- There are a number of reasons that the financial crash could not reasonably have been predicted or prevented
- The narrative about the financial crisis is wrong. What difference does it make? Quite a lot, actually
- A false narrative about the financial crisis is why we are again on our way to more trouble
The narrative that came out of the financial crisis was that it could have been prevented by better regulation. If the regulators had been diligent enough to see the build-up of risk in the mortgage system—the large number of subprime and Alt-A loans—they could have stepped in, closed down the subprime lending process and saved us all a lot of losses. But Wall Street greed and risk-taking were allowed to run wild, causing a financial crisis. This is essentially the conclusion of the Financial Crisis Inquiry Commission (from which I dissented) and it is the basis for the reforms implemented by the Dodd-Frank Act in 2010. It is also a fantasy.
Yes, of course, if the regulators had been astute enough to see what was happening in the mortgage markets in 2005 and 2006, and understood that this was the beginning of an unprecedented collapse of housing and mortgage values, they could have acted. But with the exception of a handful of market players—like John Paulson—who made lucky speculative bets, no one knew the essential facts about the mortgage market and no one published a predicted decline in housing prices in the range of 30 to 40 percent. I dare say few if any of those who are reading this today can honestly say that they sold off all their housing and mortgage assets in 2005 or 2006 because they could see the disaster coming.
There were a number of reasons that the crash could not reasonably have been predicted or prevented.
First, Fannie Mae and Freddie Mac, which were without doubt the dominant players in the mortgage market between 1992 and 2008, did not report the true number of subprime loans that they were acquiring. Until they were taken over by the government in late 2008, they had taken the position that the only subprime loans and Alt-A loans they held or securitized were those they had bought from subprime lenders or had been sold to them specifically as subprime or Alt-A loans. Since Fannie and the Freddie reported most of their mortgages as prime, the actual number of subprime and Alt-A loans outstanding was much larger than the number reported by data aggregators such as Loan Performance and the number of prime mortgages was correspondingly much smaller. For this reason, many market observers, including regulators such as the Fed (as discussed below) could not have understood the true dangers presented by the number of subprime and Alt-A loans outstanding.In 2011, the SEC sued three of the top officials of each of the GSEs for failing to disclose the number of subprime loans they had acquired and either securitized or held in their portfolios.
Second, on August 9, 2007, PNB Paribas, a French bank, suspended redemptions of securities in three of its funds that were invested in U.S. mortgages. Most historians of the financial crisis cite that day as the beginning of the crisis. Two days earlier, the Federal Reserve’s Open Market Committee had decided to leave interest rates where they were (about 5 and ¼ percent), and made no mention in their statement of any weakness in the mortgage markets. For weeks before that, Chairman Bernanke had been saying that the subprime problem was “contained.” Indeed, Fed economists around that time reported that the number of subprime loans in the financial system was 6.7 million. The actual number, as it turned out, was probably twice that.
In his new book, “Stress Test”, former Treasury Secretary Tim Geithner reported: “[E]ven when the Fed’s own bank supervisors and economists analyzed the system, they thought there was plenty of capital in the banks to absorb a substantial increase in losses. These public servants had good incentives to be tough, because a crisis would fall to us to clean up, and they were a talented, experienced and thoughtful group. Even so, they had a hard time coming up with economic scenarios that would generate losses large enough to seriously impair bank capital.”
If the Fed, with its huge data and staff resources did not know how many subprime loans were outstanding, thought the banks all had enough capital, and did not see the financial crisis coming, how were commercial banks, investment banks, rating agencies, risk managers, analysts—and, most of all, other regulators—supposed to have foreseen the crisis and taken the steps that would have prevented it? It is worth recalling that the commercial and investment banks that failed or came near to failing in the crisis got into trouble because they kept the AAA-rated mortgage-backed securities they thought were completely safe investments. In hindsight, you might say they were dumb or blind, or both, but no dumber or blinder than the Fed and its fellow regulators.
