The meltdown explanation that melts away
Although our understanding of what instigated the 2008 global financial crisis remains at best incomplete, there are a few widely agreed upon contributing factors. One of them is a 2004 rule change by the U.S. Securities and Exchange Commission that allowed investment banks to load up on leverage.
This disastrous decision has been cited by a host of prominent economists, including Princeton professor and former Federal Reserve Vice- Chairman Alan Blinder and Nobel laureate Joseph Stiglitz. It has even been immortalized in Hollywood, figuring into the dark financial narrative that propelled the Academy Award-winning film Inside Job.
As Blinder explained in a Jan. 24, 2009 New York Times op-ed piece, one of what he listed as six fundamental errors that led to the crisis came “when the SEC let securities firms increase their leverage sharply.” He continued: “Before then, leverage of 12 to 1 was typical; afterward, it shot up to more like 33 to 1. What were the SEC and the heads of the firms thinking?”
More recently, Simon Johnson, a former chief economist at the IMF, said last November that the decision “by the Bush administration, by the SEC to allow investment banks to massively increase their leverage … in terms of the big mistakes in financial history, that’s got to be in the top 10.”
It is certainly true that leverage at the investment banks zoomed between 2004 and 2007, before the near collapse. And this narrative of the rule change has plenty of appeal — it serves up villains. Stupid SEC people! Greedy bankers! It also suggests regulators were in the pockets of the big banks, and it offers support for the narrative of financial deregulation that many put at the center of the crisis.
There’s just one problem with this story line: It’s not true. Nor is it hard to prove that. Look at the historical leverage of the big five investment banks — Bear Stearns, Lehman Brothers, Merrill Lynch, Goldman Sachs and Morgan Stanley. The Government Accountability Office did just this in a July 2009 report and noted that three of the five firms had leverage ratios of 28 to 1 or greater at fiscal year-end 1998, which not only is a lot higher than 12 to 1 but also was higher than their leverage ratios at the end of 2006. So if leverage was higher before the rule change than it ever was afterward, how could the 2004 rule change have resulted in previously impermissible leverage?
The blame-the-2004-rule position made its first appearance in August 2008, when a former director of the SEC’s trading and markets division named Lee Pickard wrote an op-ed in the American Banker arguing that the SEC contributed to the crisis when it changed something known as the “net capital rule” in 2004. The net capital rule, which governs how much capital broker-dealers have to hold and how that capital is measured, is technical, but Pickard made it simple: Prior to 2004, the broker-dealers’ debt had been limited “to about 12 times its net capital,” but thanks to the change, the investment banks were now able to avoid “limitations on indebtedness.”
That October, the New York Times ran a front-page piece by Stephen Labaton entitled, “Agency’s ’04 Rule Let Banks Pile Up New Debt.” Labaton wrote that the big banks had made an “urgent plea” to the SEC that would exempt their brokerage units “from an old regulation that limited the amount of debt they could take on” and would “unshackle billions of dollars held in reserve as a cushion against losses.” This was essentially what the banks demanded in exchange for submitting their holding companies to oversight by the SEC. Previously, there had been no oversight, but the European Union was threatening to impose its own regulations unless the U.S. did so. With the loosened capital rules, billions could then “flow up to the parent company,” enabling increased leverage. Indeed, Labaton wrote, “Over the following months and years, each of the firms would take advantage of the looser rules. At Bear Stearns, the leverage ratio — a measure of how much the firm was borrowing compared with its total assets — rose sharply, to 33 to 1. In other words, for every dollar in equity, it had $33 of debt. The ratios at the other firms also rose significantly.” There were 157 comments on the Web version of this piece, most along the lines of remarks by a “Vietnam-era vet” who called the 2004 rule change “the financial equivalent of Patient Zero.”
On Jan. 3, 2009, at the annual meeting of the American Economic Association, Susan Woodward, a former SEC chief economist, highlighted the rule change in a presentation, a slide for which read: “2004 — SEC eliminated capital rules for investment banks” and “Average I-bank ratios of capital to assets: before 2004: 1 to 12. After 2004: 1 to 33.”
A number of prominent academics who were there went on to repeat a version of Woodward’s claim. They include Robert Hall, who was then the incoming president of the American Economic Association, Ken Rogoff, and Alan Blinder. In the highly regarded book This Time is Different, which Rogoff co-authored with Carmen Reinhart, the authors write that “huge regulatory mistakes” like the “2004 decision of the SEC to allow investment banks to triple their leverage ratios (that is, the ratio measuring the amount of risk to capital) appeared benign at the time.”
