Feb 11th 2010
From The Economist print edition
How banks can improve their approach to risk management
Illustration by Peter Schrank
THE annual gathering of the Global Association of Risk Professionals, held this week in New York, had as its theme “Transforming Risk in a New World Order”. Outwardly confident they may be, but those attending have been chastened by the experiences of the past two years. In the good times, it had seemed that threats to stability were being held in check by financial innovation and technological advances. It appeared that the quality of banks’ risk management had, like that of avionics, converged on a high standard, allowing banks to fly higher, more efficiently and more safely. In structured finance, however, giants such as AIG, Citigroup and UBS had been piling up “non-linear” bets, with a small—but, as it turned out, very real—likelihood of life-threatening losses. It was the equivalent of a plane that ekes out an extra few miles per tank of fuel but suffers engine failure in a storm.Banks are scrambling to convince markets, regulators and politicians that lessons have been learned. Risk is now the hottest recruitment area in finance, with some large firms, such as Morgan Stanley, doubling the size of their teams. Industry groups are pumping out self-critical reports and are working to bring centralised clearing to over-the-counter derivatives and improve underwriting and disclosure in securitisations. Pay is being aligned more closely with long-term performance, even if it still looks obscenely high to outsiders.
Yet there is scope for banks to do far more to improve their risk management. Some did better than others in coping with different aspects of the crisis, offering useful lessons. Admittedly, skill and luck are difficult to tell apart, and firms that were exemplary in some areas did badly in others. But several commonsense recommendations emerge from the crisis.
Consider risk governance: the relationships between traders, risk managers, executives and directors. The balance of power is now shifting back from risk-takers to those who police them internally. But a chief risk officer is impotent if he is nothing more than a glorified compliance chief, as is still the case at many banks. Ideally it should always be up to traders, not their risk managers, to prove their case if the two disagree. It helps if the latter command respect. Goldman Sachs, for instance, moves senior traders to risk positions, making clear that such moves are a step towards the top. Despite the firm’s swashbuckling reputation, its senior people have a conservative streak, which explains Goldman’s decision to pull back from mortgages in 2006 (though even its risk managers failed to spot the danger of being so heavily exposed to AIG).
Boards, too, need to monitor risk better. Many were not even asking the right questions—such as whether the boom in mortgage-backed markets was sustainable. A lack of relevant experience among independent directors did not help. The outsiders hired onto boards after the passage of the Sarbanes-Oxley Act in 2002 brought fresh perspectives, but could not always get their heads around collateralised debt obligations. As well as having a strong financial-services bent, boards should be pared down from the 12-16 members that are typical today, so they are less vulnerable to groupthink. Directors should also put in more time: five or six days a year is not enough to master a labyrinthine balance-sheet.
Ultimately, however, the tone on risk is set in the corner office. The bosses of banks that suffered devastating losses, such as UBS and Royal Bank of Scotland, were generally more interested in striking deals and leaping up league tables than in rooting out danger. By contrast, the top brass at banks that fared better, such as JPMorgan Chase, encouraged managers to flag up potential problems and eyed highly profitable units suspiciously. Jamie Dimon steered JPMorgan through the crisis by doggedly sticking to a few basic principles, such as not holding too much of anything. He also had a healthy scepticism of mathematical-risk models and metrics, even though his bank invented many of them. They seemed to work when asset prices were rising and volatility was low, but the meltdown exposed their inability to capture rare but catastrophic “black swan” events.
Geek drama
The answer is not to reject quantitative finance but to be honest about its limits. Models have their place, but they must be coupled with more subjective approaches to risk, such as stress tests and scenario-planning. Three years ago it might have seemed neurotic to fret about systemic liquidity shocks or the failure of a big investment bank. But the few firms that thought through the consequences of such events were better able to react when they occurred. Fixing finance will take more than sharper boards, greater scepticism towards “quants” and more powerful risk managers. But as the world awaits a regulatory panacea, those would be good places to start.
Copyright © 2010 The Economist Newspaper and The Economist Group. All rights reserved.
0 comments:
Publicar un comentario