Vertical Integration and the Production of Fraud

It is useful to explore this process by examining how two banks, Bear Stearns and Washington Mutual, came to incorporate fraudulent practices into their busi­ness models. One of the most difficult aspects of understanding the financial cri­sis is to understand the internal processes of banks.

It is here that we can see how the incentives inside the bank operated. In both banks, the people at the top wanted to fund mortgages to participate in securi­tization. Their salaries and bonuses depended on the bank growing and getting more profitable. They encouraged those who worked for them to do whatever it took to make the most money. The bond traders who sold securities were the most in touch with customers who wanted to buy securities. They discovered that it was easier to sell securities that were AAA rated and had higher returns. They also, of course, earned more money if they sold more securities, and so they encouraged their originators to find mortgages to securitize that produced the most saleable securities.

The magic of securitization allowed banks to put securities together that, while containing some risky mortgages, were able to pass tests that made them look less risky. In order to comply with the wishes of the traders in the firm, the securitization facilities put pressure on originators to produce mortgages with higher yields.

Bear Stearns

According to the Financial Crisis Inquiry Commission (2011: 200), “in mortgage securitization, Bear followed a vertically integrated model that made money at every step, from loan origination through securitization and sale. It both ac­quired and created its own captive originators to generate mortgages that Bear bundled, turned into securities, and sold to investors.”

At the top of Bear Stearns's organization were the traders who bought and sold securities and derivatives of all types. Tom Marano was the head of the Global Mortgage- and Asset-Backed Securities Department at Bear Stearns from 2003 to 2007.⁴ Marano had been with the firm since it had started trading mort­gage-backed securities in 1983. He was responsible for supervising the traders of mortgage- and asset-backed securities (MBSs and ABSs) and was involved in supervising securitization in the firm during the period when the market shifted from conventional to nonconventional mortgages.

Marano described how Bear Stearns found the loans to securitize in the following way: “We got loans from three sources. We would buy loans from third-party originators. But we also had the capacity to originate loans through firms we owned and controlled. Bear Residential Mortgage Corporation (Bear Res) originated Alt-A loans, while Encore Credit Corporation was acquired late in the game and did subprime. EMC Mortgage Corporation serviced and col­lected payments on subprime and Alt-A loans” (Financial Crisis Investigation Commission, 2010b: 6).

Bear Stearns's strategy of vertical integration had a long history back to 1990 when it first bought EMC.

Wall Street seems to have thrown out its old and trusty playbook. Instead of pulling back in 2006, several major firms went on a spending spree. Well, despite the gloomy outlook, competition is not letting up. First, cli­ents have been setting up capital-markets desks to securitize their own loans in their own version of vertical integration. Second, more players are trying to buy loans that are still for sale. That's especially true of the mortgage market, where vertical integration has been most rampant. “In 2000 we'd have maybe five or six groups bidding on a loan sale,” says Com- maroto. “Now there are 20 or more. And it's not just us banks bidding, but hedge funds and whole-loan trading boutiques.” The more bidders, the higher prices can go, and that, of course, can undermine the economics of a securitization. It also means a desk has more chance of not getting enough loans in a timely manner.

Currie (2007: 23) continues,

It would be unfair in the extreme to characterize vertical integration as little more than a headlong rush into lending as a way of staving off ABS underwriting desks' death spiral because of a lack of access to product. There are also significant advantages to owning a lender. For starters, hav­ing a captive lender eliminates the need to hunt for as many loans in the marketplace. Next, as owners, the investment banks earn all the fees from originating the loans to the gain on selling into a securitization.

