Facts are still scarce when it comes to yesterday’s reported loss of US$2 billion from the “delta-one” desk at UBS. That hasn't stopped commentators jumping to conclusions, particularly when the three letters "ETF" are involved. But what do we really know?
Initial evidence points at an unauthorised FX transaction
Several news outlets (see, for example, Bloomberg) have pointed out that Kweku Adoboli, the UBS trader arrested yesterday by City of London police, changed his Facebook status to “need a miracle” on September 6. Earlier that very same day, the Euro/Swiss franc rate had jumped by nearly 10% in a matter of minutes.
Was this a coincidence? It’s hard to think of many other market moves in the last two weeks that could have triggered such a huge loss. And the period of a week or so between the actual loss being incurred and the trade’s exposure feels about right. UBS will almost certainly have been aware for several days that something was wrong before calling in the police.
Bank risk controls failed catastrophically
It seems obvious to state that internal risk controls failed at UBS, but even so it’s difficult to fathom how such a potential loss could go unnoticed. Blogger and former trader Kid Dynamite observes that it takes some doing for a single person to lose US$2 billion. If the EUR/CHF foreign exhange rate was indeed the immediate cause of the loss then, even if the bank entered and exited its positions at the worst possible time, it must have had a notional exposure to the currency of at least US$20 billion.
How might any trader be able to enter into such a risk position without anyone raising the alarm?
One common theme between Adoboli (who, I hasten to add, is still only a suspect) and earlier rogue traders (Leeson, Kerviel) is clear—all came from a background in settlements and therefore knew their way around their banks’ computer systems.
Bank incompetence has also historically played a key role in creating the environment for big trading losses. Nick Leeson, a back office employee at Barings, used an “error account” to book unauthorised trades and was not spotted by internal auditors for years. At Société Générale, Jérome Kerviel was promoted from a job in operations to the role of trading floor assistant, then reportedly given a small trading account of his own to buy and sell warrants on German shares. The bank’s primary objective in doing this was apparently to entice external investors by boosting the screen-based trading activity in the warrants. In hindsight, this was the opening the trader needed to incur huge risks later on. Kerviel also reportedly avoided the sack when earlier unauthorised trades that were uncovered turned out to have made money for the bank.
It’s too early to blame ETFs
We’ve already seen a renewed bout of moralising about ETFs being to blame for the reported losses at UBS, with further suggestions that ETFs, like CDOs in 2007/08, risk bringing down the financial system.
“Delta-one” desks, like the one Adoboli worked on, trade in a whole list of index-related financial instruments, from futures to equity swaps, equity baskets (or programme trades) and ETFs. Those equity baskets are then often split into and traded as separate price and dividend components, creating more complex exposures (tax-driven dividend arbitrage is reportedly a key contributor to delta-one desks’ profitability in Europe). Delta-one desks also work closely with their parent banks’ hedge fund servicing units and with colleagues in equity-based secured financing. Fundamentally, though, delta-one is supposed to be a high-volume, low margin line of business for investment banks, with exposures being hedged and money being made on financing rates or things like a pick-up in effective dividend rates.
If an ETF-related trade was indeed responsible for the UBS loss then, first, a hedge was missing somewhere and, second, given the need for a very high notional exposure to incur a US$2 billion shortfall, very few ETFs could have fitted the bill. Only ETFs tracking the DAX or Euro Stoxx 50 have generated sufficient trading in the last couple of months to be even theoretically responsible for the UBS loss, one trader suggested to me this morning.
An unhedged over-the-counter equity derivatives trade, possibly also in conjunction with an FX position, is most likely to have caused the UBS loss, the trader surmised.
All this doesn’t mean there’s nothing to worry about when it comes to ETFs. Inconsistencies abound in the way exchange-traded funds are created, traded and settled in Europe. From widely differing product structures, multiple exchange listings, fragmented liquidity and inadequate trade reporting to a highly complex clearing and settlement infrastructure, there’s certainly a great deal of opacity in European ETFs. And such opacity creates plenty of opportunities for excess profits to be made by insiders.
If it does turn out that UBS’s losses were primarily as a result of someone exploiting these specific ETF market inefficiencies, then those painting exchange-traded funds as a rogue financial product will have a point. But, for the time being, other derivative instruments appear equally, if not more likely to have been culpable. In the absence of further evidence to support their claims, ETF critics are basically putting forward an argument similar to blaming futures and options, rather than poor internal controls, for the downfall of Barings in 1995.
Some standardisation of ETF product design, trading and settlement is long overdue, for sure. So is the requirement to trade derivatives on exchanges, using central counterparties. And so is the key reform that would realign incentives across the financial markets and act as a brake on excessive risk-taking.
The most obvious conclusion to be drawn from the UBS loss—one that several commentators have already made—is that taxpayers should no longer be bearing the risk of losses from such activities at banks. The existence of a government guarantee if things go wrong prevents poorly run firms from going out of business, as they should. If the latest rogue trading scandal hastens the removal of the “too big to fail” status from large financial institutions, then it will have done some good.
(Postscript, 17 September 2011. On the basis of the charges levelled yesterday against Adoboli and of further press reports, it appears that the UBS loss was caused by the trader apparently creating a series of fictitious hedges in the bank's internal accounting systems over up to three years. If correct, this case is highly reminiscent of the Kerviel fraud. Why UBS's control systems failed to pick up the fictitious trades remains to be seen. ETFs were apparently one of the instruments used by Adoboli, but it appears that it was the absence of offsetting hedges that caused UBS's losses. In summary, the main story here appears to be one of a serious management and internal audit failure at UBS.)