Chapter 2
Lifetime Achievement Award
By Dwight Cass
Risk, 44-45, January 2002
In the late 1960s, Wells Fargo Bank in San Francisco assembled a team of uniquely gifted thinkers who would go on to push the boundaries of financial theory. Working alongside William Sharpe, Myron Scholes, Fisher Black, and Robert Merton at the time was Oldrich Vasicek, who is Risk's lifetime achievement award recipient. Like his Wells Fargo colleagues, Vasicek has had a profound effect on both financial theory and practice. His equilibrium model of the term structure of interest rates is widely acknowledged as the landmark work in the field, and many credit it for setting off the series of modeling innovations that paved the way for the rapid growth of the interest rate derivatives market. Ten years later, he developed a groundbreaking credit portfolio risk model that paved the way for the approaches incorporated in the Basel II capital Accord.
Among market practitioners, he is perhaps best known for co-founding KMV, the San Francisco credit analysis firm, and for using Scholes, Black, and Merton's insights on option pricing to develop the expected default frequency (EDF) credit pricing system—a so-called Merton model approach—at the heart of KMV's product line. The company has been extremely successful, with KMV claiming more than 70 percent of the world's largest financial institutions as clients. It is hard to find a major credit derivatives dealer or loan house that does not use it. The success of the approach has prompted other companies, including Moody's Investors Service and JPMorgan Chase, to add a Merton model-based default probability estimator to their offerings.
This combination of theoretical and business accomplishments alone might be enough to warrant Risk's lifetime achievement award. But 60-year-old Vasicek has shown no interest in resting on his laurels to free more time for his enthusiasms, which range from playing classical flute music to windsurfing in the cold, windy waters surrounding San Francisco. He continues to tackle new challenges, such as the tricky problem of modeling spot and derivatives price behavior of nonstorable commodities such as electricity and telecoms bandwidth (on which he co-authored a technical article with Hélyette Geman published in the August 2001 issue of Risk). And, according to his colleagues at KMV, he remains the driving force behind the evolution of that firm's product line.
Vasicek did not originally intend to pursue a career as a financial theorist. He trained in his native Czechoslovakia as a mathematician, earning a PhD with honors in probability theory from Charles University in 1968. The first event that placed him on the road to his career in finance was the Soviet invasion in August of that year. Vasicek had been at the Czechoslovak Academy of Science in Prague, working in pure mathematics, when the Soviet tanks rolled in. He and his wife left for Vienna a few days later.
He made his way to San Francisco and began applying for jobs as a mathematician. He was interviewed for several positions—including a job at Stanford University's marine biology department doing spectral analyses of dolphins' songs. But fate lent a hand again, and he was interviewed by John (Mac) McQuown, head of Wells Fargo's Management Science Department, who was looking to hire several mathematicians. “I'm a mathematician by profession, and only went into finance because my first job here was in a bank,” Vasicek jokes.
McQuown's group already included Scholes, Merton, and Black. Also, Sharpe was consulting for the bank. This was before the Black-Scholes option pricing model had been published. “Mac hired these guys before they were famous,” Vasicek says. Wells Fargo was one of the first banks to embrace Sharpe's capital asset pricing model (CAPM), and McQuown's group was looking at ways to apply it. Vasicek worked on this project and on index fund construction.
McQuown, who would later launch KMV with Vasicek and Stephen Kealhofer, said Vasicek's talents quickly became obvious. “I vividly remember Fischer Black saying to me, on a couple of occasions, that when he had a really intractable mathematical problem, he would go to Oldrich,” he says.
Myron Scholes, the Nobel laureate in economics who is now a finance professor at Stanford University in Palo Alto, California, says: “He's got tremendous mathematical and engineering abilities, and is also a very good listener. He articulates his views and holds to his path if he thinks he is right, but he's willing to appreciate other people's views, which makes him a good scientist.” Indeed, in an arena where oversized egos are often the norm, many of Vasicek's past and current colleagues praise his tolerance and humility. “He has ideas that seem quite simple in retrospect, so much so that they're quickly borrowed by everyone else. But they reflect a deep and powerful mind.” says Kealhofer. “He has a wonderful old world/new world charm—a mixture of Prague and San Francisco,” he adds.
In 1974, Vasicek left Wells Fargo to teach at the University of Rochester (New York) Graduate School of Management, where he would stay for two years. During this period, his attention turned to the problem of interest rate behavior, which he would continue to pursue when he returned to California as a visiting professor at UC Berkeley's business school. “The pricing of bonds and behavior of interest rates was an open question at that point,” Vasicek says. “The work on the CAPM was exciting, but the research that had been done wasn't applicable to bonds.”
