by David Dreman
Though the strategies are simple and easy to use, the influence of immediate events is very powerful. We looked at why most people can’t shake off these influences and at some of the principles that can help you to do so in Part I. Next let’s look at the new, sometimes alien, world we face as investors and the strategies we can use to get through it.
Chapter 12
Contrarian Strategies Within Industries
WE’VE SEEN THAT the four contrarian strategies discussed in the previous chapter held up better than the market through both the Internet bubble and crash and the 2007–2008 market debacle.
Next we’ll examine a new strategy that works on exactly the same behavioral principles as contrarian stocks but in a very different way. This method will allow you to participate in virtually every major industry in a manner similar to an index fund. Unlike an index fund, however, it should provide well-above-market returns. As we’ve already documented, investors are entirely too confident of their ability to forecast which stocks will win and which will lose, and fashions in the marketplace play a powerful role in drawing people to popular stocks. Does that extend beyond absolute “best” and “worst” stocks, as measured by absolute contrarian indicators in the previous chapter? It should.
A New Contrarian Strategy
If trends and fashions exist in the marketplace as a whole, it is reasonable from a psychological perspective to expect that they exist within specific industries.1 Analysts’ research, expert opinion, current prospects, and a host of other variables should work on investors’ expectations almost the same within industries as in the overall market. The result again should be expectations set too high for favored stocks within an industry and too low for out-of-favor stocks. Thus, Apple might be a favorite in the computer and peripheral industry, while Dell might be thought of as a laggard. Similarly, Chubb might be a favorite in the insurance industry, while Hartford is unloved and unwanted. Overpricing and underpricing of favored and unfavored stocks within value and growth industries would thus appear to be a natural extension of contrarian strategies. Sounds great in theory, but, unlike our academic friends, we’re going to ask a rude question: does it work?
First we should ask how well the contrarian industry strategy does in the difficult environment we have recently experienced. The harsher the environment, the more accurate your judgment must be. Contrarian industry or relative strategies, like the absolute contrarian strategies we looked at in the preceding chapter, should also reduce the number of judgment errors significantly by not allowing us to overpay for stocks and not using disproved beta or other volatility measures, as well as by providing other safety checks.
To get the answer, first in collaboration with Eric Lufkin from 1970 to 19962 and then in a follow-up study with Vladimira Ilieva from January 1, 1995, through December 31, 2010, I examined the 1,500 largest companies on the Compustat database by market size. In the later study we divided the 1,500 stocks into the sixty-eight industries as classified by Standard & Poor’s Global Industry Classification Standard (GICS) for the period measured.3 The most favored stocks consisted of the 20 percent of companies in each industry with the highest P/E ratios, price-to-cash-flow ratios, and price-to-book-value ratios or lowest yields. The most unpopular were the 20 percent of stocks in each industry that had the lowest ratios by the first three measurements and the highest yield by the fourth. Returns were calculated in the same way for the remaining 60 percent of stocks in the middle quintiles.
If we used the industry strategy and took low P/E as our example, the lowest multiple in one industry, such as commercial banking, might be 10; in another industry, such as biotechnology, it might be 40. Yet, if we were right, the lowest-P/E stocks in both industries should provide well-above-market returns. Remember, in this strategy, we speak of relative P/E or the lowest 20 percent of P/Es (or price to cash flow or price to book value, or highest yield) within an industry, versus using the absolute lowest P/E measurement for the entire market, as we do for the four other contrarian strategies.
Figures 12-1 and 12-2 give the results of our most recent study for low price to earnings and low price to cash flow. Looking at Figure 12-1, for example, we see that the lowest 20 percent of stocks in each industry as measured by P/E, the first bar, returned 13.9 percent, the highest (second bar) returned 11.3 percent, and the market returned 10.2 percent. The lowest-price-earnings group outperformed the highest by 2.6 percent annually over the life of the study. Most important, the low-P/E stocks in each industry outperformed the market by 3.7 percent annually.
