Building a Bubble

Mortgage Banking

November 2008

Business strategies during the rapid expansion of mortgage credit from 2002 to 2007 worked to boost the emerging housing credit bubble fostered by low interest rates and investor demand. This is the second of a two-part series. The first part, The Origins of the House Finance Bubble, was published in the September issue.

By Robert Stowe England

The boom years' rapid and unchecked expansion of housing credit in an environment of very low mortgage rates was, as we now know beyond a reasonable doubt, a high-risk enterprise. Profits were initially high and rising as the expansion got under way, but with the benefit of hindsight, the level of risk incurred appears to have been poorly understood by many.

Never had so much credit been available to allow mortgage lenders to pursue an ever-widening array of business strategies. Rarely had leverage been more alluring, as higher leverage produced higher profits.

The seemingly boundless supply of credit from the secondary markets allowed mortgage players leeway to experiment on a large scale. Mortgage companies rushed to copy what seemed like successful efforts by other players in the market. A number of companies found themselves in uncharted waters.

Some mortgage lenders raced to gain market share in order to improve their competitive positions, with the hope and expectation that ever-greater economies of scale would boost profitability.

Subprime and alternative-A lending became the new frontier where new entrants did well and old-time players began to slug it out for a piece of this ever-growing pie. The initially low levels of delinquencies and defaults with risk-based pricing strategies encouraged more growth in these riskier markets.

The ready availability of credit from the secondary market was a green light to business-expansion strategies. It created a climate that would allow competition overall to eventually drive lending standards downward. Thus, while the business strategy of each player seemed to make sense in isolation, in the aggregate it paved the road to mortgage perdition.

The refi boom of 2002 and 2003

The business conditions created in the aftermath of the huge mortgage refinancing boom of 2002 and 2003 are identified as a primary factor behind the credit bubble by Michael Youngblood, the former managing director of asset-backed securities (ABS) research at FBR Investment Management Inc., Arlington, Virginia. Youngblood left FBR in June to form, with Steve Gaenzler, a new hedge fund firm, Five Bridges Capital, Washington, D.C., which will manage the Mortgage Opportunity Fund, which will focus on investing in undervalued mortgage assets. "When mortgage rates began to rise in early 2004, it brought an abrupt end to a two-year-long refi boom that was historically unprecedented," says Youngblood.

During the refi boom, premium non-agency pass-through securities were prepaying in excess of 90 percent of the conditional prepayment rate (CPR), Youngblood says, while agency pools were prepaying in excess of 80 percent of CPR. The weighted average life of mortgages in these pools was less than a year. Youngblood notes that the Mortgage Bankers Association (MBA) mortgage application survey refinance index peaked at 8,599 in June 2003. By May 2004, it had fallen to 1,583. "This massive up-swelling in refinancing obviously was highly profitable to the industry," Youngblood says. "And in order to serve the demand for refi credit, the industry expanded its infrastructure and staffing levels," he adds.

By the second quarter of 2004, neither refi loans nor purchase-money loans produced sufficient volume to justify the expanded capacity of the industry. "Rather than downsizing, which had been the industry's tra ditional response to refi busts, the industry began to reallocate resources to emerging sources of demand for non-traditional credit, particularly alt-A and subprime, and particularly to adjustable-rate rather than fixed-rate [product]," Youngblood says.

The industry also began to consolidate both by acquisition and by efforts to gain market share, according to Youngblood. Consolidation at times led to a bloat ing of staffing levels.

"No manager likes to downsize, and no industry likes to downsize. The desire to preserve the growth of 2002 and 2003 was perfectly understand able," Youngblood says. "But it led the industry into a direction that was unprecedented, and it also led firms without established expertise in these new markets and non-traditional products to begin to offer them and originate them in significant volume. I think that is one important cause of the mortgage finance bubble," he says.

