The housing crisis: What’s the Fed’s excuse?
Editor’s Note: As Fed chair Janet Yellen closes in on her first week in office, we return to examining the Fed. Last week, before Ben Bernanke handed over the reins, we asked Center for Economic and Policy Research co-director Dean Baker and former Fed vice chair Alan Blinder to weigh in on the departing chair’s legacy. In our Making Sense report (which you can watch below), Blinder – Bernanke’s old Princeton pal – gave him an A-. Conservative economist Charles Calomiris of Columbia agreed that Bernanke saved the day during the crisis. But on his record before the crisis – when, Calomiris contends, the bubble was obvious to a lot of people – he grades Bernanke a C-. That made us want to hear more about what Calomiris sees as the Fed’s complicity in the crisis. He and co-author Stephen Haber, of Stanford’s Hoover Institution, adapted the following post from their forthcoming book, “Fragile by Design: The Political Origins of Banking Crises and Scarce Credit.”
—Simone Pathe, Making Sense Editor
Most Americans know the Federal Reserve Board in its role as the maker of monetary policy. Through the discount rate that it sets on overnight loans to banks, and through its purchases of government bonds in the open market, the Fed plays a crucial role in influencing interest rates. In this task, the Fed is supposed to be an autonomous body that is independent of political influences.
There is evidence that the Fed has not always lived up to the standard of full independence when it comes to monetary policy, but there is at least a benchmark against which Congress and the public can judge the degree to which politics has intruded: the willingness of the Fed to permit an increase in inflation.
What most Americans do not know is that the Fed actually plays a much broader role in the economy, as a regulator and supervisor of banks. Among other things, it has the authority to decide whether banks may merge or acquire one another. In this role, the Fed is also supposed to make decisions autonomously, but there is no clear benchmark as to what would constitute independent, non-politicized behavior. Not surprisingly, there is abundant evidence that in this sphere of responsibility, politics have regularly intruded into Fed decision-making.
This is not a story of the Fed doing occasional favors for a few big banks, or succumbing to pressures from influential politicians to provide such favors. Instead, it is a story about a systematic arrangement — a political deal, if you will — that involved many parties and that played out over many years, in which the Fed used its merger approval authority to enforce that deal over and over again, and failed to use its regulatory authority to protect the financial system from the consequences of the deal, perhaps because doing so would have reduced the value of the deal to those who struck it. And, as we show in this article, that deal was a crucial factor in the creation of the subprime crisis of 2007-2009.
The Fed was at the center of an extraordinary bargain that was struck between merging megabanks, such as the Bank of America, and activist groups, such as the Association of Community Organizations for Reform Now (yes, that ACORN). That bargain produced several outcomes: the growth of megabanks that were too big to fail, the undermining of lending standards for home mortgages, and the creation of vehicles by which banks and America’s housing government service enterprises (GSEs) (Fannie Mae and Freddie Mac) were able to back those risky loans with paper thin levels of capital.
Banking crises of the magnitude of 2007-2009 do not happen without anyone seeing them coming, like a tsunami or mountain lion attack. Rather, they occur when banking systems are made vulnerable by their very construction. Two conditions have to be met in combination: there must be sufficient risk in the loans and other investments the banks are making, and there must be inadequate capital on bank balance sheets to absorb the losses associated with those risky loans and investments.
If a bank only makes solid loans to solid borrowers, there is little chance that its loan portfolio will all of a sudden become non-performing. If a bank makes riskier loans to less solid borrowers, but sets aside enough of its own money — extra shareholder capital — to cover the possibility that those loans will not be repaid, its shareholders will suffer a loss — but it will not become insolvent. These basic facts about banking crises are not a secret to bankers or government regulators; they are as old as black thread.
To understand how it was that America’s banks were stuffed full of risky loans and how they backed those loans with paper thin levels of capital, we have to go back to the early 1970s to trace the transformation of the U.S. banking system that began at that time, and to understand the economic and political opportunities that transformation created.
