line with risks. In the meantime, the FHA should use its existing authorities to tighten its insurance pricing. These steps will mean higher costs for homebuyers, but this reflects the risks borne by taxpayers. At the very least, the FHA should use its existing scope to raise annual insurance premiums until the MMIF regains its minimum required 2 percent capital ratio.
An important consideration is that Fannie Mae and Freddie Mac are taking steps to increase the pricing of the insurance they offer on mortgages and also eventually to require private capital in a first-loss position ahead of the government guarantee. Absent corresponding action by the FHA, borrowers who might otherwise qualify for conforming loans backed by Fannie and Freddie will migrate toward the FHA and its less stringent underwriting standards. This could lead to increased risk for the FHA and thus greater net exposure for taxpayers since loans backed by the FHA are riskier than those backed by Fannie and Freddie. This concern is illustrated by the experience starting in 2006 when the shutdown of subprime origination resulted in borrowers turning to FHA-backed loans for mortgages, with dire consequences for the financial condition of the FHA.
2. Require higher downpayments for additional categories of relatively risky borrowers. The FHA is unusual in allowing borrowers to have relatively modest downpayments. While this helps first-time homeowners who have not accumulated the resources for larger downpayments, a lesson of the recent crisis is that housing prices go down as well as up, and that ensuring that borrowers have equity in their home is vital to avoid foreclosures and to safeguard the stability of the housing market. As noted above, it is intrinsically pro-consumer to ensure that homebuyers get into houses and mortgages that they can sustain. The FHA now requires 10 percent downpayments for borrowers with FICO scores below 580. It would be useful to add a tier with a required downpayment of 5 percent for borrowers with FICO scores from 580 to around 620 (with the precise cutoff depending on an evaluation of risk factors). As an alternative, borrowers could be offered a choice of retaining the lower downpayment with a shorter loan term such as 20 rather than 30 years. The goal is to ensure that risky borrowers build up an sizable equity stake in their homes relatively quickly. The Committee should also consider whether borrowers with very low FICO scores such as 500 to 580 should be eligible for FHA-backed loans in the first place.
3. Reduce FHA loan limits in order to shrink the FHA market share and focus government assistance on homebuyers who most need assistance. The current loan limit of up to $729,000 means that the FHA is serving a population far outside of its mission, and this would still be the case if the FHA loan limit is allowed to revert to $629,500. It is misguided to assert that the FHA should continue to insure these high-dollar loans because they are profitable and will help to recapitalize the MMIF. As discussed below, the FHA books profits under government accounting procedures that understate the risks of its activities. Indeed, the FHA loan limit exceeds that for mortgages guaranteed by the U.S. government through support for Fannie Mae and Freddie Mac, even though the underwriting requirements for those two firms are more conservative than those of the FHA. Moreover, the FHA activity displaces the private sector to the detriment of the overall U.S. economy. Even if jumbo loans above the GSE conforming loan limits were profitable for the FHA, it would be better to allow the private sector to gauge the creditworthiness of people seeking to buy homes of three-quarters of a million dollars or more (including the downpayment on top of the maximum loan amount). It is noteworthy that the recent report from the Bipartisan Policy Center's (BPC) Housing Commission calls for loan limits for government-backed loans to be set at $275,000 in order to focus public support on families most in need (and then would accompany this with increased spending to support affordable housing for both owner-occupied and rentals).3
4. Use Fair Value accounting to evaluate the costs involved with FHA lending activities. As explained by the group of eminent academics at the Financial Economists Roundtable of the Wharton Financial Institutions Center at the University of Pennsylvania, the current accounting of FHA guarantees under the Federal Credit Reform Act of 1990 (FCRA) systematically understates the costs and risks of FHA activities.4 The understating of risk comes about because the FCRA accounting treatment ignores several types of risks that are borne by taxpayers when making guarantees on private sector activities. As an illustration, under the FCRA accounting treatment, the federal government would book a profit if it simply buys a loan from the private sector at the accurate market price. The profit comes about because the FCRA rules discount the future cash flows from the loan by the interest rate on Treasury securities rather than the higher interest rate that was used by the private lender when it made the loan. In reality, the federal government does not have any inherent advantage over a private lender at managing the risk of a loan and there is no reason to believe that an otherwise identical loan is any more valuable if owned by the federal government rather than by a private lender. The profit booked in this example is illusory; it is an artifact of the FCRA accounting treatment. This illusion of profits is the case now with the accounting treatment of FHA guarantees. The FHA books a profit when it guarantees loans to riskier borrowers and on less stringent terms than loans that private sector lenders would be willing to make.