Encore Credit Corporations was acquired in February 2007, just a few months before Bear Stearns went bankrupt. Marano describes the motivation to buy the subprime mortgage originator as follows:

Between 2005 and 2006, we experienced early payment defaults on sub­prime mortgages—loans defaulting in the first ninety days—at a fairly high rate. We set up a team to put the loans back to the originators. It would have been due to the poor origination standards by third parties. They were defaulting before they were securitized. We ended up walking away from New Century and First Long Beach in 2006 that had been providing us with loans. We decided to buy Encore in 2007 in order to secure a supply of mortgages for securitization. The mortgage markets had taken a few hits and the price of the company was trading very low. We thought we could implement our own standards in originating loans, and we could get bet­ter quality subprime loans in order to continue our business of producing MBS and CDO. (Financial Crisis Investigation Commission, 2010b: 9)

One crucial question in understanding how Bear Stearns operated is to con­sider the link between the trading desk that sold securities to customers and the originators who provided mortgages for sale to investment banks. Mary Hag­gerty was in charge of procuring loans for securitization at Bear Stearns. She was the person who linked the originators of the loans to the traders through her overseeing the procurement of loans for securitization.⁵ In her testimony to the Financial Crisis Investigation Commission, Haggerty said that the traders would provide the originators with a sheet describing the types of securities they would be able to most readily sell. They would consult with the trading desk to set up rating sheets that would be used by Bear Res and EMC to price mortgages to originators and brokers.

In her testimony, Haggerty described how this process worked:

On a day to day basis, the decision to purchase mortgage loans comes from the trading desk and it comes in two ways: one is through a flow channel and one is through a bulk channel. The bulk channel is high­ly—can be highly controlled by the trading desk because, by its nature, the traders can choose to bid on the loans or not bid on the loans or bid on them to lose, frankly. The flow channel is a situation where the trad­ers publish pricing daily and an approved seller can deliver loans that meet the criteria pursuant to those prices. So, in order to control volume, among other things, the traders control pricing where, in the case of bulk, decide whether or not to bid. So, the volume was really controlled by that pricing mechanism and what the traders wanted to buy. (Financial Crisis Investigation Commission, 2010a: 29).

Bear Stearns was one of the banks that began to aggressively enter noncon- ventional markets after 2003. It was among the largest and most important issu­ers / underwriters for mortgagese curities fromtooo to 2007. Itwasthe leading underwriter of nonagency MBSs in 2001, 2004, and 2005 and the second- or third-largest underwriter in the other years. In the same era, Bear Stearns was in the top ten in issuers, and in 2004-2006, it was in the top five (Inside Mortgage Finance, 2009: vol. 2, pp. 18-25).

In 2000, when it was the second-largest underwriter, Bear Stearns produced $9.4 billion worth of mortgage securities. At its peak in 2005, it underwrote $130.8 billion worth of mortgages. In order to keep its securitization business going, Bear Stearns needed to rapidly increase its role as a conduit for mortgages (i.e., as a wholesale buyer). It grew from wholesaling about $2 billion mortgages in 2000 to $57.2 billion in 2007.

While Bear Stearns was a player in underwriting and issuing many different kinds of nonconventional mortgages, it specialized in option ARMs. An option ARM is a fifteen- or thirty-year adjustable-rate mortgage that, after the expira­tion of a very short teaser rate period (typically one month to one year), offers the borrower four monthly payment options (only the lowest option is set forth in the loan contract): a specified minimum payment that is fully amortizing based on the teaser rate (which is no longer in effect) and that is the only payment amount set forth in the borrower's loan documents, an interest-only payment based on the sum of a margin and index that is substantially higher than the teaser rate, a fifteen-year fully amortizing payment, and a thirty-year fully amortizing payment.

When a borrower makes a monthly payment that is less than the accrued in­terest in that month, there is negative amortization, which means that the unpaid portion of the accruing interest is added to the outstanding principal balance. Many of these loans contained prepayment penalties that made it very costly for borrowers to refinance until twelve or sometimes thirty-six months had passed. This meant that borrowers could easily find themselves with mortgage loans that were experiencing substantial negative amortization while their ability to refi­nance was restricted (Kramer and Sinha, 2006).

