|US Senator Carl Levin, Michigan|
Washington Mutual Case History
Washington Mutual Bank, sometimes called WaMu, rose out the ashes of the great Seattle fire to make its first home loan in 1890. For many years, it was a mid- sized thrift, specializing in home mortgages. In the 1980s and 1990s, WaMu entered a period of rapid growth and acquisition, expanding until it became the nation's largest thrift, with $188 billion in deposits and 43,000 employees. In 2003, its longtime CEO, Kerry Killinger, said he wanted WaMu to become the WalMart of banking, catering to middle and lower income Americans and helping the less well-off buy homes.
WaMu held itself out as a well-run, prudent bank that was a pillar of its community. But in 2005, WaMu formalized a strategy that it had already begun to implement a movement from low risk to high risk home loans. That move to high risk lending was motivated by three little words: gain on sale. Gain on sale is a measure of the profit made when a loan is sold on the secondary market. This chart, which is taken from Exhibit 3, shows a slide from an April 18, 2006 powerpoint presentation entitled, Shift to Higher Margin Products, given to the WaMu Board of Directors by the President of WaMus Home Loans Division. In the upper left there is a box that lists the gain on sale for each type of loan WaMu offers. As you can see, the least profitable loans are government- backed and fixed loans; the most profitable are Option ARM, Home Equity, and Subprime loans. Subprime, at 150 basis points, is eight times more profitable than a fixed loan at 19 basis points.
Those numbers are not estimates or projections, by the way. They are the product of actual loan data collected by the bank.
Long Beach Mortgage Company (LMBC).
WaMu traditionally had sold mortgages to well qualified or prime borrowers. But in 1999, WaMu bought Long Beach Mortgage Company (LMBC), which was exclusively a subprime lender, lending to people whose credit histories didnt support their getting a traditional mortgage. Long Beach operated by having third party mortgage brokers bring proposed subprime loans to its doors, issuing financing to the borrower, and paying the brokers a fee. Even then, Long Beach made loans for the express purpose of packaging them, selling them to Wall Street, and profiting from the gain on sale. In 2003, Long Beach made and securitized about $4.5 billion in home loans. By 2006, its loan operations had increased sixfold, and Long Beach's conveyor belt sent almost $30 billion in subprime home loans into the financial system.
Subprime lending can be a responsible business. Most subprime borrowers pay their loans on time and in full. Long Beach, however, was not a responsible lender. Its loans and mortgage backed securities were among the worst performing in the subprime industry. An internal email at WaMus primary federal regulator, the Office of Thrift Supervision or OTS, stated that Long Beach mortgage backed securities prior to 2003 have horrible performance. LBMC finished in the top 12 worst annualized NCLs [net credit losses] in1997 and 1999 thru 2003. LBMC nailed down the worst spot at top loser& in 2000 and placed 3rd in 2001.
In 2003, things got so bad that WaMu's legal department put a stop to all Long Beach securitizations until the company cleaned up its act. An FDIC report noted at the time that of 4,000 Long Beach loans reviewed, less than one quarter, about 950, could be sold to investors, another 800 were unsalable, and the rest over half of the loans had deficiencies that had to be fixed before a sale could take place. Several months later, WaMu allowed Long Beach to start securitizing its loans again as well as selling them in bulk through what were called whole loan sales.
In 2005, trouble erupted again. An internal WaMu audit of Long Beach found that, relaxed credit guidelines, breakdowns in manual underwriting processes, and inexperienced subprime personnel& coupled with a push to increase loan volume and the lack of an automated fraud monitoring tool led to deteriorating loans. Many of the loans defaulted within three months of being sold to investors. Investors demanded that Long Beach repurchase them. Long Beach had to repurchase over $875 million in loans in 2005 and 2006, lost over $107 million from the defaults, and had to cover a $75 million shortfall in its repurchase reserves.
In response, WaMu fired Long Beach's senior management and moved the company under the direct supervision of the President of its Home Loans Division, David Schneider. Washington Mutual promised its regulator that Long Beach would improve. But it didnt. In April 2006, WaMus President, Steve Rotella, emailed the CEO, Kerry Killinger, that Long Beach delinquencies are up 140% and foreclosures close to 70%. & It is ugly. Five months later, in September, he emailed that Long Beach is terrible & Repurchases, [early payment defaults], manual underwriting, very weak servicing/collections practices and a weak staff. Two months after that, in November 2006, the head of WaMu Capital Markets in New York, David Beck, wrote to Mr. Schneider that, LBMC [Long Beach] paper is among the worst performing in the [market].
