Theories treated as common knowledge in today's business schools weren't always welcome ideas.
A young University of Chicago Ph.D. candidate published a short article in the March 1952 Journal of Finance that eventually revolutionized the way assets were managed.
Harry Markowitz, a brilliant but unassuming man, wrote the simply titled "Portfolio Selection," in which he presented for the first time a method by which risk could be overtly recognized in investment decision making. Previously, investment theory dealt with ways to value individual stocks and identify those with the best prospects. Risk, if it was considered at all, was considered subjectively. It had to be, as there was no recognized way of identifying, measuring and controlling risk.
Mr. Markowitz's paper, as its name implied, focused not on individual stocks, but how to build efficient portfolios -- efficient in terms of expected return and risk.
His work marked the start of an incredible 30-year period of academic research into the workings of the capital markets.
The great problem with Mr. Markowitz's approach to the construction of an efficient portfolio was the enormous number of calculations required.
Into the breach stepped William F. Sharpe, a young doctoral candidate whose dissertation was to be a simplified, and hopefully more practical, version of Mr. Markowitz's mean-variance model.
Mr. Sharpe published his paper in 1963, but continued to work on the concept, producing the capital asset pricing model in 1964. Mr. Sharpe introduced the concept of "beta" as a measure of risk; this concept alone helped revolutionize portfolio management.
A NEW HYPOTHESIS
Late in 1965 another seminal study, as important in its way as the Markowitz dissertation, was published in the Financial Analysts Journal. It contained what was to become known as the efficient market hypothesis.
Eugene Fama, then a young assistant professor of finance at the University of Chicago, concluded there are so many security analysts and investors searching out companies selling below their intrinsic values that prices adjust "instantaneously" to eliminate any discrepancy. This idea laid the intellectual foundation for the development of index funds.
The work of Messrs. Markowitz, Sharpe, Fama and several others became known as modern portfolio theory, which laid down new rules for institutional investing.
The concepts comprising MPT were not greeted with universal acclaim. In fact, the efficient market hypothesis, the capital asset pricing model and beta met with great skepticism.
But in early 1974 a new prophet appeared whose work and words helped convert many skeptics: Barr Rosenberg, a young, soft-spoken finance professor at the University of California at Berkeley.
He realized that while beta measured how a stock moved relative to the market as a whole, there must be underlying reasons for that movement. There must be fundamental reasons, related to the financial characteristics of the company, that explained why one stock moved more than the market and one moved less, and why many stocks moved together more or less independently of the market.
This was Rosenberg's concept of "extra-market covariance" -- movement not explained by the market's movements.
BARRA IS LAUNCHED
In 1974, with help of a consulting contract from American National Bank & Trust in Chicago, Mr. Rosenberg established a consulting firm called Barr Rosenberg Associates, BARRA. The assignment at American National was nothing less than the automation of portfolio management, from forecasting security returns, to estimating the risks of each security, to developing efficient portfolios.
During this work, he and his staff developed a program for predicting betas. Until this point, betas had been based on observation of how stocks had moved relative to the market in the past. Mr. Rosenberg's work looked at dozens of financial variables of companies -- size, financial condition, and others -- to determine which, if any, explained beta. He found multiple factors contributing to beta and was able to use those factors to accurately predict betas. "Barr's better betas," as Mr. Rosenberg's estimated or predicted betas became known, provided an intellectual explanation for beta as a measure of a stock's volatility.
The fact that betas could be explained by the kinds of fundamental financial data security analysts generated, and were not simply the product of some pointy-headed academic's imagination, made them more acceptable to Wall Street. Where many had resisted discussing with clients the betas of their portfolios or the stocks they bought in the first half of the '70s, "Barr's better betas" made it more acceptable to do so in the second half of the decade. Indeed, pension executives expected -- and even demanded -- it.
In mid-1975, Roger Ibbotson, a finance professor at the University of Chicago, and Rex Sinquefield, an index fund pioneer at American National Bank and Trust in Chicago, published their seminal study of the long-term historic returns of stocks, bonds, Treasury bills and inflation.