Figure 12-2 shows that the total annual return with the price-to-cash-flow strategy is 15.1 percent, which is higher than the low-P/E return. The lowest-price-to-cash-flow stocks in each industry group outperformed both the highest-price-to-cash-flow stocks and the market by more than the low-P/E group did in Figure 12-1.
Table 12-1 shows the returns of buying low-P/E stocks by this method and holding them for periods of two, three, five, and eight years. Again, the results of investing in the most out-of-favor stocks within industries are similar to those of buying the most unpopular stocks overall; however, the industry returns are somewhat lower. As Table 12-1 indicates, the returns of the laggards continue to outperform the market but fall off more rapidly against it over longer periods of time than do absolute contrarian strategies. It is best to rebalance portfolios annually when this strategy is used. The returns of price to cash flow and price to book value (not shown) are also similar.
Figure 12-3 demonstrates how soundly the relative (or industry) contrarian strategies beat the averages over the full sixteen years of the study. Low price-to-cash flow does the best. Ten thousand dollars invested by this method at the beginning of the period becomes $95,000 sixteen years later, outperforming the market by 102 percent.*69 Low relative price-to-book value and low P/E also work, while high yield also outperforms the market but lags behind the three other strategies.
Figure 12-4 once again shows that the market outperforms the highest industry-relative strategies, with the exception of high P/E, in a manner similar to how absolute contrarian strategies outgunned “best”-stock metrics. The resemblance in performance between the absolute strategies we looked at in previous chapters and the relative strategies we are looking at now is extraordinary. All of the “worst” industry-relative strategies in Figure 12-3 outperform the market.
Are these returns simply due to the superior performance of industries loaded with unloved stocks? No. The most out-of-favor stocks in an industry, regardless of whether they were dirt cheap or highly priced, outperformed the most popular stocks in each group and the market average. Low P/E beats high P/E, for example, by almost 50 percent. The evidence suggests that the relative value has a potent effect in all industries.4
So a new strategy is born. Let’s make that a Psychological Guideline to summarize the concept:
PSYCHOLOGICAL GUIDELINE 26: Buy the least expensive stocks within an industry, as determined by the four contrarian strategies, regardless of how high or low the general price of the industry group is.
This strategy will beat the market handily most of the time. The psychological reasons are identical to those behind the contrarian strategies we looked at previously. It appears that the driving force behind these surprising results is Affect, as it is with absolute contrarian strategies. Regardless of how high or low the stock is valued absolutely, it has a nearly identical influence on investors in choosing the most popular and least popular stocks in an industry, as it does on the most popular and least popular stocks in the market itself. The industry-relative price-to-value strategies appear to rely more on Affect than on fundamentals in the two periods ranging over the thirty-seven years the studies covered.
Table 12-2 shows just how different the various contrarian value measurements are for the cheapest and most expensive industries. Price is only 90 percent of book value (0.9), for example, for the lowest-price-to-book-value group (PBV) in the cheapest 20 percent of indust
ries, and this group returned 13.4 percent annually over the length of the study (above the market’s 10.2 percent). By contrast, the lowest-PBV stocks in the highest 20 percent of industries are 2.2 times PBV by this measure. This is more than 2.4 times the price-to-book-value ratio of the cheapest 20 percent. Still, the lowest PBVs in the highest-value group provide an above-market 14.1 percent return. A glance at Table 12-2 also shows that the returns of the lowest relative strategies outperform those of the highest, with the exception of PBV. So we now have two separate and distinct effects that allow you to beat the market, both of them strongly backed by statistical evidence.
Why Buy the Cheapest Stocks in an Industry?
Perhaps you’re wondering, “What is the advantage of buying the cheapest stocks within an industry rather than the cheapest stocks overall?” There are several reasons why it can make good sense. As we just saw, millions of investors, tired of being battered by bad advice, have moved into index funds, as have large numbers of their institutional counterparts. Index funds, including exchange-traded funds (ETFs), now account for over $1 trillion of investment.