Other prominent people who have blamed the SEC’s 2004 rule change for the increase in leverage at the (former) big five investment banks include historian Niall Ferguson; Joseph Stiglitz, who wrote in his book Freefall that “in a controversial decision in April 2004, [the SEC] seems to have given them [the big investment banks] even more latitude, as some investment banks increased their leverage to 40 to 1”); and Nouriel Roubini, who with his co-authors wrote in their book Crisis Economics that “investment banks reacted to this  deregulation by massively increasing their leverage … to ratios of 20, 25 or even more…”
Thus did the “fact” become part of the conventional wisdom about the crisis.
Jacob Goldfield, a Harvard physics major turned Goldman Sachs partner (he left in 2000) noticed the claim that leverage had been limited to 12 before 2004, and then soared to 33. He thought it was strange that he hadn’t heard of this when it happened. He’d also noticed what he calls “quite a few” other pieces of conventional, but inaccurate, wisdom about the crisis, so he didn’t take for granted that this one was right. Instead, he checked. He looked at the 2003 leverage of two investment banks and found that it was much higher than 12. (In fact, there’s only one firm whose leverage in 2006 or 2007 was higher than it had ever been before 2004, and that’s Morgan Stanley. Nor was the leverage for the two firms that were hit the hardest by the crisis out of historical bounds when the world went to hell: Bear’s leverage (as measured by liabilities over equity capital) at year-end 2001 was 32, versus 32.5 at year-end 2007, Lehman’s at year-end 2001 was 28.3, versus 29.7 at year-end 2007.)
Halfway across the country, a semi-retired lawyer in Chicago named Bob Lockner began reading about the 2004 rule change, too. Lockner, who specialized in commercial bank capital markets activities, was suspicious because everyone kept citing the holding company leverage — but the rule applied only to the broker-dealer subsidiaries, and so didn’t include any international business, over-the-counter derivatives, or holdings of corporate or real estate loans, for instance. He also noticed that the rule hadn’t actually been implemented in 2004. The broker-dealer subsidiaries of Merrill and Goldman began using it in 2005, but the broker-dealer subs of Bear, Lehman and Morgan Stanley didn’t begin using the rule until their fiscal 2006 years. In other words, while leverage at the holding companies had started to climb in 2004 and 2005, the rule change clearly couldn’t be the reason.
After reading the Wikipedia entry for the “net capital rule,” which mostly cited the New York Times piece, Lockner decided to rewrite it, hoping that an accurate version would force people to acknowledge the old version was wrong. When I ask why he spent his time that way, he chuckles and says: “You mean, why am I insane?”
Lockner’s rewrite of the Wikipedia entry, at least as it existed on Mar. 13, 2012, describes in great technical detail the SEC’s net capital rule and the 2004 changes. But there are a few simple points. There was never any explicit leverage limit at the holding company level before or after the rule change. Even at the broker-dealer subsidiaries, a 12:1 limit didn’t exist. Smaller broker-dealers had an early warning at the 12:1 ratio, and an actual limit of 15:1 — but even these ratios didn’t exist in the way the economists seemed to interpret them, because they were calculated in a way that excluded big chunks of debt. In any event, since 1975, the broker-dealer subsidiaries of the big five investment banks had been using a different method, which had nothing to do with 12:1 or 15:1, to calculate their leverage limit. That method was unchanged in 2004. (Interestingly enough, the holding companies for the big investment banks might actually have made it under the 15:1 limit if you calculate the ratios by excluding the debt that the SEC does.)
There is another, more subtle point. The SEC did change the way the big broker-dealers calculated their net capital in 2004 in a way that could have allowed them to reduce the capital they held (by making it easier for them to meet the minimum requirements). This could indeed have had the effect of increasing the leverage, at least at the broker-dealer level. But it’s not axiomatic that that would result in higher leverage at the holding company — and it’s not even clear what the effect of the rule change was at the broker-dealer level.
One way the broker-dealers could have reduced capital would have been, as Labaton wrote, by paying big dividends to the holding company. But when the SEC changed the rule, it also put in place a new requirement that each of the broker-dealers have $5 billion in liquid capital before the major effects of the rule change; the SEC says that would have made it harder for the broker-dealers to pay out big dividends, and in fact, it required one firm to add capital to its broker-dealer. Overall, the SEC says that capital, as measured before most of the expected impact of the rule change, stayed stable or even increased after 2004. Several people at the broker-dealers at the time also tell me that the new rule was totally inconsequential in how they managed their capital levels
Skeptics may discount what the SEC and the broker-dealers say. But the data does show that over the years, the broker-dealers, contrary to perceptions that they must have wanted their capital to be as low as possible, actually kept much more capital on hand than they were required to hold. For example, in both 2006 and 2007, Bear Stearns had seven times the amount of capital that the SEC required, or more than $3 billion in excess net capital. This might suggest that the amount of capital the broker-dealers kept was boosted by factors other than the SEC’s requirements, like business needs, or rating agency and customer demands.