It is useful to show the relative importance of different kinds of nonconven- tional mortgages fed into Bear Stearns's overall securitizations. In January 2007, the product breakdown for Bear Stearns was 48.8 percent option ARM; 12.2 per­cent Alt-A, second lien; 11.2 percent Alt-A, ARM; 6.9 percent Alt-A fixed; and 3.7 percent subprime. Bear Stearns committed itself to the option ARMs business in a significant way. The market for option ARMs exploded in 2005 and 2006 but fell from $775 billion in 2006 to $618 billion in 2007 (Inside Mortgage Finance, 2009: 140-141). Its expansion was part of the general expansion of nonconven- tional mortgages in this period:

The reason the U.S. mortgage industry profits remain high in 2004 and 2005 is tied to the rise of the nonconventional sector. The reason mort­gage companies take on riskier mortgages is because of the loan yield. If a conventional mortgage is offered at 5.5%, an A-loan would fetch 7% allowing the lender to make more profit. But something new is afoot. Sub­prime lenders such as New Century, Option One, and First Franklin have expanded their menus to include exotic mortgages such as interest only adjustable rate mortgages and payment Option ARMs that allow consum­ers low monthly payments. (Muolo, 2005: 43)

In August 2006, Friedman Billings Ramsey issued a report stating that the market for option ARM securities was quite strong:

FBR said investors were paying prices of 103 to 104.5 for payment Option ARMs. The investment banking firm says these prices can result in gain on sale margins of 150 to 200 basis points. “As a result, we would be buyers of the Option ARM lenders into second quarter earnings,” writes analyst Paul Miller. “It is one of the few products that can be put into a portfolio to­day and earn a respectable return. Despite fears in the equity markets over Option ARM credit quality, it appears the fixed income market is less con­cerned given their appetite for the product.” (Origination News, 2006: 44)

Bear Stearns (2006: 7) argued that the great increase in the sale of option ARMs was attributed to the demand for such mortgages from people who might not otherwise be able to afford to buy the house that they wanted. In the same document, they elaborate by saying, “In recent years, adjustable rate lending has increased in prominence in the U.S. mortgage market. The steepness of the yield curve in 2004 was instrumental in raising the percentage of new originations that were ARMs” (Bear Stearns, 2006: 9).

The yield curve began to flatten in 2005 and 2006. The change in the yield curve, however, did not slow down the number of option ARM originations; in fact, in those two years, the sale of those products increased (as we have already documented). Bear Stearns (2006: 9) reports, “However, even as the curve start­ed to flatten in 2004, innovation in affordability products kept ARM demand high. And with the flat or inverted curve since mid-2005, ARM market share has not dramatically declined in the non-agency sector. Thus, it appears that ARM lending has gained a leadership position in the non-agency, sector that may not appreciably change in response to possible future changes in the shape or level of the yield curve.”

This raises the question of why the option ARM market did so well in the face of a flattening yield curve. Businessweek reported,

As home prices soared, banks pushed adjustable rate loans with lower initial payments. When those got too pricey, banks hawked loans that re-

quire only interest payments for the first couple of years. And then they flogged Option ARMS as affordability tools for the masses. Banks tapped an army of unregulated mortgage brokers to do what needed to be done to keep the money flowing, even if it meant putting dangerous loans in the hands of people who couldn't handle or did not understand the risks. Wall Street greased the skids by taking on much of the new risk banks were creating. (Der Hovanesian, 2006: 73)

So where was the fraud? Option ARMs were a tool that produced predatory lending and fraud. Many people who were offered option ARMs were sold the product as a low-cost way to buy or refinance a house. Often, they could have qualified for a more conventional mortgage. But the underlying conditions of option ARMs were often obscured to buyers in order to facilitate making more profits on the part of sellers. It is useful to consider one of the most important fraud cases decided against banks in the financial crisis: Monaco v. Bear Stearns.⁶

The case was filed as a violation of truth in lending laws in California. The plaintiffs in the case, two homeowners, claimed that Bear Stearns through Bear Res and EMC hid the basic facts of the loan from them. It is worth quoting their claim at some length in order to understand how the fraud took place:

The two most important pieces of information in any mortgage loan are the interest rate and the amount of the monthly payments. For the Defen­dants' Option ARM loans, the disclosures of both pieces of information were misleading and omitted material facts. The loan documents disclosed a teaser rate, but they did not disclose that this rate would sharply increase after only one month. The loan documents disclosed a low monthly pay­ment for the first 3-5 years of the loan, which was based on the teaser rate, but did not reflect the actual amount of interest being charged or the amount Plaintiffs actually owed each month.

