170 The World of Contrarian Investing RULE 18

Invest equally In 20 to 50 stocks, diversified among 15 or more industries (if your assets are of sufficient size).

Diversification is essential. Returns among individual issues will vary widely, so it is dangerous to rely on only a few companies or industries. By spreading the risk, you have a much better chance of performing in line with the out-of-favor quintiles shown above, rather than substantially above or below this level.

In our 1970-1996 sample, for example, in each of the four contrarian strategies examined, exactly half of the 300 stocks in the lowest quintile performed better than the market over the entire period of the study.

RULE 19

Buy medium- or large-sized stocks listed on the New York Stock Exchange, or only larger companies on Nasdaq or the American Stock Exchange.

Such companies (upon which the studies have been based) are usually subject to less accounting gimmickry than smaller ones, which provides some added measure of protection. Accounting, as we have seen repeatedly, is a devilishly tricky subject and has taken a heavy toll of investors-sophisticates as well as novices.

The larger and medium-sized companies present another advantage to investors: they are more in the public eye. A turnaround in the fortunes of Chrysler (which occurred some years ago) is far more noticeable than a change in the fortunes of some publicly owned five-restaurant steak franchise buried, say, in the sands and winds of Death Valley. Finally, the larger companies have more "staying power." Their failure rate is substantially lower than among smaller or start-up companies.*

Should We Abandon Security Analysis Entirely?

As we have seen, selecting stocks by their contrarian characteristics and placing no reliance on security analysis has given better-than-average returns over long periods of time. Should we, then, consider abandoning security analysis entirely? The evidence weve seen certainly shows it doesnt help much. However, I would not go quite this far (and not just

An Eclectic Approach

In my own application of the low P/E approach, I use the bottom 40% of stocks according to P/Es for stock selection. The two lowest quintiles provide plenty of scope for applying the ancillary selection indicators that follow.

If, after what youve seen, you are brave enough to dabble in security analysis, here are the ancillary indicators I believe most helpful:

Indicator 1. A strong financial position

This is easily determinable for a company from information contained within its financial statements. (The definitions of the appropriate ratios-current assets versus current liabilities,* debt as a percentage of capital structure, interest coverage, etc.-can all be found in any textbook on finance, as well as in material provided free of charge by some of the major brokerage houses.)

A strong financial position will enable the company to sail unimpaired through periods of operating difficulties, which contrarian companies sometimes experience. Financial strength is also important in deciding whether a companys dividends can be maintained or increased.

* Previously defined in chapter 3, and in glossary.

t Martin S. Fridsons book Financial Statement Analysis (Wiley Press, 1995) provides a brief introduction to co orate accounting and outlines most of the important financial ratios the average investor would require.

because Ive been thoroughly steeped in the doctrines of the Old Church).

I believe parts of it can be valuable within a contrarian framework. Contrarian methods eliminate or downgrade those aspects of traditional analysis, such as forecasting, that have been shown to be consistently error-prone. By recognizing the limitations of security analysis, you can, I believe, apply it to achieve even better results within the contrarian approach.

In the next sections, Ill attempt to show you how five fundamental indicators can be used to supplement the three A-B-C rules of contrarian selection we just looked at. Following this analysis, well examine other contrarian methods that do not depend on security evaluation at all. These, too, should provide above-average market results. After reviewing the various methods, you can choose which suits you best.

Indicator 5. A higher rate of earnings growth than the S&P

500 in the immediate past, and the likelihood that it will not plummet in the near future

Such future estimates are not an attempt to pinpoint eamings, but only their general direction. Remember that we are dealing with stocks in the bottom quintiles, for which only the worst is expected. Unlike conventional forecasting methods, we do not require precise eamings estimates, but need merely note their direction, and only for short periods, usually about a year or so.

In my initial contrarian work, I was more of a purist on this subject. I thought, since contrarian strategies worked at least in part because of analysts errors, why bother with forecasting at all? Some rather harsh experiences have caused me to modify this position.

If, for example, the Street estimates that a companys eamings are likely to be down for some time, I would not msh in to buy, no matter how positive my indicators appeared to be. Often, as weve seen, analysts are overoptimistic. All too frequently, an estimated moderate decline in eamings tums into a drop off a cliff. This was the case with Chrysler in the early 1990s when the initial estimates were for relatively small declines. The actual drop in eamings brought the company to the brink of Chapter 11. The stock fell over 80% before rebounding. If eamings are declining, stock prices often follow them down for some time.

The important distinction between forecasting the general direction of eamings and trying to derive precise eamings estimates is that the former method is far simpler and thus more likely to succeed.

Indicator 4. Earnings estimates should always lean to the conservative side

This ties in to Graham and Dodds "margin of safety" principle. By relying on general directional forecasts and keeping them conservative, you are reducing the chance of error even furtiier. If you do this, and the company still looks as though its eamings will grow more quickly than the S&P for a year or so, you have a potentially rewarding investment.

Indicator 2. As many favorable operating and financial ratios

as possible

This helps to ensure that there are no structural flaws in the company. The definition of such ratios again can be found in standard financial textbooks.