David A. Viniar – Goldman Sachs and Co. CFO – Statistical Data Included
THE ART OF BALANCING RISK AND REWARD AT GOLDMAN SACHS.
DAVID VINIAR SPENT THE FIRST day of his summer vacation dealing with a ruptured steam pipe at the Manhattan offices of Goldman Sachs & Co. As CFO of one of Wall Street’s most venerable investment banks, Viniar does not actually have “plumbing” listed in his job description. But he is responsible for risk, whether it comes in the form of widening credit spreads, capital-intensive block trades, or–as happened in August–an all-weekend scramble to make sure a faulty pipe didn’t interfere with Monday’s opening trades.
As Goldman’s top risk-management officer, of course, Viniar has few challenges that are quite as straightforward as a broken pipe. “Being CFO of a global securities firm is one of the toughest jobs out there,” notes analyst Mark Constant of Lehman Brothers. “The risks he is asked to manage are incredibly complex, and there are variables that are almost impossible to predict.”
Viniar became CFO in 1999, but he has seen his share of risks and variables since becoming a partner at the investment bank in 1992. There was the infamous Wall Street crisis in the fall of 1998. There was also the residual effect of having to postpone the company’s initial public offering from September 1998 to May 1999. More recently, questions regarding analyst independence have threatened Goldman’s venerable reputation as well as those of its sister firms.
To better manage such risks, Viniar has developed enhanced assessment models and instituted additional checks and balances during the past three years–efforts that earned the 45-year-old finance chief the 2001 CFO Excellence Award for Risk Management. And although Viniar dubs risk management a team effort, he is quick to note that the buck stops with him. “It is my responsibility to know what is going on everywhere in the firm,” he says. “If [CEO] Hank Paulson wants to know what our risk is in any area, he shouldn’t have to make 10 calls; he should call me.”
The Russian Bear
The trigger for many of the changes, of course, was the crisis of 1998. That summer, Russia’s devaluation of the ruble and the country’s subsequent debt crisis led to cascading liquidations in Western markets. Most infamously, Long-Term Capital Management, the supposedly invincible hedge fund, began a spectacular meltdown as widening credit spreads made a mockery of its risk models. Goldman was also badly stung in the fallout: Industry observers estimate it lost almost $1 billion.
“It is hard to really discuss risk with anyone on Wall Street without thinking–at least a little bit–about 1998,” says Viniar. And while he insists that the crisis did not change Goldman’s underlying risk philosophy, he admits, “we learned some things–some of which we knew, some of which we should have known. And a couple of things we do a little differently.”
What Viniar learned during the crisis was that the tools being used by Goldman and most other firms to analyze day-to-day risk–specifically value-at-risk–were inadequate in crisis situations. VAR does not take liquidity risk into account, and it ignores outlying catastrophic risks. “VAR is a tool, but only one tool [of many],” notes Viniar. Even scenario analysis, which can address some of VAR’s deficiencies, came up short in 1998, he adds. The crisis “caused us to look at more-extreme scenarios than we used to look at,” says Viniar. “It made us expand out the tails of what we deemed a realistic possibility.”
In response, Viniar helped Goldman develop a new generation of risk-management models, including hybrid VAR measures for relative value and liquidity risks, a firmwide credit-spread-scenario analysis, a firmwide emerging-markets stress test, and economic scenario analyses.
The results speak for themselves. For example, as the economy slowed in late 2000 and throughout this year, credit spreads responded by growing wider than they had back in the disastrous fall of 1998 (albeit not as rapidly). But the effect was minimal, notes Viniar, thanks to the decision–driven in part by the new scenario analyses–to hedge credit spreads with total-return swaps.
Ultimately, however, “there is no way to eliminate risk, only to control it,” says Constant. That’s particularly true at Goldman, he notes, which relies more on capital markets, institutional sales and trading, and investment banking, and less on asset-management fees and other activities that produce more-predictable earnings streams. Compared with other leading Wall Street investment banks, he adds, “Goldman has much more complex economic benefits and risks.”
To better control those risks, Viniar sits on or chairs no less than four committees responsible for risk management. In addition to sitting on the executive management committee–“the ultimate controlling body,” says Viniar–he is co-chair of a firmwide risk committee, which determines Goldman’s risk appetite and approves overall market and credit risk limits.
But while market and credit risks garner the most press, he says, they pale in comparison with liquidity risk. “The way companies in this industry get in real trouble is when they actually run out of money,” says Viniar. Liquidity, of course, is a traditional CFO concern, and one that he watches closely as chair of Goldman’s finance committee.
But controlling risk, he insists, extends well beyond any one committee–or any one person, including him. “That any one person could be solely responsible for risk management is not even within the realm of possibility,” says Viniar, who describes Goldman’s trading managers and other revenue producers as “the first line of defense. They are responsible for analyzing risk versus reward.”
That responsibility, he says, is reinforced by an organizational structure that is carefully designed to avoid potential conflict. “We believe very strongly in segregation and separation of duties. You can’t, for example, have anyone who trades also able to move money.” Such separation also applies to those who control risk. “No one in a control function reports to anyone in a revenue function,” says Viniar. “I set their compensation, I am responsible for their promotions.” In the infamous 1995 case of Kidder Peabody & Co. bond trader Joseph Jett, who was accused of losing $300 million, recalls Viniar, “the person monitoring [Jett’s] risk reported to him. One of the fastest ways to get into trouble in the securities industry is not to have independent [risk monitors].”
Of course, as that steam pipe demonstrated, not all risk can be quantified or modeled. Recently, Viniar has had the more-traditional CFO headache of dealing with the impact of a market downturn. “There is clearly a risk of being too big for the environment, but there is also a risk of being too small,” he notes. Although Goldman has had some layoffs, Viniar is cautious about downsizing. “If you are not appropriately sized when the [business] environment gets better,” he says, “you will lose clients to your competitors.”
Other risks can be even tougher to define. Viniar declines to comment directly on questions about analyst independence that have plagued Goldman and other investment banks this past year, nor will he discuss recent lawsuits brought by disgruntled investors, but he admits to staying closely involved in those issues. “Making sure that we are not taking undue reputation or legal risks is just as much my responsibility as looking at new products, trades, or commitments of capital,” he insists. “There is nothing more important to us than our reputation.”
Obviously, there’s a solid financial reason for Goldman’s CFO to take reputation seriously. “The return on equity that Goldman Sachs generates in its advisory businesses is traditionally higher than it is for some of its peers,” notes Constant. “The higher-return segments are those where you sell the perception of wisdom and expertise. A lot of people hire Goldman Sachs because it is Goldman Sachs.”
Fall 1998 taught Goldman to hedge under extreme scenarios, says Viniar.
Despite wider credit swap spreads in 2000, the impact on the firm was
Fall 1998 [*] credit Y2000 [**]
AAA 20 36
AA 30 55
A 45 80
BBB 70 116
BB 215 154
B 320 413
Swap Spread (5yr) 30 30
Swap Spread (10yr) 25 24
(*)Fall 1998 covers the period from 7/31/98 through 9/30/98.
(**)Y2000 covers the period from 1/03/00 through 12/29/00.
Source: Goldman Sachs
Note: Table made from bar graph
COPYRIGHT 2001 CFO Publishing Corp.
COPYRIGHT 2001 Gale Group