The Defensive Investor and Common Stocks

Investment Merits of Common Stocks

In our first edition (1949) we found it necessary at this point to insert a long exposition of the case for including a substantial common-stock component in all investment portfolios.* Common stocks were generally viewed as highly speculative and therefore unsafe; they had declined fairly substantially from the high levels of 1946, but instead of attracting investors to them because of their reasonable prices, this fall had had the opposite effect of undermining confidence in equity securities. We have commented on the converse situation that has developed in the ensuing 20 years, whereby the big advance in stock prices made them appear safe and profitable investments at record high levels which might actually carry with them a considerable degree of risk.t

The argument we made for common stocks in 1949 turned on

* At the beginning of 1949, the average annual return produced by stocks over the previous 20 years was 3.1%, versus 3.9% for long-term Treasury bonds-meaning that $10,000 invested in stocks would have grown to $18,415 over that period, while the same amount in bonds would have turned into $21,494. Naturally enough, 1949 turned out to be a fabulous time to buy stocks: Over the next decade, the Standard & Poor's 500-stock index gained an average of 20.1% per year, one of the best long-term returns in the history of the U.S. stock market.

t Graham's earlier comments on this subject appear on pp. 19-20. Just imagine what he would have thought about the stock market of the late 1990s, in which each new record-setting high was considered further "proof" that stocks were the riskless way to wealth!

two main points. The first was that they had offered a considerable degree of protection against the erosion of the investor's dollar caused by inflation, whereas bonds offered no protection at all. The second advantage of common stocks lay in their higher average return to investors over the years. This was produced both by an average dividend income exceeding the yield on good bonds and by an underlying tendency for market value to increase over the years in consequence of the reinvestment of undistributed profits.

While these two advantages have been of major importance— and have given common stocks a far better record than bonds over the long-term past—we have consistently warned that these benefits could be lost by the stock buyer if he pays too high a price for his shares. This was clearly the case in 1929, and it took 25 years for the market level to climb back to the ledge from which it had abysmally fallen in 1929-1932.* Since 1957 common stocks have once again, through their high prices, lost their traditional advantage in dividend yield over bond interest rates.t It remains to

* The Dow Jones Industrial Average closed at a then-record high of 381.17 on September 3, 1929. It did not close above that level until November 23, 1954-more than a quarter of a century later—when it hit 382.74. (When you say you intend to own stocks "for the long run," do you realize just how long the long run can be-or that many investors who bought in 1929 were no longer even alive by 1954?) However, for patient investors who reinvested their income, stock returns were positive over this otherwise dismal period, simply because dividend yields averaged more than 5.6% per year. According to professors Elroy Dimson, Paul Marsh, and Mike Staunton of London Business School, if you had invested $1 in U.S. stocks in 1900 and spent all your dividends, your stock portfolio would have grown to $198 by 2000. But if you had reinvested all your dividends, your stock portfolio would have been worth $16,797! Far from being an afterthought, dividends are the greatest force in stock investing.

t Why do the "high prices" of stocks affect their dividend yields? A stock's yield is the ratio of its cash dividend to the price of one share of common stock. If a company pays a $2 annual dividend when its stock price is $100 per share, its yield is 2%. But if the stock price doubles while the dividend stays constant, the dividend yield will drop to 1%. In 1959, when the trend Graham spotted in 1957 became noticeable to everyone, most Wall Street be seen whether the inflation factor and the economic-growth factor will make up in the future for this significantly adverse development.

It should be evident to the reader that we have no enthusiasm for common stocks in general at the 900 DJIA level of late 1971. For reasons already given* we feel that the defensive investor cannot afford to be without an appreciable proportion of common stocks in his portfolio, even if he must regard them as the lesser of two evils—the greater being the risks attached to an all-bond holding.