Youngblood notes that the mortgage lending industry has traditionally been commission-based when it comes to originations. "Labor flexibility is a keynote of mortgage finance, and certainly managers had the ability to downsize their commission sales force but elected not to do so," he argues. "I think the burden of that infrastructure and staffing led them to be less attentive to the potential pitfalls of new borrowers in alt-A and subprime markets," as well as to newer instruments, such as the hybrid adjustable-rate mortgages (ARMs); payment-option ARMs; and new forms of underwriting and various forms of light-, alternative- or no-documentation.

Such loan types had been offered in the market before, but "had been used judiciously on a very small portion [of originations]," Youngblood says.

Subprime and alt-A specialty lenders

The emergence of alt-A and subprime specialty lenders is another major cause of the housing credit bubble, according to Youngblood. In the aftermath of the failure of Long-Term Capital Management in 1998, there was a shake-out of specialty mortgage lenders, he notes. Many did not survive the rationing of credit that followed-including Conti Financial, Southern Pacific Funding Co. and First Plus, among others, Youngblood says.

At the same time, several of them were recapitalized, including Delta Financial Corporation, San Francisco; New Century Financial Corporation, Irvine, California; Accredited Mortgage Services, Little Ferry, New Jersey; Fremont Mortgage, a division of Fremont National Bank and Trust Company, Fremont, California; First Franklin Financial Services, Tallahassee, Florida; and Option One Mortgage Corporation, Irvine, California. "These lenders redeployed their new business models successfully for several years," Youngblood says, "penetrating new markets using the underwriting tools of more established lenders."

Once the Fed started tightening in June 2004, these recapitalized subprime lenders saw a tightening of margins. "They were largely portfolio lenders and not pure mortgage banking companies as they had been before," Youngblood explains. "It was thought [that the] com- bination of stable earnings from the mortgage portfolio would offset the cyclical boom-and-bust earnings of the ortgage banking side of the company and the tax-advantaged dividends [of the real estate investment trusts REITs), such as New Century; Saxon Capital Inc., Glen Allen, Virginia; and Accredited, for example] would further contribute," Youngblood says.

Established lenders after 2004 began to copy the specialty lenders' product lines and underwriting practices, and emulate their efforts to serve the underserved elements of the housing market, according to Youngblood. The specialty lenders did well during the first year of Fed tightening, but as the established lenders began to compete in the same markets with them, the subprime lenders initially sought to maintain their origination volume by moving up the credit curve.

"In 2005 they originated their lowest [average mortgage] rate ever, their highest [average] FICO® ever and one of their lower loans-to-value [LTVs] ever," Youngblood says. "Then profit margins began to collapse," he says.

"By the third quarter of 2005, the subprime and alt-A industries were running red ink," says Youngblood. With their business model under pressure, "their response was to move down the [credit] curve to originate mortgage loans with higher yields to rebuild their margins," Youngblood says. "Again, this is a perfectly rational strategy." However, the strategy was compromised by the fact that many companies simultaneously liberalized underwriting practices, he notes.

The early signs of this weakness could be seen in the rising number of repurchase requests, "primarily by dealer counterparties, but also by non-dealer counterparties," explains Youngblood. The lenders were, indeed, having to buy back more of the loans because the loans "did not match" the warranties and representations made by the lenders. "The rising tide of repurchase requests hindered profitability," Youngblood says, "and raised accounting issues" about whether firms were adequately reserved for repurchase requests.

By the third quarter of 2006, the deterioration of credit quality began to show up in the statistics for delinquencies and defaults. Mortgage bond pioneer Lewis Ranieri "gave his famous speech on the subject in December 2006," says Youngblood, who also issued a warning around the same time. Ranieri rang the alarm by saying that the "mandated level of disclosure for capital markets has not kept up with the mortgage-product innovation, making the risk level in residential mortgage-backed securities [RMBS] not readily available or easily transparent."

