The Little Book of Value Investing

One of the first studies done on PE ratios and performance was authored by Professor Sanjoy Basu of McMaster University. In “Investment Performance of Common Stocks in Relation to Their Price Earnings Ratios” he looked at stocks listed on the New York Stock exchange from 1957 to 1971. For each year he divided all the listed stocks into five equal groups or quintiles and examined their future performance. He found that value stocks far outperformed their growth peers with a hypothetical $1 million investment growing to more than twice what the higher PE stocks would have achieved.

In his study “Decile Portfolios of the New York Stock Exchange, 1967-1985,” Yale professor Roger Ibbotson ranked all the stocks in 10 equally weighted groups (deciles) according to their P/E ratios. He examined all listed stocks from 1966 to 1984 and found that the cheaper, less popular stocks gave far greater returns. In fact, $1 invested in the cheapest stocks grew to over six times as much as the highest P/E ratio companies and twice as much as those in the middle of the pack. He also looked at how stocks selling at very low multiples of book value compared with growth stocks selling at much higher multiples of asset value. He sorted all the stocks on the New York Stock Exchange (NYSE) into deciles (groupings of 10) for each year and compared the performance of each group. He looked at stocks from 1967 to 1984 and found that stocks priced very low compared to book value outperformed the glamorous names by better than two to one and the market as a whole by better than 75 percent.

One of my favorite studies that I often refer to when discussing the merits of the value versus growth approach to investing was done by Josef Lakonishok, Robert Vishney, and Andrei Shliefer entitled “Contrarion Investment, Extrapolation and Risk.” It ranks all the stocks on both the New York and American Stock Exchanges by P/E ratio in deciles. Each portfolio was held for five years and then sold. They found that across the range of five-year holding periods, the low P/E stocks offered almost twice as much return. Imagine—twice as much in as short a time as five years! They also ranked stocks by price-to-book ratio, also in deciles, and held them for five years. They examined stock prices from 1968 to 1990. Once again, those selling cheapest when compared to book value outperformed by a very wide margin, almost three times the more glamorous stocks, over the five-year holding period. In this same study, they found that the low price-to-book stocks outperformed growth selection in 73 percent of one-year holding periods, 90 percent of three-year, and 100 percent of the five-year holding periods.

One of the most exhaustive examinations of the performance of value stocks was done by Richard Thaler and Werner FM De Bondt, then professors at the University of Wisconsin and Cornell University, respectively. In a 1985 edition of the Journal of Finance they published a paper, “Does the Stock Market Overreact?” that looked at the idea of buying stocks that had gone bump in the night and performed poorly against those that had shone in the sun and performed the best. They examined stock prices starting in December 1932 through 1977, a period covering 46 market years. They looked at the 35 stocks on the New York Stock Exchange that performed the worst over the prior five years against the 35 listed stocks that had been the brightest stars. They compared the results of investing in each basket with an index made up of an equally weighted portfolio of all stocks on the NYSE. They found that, on average, over the next 17 months, the worst stocks gained about 17 percent more than the index, and the bright stars of the past faded quickly returning about 6 percent less than the index over the time period. They also studied holding the portfolios of stocks over three years and found that the prior “bad” stocks continued to far outperform the best past performers. In 1987, Werner FM De Bondt and Richard Thaler further sorted stocks into quintiles (groupings of 20) in their research paper “Further Evidence on Investor Overreaction and Stock Market Seasonality” and found that the stocks selling below book value outperformed the market by more than 40 percent, or almost 9 percent a year.

In a study that compared PE ratios within industry groups, Professors David Goodman and John Peavy of Southern Methodist University ranked stocks within industry groups across more than 100 different industries according to PE ratios. They sorted all the different groups into quintiles and found that even within more specific groupings, the stocks with lower price-to-earnings ratios far outperformed the higher priced stocks. A dollar invested in the bottom quintile of each group, rebalanced annually, grew to over 12 times the highest P/E group and more than twice those with the second lowest P/E ratios.

In their 1992 study “The Cross Section of Expected Stock Returns,” Eugene Fama and Kenneth French examined all nonfinancial stocks included in the Center for Research in Security Prices files, perhaps the most comprehensive database of stock prices. Their study covered the period from 1963 to 1990. They used deciles of stocks ranked according to price-to-book value. The lowest price-to-book value stocks returned almost three times as much as the highest over the 27-year time period. They also looked at holding the portfolios of stocks over three years and found that the prior “bad” stocks continued to far outperform the best past performers.

Global Stocks

As discussed, stocks that have value characteristics perform well in and outside the United States. It was a delight to serendipitously stumble on some value opportunities in Japan when we found insurance companies selling for one-third of book value in the 1980s, but independent research confirms that all around the globe buying stocks selling below book value is a sound idea.

I found of particular interest a study done by Mario Levis, a professor at the School of Management, University of Bath in the United Kingdom, that looked at all the stocks in the London Share database. He looked at stocks from 1961 to 1985 and sorted them into quintiles. Once again, the lower P/E ratio stocks outperformed more exciting growth companies by an extraordinary margin. Over that time period, $1 invested in the lowest price-to-earnings ratio group returned more than five times the highest ratio stocks and double that of those in group two. Performance was three times as high as the companies in the middle group of P/E levels.

In a Morgan Stanley research paper titled “Ben Graham Would Be Proud,” Barton Biggs examined the return for low price-to-book value investing around the world. About 80 percent of the stocks in the study were outside the United States and, once again, the cheap stocks outperformed the more expensive as well as the world market indexes. Nobel Prize winner William Sharpe looked at stocks in Germany, France, Switzerland, the United Kingdom, the United States, and Japan in his Financial Analysts Journal article in1993 titled “International Value and Growth Stocks.” He examined stocks in the S&P 500 in the United States and stocks included in the Morgan Stanley Capital International Index for the other nations. He ranked the stocks every six months. The top 50 percent of stocks in price-to-book value were the growth portfolio, and the 50 percent that sold lowest compared with asset value were the value portfolio. From 1981 through 1992, the value stocks outperformed the growth stocks in each and every country by a substantial margin.

