It was a case of déjà vu all over again for anyone who remembered the Barings, Daiwa and NatWest Markets case. A lone trader – not a star performer, but a minor operative with a reliable but unglamorous track record – is suddenly discovered to have gone off the rails. The consequent losses are bigger than anyone had thought possible from a seemingly unexciting and straightforward business line and, in some cases, they are big enough to threaten the bank’s very existence.

Apart from the fact that the bank concerned faces no serious threat to its survival, the latest ‘rogue trader’ scandal at Allfirst, and its parent Allied Irish Bank (AIB), fits neatly into the established pattern. John Rusnak, a 37-year old currency trader who by all accounts seems to have lived an average, suburban American lifestyle, turns out to have lost $750 million of his employers’ money on unauthorised trading.

It may be too early to draw hard and fast conclusions about what exactly went wrong in the AIB/Allfirst case, since various internal and external investigations are still under way. It’s not even clear yet if alleged ‘rogue trader’ Rusnak had an accomplice – although given the nature of the alleged activities, it is hard to see how anyone could have managed them alone. But there are several implications that can be identified even at this early stage – not least, the striking parallels with the way in which similar scandals have unfolded.

The most obvious similarity concerns the future of AIB and its Allfirst subsidiary. Although the bank is not going to go bust (there are even suggestions that $750 million may be an overestimate of its total losses), there is now a strong possibility that AIB will not remain independent. Certainly, Allfirst’s degree of operational independence will be severely curtailed, although apparently a plan to centralise its treasury operations in Dublin had already been in place since last year. Heads are likely to roll at the top of Allfirst, and near the top of AIB.

While it may be too soon to say what caused the problems and who was to blame, it is fairly safe to conclude that there were serious flaws in the oversight and risk management systems at Allfirst. It’s also clear that the main reason that AIB will not share with Barings the dubious distinction of having been destroyed by a rogue trader is that it was strongly capitalised. In other words, it has been saved from calamity by strong capital reserves, not by good risk controls.

That conclusion may eventually be used as an argument for the concept of holding capital against operational risk, an idea that has proved controversial since it was proposed by the Basle Committee as part of its regulatory capital reform package. Opponents of this idea will continue to argue that the AIB case changes nothing. After all, the bank had not specifically set aside capital against the possibility of a rogue trader losing money on currency deals, and it could not possibly have foreseen exactly how much it would need to allocate. If Rusnak had lost 10 times this amount, the opponents of an op risk charge might ask, would that have been an argument for holding 10 times as much capital?

Whatever the merits of that debate, op risk is emerging as the key risk factor in this case. This is one of several parallels between the AIB losses and the disastrous losses generated by rogue traders at Barings and NatWest Markets (NWM). Another conclusion that may be drawn concerns the type of activity where the losses were made. Sometimes described as ‘risk-free arbitrage’, the type of trading in which Rusnak was supposed to be engaged might more accurately be termed ‘low-risk arbitrage’. It involves each transaction – in this case, on the forward foreign exchange markets – being hedged by equal and opposite trades, so that both the profits from the trade and the risks of loss are low. Money is made by exploiting small differences between the price of the different transactions, caused by imperfect market information or other discrepancies.

In theory, this type of business is low on market risk. If each transaction is hedged – using options, for example – then the market risk of any trade should be minimal. This is reassuring to those who run the individual business lines, to risk managers and, ultimately, to the board of directors. But every time a rogue trader comes along to disrupt this comfortable world-view, it becomes increasingly clear that such activity is actually not low risk. Arbitrage trading may be low in market risk when conducted properly, but it also involves a very high degree of operational risk – whether this takes the form of deliberate deception or simple mispricing of deals. When the two are combined, as they appear to have been at Allfirst, the losses from this op risk can be massive.

The key point here is that arbitrage strategies are high in op risk because they are low in market risk. To make money by exploiting small market discrepancies, you have to take huge positions, and therefore leverage your operational risk – which in most banks is not quantified or frequently reported.

The danger in this kind of seemingly low-risk business line lies partly in the complacency it engenders, and partly because it can be difficult for traders to confess to market losses in activities in which they are not supposed to be making losses. A truly ‘directional’ trading strategy – of the kind that might be employed by certain hedge funds, for example – is known to be risky, and trading losses are part of the realities of doing business for this kind of operation. In such a business, even a substantial loss does not undermine the purpose of the business. A proprietary trader who tells his or her boss about trading losses may not expect a pat on the back, but is unlikely to lose their job if they make profits more regularly than losses. It’s acceptable, at least internally, to talk about losses, examine why it happened, and move on from that particular strategy after having absorbed its lessons.

But in a business that is not supposed to take on such a level of risk, such an admission could well result in the whole point of the business being reassessed. On a ‘low-risk arbitrage’ trading desk (of the sort that Rusnak was employed on), sizeable losses are not really supposed to occur and, consequently, it’s much harder for traders to acknowledge them. The temptation is for traders who have made losses to cover them up then engage in a high-risk (but potentially high-reward) trading strategy to recoup the losses before anyone notices them.

This is what happened with Nick Leeson at Barings and Toshihide Iguchi at Daiwa Bank, and it appears to be what went wrong at Allfirst. As Iguchi, the man who lost $1.1 billion of Daiwa’s money by illicit trading in US Treasury bonds, told a journalist while serving his four-year prison sentence: “I think all traders have the tendency to fall into the same trap. You always have a way of recovering the loss. As long as that possibility is there, you either admit your loss and lose face and your job, or you wait a little – a month or two months, or however long it takes.”

It’s a dilemma that Rusnak would doubtless recognise, and one to which top management may need to give serious thought. Time and again, it has transpired that a seemingly low-risk activity, such as forex dealing by a relatively small desk at an overseas subsidiary, has contained massive unforeseen operational risks.

Given that history, now may be a good time to ask how much a bank’s board of directors actually understands about the hidden operational risks of supposedly low-risk business lines. Would AIB’s board have wanted to maintain this relatively small business line if it had known of the operational risks embedded in it? Presumably not. And are there other banks with business lines that contain similarly high levels of op risk, hidden in an activity that seems reassuringly low in market or credit risk? Almost certainly.

As we observed in last month’s Opinion column, even the best-run companies may have one or two business lines they don’t fully understand. If this lack of understanding extends to business lines where there is virtually no limit to the amount of money that can be lost, the very existence of the company is in jeopardy.