It’s probable that half of all the low-FICO score and high-loan-to-value loans that were backed by the Federal Housing Administration during the peak of the housing bubble will default.
More than one out of every four loans insured by FHA in 2007 will have resulted in an insurance claim. And that’s on top of the $10 billion the agency will ultimately lose as a result of its 2008 book of business.
Those scenarios will happen. They were confirmed by an actuarial review contractor, Integrated Financial Engineering, and supported and then delivered to Congress by the Department of Housing and Urban Development, which oversees FHA.
But they might not stop there.
The federal debt crisis might deepen, with yields on Treasury and corporate bonds rising significantly, causing business investment and hiring to fall. The European debt crisis could expand beyond Greece and other troubled nations, resulting in a significant European recession that causes American business confidence and exports to fall.
Foreclosures that were delayed could accelerate and set off a cycle of self-reinforcing house price declines. As measured by the National Association of Realtors, prices on existing homes may ultimately tumble 46% from their 2006 peak by the third quarter of 2012. Prices are already down 30% through the end of 2011.
Then housing starts decline and ultimately plummet more than 85% from their 2005 peak. And although they finally bottom out in the third quarter of 2012, gains come at a snail’s pace for several years.
And then there’s the unemployment rate, which dropped to 8.6% in November. It reaches a high of 13.7% in the fourth quarter of 2012 and remains in double digits until 2015.
These aren’t contrived doomsday scenarios. In fact, Moody’s Analytics says there’s a 4% chance the U.S. economy could experience something worse.
The pessimistic, worst-case scenario was one of many forecasts presented in a July Moody’s report that Integrated Financial Engineering used in forecasting the financial health of the FHA’s mutual mortgage insurance fund. The agency uses the fund to pay banks claims on loans it insures, including the $10 billion from 2008 it will ultimately pay.
Of course, a measly 4% chance indicates these forecasts are highly unlikely. However, if they did come to fruition, the FHA’s reaction along the way in trying to maintain the fund would be similar to the agency’s reaction during a second recession, which some say is highly likely.
FHA officials won’t hesitate to remind anyone that it can still hike premiums and down payments to ensure the agency can cover claims on loans it insures. But at what point do the premiums get so high banks flee and turn to private insurers? That would help to accomplish FHA goals of reducing market share and returning capital to the private markets, but would leave the agency with a heavily undercapitalized insurance fund that needs those bank payments to stay afloat.
Three years after the financial crisis, the FHA remains under fire from critics who say the agency is undercapitalized to the point it will need a multibillion-dollar bailout from the Treasury in the coming years.
The FHA rigorously rejects the notion it has become an increasingly and endlessly risky organization and that it will inevitably need to ask for special assistance from the Treasury.
However, a September report from the U.S. Government Accountability Office revealed that an FHA budget official told the GAO that a depletion of the mutual mortgage insurance fund was a possible scenario within the next few years. Such a case would force the FHA to draw on its permanent and indefinite budget authority to cover additional increases in estimated credit subsidy costs.
If the FHA were a private entity, that revelation might alarm investors exposed to the risk and force management to adjust.
Some say the FHA is not properly, or even at all, considering borrower default and unemployment risks in determining its ability to cover future mortgage insurance claims.
The 78-year-old mortgage insurer has never asked for a federal bailout and officials say it won’t need to. But the agency and its auditor continue to underestimate the mutual mortgage insurance fund’s capital ratio.
Additionally, when the FHA says the fund will grow at an adequate pace based on its claim that housing prices won’t fall another 4% and then acknowledges that future housing prices are, in fact, the principle unknown in its assessment, the agency’s declaration of not needing a bailout in a few years begins to tremble, giving credence to critics.
QUESTIONABLE FORECASTING
Legislation passed in 1990 requires the mutual mortgage insurance fund to retain a capital ratio of at least 2%. The ratio, a measure of the financial soundness of the fund, compares the capital resources to the FHA’s unamortized issuance-in-force, or the dollar amount of mortgages it insures.