Circa 1970, it was illegal for banks to branch across state lines, and the vast majority of states (38 out of 50, to be exact) limited the ability of banks to open branches even within the state. Some states, such as Texas, outlawed branches entirely: all banks were single office, “unit banks.” In short, the megabanks that today have a branch on every corner and an ATM anywhere they can rent six square feet of space did not yet exist. The Bank of America existed as a legal entity, but it was a California-only bank, and circa 1970 it was actually shedding branches.
As the result of a variety of influences, the limits on bank branching became increasingly unsustainable in the 1980s. By the mid-1990s, subject to approval by regulators, American banks were free for the first time in their history to merge and branch wherever they liked. They had strong incentives for doing so: growing large could allow them to take advantage of economies of scale in administration, permit them to spread risk across regions, push back against big foreign banks that were grabbing U.S. market share, and perhaps become large enough to obtain too-big-to-fail (“TBTF”) protection from the U.S. government. As of 1984, they had reason to expect such TBTF protection; in that year, the government intervened to save Continental Illinois on the basis of its alleged “systemic importance.”
There was a crucial catch to the legal changes that permitted bank consolidation, however: acquisitions had to be approved by a number of regulators, most particularly the Fed. There were several criteria that the Fed was supposed to take into account, but as a practical matter there was only one that really mattered: whether the merging banks were good citizens as defined by the Community Reinvestment Act (CRA) of 1977.
The CRA was passed by Congress with good intentions — because in the context of America’s 1970s system of small, geographically segmented banks, the incentives of banks to look for strong borrowers in low income neighborhoods may have been weak — but the CRA had little impact on banking until the rapid-fire mega-mergers of the 1990s got underway. From 1977 to 1991, total CRA commitments by banks equaled only $8.8 billion.
In the context of the creation of mega-banks, however, this largely moribund piece of legislation became a powerful lever for activists to negotiate directed-credit deals with merging banks. Activist groups that one would have expected, on ideological grounds, to oppose the creation of too-big-to-fail banks, began to show up at Fed hearings to testify on behalf of merging banks.
For example, when NationsBank sought approval from the Federal Reserve Board to acquire the Bank of America in 1998, creating the largest bank in the United States, Bruce Marks, the executive director of the Neighborhood Assistance Corporation of America (NACA), who portrays himself as a populist defender of low income Americans against predatory bankers, sent a letter to the Fed stating that: “They [NationsBank and Bank of America] need to be applauded and supported. The regulators need to approve the application immediately and get back into the business of regulating the many discriminatory and predatory financial institutions that prey on working people.”
George Butts, the president of ACORN Housing Corporation, also sent a letter and offered testimony to the Fed in support of the merger.
NACA and ACORN’s support for the merger that allowed Bank of America to operate across the entire United States was not a one-off event. Six years later, when Bank of America acquired FleetBoston, Marks and Butts were there again, testifying in support of the merger. They were joined by Maude Hurd, the national president of ACORN, who echoed their strong support: “You probably know that when banks are merging, ACORN looks at the record and usually say [sic] the banks should do better and usually don’t support mergers. But ACORN and ACORN Housing have a long history of working with both Bank of America and Fleet. And I am here to tell you that we support this bank merger.”
Nor was it the case that ACORN’s support was confined solely to Bank of America mergers. When J.P. Morgan Chase acquired Bank One in 2004, creating an enterprise whose assets currently exceed $2.5 trillion, representatives from ACORN Housing testified on behalf of the merging banks.
Activist groups were supporting megamergers because the resulting megabanks were willing to share some of the economic benefits that they obtained from growing massively large with the activists. The banks were willing to do so because this was necessary under the Fed’s merger approval process.
Consider Hurd’s testimony in front of the Fed at its hearing about the Bank of America-FleetBoston merger: “Bank of America is our largest lender. This year the ACORN Housing Program had done over 4,850 mortgages with Bank of America, worth over $665 million dollars. Since 1991, we have done over 30,000 mortgages with Bank of America, and these mortgages are worth over $3 billion.”