Rules for the Common-Stock Component

The selection of common stocks for the portfolio of the defensive investor should be a relatively simple matter. Here we would suggest four rules to be followed:

1. There should be adequate though not excessive diversification. This might mean a minimum of ten different issues and a maximum of about thirty.t

2. Each company selected should be large, prominent, and conservatively financed. Indefinite as these adjectives must be, their general sense is clear. Observations on this point are added at the end of the chapter.

3. Each company should have a long record of continuous dividend payments. (All the issues in the Dow Jones Industrial Aver pundits declared that it could not possibly last. Never before had stocks yielded less than bonds; after all, since stocks are riskier than bonds, why would anyone buy them at all unless they pay extra dividend income to compensate for their greater risk? The experts argued that bonds would outyield stocks for a few months at most, and then things would revert to "normal." More than four decades later, the relationship has never been normal again; the yield on stocks has (so far) continuously stayed below the yield on bonds.

t For another view of diversification, see the sidebar in the commentary on Chapter 14 (p. 368).

age met this dividend requirement in 1971.) To be specific on this point we would suggest the requirement of continuous dividend payments beginning at least in 1950.*

4. The investor should impose some limit on the price he will pay for an issue in relation to its average earnings over, say, the past seven years. We suggest that this limit be set at 25 times such average earnings, and not more than 20 times those of the last twelve-month period. But such a restriction would eliminate nearly all the strongest and most popular companies from the portfolio. In particular, it would ban virtually the entire category of "growth stocks," which have for some years past been the favorites of both speculators and institutional investors. We must give our reasons for proposing so drastic an exclusion.

Growth Stocks and the Defensive Investor

The term "growth stock" is applied to one which has increased its per-share earnings in the past at well above the rate for common stocks generally and is expected to continue to do so in the future. (Some authorities would say that a true growth stock should be expected at least to double its per-share earnings in ten years—i.e., to increase them at a compounded annual rate of over 7.1%.)+ Obviously stocks of this kind are attractive to buy and to own, provided the price paid is not excessive. The problem lies there, of

* Today's defensive investor should probably insist on at least 10 years of continuous dividend payments (which would eliminate from consideration only one member of the Dow Jones Industrial Average-Microsoft-and would still leave at least 317 stocks to choose from among the S & P 500 index). Even insisting on 20 years of uninterrupted dividend payments would not be overly restrictive; according to Morgan Stanley, 255 companies in the S & P 500 met that standard as of year-end 2002. t The "Rule of 72" is a handy mental tool. To estimate the length of time an amount of money takes to double, simply divide its assumed growth rate into 72. At 6%, for instance, money will double in 12 years (72 divided by 6 = 12). At the 7.1% rate cited by Graham, a growth stock will double its earnings in just over 10 years (72/7.1 = 10.1 years).

course, since growth stocks have long sold at high prices in relation to current earnings and at much higher multiples of their average profits over a past period. This has introduced a speculative element of considerable weight in the growth-stock picture and has made successful operations in this field a far from simple matter.

The leading growth issue has long been International Business Machines, and it has brought phenomenal rewards to those who bought it years ago and held on to it tenaciously. But we have already pointed out* that this "best of common stocks" actually lost 50% of its market price in a six-months' decline during 1961-62 and nearly the same percentage in 1969-70. Other growth stocks have been even more vulnerable to adverse developments; in some cases not only has the price fallen back but the earnings as well, thus causing a double discomfiture to those who owned them. A good second example for our purpose is Texas Instruments, which in six years rose from 5 to 256, without paying a dividend, while its earnings increased from 40 cents to $3.91 per share. (Note that the price advanced five times as fast as the profits; this is characteristic of popular common stocks.) But two years later the earnings had dropped off by nearly 50% and the price by four-fifths, to 49.+

The reader will understand from these instances why we regard growth stocks as a whole as too uncertain and risky a vehicle for the defensive investor. Of course, wonders can be accomplished with the right individual selections, bought at the right levels, and later sold after a huge rise and before the probable decline. But the average investor can no more expect to accomplish this than to find money growing on trees. In contrast we think that the group of

