this method at the beginning of the period becomes $626,000* in 27 years. Low relative P/E and low price-to-boolc value also work well, while low relative price-to-dividend easily outperforms the market and lags behind the other three strategies by much less than on an absolute basis.

Are these retums 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 80-90% of the time. The evidence suggests that the relative value has a potent effect in all industries.

So, a new strategy is bom. Lets make that Rule 20 to summarize the concept.

RULE 20

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.

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.

We also tested to see if the relative industry effect is independent of the absolute value effects discussed in the last two chapters, and found that it was. Table 9-2 shows just how different the various contrarian value measurements are for the cheapest and most expensive industries. Price, for example, is only 70% of book value (.7) for the lowest price-to-book value group in the cheapest 20% of industries, and this group

* Dividends are reinvested.

Table 9-1

Industry-Relative Price/Earnings Buy-and-Hold Annual Returns January 1,197a-December 31,1996

$600,000 .

$500,000

$400,000

$300,000

$200,000

$100,000

Average Annual Retums

Low Rel. P/CF 18.4

LowRel.P/E 17.7

Low Rel. P/BV 17.8

Low Rel. P/D 17.0

Market 14.9

Low P/CF

........ a

$ 626,000

Low P/E

$ 572,000

Low P/BV $ 547,000

Low P/D d

$ 492,000

Market

$ 289,000

-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-r-1-1-1-1-1-1-1-r-1

1971 1976 1981 1986 1991 1996

YEAR

♦ Initial Investment of Ten Thousand Dollars in 1970 with Annual Rebalancing

returns 18.4% annually over the length of the study (well above the markets 14.9%). By contrast, the lowest price-to-book value stocks in the highest 20% of industries are 1.4 by this measure, double the price-to-book value ratio of the cheapest 20%. Still, they provide a well-above-average 17.4% retum. Thus, the industry effect is very different than that of simply buying the lowest valued stocks in the averages. So we now have two separate and distinct effects that allow you to beat the market, both strongly backed by statistical evidence.

Why Buy the Cheapest Stocks in an Industry?

Perhaps youre 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. Hundreds of thousands of investors, tired of being battered by bad advice, have moved into index funds, as have large numbers of their institutional

Figure 9-3

Another Flavor of Value: Industry-Relative Strategies

1970-1996

* The P/D is listed as dividend yield.

8. Index funds now account for several hundred billion dollars.

The excidng development about buying the cheapest stoclcs in a group of major industries is that it leads to excellent long-term retums, while allowing you the chance to participate in stocks across the board. Our study indicates the retums dwarf those of an index fund. While its not a strategy for everybody, it will work for investors who can afford to own a 40- or 50-stock portfoho across 30 or 40 major industry groups.

Our research also shows that once the portfolio is in place, as Table 9-1 demonstrated for low P/E, it needs htde 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-results in higher annual retums over five-year holding periods than holding the portfohos for only one year. Retums, in statisdcians terms, do not "decay" over five-year periods; they actually improve.

Why this works is speculative, but it appears that company fortunes do change over time. Industry laggards often tighten their belts, improve their management, and find ways of increasing their market share or developing new products, which results in their continued outperformance of the market for long periods. Analysts and investors slowly change their opinions of these laggards. Now when eamings su rise pleasantly, the market applauds and awards higher prices.

For favored-industry stocks, the process is exactly the opposite. Expectations are too optimistic-so high that even a brilhant management cannot meet them. Something has to go wrong. A glitch occurs because of increased competition (which builds rapidly in high-profit businesses) or a slight lag in the introduction of new products. And so it

Table 9-2

CoNTRARUN Industry Ratios 1970-1996

The price-to-value ratio of the cheapest stocks in the lowest 20% of industries is much lower than the price-to-value ratios of the cheapest stocks in the highest 20% of industries. Yet the retums of the cheapest 20% of stocks in each group are nearly identical.