Much the same was true of NACA. As Bruce Marks put it in his Fed testimony about the 2004 Bank of America-FleetBank merger, “So when they [Bank of America] say they’re going to commit $750 billion [in affordable housing lending] we take that to the bank.” As of 2012, NACA had obtained more than $13 billion in bank commitments for its loan programs via partnerships with Bank of America, Fleet Financial, Citibank, First Union Bank, NationsBank and others.
The terms of the mortgages granted to ACORN and NACA constituents provide clear evidence of an exchange of favors. To use NACA’s phrase, the mortgages its members received seemed “too good to be true.” In the case of NACA, borrowers received a one-size-fits-all mortgage with a fixed-rate, 30-year term, no down payment, no closing costs, no fees, no credit check, and no mortgage insurance premium. This is especially remarkable considering that NACA’s borrowers had low average credit ratings: 65 percent of NACA homeowners had a credit score that would categorize them as high-risk borrowers (a FICO score of less than 620), while nearly 50 percent had a score that would characterize them as very high-risk borrowers (a score of less than 580).
Mortgages directed through ACORN Housing had similarly generous terms. As Butts observed at the 1998 Fed hearing on the Bank of America-NationsBank merger, “Virtually all these loans [through a NationsBank-ACORN partnership] were to lower income households, with small down payments, with nontraditional credit, with cash on hand, and with older, urban housing stock.” He went on to note that borrowers in the program were allowed to qualify on the basis of “…food stamps as income, voluntary child support, cash on hand, or steady income rather than the same job for two years.” Banks also provided support to activist groups in the form of origination fees for administering the directed-credit programs, or philanthropic contributions, to those groups.
These arrangements with NACA and ACORN were just the tip of the iceberg. An accounting conducted by an activist umbrella group, the National Community Reinvestment Coalition (NCRC), estimated that America’s banks contractually committed $858 billion in 187 agreements with activist groups between 1992 and 2007.
As large as this number is, it represents a lower bound estimate of the total amount of credit directed through activist groups. The NCRC data indicate that banks committed an additional $3.7 trillion over that same period in “voluntary” lending programs to low income or urban homeowners, and a comparison of that data and public statements by activist groups suggests that some of those funds were channeled through the activists.
For example, the 1998 merger between Bank of America and NationsBank resulted in a $350 billion community lending commitment by the banks that the NCRC codes as “voluntary,” but NACA’s website states that Bank of America “looked to NACA to place $3 billion of those loans.” Similarly, the 2004 merger of Bank of America and FleetBank occasioned a $750 billion lending commitment that the NCRC codes as “voluntary, upon merger,” but NACA’s website states that $3 billion of these funds were directed through NACA, and that “Ken Lewis [Bank of America CEO] announced this huge commitment at NACA’s headquarters.”
Not only did the Fed bless these arrangements, but along with other regulators, the Fed gave very favorable treatment to risky mortgages and the securities backed by them when measuring the amount of capital banks had to maintain as a cushion against those risks. Such concessions, however, had their limits: banks would be willing to originate more risky loans if they could find a way to sell some of those risky loans. But Fannie Mae and Freddie Mac, the mortgage purchasing giants, were reluctant to purchase these loans precisely because they realized that these loans were risky.
Activist groups found a way to solve that problem: they enlisted political allies in the House and Senate to force Fannie and Freddie to buy bank mortgages generated through CRA commitments.
Sen. Alan Dixon, D-Ill., convened a Senate Banking Committee hearing in 1991 in which activists testified about the need for Fannie and Freddie’s underwriting standards to be relaxed: “It is ACORN’s observation that the underwriting criteria employed by Fannie Mae and Freddie Mac have been developed not for the general mortgage market, which includes low and moderate income homeowners, but for a middle income and substantially suburban mortgage market. As a result, it is our firm belief that the underwriting standards dictated by the secondary mortgage market are, at a minimum, income discriminatory and may, by extension, be racially discriminatory.” (Italics are ours.)
