Monday, January 28, 2008

Kerviel is just part of a rogues’ gallery

(Frank Partnoy in the FT) Until last week, many surveys ranked Société Générale as the top equity derivatives trading house in the world. The French bank’s reputation had soared since the early 1990s, when it began hosting an extravagant Super Bowl party to showcase its financial sophistication. Each January, SocGen’s traders converted a Manhattan auditorium into a massive trading pit and invited corporate executives to join them to trade derivatives based on the game’s score. The party symbolised the bank’s sharp and aggressive culture.

Last Thursday, SocGen announced it had been tackled by an expert in a different sport. Its press release warned it had “uncovered an exceptional fraud in a sub-section of its market activities”. As word spread that Jérôme Kerviel, a 31-year-old trader from the bank’s Delta One equity derivatives team had lost €4.9bn , the bank’s managers tried to assuage shareholders by announcing they would sue Mr Kerviel, a judo expert and junior trader whose total income last year, including bonus, was €100,000. Mr Kerviel’s assets are unlikely to cover the loss.

Mr Kerviel took SocGen to the mat with classic judo technique, using his employer’s strength against it. The size of his trading positions was in the range of €50bn, more than the big bank’s market capitalisation. SocGen defended the loss as rogue trading. The chairman described “a lone man who built a concealed enterprise within the company”. The media cited Nicholas Leeson’s $1.4bn derivatives loss at Barings in 1995. But the truth this time is more complex and troubling. Mr Kerviel is not the only controls problem at SocGen. Nor is SocGen the only bank with an exceptional fraud. Instead, his story is deeply connected to the numerous recent revelations of massive losses from derivatives at banks and an epic global risk management failure.

Although SocGen has emphasised that Mr Kerviel was responsible only for “plain vanilla futures hedging on European equity market indices”, the bank’s own documents suggest he was involved in more complicated strategies, including quantitative trading, swaps and equity derivatives arbitrage. SocGen’s initial investigation revealed that Mr Kerviel hid large speculative bets on European stock indices by entering false hedges into SocGen’s risk management system. Like most banks, SocGen assesses its traders’ risks on a net basis. Apparently, even a junior trader can buy €50bn of equity derivatives, as long as he also takes offsetting short positions.

This approach to risk management, focusing on net positions, is prone to abuse. Even traders who do not falsify their hedges can make massive bets appear to involve small risks. This is particularly true if risk management assumptions are based on historical data or do not account for worst-case scenarios. The problem extends beyond equity derivatives: few people are confident that banks have accurately assessed the risks associated with $45,000bn of credit default swaps, many of which are recorded as hedges.

Moreover, to be fair to Mr Kerviel, most of the losses were not his fault. His trades were down “only” about €1.5bn when SocGen discovered them 10 days ago. Then SocGen’s management lost an additional €3.4bn from abruptly unwinding those positions in a declining market.

In addition, the bad news from SocGen last Thursday was not limited to Mr Kerviel. The same press release that blamed “one trader” for a €4.9bn loss disclosed another risk management failure: write-downs of €2.05bn related to “unhedged super senior CDO [collateralised debt obligation] tranches” and other subprime-related derivatives. A proper ranking of the losses SocGen announced would go: first, trading losses by management; second, CDOs; third, Mr Kerviel. The leading rogue trader in history was a distant third on SocGen’s list of bad news that day.

Many other banks, including Merrill Lynch, Citigroup, Morgan Stanley and others have disclosed even bigger losses from subprime-related derivatives and CDOs. As was the case at SocGen, these banks’ risk management systems did not alert managers, directors, or shareholders to the risk that a handful of people could bring them to their knees. The fact that the SocGen scandal involves one person and a more brazen scheme makes it different in degree only, not in kind. When the bank’s 120,000 employees watch the Super Bowl this year, their only solace will be that they are not alone.

The writer is a law professor at University of San Diego and author of ‘The Match King: One Bullet and the Financial Scandal of the Century’, forthcoming later this year from Profile Books

No comments: