Confessions of a Subprime Lender

by Richard Bitner

  The Demise of the Industry

  As early as eight years ago, a series of events and trends began that promoted the rapid growth and contributed to the eventual decline of subprime lending. I’ve broken them down as follows:• Three industry-specific events that helped subprime lending to expand.

  • What happened when lenders experienced lapses in judgment.

  • The decline in profitability.

  • How new and incompetent brokers impacted the industry.

  • The development of new and riskier mortgage products.

  The Keys to Growth

  Beginning in 2000, several events occurred that promoted the expansion of subprime lending. While each one was independently significant, the collective impact served as a major catalyst for growth.

  First, Standard and Poor’s (S&P) concluded the piggyback mortgage (where the customer took out a simultaneous second loan in lieu of a down payment) was no more likely to default than a single loan. The other agencies took the same position shortly thereafter. While the event went largely unnoticed outside of the industry, its impact was important.

  Bill Dallas, viewed by many as a mortgage innovator, was the first to offer a 100 percent mortgage in partnership with Freddie Mac. He also believed the piggyback, or 80-20 mortgage, would perform as well as the single 100 percent loan provided the FICO score was over 680. By adopting a more liberal set of credit standards, the rating agencies effectively gave birth to the subprime piggyback mortgage. Within a few years, this product became an industry staple.

  It wasn’t until six years later that S&P reversed its decision, determining that piggybacks had a much higher probability of default. Their initial decision proved disastrous. In July 2007, Moody’s made the unprecedented move to downgrade every second-lien securitization it had rated from 2005 to the present. It’s clear the rating agencies’ initial announcement had been based on faulty assumptions.

  Second, the decision on piggybacks also affected the mortgage insurance (MI) industry. MI companies provide coverage to lenders for loans over 80 percent loan-to-value (LTV) in case of borrower default. Since the ruling allowed second-lien mortgages to be used in place of mortgage insurance, the decision effectively neutered the MI industry and created a void.

  Before S&P’s announcement, a lender that wanted to offer a loan product over 80 percent LTV required MI to securitize the product. Since profitability was tied to effective risk management, most MI companies would err on the side of caution. They performed a check-and-balance function, which kept the industry thinking rationally and restricted lenders from implementing products that were poorly conceived.

  Third, the GSEs (government-sponsored enterprises)—Fannie Mae and Freddie Mac—experienced their own problems a few years later. A deep look into their accounting practices revealed that, unlike other business scandals in which companies tried to hide losses, the GSEs made so much profit they were attempting to spread their income out over time.

  Once their accounting practices became headline news, auditors were brought in to sort things out. That fact that the GSEs tried to alter their financials was enough incentive for Congress to impose restrictions. The decision to place caps on their portfolios ultimately hindered their ability to grow. At a time when the industry was experiencing record volume, the GSEs were made to sit in the penalty box. If they hadn’t been restricted, the GSEs could have played a more active role in the secondary market.

  The combination of all three events—the growth of piggyback mortgages, the neutralization of the MI companies, and the punishment of the GSEs—removed the last barriers to growth for the subprime industry. With the investment banks and rating agencies left to serve as the industry’s moral compass, effective risk management gave way to reckless behavior.

  Lapses in Judgment

  Lapses in judgment are nothing new in the industry. Occasionally a subprime lender was too aggressive with a product offering, thinking they’d found an ingenious way to recreate the risk model. Inevitably, the product performed poorly and at Kellner we’d attribute the lender’s action to temporary insanity.

  After the first subprime meltdown in 1997, The Associates (purchased by Citigroup in 2000) offered a 95 percent loan-to-value program (5 percent down payment) for borrowers with 540 credit scores. Historically, the product required a minimum credit score of 560 to 580, since default models indicated loan performance dropped precipitously below that level. Their decision to break from conventional thinking reminded me of the Roadrunner cartoons, with The Associates playing the part of Wile E. Coyote, super genius.

  When The Associates paid us 600 basis points for using the product, we thought someone in their trading department had spiked the water cooler. As a new company struggling to survive, we were happy to use this product and it was instrumental in helping us to get over the initial hump. In our first four months, it represented nearly 40 percent of our business. It had the two things every subprime wholesale lender wanted—a unique niche and a high margin. Given the absurdity of the product, it seemed only fitting that the first mortgage we ever closed, which fit its guideline, went to foreclosure less than a year later. There was nothing fraudulent or deceptive about the deal. It was just a high-risk loan based on a flawed risk model.

  The product offering was short lived. When Citigroup purchased The Associates, they immediately discontinued the program. Some time later, a colleague confirmed what many of us had already expected—the product performed poorly.

  Another profound lapse in judgment occurred in 2003 when RFC offered 100 percent financing for borrowers with 560 credit scores. Until that point, it was generally accepted that 580 was the minimum score. Writing a 100 percent loan with a 560 score was like swimming with sharks—it was only a matter of time until you were bitten.