Piffnwfir were not informf d ofthe sharp increase in theinterest rate and the fact that their monthly payments were not enough to pay the in­terest accruing on the loan until they made multiple payments following the close of the loan. Had Defendants disclosed this material information, Plaintiffs would not have purchased the Option ARM loans. (Monaco v. Bear Stearns: 13)

In 2013, the case was settled in favor of the plaintiffs. They received a payment of $18.3 million. What this case proves is that Bear Stearns knowingly omitted information about loans to mortgagors. There is good evidence that Bear Stearns then turned around and made securities out of these mortgages. To sum up, Bear Stearns aggressively entered the option ARM market, worked hard to convince mortgagors to take the mortgages, and, in order to keep making these loans, mislead them about the actual terms of the mortgages.

The last issue to consider is the degree to which the loans that were given to people with less-than-stellar credit failed. The banks were certainly complicit in working to sell as many mortgages as they could. As the ability of potential mort­gagors to repay their loans increased, banks became more desperate in finding new loans to make. The option ARM loan was designed to allow mortgagors to get into a house because of the low teaser rates at the beginning. But because these rates adjusted quickly, many of those who got these loans found themselves unable to pay. With loans that had negative amortization and with loan condi­tions that meant that owners could not easily refinance, many mortgagors ended up in foreclosure.

Table 7.1 shows how various vintages of mortgages issued from 2002 to 2006 performed as of 2010. There are two important patterns to note. First, over time, loans of both conventional and nonconventional mortgages increased their rates of delinquency. Mortgages that were originated in 2002-2003 were less likely to have problems than those issued in 2004-2007. This suggests that there was a clear deterioration in loan standards from 2004 to 2008 for all mortgages. This reflected the need of banks to originate mortgages in order to securitize. Overall, the loans that were the most likely to have trouble were those issued in 2006-2007.

Table 7.1 : Loan performance of nonconventional mortgages issued in 2002-2008

Table 7.1 : (continued)

Source: J.P. Morgan (2010:5)

Note: LTV = Loan to Value; FRM = Fixed Rate Mortgage; REO = Real Estate Owner, refers to properties that have been foreclosed on that have come into the possession of a mortgage lender.

The second important pattern is that the worse-performing loan categories were Alt-A loans. In this category, the option ARMs were the ones most likely to have problems: 46 percent of the Alt-A option ARMs issued in 2006, the peak year for that type of mortgage, were delinquent in some fashion. Almost 20 per­cent were ultimately foreclosed on. This compares with foreclosure rates of less than 4 percent for those who got conventional mortgages in 2006 (up from less than 1 percent in 2002). The leader in originating this category was Bear Stearns. Bear Stearns, by committing fraud through lying to people about how much they would have to pay for their mortgages, actually made the financial crisis worse. Many of the people who got those loans found the conditions so onerous that they felt it necessary to walk away. The Bear Stearns case shows how fraud and the ultimate disaster that was the financial crisis were intertwined. In their desire to continue to securitize mortgages, Bear Stearns found it necessary to originate any mortgage they could, including many that no one would be able to pay back. This made them cross the line in their dealings with mortgagors by engaging in predatory lending and fraudulent truth in lending. Those who were on the receiving end of these too-good-to-be-true mortgages found themselves going deeper into debt with little ability to refinance. Being unable to pay their mortgages, they lost their homes at high rates.