Third, there were other reasons that few people foresaw the crisis. As discussed below, a giant housing bubble—the largest in US history—had developed between 1997 and 2007. When bubbles grow, they tend to suppress delinquencies and defaults. Borrowers who bought their homes with 3 percent or zero downpayments but could not meet the payments found that the increased prices allowed them to refinance, or to give the property back to the lender with no loss on either side, because the value of the home had risen. So despite the fact that there were so many subprime loans around, the defaults were relatively few. This odd phenomenon was explained in part by a touching faith in the capacities of automated underwriting, newly put in place by the GSEs and others. The computer algorithms that looked at mortgages more carefully than any human could, and knew the histories of similar mortgages over the last few years, were cited as a reason why “this time it’s different.” In this new computer age, many concluded, we finally had figured out how to take the risk out of subprime loans.
Fourth, the homeownership rate was rising, much to the delight of Washington, and what’s more it was rising among underserved groups. For both Democrats and Republicans, this was proof that their policies were working. The Democrats saw the affordable housing goals (discussed below) as responsible for the growth of home ownership among underserved groups, and the Republicans saw it as a success for Bush’s ownership society. Even if a regulator was perceptive enough to see the future, what could he or she have done? At a time that there were few defaults and the home ownership rate was finally growing, would any regulator—no matter how persuasive and armed with data—be able to persuade Congress that it made sense to stop this boom before everyone who wanted a home had actually got one? I doubt it.
Does it matter if the narrative is wrong?
Okay, so the conventional narrative about the crisis might have been wrong. To quote Hillary Clinton in another context, “What difference, at this point, does it make?” Quite a lot, actually. The idea that the financial crisis was caused by private greed and risk-taking, and could have been stopped by more and better regulation—focused attention on the private sector’s role in the crisis. Yet, while the private sector suffered most of the dramatic losses, it was only a small part of the story.
My AEI colleague, Ed Pinto, and I have estimated that on June 30, 2008, there were 32 million subprime and Alt-A mortgages in the U.S. financial system—58 percent of all the mortgages outstanding. Of these, 76 percent were on the books of government agencies, primarily Fannie Mae and Freddie Mac. This shows without question that the demand for these low quality mortgages came from the government, not the private sector. In addition, it makes clear that the private sector—with only 24 percent of the subprime and Alt-A mortgages—was not the principal cause of the crisis.
Why did Fannie and Freddie—and other government agencies such as FHA—acquire so many subprime and Alt-A loans? The answer is the affordable housing goals, which were imposed on the GSEs in 1992 and required them to buy a certain quota of mortgages that had been made to borrowers at or below the median income where they lived. The goals were increased by HUD over the years, and subgoals added for home purchase loans and very low income groups, through the Clinton and George W. Bush administrations. By 2007, they required that 55 percent of all loans the GSEs purchased be made to low income borrowers.
One of the chief obstacles to the purchase of homes by low income and underserved borrowers is the required downpayment; in order to find goals-qualifying loans, by 1995 Fannie and Freddie were accepting 3 percent downpayments and by 2000 they were accepting zero downpayments. Low downpayments increase homeownership, but they also increase housing prices. If a potential buyer has $10,000 for a downpayment, he or she can buy a $100,000 house if the downpayment is 10 percent, but a $200,000 house if the downpayment is 5 percent.
The GSEs’ reduction in downpayment requirements, which began shortly after the affordable housing goals went into effect, eventually spread to the market as a whole and—together with the government’s support for low mortgage interest rates—built an enormous housing price bubble between 1997 and 2007. The chart above, based on Robert Shiller’s data, shows the size of this bubble in relation to previous bubbles. The vertical lines show the high points of previous bubbles and where the bubble already stood in 2000 and 2003, when some argue that the Fed’s monetary policy caused the bubble. It’s clear that while low interest rates in and after 2003 might have accelerated the bubble, it was already at historic levels well before that.
When the bubble collapsed in 2007 and 2008, driving down housing prices throughout the United States, it caused an unprecedented number of delinquencies and defaults among the subprime and Alt-A mortgages that then formed a large majority of the outstanding mortgages. Investors fled the market for mortgage-backed securities, forcing huge asset writedowns among investment and commercial banks and other holders of these instruments. When Lehman Brothers declared bankruptcy in September 2008, a full-scale market panic erupted, with banks hoarding cash and refusing to lend to one another, even overnight.