  At Kellner we viewed RFC’s program as a desperate act. Always the conservative stalwart, RFC seldom pushed the risk envelope. When I worked for them in 1998, one of their more unusual product offerings was a 125 percent loan, which was a second-lien mortgage that allowed consumers to borrow up to 125 percent of the value of their homes. When this industry segment imploded, RFC was the only major investor left standing. They built a reputation as a smart and conservative company because they understood how to manage risk.

  When the Wall Street investment banks started capturing a larger share of the subprime market, RFC quickly fell behind. In a few years they went from being a top five investor to barely making the top 20. The 100 percent product was intended to help them reclaim market share.

  Offered only to select customers, the product proved to be a disaster. Seldom in the history of mortgage lending had a new product been so quickly pulled from the market. It showed how the pressure to compete for market share could wear down even the smartest lenders.

  This should have sounded some alarm bells. If a company widely regarded as the leader in managing risk for nonagency mortgages experienced such a profound lapse in judgment, how would other, less-skilled investors respond to the pressure?

  Profit Margins

  From 2000 to 2002 we were paid between 450 to 500 basis points (bps) for each loan sold. In some cases the figure exceeded 500 bps, as evidenced by The Associates example, but that was the exception. By 2003 we started experiencing a marked decline in profitability. With over 100 subprime wholesale mortgage companies competing for business, lenders grew volume at the expense of profit margin.

  Table 6.1 shows what happened to our profits from 2003 through 2005. The numbers are strictly for illustration purposes and don’t represent actual revenue. The first year serves as a baseline with 100 loans equaling $100 in revenue. This chart shows how the following years stacked up relative to 2003. It doesn’t take a Wharton graduate to realize the business model was headed for disaster. Although volume was growing, net revenue per loan was dropping fast. Even though expenses increased as result of growing the business, the decline was largely a result of being paid less for the product. Conversat
ions with our competitors indicated they were experiencing a similar trend.

  Table 6.1 Net Revenue Comparisons

  Several things contributed to this decline. First, the largest subprime lenders started a price war. Companies like New Century and Argent offered rates that weren’t compatible with the risk levels. We tried to win customers by offering stellar service and for a while it worked. But once technology leveled the playing field, our competitors improved their service. We had to shrink our profit margins to remain competitive.

  The pricing pressures meant small and medium-sized lenders were hit the hardest. The same investors who purchased our loans had wholesale divisions that undercut our pricing. Since the biggest lenders put loans directly into a security, their margins were higher, which enabled them to compete better in a price war. As a pass-through lender, another layer in the food chain, we didn’t have that luxury.

  Second, it’s no coincidence our revenue peaked just as the fed funds rate bottomed out. While that indicator doesn’t dictate fixed mortgage rates, it influences the overall cost of money, which impacts interest rates for ARMs. Keep in mind that most subprime loans are adjustable rate, not fixed. Kellner’s ARM to fixed product ratio was 80/20, similar to most of our competitors. As the Fed increased the funds rate by more than 4 percent from 2004 to 2006, interest rates for subprime ARMs remained flat. The only way for revenues to keep pace was to increase loan production.

  Watching this scenario unfold, I realized the industry was losing touch with reality. I frequently spoke with a colleague and competitor who owned Concorde Acceptance Corporation and we would talk about the state of the industry. We often discussed the risk-reward curve, which helped analyze the effectiveness of our business model. By late 2004 we both felt the business had reached a point where the risk of being a wholesale subprime lender outweighed the financial rewards.

  Rational thinking dictates that when the cost of money goes up, interest rates should follow. While some reduction in margin is acceptable and expected in a highly competitive market, the leading subprime companies took it to the extreme. Unfortunately, as margins were getting squeezed, the most critical factor was being ignored—risk.

  At a basic level, mortgage lending is nothing more than effective risk management. If a lender offers a high-risk product and profit margins continue to drop, one of two things must happen. The lender either increases interest rates or tightens underwriting guidelines to compensate for the reduced margin and subsequent risk. Not only did the industry choose to ignore both principles, it went in the opposite direction by developing more aggressive products.

  New Products—A Meltdown of Epic Proportions

  If the subprime industry was teetering on the edge of a cliff, relaxed underwriting standards pushed it over the edge. Before discussing the particulars, here is a quick recap of the events leading to the industry’s demise.• With the advent of the piggyback mortgage and the neutralization of mortgage insurance companies and government-sponsored enterprises, investment firms and rating agencies were left to regulate the industry.

  • Interest rates fell to record lows, creating a frenzy among consumers to acquire investment properties, treat their homes like ATM machines, or achieve the American dream by owning a home regardless of whether they could afford it.

  • By 2003 brokers were originating the majority of all subprime loans. By 2006, the figure peaked at 63 percent. With many loan officers new to the business, this unregulated and unsupervised group of originators took a bad situation and made it worse.