Fannie and Freddie resisted, but ultimately they were also given an inducement to participate in the deal: in return for agreeing to lower their underwriting standards in order to purchase CRA-commitment loans, they obtained the right to back their mortgage portfolios with paper-thin levels of capital, and be regulated by an office within the Office of Housing and Urban Development (HUD) that had little power and no experience in prudential regulation. And, of course, their debts were implicitly guaranteed by the U.S. Treasury, which is to say, by taxpayers.
The resulting deal was embodied in the Federal Housing Enterprise Safety and Soundness Act (usually called the GSE Act), signed into law by President George H. W. Bush just prior to the 1992 election. The election of Bill Clinton in 1992 allowed activist groups to apply vastly more pressure on merging banks and Fannie and Freddie. As President Clinton said in a 1999 speech, the banking-reform legislation of that year “establishes the principles that, as we expand the powers of banks, we will expand the reach of the [Community Reinvestment] Act.” Affordable lending mandates for Fannie and Freddie were increased throughout the Clinton administration.
When George W. Bush came into office in 2001, he further increased the affordable-housing mandates that the Clinton Administration had imposed on Fannie and Freddie. By 2008, roughly 80 percent of Fannie or Freddie mortgage purchases in the secondary market were in one of the mandated categories.
Meeting these ever-escalating targets became increasingly difficult. A first, crucial step in GSE relaxation of underwriting standards to meet HUD targets was a 1994 decision to purchase mortgages with loan-to-value ratios of 97 percent (a 3 percent down payment). The increasing tolerance for low down payments was not enough. Fannie and Freddie also had to buy increasing numbers of loans to borrowers with weak credit reports. They also aggressively moved into the markets for adjustable-rate mortgage loans (ARMs) and interest-only loans. Even these steps were not sufficient. In 2004, Fannie and Freddie removed the limits that they had earlier placed on their purchases of so-called no-docs loans (loans in which the applicant simply stated his or her income and employment history, with no independent documentation).
How did these arrangements give rise to the subprime crisis? Most importantly, although the mandates driving the debasement of GSE underwriting were intended to be selective, in meeting those mandates, Fannie and Freddie found it more politically convenient to relax standards for everyone –inner city dwellers seeking to purchase a modest home, as well as suburbanites who aspired to trade up to a McMansion. Curiously, this is a point that has been missed by many academics who argue that the Community Reinvestment Act played no role in the subprime crisis because they detect no difference in default rates between loans made to satisfy the CRA and other loans made at the same time. Whether or not that claim is true — and it’s been hotly contested — is beside the point.
Community Reinvestment Act loans and the mandates imposed on Fannie and Freddie set in motion a process in which mortgage underwriting standards became debased for everyone. When Fannie and Freddie — who, by virtue of their size and special status set the standards for the entire mortgage industry — agreed to purchase loans that required only a 3 percent down payment, no documentation of income or employment, and a far-from-perfect credit score, they affected the borrowing terms of millions of American families, not just the urban poor.
The increasing riskiness of mortgages, however, is not the whole story. Had banks and the GSEs backed these loans with adequate capital, high-risk mortgages would have caused shareholders to lose money, but that would not have threatened the entire financial system. But the banks, and Fannie and Freddie, were not forced to maintain adequate capital. The same political forces that pushed for the explosion of high-risk mortgages also pushed regulators to understate mortgage risk, and to permit banks and GSEs to maintain paper-thin capital buffers.
How much of this was the fault of the Fed, through its willingness to broker the political deals between the merging mega banks and the activist groups? None of the sad history of massive increases in risky mortgage origination and paper thin bank and GSE capital requirements would have been possible without the Fed’s willing participation. Yet scapegoating the Fed misses a deeper point: Congress and the administration, over many years and under both political parties, put the Fed in the position of deal-broker to lend legitimacy to these deals, knowing that the Fed had little real power to oppose them. The Fed is a product of legislation and is subject to payback if it does not do the bidding of powerful politicians. The deeper problem is the structure of our democracy, which gives rise to destructive deals that make the banking system fragile by design.
Making Sense interviewed Alan Blinder and Charles Calomiris about Bernanke’s tenure at the Fed.