Why I am personally responsible for Societe Generale’s $7 Billion dollar loss and what Societe Generale needs to do about it. I hate to admit it, but Societe Generale’s losses are all my fault. In 1983, I wrote Commodity Market Money Management, which introduced quantitative technique to what was then the derivatives industry. Judging by recent articles on Societe Generale’s disaster, they were using techniques similar to the ones I advocated, only radically more sophisticated, to manage their risks. I don’t suggest that Societe Generale’s management or traders read my writings or even know who I am. But their attitudes and techniques they used were ones I advocated. Attitudes and techniques, not incidentally, that have become standard operating procedures in the futures, derivatives and hedge fund industries. Quantitative techniques have several virtues. One is that these techniques are radically more precise than non-quantitative techniques. A second is that when quantitative techniques are used, character does not matter. As long as you wave your magic wand and chant the right invocation, the lead will turn into gold. It doesn’t matter what kind of person you are; whether you are good or bad, rich or poor, handsome or ugly, neo-conservative or sensible. Technique is all that matters. But the problem at Societe Generale was not that they used the wrong techniques, although sometimes, surely, they did, but that the techniques were not really used.
The proper techniques were not really used because Societe Generale had the wrong culture; it had what the New York Times calls a “culture of risk.” The New York Times quotes an anonymous trader as follows, “During appraisals by bosses, they made it clear you were judged by how well you did your basic job, as well as how much money you made on prop trades.” As presented, this idea is stupid. Not everyone can trade. Not everyone can learn to trade. Societe Generale had taken a sound idea and turned it cancerous. While not everyone can trade, everyone can innovate, at least, a little bit. Everyone can look at his or her job and say, “what can I do better?” In this sense, every business wants a culture of risk; innovation is always risky. More to the point, perhaps, while you want your traders to take risks, you want them to think about their trades, too. You want your traders to take only risks that make sense.
Unfortunately, this is not all of the work that must be done. No reasonable person expects traders to keep their risks under control. Some do, of course, but not all. For many of the good ones, it is not in their character. I do not want to suggest that traders as a group are irresponsible or dishonest. I have had traders over for dinner and I never once counted the silverware after they left. But good traders often live life on the edge. They see and take risks where you and I would not. And in so doing, they often take risks that they should not. As a consequence, as a group, traders need someone to keep them under control. They need a designated adult.
It was precisely here, that Societe Generale failed. As the New York Times noted, with traders making so much money, “they were untouchable; they had the power.” Part of the solution is programming and quantitative technique, of course. I have read and expect to read more articles about this or that flaw in the system. But judging on the basis of what I have read, the problems at Societe Generale were not with their systems, but with the people running their systems. Societe Generale had programs and technique by the ton. But programs and techniques are of no use unless someone enforces them. And the programs and techniques a person can enforce depend, in large part, upon his or her character. You and I are flawed vessels. We can do two or three or four things well and everything else poorly, if at all. Unfortunately, Societe Generale had the wrong people running risk management. Without doubt, the people Societe Generale had were intelligent, well-educated, dedicated and hard working. They were still the wrong people. Unfortunately, Societe Generale had Frank Burns (of MASH) running risk management, when they should have had a paranoid control freak, when they should have had Dick Cheney.
Consider, as evidence the following quote from Daniel Bouton, Societe Generale’s CEO. According to the New York Times, Mr. Bouton said, “…while our derivatives business was gong 130 miles an hour, risk control was only going 80.” But that just wasn’t true. The New York Times also notes, “It was not uncommon for traders to briefly exceed limits imposed on their trading before pulling back, despite controls meant to prohibit this.” This is not a subtle error. The first time this happens the CEO needs to call the directors of risk management into his office and ask exactly what went wrong and exactly what is necessary to prevent it from ever, ever happening again. The CEO needs to make sure the director of risk management knows that much more than his job is on the line. The second time this happens, the CEO needs to call the director of risk management into his office and place two or three warning shots between the director of risk management’s eyes. I recommend the .577 Nitro Express. It kicks a bit and the grip may retain fingerprints, so use with caution.
A paranoid control freak like Cheney would never allow traders to run the business. He would insist on establishing and enforcing control. He would divide and conquer, which in this instance means creating a risk management department as paranoid as he is. He and his department would have verified the data as entered, which Societe Generale’s system reportedly did not. And he would have instituted controls that were inherently difficult to subvert. The New York Times notes that, “Mr. Kerviel knew the timing of the nightly reconciliation of the day’s trades by Eliot, so he was able to expertly delete and then re-enter his unauthorized transactions without being caught.” Eliot was the name of the bank's proprietary system for booking trades. A paranoid would have randomized the times the reconciliations were done. If that was not technically possible, and it probably was not, he would have designed the system to flag trades that were not reconciled for special scrutiny.
Mr. Cheney is reportedly secretive, an empire builder and does not trust anyone else’s judgment. As far as managing risk is concerned, these are all character flaws. More likely than not, Mr. Cheney knows nothing whatever about managing risk. But a lack of knowledge is not as important as risk managers invariably make it seem. If a deep understanding of what your employees do is necessary to manage them, modern business would not be possible. Mr. Cheney is intelligent enough to learn and willing to learn and that is all that is necessary.
I have met many risk managers and, as a group, they are highly intelligent and they know an enormous amount about risk management. Among the risk managers I have met, I have never met a single paranoid control freak. I think that explains a lot about the recent failures in our industry. As an industry, we do not always hire the right people. We do not always consider character as seriously as we should. And that can sometimes lead to disaster.
|
P