Vasicek realized that arbitrage would link bond prices up and down the term structure, so, for example, there had to be a relationship between investing in a one-year bond twice in succession and investing in a two-year bond straight off. He found the common denominator to be the short-term interest rate. “If you postulate that the pricing of a long bond is a function of the short rate over the term of the bond, or more accurately, of your probabilistic description of the short rate's behavior over the term of the term bond, you have a common variable for the pricing of bonds,” he notes. “In other words, you have a state variable for the pricing of bonds with all terms from the current value of the short rate—that was the point at which the idea really broke for me at the time,” he says.
Inspiration
Vasicek's paper, titled “An Equilibrium Characterization of the Term Structure,” was published in the Journal of Financial Economics in 1977. It was either the basis for, or inspired many of the theoretical advances that came in the years that followed. Among these was the influential 1985 Cox-Ingersoll-Ross model. One of that model's authors, Stephen Ross, professor of finance and economics at MIT's Sloan School of Management in Boston, says of Vasicek's work: “It is a wonderfully simple, empirically amenable model. It guides us in a lot of our intuitions about the subject.” McQuown argues that Vasicek's model was the critical catalyst that spurred development of the interest rate derivatives market. “The interest rate swap market relies on that model,” he says. “If it wasn't for Oldrich's contribution, the interest rate swap market may have taken longer to develop. I dare say the problem would have been solved sometime, but Oldrich solved it first.”
“It was like opening a Pandora's box as far as academic research goes,” Vasicek says. But, he notes with some modesty that he had not actually aimed to redefine how the world viewed interest rate products. “But what's kind of funny is that my paper focused on the theory,” he says. “What has become known as the Vasicek model was just an example. I developed the theory and I wanted to illustrate it on a particular type of process and go through the calculation and determine the final equation. But it's the example I used that's been remembered.”
The shift in his emphasis to credit risk came when McQuown recruited Vasicek for an ill-fated scheme in the early 1980s. “I persuaded him to join me in a venture that ultimately went down the tubes called Diversified Corporate Loans in 1983. We wanted to create a pool of credit from major US banks where the banks could swap qualified loans into a pool in return for a pool interest.” The idea was to give the banks liquidity and portfolio diversification, but to do so, the firm needed a way to value credit risk, to aid the participating banks in valuing the loans they put into the pool, and the pool itself.
“So,” Vasicek says, “I started to work on credit. Up to then, credit was strictly a judgment call. So I developed an application of option pricing theory—the Black-Scholes and Merton work.” Kealhofer says: “He laid down the theoretical footprint in a short time that we've been using for 17 years now. He laid the groundwork for both the basic credit technology and the portfolio technology. We've been laboring in those two veins ever since.”
When DCL went out of business in 1989, Kealhofer (who had joined several years before), McQuown and Vasicek launched KMV to further develop and market the credit evaluation tool. Myron Scholes says: “It set the stage for using more modern technology than the rating agencies have used, and it led to more people thinking about using option technology to do credit pricing. Others had used the option framework to price debt. But his work at KMV took the lead in developing something that was usable by a vast number of people.” Indeed, having a widely available set of reliable credit pricing tools was a necessary precondition for the development of the credit derivatives market, Scholes notes.
Pioneering
MIT's Ross says: “Vasicek's work on credit risk is a different kind of pioneering. It's a demonstration of how one can take high-quality academic work and turn it into a solid business without compromising the academic stuff or the understanding of the needs of the marketplace. Marrying those two is never easy.”
Vasicek says the launch of competing Merton-model products—JPMorgan Chase's being the most recent, the specifications of which were published in the November 2001 issue of Risk—is good for the markets. “I'm glad that people are becoming interested, because any effort in this direction will make the market for these securities more liquid and more efficient,” he says. “There has been a bit of confusion about what a Merton model is. I guess people keep forgetting it is not a formula, it is a framework. It is a structure that allows you to get a specific mathematical solution to the value of a firm. But it would depend on the assumptions you make about the firm's financial structure and the capital statements and the payments the firm is making—coupons, dividends, and the nature of the debt, convertibility and optionality. It's a very complicated thing. We give a lot of attention to how we characterize the firm, on top of the mathematical problem, then it's a fair amount of work.”
As co-head of KMV's research group, Vasicek continues to guide development of the firm's products. “We just rolled out a new method of calculating a loss on credit instruments that uses the empirical distribution of default risk,” Kealhofer says. “That was suggested by Oldrich.” As factors such as the Basel Accord and the rapid growth of the credit derivatives and synthetic collateralized debt obligation markets have kept the credit risk modeling arms race in full swing, and new pricing challenges such as those in the electricity and bandwidth markets have arisen, Vasicek's work has remained at the vital crux between theory and practice.