Ethics for pleasure and profit
Say, here’s a good way to predict the US stock market. It has nothing to do with women’s hem-lines, baseball’s world series or the whereabouts of Louis Rukeyser, host of a popular American television show called “Wall Street Week”.
All eyes were on Alan Greenspan this October 15 as Wall Street skidded to the end of its worst week in a decade. The US Federal Reserve chairman had given an impeccably even-handed speech at dinner the previous night to a conference of bankers studying the ins and outs of risk measurement. Indeed, several of those present remarked that, after a day of debating the intricacies of fancy models, the chairman had scarcely said anything at all. It was another classic performance of mumbling from the Fed. Yet, the word went out, from that time and place, to sell. Asian markets dropped almost immediately, before the sell-off spread overnight to Europe. Then, pessimism turned into near panic when a spurt in the US producer price index greeted traders just before Wall Street opened on Friday morning. The Dow sank 266 for the day and 640 for the week. But everybody knew that the PPI number was a fluke.
So, why was the market in such a tizzy? Back at dinner, the fish was too old and the wine too young. It must have been the food. The real bombs had exploded earlier that Thursday, as prices were even rising a bit on Wall Street. John Reed, Citigroup’s chairman, told the meeting that either big banks like his used accounting which made it impossible for the market to evaluate risks or, worse, the market thought that even the banks weren’t able to evaluate them. Reed reminisced, for good measure, about last year’s bailout of Long-Term Capital Management – the hedge fund run by a group of Nobel prize-winning Wunderkinder. He recalled that people began phoning him over the weekend to ask whether he wanted to provide LTCM with a voluntary infusion of capital. His reply: “No, thank you. We don’t have any exposure. Sorry about that.” But then Reed noticed a funny thing. Citi was the lead bank for half the people – Goldman, Merrill and so forth – who were going to that meeting at the Fed. Citi, in fact, was in over its head along with every one of the firms which actually had direct exposure to LTCM. “This idea that you can look at only the first-order and not the second-order effects is simply wrong in my estimation,” Reed told the amused, but somewhat edgy group. Reed’s general rule of thumb is that any hit will be twice as big as the models predict. “You generally find out that your exposure is much higher than people had said when you were making the loans,” he warned.
Reed noted that there had been few and maybe no examples, in his experience, where an individual firm successfully avoided major risk issues that snared the industry as a whole. “That suggests to me,” he said, “that we tend to succumb to group thinking. Attitudes become communal. That’s a little scary because it says ‘how much can we really do when dealing with this subject?'”
Banks, however, can at least make sure that risk assessments are independent. That approach has been a tough sell for Reed following Citi’s merger with Travelers. But Reed now thinks that he has won the debate. “I don’t want the guys running the big trading portfolios to be the principal participants in deciding what appropriate trading risks should be,” he said. “They will inevitably tell you not to worry about this or that; there is no risk and we shouldn’t change what we are currently doing.”
Reed says that he’s still busy thrashing out the nitty gritty of capital adequacy with Sandy Weill and Citigroup’s board. “The real question that Sandy and I talk about,” he says, “is how frequently would we be willing to have zero earnings in a quarter. Do we want to do that every third year, every five years, once a year?” Citigroup should earn about $2 billion or $2.25 billion in a typical quarter, according to Reed. “You assess capital adequacy by asking how big a hit do you think you might have to take.” Reed then underscored the importance of being able to maintain the credit rating and the institution’s ability to operate effectively in the marketplace. Citibank suffered five downgrades in its credit rating during the 1990s, but subsequently recovered. “Anybody who thinks that, at the most senior level, we game the regulatory environment to see if we can operate with less capital than we need hasn’t been there,” he said. Reed thinks that how the financial industry deals with the internet may turn out to be one of the biggest risk categories. He reminded the group that risk assessment should not overlook the firm’s strategic position. A high cost-base can be particularly unhealthy. “You get yourself into a narrow compartment strategically,” he said, reflecting on some of his experiences at Citi. “You have no place to go. Your earnings start going and you begin to take risks you shouldn’t take.”
Leslie Rahl, president of Capital Market Risk Advisors in New York, told the group that too many people are becoming mesmerized with value-at-risk (VAR) and other quantitative techniques. Rahl agrees that these tools are valuable. But she thinks that we will look back on them 10 years from now with the same amusement which “state-of-the-art” approaches from the 1980s now inspire. Rahl pointed out that the quantitative part of risk management represents only about one third of a comprehensive risk management programme. “Senior managers – with practical wisdom – definitely need to get involved in helping to set the assumptions behind some of these complex models,” she said. For now, Rahl thinks that basic checks and balances, the operational controls, are key to managing risk. Rahl should know. Rahl’s firm has surveyed major financial institutions around the world about how the Russian crisis and the LTCM problem affected risk management. There has clearly been some progress. Rahl’s latest survey found that the greatest impact was the integration of market risk and credit risk functions. Only 9% of the firms had combined them before the crisis, but 64% have done that now. More firms are also focusing on documentation basis risk. That’s up from 60% before the crisis to 90% now. But Rahl is concerned that not enough institutions are stress-testing the correlations among risks. That proportion went up by only 9%, from 45% before the crisis to 54% now. Perhaps traders should dump some stock after a revelation like that. But to parse what really moves the market, tune in whenever Greenspan gives a dinner speech in Washington. Check the menu and test the food.
James Smalhout