Use of the fair value accounting treatment would not reduce the merits of FHA activities in any way. Fair value accounting would simply measure the risks involved more accurately. The CBO assessment of these risks indicates that the 1.9 percent profit rate (that is, the negative 1.9 percent subsidy rate) for FHA loan guarantees calculated under the FCRA accounting treatment is more accurately measured as a 1.5 percent cost (a positive subsidy rate) using fair value accounting.5
Opposition to the use of fair value accounting seems to reflect the outcome of a higher cost rate and the concern that assigning a cost rather than a profit to FHA activities would lead to less government support for the FHA. I do not agree. The activities of the FHA are a vital part of the overall government support for housing. These activities are so important that they should be undertaken and paid for with a clear recognition of the costs and risks. Use of fair value accounting is not a mechanism by which to reduce the scope of FHA activities, but rather a move toward a transparent understanding of their costs. Indeed, the experience of the past several years illustrates that the risks of FHA activities have been consistently underestimated.
Fair value accounting is used to accurately gauge the costs of government support for Fannie Mae and Freddie Mac, and a variant of fair value accounting has been used to measure the costs and risks of government support through the TARP. It is no less important to accurately measure the costs of FHA guarantees.
5. Expand indemnification authority so that the FHA can pursue claims against all lenders. The FHA has requested improved indemnification authority along several dimensions, as well as authority to terminate origination and underwriting approval, to change the compare ratio requirement to provide greater flexibility, and authority to transfer servicing. These are useful steps to take. Indeed, the FHA should fully pursue inappropriate actions on the part of originators and servicers. At the same time, it should be recognized that greater indemnification authority is an important step, but not a cure-all for the FHA solvency issues, present or future.
A useful accompanying step on the part of the FHA would be to provide clearer information on the rules and procedures governing underwriting standards and quality control reviews, and on the factors that lead to an indemnification request. The FHA should provide feedback to lenders on an ongoing basis to minimize defects and losses to the FHA and to reduce the risk borne by lenders. Taking these clarifying steps could reduce the use of so-called lending overlays under which originators impose more stringent lending standards than required by the FHA.
6. Require increased capital at the FHA and plans to maintain capital levels. The current requirement of a 2 percent capital ratio has proven inadequate, suggesting that reform should increase the capital maintained against losses. This would mirror developments in other parts of the financial sector, where firms such as banks are appropriately being required to maintain increased capital levels. Any time the capital level is projected to dip below the required level, the FHA should be required to provide a plan to Congress on how it will restore the needed capital level. This added accountability of the FHA is important for providing Congress with a mechanism with which to monitor FHA performance.
7. Further reduce maximum allowable seller concessions to avoid inflated appraisal values. FHA has taken important steps through administrative actions to address seller concessions. It would be useful to codify these changes in legislation.
8. Improve transparency with better information on foreclosure risks and costs. FHA reporting to Congress and the public could usefully provide more detailed information on the factors and combinations of factors that are associated with increased risk of loss for the MMIF. It would be especially useful to establish a program to review the cause of early period delinquencies of FHA-backed loans.
9. Make the FHA risk office permanent. This useful innovation should be codified in legislation.
10. Impose consequences in the event that the FHA requires a bailout from the Treasury. At the minimum, the necessity of drawing money from the Treasury should trigger an automatic report to Congress with an explanation for the bailout and a plan to avoid future bailouts. More frequent independent actuarial studies of the MMIF should be required when the fund is below its required capital ratio.
