Fifty years ago, Harry Markowitz was looking for a dissertation topic. The investment world hasn't been the same since.
Now known as the father of modern portfolio theory, Mr. Markowitz laid the foundation for investment management in the second half of the 20th century.
His application of mathematics to portfolio management related risk to return for the first time.
"Without his ideas, there would be no modern investment management," said Andy Turner, managing director, investment policy and research, Frank Russell Co., Tacoma, Wash.
Said Rob Arnott, managing partner of First Quadrant LP, Pasadena, Calif.: "He really shattered the paradigm of the first half of the century, where the goal of the investment community was to pick the best investment. He recognized the power of diversification and demonstrated it mathematically."
Winner of the 1990 Nobel Prize in economic sciences, along with William F. Sharpe of Stanford University and Merton Miller of the University of Chicago, Mr. Markowitz's pioneering work in understanding the nature of risk in portfolio management laid the foundation for much of modern finance.
His influence extends to such major developments as the Capital Asset Pricing Model, the creation of indexed portfolios and, more recently, concepts of budgeting risk within a portfolio.
In 1986, he wrote that the probability of a stock market crash was much higher than was revealed in a standard returns distribution, Mr. Arnott said. The Journal of Finance rejected that article; a year later, the stock market collapsed.
A chance conversation
A chance conversation in 1950 led Mr. Markowitz to develop what has become his best-known contribution to investment theory: how risk and return work hand in glove in portfolio construction.
Waiting to discuss possible dissertation topics with Jacob Marshak, his academic adviser at the University of Chicago, Mr. Markowitz started chatting with a broker who also was waiting in Mr. Marshak's anteroom.
"This broker suggested that I try applying mathematics or statistics to the stock market. I mentioned that to Professor Marshak," Mr. Markowitz said in an interview.
Mr. Marshak, an ex-director of the Cowles Commission for Research in Economics, responded that "Alfred Cowles himself had been interested in the application (of mathematics) to the stock market," and that notion had been "part of his motivation when he founded the Cowles Commission," Mr. Markowitz said.
Armed with a reading list on financial theory, Mr. Markowitz arrived at his revelation on the roles of risk and return in one fateful afternoon in the library. He was reading John Burr Williams' "Theory of Investment Value," a classic text that posits that the value of a stock should be the present value of its future dividends. But to maximize the return on a portfolio as Mr. Williams suggested, Mr. Markowitz realized, "then you would put all your money into one security, and I knew that didn't make sense."
That meant investors must be interested in two things: risk and return. "The first idea that popped into my head: Measure risk in terms of variance," he said. Some believe Mr. Markowitz's most original insight was that the riskiness of a portfolio depends on how the invested securities move in relation to each other.
"The way you get diversification is not just through stocks with low variances but through low covariances," Mr. Markowitz said.
As an economics student, it was "just very natural to draw a curve" showing the tradeoffs between risks and rewards available to investors. "So I drew the world's first efficient frontier, reflecting on what I had just read. So the basic idea all came in that one afternoon," he said.
"I figured I could get a dissertation out of that, I didn't realize I could get a Nobel Prize."
Mr. Markowitz's seminal work was published in 1952 as a paper called "Portfolio Selection." By the time he expanded the paper into a book seven years later, he realized there was a major obstacle to implementing his thesis: creating an efficient portfolio of a large number of securities requires an enormous number of calculations.
Mr. Markowitz suggested a simplified approach, relating the returns of securities to a single factor. But he left that task to Bill Sharpe, who had turned to Mr. Markowitz for help in developing his own dissertation topic. Simplifying the model, Mr. Sharpe related stock prices to the market as a whole.
With Mr. Sharpe's simplification, the time required to create an efficient portfolio was cut to 30 seconds on an early 1960s state-of-the-art computer, compared with 33 minutes for the full Markowitz program.
Today, at age 72, Mr. Markowitz serves as "a quant guy's quant," with a number of consulting relationships.