Washington Mutual

Washington Mutual Inc. (known as WaMu) called itself “the bank for everyday people” because it focused on consumers and small-to-medium-sized business­es. WaMu grew by the end of 2007 to be the nation's largest savings and loan in both assets ($328 billion) and revenues ($25.5 billion). The CEO at WaMu was Kerry Killinger, who joined the company in 1983 and became chief executive in 1990. He inherited a bank that was founded in 1889 and had survived the Depression and the savings and loan scandal of the 1980s. An investment ana­lyst by training, he was attuned to Wall Street's hunger for growth. Between late 1996 and early 2002, he transformed WaMu into the nation's sixth-largest bank through a series of acquisitions. A crucial deal came in 1999 with the purchase of Long Beach Financial, a California lender specializing in subprime mortgages. From 2000 to 2003, WaMu's retail branches grew 70 percent and reached twen­ty-two hundred across thirty-eight states.

As the conventional mortgage market dried up in 2003, WaMu found itself in the same position as the other vertically integrated banks. In order to turn around its lagging mortgage division, WaMu had to shift its mortgage origina­tion strategy. It did so by following Bear Stearns and pushed nonconventional mortgages, most notably ARMs but also subprime mortgages. In prepared re­marks for the 2004 third-quarter conference call, CEO Killinger said, “We are paying closer attention than ever to product mix to assess our profit by product and distribution channels, and exercise stronger controls than ever. The goal is to ensure that we are disciplined about originating higher margin product whenev­er we can. In this market our emphasis is ARM product origination, principally for our balance sheet. These ARMs helped the balance sheet to grow by $10.3 billion this past quarter” (Kennedy and Bowen, 2008: 2).

During the same third-quarter 2004 conference call, Killinger said, “The op­tion-ARM product is the key flagship product for our company” (Kennedy and Bowen, 2008: 2). Between April 2004 and the end of 2007, WaMu underwrote $184.8 billion in option ARMs. Most of these mortgages were securitized, and many remained on WaMu's balance sheet. WaMu's ARMs expanded from about 25 percent of new home loans in 2003 to 70 percent by 2006.

Although WaMu appeared to exit the subprime lending business in 2003 when it sold Washington Mutual Finance to Citigroup, it retained a significant pres­ence in the subprime market through its 1999 purchase of Long Beach Mortgage. As of early 2008, Long Beach Mortgage was one of the country's largest lenders to people with damaged credit. While Long Beach operated in all fifty states except Mississippi, its largest market by far was California. In 2005, Long Beach Mortgage made more than one-quarter of all WaMu home-purchase loans.

As the conventional mortgage market turned down, Mr. Killinger expressed a desire to see that business grow faster than WaMu's traditional mortgage lending because it was more profitable: “We earn better margins in the subprime business because we're very efficient and have an advantage over some competitors” (Mayo and Allison, 2005). In late 2005, WaMu planned to expand subprime lending along the East Coast, taking subprime mortgages directly to consumers through its retail branches and home loan offices. In prepared remarks for a conference call discussing WaMu's 2004 results, Senior Vice President and CFO Tom Casey said, “First, we remain comfortable that we can achieve average asset growth in the 10 to 12% range. While we sell a significant portion of our ARM production into the secondary market, we still expect ARM retention to be a driver of asset growth. We also expect growth to come from planned increases in the home eq­uity, multi-family, and nonprime lending” (Kennedy and Bowen, 2008: 5).

Pressures were brought on employees of WaMu to increase the sale of ARMs and subprime mortgages. This pressure pushed sales agents to pump out loans while disregarding borrowers' incomes and assets, according to former employ­ees. The bank set up what insiders described as a system of dubious legality that enabled real estate agents to collect fees of more than $10,000 for bringing in borrowers, sometimes making the agents more beholden to WaMu than they were to their clients. WaMu gave mortgage brokers handsome commissions for selling the riskiest loans, which carried higher fees, bolstering profits and ulti­mately the compensation of the bank's executives. WaMu pressured appraisers to provide inflated property values that made loans appear less risky, enabling Wall Street to bundle them more easily for sale to investors. “It was the Wild West,” said Steven M. Knobel, a founder of an appraisal company, Mitchell, Maxwell & Jackson, which did business with WaMu until 2007. “If you were alive, they would give you a loan. Actually, I think if you were dead, they would still give you a loan” (Goodman and Morgenson, 2008).

