When I set out to write Mortgage Market Mayhem in late 2007, I could see that the avalanche had already started, and that nothing was going to stop it. The chapter below was written in March of 2008, at about the time Bear Stearns collapsed.
As we now watch the 21st century version of the Pecora Hearings, we are treated to the spectacle of such 9-figure luminaries as Robert Rubin telling us no one could have predicted the collapse, or the discredited “maestro” Alan Greenspan looking to foist the blame off onto the housing finance Agencies, Fannie and Freddie and their government-guaranteed cousin, Ginnie Mae.
Never mind that the Agencies had seen their market shares drop precipitously, or that the snow mass began rolling with purely non-Agency paper, or that Greenspan and Rubin were instrumental in hiding the massive levered bet embodied in Credit Default Swaps.
What seems to be missing in most of commentary on the topic is the big picture – how the policies of the time set the incentives so that correction was inevitable, and the bears in the trade had the means to guarantee that any correction became a collapse.
In The Blame Game, I laid out the differences between the mortgage brokers and borrowers (the micro players) and the policy-makers and regulators (the macro players).
In this book excerpt published today on Bloomberg.com, Roger Lowenstein lays out the case for re-regulation and the unfortunate fact that most of the crucial changes have already been defeated by the money crowd.
It almost seems like we needed the hard lessons of a Hoover style administration and a real depression to get to the point that the public is no longer willing to support policies that hurt them. If the worst of the first Depression had been avoided, what are the chances we would have seen Glass-Steagall and the SEC established?
Just something to think about.
With that, here’s The Blame Game:
The Blame Game
(There’s Plenty to Go Around)
As problems with subprime mortgages began to appear, attention zeroed in on mortgage brokers, banks, lenders, appraisers and mortgage borrowers. Certainly, they all deserve to take their share of the blame, if only for being overly optimistic.
But there were bigger players at work. They created the systemic demand that may have had more to do with the size of the problem than all the smaller participants put together.
In this macro category we can put the Federal Reserve, the investment banks, CDO managers and other securitizers, portfolio managers, Rating Agencies, hedge funds, the Financial Accounting Standards Board (FASB) and the media. The government, the government-sponsored housing Agencies (Ginnie Mae, Fannie Mae and Freddie Mac) and state mortgage lending regulators also get to take partial credit for the macro conditions.
[CDO – Collateralized Debt Obligation. A securitized bond deal that uses bonds or loans as its collateral for a new multi-class bond deal. Initially executed with junk bonds by Drexel in the 1980’s and used to fund bank loans off balance sheet by NatWest Bank in 1996.]
Together, these macro players set the policies and took the actions that created an environment where the originate-to-securitize business model was terrifically profitable. Unfortunately, the macro players also provided the accelerant that turned what could have been a market correction into a market calamity.
The macro players set up the largely unregulated mortgage lending business that steered market share for subprime and non-conforming lending away from the banks, making it easy for the micro players so inclined to take advantage of lax oversight.
The Fed lowered rates to 1% and kept them there, allowing lenders to extend low rates to their borrowers, who could “lock in” those low rates for two or three years. Some lenders, especially the big securitization machines, offered loans with all the “innovations” – easy documentation standards, very low down payments and low initial rates with prepayment penalties that made those loans profitable even if they were quickly paid off.
When mortgage originator/securitizers started making and buying loans with very small down payments and no income verification, they were essentially betting that the value of the houses would rise so fast that the mortgages would still be paid even if the borrowers really couldn’t afford to pay them. A house and its expected appreciation became a substitute for the true equity of a traditional down payment.
Only steady house price inflation could make such loans safe. To be fair, until 2007, this wasn’t such a stretch, as there had never been a year since the Great Depression in which house prices in the U.S. had not gone up by at least a little bit.
With house price appreciation in the double digits, more and more lenders found they could be even more profitable if they sold their loans into securitizations. In this environment, making 20% “second mortgages,” often as part of a home loan equal to or close to 100% of the home’s value, didn’t feel so risky.