The new research about buying the cheapest stocks in a group of major industries leads to excellent long-term returns, while allowing you the chance to participate in stocks across the board—any investor can use this formula. Our study indicates that the returns are well ahead of those of an index fund. Though it’s not a strategy for everybody, it will work for investors who can afford to own a broadly diversified portfolio of stocks: two in each industry across sixty-eight industries.
This is not an index fund as such, because it does not own hundreds of stocks, but it does own the most unfavored stocks in each of the sixty-eight major industries. As we saw, the returns have been well ahead of the S&P 500 over time. It can be used in lieu of an index-plus fund, a fund that attempts to outperform the S&P 500 with a somewhat similar portfolio. Unlike an index-plus fund, industry-relative portfolios have long records of outperforming this way.
Our research also shows that once a portfolio is in place, as Table 12-2 demonstrates for contrarian stocks, it needs some fine-tuning. Buying a portfolio of the lowest-valued stocks in an industry and holding it without any changes—regardless of the contrarian strategy you prefer—unlike buying absolute contrarian strategies, results in some “decay” in results in statisticians’ terms, but they are still well ahead of the market. You are better off rebalancing the portfolio every year or two.
Though we continue to research the industry strategy work, we believe, as previously stated, it is caused by Affect and other psychological influences.
It also appears in relative industry strategies, as in absolute ones, that company fortunes change over time, as Graham and Dodd noted in Security Analysis.5 Industry laggards often tighten their belts, improve their management, and find ways of increasing their market share or developing new products, resulting in their outperformance of the market for fairly long periods.
However, such changes could certainly be Affect-related. Analysts and investors slowly change their opinions of these laggards. Now, when companies’ performance surprises pleasantly, the market applauds and awards them higher stock prices. For favored-industry stocks, the process is exactly the opposite: expectations are too optimistic—so high that even a brilliant management cannot meet them. Something has to go wrong and usually does.
Moving back from the “why” to the “how,” let’s look at another reason this approach can help you. Buying the lowest-valued stocks in each major industry opens a much larger investment universe than is available with an “absolute” contrarian strategy. Investing in the absolutely cheapest stocks, no matter which of the previous methods you pick, gives the investor only the bottom 20 percent of stocks in the marketplace from which to select. With an industry-relative strategy, you can get a crack at every industry in the entire market.6
The advantage of the contrarian industry strategy is that you have far more diversification by industry than you do in the original contrarian strategies. This diversification should protect you from the underperformance that occurs when the most out-of-favor stocks and industries in the market are taboo. Thus, if industries such as communications equipment or biotechnology are headed for the stars, you will not feel left out. You will also not be positioned only in the most disliked groups, which may underperform for months or sometimes years.
Although the returns of the industry-relative strategies are below those of the absolute strategies, they still significantly outperform the market. As we’ve learned, it is difficult for individual investors, and even more for professionals, to take unpopular positions for long periods, even if they are right in the end.
With the inane focus on quarterly performance by too many consultants and clients continuing to increase, a manager who lags behind for too long can easily get the ax. True, he or she may prove right, but all too frequently it is “dead right.” For individuals, the psychological pressure to run for the hills is difficult or impossible to resist. After all, it is psychological pressures that keep people from following contrarian strategies in the first place.
The Defensive Team
“Do I still get to eat my cake and have it too?” many investors might ask of the industry-relative strategies. “We see that they provide impressive returns over time, but how good are they in down markets?” Pretty good, as Figure 12-5a demonstrates. As with the absolute contrarian strategies in Figure 10-5a, we measured the bear market returns for all the down market periods. Relative value, like the other four absolute contrarian strategies, outperformed the market in the twenty down quarters in this study. The high-yield method again shines, declining only 5.1 percent quarterly on average in down quarters (the best of the litter) against –8.1 percent for the overall market. Similarly, Figure 12-5b shows that all industry-relative strategies outperformed the market in the forty-four quarters of our study when stocks were heading up.