A New York Times article written by Peter Goodman and Gretchen Morgenson (2008) included interviews with a large number of people who worked for WaMu in its offices in California, Florida, Texas, and New Jersey. It is useful to tell a few of the stories, as they capture the flavor of how WaMu was aggressively pursuing loans:

An employee at WaMu's San Diego processing office, Sherri Zaback's job was to take loan applications from branches in Southern California and make sure they passed muster. Most of the loans she said she han­dled merely required borrowers to provide an address and Social Secu­rity number, and to state their income and assets. She ran applications through WaMu's computer system for approval. If she needed more infor­mation, she had to consult with a loan officer—which she described as an unpleasant experience. “They would be furious,” Ms. Zaback said. “They would put it on you that they weren't going to get paid if you stood in the way” (Goodman and Morgenson, 2008: 4)

The sheer workload at WaMu ensured that loan reviews were limited. Ms. Zaback's office had 108 people and several hundred new files a day. She was re­quired to process at least ten files daily. “I'd typically spend a maximum of 35 minutes per file,” she said. “It was just disheartening. Just spit it out and get it done. That's what they wanted us to do. Garbage in, and garbage out” (Goodman and Morgenson, 2008).

WaMu flourished in Southern California, where housing prices rose so rapid­ly during the bubble that creative financing was needed to attract buyers. To that end, WaMu embraced so-called option ARMs, adjustable-rate mortgages that enticed borrowers with a selection of low initial rates and allowed them to decide how much to pay each month. But people who opted for minimum payments were underpaying the interest due and adding to their principal, eventually caus­ing loan payments to balloon. Customers were often left with the impression that low payments would continue long term, according to former WaMu sales agents. For WaMu, variable-rate loans, option ARMs in particular, were espe­cially attractive because they carried higher fees than other loans and allowed WaMu to book profits on interest payments that borrowers deferred.

WaMu's retail mortgage office in Downey, California, specialized in selling option ARMs to Latino customers who spoke little English and depended on advice from real estate brokers, according to a former sales agent who request­ed anonymity because he was still in the mortgage business. According to that agent, WaMu turned real estate agents into a pipeline for loan applications by enabling them to collect “referral fees” for clients who became WaMu borrowers. Buyers were typically oblivious to agents' fees, the agent said, and agents rarely explained the loan terms (Goodman and Morgenson, 2008: 6).

As the boom faded in 2006 and 2007, nearly half of all option-ARM borrow­ers made minimum negative amortization payments. WaMu recorded a net loss for 2007 of $67 million, compared with a net income of S3.56 billion in 2006. The decline was primarily the result of significant deterioration in the company's residential mortgage loan portfolio and a sudden and severe contraction in sec­ondary mortgage market liquidity for nonconforming residential loan products such as ARMs (Kennedy and Bowen, 2008: 4). By April 2008, WaMu had posted a first-quarter loss of $1.14 billion and increased its loan loss reserve to $3.5 bil­lion. Its stock had lost more than half its value in the previous two months.

In September, CEO Killinger was forced to retire. Later that month, with WaMu buckling under roughly $180 billion in mortgage-related loans, the FDIC seized the bank and sold it to JPMorgan Chase for $1.9 billion, a fraction of the $40 billion valuation the stock market gave WaMu at its peak. Because the bank was taken over by the government and liquidated, it has been difficult for peo­ple who were harmed to sue. As a result, victims of the predatory lending and mortgage fraud the bank committed have not been compensated. It is clear that WaMu behaved very much like Bear Stearns. It promoted mortgages to people with little chance to pay them back. It made money off of fees and the securities it made using the mortgages. It ended up insolvent as those mortgages ended up in foreclosures. It continued to hold many of the securities based on those mortgages and as a result suffered huge losses.