Investors can identify the factors that determine stock prices and outperform stock markets worldwide, according to a new academic study that refutes the efficient market theory and other tenets of modern finance.
Robert A. Haugen and Nardin L. Baker, who co-authored the study, developed a model that found the highest returns in stocks that are generally the least risky and have the largest capitalization and the most liquidity.
"I call these superstocks," said Mr. Haugen, professor of finance at the University of California, Irvine.
These are the types of stocks into which large institutions can put a lot of money, he noted.
In their study, Messrs. Haugen and Baker found the predictions of expected returns from the model "are strongly and consistently accurate."
The research points to major failures in modern portfolio theory and the efficient market hypothesis. The widely accepted portfolio theory suggests the riskier the stocks, the higher returns and vice versa. The efficient market hypothesis suggests the best return one can earn in the long run is the market average, because no one can have an information advantage to beat the market.
"That (efficient market hypothesis) is a catchy phrase and this conclusion is controversial," said Mr. Baker, a portfolio manager at Grantham, Mayo, Van Otterloo & Co., Boston.
Mr. Haugen said the Journal of Financial Economics will publish the paper, "Commonality in the Determinants of Expected Stock Returns," in July.
According to the paper, the factors that determine stock prices are stable over time and are "strikingly common" in major equity markets around the world. Those factors can be identified by investors to determine expected stock returns.
They show evidence that "stock returns can be predictable with factors that are inconsistent with the accepted paradigms of modern finance" such as the efficient market hypothesis.
The study found "stocks with higher expected and realized rates of return are unambiguously (their emphasis) of lower risk than the stocks with lower returns."
The results show a consistent risk-return pattern contrary to modern financial theory.
"Our high return deciles without question have lower risk," Mr. Haugen said.
"If the market were efficient, it would be impossible to construct a portfolio like that," of high returns and low risk, Mr. Haugen said.
The authors did monthly regressions on the stocks in the Russell 3000 stock index, which consists of the 3,000 largest U.S. stocks in terms of market capitalization. They tested the stocks over the period 1979 through 1993.
In addition, they tested their assumptions on stocks in France, Germany, Japan and the United Kingdom for a shorter period, 1985 through mid-1994, because of limitations in commercially available databases.
The model divided the Russell index into performance deciles; it found the higher the expected return, the lower the risk.
"The high-return deciles are safer (than the low-return deciles) and less risky," Mr. Haugen said.
"Low-return deciles in volatility, beta, in all dimensions are worse (in terms of risk)."
In their paper, Messrs. Haugen and Baker wrote they consciously designed their analysis to minimize the problems in testing data "so that we are confident that our results are real."
Mr. Haugen is selling software to money managers and pension funds that are setting up investment portfolios based on the research.
Among the money managers: Harbor Capital Management, Boston, which is planning a quantitative midcapitalization equity portfolio; the analytic trading division of Jefferies & Co., Los Angeles, which is developing a market-neutral global portfolio and where Mr. Haugen also plans to place some money; Montag & Caldwell Inc., Atlanta; Northwest Quadrant Inc., Newport Beach, Calif.; George Vahanian, an adviser at Merrill Lynch Inc., Pittsburgh; and Analytic/TSA Capital Management, Los Angeles.
Industrifinans, a Norwegian investment manager, has been running a $35 million portfolio based on the model since November 1994 using Mr. Baker and Grantham, Mayo as its investment adviser.
Mr. Haugen said that is the only portfolio with any real track record.
According to Mr. Haugen, the Industrifinans portfolio outperformed the Morgan Stanley Capital International World Index by 140 basis points for the last two months in 1994; by 540 basis points in 1995; and by 100 basis points in the first two months of 1996.
The determinants are related to risk, liquidity, price level, growth potential and stock price history. Among the 52 factors, each given various weights in the process they used, are:
For risk: market beta, volatility of returns, debt-to-equity ratio;
For liquidity: market capitalizations, trading volume;
For price level: earnings-to-price, book-to-price, dividend-to-price, cash flow-to-price ratios;
For growth potential: profit margin, return on assets, return on equity, earnings growth; and
For stock price history: excess return, relative to the Standard & Poor's 500 Stock Index, in periods ranging from one month to five years.
Generally, the stocks that emerge in the optimum portfolio have high book-to-price, cash flow-to-price and earnings-to-price ratios and good return on equity and growth rates, Mr. Baker said.
The model portfolios have a turnover of 60% to 80% annually.
No single stock has the ideal factors of the model, Mr. Haugen said.
"If you use a conventional screen to profile, there is no stock that fits the profile," he said. "But the market is inefficient enough to allow you to find a portfolio that fits the screen the best."
For example, he said the top decile amounted to 300 stocks the model screened from the Russell 3000. He said a real portfolio would need at least 100 stocks.
"Hardy anybody screens to get an aggregate portfolio," as his model does, Mr. Haugen said. "They screen to get individual stocks to meet a particular profile."
Their research also shows a flaw in the small-stock theory. Pension funds and other large institutional investors can't take advantage of the theory that contends small stocks outperform large stocks in the long run, because the cost of trading small stocks outweighs any return advantage, Messrs. Haugen and Baker said.
But institutional investors will be able to profit from this new research, they contend. One reason: the stocks in the portfolios developed from it are among the biggest and most liquid, minimizing trading and market impact costs.
Because these are big stocks, the determining factors won't be easily arbitraged away, erasing any return advantage.
"You need major players in the market to close it," Mr. Haugen said. "You need fiduciary investors (which are big players), and fiduciary investors are slow to move. They can't change their process and remain successful, because they'll lose their track record, and a track record is their ticket to the finals (presentations in hiring money managers)."
He believes managers will develop new products to use the research.
"These excess return (found in the research) can be reduced" as more investors discover them, Mr. Baker added. "But that can take a long time" with these large stocks.
