Why did Lehman Brothers fail five years ago? Oh I don't know, but you know who'll tell you? Pretty much everyone on the Internet, this week! With the five-year anniversary of Lehman's bankruptcy coming up on Sunday, if you want to be told what went wrong at Lehman and What It All Means, you are in, like, certainly one of the top five weeks for that ever. So that is nice for you.

Today, though, you can read DealBook addressing a related topic, which is why regulators didn't sue or prosecute or jail or guillotine anyone at Lehman for blowing it up. The answer you might expect -- well, the answer I might expect, being me, is "because no one broke any laws." I just met you, and it's at least possible that you might expect something more like "because the system is rigged and evil banksters never get in trouble."* But those would have seemed like the leading contenders: "No one at Lehman broke laws" or "The corrupt and craven SEC isn't enforcing the laws Lehman broke."

But the interesting news in that DealBook article is that neither is quite right! Here is sort of what happened with Lehman:

  1. It placed its money in rather riskier and more illiquid places than its competitors.
  2. For that and other reasons, it lost the confidence of its funding sources.
  3. You don't last that long, as an investment bank, without the confidence of your funding sources.
  4. It didn't.
  5. Also, it did some shady accounting stuff.

Thing 1 is bad, but it's not really illegal to take risks, even dumb risks. Investment banking requires a lot of choices under uncertainty, and sometimes those choices are wrong; you can't really outlaw dumb. Things 2 through 4 were more or less acts of nature; you can't get mad at Lehman's executives for losing the confidence of the market.**

But the fact that blowing up a big business is not in itself illegal troubles people endlessly, even people at the Securities and Exchange Commission who should know better. Here is DealBook:

But Mary L. Schapiro, the S.E.C. chairwoman ... pushed George S. Canellos, who supervised the Lehman investigation as head of the S.E.C.'s New York office, to explain how executives who presided over the biggest bankruptcy in United States history could escape without a single civil charge.
"I don't get it," she said during a tense exchange with Mr. Canellos in her private conference room in Washington, according to the officials, who were not authorized to speak publicly. "Why is there no case? ... The world won't understand."

Now the penalty for presiding over a big bankruptcy is bankruptcy; it's not also separately illegal to go bankrupt. But, like she said: The world won't understand.

On the other hand, Lehman's massaging of its balance sheet really was pretty bad. The most notorious bit of massaging, and the apparent focus of the SEC's investigation, is Lehman's use of Repo 105. The quick version: In 2000, U.S. accounting authorities adopted a new rule about certain financing transactions, and Lehman quickly found a way to exploit it to make its balance sheet look less risky. Lehman discovered that it could do secured borrowing but characterize it as a sale, moving assets off its balance sheet to reduce the leverage that it reported to the market. In 2007 and 2008, as investors became more focused on investment bank leverage, Lehman did more and more of this, concealing its true financial position from investors. By the end, Lehman had hidden $49 billion in Repo 105 deals, reducing its reported net leverage from 13.9x (7.2 percent) to 12.1x (8.3 percent).

There is a decent argument that this was technically legal, but there is also a pretty good argument that it was not.***

The interesting thing is that the SEC's decision not to sue anyone doesn't seem to have been based on concluding that Repo 105 was legal . Rather it was based on the SEC's conclusion that:

  • Senior Lehman people didn't, like, tent their fingers and laugh diabolically about how they were pulling a fast one with Repo 105, but instead had only a dim understanding of what was going on, so it would be hard to prove that they meant to defraud anyone.
  • And, as DealBook puts it, prosecutors "discovered that Repo 105 had nothing to do with Lehman's failure" and the SEC "concluded that Repo 105 would not have been 'material' to investors because the firm's leverage ratio was trending downward regardless of Repo 105."

That first part is ... strange; there were a lot of people at Lehman running around freaking out about Repo 105 to anyone who would listen,**** so it's odd that the senior executives are off the hook because, basically, they wouldn't listen.

The second part -- the SEC's apparent conclusion that the Repo 105 shenanigans weren't actionable fraud because they didn't cause Lehman's collapse and weren't material to investors -- is more interesting. It's debatable, for one thing; Lehman's bankruptcy examiner certainly thought that Lehman's artificial deleveraging via Repo 105 was material.***** But you can understand the SEC's view: Lehman's problems were not really about its poorly disclosed shady financing of certain mostly low-risk and liquid investments.****** Lehman's problems were rather the traditional ones: short-term financing of long-term investments, which required Lehman to retain the confidence of the markets, coupled with bad decisions about those long-term investments that caused Lehman to lose that confidence.