A May 2009 cover story of HousingWire magazine titled “A ticking time bomb?” asked if the FHA was the next mortgage crisis. If the bomb exists, it hasn’t detonated — yet. But if the falling capital ratio is any indication of the ticking clock, the Treasury may feel a multi-billion dollar financial explosion in the next few years.
But perhaps critics of the FHA just don’t get it.
“The difficulty is that it’s hard for outsiders to get a hold of the data so the claim can always be made that you’re missing the something that we see, but that’s not exactly a strong defense,” Andrew Caplin, New York University economics professor, told HousingWire.
Caplin testified to the House Financial Services Committee in December on how to enhance the FHA. He denounced the way IFE treated the agency’s popular “streamline refinance” program (refinancing FHA-backed mortgages to prevailing lower rates without new underwriting) as if it represented final termination of FHA’s insurance obligation. In truth, there is no cancellation of the underlying insurance and little by the way of additional fees to the FHA, he said.
“By lumping refinancing together with mortgage terminations in which the FHA’s insurance obligation is extinguished, (IFE) overestimates both FHA’s past success rates and its projected future success rates,” Caplin told the committee.
The future success of the capital ratio of the FHA fund is a heated point of contention in the housing industry.
HousingWire reported in the 2009 cover story that IFE projected the capital ratio would fall to 2.28% by fiscal year 2011 and then grow to 2.9% by 2015. At that time, the FHA’s latest fiscal year ended Sep. 30, 2008, when the ratio stood at 3%.
“We conclude the capital ratio is and will remain above 2% in future years under our base-case scenario; and the capital ratio could drop below 2% in future years under more pessimistic economic scenarios,” IFE president Tyler Yang said in a 2008 letter to the FHA commissioner.
The ratio fell below that threshold to 0.53% the following year. The fall was due to a 5% plunge in home prices, which further impaired the value of FHA books and loans already underwater, particularly from the years of the housing bubble from 2006 to 2008. It fell to 0.50% during the next year.
A collapse in home prices could, for example, trigger a new wave of strategic defaults, significantly reducing the billions of dollars of borrower payments that flow into FHA’s insurance fund by way of premiums from banks.
It’s 2012 and those more pessimistic economic scenarios are upon us.
The FHA’s capital ratio now stands at 0.24% ($2.6 billion in estimated economic value against an active portfolio of $1.08 trillion), some 90% under IFE’s 2008 projection of 2.28%. And now the agency projects the ratio to grow to 2.72% by 2015.
But FHA continues to dramatically underestimate the future financial status of its mortgage insurance fund.
In its 2009 report to Congress, the FHA told Congress its ratio would hit 1.74% in 2011 and then rise back above the legal minimum to 2.27% in 2012. In 2010, FHA lowered its 2011 forecast to 0.99% and its fiscal 2012 ratio to 1.24%, back below the statutory minimum.
The updated 2012 forecast now stands at 1%, and the FHA optimistically expects the ratio to bounce back above 2% by 2014 and increase to 4.18% by 2018. HUD spokesperson Brian Sullivan attributes the optimism to the top-notch credit caliber of new FHA borrowers with average credit scores of 700 and the projection that it will add billions of dollars to its capital resources in 2012.
“The recovery of MMI capital is well assured by the historically high premium rates charged today, and by controls put in place over the past two years,” FHA said in its 2011 report to Congress. “The principle unknown for the future remains how and when housing markets will recover.”
Increases in premiums over the last two years have increased in the economic value of the fund by $1.37 billion, according to the FHA.
Tom Lawler, founder of Leesburg, Va.-based Lawler Economic and Housing Consulting, says the 2014 forecast is “plausible in a very benign scenario, but it is optimistic.”
