By Thomas K. Grose

THE STOCK MARKET MAY BE UNPREDICTABLE, BUT THE DEMAND FOR ENGINEERS WHO CAN DEVELOP SOFTWARE THAT PREDICTS RETURNS IS ANYTHING BUT.

Most investors would find a crystal ball useful for mitigating risk: It wouldn't hurt to know in advance which stocks are going to soar and which are doomed to plummet, or the price of soybeans and sow bellies three months hence. Engineers are likely to look askance at crystal balls, however, especially when there are more accurate ways to channel the future. In recent years, a new breed of Wall Street soothsayers—financial engineers—has assembled a variety of quantitative techniques based on computer models, advanced math formulas, and financial theory that predict returns and assess risk.

When quantitative technology proved its effectiveness and began reshaping Wall Street more than a decade ago, the industry often turned to industrial or operations research engineers to design new investment tools: derivatives, which are complex financial products whose value is derived from other financial products, often futures; and arbitrage, an investing method that exploits market inefficiencies. These tools were put to use in hedge funding, as the engineers sought to reduce risk and deliver positive returns regardless of market conditions.

Following the success of these early financial engineers, a growing number of universities began offering master's degrees in financial engineering. By one estimate, there are 40 to 50 such programs today. Most are multidisciplinary programs that combine engineering, math and statistics, and business and economic courses.

Typical of these new financial engineering programs is the University of Michigan's (UM), based in its College of Engineering but run jointly with its School of Business and its School of Literature, Science, and the Arts. The growth of Michigan's financial engineering program has been swift, paralleling the rapid advancement of quantitative technologies in financial services. Michigan launched the program in 1997 with six students; today 95 are enrolled. In 2000, it graduated 20 students; this year, it expects to graduate 65.

UM's degree grew out of its industrial and operations engineering program. "That's the closest thing we have to a business school in engineering," explains Stephen Director, dean of the College of Engineering. Because the engineering college had a long history of running interdisciplinary programs, while Michigan's business school was less quantitative than most, "it made sense for it to grow out of here," Director says.

Princeton also runs a successful new program based in its six-year-old operations research and financial engineering department and run jointly with its business school. With 30 students currently enrolled, "We feel like we are blazing new ground, creating a new engineering discipline," says Erhan Cinlar, department head. Columbia University's industrial engineering and operations research department also runs its program with the business school, and has 65 students enrolled. At some schools, such as Stanford and Columbia, the program is called financial mathematics and is based in the math department. At others, including the University of California-Berkeley and MIT, it's anchored in the business school.

Roots: Square and Academic

Stephen Pollock, director of Michigan's program, is skeptical of programs that are primarily financial math or focused on empirical business theories. The engineering perspective is the glue that binds the complex formulas and economic theories together, he insists. "It is financial engineering," Pollock says, stressing the e-word. Indeed, Director adds, the finance industry began hiring engineering graduates because people with the skills to devise complex financial instruments "were not coming out of the business schools."

Problem solving and an ability to deal with uncertainty, Pollock and Cinlar say, are key engineering skills that financial engineers must learn. "To me," Cinlar explains, "an engineer is someone who has a problem and solves it. We concentrate on the problem, and we bring and use whatever tools we need to solve it." In the world of finance, helpful tools include modeling, complex equations, and such engineering problem-solving techniques as optimization and simulation. Guillermo Gallego, chairman of Columbia's industrial engineering and operations research department, says mathematicians too often look for abstract elegant solutions, while engineers seek to remedy real-world problems. As for uncertainty, Cinlar notes: "Most people want to avoid risk. We love it. And people are willing to pay us to show them how to avoid it."

Students attracted to financial engineering tend to be career-oriented—and interested in a fat paycheck. "There are many students who want this degree, not for the education, but for the perks," Pollock says. Gallego agrees. "They are bright students who like math but want it relevant." Pollock says some also view it as an M.B.A. shortcut, "a way to get into finance without the years of experience required for an M.B.A."

Roughly half of Michigan's financial engineering students have undergraduate degrees in engineering, with most coming from the industrial/operations research, electrical, and mechanical engineering disciplines. About 25 percent are physics and math majors, and about a quarter come from economics and business. Princeton's students tend to be equally divided among engineering and math undergraduates; Columbia's mostly have engineering and math degrees, though a few have degrees in economics.

Michigan, Princeton, and Columbia get many more applicants than they have places available. UM reports that total applications are down over the past two years. Nevertheless, Pollock admits that UM's program has about 25 percent more students than it should have. In both 2002 and 2003, it enrolled about 70 students. Tightened admissions standards will keep future classes in the 45 to 50 range. It is likely, however, that the number of universities offering the degree will increase because there is so much demand—much of it coming from foreign students. The overwhelming majority, 90 percent, of Michigan's applicants are from overseas, and they comprise 50 to 55 percent of those accepted. "Schools are going to go where the money is," Director says. Gallego says that "things usually move slowly in academia, but finance engineering is moving fast."

Nonetheless, Director cautions, running an interdisciplinary program can be a challenge for many schools, and turf wars have been known to break out. His college has a history of working with other schools and departments, but that's not true at all schools. "The first question is always, ‘How do we share the revenues?' " And faculty from different areas need to feel they are on equal footing. "It can't be one school pushing it on the others or trying to do it all itself."

Just as there is demand for the programs, there is usually a high demand for the graduates. "They are heavily recruited," Director says. Cinlar and Gallego agree that industry demand is currently strong. But, Gallego adds, two years ago students struggled to find jobs, "It goes with the economy." Pollock stresses that graduates must have strong communications skills. Some early Michigan graduates were foreign students who were expert at crunching numbers but whose English skills were weak, and they had a hard time finding jobs. Michigan's admissions process now looks for applicants who not only can handle the math but handle themselves in a business environment, Pollock says. "You will not get hired if you present yourself as a math nerd who hopes only to be asked to solve an equation."

Once grads do land jobs, the pay is good. Starting salaries tend to range from $50,000 to the low six figures. Companies that have hired UM students are a veritable Who's Who of blue-chip firms, including Lehman Brothers, T. Rowe Price Associates, J.P. Morgan Chase, Merrill Lynch, Morgan Stanley, and Goldman Sachs.

No one foresees a doctorate in financial engineering since the recipients are overwhelmingly career-minded and not looking for a life in academia. "But," Pollock says, "new models and methods need to be developed, so research needs to be done, and that will require Ph.D.'s." That said, these doctorates may not be in financial engineering; they'll likely be in industrial engineering or mathematics, with dissertations on subjects relevant to the field, such as financial math or modeling.

Michigan's program has experienced a few growing pains. Placement was initially somewhat hampered, Director says, because the engineering college was suddenly dealing with companies it had never dealt with before. Finding the right mix of skills among faculty members can also be a daunting task, though it helps that UM is a large school with a good-sized pool of qualified instructors. Most of the problems have been bureaucratic, Pollock explains, because faculty members have to deal with colleagues who might be located across campus, not across the hall. "But there haven't been any intellectual problems." Clearly a crystal ball isn't needed to predict that financial engineering has a bright future.

Thomas Grose is a freelance writer based in Great Britain.