Further, Ranieri stated, "This stuff doesn't just get sold to money managers. It gets sold to the public and to foreign investors who don't have a clue what to look for." The "word was out" on subprime and alt-?, Youngblood says. "Certainly after Ranieri's speech, every analyst on [Wall] Street knew a train wreck was coming; there's no doubt about that."

Credit-rating agencies

Youngblood identifies the credit-rating agencies as another major contributor to the housing credit bubble. As the volume of alt-A and subprime mortgage loans began to surge and these loans entered into securitized pools in ever greater numbers, the three major credit-rating agencies - Standard & Poor's (S&P), Moody's Investors Service and Fitch Ratings, all based in New York - continued to use their established credit criteria and credit-enhancement criteria, he says. These were not revisited until well into 2006, he says, noting that in May 2006 S&P revisited its credit criteria, including the criteria for second-lien loans. In September 2006 Moody's released its new metrics, and in October 2006 Fitch followed suit, he notes.

"The sequence of events tells me that the rating agencies were aware of the deterioration in credit quality belatedly and, as hindsight tells us, the changes they made were utterly inadequate to protect holders of investment-grade classes of these securities from loss," says Youngblood.

The mortgage banking industry "was evolving rapidly from 2001 - evolving in its capital structure, evolving in mortgage market participants, evolving in mortgage market products, evolving in underwriting practices, evolving in borrowers - and [virtually] no one really kept pace with this rapid evolution - in thinking through the implications of, for example, layered risk," says Youngblood.

Mortgage insurance companies

Not everyone missed the boat, Youngblood notes. "There were absolutely some far-sighted participants," such as some of the mortgage insurance (MI) companies. "Clearly, a number of MI companies sharply reduced their insurance books, really starting in late 2005 and certainly by mid-2006," he says. Mortgage insurance companies had "the tapes" on early delinquency and default data that were not available to analysts until later, he maintains. As one MI source explains, "Lenders used to literally send a magnetic computer tape to the MIs to report payment defaults." Today, 90 percent of reporting is handled by electronic data interchange (EDI), "but the old-timers still refer to receiving 'the tape' each month," the source says.

Analysts, including Youngblood, had seen favorable results from many of the lending segments where the growth in credit extension was the greatest. The underwriting of interest-only loans by public lenders in 2006 was openly praised by Youngblood, who had studied them closely, he recalls. "The underwriting criteria were never altered, but the underwriting practices were altered by a huge magnitude," he says.

Analysts could not determine the significant deterioration in underwriting practice until they had the data on loan performance, he explains. "Those not getting tapes to structure a deal or tapes to structure an MI were behind the information curve." Early warnings from mortgage insurance companies were "one of the key signs that those inside the curve were raising red flags," Youngblood says.

Youngblood says he had called on publicly traded lenders in June 2006 and asked them to provide tapes of loan performance, with no success. "It wasn't until I got the September 2006 tape in early November from [San Francisco-based] LoanPerformance that I was able to finally see what was going on," he recalls. The tape showed non-agency RMBS delinquencies for subprime jumping dramatically higher, while alt-A was also rising.

Youngblood says within two days of seeing those tapes he sounded a warning as to how bad loan performance had become in a presentation at the FBR Investor Conference in New York on Nov. 28, 2006. "I do take pride in having moved from being a bull to being a bear. I am not ashamed for my own part for waiting," he says. "I am not a quantitative analyst and not one to voice unsubstantiated claims."

Regulatory perfect storm

There is plenty of blame to go around for not properly monitoring the mortgage market and thereby making it possible for the bubble to form and get so large. Certainly market regulators such as the credit-rating agencies and some mortgage insurance companies could have done more to disclose the problem. Yet government regulators also played a role in the buildup of the credit bubble.

Indeed, the policy and regulatory forces at work from 2001 onward represent a "perfect storm" for steps in Washington that created the climate for the housing credit bubble, according to Bob Eisenbeis, chief monetary economist with Cumberland Advisers, Vineland, New Jersey, and former executive vice president of the Federal Reserve Bank of Atlanta. Eisenbeis, too, identifies "extremely low interest rates for a long period of time" as the key underlying factor in the series of policy steps that occurred. (See "The Origins of a Housing Credit Bubble, Mortgage Banking, September 2008, for a discussion of the role of low interest rates in the creation of the housing credit bubble.)

