well capitalized, co-chief executive officer John Reed vowed in a speech Thursday.
Recalling a close brush with insolvency in the early 1990s and the ensuing struggle to regain the market's confidence, Mr. Reed, who headed Citigroup predecessor Citicorp at the time, said he aims well beyond what regulatory minimums or risk management models would dictate.
"My general rule of thumb is it will be twice as big as any of the models say," Mr. Reed said at a conference on risk management sponsored by the Office of the Comptroller of the Currency.
"Our objective is to operate with more capital than we're going to need -- ever -- so that you never have to deal with the market's perception of not having enough capital," Mr. Reed said. "Let me tell you, I'm going to be long dead before we're ever undercapitalized again."
He said that when he and Sanford Weill, his co-chairman and co-CEO at the company that resulted from last year's Citicorp-Travelers Group merger, decide how much risk the $690 billion-asset Citigroup is willing to take, "the real question that (we) talk about is, 'Hey, how frequently would we be willing to have zero earnings in a quarter?' How big a hit could we take" without losing investors' confidence?
The sophisticated models and in-depth analyses of risk exposures all come down to one thing: capital adequacy.
"Necessary capital is that which is necessary to take a hit and still be able to operate effectively within the marketplace," Mr. Reed said, adding that the number is well above regulatory minimums.
The models often overlook what Mr. Reed called "second-order effects." As an example, he said that Citicorp had no direct exposure to Long-Term Capital Management, the hedge fund that flirted with failure a year ago. But Citibank was the primary lender to half the companies that came to Long-Term Capital's rescue.
"We were in over our heads with every one of the firms that did have direct exposure to Long-Term Capital," Mr. Reed said. "It would have been reverberated through the system in about two nanoseconds. None of these firms had the capital to sustain the losses."
Mr. Reed also noted that even with the risk management field's technological progress, few institutions manage to escape losses when big problems occur. "We all succumb to group thinking," he said. "That's a little scary."
Though the nuts and bolts of risk management are important, so too is an overall strategy.
"There have been more financial institutions ... that have disappeared by getting their strategy wrong than by being overexposed to risk," he said. "How you deal with the Internet is one of those strategic risks."
Citigroup also spends a significant amount of time looking at what is happening in the world, not just economically but politically as well, Mr. Reed said. It is also crucial that risks be assessed independently, he said.
"I have had some difficulty in our new merged situation making the point that I don't want the guys who are running the big trading portfolios to be the principal participants deciding what we think the appropriate trading risks should be," Mr. Reed said. "I think I've won the discussion, but it wasn't intuitively obvious to everyone."
Also speaking during the two-day OCC conference, Treasury Secretary Lawrence H. Summers said risk management has a particular urgency in the context of recent worldwide upheavals such as the Asian crisis and the near-failure of Long-Term Capital.
Mr. Summers said the frequency of such financial events appears to be increasing. Part of the problem, he said, lies in the inability of regulators to keep up with the pace of changes in the financial world.
"It is very difficult -- I think impossible -- to make the case that the progress in increasing stability through better regulation is even keeping up with the potential for instability created by innovation," he said.
Mr. Summers praised the capital proposal released by international regulators in June, commenting specifically on its intention to use market discipline to supplement regulatory efforts to measure credit risk. Creating greater transparency and allowing market forces to assess the credit risk of financial institutions, he said, "will increasingly be a center of any effort to maintain financial stability."
Future risk measurement efforts, he warned, must guard against the "false sense of security" engendered by some quantitative models. Risk managers should beware the tendency to forget that models typically assume a relatively stable economic environment, not the kind of financial crises that he sees as more common today.