Shedding Light on the Crisis' Dark Corners
10/01/2009 2:08 pm EST
As more and more experts declare the financial crisis of 2008-2009 is over, the post mortems have begun.
Publishers have churned out books on the subject by noted journalists like David Wessel and Gillian Tett, as well as economists like Mark Zandi and pundits like Barry Ritholtz.
And enterprising academics are weighing in with their own take on what caused the crisis that pushed the world economy to the brink of catastrophe.
Some recently published studies have shed unusual light on the murkiest parts of the housing market, where the crisis began, and come to some striking conclusions on what went wrong and who was to blame.
They may be full of impenetrable jargon and mind-numbing equations, but the authors have access to huge, unique sets of data, and they’re thorough and pretty objective. So, no shouting and political posturing, for a change.
At the heart of the housing bubble and subsequent crash was a near-total breakdown in the standards that had guided mortgage lenders for decades. Structural changes in the industry, particularly the dominance of securitization, caused key players to eschew meaningful due diligence. Incentives were strong and opportunities abounded for lenders and borrowers to commit what can only be described as fraud. Given all that, disaster was inevitable.
Of course, Alan Greenspan’s Federal Reserve kept interest rates too low for too long early in the decade. That was the backdrop for an explosion in mortgage credit that soon spilled over into subprime borrowers, as Atif Mian and Amir Sufi of The University of Chicago Booth School of Business wrote in their 2008 study.
The two scholars suggest that securitization played a key role.
And indeed, securitization—the packaging of mortgages, credit card debt, auto loans, and other financial assets into securities that are then sold to investors—had become pervasive by the turn of the Millennium. Spearheaded by government sponsored entities (GSEs) like Fannie Mae and Freddie Mac, securitization helped banks and other lenders keep liabilities off their balance sheets and share the risk with others.
About $3.6 trillion of US mortgage loans were securitized as of 2006, and securitization rates (the dollar value of securitized loans divided by the dollar value of loan originations) shot up from less than 30% in 1995 to over 80% in 2006.
That caused a relaxation of standards, to say the least. Tracking more than a million mortgage loans originated between 2001 and 2006, four academics—Benjamin Keys, Tanmoy Mukherjee, Amit Seru, and Vikrant Vig—concluded that “existing securitization practices did adversely affect the screening incentives of lenders”—leading to default rates 10%-25% higher than portfolios that weren’t likely to be securitized.
Professor Seru—also from the University of Chicago—explained that earlier, Freddie and Fannie would typically take a sample of conforming mortgages from banks and compel them to keep a good number in their own portfolios.
Those requirements eased as the housing bubble grew and Wall Street firms started competing with the GSEs, particularly in the subprime market. Demand shot up for anything yielding more than Treasuries as insurers, pension funds, and other large institutions clamored for “tranches” (pieces) of these securitized mortgage pools. Even subprime was fair game, because ratings agencies had given AAA ratings to many of those tranches, too.
With so many willing buyers—and fewer loans on their balance sheets—banks had little incentive to do real due diligence.
“As soon as the loan is outside the bank’s books, they are a free bird,” Seru told me. Instead, lenders and investors relied on one of the mortgage industry’s great rules of thumb: the FICO score.
FICO scores are rough measures of the likelihood a borrower will repay a loan based on his or her credit history. Most scores range between 500 and 800, but the key benchmark is 620. Prospective borrowers with FICO scores above that level are more likely to get mortgages than those who have lower scores.
Unfortunately, the study showed, lenders, investors, and rating agencies relied almost exclusively on that score and ignored “soft” information, such as how much supporting documentation the borrowers had. Result: Low-documentation loans to borrowers with FICO scores above 620 defaulted more than similar loans issued to borrowers with FICO scores below that threshold, because lenders were more likely to scrutinize loans to the lower-scoring borrowers.
That finding dovetails nicely with another study by Wei Jiang, Ashlyn Aiko Nelson, and Edward Vytlacil, called “Liar’s Loan?”
The three studied all 721,767 loans originated by an unidentified “major national mortgage bank” between January 2004 and February 2008. The bank relied on low- and no-documentation loans often produced by mortgage brokers to pursue “an aggressive expansion strategy” that helped it grow by 50% a year from 2004 to 2006. Low-doc loans produced by brokers accounted for an astonishing 75% of the bank’s loan originations by late 2006, the top of the market.
The bank’s loans had a cumulative delinquency rate of 28% by 2009, the study reported. Loans issued by brokers were 50% more likely to go into delinquency than the bank-generated mortgages, because brokers pursued “marginal borrowers.”
Ironically, many of these borrowers had pretty good FICO scores, so their loans could be easily securitized and sold. But the study estimates that borrowers exaggerated their income by a median of about 20% to get low-doc loans—which, of course, defaulted much more frequently.
“People wound up buying houses well beyond the limits of their incomes,” Professor Wei Jiang of the Columbia Business School told me.
So, who was to blame? Individuals? Brokers? Or both?
“You can’t separate the two,” she said. “They have to be in collusion to make that happen.”
Indeed. It’s too easy to blame greedy borrowers for all of this. They had lots of help and encouragement from corrupt mortgage brokers. Plenty of others were making too much money to pay attention—and regulators were nowhere.
“The development of financial markets and financial products separate people who take action from those who face consequences,” says Jiang. In other words, incentives were out of whack, and the taxpayer wound up footing the bill.
Until you change those incentives—and there are some good ideas out there—we risk a repeat of the crisis, albeit in another form.
As Professor Seru put it: “You want more skin in the game for these banks.” And for borrowers, investors, Wall Street firms, rating agencies, and regulators, too.
It’s called accountability and responsibility. Maybe we ought to try it some time.
Howard R. Gold is executive editor of MoneyShow.com. The views expressed here are his own.