HUD Secretary Shaun Donovan testified in December to the House Financial Services Committee that as a result of IFE’s analysis, HUD is not only having discussions with Congress about whether additional premiums are necessary, but that the department intends to announce how it will address premium prices in its fiscal 2013 budget proposal. The proposal will be published in February.
Ruth Lee, executive vice president of Denver-based Titan Lenders, doesn’t think FHA will need a bailout. She says officials have done a fine job of teetering on a narrow fulcrum trying to resolve its weaknesses without tanking the housing market.
“They have shown a very strong willingness to manipulate mortgage insurance premiums to help them recover that capital requirement,” Lee says. “If they had gone into some kind of draconian MIP increase, it would have harmed the housing market, which exacerbates their problem.”
A good barometer of whether the FHA adequately gauged the health of the mortgage insurance fund in its 2011 report to Congress is looking at how the agency responded to three changes the GAO recommended in the 2010 report.
1) Switch from a model that uses a single economic forecast to produce the base-case estimate of its capital ratio to a stochastic one, which involves running simulations of hundreds of different economic paths to produce a distribution of ratio estimates (an approach the Congressional Budget Office in 2003 also said was more accurate).
2) Revise the manner in which the actuarial reviewer imputes LTV ratios for the aforementioned “streamline refinance” mortgages in assessing the default risk of this growing segment of borrowers.
3) Use localized indexes, rather than a national house price index, to capture the impact of future house price movements on loan performance.
Mathew Scire, director of Financial Markets and Community Investment at the GAO, said the FHA and Integrated Financial Engineering have addressed issues No. 2 and No. 3 in “a reasonable way.” An examination by HousingWire of the 2011 actuary report and the subsequent FHA report shows the agency has, indeed, addressed those issues.
However, it has not addressed the first, and perhaps most important, issue.
Scire said the base-case economic scenario, or the one that FHA thinks is most likely, isn’t considering the dozens of possible scenarios that might produce negative results — scenarios that a stochastic model would detect. Running only that one scenario tends to underestimate insurance claims and mortgage prepayments and, therefore, overestimate cash flows.
“By not doing it, they’re likely overstating their positive cash flows,” Scire says. “Right now they’re only using one set of housing prices and interest rates as opposed to hundreds.”
FHA expects positive cash flows in 2012 to reach $9 billion, a figure Sullivan points to when critics claim the chance of the agency needing a bailout is 50%. But those positive cash flows are included in the insurance fund’s economic value, which is used as the numerator in the ratio. Considering Scire’s contention that the FHA is likely overstating its cash flows, perhaps critics aren’t far off.
Sullivan says that newspapers such as The Wall Street Journal, which ran a story with the sub-headline, “Report Sees Nearly 50% Chance Federal Housing Administration Will Need Bailout,” are misinterpreting a specific line in the 2011 congressional report. The sentence in the report states: “With economic net worth being very close to zero under the base-case forecast, the chance that future net losses on the current, outstanding portfolio could exceed current capital resources is close to 50%.”
Sullivan says that sentence alone presumes the FHA will stop doing business, which of course, it will not. He repeatedly referred to the base-case scenario of adding $9 billion to mortgage insurance fund’s economic value.
Brian Chappelle, partner at Washington-based mortgage banking consulting firm Potomac Partners and a former FHA official, says there are two different, but equal, stories when gauging the administration’s financial health.
“You have modeling and projections on one hand and you have actual performance on the other, and it’s a tug-of-war between the two sides,” Chappelle says.
He said just looking at the projections leads some people to believe the FHA and banks need to tighten up. “But if they tightened up, that would just exacerbate the problem.”
Another issue to consider whether the FHA’s forecasting of future losses is realistic. Lawler warns that foreclosure delays have gotten so long that the severity of loss to FHA books will be a lot worse than what has been modeled. Increasing foreclosure delays could lead to deteriorating property value.
“They’re not ridiculous, but they are probably a little on the low side,” Lawler says of the FHA’s future loss predictions.