It's a tempting story to say of Lehman that what went wrong is that they did some dodgy accounting to reduce their leverage ratio and hide their true financial condition from the public. But that's silly. What went wrong is that their business didn't work out, at a pretty unforgiving time, and their funding went away. Knowing that Lehman's "real" leverage was 14x, rather than 12x, wouldn't have helped you predict that. The problem wasn't leverage; it was liquidity.

You can disagree with the SEC's decision not to go after Lehman's gimmickry: For one thing, as my colleague Jonathan Weil points out, it's not particularly well explained; for another, it couldn't be that hard to convince a jury that Lehman was up to some naughty stuff. But I sort of like it. The way to prevent another Lehman is -- well, lots of people will tell you that this week too, why should I get ahead of them? But it's something like subjecting systemically important investment banks to banking-type regulation, or reforming short-term funding markets to reduce the risk of runs, or requiring banks to hold more capital and liquidity reserves, or making bank failures less disastrous through resolution planning or bank breakups. That is: It's something systemic, and directed to the actual business risks of the investment banking model.

Focusing on those sort of systemic issues -- as regulators have mostly done since 2008 -- makes sense. Pretending that the 2008 financial crisis was about a series of idiosyncratic technical accounting decisions obscures that focus. The SEC could have tried to shoehorn Lehman into a story of obscure accounting fraud, and it chose not to just because that's not the real story, and because the real story is more important. That seems like a good choice.

* But I will pay you the compliment of assuming that you don't actually go around using the word "bankster" in public. If you do, though, I mean, we have a comment section, and my understanding of the Internet is that blog comments are mostly for saying "bankster" a lot.

** You could, if you wanted, get mad at some of them for going perhaps a bit beyond the call of duty to try to retain that confidence, by getting on conference calls towards the end and saying "things are great!" when things were in fact objectively less than great. That's generally a no-no, though you have to cut them some slack: the only way to save the firm is to project confidence; if you get on a conference call and say "yeah, we're doomed," then you are.

*** I mean you don't really want to talk about Repo 105 down here do you? The argument that it was legal is that it seems to be allowed by the letter of SFAS 140, the applicable accounting standard, and that Lehman got a law firm to sign off on it, albeit a law firm in London because no U.S. firm would do it. The argument that it wasn't legal is that it was obviously intended to deceive investors and ratings agencies, and you're not supposed to do transactions just for their deceptive accounting effect even if they technically follow the letter of the rules. (A further argument is it isn't even that clear that it's allowed under the letter of the rules -- thus the law firms' disagreement.) The Valukas Report on Lehman -- incidentally, the week that was released was the single best week for being told what happened at Lehman; it's a good report, by Lehman's bankruptcy examiner; go read all 2,200 pages! -- is chock full of people at Lehman saying "oh yeah that was some shady deceptive stuff"; a sampling:

In an April 2008 e?mail asking if he was familiar with the use of Repo 105 transactions to reduce net balance sheet, Bart McDade, Lehman's former Head of Equities (2005-2008) and President and Chief Operating Officer(June-September 2008), replied: "I am very aware ... it is another drug we r on." A week earlier, McDade had recommended to Lehman's Executive Committee that the firm set a cap on the use of Repo 105 transactions. A senior member of Lehman's Finance Group considered Lehman's Repo 105 program to be balance sheet "window?dressing" that was "based on legal technicalities." Other former Lehman employees characterized Repo 105 transactions as an "accounting gimmick" and a "lazy way of managing the balance sheet."

Also the extent of the use of Repo 105 was undisclosed to investors, ratings agencies, the SEC, the Fed, etc. I dunno, I tend to have a pretty good sense of humor about these things but even I think the Repo 105 stuff looks pretty shady.

**** Led by controller Martin Kelly. From the Valukas Report again:

Lehman failed to disclose its Repo 105 practice even though Kelly believed "that the only purpose or motive for the transactions was reduction in balance sheet;" felt that "there was no substance to the transactions;" and expressed concerns with Lehman's Repo 105 program to two consecutive Lehman Chief Financial Officers -- Erin Callan and Ian Lowitt -- advising them that the lack of economic substance to Repo 105 transactions meant "reputational risk" to Lehman if the firm's use of the transactions became known to the public.

***** I told you, go read the whole Valukas Report, like 1,000 of its pages are on the materiality of Repo 105. (Actually just 30, pages 884-914.)

****** Repo 105 seems to have been mostly used for Treasury bonds and such.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Matthew S Levine at mlevine51@bloomberg.net