I don't see any bribes. Photographer: Suzanne Plunkett/Bloomberg News
I don't see any bribes. Photographer: Suzanne Plunkett/Bloomberg News

One thing that I used to do for a living was build derivative products. This, it turns out, is actually not much of a living; all the money is in selling the derivative products. So once I'd built a thing, I'd wrap it up in a pretty package, throw it in my bag, get on an airplane, and go see people to try to sell it to them. But spending your days building derivatives does not particularly suit you for human interaction, and showing up uninvited at someone's offices with a bag of derivatives is rarely a good way to ingratiate yourself.

So what I'd do is, I'd get on the airplane with a relationship banker. The relationship banker did not spend his days in the poorly-lit derivatives lab, tinkering with tax regulations and one-day lookbacks. The relationship banker would spend his days in socially acceptable pursuits, such as golfing and remembering the names of executives' children. (Also: lots of time on planes.) So when I'd arrive at a company with my bag of derivatives, the chief executive would greet us warmly, me and the relationship banker. "Hey John," the banker would say, "how are little Humphrey and Petunia? Petunia's off to Rice next year, right? Oh, by the way, I'd like you to meet Matt, he has a bag of derivatives."

Then off we'd got to a conference room, where I would dump my bag of derivatives on the chief executive. Then he'd say to me, "thanks Mark, lot of great ideas here, I will be sure to discuss them with my board," and would turn with relief to gossiping about his competitors with the relationship banker.

For reasons that are probably becoming apparent, I have repressed a lot of these memories, but they were dredged back up for me by the Libyan Investment Authority's lawsuit against Société Générale, which depending on whom you believe is either an extreme example of the corruption that can wrap itself in the cloak of banking as usual, or just banking as usual.

The story is that the old LIA of Muammar Qaddafi's government invested a lot of money with SocGen and lost most of it. The new, post-revolutionary LIA seems to invest mainly in animosity against global banks, which, bull market. So the new LIA is suing Goldman Sachs for bamboozling the old LIA, and now it's also suing SocGen for corrupting the old LIA.

The claim is that SocGen sold old-LIA a bunch of terrible derivatives -- which, sure1 -- by hiring a fixer company called Leinada Inc., which was owned by Walid Giahmi, a friend of the Qaddafi family. Leinada, according to the claim, didn't do much, but nonetheless got paid $58.5 million by SocGen for its services or, as the LIA's claim puts it, "services."

The LIA thinks that these services consisted mainly of bribing LIA representatives, or perhaps of bribing "members of the Gaddafi family and/or their representatives" to get them to "influence the LIA's decision to enter into the Disputed Trades, by making, or causing to be made, intimidatory threats to representatives of the LIA." It has no direct evidence for these claims -- that's why it's going to court -- but I guess there are some good circumstantial reasons to think that Leinada didn't do $58.5 million worth of derivative structuring:

(1) The opaque nature and inconsistent description of the services provided by Leinada, and the scale of its supposed remuneration. ...
(4) The fact that the SocGen Defendants had no need of the services allegedly provided by Leinada, but could themselves have structured and devised an appropriate investment solution for the LIA without the involvment of Leinada.
(5) The fact that Leinada was ultimately owned and/or controlled by Mr Giahmi, an individual who:
(i) had no discernible expertise in advising on or structuring financial derivative transactions; and/or
(ii) had connections both with the Gaddafi family and with representatives of the LIA, which he was in a position to exploit.

But all of those things would be equally true whether Leinada provided legitimate relationship-banking services or just a conduit for bribes and intimidation.2 SocGen could easily have structured and devised its derivatives without Leinada's help, but structuring and devising derivatives is worthless if you're not going to meetings with people who might want to buy them. Nobody thinks Leinada did $58.5 million worth of derivative structuring, but nobody really disputes that Leinada did at least $58.5 million worth of introducing. Introducing, as a business, is a high-margin business.

