Benign View of Loans Helped Create Bonds, Led to More Lending
In 2000, Standard & Poor's made a decision about an arcane corner of the mortgage market. It said a type of mortgage that involves a "piggyback," where borrowers simultaneously take out a second loan for the down payment, was no more likely to default than a standard mortgage.
While its pronouncement went unnoticed outside the mortgage world, piggybacks soon were part of a movement that transformed America's home-loan industry: a boom in "subprime" mortgages taken out by buyers with weak credit.
Six years later, S&P reversed its view of loans with piggybacks. It said they actually were far more likely to default. By then, however, they and other newfangled loans were key parts of a massive $1.1 trillion subprime-mortgage market.
Today that market is a mess. As defaults have increased, investors who bought bonds and other securities based on the mortgages have found their securities losing value, or in some cases difficult to value at all. Some hedge funds that feasted on the securities imploded, and investors as far away as Germany and Australia have suffered. Central banks have felt obliged to jump in to calm turmoil in the credit markets.
It was lenders that made the lenient loans, it was home buyers who sought out easy mortgages, and it was Wall Street underwriters that turned them into securities. But credit-rating firms also played a role in the subprime-mortgage boom that is now troubling financial markets. S&P, Moody's Investors Service and Fitch Ratings gave top ratings to many securities built on the questionable loans, making the securities seem as safe as a Treasury bond.
Also helping spur the boom was a less-recognized role of the rating companies: their collaboration, behind the scenes, with the underwriters that were putting those securities together. Underwriters don't just assemble a security out of home loans and ship it off to the credit raters to see what grade it gets. Instead, they work with rating companies while designing a mortgage bond or other security, making sure it gets high-enough ratings to be marketable.
The result of the rating firms' collaboration and generally benign ratings of securities based on subprime mortgages was that more got marketed. And that meant additional leeway for lenient lenders making these loans to offer more of them.
The credit-rating firms are used to being whipping boys when things go badly in the markets. They were criticized for being late to alert investors to problems at Enron Corp. and other companies where major accounting misdeeds took place. Yet they also sometimes get chastised when they downgrade a company's credit.
The firms say that since first asked to rate securities based on subprime loans more than a decade ago, they've done the best they could with the data they've had. "The housing market has proven to be weaker than a lot of expectations," says Warren Kornfeld, co-head of residential mortgage-backed securities at Moody's. This summer, the firms downgraded hundreds of mortgage bonds built on subprime mortgages. They say those bonds represent only a small part of the subprime-mortgage market.
The subprime market has been lucrative for the credit-rating firms. Compared with their traditional business of rating corporate bonds, the firms get fees about twice as high when they rate a security backed by a pool of home loans. The task is more complicated. Moreover, through their collaboration with underwriters, the rating companies can actually influence how many such securities get created.
Moody's Investors Service took in around $3 billion from 2002 through 2006 for rating securities built from loans and other debt pools. This "structured finance" -- which can involve student loans, credit-card debt and other types of loans in addition to mortgages -- provided 44% of revenue last year for parent Moody's Corp. That was up from 37% in 2002.
When Wall Street first began securitizing subprime loans, rating firms leaned heavily on lenders and underwriters themselves for historical data about how such loans perform. The underwriters, in turn, assiduously tailored securities to meet the concerns of the ratings agencies, say people familiar with the process. Underwriters, these people say, would sometimes take their business to another rating company if they couldn't get the rating they needed.
"It was always about shopping around" for higher ratings, says Mark Adelson, a former Moody's managing director, although he says Wall Street and mortgage firms called the process by other names, like "best execution" or "maximizing value."
Executives at both ratings firms and underwriters say the back-and-forth stopped short of bargaining over how to construct securities or over the criteria used to rate them. "We don't negotiate the criteria. We do have discussions," says Thomas Warrack, a managing director at S&P, which is a unit of McGraw-Hill Cos. He says the communication "contributes to the transparency" preferred by the market and regulators.