This, he says, was driven in part by the desire to avoid so-called "unwanted deflation" - a phrase former Federal Reserve Chairman Alan Greenspan and others used at the time, and a matter of great concern to some on the Federal Reserve Board and among the Reserve Bank presidents. "They were concerned about a Japanese-type downward-spiral situation," he adds.

There were two forces at work driving the economy-spending in general and spending on housing. The economic slowdown was driven by a "corporate investment-led decline, triggered in part by the dot-com collapse," Eisenbeis says.

"Because of housing's special position in the economy, I believe it would have been extremely difficult for policy-makers [at the Fed] to stand up and say, 'We're going to tighten, because too many people are getting houses,'" Eisenbeis says. "That would have been politically a very difficult situation to be in . . . ," he adds. "It was being widely touted that homeownership rates were really up. And people who did not have access to housing in the past were suddenly getting a chance to buy their own homes, and so on. If the Fed had stepped in at that point in time and tried to reverse policy, they would have been as welcome as a skunk at a garden party - on Capitol Hill, that's for sure," says Eisenbeis.

At the same time the Fed was sticking by a low interest-rate policy, regulatory policy on capital requirements gave favorable treatment to moving assets off the balance sheet, Eisenbeis says. "As the Europeans were starting to up the capital requirements, U.S. banks were already more highly capitalized than European banks; and so there was a reluctance on the part of the regulators to pile on [and] increase capital requirements further on U.S. banks," he says.

The unwillingness of regulators to set capital limits for off-balance-sheet activities set the stage for a dramatic expansion of the originate-to-distribute model of mortgage finance, according to Eisenbeis.

Once the private-label securitization model gained traction, the volume of subprime and alt-A mortgages being securitized began to grow rapidly. In place of Freddie's and Fannie 's guarantee, these new private MBS market players substituted guarantees from monoline insurers, Eisenbeis says.

The next element of this perfect storm was an increasing role for the credit-rating agencies, Eisenbeis notes. The institutions that created off-balance-sheet vehicles or participated in the creation of new financial instruments to sell to investors turned to the credit-rating agencies rather than do their own homework on the credit quality of the assets in the pools and the structures being developed, he says. There was, in fact, an "outsourcing of due diligence on the part of institutions, all driven by fee-generation," he explains. Here was a new profit source that institutions were eager to exploit - so eager that they lost sight of the risks, Eisenbeis contends. "When you look at the fee income and trading incomes for large institutions, those were the growth areas," he says.

Unfortunately, the appetite for more fee income and the associated profits began to take off at financial institutions just at the wrong moment. "It just happened to come along about the time - 2005 - when housing started to slack off and construction started to slack off, although prices had not yet started to slacken off," says Eisenbeis.

"So, when you get to 2006, when the prices start to hit [their peak, that's] when this derivative activity exploded on these questionable [instruments], and a lot of investors, including the investment banks, got caught holding the hot potato," he says. "Assets that banks thought were off their balance sheet turned out to be on their balance sheet, because of reputational risk and reputational concerns," he explains. "That resulted in the holders of these questionable assets - I won't say getting bailed out, exactly, but at least their losses were cushioned by commitments, [and] not necessarily legally binding commitments, but reputationally binding commitments - to take a lot of assets back on the balance sheet," he says.

Non-bank financials headed the list of the big institutions that sponsored off-balance-sheet vehicles, according to Eisenbeis. He identifies the non-bank firms of New York-based Merrill Lynch & Co. Inc. and Bear Stearns as among the biggest sponsors in the United States, along with New York-based Citigroup Inc., a bank holding company. "And foreign banks, both European and U.K., were very significant players in this market as well," he says.