Subprime lending was booming. By 2006, there were banks, Wall Street dealers, specialized finance companies, hedge funds, and even well-known companies like General Electric, General Motors and H & R Block all competing for pieces of the subprime mortgage pie.
Companies securitizing subprime mortgage loans were making equity returns of 25% to 40%, even after paying overhead. Wall Street firms were making over 60% on their investment in subprime mortgages before paying overhead and bonuses. Even if losses went back up to the 4% to 6% levels that accompanied the prior two recessions, these bulk mortgage purchasers would still enjoy a handsome profit from their securitizations.
In a unique case of unregulated madness, Wall Street firms provided credit to CDO asset managers to buy the lower credit portions of mortgage securitizations and provided credit to hedge funds to buy the unrated portions.
While Wall Street had to follow certain regulations, the entities it funded to buy its products did not. These entities ended up having an enormous impact on the events that unfolded, since they were instrumental in the expansion of securitization beyond sensible or reasonable bounds, with disastrous effects for the market as a whole.
It was the CDO managers who really drove demand (and prices) up for subprime mortgage bonds in 2005, 2006 and the first half of 2007. They were the largest buyers of subprime “mezzanine” bonds (the bonds rated below AAA and above junk) from new securitizations.
In putting together CDO deals, CDO managers typically purchased fifty percent subprime bonds, and built the rest of their deals from a mix of commercial real estate bonds, auto loan bonds, student loan bonds, and even CDO’s from other managers.
The raw ingredients for a successful CDO recipe often came directly from the Wall Street dealers, who recommended and then sold the bonds to the CDO managers. The CDO managers then combined those bonds to create new bonds rated from AAA down through AA, A, BBB, and unrated.
What did they do with these second-stage securitizations? Why, they sold them to the Wall Street dealers who had supplied the raw materials to make them in the first place! Wall Street, in turn, sold the CDO’s to investors, and sometimes to Structured Investment Vehicles (SIV’s). The SIV’s which weren’t under the direction of banks or other lenders were frequently created and controlled by the very same Wall Street dealers.
If the CDO managers didn’t have the capacity to finance the bonds they used as raw materials while they held them, they were, in essence, captive entities of Wall Street. They got their raw materials from a Wall Street dealer who also supplied their financing, and they were obligated to have that same Wall Street dealer sell the bonds they would ultimately create.
The Wall Street dealers themselves were doing about a deal a week, and had access to the Rating Agency models, which would determine the credit rating of the tranches that came out of the deals.
The Rating Agencies had a forty-year track record of rating first-stage securitizations with excellent results. It was all based on evaluating risk and the precept that the credit performance for one borrower in the deal should not be dependent on how other borrowers perform. In a successful securitization, bond builders attempt to isolate risk, so the poor performance of some borrowers in the collateral pool will have limited correlation to the performance of the other borrowers.
Unfortunately, the Rating Agencies had no track record to speak of when it came to evaluating re-securitizations that were structured as CDO’s. When rating the tranches of these re-securitizations, the Rating Agencies felt they could model the probability of loss for a pool of CDO mortgage bonds the same way they could with any other pool of credits. They assumed that one BBB rated subprime mortgage bond would have distinctly different credit losses than other BBB rated subprime mortgage bonds.
They were wrong, of course. They had overlooked an important fact: If history showed that all BBB rated subprime bonds had behaved uniformly well in the past, it might be equally probable that they could all behave uniformly poorly in the future. Instead, the Rating Agencies chose to focus on the differences between lenders in quality control, servicing diligence, and regional marketing, and based their analysis on those factors.
With separate households paying their mortgages or companies paying their bank loans, this analysis works. But with BBB bonds from mostly similar pools of mortgage loans, the fact that each pool has thousands of borrowers actually makes the bonds more alike than different.
CDO managers acquired bonds from structured finance deals that had already stratified the probabilities of default among relatively uniform pools of loans. CDO’s built from these bonds were essentially super-levered concentrations of credit performance.
In other words, all BBB bonds behave more or less like other BBB bonds. Ironically, this is what the ratings, which establish whether a bond is rated AAA or BBB, are based on. If you buy a BBB corporate bond, it is supposed to be just as likely to default as any other BBB corporate bond.
