Until just a few years ago, the investment business resounded with genteel arguments between managers with different investment approaches. Each would point to compelling reasons why his or her methods were best.

Growth managers, for one, assumed that rapidly increasing sales, profits, and/or market share would lead to a rapidly growing stock price. Meanwhile, value managers argued convincingly that overlooked or out-of-favor companies would provide steady growth while high dividends and a large asset base would ensure downside protection. Small-company managers spoke fondly of discovering just one or two of tomorrow’s Microsofts. Large-company managers favored liquidity and well-established companies. Mid-sized-company investors argued that second-tier companies offered stability, growth potential, and the opportunity to exploit market inefficiencies. The few foreign-stock managers were busy trying to convince Americans that international investing wasn’t just plain crazy. (Emerging-market managers were all still in diapers.)

While these arguments made for wonderful entertainment, they were unresolvable. Debaters lacked even common definitions; their discussions were devoid of appropriate yardsticks, and they lacked the necessary tools to measure performance or risk. Each management approach yielded “acceptable” positive results, but each excelled at different times. Comparisons were necessarily difficult, and managers would each stress the time frames in which their own approach excelled.

Investors could hardly be blamed if they didn’t find solid guidance from Wall Street’s competing gurus. The truth is, all the gurus were just blowing smoke. Nobody knew what was going on! (The idea that various types of investments might be complementary hadn’t even been considered.)

A Modest Proposal

Two University of Chicago professors, Eugene Fama and Kenneth French, found an elegant way to help resolve the above problem. But in doing so, they touched off one of the liveliest debates — and biggest dogfights — finance has seen in years. While the results are surprising, their arguments are compelling. In fact, other data solidly supports their original study: Markets appear to have a sweet spot where higher returns can be expected without additional risk!

Under the Capital Asset Pricing Model (CAP-M), stock prices and expected future returns are related to both market risk and a unique risk that each stock has, called “Beta.” Beta is a measure of the volatility of the individual stock in relation to the market as a whole.

Everyone in finance loved CAP-M. It was elegant and relatively easy to understand and explain. There was just one small problem: Beta didn’t do a very good job of explaining either price or returns. In particular, CAP-M and Beta left large anomalies in two areas: small companies and low-priced companies had higher-than-expected returns.

In their June 1992 article, “The Cross-Section of Expected Stock Returns” (published in the Journal of Finance), Fama and French set out to find a better way to explain prices and returns.

Beta is a single-factor variable. Fama and French tried a number of other factors in combination to see if they could provide a better fit. They found that together, size and book-to-market (BTM) ratio did the best job of explaining stock performance. BTM is the ratio of a firm’s book value per share to its stock price. If you are particularly observant, you may have noticed that BTM is the inverse of price-to-book (P/B). This alternative figure was created out of necessity, because book value may sometimes be zero, and a ratio with zero on the bottom is impossible to use in calculations.

A firm with a high BTM has lots of assets per share compared to a low-BTM firm. As it happens, high-BTM firms have characteristics associated with “value” and low-BTM firms tend to be “growth” firms. Growth and value are somewhat fuzzy terms. Everyone seems to agree that Microsoft is a growth company, but value seems to be in the eyes of the beholder. BTM provides an objective measure.

“Value investing” may be one of the world’s greatest public-relations terms. Value firms are sick puppies. High-BTM firms (low P/B) tend to have low P/E’s (price-to-earnings ratios), low return on equity, low return on assets, slow or no growth of sales, disappointing profits, and other discouraging financial results. Even though they have large assets, the market has driven down the price of their stock. Because management often has no clear idea how to generate additional business growth, many high-BTM firms pay large dividends. They are troubled firms. Usually they have been troubled for a while, and will continue to be troubled for some time in the future. The risk of business failure is higher than that for healthy, growing firms. They are companies under stress.

Low-BTM firms (high P/B) are just the opposite. They have high P/Es, return on equity, and assets. Usually, they have histories of exponential growth of profits, sales, market share, and other healthy, desirable attributes. Generally they have so many investment opportunities internally that they do not pay high dividends. They are healthy companies.

Dividing the Market by Size and BTM Ratio

Fama and French took all the stocks in the N.Y. Stock Exchange and divided them into 10 groups, or deciles, by market capitalization. Market capitalization is the total value of all the securities of a firm. It is found by multiplying the price of a share by the number of shares outstanding.

Having now established arbitrary size groups, the pair took all stocks traded on all exchanges and distributed them into the appropriate size groups. Because of the smaller, average size of the non-NYSE stocks, the groups now contained many more stocks listed in the smaller deciles than seen in the equal distribution of the original NYSE deciles.

If you think of the size groups as being listed from top to bottom, Fama and French then horizontally sliced the result into ten groups (deciles) by BTM. They now had 100 portfolios or styles. Each portfolio was followed for one year, and then the procedure was done all over again. The performance of each of the 100 portfolios, as annually redefined, was followed from 1964 to 1992, a 28-year period.