Some critics, such as Ohio Attorney General Marc Dann, contend the rating firms had so much to gain by issuing investment-grade ratings that they let their guard down. They had a "symbiotic relationship" with the banks and mortgage companies that create these products, says Mr. Dann, whose office is investigating practices in the mortgage markets and has been talking to rating firms.
In assembling a security such as a mortgage bond, an underwriter first pulls together thousands of loans that will serve as collateral. Before marketing the security, the underwriter slices it into perhaps 10 "tranches" with varying levels of risk and return.
The riskiest tranche has the highest potential return, but it ought to, because the buyer is taking a great risk: This tranche will absorb the first defaults that occur in the pool of mortgages. The next-lowest tranche is the second-hardest-hit by any defaults. Because of this structure, most of the higher tranches traditionally were considered well-enough insulated from defaults to merit investment-grade ratings -- in some cases, triple-A ratings.
The process, in a bad market, is like prisoners walking the plank on a pirate ship. The holders of the riskiest securities are at the front of the line and go overboard first. What's happening in the subprime-mortgage market is that investors further back than many imagined possible are going overboard as well.
Had the securities initially received the risky ratings that some of them now carry, many pension and mutual funds would have been barred by their own rules from buying them. Hedge funds and other sophisticated investors might have treated them more cautiously. And some mortgage lenders might have pulled back from making the loans in the first place, without such a ready secondary market for them.
Many money managers lacked the resources to analyze different pools of assets and relied on ratings companies to do so, says Edward Grebeck, chief executive of a debt-strategy firm called Tempus Advisors. "A lot of institutional investors bought these securities substantially based on their ratings, in part because this market has become so complex," he says.
Back in 2000, piggyback mortgages were just one among a handful of new loan varieties that credit analysts were having to evaluate. Until that point, few borrowers used piggyback loans to stretch beyond their means. But lenders began proposing these structures as a way to make homes affordable as their prices rose.
Because buyers putting less than 20% down may have less incentive or ability to avoid default, they normally had to buy private mortgage insurance to protect the lender if they fail to make the payments. But as interest rates slid and home prices rose, plenty of lenders were willing to provide a second, piggyback mortgage for all or part of the 20%, without insisting on mortgage insurance.
The big mortgage buyers Fannie Mae and Freddie Mac wouldn't purchase these piggyback deals, which didn't meet their standards. But Wall Street firms would, because they found they could turn them into high-yielding securities. And there were plenty of buyers for such securities: With interest rates low, many investors were in search of higher-yielding instruments.
Data provided by lenders showed that loans with piggybacks performed like standard mortgages. The finding was unexpected, wrote S&P credit analyst Michael Stock in a 2000 research note. He nonetheless concluded the loans weren't necessarily very risky.
S&P didn't let loans with piggybacks completely off the hook. S&P said in 2001 that it wouldn't penalize a subprime mortgage pool so long as the value of loans with piggybacks didn't exceed 20% of the overall value. Any more than that, and it would impose a rating penalty, S&P said. The firm notes that its assumptions "remained appropriate for several years."
Despite this limit, S&P's stance was good news for underwriters and lenders. For underwriters, the S&P decision made it easier to create investment-grade securities based on pools of subprime loans. And underwriters' appetite for the loans, in turn, made it easier for lenders to originate them.
Trends then converged to create explosive mortgage-market growth. Falling interest rates -- as the Federal Reserve sought to prop up the economy after the tech-bubble burst -- made home financing less expensive. New technologies let bankers construct bonds from the payments of thousands of different mortgages. The fastest-growing segment was subprime loans. Lenders brought out loans in which borrowers didn't have to document their income, or could at first pay only interest and no principal -- or could use a piggyback to, in effect, borrow the whole cost of the home.
At first, underwriters creating mortgage securities made sure the loan pools they based them on didn't have more than 20% with piggybacks. But by 2006, some were willing to accept a ratings penalty. They created securities like those structured from a pool of 14,500 loans from Washington Mutual Inc.'s mortgage arm. About 52% of the pool's value consisted of loans with piggybacks, a prospectus showed.
