Ever since Nick Leeson spectacularly bankrupted Baring's Bank in 1995 as a result of $1.4bn losses on his unauthorised derivative trades there has been a steady stream of similar unauthorised trades and large losses. In 1996 Sumitomo lost $2.6bn on unauthorised copper trades, in 2006 Austria's BAWAG lost $2.4bn from unuthorised currency trading.
In September 2007 Credit Agricole reported a loss of €250m from unauthorised trades and in January this year Societe Generale lost €4.9bn as a result of €50bn unauthorised trades by Jerome Kerviel. The latest additions to this list came last week when Groupe Caisse D'Epargne, one of France's biggest mutual savings banks announced losses of €600m from unauthorised equity derivatives trading and Citic Pacific, based in Hong Kong announced losses of over $1bn from unauthorized currency trading.
It is tempting to say "serves them right" if the banks were foolish enough to allow a bunch of rogue traders to run up such massive losses but the huge irony in these losses, and the much larger recent losses from authorised trading in mortgage and credit derivatives is that all the major banks have in place "risk control" departments that are meant to monitor all positions and ensure that banks are not exposed to excessive levels of risk.
The clear implication of both the rogue traders and the large losses on authorised trades is the banks' risk control departments either do not know what they are doing or have no fool proof way of avoiding either systemic trading risk or rogue traders. Rogue traders capture attention when markets are normal but the systemic risk problem is clearly the more serious of the two, not least because of the huge sums of money involved in recent losses and write-downs. Most, if not all, the banks involved believed that the business and the trades they were engaged in were essentially sound and, if not risk-free, at least that the business risks were understood and controlled for. This confidence must now seen to be fundamentally misplaced.
Although they had sophisticated mathematical models in place which estimated the level of risk under a range of different possibilities, the range was clearly inadequate. What has happened is that the banks had models which allowed for defaults or losses of a specified level, say 10%, 15% or 20%, but failed to incorporate the possibility that the level of defaults could run to 40% or 50% and the losses to 70% or 80% or more. In the auction to unwind the credit default swap trades undertaken by Lehman Brothers, the average sum estimated to be recoverable from the trades by creditors was under 9% of the value of the swaps. Someone is going to have to pick up the other 90%+ of potential losses.
They may, of course, argue like John Meriwether and Long Term Capital Management, who lost $4.6bn in derivative trading in 1998 and nearly brought down the financial system, that the events which took place could not have been reasonably anticipated. In the case of LTCM it was an unexpected combination of Russian defaults and credit crisis. But this is precisely the problem. The models in place only allowed for a restricted range of foreseen risks. Indeed, it is doubtful whether they can allow for the range of possible outcomes. What they have done is to model the predictable, not the unpredictable, such as a major counter-party going under.
All this points in one direction. Most banks and trading houses are incapable of operating risk control systems of sufficient sophistication to minimise systemic trading risk. This is because the really severe risks are potentially unknowable and complex derivative trading simply magnifies the risks, creating what Warren Buffett perceptively termed "weapons of mass financial destruction". This suggests that if banks are unable to do it then risk controls will have to be put in place by national and international regulators, primarily in the form of limiting the degree and scale of risk that can be undertaken and the nature of the instruments that can be created and traded.
The mathematicians and rocket scientists in the banks have had their fun. The central banks are now picking up the pieces of their failed ventures. It is time to consider how to limit the risks of some forms of derivative trading. It may be that all trades should be notified to a central trading registry, some derivatives should be banned outright, and others should have to have a risk premium paid to the regulatory agencies as an insurance against potential default. What is certain is that risk assement is too important to be left to the market participants alone. We are living with the legacy of that failure.