WHARTON REPORT VS. HUD
The debate over the financial health of the FHA received another quarrelsome injection when a study release in October claimed the agency would need a bailout from the Treasury of $50 billion to $100 billion in the next several years. Joseph Gyourko, a real estate finance professor at the University of Pennsylvania’s Wharton School, conducted the study, commissioned by the conservative think tank American Enterprise Institute.
Gyourko concluded that the combination of increasingly less capital per dollar of insurance guarantees and a growing number of underwater homeowners has made the FHA a very risky proposition for taxpayers. He writes that the actuarial analysis does not fully control for certain credit risks, including unemployment.
“This leaves a quick and substantial economic and housing market recovery as the primary way for FHA to avoid generating substantial losses for American taxpayers,” Gyourko said in his report. “That certainly is to be hoped for, but hope is not a sound foundation on which to run what is now a trillion-dollar entity.”
Raphael Bostic, HUD’s assistant secretary for policy development and research, fired back days later on the department’s official blog, calling Gyourko’s claims not just false, but outrageous.
Bostic rejected Gyourko’s notion that the FHA is the next housing bailout, saying the agency’s total liquid assets are at the highest point ever and $400 million higher than one year earlier. Over the last three years, HUD paid out a record $35 billion in claims while increasing its dedicated loss reserves by $20 billion. And although the actuary predicts a record payout of claims in 2012, it also estimates the new 2012 book of business will add $9 billion to the economic value of the fund, Bostic says.
“This is true, but it is only half the story, of course,” wrote Gyourko on AEI’s website in response to Bostic. “Any conclusion about how risk changed necessarily involves a comparison of how FHA’s potential liabilities grew relative to its capital.”
FHA issued about $213 billion in new guarantees on single-family mortgages in fiscal 2011 along with the additional $400 million in liquid assets. That means for every dollar the FHA added in total liquid assets, it added $532 of potential liabilities.
The 2011 actuarial analysis concluded that home prices would decline an additional 5.6% in 2011 and then bottom out in mid-2012, a pattern supported by housing analysts at JPMorgan Chase and the Federal Reserve Bank of San Francisco. But in his online exchange with Gyourko, Bostic said the “actuarial estimates imply that home prices would have to fall 4% to 5% in 2012 before FHA would require outside assistance.”
FHA Acting Commissioner Carol Galante said in a November conference call that even if home prices fell 4% to 5% beyond the base-case scenario, would still not need to resort to assistance from the Treasury.
The widely followed Standard & Poor’s/Case-Shiller 20-city composite housing price index fell 3.6% for the 12-month period ended Sept. 30, which is the fiscal year-end for the FHA. Annual home prices according to the index have worsened each month since October 2010, although the rate at which the declines are occurring have slowed each month since May when the rate was a negative 4.16%.
Applying that to FHA’s economic projections, home prices will increase 1.3% in 2012.
David Lykken, managing partner at Austin, Texas-based Mortgage Banking Solutions, says the massive supply relative to demand, which has suppressed house prices, spells big problems for the FHA. “I think the reality is that we’re going to see the FHA needing a bailout that did not need to happen.”
Chappelle, from Potomac Partners, says his primary problem with Gyourko’s findings is that he diminishes the economic value of FHA loans originated after the financial crisis.
“His fundamental conclusion is that the loans originated from 2006 to 2008 are going to be FHA’s demise, whereas, to me, the loans originated in the last three years are FHA’s salvation,” Chappelle says. “With all that being said, if house prices drop 20%, we’re all in trouble — FHA included. You can’t blame FHA for that.”
Bostic refuted Gyourko’s contention that the FHA has become a much riskier organization, saying policy changes and record premium increases have kept FHA finances positive.
