The human condition

This magazine has thrived because it has covered a critical debate in banking regulation: How can operational risk be measured - if at all? Matthew Crabbe, a former editor of Risk and now managing director of the risk management division of Incisive Media, looks at how the crisis of confidence and recent record-breaking financial frauds have taken that debate to another level altogether

In 1998, global regulators started an earnest public consultation on the revision of the Basel Accord on bank capital. This turned out to be a genuine debate. Much of what was proposed for Basel II was highly controversial, but the banking industry certainly had its chance to shape and to change it.

Remember, for example, the Basel Committee's proposal for a specific 15% charge for credit derivatives positions - the so-called W charge? No sooner had we poor hacks worked out what it was and how it might work, the W charge idea was dropped for fear that it might damage liquidity in the still-nascent credit derivatives business.

But the W charge proposal was only ever a side show. The Committee's proposal for a charge on operational risk, which it calculated to be some 20% of banking risk, sparked more criticism than anything else in the package. The International Swaps and Derivatives Association unsuccessfully fought to have the charge removed from the first draft of Basel II. The widely held criticism was that the regulators were simply looking for a catch-all category of risk that wasn't covered by their understanding of credit risk and market risk.

"How on earth do you calculate the risk of all the lights going out at the same time?" sniffed one senior investment banker to us at the time. The regulators themselves were clear that the key risk they wanted to capture with the charge was in the area of trades processing. But other risks fell into the operational risk category, such as fraud, that could not be quantified in the same way. So, as a result, the Basel Committee was from the outset open to the accusation that in the operational risk charge it had created the opportunity to make things up as it went along.

The regulators have seemed at times to encourage that view. Susan Schmidt Bies, a member of the board of governors of the US Federal Reserve System, told a regulators' conference back in 2005: "To address the difficulties presented by the very nature of the risk, the designers of operational risk measurement frameworks have had to be more innovative, take bigger steps into new territory, and be more willing to step away from traditional - and comfortably familiar - techniques than their counterparts in the market- and credit-risk arenas."

A decade on from the first consultation on Basel II, we're seeing a return to calls for slap-on, catch-all, capital charges and it's not a good environment for bankers to complain about any kind of regulation.

More interestingly, however, there is a reinvigorated debate about how human behaviour has contributed to the current financial crisis. It's a debate that possibly marks the start of a new epoch in risk management; one in which behavioural economics takes centre stage.

Animal Spirits, a new book by economics professors Georg Akerlof and Robert Shiller, invokes the phrase used by John Maynard Keynes to describe the emotional reactions of investors and other participants in the economy: reactions that exacerbate both downturns and up-turns.

According to Akerlof and Shiller, the reason the current crisis was not foreseen, and has not been fully understood, is because conventional economic theory does not take account of these animal spirits: "Conventional economic theories exclude the changing thought patterns and modes of doing business that bring on a crisis. They even exclude the loss of trust and confidence. They exclude the sense of fairness that inhibits the wage and price flexibility that could possibly stabilise an economy. They exclude the role of corruption and the sale of bad products in boom, and the role of their revelation when the bubbles burst."

There's little doubt that the sweeping changes in financial markets regulation that will be a consequence of the crisis will bring what we know as operational risk management to the fore.

Financial regulators have been telling audiences at our OpRisk & Compliance conferences over recent months that they want to investigate how bonus deals can be structured to avoid incentivising traders to take on too much risk.

And they have also been telling audiences that recent headline frauds have prompted them to review the systems banks have for controlling that specific risk.

Hard to believe it, with all that has happened since, but Jerome Kerviel's record-breaking EUR4.9 billion trading losses at Societe Generale were revealed only at the start of last year.

Kerviel's alleged fraud was apparently motivated simply by greed and the desire to look good. Like John Rusnak, the Allied Irish Bank foreign exchange trader whose fraudulent trades over a period of years up to 2001 cost his employer $692 million, Kerviel was apparently trying to boost his bonus payments by faking trading profits. This was facilitated by SG's short-sighted focus on market risk - so long as the trader's books appeared to balance, his line managers were satisfied. SG did not have adequate systems to check that the hedge trades Kerviel had placed were not fakes - and that's a basic operational risk failure.

All will be revealed during the course of Kerviel's trial, which is due to start early next year. Including of course, the degree of guilt that the French legal process will attach to Jerome Kerviel. And if not during the trial, then in the subsequent movie that is being made.

For operational risk managers at banks such as SG, the Kerviel affair raises far more complex questions than how the losses went unnoticed and who was directly responsible. Very soon after the losses were uncovered, a senior Societe Generale executive admitted in conversation that what Kerviel had done would be likely to prompt some fundamental rethinking of the bank's culture.

SG was a bank that had prided itself on its French identity. While its major rival, BNP Paribas, had moved its investment banking and trading hub to London years before, SG had kept its headquarters in Paris.

There were practical reasons for this. Salaries in Paris have been lower than in London (Kerviel was on a basic of EUR74,000). And one cannot blame SG's management for believing that many of its French staff would rather live in Paris than London, and that in Paris they would be less open to approaches from headhunters working for London-based rivals.

There was more of a familial atmosphere on SG's trading floors than at other investment banks. One could even argue that a little Gallic chauvinism gave SG an edge in marketing - set it apart from the Anglo-Saxon herd of competitors when it came to pitching products on the Continent (the largest market for its core equity derivatives products).

How Kerviel's losses on the Delta One derivatives desk were missed, how he was promoted to a position in which he could make them - that's the bread and butter of operational risk. But the cultural dimension to the Kerviel story - how his bank, for a number of superficially good reasons of personnel and broader business management, put itself in that position of profound vulnerability - is the most challenging one for risk managers.

We started trying to quantify the possibility of trade-processing failures. Right now, today, the focus is on disentangling the human reasons behind the collapse in confidence in world markets, and the reasons individuals and major firms alike have been tricked so easily by Kerviel, Madoff and Stanford.

There's no shortage of stories for OR&C to pursue as it marks this 10th anniversary.

Matthew Crabbe is managing director of the financial risk management division at Incisive Media, the company that publishes OpRisk & Compliance. He was the editor of Risk magazine between 1999 and 2001. Email: [e-mail].