.
.--.
Print this
:.--:
-
|select-------
-------------
-
WHAT REALLY HAPPENED AT UBS?

Former UBSers argue what's right and wrong in “Der Fall Der UBS.”

On January 30, 1998, when Union Bank of Switzerland announced losses of 350 million Swiss francs ($239,480,000) in its global equity derivatives group, rumors were rife and hard facts were scarce. Already in late 1997, however, a number of articles speculating on what had caused the equity derivatives losses that ultimately forced the merger of UBS with Swiss Bank Corp. had appeared.

“There was strong recognition of the problem—and there were plans to address it—but the effort was too little too late. While the bank was moving toward developing an independent structure, the GED was pushing ahead and making money.”

But the story of what actually happened remained tantalizingly out of reach. Now Dirk Schütz, deputy editor of Bilanz, a leading Swiss financial magazine, has written a detailed account of those events. “Der Fall Der UBS,” recently published in German by the Swiss publisher Bilanz, quickly became a best-seller in Switzerland. Rights to publish an English-language edition of the book are now being negotiated.

Although the book's focus is on the rise of Mathis Cabiallavetta, the CEO of UBS, two riveting chapters concentrate on the story of Ramy Goldstein and his contribution to the losses that led to the bank's merger with SBC.

Schütz's account tells how Goldstein, the brilliant and aggressive equity derivatives maven, set up a global equity derivatives group within UBS as a world unto itself—outside the bank's risk management controls. He earned enormous profits for the bank, and multimillion-dollar bonuses for himself. The book describes how UBS created two separate and overlapping risk control functions within UBS, both of which reported to the heads of business units and not to the bank's senior management. Goldstein's success and the inherent conflicts of interest within the bank insulated him from efforts to control his trading group. The risky trades and lack of controls caught up with him in 1997, however, when a series of losses in his department forced UBS into a merger with SBC.

But how accurate is Schütz's version of the events?

A number of former employees at UBS who read a translation of the book agreed to speak off-the-record for this article. Their responses fall into two groups. Former employees of Goldstein's global equity derivatives (GED) group argued that Schütz's main premise—that the losses were caused by faulty risk management controls—was incorrect. They asserted, moreover, that the risks Goldstein took on were well-known and fully authorized, and that the best risk systems would not have prevented the losses. (See box on page 24.)

Other former UBS employees who served as risk managers or who worked in the bank's fixed-income groups disagreed with several important assertions in the book, but uniformly said they believed that the two central theses of the book—the insularity of Goldstein's department and the failure of UBS's risk management effort—were right on target. According to these sources, the book is correct in ascribing the seeds of the problem to Goldstein's independence within the bank. But they added that many people at the bank actively voiced doubts about the conflicts of interest and overlapping responsibilities in UBS's risk management efforts.

“Many of us argued that this was crazy,” said one risk manager, who gave this impassioned analysis of the problem: “There were lots of meetings and issues raised about GED by controllers in London and Zurich, and lots of efforts to try to allow the infrastructure to catch up. But the people you had to talk to were in a conflict-of-interest role. Although they were sensitive to these issues and wanted good risk management, they did, as the book points out, give the business the benefit of the doubt. There wasn't a forum at the highest level of the bank where this issue could be debated, where Cabiallavetta or Werner Bonadurer [head of the group trading risk management division] had to defend in front of the chairman what was going on. There was strong recognition of the problem—and there were plans to address it—but the effort was too little too late. While the bank was moving toward developing an independent structure and struggling with how to get there, the GED was pushing ahead and making money.”

“If you correlate profitability with risk, they were commensurate. You can't have your cake and eat it too. Risk materializes when conditions go haywire.”

This source said, for example, that an effort to create a global value-at-risk system began in late 1994 and was largely completed by 1997. But while risk managers were given a clear mandate to integrate the bank's various systems, the equity derivatives exposures were the last to be brought in, because of technology challenges and broken promises of cooperation from Goldstein's group.

Most of the people interviewed for this article were convinced that the bank and its chief traders were well aware of the riskiness of the trades the bank was undertaking. Indeed, one passage in the book evoked snorts of disbelief. When describing the losses associated with Japanese convertible bonds, the book says, “Although Neil Thalheim and Ronny Apfel…no doubt understood the complexities of the convertibles, Goldstein and Bauer reportedly received only an oral synopsis from them.” “That's nonsense,” said one senior manager. “Goldstein knew everything verbatim.” According to another person, “Cabiallavetta and Hans-Peter Bauer [global head of fixed income, foreign exchange and derivatives] and the bank knew the risks they were taking. The deals were not that complicated relative to the rest of the derivatives world.”

Too risky?

The book's second principal argument is that Goldstein took on enormous risks that led to a series of losses that ultimately brought down the bank. Goldstein's former colleagues outside the GED differ on just how risky his operations were vis-à-vis the rest of the market.

“The positions we took were much bigger and the trades were more complicated,” said one senior manager. Another agreed that Goldstein's trades were way over the top. “If you correlate profitability with risk, they were commensurate,” said this former senior UBS official. “You can't have your cake and eat it too. Risk materializes when conditions go haywire.”