* Graham makes this point on p. 73.

t To show that Graham's observations are perennially true, we can substitute Microsoft for IBM and Cisco for Texas Instruments. Thirty years apart, the results are uncannily similar: Microsoft's stock dropped 55.7% from 2000 through 2002, while Cisco's stock—which had risen roughly 50-fold over the previous six years—lost 76% of its value from 2000 through 2002. As with Texas Instruments, the drop in Cisco's stock price was sharper than the fall in its earnings, which dropped just 39.2% (comparing the three-year average for 1997-1999 against 2000-2002). As always, the hotter they are, the harder they fall.

large companies that are relatively unpopular, and therefore obtainable at reasonable earnings multipliers,* offers a sound if unspectacular area of choice by the general public. We shall illustrate this idea in our chapter on portfolio selection.

Portfolio Changes

It is now standard practice to submit all security lists for periodic inspection in order to see whether their quality can be improved. This, of course, is a major part of the service provided for clients by investment counselors. Nearly all brokerage houses are ready to make corresponding suggestions, without special fee, in return for the commission business involved. Some brokerage houses maintain investment services on a fee basis.

Presumably our defensive investor should obtain—at least once a year—the same kind of advice regarding changes in his portfolio as he sought when his funds were first committed. Since he will have little expertness of his own on which to rely, it is essential that he entrust himself only to firms of the highest reputation; otherwise he may easily fall into incompetent or unscrupulous hands. It is important, in any case, that at every such consultation he make clear to his adviser that he wishes to adhere closely to the four rules of common-stock selection given earlier in this chapter. Incidentally, if his list has been competently selected in the first instance, there should be no need for frequent or numerous changes.t

* "Earnings multiplier" is a synonym for P/E or price/earnings ratios, which measure how much investors are willing to pay for a stock compared to the profitability of the underlying business. (See footnote t on p. 70 in Chapter 3.) t Investors can now set up their own automated system to monitor the quality of their holdings by using interactive "portfolio trackers" at such websites as,,, and Graham would, however, warn against relying exclusively on such a system; you must use your own judgment to supplement the software.

Dollar-Cost Averaging

The New York Stock Exchange has put considerable effort into popularizing its "monthly purchase plan," under which an investor devotes the same dollar amount each month to buying one or more common stocks. This is an application of a special type of "formula investment" known as dollar-cost averaging. During the predominantly rising-market experience since 1949 the results from such a procedure were certain to be highly satisfactory, especially since they prevented the practitioner from concentrating his buying at the wrong times.

In Lucile Tomlinson's comprehensive study of formula investment plans,1 the author presented a calculation of the results of dollar-cost averaging in the group of stocks making up the Dow Jones industrial index. Tests were made covering 23 ten-year purchase periods, the first ending in 1929, the last in 1952. Every test showed a profit either at the close of the purchase period or within five years thereafter. The average indicated profit at the end of the 23 buying periods was 21.5%, exclusive of dividends received. Needless to say, in some instances there was a substantial temporary depreciation at market value. Miss Tomlinson ends her discussion of this ultrasimple investment formula with the striking sentence: "No one has yet discovered any other formula for investing which can be used with so much confidence of ultimate success, regardless of what may happen to security prices, as Dollar Cost Averaging."

It may be objected that dollar-cost averaging, while sound in principle, is rather unrealistic in practice, because few people are so situated that they can have available for common-stock investment the same amount of money each year for, say, 20 years. It seems to me that this apparent objection has lost much of its force in recent years. Common stocks are becoming generally accepted as a necessary component of a sound savings-investment program. Thus, systematic and uniform purchases of common stocks may present no more psychological and financial difficulties than similar continuous payments for United States savings bonds and for life insurance—to which they should be complementary. The monthly amount may be small, but the results after 20 or more years can be impressive and important to the saver.