When times were good, all the supbrime bonds performed better than expected. In fact, from 1984 through 2004, subprime mortgage bonds which were rated BBB performed about as well as municipal or corporate bonds rated AAA.
The new CDO managers were able to set themselves up in small offices and, within a short time, have billions of dollars in assets under management while they let the Wall Street dealers supply the raw materials, do the structuring and marketing, provide the financing for the assets during the warehousing period, and sell the re-securitizations.
Even a senior management fee of as little as 10 basis points equates to $1 million per year on a $1 billion CDO. Once you get half a dozen deals under management, a “two guys and a Bloomberg” operation could make really nice money.
The best payoff was for the Wall Street firms, though. They quoted fees of $5 million for those $1 billion CDO deals, and they also made fees off the first tier deals that created the BBB bonds the CDO managers were buying. All the fees owed to Wall Street were paid up front, at the closing table for the deals.
Sometimes, CDO managers also held the first loss equity tranche of their deals, so they had capital at risk in a position that would take losses before all the other CDO investors. But even this investment paid healthy returns of 15% to 20% when all the bonds were paying, which almost all mortgage bonds do for at least a few years.
For the CDO managers, the biggest perceived business risk was not the credit risk they were taking. They had good reason to minimize that risk in their minds, because they were typically buying bonds rated BBB or higher. There had never been meaningful losses in those bonds in their lifetimes.
The real risk they saw was that the profit (the “spread”) they hoped to capture would shrink before they could buy all the bonds they needed to make their new bonds.
If profit dissipated (spreads “tightened”), the deal might not work any more. The anticipated returns to the first loss bond or even to the rated bonds in the CDO structure might disappear.
For this reason, CDO managers wanted to jump into a fully invested position as quickly as possible. Whenever they could get a few extra basis points by buying one BBB bond instead of another, it didn’t take long to make the decision. Bonds that traditional investors would have bought only after lengthy due diligence were flying off the shelves.
Long-term investors that had been in the market for years suddenly found themselves outbid by CDO managers, on every deal, often within minutes of the deal being announced. The frantic rush to buy left no time for proper analysis of the bonds before a commitment had to be made.
The CDO managers made the business of analyzing and purchasing bonds more like a TV quiz show than a business transaction. Speed at putting in the buy order was the thing that separated the winners from the losers. Competition to get all the bonds quickly and grab a little extra spread eliminated any meaningful relationship between price and quality for bonds in the marketplace.
It was all made worse by the fact that some Wall Street houses had bought and securitized pools of subprime mortgages containing at least 20% to 30% of the riskiest subprime loans.
At the same time, mortgage brokers had learned how to submit a “clean” looking file to lenders, and borrowers were more than happy to do whatever was necessary to qualify for a loan. Generous appraisals became the rule of the day for appraisers who relied on referrals and knew the numbers needed to make the mortgage brokers and real estate agents happy.
At real estate closings, the mortgage brokers and real estate agents walked away with thousands of dollars. The mortgage bank walked away with several hundred dollars of profit on each loan when they sold them to securitizers, selling the loans in blocks of several hundred at a time. The securitizers also made several hundred dollars on each loan by turning the mortgages into bonds. And many of the mortgage bonds went on to be re-securitized into new bonds by the CDO managers.
Structured finance CDO issuance grew from a few billion a year in 2002 to over $200 billion in 2006. Several Wall Street shops were issuing or underwriting $40 to $80 billion a year of subprime mortgage bonds, and then selling pieces of the deals to fund managers to whom they offered liberal credit. Wall Street had created its own customers.
[Underwriting – A commitment by a securities dealer to purchase a stock or bond issue from its client. These issues are meant to be sold to the public, but the investment bank will buy the securities even if the public does not buy the offering.]
Perhaps worst of all was the fact that these customers were beyond the reach of regulators, as nearly every CDO and hedge fund legally resides in a tax haven like the Cayman Islands, even though almost all the people making the investment decisions live and work inside a fifty-mile radius of Wall Street.
A vast financial empire was being constructed on a very shaky foundation.