At the end of 2006, Long Beach saw another surge in early payment defaults. Mr. Schneider sent an email to his subordinates that, [w]e are all rapidly losing credibility as a management team. 2007 was no better. Audit after audit detailed problems. WaMus chief risk officer, Ron Cathcart, forwarded an email from a colleague about Long Beach noting: Appraisal deficiencies &. Material misrepresentations ... Legal documents were missing or contained errors or discrepancies & loan decision errors &. [D]eterioration was accelerating in recent vintages with each vintage since 2002 having performed worse than the prior vintage.
In June 2007, WaMu shut down Long Beach as a separate entity, and took over its subprime lending operations. It issued several subprime securitizations. The subprime market froze in the fall of 2007, and WaMu ended all of its subprime lending. By then, as shown in this chart, from 2000 to 2007, Long Beach and WaMu had together securitized at least $77 billion in subprime loans. Today, although AAA rated securities are supposed to be very safe with low default rates of 1-2%, Long Beach mortgage backed securities report loan delinquency rates of 20, 30, 40, even 50%, meaning more than half of their underlying loans have gone bad.
Washington Mutual Retail Lending.Washington Mutual's problems were not confined to its subprime operations. In August of 2007, more than a year before the collapse of the bank, WaMus President, Steve Rotella, emailed CEO Kerry Killinger saying that, aside from Long Beach, WaMu's prime business was the worst managed business I had seen in my career.
When Washington Mutual talked about its prime mortgage business, it used the term loosely. While the borrowers who received loans from WaMus loan officers tended to have better credit scores than Long Beachs subprime borrowers, that was not always the case. WaMu loan officers routinely made very risky loans to people with below average credit scores.
And just like at Long Beach, in WaMus loan business, volume was king. Loan officers got paid per loan, and got paid more per loan if certain volume targets were met. Loan processors were given volume incentives as well as were entire loan processing centers. Even risk managers were evaluated, in part, on the extent to which they supported revenue growth targets. Loan officers also got paid more for closing high risk loans than low risk loans.
Not surprisingly, people cut corners to keep the conveyor belt moving and increase their pay. For example, an April 2008 memo from a WaMu internal corporate fraud investigator states: One Sales Associate admitted that during that crunch time some of the Associates would manufacture asset statements from previous loan doc[ument]s, because the pressure was tremendous, and they had been told to get the loans funded, whatever it took.
In fact, WaMu personnel regularly identified fraud problems with its so-called prime loans, but the problems received little attention from management. Perhaps the most compelling evidence involves two top loan producers at two different WaMu offices, called Montebello and Downey, in southern California. Each of those loan officers made hundreds of millions of dollars in home loans each year and consistently won recognition for their efforts. In 2005, an internal WaMu review found that loans from those two offices had an extremely high incidence of confirmed fraud (58% for [Downey], 83% for [Montebello]). The review found that virtually all of it stemm[ed] from employees in these areas circumventing bank policy surrounding loan verification and review. The review went on: Based on the consistent and pervasive pattern of activity among these employees, we are recommending firm action be taken to address these particular willful behaviors on the part of the employees named.
This review had taken over a year to complete and was discussed with senior management at the bank, including Home Loans President David Schneider. But virtually none of the proposed recommendations were implemented. The fraud problem was left to fester until two years later, when in June 2007, one of the banks mortgage insurance companies refused to insure any more loans issued by the loan producer from the Montebello office, and complained to WaMu's state and federal regulators about fraudulent borrower information.
WaMu then conducted another internal investigation, this one lasting ten months. In April 2008, a WaMu audit and legal team produced an internal memorandum which, at first, WaMu tried to keep from its regulator, OTS. But the OTS Examiner In Charge demanded to see the memorandum, and it was eventually turned over. He told us that once he read it, he considered it the last straw that changed his view of how the bank dealt with fraud.