The results are both robust and remarkably similar to those that Eric Lufkin and I obtained in Contrarian Investment Strategies: The Next Generation in 1998 (for 1970–1996), although the databases and time periods are very different. The main point is that these strategies are time-tested and work.
Additional Portfolio Considerations
WHERE DO I GET MY STATISTICS?
You might ask how one determines quintiles. Brokerage firms, advisory services, and financial publications often advertise long lists of contrarian stocks (the Value Line Survey, for example, presents weekly tables of the hundred lowest P/E ratios, price-to-cash-flow ratios, price-to-book-value ratios, and highest-yielding stocks of the 1,700 companies in its universe). Two other statistical database providers, the American Association of Individual Investors (800-428-2244) and Investors Alliance (866-627-9090), also offer discs of company information that allow screening for a minimal fee.
To select contrarian stocks on your own, some simple rules should suffice. First, take a broad market index, like the S&P 500, for which the current P/E, price-to-cash-flow, and price-to-book-value ratios and dividend yield can easily be found in a variety of sources. The current P/E ratio of the S&P 500, in September 2011, was about 12х, its price-to-cash-flow ratio 7х, its price-to-book-value ratio 1.9х, and its yield 2.3 percent.
Pick well-established companies. A rule of thumb might be to use a 20 percent discount or more from the S&P 500 for any of the first three measures and a yield of at least 1 percent above the market for the fourth. The deeper the discount from the S&P 500, the further into contrarian land you go.
There is nothing magical about picking the bottom 20 percent by any of the first three measures. It’s simply a good cutoff point for a computer. As we found in virtually all of the contrarian studies, this group consistently outperformed the market.
A simple method should work fairly effectively if you do not have the quintiles for these strategies. How can you find the ratios? Your broker should be able to procure all the ratios for you very easily. The Va
lue Line Investment Survey, among others, normally runs a list of these metrics weekly, as well as one of high-yielding stocks. Yields are also published daily in the financial section of any large newspaper. Once again, you can compare them with the average yield of the S&P 500, which is readily available from a number of sources. As indicated, the price-to-cash-flow and price-to-book-value ratios can be obtained at low cost from the Morningstar or Value Line Web sites or a variety of other services.
Though I don’t believe contrarian investing need necessarily be the final answer to stock selection for everyone, it has consistently performed better in both good and bad markets. It is also the only strategy that I know of that effectively and systematically checks investor overreaction—by far the largest and most important source of investor error.
WHAT CONTRARIAN STRATEGIES WON’T DO FOR YOU
Whether you opt for the eclectic contrarian strategy or take the contrarian bit between your teeth and attack the market without any form of security analysis, keep in mind that the strategies are relative rather than absolute. This means that they won’t help you decide when to get into or out of particular stocks. But no strategy that I am aware of does so successfully. We saw in chapter 4 that the technical analysis that claimed to be able to do so crashed and burned. Whether the market is high or low, you will receive no warning signals to sell in the first place or buy in the second.
What contrarian strategies should do is give you the best relative opportunities for your capital, particularly in the present difficult environment. This means that in a falling market, your stocks ought to decline less than the averages, and in a rising market, they should perform somewhat better. But remember that averages are deceptive. Even if contrarian stocks provide better returns in most bear markets, they don’t do so for all bear markets. Similarly, though contrarian strategies provide far-above-average returns over time, they don’t do so every time. Think of yourself as the owner of a casino. The odds in your favor are extremely high, but in this financial casino, unlike a regular casino, you will have a few years of poor results, as was the case from 1997 to 2000 or 2007 to 2008, eventually followed by some years of staggering returns. Your greatest enemy is the psychology highlighted in the first section of the book. If you can control the strong urge to play the winners, your chances of success even in an alien environment are high.