  • The intense pricing war among subprime lenders caused a reduction in profit margins. For revenues to keep pace required lenders to finance more borrowers, which led to the development of less restrictive underwriting standards.

  • Since the agencies were overly aggressive in how they rated subprime securities, the stage was set for riskier loan products to enter the mix.

  There are enough examples of foolish product offerings and guideline changes to fill an entire chapter. I’ve detailed a few of them to provide some insight on how the industry lost its sensibilities. Looking back, it’s clear that each was a disaster waiting to happen.• 100 Percent Stated Income Loan. Even though Countrywide wasn’t the only lender to offer stated income loans, their offering was risky. Lending to borrowers with no down payment and no proof of income had merit as long as their credit scores were high (700+). Countrywide offered this product to self-employed borrowers with 620 scores and wage earners with 640 scores.When stated income loans were developed in the 1980s, they were designed specifically for the self-employed borrower. They required a sizable down payment, excellent credit history, and intense scrutiny of the appraisal. Allowing a borrower who earned a set wage to qualify for this program was not an option. But once borrowers with mediocre credit could finance with no down payment and no income verification, it was the beginning of the end. Since the interest rate was only slightly higher for stated income loans compared to full income documentation, brokers opted for the path of least resistance. Was there any need to bother with collecting tax returns and pay stubs when the interest rate for a stated income loan was only three-eighths of a percent higher?

  • Investment Property Loans. While low interest rates fueled the market for investment properties, riskier mortgage products took the demand to another level. A subprime product that historically required a minimum down payment of 10 percent and proof of income was being offered with no money down and no income verification. At one point, lenders advertised the loan for borrowers with 660 credit scores, enabling speculators to simultaneously purchase multiple properties. As a result, speculative buying in markets like Las Vegas and Miami artificially inflated property values to unsustainable levels.

  • Guideline Changes. While credit score was an excellent indicator of loan performance, its reliability was predicated on holding other credit factors (housing history, bankruptcies, and so on) constant. This was another area where logic failed. For example, by no longer requiring solid rental verification (allowing private verifications in its place), the risk models were skewed even further. When borrowers with bad credit, no money, no verifiable income, and no history of paying rent were approved for mortgages, why would anyone be surprised that the loans defaulted?The only product Wall Street didn’t create was a stated credit loan. Could you imagine the conversation? So Mr. John-son, what’s your credit score? What’s that you say, 750? Congratulations, you’re approved!

  • A Classic Screw-Up. As the implosion of 2007 drew near, HSBC (previously Household Finance), the company that led the industry for years in 100 percent financing with 580 credit scores, suffered a psychotic break. For nearly a decade, this product had always required a 12-month housing history of no more than one 30-day late payment. When HSBC announced that borrowers with 580 scores and a 90-day late payment for housing history could qualify for 100 percent financing, it was clear that someone in the risk department had lost his mind. It didn’t matter if a borrower was one step away from foreclosure, HSBC would finance the purchase of a new home with no down payment. Even though that product lasted only a few months, it still ranks as one of the worst offerings in the history of subprime lending.

  The Walls Come Down

  As the 2006 subprime book of business started showing abysmal levels of performance, investors conducted a mass exodus from the secondary market. With no appetite for the product, the subprime industry experienced a total meltdown. One person started a web site, www.lenderimplode.com, to keep track of companies that went out of business or filed bankruptcy. He referred to the list as the lender implode-o-meter. In the first five months of 2007, more than 80 lenders (or divisions of companies) had shut down or gone out of business. By April 2008, the list topped 250.

  In perhaps the cruelest of ironies, the fallout from subprime carried over a few months later to the prime side. With the investment community retreating from all nonagency mortgage-backed securities, the Alt-A market collapsed as wel
l. Conservative lenders that never originated a subprime mortgage were left with no buyers for their products, so even they started going out of business.

  Like dozens of other lenders, Kellner closed its doors in spring 2007—#44 on the implode-o-meter. Though Ken Orman was adept at staying ahead of the curve, even he couldn’t envision just how bad things would get. With no investors willing to pay a premium for our product, he had no alternative but to shut down the company.

  Having left 16 months earlier, I managed to bypass most of the destruction. I never could have imagined that my house catching fire would be a blessing in disguise, but that’s how I’ve come to view the experience. Although Mike Elliott, our third partner, also left months earlier, Ken managed to find a few outlets to sell his remaining loans and walked away relatively unscathed. Selling them on the scratch and dent market was an expensive proposition, but nothing compared to the losses suffered by lenders who stuck it out a little longer.

  As attorneys and politicians spend months, possibly years, trying to sort out this mess, there are some pressing questions still to address. At the top of the list is whether anything can be done to minimize the damage. Certainly there are no easy answers. The next and final chapter of this book addresses the current crisis and discusses some of the solutions under consideration. It is crucially important that all of us consider the future of the industry and the systemic changes that need to be implemented.