In addition to these steps, Congress should consider other actions in the context of FHA reform legislation, including:
1. Income tests on FHA borrowers to focus the mission. An income test would ensure that the FHA serves borrowers who most need assistance. It is hard to understand why a family with a high annual income purchasing a home that costs over $700,000 should be eligible for an FHA-backed loan. To be sure, most FHA-backed borrowers are not in this category, but there is no reason for such homebuyers to be eligible in the first place. It is especially worrisome that the FHA views these borrowers as a way to book profits rather than as the diversion of government resources away from families who truly need assistance to become homeowners.
2. Increase the role of private capital and reduced the government exposure to credit risk. It would be useful to examine possibilities for FHA-backed loans to be made in which there is first-loss private capital ahead of the government guarantee. This could be at the individual loan level such as by including private mortgage insurance (PMI), or at the level of the mortgage-backed security (MBS). Having private capital ahead of the government guarantee would protect taxpayers while providing incentives for improved origination quality (since investors will require this to take on the first-loss risk of FHA-backed loans). An additional step to reduce the government exposure to risk would be for the FHA guarantee to cover less than 100 percent of potential mortgage losses. This is already done by the Veterans Administration (VA) in providing guarantees for mortgages, and VA-backed loans have considerably more favorable performance in terms of fewer delinquencies than FHA-backed loans. This is not surprising since private investors exposed to credit risk in the VA loan program have "skin in the game" and thus incentives to focus on careful loan origination.
3. Require private capital in a first-loss position ahead of FHA guarantees on multi-family and health-related mortgages. The multi-family divisions of Fannie Mae and Freddie Mac both require private capital ahead of their guarantees, and have considerably better loan performance than for multi-family residential mortgages bundled into commercial mortgage backed securities. This again reflects the beneficial incentives brought about when market participants have their own capital at risk.
4. Restrictions on FHA backing for borrowers with recent foreclosures. The FHA might specify that borrowers who have been foreclosed on in the past several years are not eligible for FHA-backed loans.
Putting these reforms into statute is important for providing certainty to borrowers and to all participants in the housing industry. The details of each of these ideas should be tested by the committee against careful underwriting analysis.
An important broad point is that changes in other parts of the housing finance system will affect the FHA. The qualified mortgage (QM) rule from the Consumer Financial Protection Bureau is likely to lead private originators to avoid non-FHA loans with debt-to-income ratios above 43 percent. With the FHA still willing to provide a guarantee for borrowers with higher debt service burdens, it would be natural to expect a migration of risky borrowers to the FHA. This illustrates the importance of strengthening underwriting standards at FHA. Careful underwriting helps protect taxpayers against risk, but strong underwriting standards are also intrinsically pro-consumer in that they help ensure that borrowers get into homes that they can sustain.
The solvency challenge facing the FHA requires legislative action. Indeed, reform of the FHA is essential to ensure that the FHA remains effective at carrying out its mission; to protect taxpayers; and to ensure that government actions do not unduly interfere in the allocation of capital by the private sector and thereby detract from U.S. economic growth.
Several years into the economic recovery, the FHA market footprint is still enlarged relative to the historical norm and the agency faces challenges managing risk. FHA activities are evaluated using an accounting framework that understates the potential costs to taxpayers from the risks taken on through the FHA guarantees. And the FHA serves buyers of homes that are too costly to be plausibly related to the mission. It is vital to take immediate steps to stabilize and refocus the FHA. These steps should not wait for other developments.
FHA reform further can be helpful for setting the stage for subsequent reforms of the housing finance system, notably including reform of the government-sponsored enterprises (GSEs) of Fannie Mae and Freddie Mac. Looking ahead, as the FHA returns to its historical market share of 10 to 15 percent, and as the GSEs eventually are reformed, this will put in place the conditions for a larger share of mortgage financing to be performed without a government guarantee. To avoid the prospect of an immense amount of new lending landing on bank balance sheets and making big banks even larger, it will be useful instead to have a larger scale restart of private label securitization. It would be useful in this regard for the Committee to examine the impediments to non-guaranteed securitization.
It is important to move forward immediately with FHA reform, but not to stop there. Over time, reforms of the U.S. housing finance system are needed to best serve American families, to protect taxpayers, and to once again make it possible for the housing sector to make a strong and sustained contribution to U.S. economic growth and job creation.