But others argued that Goldstein's trades were not much riskier than what other firms were doing at the time. Schütz, for example, says that UBS suffered more losses than other banks from U.K. tax law changes in 1997, because it wrote more long-dated options than its competitors. But others pointed out that the tax law changes had taken all banks by surprise and that losses weren't too far off the norm.

There is also some disagreement about the losses associated with long-dated equity options that UBS wrote on European equities. Schütz gives passing notice to a loss of 100 million Swiss francs from a model error in 1997. The losses from those instruments, however, might have been even larger. One source described them as “you choose” options that allowed investors to choose the higher of two European stock indices. Another described the long-dated options as “a trading instrument unique to UBS and one or two other players.” According to these sources, the modeling for these options fell apart when volatility went through the roof in the summer of 1997.

The cause of UBS's difficulties in this area appears to have been a relatively subtle problem in the analytics of long-dated basket options. According to one source, the models received a thorough review by Federico Degen's model-certification group, and then another examination by Werner Zimmerman's risk control group, which did a more detailed review of the code itself. But neither group, apparently, caught the problem.

“In some leading-edge theories, traders, quants and academics make refinements in the model after it's first made,” explained one source. “You can't really say the model was wrong. Black-Scholes-Merton went through many iterations, but you can't really argue that the first version was wrong. If [the model for long-dated options] were a completely bad mistake, it would have stuck out like a sore thumb, but this was a more subtle issue.”

“I don't want to give the impression that I think he was a crook. He did not violate his limits. But on the other hand, the limits were not in sync with what was going on, and nobody told him to stop.”

The book also argues that UBS's losses in Japanese convertibles were the final catastrophe that brought down the bank. Schütz believes that UBS took far more aggressive positions than other banks. While some sources agreed with Schütz's account, others noted, again, that UBS had been only a little more aggressive than others.

Another important argument involves the Next technology platform that Goldstein's group used—a platform that made integration efforts by risk managers difficult or impossible. One of Goldstein's defenders pointed out that Goldstein had insisted on controlling his own systems because he doubted that the bank had the technical sophistication to support his high-tech efforts. According to this account, in fact, Goldstein traded a guaranteed compensation for his right to make his own systems decisions.

The same technologist argued that in the early nineties, Next was the most sophisticated platform available and that the regional technology incompatibilities Schütz describes in the book were not unique. “Goldstein definitely used the Next system to scare people off,” admitted this source. “He'd say, ‘We have this very different system that you don't understand, and it would be disruptive for you to come in and have completely free run of the stuff.' That worked many times.” But after 1996, the source pointed out, risk controllers had terminals on their desks that gave them access to all of Goldstein's applications and deal data. How they ran their valuations after that point was up to them. (Goldstein used the Next system until the very end—and ironically, say sources, the new equity derivatives team at SBC is still using a version of Goldstein's system.)

Bigger picture

Those familiar with the Goldstein saga suggest that the story ultimately points to the limitations of the risk management function itself. “If the bank had the risk appetite, the tax law change would have clobbered you anyway,” said one source. “Risk management could have been a stronger voice in saying, ‘There is too much risk here. Is this really your risk preference?' But hindsight is 20/20. If you were afraid, at the time, that Japanese bank stocks would completely collapse to an absurdly low level, you wouldn't have found a lot of people who would have taken you seriously.”

“The risk manager's job is not an easy role,” this source added. “He or she has to call fire repeatedly and still be credible. Look at Long Term Capital Management. Are the managers all fools? No. It's easy to fault the risk management infrastructure, but it could be a function of risk appetite as well.”

For all their criticisms, none of Goldstein's former colleagues accuses him of fraud. “I don't want to give the impression that I think he was a crook,” said one source. “He was thoroughly professional. He was playing according to the rules. Factually, he's right: He did not violate his limits. But on the other hand, the limits were not in sync with what was going on, and nobody told him to stop. He was also aggressive in recognizing income on complex, long-term deals, where you could book profits on everything up front without putting up the proper reserves. Somebody should have stopped him.”

Many colleagues expressed feelings of betrayal and bitterness about Goldstein's personal conduct. “He was a shrewd operator,” said one source. “He tried to sell Apfel and Thalheim down the river. He'd sell everyone down the river.” “For all the loyalty shown to him, people did not get the sense that he showed loyalty in turn,” said another.

Despite this, however, former UBSers focus more of their ire on Mathis Cabiallavetta for agreeing to a merger rather than taking the losses on the chin. “The issue is really Cabiallavetta's lack of judgment or integrity, and his weakness in facing situations,” said one source. “So what if they lost a billion dollars. It's not a big deal for a bank that had $23 billion in capital. He destroyed a bank for peanuts.” According to another, “When you're a running bank, you should be able to stand in the wind and take the consequences. We played hard and we lost hard. But he sold the bank in order to cover his tracks.”

Was this information valuable?
Subscribe to Derivatives Strategy by clicking here!

--