Various measures of leverage at the broker-dealer level do reach fresh peaks subsequent to the rule change. But it’s not obvious that that’s because of the rule change. The increases aren’t big in all cases, nor do they follow immediately after the firms implemented the new rules, and there’s enormous volatility from year to year, which may argue that the driving factor wasn’t the loosening of a preexisting constraint. Interestingly enough, one measure of leverage at the best-performing firm’s broker-dealer — Goldman Sachs — was higher before the rule change than at any other firm’s broker-dealer after the rule change.
The funny thing is that this is a mistake that no one has corrected. Although Erik Sirri, who was then the director of the SEC’s division of trading and markets, rebutted the claim in 2009, the New York Times didn’t cover it. Lockner says he wrote to a handful of economists; only Niall Ferguson responded and was chagrined to find out he was wrong. Of the people I cited earlier, only Blinder, Johnson, Kwak and Susan Woodward responded to my calls or emails. Blinder now says: “It’s true that very high leverage was a big source of the problem, but the net capital rule does not appear to have changed that much.” (The New York Times hasn’t issued a correction to his op-ed.) Woodward now says that while she doesn’t think the 2004 change is even on the top ten list of the most important contributors to the crisis, it doesn’t really matter, because “everyone agrees that too much leverage was a key cause.” Pickard, for his part, believes that the rule change hasn’t gotten enough blame yet, and he says that if leverage at the holding companies was higher in the 1990s, then the investment banks must have been playing games with their books.
More recently, Andrew Lo, the director of MIT’s Laboratory for Financial Engineering, wrote a paper analyzing 21 books on the financial crisis. In his paper, he pointed out the fallacy of blaming increased leverage on the 2004 rule change. Although Lo’s paper was picked up by the Economist, even that didn’t spur any of the academics who made the mistake to correct it. Why?
The best reason was voiced by James Kwak, who co-authored the book Thirteen Bankers with Simon Johnson. The book also links the increased leverage at the holding companies to the SEC’s rule change (although Johnson and Kwak never say the leverage was limited to 12:1 beforehand.) In a blog response to Lo’s paper, Kwak argues that Lo is making too big a deal out of this, because the rule change “very well might” have played a role in the increased leverage even if “we can’t tell how much.” In a conversation with me, Simon Johnson, Kwak’s co-author, argued that even if the broker-dealers kept excess capital on hand, well, they might have kept more excess capital had the rule change not occurred. This is indeed possible, although over time, the excess capital that the broker-dealers kept varied wildly, making it hard to see that they were targeting a specific amount of excess. In any event, Kwak and Johnson have a point: What happened at the broker-dealer level is murky and should be better understood. But the problem is that saying the rule change “might” have caused increased leverage just at the broker-dealer level is very different from saying it was an important cause of the crisis (especially since Lehman’s broker-dealer stayed solvent after its bankruptcy — it wasn’t the root of Lehman’s problems). At this point, the burden of proof should be on those who have claimed that the rule change was seminal.
Another reason that was suggested to me is that it’s politically incorrect to challenge the conventional wisdom about the rule change, because doing so might be construed as a defense of the SEC or the investment banks. That’s ridiculous: Facts are facts, and those who supposedly traffic in them should have respect for them. In addition, it’s far from a defense of the SEC to say that the rule change has been misrepresented. It may well have had pernicious effects that aren’t well understood yet. The broader context of the 2004 rule change was that, as Labaton pointed out, the SEC agreed to supervise the investment bank holding companies, and it clearly failed in those responsibilities. While the 2004 rule change offered a lovely explanation for that failure — blind incompetence is easily fixable — the real failings might be harder to fix, especially if we’re not looking for them.
A third reason I was given for why this mistake is no big deal is that because high leverage was surely to blame for the crisis, it’s beside the point whether the 2004 rule change made things worse or not. That’s silly — saying cancer killed the patient and saying the water he drank gave him the cancer are two very different claims. And it’s also dangerous, because if the rule change wasn’t behind the increased leverage at the investment banks, or the broker-dealers, then what was? If our goal is to prevent another crisis, isn’t it important to understand what actually happened? Or as Lo said to me: “If we haven’t captured the killer, then the real killer is still out there somewhere.”
PHOTOS: The U.S. Securities and Exchange Commission logo adorns an office door at the SEC headquarters in Washington, June 24, 2011. REUTERS/Jonathan Ernst; Princeton University Professor of Economics Alan Blinder speaks during a presentation at the American Economic Association Conference in Atlanta, January 3, 2010. REUTERS/Tami Chappell