Losers to Winners

Academic research also supports many value investing techniques. James Porterba of the Massachusetts Institute of Technology and Lawrence Summers of Harvard (who later went on to become Secretary of the Treasury and the controversial president of Harvard) in March 1998 published a paper entitled “Mean Reversion in Stock Prices, Evidence and Implications.” They looked at monthly stock prices on the NYSE from 1926 to 1985 to determine if large price increases or decreases were followed by reversals or continued in the original direction. They found that current high investment returns tended to be followed by lower returns, and low investment returns tended to lead to higher performance. In total, they examined stock price reactions in 17 nations including the United States, the United Kingdom, Switzerland, Canada, Japan, Belgium, and the Netherlands. They found that stock prices tended to act the same all over the globe. Today’s worst stocks became tomorrow’s best, and the darlings of the day becoming the spinsters of the next day.

Insider Buying

In “What Has Worked in Investing,” a paper authored by Tweedy, Browne, we examined several key studies that show the tremendous outperformance of stocks with insider buying. We looked at five key studies that showed that stocks with insider buying outperformed the stock market by at least a two-to-one margin. We also looked at several studies that examined insider buying in countries around the world and found that insider buying was predictive of higher returns on a global basis as well. Only a few countries outside the United States require insiders to report transactions, so the information is not as useful.

Fortunately, the field of academic research into financial markets is ongoing and prolific. Many papers have looked at the relationship between insider buying and future returns. One such study by Thomas George and Nejat Seyhun of the University of Michigan looked at over 1 million transactions over a 21-year period. They found that stocks with insider buying outperformed the market by over 6 percent over the next 12 months. The conclusions shared by Professor Seyhun and myself are further borne out in a study by Fuller Thaler Asset Management entitled “Extrapolation Bias, Insider Trading.” In 2001, Andrew Metrick of the Wharton School, Leslie Jeng of Boston University, and Richard Zeckhauser of Harvard released their paper “Estimating the Returns to Insider Trading, a Performance Evaluation Perspective” that confirmed these findings. They looked at insider activity and stock prices from 1975 to 1996 and found that those companies with insiders buying stock outperformed the overall stock market by about the same 6 percent. A 2003 paper by Joseph Piotroski and Darrell Roulstone of the University of Chicago found that insider buying signaled that earnings and cash flow would improve over the next 12 months, leading to a higher stock price. They looked at stocks selling at low earnings multiples or below book value between 1984 and 1995 and found that companies with these characteristics with heavy buying by insiders dramatically and substantially outperformed the market. They also found that those with high multiples and insider selling tended to underperform by a wide margin.

The same held true for companies that bought back stock. One of the first studies into the effect of stock buybacks was done in a Fortune magazine article by Carol Loomis in 1985. She looked at all the stocks in the Value Line universe from 1974 to 1983 and found that companies that bought back stock earned 50 percent more annually than those that did not. A study by University of Illinois professors David Ikenberry and Josef Lakonishok in 1994 looked at companies that bought back stock from 1980 to 1990 and found that over the next four years they outperformed the market by 12.1 percent. For those companies that had other value traits selling at low prices to earnings or book value the professors found that the outperformance was over 45 percent.

Another study done by Professor Ikenberry, with Konan Chan and Inmoo Lee, found that companies that bought back stock between 1980 and 1996 averaged 6 percent more than the market over 12 months and 23 percent over four years. In his most recent study released in 2005, Ikenberry looked at stock buybacks by companies with good earnings and low valuations. He found that, between 1980 and 2000, companies that repurchased stock outperformed by better than 35 percent over four years.

The Latest Look

Lest you think that we rely only on older studies to prove the worth of buying stocks with low prices when compared to earnings, a study from the Brandes Institute, a part of Brandes Asset Management, a venerable value firm, repeated the work of Lakonishok, Vishney, and Shliefer on U.S. stocks, updated it through 2004, and also conducted a similar study of international stocks. Its research showed that the low PE ratio stocks, when tested from 1969 all the way through 2002, have far outperformed the higher priced growth issues. In addition, Professor Lakonishok, along with Louis Chen at the University of Illinois, updated his studies through the year 2002 and found that the value strategy of buying stocks cheaply based on earnings continued to vastly outperform other stocks. They also released a study that examined returns on U.S. stocks from 1986 through 2002. They looked at stocks that they called falling knives, a play on the old Wall Street adage of never trying to catch a falling knife. They defined falling knives as stocks that had fallen 60 percent in price over the prior 12 months. They found that although these stocks did indeed represent a risky proposition with a bankruptcy and failure rate four times that of the market as a whole, as a group they far outperformed the market over one-, two-, and three-year holding periods. Not surprisingly they found that the larger the market capitalization of the company, the higher the outperformance and the less the chance of corporate failure. As one of the chief tenets of our value investing approach is to always maintain a margin of safety, the likelihood of buying into an undercapitalized or poorly financed falling knife would seem to be lessened, giving us an opportunity for further outperformance of the market averages. The Brandes Institute updated this work and took a global look at falling knives in a 2004 paper titled “Falling Knives Around the World,” examining stocks from around the world from 1980 to 2003. As with the previous study, it looked at companies with a market capitalization of over $100 million that had fallen 60 percent in price after the price collapse. Not only did the falling knives in the United States continue to show marked outperformance over the market as a whole, this outperformance held true around the globe.

U.S. Stocks

Global Stocks

Losers to Winners

Insider Buying

The Latest Look