One of the most seductive narratives about the recent financial crisis is that it was caused by dizzying increases in the amount of leverage on the balance sheets of Wall Street firms, leaving the financial system virtually no margin for error. Leverage, we’ve been told repeatedly, went from about 12-to-1 in 2004 to 33-to-1 in 2008. (Leverage is the ratio of debt or assets to equity; at 33-to-1 leverage, a mere 3 percent drop in the value of a firm’s assets can wipe out its equity.) The reason for the increase, so the story goes, was an underappreciated change, in April 2004, to an obscure Securities and Exchange Commission rule, which let Wall Street off its short leash and allowed unprecedented risk-taking. If not for that, according to the popular press and many accomplished scholars, the crisis might not have happened. The acceptance of this thesis has colored not only how we think about what happened but also the new laws that were designed to prevent the next crisis. The problem is, it’s flat wrong. And because we have misunderstood the facts, we may now be trying to cure the wrong disease.
The spread and evolution of the idea that the financial crisis was caused by a giant increase in leverage, enabled by the SEC, bears a passing resemblance to the old-fashioned, elementary-school game of telephone. While the change to the SEC’s so-called net-capital rule in 2004 was plenty esoteric, in the main, it did not allow big securities firms to take on more leverage. The SEC did two things in 2004: First, it assumed the added responsibility of regulating Wall Street’s larger holding companies—as opposed to just the broker-dealer subsidiaries within them. That’s where more and more funky and risky assets, such as derivatives and mortgages, had been housed over the years. Second, the SEC required the holding companies to report their capital adequacy in a way that was consistent with international standards, and to discount their assets for market, credit, and operational risks. Clearly, the SEC did a poor job of monitoring Wall Street once it obtained this increased regulatory authority. But the rule change increased rather than decreased the SEC’s oversight of the financial sector, and did not suddenly permit a dramatic increase in leverage.
Yet that’s not how the rule change got interpreted. In the aftermath of the collapse of Bear Stearns, in March 2008, people were eager to know how a company that had thrived for 85 years, and that had $18 billion in cash on its balance sheet, could evaporate in a week’s time. Enter Lee Pickard, a former director of the SEC’s trading-and-markets division and one of the architects of the net-capital rule in 1975. In an August 2008 essay in American Banker, Pickard lambasted the 2004 change, which he believed had allowed Bear Stearns to incur “high debt leverage” without “substantially increasing [its] capital base.” He argued that the original net-capital rule required securities firms to discount, or “haircut,” the value of their assets depending on the assets’ perceived risk, and that it limited the amount of debt they could incur “to about 12 times [their] net capital.” After the SEC’s 2004 rule change, he wrote, the large securities firms were permitted to avoid the haircuts and the limitations on indebtedness. According to Pickard, “The losses incurred by Bear Stearns and other large broker-dealers” were caused “by inadequate net capital and the lack of constraints on the incurring of debt.”
Pickard’s criticism appealed to journalists eager to understand the causes of the crisis. On September 18, 2008, The New York Sun ran an article summarizing Pickard’s assertions and quoted him as saying “The SEC modification in 2004 is the primary reason for all of the losses that have occurred.” The SEC’s trading-and-markets division tried to refute Pickard’s critique in a little-read appendix to a report issued on September 25 on the collapse of Bear Stearns. “[Pickard] says that broker-dealers were formerly subject to a leverage ratio limit of 12x net capital,” the commission wrote, but they “were not.”
Nonetheless, the idea kept picking up steam. At the end of September, TheNew York Times began a series about the causes of the financial crisis. One headline blared: “Agency’s ’04 Rule Let Banks Pile Up New Debt.” On December 5, the Columbia Law professor John Coffee added his imprimatur. In the New York Law Journal, Coffee wrote of the 2004 rule change, “The result was predictable: all five of these major investment banks increased their debt-to-equity leverage ratios significantly in the period following” the change. Around the same time, Coffee’s esteemed colleague, Joseph Stiglitz, writing in Vanity Fair, described five key “mistakes” that had helped cause the financial crisis. Sure enough, the 2004 rule change got prominent play. And for the first time, Stiglitz explicitly mentioned the extent to which the leverage ratios had increased—“from 12:1 to 30:1, or higher,” he wrote, allowing the banks to “buy more mortgage-backed securities, inflating the housing bubble in the process.” The idea that the SEC’s rule change had allowed leverage to balloon now had the backing of a Nobel Prize winner.
On January 3, 2009, Susan Woodward, who was the SEC’s chief economist from 1992 to 1995, spoke at the American Economic Association. Her slides repeated Pickard’s thesis and used the same numerical ratios that Stiglitz had used. One of her fellow panelists that day was Alan Blinder, a former vice chairman of the Board of Governors of the Federal Reserve System and an economics professor at Princeton.
Three weeks later, Blinder wrote an opinion column for TheNew York Times about the “six errors on the path to the financial crisis.” Error No. 2, Blinder wrote, was “sky high leverage” enabled by the 2004 rule change. He, too, noted how securities firms’ leverage had grown, this time to 33-to-1, from 12-to-1. “What were the S.E.C. and the heads of the firms thinking?” he wondered. Blinder’s column firmly established the conventional wisdom, which proved increasingly difficult to dislodge.
Both Stiglitz and Blinder were right to point out that Wall Street was highly leveraged before the crash, on the order of 33-to-1 or more. But the truth is that in recent decades, Wall Street firms have almost always been highly leveraged. For instance, according to a 1992 study by the U.S. General Accounting Office (now the Government Accountability Office), the average leverage ratio for the top 13 investment banks was 27-to-1 midway through 1991 (up from 18-to-1 in 1990). A subsequent GAO report, in 2009, noted that the big Wall Street investment banks had higher leverage in 1998 than in 2006. According to SEC filings, in 1998, the year before it went public, Goldman Sachs was leveraged at nearly 32-to-1, while in 2006 it was leveraged at 22-to-1. In 1998, Bear Stearns’s leverage was 35-to-1; in 2006, its leverage was 28-to-1. Similar patterns applied at Merrill Lynch and Lehman Brothers. To be sure, leverage has fluctuated over time: In the early 1970s, for instance, it was generally below 8-to-1. But in the 1950s, it sometimes exceeded 35-to-1.