So, just as the housing market was heading down, the volume of mortgage derivatives activity exploded, which was just about the worst thing that could have happened, according to Eisenbeis. It kept the flow of mortgage credit going artificially, until even that could not be sustained and it all collapsed.

Should the Fed have stepped in sooner, as some suggest, and set stricter rules to govern mortgages - such as a rule that lenders had to qualify mortgage borrowers on the rate that a loan could adjust to and not on the teaser rate? Eisenbeis says emphatically no. "I think it was a huge mistake to have the Federal Reserve involved in that kind of micro-regulation in that kind of loan market," he says. Why? "They have no particular expertise in that area. It's the worst kind of regulatory responsibility for a central bank to have, because all you have is a downside associated with it to begin with. And secondly, it's not monetary policy," he says. "Carry that to its logical conclusion, and it says government knows how to allocate credit better than individual competitive markets. The Russians tried that, and it didn't work."

The decision to give the Federal Reserve authority in the area of mortgage lending has clearly been a mistake, argues Eisenbeis, and one that has been ongoing for many years. "It has its roots in other regulatory failures, [such as] the failure of the Home Loan Bank system to do what it was designed to do," he says. In the wake of the failure, "The Fed was there to take up some of the slack," Eisenbeis says.

The decision to give the Fed what were essentially conflicting assignments "goes all the way back to the Community Reinvestment Act [CRA] . . . ," says Eisenbeis. CRA regulation is "credit allocation, really," and such activities by governments tend to have undesirable outcomes, he adds.

Risk-management monetary policy

The other failure in regulation is the decision by the Fed to engage in "so-called risk-management policies when it comes to monetary policy, Eisenbeis says. "That's essentially what we were describing coming out of the 2000-2001 recession," he says. Officials at the Fed saw the potential for a "low-probability event, but one that has potentially high costs that you want to avoid" - namely a downward deflationary spiral.

A risk-management policy in this situation suggests that "you essentially take steps to avoid that problem from occurring, Eisenbeis says. "What that ignores is that then by having a policy - in this case a very-low-interest-rate policy for a long period of time - [the risk-management policy] essentially has second-round consequences because it can then be destabilizing in the next round," he says.

He points out, for example, that "it's much more difficult to reverse interest-rate cuts than it is to reverse interest-rate increases." This was the problem with lowering interest rates too much in 2001, then being unable to raise them again in a timely manner.

One of the remedies that people have been recommending in the aftermath of the bursting of the housing credit bubble is that the Fed should act pre-emptively to burst bubbles in the future. What does Eisenbeis think of this approach?

"I think that's a bad idea, because the Fed says the Federal Reserve has a better idea of what normal prices are than the market does. And if that's the case, they might as well set all the prices and interest rates. The logical consequence of that is not something we want to experience," he says.

"I think the better policy is to essentially concentrate on making policy consistent with longer-run growth and equilibrium situations in the economy as a whole, and really try to avoid getting sucked into dealing with every little problem," Eisenbeis says.

Conflicting regulatory goals

The involvement of the Federal Reserve in promoting homeownership is not an appropriate function for a central bank either, according to Eisenbeis. This assignment draws the central bank into the kind of political quagmire that makes it difficult for it to perform its necessary role in monetary policy, he argues.

"They were drawn into the S&L [savings-and-loan] crisis. And then they got responsibility for enforcing the Community Reinvestment Act. And then they got the Home Mortgage Disclosure Act [HMDA], because that was put on them because of [the] Truth in Lending [Act]. It just goes on and on," he says.

Excessive leverage

In reviewing the root causes of the housing credit bubble, one should also point out that a lot of companies that got into trouble had flawed business models, Eisenbeis says. "There wasn't accountability in the right places, and everybody thought everything was all right. But it got exposed. Sometimes these things happen and that's the way it is," he says.