By 2006, S&P was making its own study of such loans' performance. It singled out 639,981 loans made in 2002 to see if its benign assumptions had held up. They hadn't. Loans with piggybacks were 43% more likely to default than other loans, S&P found.
In April 2006, S&P said it would raise by July the amount of collateral underwriters must include in many new mortgage portfolios. For instance, S&P could require that mortgage pools have extra loans in them, since it now expected a larger number to go bad.
Still, S&P didn't lower its ratings on existing securities, saying it had to further monitor the performance of loans backing them. It thus helped the market for these loans hold up through the end of 2006.
Some investors, however, grew concerned, as newer mortgage securities appeared that were based not just on piggyback loans but on loans with other risky attributes as well. One money manager, James Kragenbring, says he had five to 10 conversations with S&P and Moody's in late 2005 and 2006, discussing whether they should be tougher because of looser lending standards. "I'd think there would be more protection to guard against defaults," Mr. Kragenbring, from Advantus Capital Management, says he said to the rating companies.
He says he was told that for much of 2005 and 2006, subprime loans were performing about the same as in previous years. Other analysts recall being told that ratings could also be revised if the market deteriorated. Said an S&P spokesman: "The market can go with its gut; we have to go with the facts."
In the second half of 2006, Mr. Kornfeld at Moody's noticed a troubling trend. In an unusually large number of subprime loans, borrowers weren't making even their first payments. The market's great strength "could not continue," Mr. Kornfeld recalls thinking at the time. He called staff meetings to discuss his concern, and in November Moody's said publicly it saw signs of deterioration.
In March 2007, S&P said it expected home prices to be stagnant this year but grow 3% to 4% in 2008. By early July, S&P had lowered this forecast. It said its chief economist projected that home prices would fall 8% from the 2006 peak to a trough expected in the first quarter of 2008.
Defaults and delinquencies rose. Hard-pressed borrowers found it harder to get a new loan to bail them out or to sell their homes and pay off the loan that way. By July, almost a third of the loans in Washington Mutual's subprime pool were delinquent or in foreclosure. This performance, much worse than what credit-rating firms had expected, forced Moody's and S&P to slash their ratings on several securities backed by those loans. On some, S&P cut an initial A-minus investment-grade rating by five notches, to a below-investment-grade BB.
The downgrading, begun late last year, became an avalanche this summer. On July 10, Moody's cut ratings on more than 400 securities that were based on subprime loans. S&P put 612 on review, and downgraded most two days later. The moves jolted financial markets and prompted some investors to criticize the ratings firms for misjudging the market.
The firms said that the soaring market of 2005-06 had reduced the relevance of their statistical models and historical data.
Money mangers unloaded on a July 12 conference call with Moody's analysts. "You had reams upon reams of data," said Steve Eisman, a managing director of hedge fund Frontpoint Partners, which had made bets against the subprime market. "Despite all that data, your original predictions of the performance of 2006 loan pools have proven to be completely and utterly wrong." He asked why the rating firms waited to take major steps.
The chief credit officer at Moody's, Nicholas Weill, replied that some of the original subprime data provided to rating firms weren't "as reliable as expected." He also said Moody's put out "early warnings" of downgrades as far back as November 2006. Instead of cutting ratings right away, he added, Moody's needed time to see whether the loans would start to recover. "What we do is assess information available at the time," Mr. Weill said.
S&P, Moody's and Fitch Ratings have reacted by repeatedly toughening their ratings methodology for new subprime bonds, requiring significantly bigger cushions. They now assume more and quicker defaults among pools of loans, especially those with piggybacks.
The changes have had an effect. About 27% of loans made in the first quarter of this year had piggybacks attached, down from 35% a year earlier, according to S&P research. Overall, issuance of subprime-mortgage bonds is down 32.5% this year through June, according to Inside Mortgage Finance. That is resulting in lower Wall Street profits and tighter lending standards for consumers.
Committees in the U.S. House and Senate are broadly examining the mortgage market, as are various state and federal agencies. It's not clear whether ratings firms will become a focus of the inquiries.