On this set of facts, it doesn’t seem as clear that private greed and risk-taking caused the financial crisis. Indeed, if Fannie and Freddie had been encouraged to maintain the underwriting standards that they had traditionally followed—a downpayment of 10 to 20 percent, a FICO score above 660, and a back-end debt-to-income ratio of no more than 38 percent—the financial crisis might never have occurred. These underwriting standards had generally produced a default rate of less than 1 percent, but variance from these standards produced much higher rates. Yet, the government’s role in forcing a decline in underwriting standards through the affordable housing goals has not been cited as a significant contributor to the crisis. Indeed, the Financial Crisis Inquiry Commission specifically rejected the idea that Fannie and Freddie or the government’s housing policies had any more than a marginal role in the crisis.
The consequences of a false narrative
This reading of history had consequences. If the reason for the crisis was private greed and risk-taking, then the proper response was to adopt the Dodd-Frank Act—which tightened regulation throughout the financial system— and to punish the lenders and securitizers who had led the country into perdition. We see these consequences every day, with the left now demanding criminal actions against the banks that have already disgorged billions for their sins. In addition, according to the conventional narrative, the large number of subprime loans in the financial system were the result of predatory lending, stimulating support in the Obama administration and Congress for a Consumer Financial Protection Bureau (CFPB). Above all, there was no need to change government housing policies; these bore no responsibility for the crisis.
The results of this narrative were fully predictable. Subprime lenders are now advertising FHA loans for borrowers with FICO scores of 580 or below. The Qualified Mortgage (QM) rule adopted by the CFPB in January 2013 contained no underwriting standards at all, with the exception of a maximum debt-to-income ratio of 43 percent—and even that could be waived if the loan was sold to the GSEs or FHA. Mortgages with 580 FICO scores and 3 percent downpayments, which had default rates of more than 20 percent in the crisis, were permitted under the QM rule, as long as the lender could certify that the borrower could afford the mortgage at the time it was made.
The Johnson-Crapo bill recently passed by the Senate Banking Committee adopted the QM rule as the basic standard for mortgages. The legislation also contained provisions that offered strong incentives for lending to underserved communities, although it did not impose numerical affordable housing requirements like those imposed on the GSEs. Reportedly, however, it was opposed by the “progressives” on the committee because it did not go far enough in assuring a “duty to serve” these communities. The new director of the Federal Housing Finance Agency (FHFA), Mel Watt, has promised to ease underwriting standards and not reduce conforming loan limits.
Most remarkable of all, the six agencies that were designated by Dodd-Frank to develop a high quality mortgage that could be securitized without the need for risk retention by the securitization sponsor, simply gave up on the task last August and decided that the high quality mortgage would be the same as the QM. In doing so, the agencies—the Fed, OCC, FDIC, CFPB, FHFA and HUD—noted that 23 percent of the loans originated from 2005 to 2008, that met the standards of the QM rule, “experienced a spell of 90-day or more delinquency or a foreclosure by the end of 2012.” In other words, the agencies that are supposed to be the guardians of the U.S. financial system agreed that a mortgage underwriting system that allowed a 23 percent rate of default did not trouble them.
This could never have happened if the narrative about the financial crisis had properly located the problem in the reduction of mortgage underwriting standards brought on by the government’s housing policies and implemented largely through the affordable housing goals. In that case, the choice for policymakers would have been seen as a choice between increasing the availability of mortgage credit and the possibility of a return to the conditions that produced the financial crisis. Instead, the choice the agencies saw was much narrower, simply a choice between credit availability and mortgage defaults: “Academic research and the agencies’ own analyses indicate that credit history [FICO score] and the LTV ratio [downpayment] are significant factors in determining the probability of mortgage default. However, these additional credit overlays may have ramifications for the availability of credit that many commenters argued were not outweighed by the corresponding reductions in likelihood of default.”
That, in a nutshell, is why we are again on our way to a mortgage financing system that will one day bring on another financial crisis.
Peter J. Wallison, a co-director of the American Enterprise Institute’s program on financial policy studies, researches banking, insurance, and securities regulation. As general counsel of the U.S. Treasury Department, he had a significant role in the development of the Reagan administration’s proposals for the deregulation of the financial services industry. He also served as White House counsel to President Ronald Reagan and is the author of several books, including Ronald Reagan: The Power of Conviction and the Success of His Presidency (Westview Press, 2002). He also writes for AEI’s Financial Services Outlook series.