In addition to the obvious failures in risk management at places like Bear Stearns, there is the fundamental mistake of excessive leverage. Bear Stearns, for example, was essentially "a big thrift," says Eisenbeis. "And what did the thrifts do? They borrowed short and lent long, right? It just happened to be [that] short was overnight and long was probably three weeks, but it's still the same problem," he says.

Throw in another fundamental flaw in Bear Stearns' business model-excessive leverage-and you have an accident waiting to happen, he says. "That's one of the many reasons that many of us favor just the leverage constraint more than anything, particularly tying it into prompt corrective action and early intervention - as opposed to this fussing around with Basel I and Basel II, [which] is the ultimate fiddling while Rome burns," he says. "These guys were worrying about allocating capital at healthy institutions, and this is not what they should be worrying about," he says. "You don't find regular [non-financial] companies [that] leverage anywhere approaching the kind of leverage you see in financial institutions," Eisenbeis says.

Washington is still a long way from understanding the dangers of high leverage, according to Eisenbeis. "Look at what the response of Congress has been to Fannie and Freddie: Let them increase their leverage," he says. The Office of Federal Housing Enterprise Oversight (OFHEO) temporarily raised the capital requirements in the wake of accounting scandals at the government-sponsored enterprises (GSEs), but in the wake of the meltdown in mortgage financing for homeowners, the agency "got browbeaten" into lowering the capital levels, he says.

Eisenbeis says he, Larry Wall and Scott Frame-financial economists in the research department at the Atlanta Fed-were among the ones who had argued that OFHEO should have receivership authority and a closing rule just like the banking authorities.

"When you have only a few institutions you're regulating, it's awfully hard to avoid regulatory capture"-meaning that the regulator falls under the sway of the institutions it regulates rather than serving as an independent regulatory authority. "If you wanted to solve problems at Fannie and Freddie, the first thing you would do is outlaw lobbying on their part," he says. And that is exactly what happened when the government took the GSEs recently into conservatorship.

Eisenbeis also recommends some change in the tax laws. He cites what he sees as shortsighted rulings by the Securities and Exchange Commission (SEC) and others that prohibited banks from reserving against a protfolio of loans. The rulings required banks to "closely identify prospective losses on individual assets before you could reserve," he says. The purpose of this ruling was not to have good banking regulation, but to limit the ability of banks to shelter income from taxation and thus limit tax-revenue losses. "Well, it turns out that's a bad policy," he says, because it delays banks' efforts to deal with emerging problem loans.

Lessons learned?

So, where does all this leave the debate? What caused the biggest financial bubble since World War II? One could make the case that the weight of arguments falls into a pattern, even as individual economists or market analysts may strongly disagree in terms of identifying which causes were paramount. First, the Fed lowered interest rates too much and kept them low for too long, lighting the match for the hot-air balloon of the credit bubble. Then, a regulatory regime gave the Fed conflicting goals while also leaving non-banks free to engage in a profit-driven campaign of mortgage credit expansion built on risk models and assumptions that were untested and faulty.

Global demand for investments was so high that investors became lazy and even incautious in reviewing investments. Credit-rating agencies used models that were not sufficiently robust in terms of the history of loan performance for alt-A and subprime loans indifficult times, and did not even take into consideration a time of falling house prices. Super-low interest rates at a time of sharply rising home prices set homebuyers into a buying frenzy and created a whole new class of speculators.

Mortgage lenders, a final burst of creativity, loosened underwriting standards in a seemingly mad effort to keep markets share as profit margins vanished.

Finally, policy-makers in Washington applauded the growing homeownership rates, applauded the conflicting assignments they were giving to the Fed and ignored the potential that homeownership at any price could lead to widespread financial fallout for first-time homebuyers.

There's a common-sense rule that applies here: Too much of a good thing can be had for you. In this case, it was too much of a push to achieve homeownership regardless of the level of risk-taking. And it has turned out to be a very painful experience for many homeowners, financial institutions, investors and global financial markets.

END