This is a little off my normal beat, but last night just before I left my office I took one last quick look at Bloomberg, and was shocked to see that Société Générale had just announced that a rogue trader had knocked a $7.1 billion hole into its capital base. Ouch! This was an astounding loss, and the frantic attempts by Société Générale to close out its huge underwater positions may have actually added to the market panic early this week – one which saw the Fed suddenly cut rates by 75 bps.
The good news is that Société Générale seems to have responded very quickly to the problem. Apparently the fraudulent trades were first discovered on January 19, and within a week the bank had closed out all the positions as they understood them to be, which is not to say they have discovered all the potential losses – apparently the trader was fairly sophisticated about hiding trades.
Nonetheless this quick response stands in sharp contrast to the actions (at least insofar as we were able to interpret them) of the authorities here in China when rumors in October 2006 that a trader acting on behalf of the copper reserve fund in China had taken huge trading losses on short copper futures on LME (the initial reports, if I remember correctly, suggested losses of around $500 million). Rather than immediately close out the losing positions, it seemed that the authorities let them roll over continuously in the hopes that copper prices would soon drop and some of the unrealized losses would be reversed. That didn’t happen. Copper prices continued to rise, and it seems that the final losses may have been substantially higher. How much did the copper fund eventually lose on the position? Of course we don’t know – state secret.
Although it is hard to get details about the copper trade, it seems that not only did the authorities not immediately close out the position, they may have actually added to their short by selling physical copper out of their reserves in China in an effort to coax spot prices down (and, perhaps, effectively to transfer realized losses from the very visible LME account to a much less visible reserve fund). In so doing they made two classic risk management mistakes. First, they made an unwarranted distinction between unrealized and realized losses, wildly hoping that if they did not realize the losses, the losses weren’t real, and if they held on long enough they could even reverse them. Second, they increased their position in an attempt to gamble their way out of the initial loss. These are how bad trading losses become trading disasters, and they are classic mistakes of inexperienced traders and risk managers.
It seems that Société Générale understood this and moved quickly to close out the losing position. Of course I don’t want to praise them too much, given that unauthorized and/or fraudulent trading losses should be discoverable long before they reach the multi-billion dollar mark, still, once they discovered the problem, they moved quickly to limit the loss, unlike the Chinese copper authorities who tried to gamble their way out of it. If you remember the China Aviation Oil disaster in Singapore the previous year you can pretty much guess that the same thing happened – small losses were hidden and attempts were made to gamble back into profitability.
I think there are two important lessons we should be reminded of from the Société Générale fiasco. The first is that when things go wrong in the financial markets there is often a tendency for a lot of other seemingly-unrelated things going wrong at the same time. Just as the French bank seemed to be getting its arms around its sub-prime-mortgage-related losses and the very difficult market conditions of the past few months, it got blindsided by this loss (and I think I remember that a few months ago another large French bank, Crédit Agricole, also reported several hundred million dollars in losses on unauthorized trades).
This kind of coincidence is usually not a coincidence. Warren Buffet famously said that it’s only when the tide goes out that you know who is swimming naked, and I suppose it is usually in a crisis that you discover that your risk systems weren’t as good as you thought. I discussed a somewhat-related topic in earlier blog entries, on the tendency of developing countries in general and China in particular to combine a number of pro-cyclical and self-reinforcing balance sheet mechanisms, so that problems in one part of the balance sheet bring out problems in another.
The second lesson is that a quick, and sometimes brutally firm, response to a financial disaster is, while often difficult to do, the best way to prevent it from getting completely out of hand. Rather than employ a gradualist approach, or take steps that require at least some good luck in order to be successful, a good risk manager often does best by stepping in quickly and closing down and cleaning up the mess. Among other things that means that there needs to be a clear incentive for managers not to postpone a resolution – and of course that requires a separation between the execution of a strategy and the risk management function.
Talking about fraud, according to an article in today’s South China Morning Post, China’s foreign exchange regulator, SAFE, could liberalize rules allowing firms to buy assets outside the mainland. Zou Lin, a director in the capital project department at SAFE, was quoted by Shanghai Securities News as saying that SAFE is planning to scrap background checks on companies investing overseas.
At first this seems like an obvious move. The PBoC is eager to get as much money as possible out of the country in order to help it regain control of monetary policy, and liberalizing the outward investment process is clearly aimed at accelerating cash outflows by making it easier for companies to make acquisitions or investments abroad.
But in fact China has long grappled with corruption, dirty money, and illegal money flows, and what little control the authorities have had is likely to be further eroded by throwing out background checks. Of course on the other hand China’s informal banking system has done a pretty good job, according to anecdotal evidence, of helping dirty money leave the country, so maybe these new measures will at least bring the illegal flows back into the formal banking system, where they can be measured and monitored.