In 2007, the breaking of the bubble showed up initially in second mortgages. Securitizers like the Household Finance division of HSBC Bank had been ready buyers of “piggyback” loans. Early in 2007, HSBC was the first major bank to acknowledge large expected credit losses from subprime loans. It announced that it was reserving $9 billion for future losses in its portfolio of second mortgages.
Soon after, the market suffered repeated additional shocks. Subprime lender New Century admitted that it had overstated earnings for several reporting periods. H&R Block said their Option One division was taking a $100 million dollar loss on loans that defaulted in the first three months. While H&R Block’s subsidiary experienced the worst of these early payment defaults (EPD’s), other lenders were also seeing an uptick in early defaults.
At the same time, homeowners who had borrowed nearly the entire purchase price of their properties were going into foreclosure in unprecedented numbers.
Losses from early defaults on presumably fraudulent loans generally remained the burden of the mortgage banks (originators). At least that was the case until virtually all the large independent mortgage banks went bankrupt. The servicing of those mortgages then had to be transferred, and the confusion factor was enough to cause more delinquency all by itself. Borrowers sometimes didn’t even know where they were supposed to send their payments. Sometimes their payments were “misplaced.” Greater delinquency gave us more borrowers in the hole, which led to more foreclosures.
The losses that hit hard in the bond market came first in the CDO market, as participants realized that the reversal of house price appreciation could lead to much larger losses than anticipated in the subprime mortgage pools backing the bonds CDO managers had bought. The bonds the CDO managers held started getting downgraded in anticipation of future credit losses.
Most CDO’s have provisions that define a default condition if a significant fraction of the debt in the CDO gets downgraded. So even before there were actual losses on the loans that affected the subprime bonds in the deals, CDO deals were technically in default. Credit losses in the BBB subprime bonds used as collateral could result in credit losses even in the AAA tranches issued by the CDO’s. Half or more of the 2006 vintage subprime BBB bonds were in danger of credit losses.
Managers of mutual funds heard that anticipated losses from subprime bonds were expected to hurt some CDO bonds that were originally rated AAA, and some jumped to the conclusion that no bond backed by subprime mortgages was worth anything, even if it was AAA rated. They reacted by selling the stock in any company that had any investment in subprime bonds, and the small investor followed.
The knee-jerk reaction that concluded all AAA bonds with subprime exposure were actually junk propelled a massive sell-off. The lack of due diligence and inability to properly evaluate those AAA bonds was even more unfortunate given the fact that the AAA subprime mortgage bonds made up nearly 80% of the deals, on average. Most subprime deals had only 2% or 3% of the BBB bonds that were the problem.
Unfortunately, subprime mortgages aren’t the only ones in trouble.
Things aren’t looking so rosy in the alt-A mortgage sector, either, despite the media’s early assertions that this sector was insulated from the problems plaguing subprime borrowers. Unlike the subprime borrowers, alt-A borrowers tend to have “prime” credit scores but they do not qualify as prime borrowers. This is most often due to the fact that they have their own businesses or work as independent contractors, and thus have no established salaries.
The alt-A mortgage loans certainly seemed better. They seemed better because of the high credit scores of the borrowers. But some deals were backed by pools of loans in which half came with no income verification. A few alt-A deals were composed entirely of no-doc loans.
This sector of the market tended to draw the most aggressive house-flipping investors. Pools of alt-A mortgages could be comprised of 80% primary residences with the rest of the pool comprised of “investor properties,” while pools of subprime deals averaged around 5% investor loans.
Many of these investors never thought of themselves as landlords. They were not investing for income like traditional buyers of investment properties, but solely for price appreciation. They bought properties that couldn’t rent for enough to pay the mortgage and maintenance, hoping to make a killing by flipping them, sometimes buying multiple properties at the same time so multiple lenders would not “see” each others’ loans on their credit reports.
In 2007, alt-A borrowers took out nearly $400 billion in mortgage loans, not too much less than the $600 billion peak that subprime lending reached in 2006. In early 2008, some pools of alt-A loans were performing just as badly as subprime deals of the same vintage. Without the extra structural protections built into the structures that subprime deals have, AA rated alt-A bonds turned out to be just as risky as single A or BBB subprime bonds, or even worse.