The April 2008 memorandum, Exhibit 24, stated that employees at the Montebello loan center consistently described an environment where production volume rather than quality and corporate stewardship were the incented focus. At this loan center, 62% of the sampled loans from two months in 2007 contained misrepresentations and suspected loan fraud. The memorandum noted that similar levels of fraud had been uncovered at the same loan center in 2005, and that no action had been taken in response. The memorandum raised the question of whether the billions of dollars in loans from that center should be reviewed, given the longstanding fraud problem and the fact that the loans may have been sold to investors.
These fraudulent loans, shocking in themselves, were symptomatic of a larger problem. WaMu failed to ensure that its employees issued loans that met the banks credit requirements. Report after report indicated that WaMu loan personnel often ignored the banks credit standards. December 2006 minutes from a WaMu Market Risk Committee meeting stated, for example: [D]elinquency behavior was flagged in October  for further review and analysis &. The primary factors contributing to increased delinquency appear to be caused by process issues including the sale and securitization of delinquent loans, loans not underwritten to standards, lower credit quality loans and seller servicers reporting false delinquent payment status.
A September 2008 review found that controls intended to prevent the sale of fraudulent loans to investors were not currently effective and there was no systematic process to prevent a loan & confirmed to contain suspicious activity from being sold to an investor. In other words, even where a loan was marked with a red flag indicating fraud, that didnt stop the loan from being sold to investors. The 2008 review found that of 25 loans tested, 11 reflected a sale date after the completion of the investigation which confirmed fraud. There is evidence that this control weakness has existed for some time.
Sales associates manufacturing documents, large numbers of loans that dont meet credit standards, offices issuing loans in which 58, 62, or 83% contain evidence of fraudulent borrower information, loans marked as containing fraud but then sold to investors anyway. These are massive, deep seated problems. And they are problems that, inside the bank, were communicated to senior management, but were not fixed.
Option Arms.WaMu's flagship mortgage product, the Option ARM, was also marked by shoddy lending practices. The Option ARM is an adjustable rate mortgage which typically allowed borrowers to pay an initial teaser rate, sometimes as low as 1% for the first month, and then imposed a much higher floating interest rate linked to an index. The Option in the loan name refers to an arrangement which allowed borrowers to choose each month among four types of payments: payments that would pay off the loan in 15 or 30 years, an interest only payment, or a minimum payment that did not cover even the interest owed, much less the principal. If the minimum payment were chosen, the unpaid interest would be added to the loans principal, causing the loan amount to increase rather than decrease over time. In other words, the borrower could make payments as required, but still owe the bank more money on the principal each month. It was a negative amortizing loan.
Option ARMs allowed borrowers to make very low minimum payments for a specified period of time, before being switched to higher payment amounts. Most borrowers chose the minimum payment option. After five years, or when the loan principal reached a specified amount of negative amortization such as 110% , 115% or 125% of the original loan amount -- whichever came first -- the Option ARM would recast. The borrower would then be required to make the fully amortizing payment needed to pay off the loan within the remaining loan period. The required payment was typically much greater often double the prior payment -- causing payment shock and increasing loan defaults.
WaMu was eager to steer borrowers to Option ARMs. Because of the gain from their sale, the loans were profitable for the bank, and because of the compensation incentives, they were profitable for mortgage brokers and loan officers. In 2003, WaMu held focus groups with borrowers, loan officers, and mortgage brokers to determine how to push the product. A 2003 report summarizing the focus group research stated: Few participants fully understood the Option ARM. ... Participants generally chose an Option ARM because it was recommended to them by their Loan Consultant. & Only a couple of people had any idea how the interest rate on their loan was determined. It said that, while borrowers, generally thought that negative amortization was a moderately or very bad concept, that perception could be turned around by mentioning that price appreciation would likely overcome any negative amortization. The report stated: [T]he best selling point for the Option ARM loan was [borrowers] being shown how much lower their monthly payment would be & versus a fixed-rate loan. That year, 2003, WaMu originated $30 billion in Option ARMs.
To increase Option ARM sales, WaMu increased the compensation paid to employees and outside mortgage brokers for the loans, and allowed borrowers to qualify for the loan by evaluating whether they could pay a low or even the minimum amount available under the loan, rather than the high payments following recast. In 2004, WaMu doubled its production of Option ARMs to more than $67 billion. WaMu loan officers told the Subcommittee that they expected the vast majority of Option ARMs borrowers to sell or refinance their homes before their payments increased. As long as home prices were appreciating, most borrowers were able to refinance. Once housing prices stopped rising, however, refinancing became difficult. At recast, many people became stuck in homes they could not afford, and began defaulting in record numbers.