The Investor's Personal Situation

At the beginning of this chapter we referred briefly to the position of the individual portfolio owner. Let us return to this matter, in the light of our subsequent discussion of general policy. To what extent should the type of securities selected by the investor vary with his circumstances? As concrete examples representing widely different conditions, we shall take: (1) a widow left $200,000 with which to support herself and her children; (2) a successful doctor in mid-career, with savings of $100,000 and yearly accretions of $10,000; and (3) a young man earning $200 per week and saving $1,000 a year.*

For the widow, the problem of living on her income is a very difficult one. On the other hand the need for conservatism in her investments is paramount. A division of her fund about equally between United States bonds and first-grade common stocks is a compromise between these objectives and corresponds to our general prescription for the defensive investor. (The stock component may be placed as high as 75% if the investor is psychologically prepared for this decision, and if she can be almost certain she is not buying at too high a level. Assuredly this is not the case in early 1972.)

We do not preclude the possibility that the widow may qualify as an enterprising investor, in which case her objectives and methods will be quite different. The one thing the widow must not do is to take speculative chances in order to "make some extra income." By this we mean trying for profits or high income without the necessary equipment to warrant full confidence in overall success. It would be far better for her to draw $2,000 per year out of her principal, in order to make both ends meet, than to risk half of it in poorly grounded, and therefore speculative, ventures.

The prosperous doctor has none of the widow's pressures and compulsions, yet we believe that his choices are pretty much the same. Is he willing to take a serious interest in the business of investment? If he lacks the impulse or the flair, he will do best to

* To update Graham's figures, take each dollar amount in this section and multiply it by five.

accept the easy role of the defensive investor. The division of his portfolio should then be no different from that of the "typical" widow, and there would be the same area of personal choice in fixing the size of the stock component. The annual savings should be invested in about the same proportions as the total fund.

The average doctor may be more likely than the average widow to elect to become an enterprising investor, and he is perhaps more likely to succeed in the undertaking. He has one important handicap, however—the fact that he has less time available to give to his investment education and to the administration of his funds. In fact, medical men have been notoriously unsuccessful in their security dealings. The reason for this is that they usually have an ample confidence in their own intelligence and a strong desire to make a good return on their money, without the realization that to do so successfully requires both considerable attention to the matter and something of a professional approach to security values.

Finally, the young man who saves $1,000 a year—and expects to do better gradually—finds himself with the same choices, though for still different reasons. Some of his savings should go automatically into Series E bonds. The balance is so modest that it seems hardly worthwhile for him to undergo a tough educational and temperamental discipline in order to qualify as an aggressive investor. Thus a simple resort to our standard program for the defensive investor would be at once the easiest and the most logical policy.

Let us not ignore human nature at this point. Finance has a fascination for many bright young people with limited means. They would like to be both intelligent and enterprising in the placement of their savings, even though investment income is much less important to them than their salaries. This attitude is all to the good. There is a great advantage for the young capitalist to begin his financial education and experience early. If he is going to operate as an aggressive investor he is certain to make some mistakes and to take some losses. Youth can stand these disappointments and profit by them. We urge the beginner in security buying not to waste his efforts and his money in trying to beat the market. Let him study security values and initially test out his judgment on price versus value with the smallest possible sums.

Thus we return to the statement, made at the outset, that the kind of securities to be purchased and the rate of return to be sought depend not on the investor's financial resources but on his financial equipment in terms of knowledge, experience, and temperament.

Note on the Concept of "Risk"

It is conventional to speak of good bonds as less risky than good preferred stocks and of the latter as less risky than good common stocks. From this was derived the popular prejudice against common stocks because they are not "safe," which was demonstrated in the Federal Reserve Board's survey of 1948. We should like to point out that the words "risk" and "safety" are applied to securities in two different senses, with a resultant confusion in thought.