Nearly half of FHA borrowers in 2007 had credit scores below 670, whereas only 3% had credit scores below that threshold in 2010 and 2011. Bostic acknowledges the FHA is still troubled by subprime loans it insured from 2006 to 2008, but says that 2009 to 2011 books (about 75% of FHA’s portfolio) are performing substantially better than those earlier books. FHA grew from insuring 6% of all new home purchases in 2007 to 30% in 2010.
Defending IFE’s use of home prices as a factor in determining borrower default potential, Bostic said Gyourko’s assertion that the actuary should have instead used the unemployment rate was spurious and that home prices are a far more valuable indicator.
Gyourko referred to HUD’s own reporting that income shock such as unemployment is the No. 1 reason servicers give as the cause of default on a FHA-guaranteed mortgage.
“This clearly indicates that it is the combination of high loan-to-value ratios and unemployment that matter, not just loan-to-value alone,” Gyourko said.
ASSESSING ITS OWN RISK ASSESSMENT
Over the past two years, the FHA has reformed its credit policy, risk management, lender enforcement and consumer protections, collectively representing the most sweeping changes in its history. For example, in addition to the premium increases, FHA increased down payment requirements to 10% for borrowers with credit scores of 500 to 570 and made anyone whose credit score is below 500 ineligible for FHA-insured loans.
“While reasonable people can disagree over predictions of what the future will hold, what isn’t in dispute is what FHA is doing right now,” Bostic says.
The GAO agreed, but reported in November that FHA’s current risk assessment strategy lacked integration and specific annual mechanisms to ensure it identified all emerging risks caused by changing conditions.
Back in 2010, the FHA established a risk office and hired a consultant to help it develop a strategy for identifying and addressing risks related to the rapid increase in single-family business volume. However, the GAO analysis found implementation of the consultant’s recommendations — one of them was creating a “template” to identify said risks — has been slow because of delays in defining the new office’s authority, changes in FHA leadership and difficulty in filling new staff positions.
Without ongoing internal and external mechanisms in place to anticipate and address new and emerging risks, FHA lacks a systematic approach to help the agency identify, analyze and formulate timely plans to respond most effectively to new and changing conditions and risks, GAO said in the report.
HUD agreed, stating it was either currently working toward achieving the various recommendations or had plans to do so soon.
“HUD has agreed that they will take action on all the recommendations, so I’ll take them at their word that they’ll be moving out on these,” the GAO’s Scire says, who added that he’d like to see action within a year. HUD has to report to Congress to explain how it plans to implement any GAO recommendations.
Scire says FHA staff shortages and high turnover make this difficult, though.
While the FHA’s single-family staff rose 8% to 1,011 employees in 2010 from 932 in 2006, the number of new loans endorsed by the agency each year spiked 317% to nearly 1.7 million from almost 400,000. Officials have taken steps to address the considerable increase in volume-driven loan review and the management of foreclosed homes. It is too soon to determine the effectiveness of these process changes, the GAO said.
When the FHA reorganized in 2010, it moved the Office of Evaluation into the Office of Risk Management and Regulatory Affairs. The Office of Evaluation’s responsibilities include overseeing the annual independent actuarial studies that determine the value of the mortgage insurance fund and conducting ongoing portfolio analysis designed to assess risks to the insurance fund.
Also, 63% of homeownership center staff (who conduct most day-to-day functions) is eligible to retire in the next three years, but the FHA has not developed a plan to manage retirements or hire staff with necessary skills. Without a workforce planning process that includes succession planning, the GAO says the FHA’s ability to systematically identify staffing needs is limited.
In response to a GAO analysis, FHA officials told the government watchdog in 2010 that they were planning to require the actuarial review contractor to use the stochastic simulation model for its 2011 review. But the officials said the model would be used to examine the implications of extreme economic scenarios on the mutual mortgage insurance fund and that decisions about using the model to estimate its capital ratio had not been made.
So, perhaps in at least one sense, Moody’s Analytics’ extreme 4% forecast isn’t that highly unlikely after all.