The Ibbotson-Sinquefield study examined the returns of the three asset classes from the beginning of 1926 to the end of 1974. They found that over that period, which encompassed the Great Depression as well as the bull market of the 1950s and 1960s, common stock total returns (market gains plus dividends) had averaged 8.5% compounded annually, while bonds had compound annual total returns of 3.6%, and Treasury bills had compound annual returns of just 2.2%, equaling the rate of inflation for the period.
The Ibbotson-Sinquefield study updated, and differed in significant ways from, a 1965 study by Lawrence Fisher and James Lorie into the long-term investment returns produced by common stock.
While the Fisher-Lorie study used as its stock market proxy the equal-weighted New York Stock Exchange index of all stocks listed on the exchange, the Ibbotson-Sinquefield study used the capitalization-weighted Standard & Poor's 500 index. This partly explains the difference in equity returns between the two. The Fisher-Lorie study put the long-term return at 10.3% between 1926 and 1965, compared with Ibbotson-Sinquefield's 8.5% for the longer period of 1926 to 1974. The bear market in the last two years of the Ibbotson-Sinquefield study also would have contributed to the lower returns.
Messrs. Ibbotson and Sinquefield used their historic data, efficient market theory and probability analysis to project returns on the three asset classes out to 2000. They projected that equities would remain the best long-term investment with annual total rates of return of 13.6% compounded. But there was a 5% chance those returns could be as low as 4.9% or as high as 22.9%. Long-term government bonds were expected to return 8.4%, while Treasury bills were expected to give 7.1%, compared with an expected inflation rate of 5.9% per year. Although the projections were criticized at the time, they were not far off the mark through mid-1998.
Nevertheless, the Ibbotson-Sinquefield study was timely and significant for two important reasons. One, it helped re-establish confidence in equity investing, which had been badly shaken by the 1973-'74 bear market. If equities could still show a large long-term edge over bonds after a bear market that cut the value of the Dow Jones industrial average in half, then the case for more equity investing was strong. In addition, the study provided academically rigorous data to be used in asset/liability planning models of the kind consultants were marketing to pension executives.
A controversy erupted in 1980 when a finance professor at the University of California-Los Angeles challenged the apparent alphas being produced by many money managers, and indirectly questioned the capital asset pricing model.
Professor Richard Roll had noticed that many managers in the late 1970s appeared to be producing positive alphas. That is, they were producing returns above that which would be expected from the level of risk, as measured by beta, they were taking. According to the CAPM, that extra return, known as alpha, was the result of taking risk specific to each stock and reflected the manager's ability to select stocks that outperformed the market. That is, it was a measure of a manager's stock-picking skill.
Mr. Roll argued that this alpha, in most cases, was a result of using the wrong benchmark to reflect the market. An example of this was when the S&P 500 was used as a benchmark for the calculation of beta, and the performance of money managers. The S&P 500 is not truly representative of the whole market and is an inefficient portfolio, and therefore gives a false impression of the risk being taken by money mangers, according to Mr. Roll's analysis.
He argued the positive alphas of many managers in the late 1970s were the result of their overweighting small-cap stocks when those stocks, which are underweighted in the S&P 500, were doing well. That is, the managers were taking advantage of a weakness in the index.
Measured against an efficient portfolio that is a better index, their stock-picking ability might well disappear.
Other finance professors defended the utility of CAPM and beta. Burton Malkiel, a professor at Princeton University, said beta's measurement of the risk of individual stocks was not very accurate, but the beta of a large and diverse portfolio was accurate.
Barr Rosenberg said CAPM made unrealistic assumptions about the world, but nevertheless it remained a useful theory.
Other defenders of CAPM said no one believed beta was a perfect measure of risk, but it was the best available at the time and would not be abandoned until a better measure was developed.
But Roll's criticism of CAPM and beta prompted consultants and pension executives to look more skeptically at managers' alpha.
It made them more aware of the fact that the S&P was not a very good replica of the market and accelerated the search for better indexes.
Another professor, Stephen Ross of Yale University, developed a possible replacement for CAPM -- the arbitrage pricing theory. APT never was as widely accepted or adopted as CAPM, although Messrs. Ross and Roll founded an investment management firm -- Roll and Ross Asset Management Corp. -- to invest money based on the theory's insights.