A bond is clearly proved unsafe when it defaults its interest or principal payments. Similarly, if a preferred stock or even a common stock is bought with the expectation that a given rate of dividend will be continued, then a reduction or passing of the dividend means that it has proved unsafe. It is also true that an investment contains a risk if there is a fair possibility that the holder may have to sell at a time when the price is well below cost.

Nevertheless, the idea of risk is often extended to apply to a possible decline in the price of a security, even though the decline may be of a cyclical and temporary nature and even though the holder is unlikely to be forced to sell at such times. These chances are present in all securities, other than United States savings bonds, and to a greater extent in the general run of common stocks than in senior issues as a class. But we believe that what is here involved is not a true risk in the useful sense of the term. The man who holds a mortgage on a building might have to take a substantial loss if he were forced to sell it at an unfavorable time. That element is not taken into account in judging the safety or risk of ordinary real-estate mortgages, the only criterion being the certainty of punctual payments. In the same way the risk attached to an ordinary commercial business is measured by the chance of its losing money, not by what would happen if the owner were forced to sell.

In Chapter 8 we shall set forth our conviction that the bona fide investor does not lose money merely because the market price of his holdings declines; hence the fact that a decline may occur does not mean that he is running a true risk of loss. If a group of well-selected common-stock investments shows a satisfactory overall return, as measured through a fair number of years, then this group investment has proved to be "safe." During that period its market value is bound to fluctuate, and as likely as not it will sell for a while under the buyer's cost. If that fact makes the investment "risky," it would then have to be called both risky and safe at the same time. This confusion may be avoided if we apply the concept of risk solely to a loss of value which either is realized through actual sale, or is caused by a significant deterioration in the company's position—or, more frequently perhaps, is the result of the payment of an excessive price in relation to the intrinsic worth of the security.2

Many common stocks do involve risks of such deterioration. But it is our thesis that a properly executed group investment in common stocks does not carry any substantial risk of this sort and that therefore it should not be termed "risky" merely because of the element of price fluctuation. But such risk is present if there is danger that the price may prove to have been clearly too high by intrinsic-value standards—even if any subsequent severe market decline may be recouped many years later.

Note on the Category of "Large, Prominent, and Conservatively Financed Corporations"

The quoted phrase in our caption was used earlier in the chapter to describe the kind of common stocks to which defensive investors should limit their purchases—provided also that they had paid continuous dividends for a considerable number of years. A criterion based on adjectives is always ambiguous. Where is the dividing line for size, for prominence, and for conservatism of financial structure? On the last point we can suggest a specific standard that, though arbitrary, is in line with accepted thinking. An industrial company's finances are not conservative unless the common stock (at book value) represents at least half of the total capitalization, including all bank debt.3 For a railroad or public utility the figure should be at least 30%.

The words "large" and "prominent" carry the notion of substantial size combined with a leading position in the industry. Such companies are often referred to as "primary"; all other common stocks are then called "secondary," except that growth stocks are ordinarily placed in a separate class by those who buy them as such. To supply an element of concreteness here, let us suggest that to be "large" in present-day terms a company should have $50 million of assets or do $50 million of business.* Again to be "prominent" a company should rank among the first quarter or first third in size within its industry group.

It would be foolish, however, to insist upon such arbitrary criteria. They are offered merely as guides to those who may ask for guidance. But any rule which the investor may set for himself and which does no violence to the common-sense meanings of "large" and "prominent" should be acceptable. By the very nature of the case there must be a large group of companies that some will and others will not include among those suitable for defensive investment. There is no harm in such diversity of opinion and action. In fact, it has a salutary effect upon stock-market conditions, because it permits a gradual differentiation or transition between the categories of primary and secondary stock issues.

* In today's markets, to be considered large, a company should have a total stock value (or "market capitalization") of at least $10 billion. According to the online stock screener at, that gave you roughly 300 stocks to choose from as of early 2003.

0 0

Post a comment