It's also a business that is notably prone to corruption. In particular, when banks operate in countries where they don't have deep roots, they tend to try to grow roots fast by hiring (as employees or consultants) people who already have strong relationships. And the dumb easy ways to do this -- hiring the children of the country's political elite, paying giant consulting fees to well-connected brokers -- seem to be pretty tempting despite their potential pitfalls.

SocGen thinks everything it did was above board:

Société Générale wishes to underline the fact that it works occasionally with financial intermediaries in countries where it does not have local teams in place. Such arrangements are systematically disclosed to the bank’s client, and are reviewed in detail by its compliance department.

The disclosure part is more or less true: The term sheets for these trades all disclosed that Leinada "has collaborated [with SocGen] and has been remunerated for structuring and advising this investment solution," which, fine, you could quibble with the word choice, but it's reasonably disclosey.

On the other hand, according to the LIA, at the time of these trades, "the LIA's board of directors had issued standing orders to the effect that no intermediaries should be used when the LIA entered into transactions with financial institutions, and that the transactions should include a written statement to that effect." These transactions included a written statement to the opposite effect, that there was an intermediary. The LIA did not follow its own board's rules.

And so now it's ... suing SocGen? It is a little hard to blame SocGen for not following the LIA's -- apparently undisclosed? -- internal rules, but that is life in the relationship business that is banking. Your client is an entity -- a company, a country, an investment authority -- but your relationships are with humans. Sometimes those humans do things that are not in the client entity's best interests, particularly in hindsight, and every so often that comes back to bite you. You don't have to go as far as Libya for examples; Royal Bank of Canada could be on the hook for hundreds of millions of dollars because it helped Rural/Metro's board of directors do stuff that a court later found was not in Rural/Metro's best interests.

Even if everything that LIA says is true, it's sort of an embarrassing lawsuit for them to bring: LIA is basically suing SocGen for LIA's own corruption.

On the other hand, even if everything SocGen says is true, it's still sort of embarrassing for the bank. SocGen was selling very silly derivatives3 on its own stock to the LIA. That sure sounds like a business built on principal-agent problems: You work your connections (or whatever!), you build a relationship with the humans at the LIA, and you ultimately use that relationship to sell them on a goofball trade that ... that might be more amusing to the people approving the trade than it is beneficial for the entity whose money is invested in it. If later humans, at a later incarnation of the LIA, don't feel bound by that relationship, and want to get out of those goofball trades, that can't come entirely as a surprise.

1 The Particulars of Claim come with a handy Schedule I describing all the trades, which you can read if, like me, you miss your days in the derivatives lab. Were there lookbacks? Hoo boy were there lookbacks:

A Look-Back Feature pursuant to which the initial SocGen share price for the purpose of determining the Final Return ("the Entry Price") was the lowest of a series of 130 "Initial Reference Values" which were each themselves averages of the SocGen share price over a 20-day period ("the 20-Day Averages"). However, the potential value of this Look-Back Feature to the LIA was limited by virtue of the fact that the Entry Price was (1) subject to a floor of 75% of the first of the 20-Day Averages; and (2) calculated by reference to 20-Day Averages rather than the lowest SocGen share price during the relevant period.

The question for you is: Did the LIA have the tools to price this trade correctly? And the answer is: Come on, obviously not. I'd say the probability that SocGen had the tools to price it correctly is only like 80 percent.

2 On the other hand, the LIA also lists "(7) The notorious fact that corruption pervaded both government and business activities in Libya, both prior to and throughout the period of time when the Disputed Trades were being executed."

3 I'm not going to tell you these were "complex derivatives," since I've got a big note taped to my computer saying "NEVER TYPE 'COMPLEX DERIVATIVES,' " but come on, nobody needs to set their initial price based on the lowest of 130 20-day averages. These thing were built to confuse.

To contact the writer of this article: Matt Levine at mlevine51@bloomberg.net.

To contact the editor responsible for this article: Tobin Harshaw at tharshaw@bloomberg.net.