Once the credit issues became apparent in the alt-A sector, the entire macro system worked in reverse, pushing asset values down faster and faster. The Fed kept raising rates until the middle of 2006, buoyed by a false confidence that a major correction in the largest asset held by Americans would somehow be “contained” in the realm of subprime borrowers.
During the bubble years, extraction of equity from American homes amounted to hundreds of billions of dollars every year. Some estimates say a full percentage point of our GDP came from the refinancing boom.
Construction, finance and real estate companies added millions of workers to handle the increased volume of new homes and the turnover of existing homes. A sudden reversal of the lending environment was sure to eliminate those jobs, but our policy makers pretended that couldn’t happen, and kept tightening the screws so they could fight inflation.
In an ironic twist, foreigners holding U.S. mortgage bonds felt the losses as well, and they soon decided that the cornerstone of the United States’ economic system was hopelessly fragile, so they avoided U.S. dollar investments.
When the dollar dropped as a result, inflation resulted from the currency decline in spite of any attempt to control it through monetary policy. In a sense, the Fed’s attempt to be the Great Inflation Fighter led directly to inflation in all of the goods we import, which includes commodities and most manufactured goods.
Accounting standards adopted after the corporate scandals in 2001 and 2002 added more fuel to the fire. Those standards forced investors and other holders of mortgage bonds to realize market value losses as immediate declines in capital or outright losses, even if the bonds were still in portfolio.
The result of this short-sighted regulation was that when one weakened levered holder collapsed, that single event automatically led every other holder to take an immediate loss to capital. Each time the prices of the bonds declined, other bond investors who borrowed money on margin to finance their mortgage bonds were that much closer to being forced to sell.
Large supplies of high-quality mortgage bonds were being sold under duress into a market that was having real trouble absorbing them. Making new loans to securitize was a money-losing exercise because the secondary bonds were being sold so cheaply.
Lenders were “losing” capital, so they couldn’t lend if they had to keep the loans in their investment portfolios. With the securitization market effectively shut down, any loan that couldn’t be guaranteed by one of the Federal mortgage Agencies would have to be kept in the lender’s portfolio.
Homeowners who might have been able to refinance their adjustable rate mortgages when rates increased were suddenly left with no refinancing options that didn’t require an additional down payment, even though fixed mortgage rates had dropped far enough that many borrowers could have easily handled the monthly payments. As a result, millions of homeowners were left with only one option: to try to sell their homes, which just lowered prices further.
The macro players added fuel to the fire by tightening credit standards far past where they were before the bubble in debt developed. They overshot, in other words.
Other macro players like the press did their part to make the situation worse, coming up with frightening headlines like “House Price Growth Drop Fastest on Record.” Reading that, it was easy for nervous homeowners or potential buyers to miss the word “Growth” and come away believing that prices were dropping, even before they had.
The Fed stayed the course with inflation-fighting high interest rates until September of 2007, when the collapse of the credit markets was all but assured. Investors avoided any exposure to any fund or bank that had any exposure at all to subprime mortgage bonds, no matter how small or how safe the investment might be.
Rating Agencies left their high ratings on subprime mortgage bonds until it was obvious that the problems were coming, and then they devised new and tougher tests, which suddenly ratcheted all the ratings lower.
Regulators and politicians began to announce investigations into mortgage lenders, accelerating lenders’ downward spiral toward bankruptcy, as shareholders dumped their stock into a fearful market.
To complete the picture, hedge funds that wanted to bet against housing and mortgage bonds applied tremendous pressure, with leverage, to drive down the prices of mortgage bonds and the stocks of mortgage lenders.
In essence, a huge interconnected system existed to create mortgages, create mortgage bonds, and then to create new bonds from bonds that had just been made, all in very big hurry, with almost every entity along the line unloading the risks as they took home the profits. None were taking home profits as fast as the Wall Street dealers, who claimed that every bond was sold not on the traditional basis of yield-to-maturity or yield-to-call but rather on the basis of yield-to-bonus.