The results were surprising. Small-company stocks had higher rates of return than larger-company stocks, but they had a much higher risk as measured by standard deviation of returns. However, high-BTM stocks (value) had higher rates of return than low-BTM stocks (growth) without any higher risk, as measured by standard deviation. This occurred at every size level. The value guys were right all along!

Investors in the bottom three deciles by size might expect a total return of about 5% (compounded annually) higher than the top three deciles. However, they will experience greatly increased volatility. At every size level, investors in the highest three deciles by BTM will receive about 5% greater compounded return than the bottom three. Value investors will not experience any significant increase in risk, at least as measured by volatility.

New Study, New Problems

The implications of this study, if validated, are staggering for both economists and investors. CAP-M and many of its implications are discredited. Investors can now construct portfolios with better performance than the market as a whole. Economists are stuck with the problem of explaining how value stocks can provide higher total returns without being subject to additional risk.

The author of CAP-M, William F. Sharpe, seems to be enjoying the debate. He has stated that he thinks Fama and French are on to something. He has also said that he thinks CAP-M, for which he won the Nobel Prize in Economics, was a pretty good first effort, and he is glad that the committee can’t take back the prize. The rest of the academics seem to have worked themselves into a frenzy either attacking or defending CAP-M. You can find plenty of papers posted on the Net at various universities if you care to follow the battle.

One of the implications of CAP-M was that the “super efficient” portfolio, the one which generated the most return per unit of risk, was the total world-market basket. An investor who wanted more or less risk could take this global-market index and either leverage it or water it down with a “risk-free” asset. This led to the spread of global indexing as an investment technique. Now it turns out that investors can do considerably better than the world-market index by heavily weighting their portfolios with value stocks.

Economic Justification for the Three-Factor Model

The idea that investors can expect additional returns without additional risk has even Fama and French struggling. It smacks too much of a free lunch. As faculty members of the University of Chicago, we would expect them to cheerfully die before they would admit to the existence of a free lunch. Consequently, they are trying to identify factors other than volatility that might explain the paradox.

Fama and French believe that their findings are consistent with an efficient market. They relate the differences in pricing and performance to cost of capital. If you run a large company and either borrow money from the bank or issue bonds, you will generally have to pay a lower interest rate than a small company because of the lower risk you appear to offer. In the same manner, if you issue stock, you will generally command a higher price than a small company. As we might expect, large companies have a lower cost of capital.

In a like manner, well-run firms have a smaller cost of capital than poorly run or stressed firms. High cost of capital means depressed stock prices and translates into higher expected returns.

A Bagful of Sick Puppies

I must admit that it is difficult to get too excited about an investment philosophy that advocates buying sick firms. It goes counter to the grain, and the whole idea takes a little getting used to. It’s hard to imagine generating much envy as you describe your portfolio of downtrodden losers. However, the returns generated by a diversified portfolio of distressed companies more than make up for the glamour of trying to uncover tomorrow’s Microsoft. It appears that investors have been paying too much for growth firms and too little for value firms.

Of course, the Fama-French research was subjected to all the normal indignities of any revolutionary study. However, enough studies in other markets and other time frames have validated their original work. Value stocks appear to perform equally well in global markets.

The three-factor model goes a long way toward explaining the returns of many mutual funds and portfolio managers. By examining managers’ styles (as defined by the size and BTM ratio of their portfolios), we have another powerful tool to evaluate management effectiveness. It’s even possible to examine the pattern of a fund’s past performance and make a very close guess as to the portfolio composition. In most cases, style accounts for far more of the performance than does skill, cunning, or luck.

Investors receive another benefit from the Fama-French three-factor model. By incorporating explanations of stock returns based on size and BTM ratios, we are able to more confidently predict expected returns when modeling portfolios. This methodology represents a measurable improvement over using unadjusted, raw-data past returns as the expected future rates of returns. Historical raw data is subject to unusual non-recurring, non-economic events that can dramatically distort its usefulness as a forecasting tool. Improved rates-of-return forecasts will lead to much-improved optimization models and better-performing, lower-risk portfolios.

While long-term data would strongly suggest the superiority of small-company and value investing to maximize returns, we must still be aware that growth and larger companies may experience extended periods of market favoritism. For instance, small companies did far below average in the 1980s. In the short run, we can expect significant year-to-year variation. Accordingly, it appears wise to continue holding some of both in a well-constructed plan to minimize risk at the portfolio level. However, the best available data would indicate that a strong tilt to value and a higher representation of small-company stocks in equity portfolios will handsomely reward long-term investors.

Next: Fun with Numbers

In the next chapter, we will shift gears and examine some basic techniques investors should utilize to build an investment strategy that will carry them into the 21st century. I call these techniques “no-brainers”: the magic of compounding, dollar-cost averaging, the joys of tax deferral, and why the best time to invest is the time when you have money.