WaMu became one of the largest originators of those types of loans in the country. From 2006 until 2008, WaMu securitized or sold a majority of the Option ARMs it originated, infecting the financial system with these high risk mortgages. Like Long Beach securitizations, WaMu Option ARM securitizations performed badly starting in 2006, with loan delinquency rates between 30 and 50%, and rising.
Destructive Compensation.Destructive compensation schemes played a role in the problems just described. Hearing exhibits will show how Washington Mutual and Long Beach compensated their loan officers and processors for loan volume and speed over loan quality. Loan officers were also paid more for overcharging borrowers obtaining higher interest rates or more points than called for in the loan pricing set out in the banks rate sheets and were paid more for including stiff prepayment penalties. Loan officers and third party mortgage brokers were also paid more for originating high risk loans than low risk loans. These incentives contributed to shoddy lending practices in which credit evaluations took a back seat to approving as many loans as possible.
The compensation problems didn't stop in the loan offices. They went all the way to the top. WaMus CEO received millions of dollars in pay, even when his high risk loan strategy began losing money, even when the bank began to falter, and even when he was asked to leave his post. From 2003 to 2007, Mr. Killinger was paid between $11million and $20 million each year in cash, stock, and stock options. Thats on top of four retirement plans, a deferred bonus plan, and a separate deferred compensation plan. In 2008, when he was asked to leave to leave the bank, Mr. Killinger was paid $25 million, including $15 million in severance pay. $25 million for overseeing shoddy lending practices that pumped billions of dollars of bad mortgages into the financial system. Another painful example of how executive pay at U.S. financial firms rewards failure.
Mortgage Time Bomb.The information uncovered by this Subcommittee is laid out in over 500 pages of exhibits. These documents detail not only the shoddy lending practices at Washington Mutual and Long Beach, it shows what senior management knew and what they said to each other about what they found. Senior executives described Long Beach as terrible and a mess, with default rates that were ugly. With respect to WaMu retail home loans, internal reviews described extensive fraud from employees willfully circumventing bank policy. Controls to stop fraudulent loans from being sold to investors were described as ineffective. WaMus President described it as the worst managed business he had seen in his career. That was the reality inside Washington Mutual.
To keep the conveyor belt running and feed the securitization machine on Wall Street, Washington Mutual engaged in lending practices that created a mortgage time bomb. This chart, Exhibit 1(b), summarizes the lending practices that produced high risk mortgages and junk securities: targeting high risk borrowers; steering borrowers to higher risk loans; increasing sales of high risk loans to Wall Street; not verifying income and using stated income or liar loans, accepting inadequate documentation loans; promoting teaser rates, interest only and pick a payment loans which were often negatively amortizing; ignoring signs of fraudulent borrower information, and more.
The last two bullet points on the chart deserve particular scrutiny. Were going to hear today how, at a critical time, Washington Mutual securitized loans that had been selected specifically for sale because they were likely to go delinquent, without informing investors of that fact. Getting them sold became an urgent goal. We will also hear that at times, Washington Mutual securitized loans that had already been identified as being fraudulent, also without informing investors.
WaMu built its conveyer belt of toxic mortgages to feed Wall Streets appetite for mortgage backed securities. Because volume and speed were king, loan quality fell by the wayside, and WaMu churned out more and more loans that were high risk and poor quality. Once a Main Street bank focused on financing mortgages for its customers, Washington Mutual was taken in by the short-term profits that even poor quality mortgages generated on Wall Street.
Washington Mutual was not, of course, the only one running a conveyor belt dumping high-risk, poor-quality mortgages into the financial system. Far from it. Some of the perpetrators, like Countrywide and New Century, have already been hit with federal enforcement actions and shareholder lawsuits; others may never be held accountable. But all of us are still paying the price.
Link to Hearing Exhibits 1 - 86 [PDF 51 MB]
Link to Permanent Subcommittee on Investigations hearing page with list of witnesses and links to witness testimony