UBS fraud likely to reveal management failures, again

The recent rogue-trading scandal at UBS has again raised doubts about managerial competence and the efficacy of supervisory systems at investment banks.

Bonus-hungry traders operating in an environment where taking risk is encouraged and where there is enormous, competitive pressure between individuals and divisions to make profits, can too easily defeat restrictions if they have a mind to do so.

“These types of frauds are likely to continue to happen,” said Richard Abbey, senior managing director of financial investigations at Kroll, a leading global consultancy firm. “It is impossible to guarantee that someone committed to finding a way to breach systems and controls won’t find a way to do so.”

Kweku Adoboli was charged in London last week with causing a $2.3 billion loss at UBS. He worked on the Swiss bank’s ETF (exchange-traded fund) desk, part of its delta-one division, where he would have executed trades for clients while offsetting or hedging any additional risk. Delta-one desks earn fees by using the bank’s own capital to create exposure for customers and from innovative hedges.

But, during the summer, Adoboli allegedly started taking open positions in several major market indices, including the S&P 500, DAX and EuroStoxx, and then entered fictitious offsetting hedges, possibly for forward settlement, and somehow circumventing back office systems.

“At a minimum, there should be quite basic systems in place for managing the risk of trading positions,” Abbey told FinanceAsia in an interview. “These are strict limits on intra-day and overnight positions, monitored by sophisticated electronic and human supervisory functions; and checks on those positions and their validity made by the back office.”

Abbey knows what he’s talking about. He is head of Kroll’s London financial investigations practice and has examined instances of corporate fraud for the past 16 years. These include notable cases such as Barings Bank rogue trader Nick Leeson and Morgan Grenfell Asset Management’s gender-confused Peter Young in the 1990s, the collapse of Italian company Parmalat, the alleged misappropriation of $130 million in a complex commodity fraud in Asia by rogue traders and, recently, he led the forensic investigation into alleged irregularities at Glitnir Bank in Iceland prior to its failure in October 2008.

In the case of Adoboli, of course, there are still a host of unknowns. Was he trying to recoup losses or simply make big profits? How was the fraud actually discovered?

Perhaps most bewildering is why none of his colleagues suspected anything was wrong — unless, of course, they did? Adoboli worked on a five-person ETF desk, and anyone who has spent time battling it out on a trading floor will know how difficult it is to hide anything. The atmosphere is competitive, aggressive and noisy, with traders highly sensitive to vulnerabilities, weakness and, most of all, to BS. 

“It is still too early to tell,” said Abbey. “You have to assume that someone else saw Adoboli’s trades. Until we know how the fictitious trades were booked and whether somehow he managed to circumvent controls — possibly using knowledge he had gained from working in the back office — we won’t know whether he was acting alone although there is a high possibility he was.”

Jerome Kerviel, who was charged with losing €4.9 billion at Societe Generale in 2008, shared some similarities with Adoboli — he was also a delta-one ETF trader and started his career in his bank’s back office.

But, there is also the wider issue: “The question remains — despite the dramatic instances of fraud and risk management failures in the past — just how much effort is invested by senior management in understanding how banks’ traders actually make money,” said Abbey.

“Perhaps, if a trader is making profits there is still a tendency to avoid rocking the boat — when, in fact, that should be the time to make more detailed checks and ensure that his activities are legitimate. Arguably, there is too much reliance on electronic controls and too little on proactive human supervision. Real-time management reports generated by electronic systems are only as good as the data and information put into those systems. It is the familiar adage: garbage in, garbage out.”