As was shown in our chapter on Trade Cycles and Price Fluctuations, when commodity prices fall there is a tendency for fixed rate investments to rise, because money becomes plentiful and cheap and pushes them up, but for the profits and consequently the prices of ordinary shares to be diminished, because the lower prices make the finished article fetch less than would otherwise have been the case by the time it has gone through the process of manufacture, while food and materials are losing value all the time that they are moving through the hands of merchants and middlemen. If these tendencies act with sufficient strength and are not balanced by counteracting influences, one would therefore expect, in periods of falling commodity prices, fixed interest securities to rise and ordinary shares to fall and to disappoint their holders in the matter of income.
Mr. Edgar Lawrence Smith, author of Common Stocks as Long Term Investments,[1] relates in his Introduction that his book is the record of a failure—the failure of facts to support a theory. The theory was the one just referred to, that at a time when the buying power of money is rising, that is, when the prices of goods and services are falling, bonds are a better investment than common shares, as they were, in the belief of the theorist, from the close of the American Civil War in 1866 up to 1896, a time when the buying power of the dollar rose steadily.
Mr. Smith subjected this theory to the test of historical investigation, covering the period from 1866 to the end of the century, and found that high grade bonds failed to demonstrate themselves as having been the better investment. So he went on with his tests because he felt that "the facts assembled seemed worthy of further investigation. If they would not prove what we had hoped to have them prove, it seemed desirable to turn them loose and to follow them to whatever end they might lead." So he made a series of tests to see "how it would work out, if our fathers had invested solely in common stocks, not with the thought of immediate speculative gain, a thought that is unfortunately most difficult to eliminate from the choice of common stocks, but with the sober-minded purpose of providing (1) constancy of income, and (2) safety of principal."
In making these tests he had to be careful to avoid what the Stock Exchange calls "jobbing backwards"—that is applying what we know to-day to problems of yesterday or yesteryear. If he had attempted to let his imaginary investor in the past choose stocks that seemed to be the best to buy, he would almost certainly have been influenced by what is known now about their subsequent history. So the only principle of sound investment that he could apply to the choice of stocks was that of diversification. Each test assumed the investment of approximately $10,000 (£2,000) in ten different common stocks of large companies and of an equal amount in "high grade bonds." The stocks chosen were selected by some quite mechanical method—those in which there had been the largest number of transactions during the week in which the investment was supposed to have been made, or those which showed the highest yield on the basis of the previous year's dividend, with the most consistent dividend record.
It was also thought necessary to assume that the supposed investor had no further funds beyond his $10,000 for subscribing to additional stock if offered by his companies, and therefore had to sell his "rights" on these occasions at the average price that they would have fetched during the first thirty days in which they were quoted; and for the same reason he was forced to sell any fractional shares that might come to him by way of stock dividend; the proceeds of these sales were added to current income. Whole shares received by way of stock dividend were retained and increased the value of his capital holding and the amount of his income in after years. As he has no funds out of which to meet "assessments," the American term for further capital demanded from stockholders, in the event of reconstructions, he has to sell out whenever these unfortunate experiences occur, and he naturally does so at considerable losses. Otherwise he makes no change in his holding of common stocks during the periods which range from seventeen to twenty-two years; when his bonds are paid off he has to reinvest, but he leaves his shares alone and they are only changed by consolidations and amalgamations.
The remarkable result of these tests is that in every case out of the twelve tests given the investor would have had a higher income from common shares than from bonds; in every case except one the advantage is on the side of the common shares when income and increase in capital value are added together. In one case only is the advantage on the side of the bonds, and that is because the rise in capital value of the bonds was great enough to offset the larger income from the shares.
The first test is made by the investment of $10,000 in the first full week of January, rgor, on the top of a quick rise in security prices during the last months of 1900. The investor is supposed to have put as nearly as he could $1,000 each into the ten industrial stocks in which there had been most transactions. These were American Sugar, American Tobacco, Continental Tobacco, People's Gas, Tennessee Coal and Iron, Western Union Telegraph, Federal Steel, Amalgamated Copper, American Smelting and American Tin Plate.
At that time fifteen high grade railroad bonds were selling at prices yielding 3.95 per cent. But in order to give the bond every chance, the investor was supposed to have made his bond investment so as to earn 4 per cent. and to have chosen his bonds so well that during the period from January, 1901, to December 31, 1922, none of them lost any value owing to poor credit or high interest rates.
The result of this test is that the holding of common stocks, which were bought for $10,002 in January, 1901, had a market value on December 31, 1922, of $15,422—an increase of over 50 percent. The total income received from them was $19,781, averaging a shade less than 9 per cent., as compared with 4 per cent. on the bonds, and the total advantage on the side of the shares, capital increase and extra income together, was $16,401. And this is on the assumption, probably highly flattering to them, that there had been no decline in the prices of the bonds in a period during which the biggest war that ever happened had raised the rate of interest on gilt-edged securities.
It was to be expected that in this period, 1901-22, ordinary shares would be a more profitable investment than bonds, because it was a time of rising commodity prices and interest rates and of feverish industrial activity. Mr. Smith, while freely admitting this, maintains that "regularity of income in this first test was maintained through two severe industrial depressions and the greatest war in history." In claiming regularity of income, however, he seems rather to strain the meaning of the word, as is shown by his own very clear statement year by year of the cash income received from the shares. For the first year it was just over 8.2 per cent., and it went on as follows year by year: 5.9, 5.6, 5.1, 7, 10, 8.8, 8.6, 9.1, 10.7, 8.7, 10.4, 10.5, 8.8, 7.4, 10.5, 14.1, 14.1, 9.3, 9.6, 7.8 and 7.4. The "peak years," when 14 per cent. was earned, were 1917 and 1918, when American industry was making abnormal war profits, and an income which starts at 8 per cent. and leaves off at 7½ per cent., having touched 14 and 5 per cent. in the meantime, can hardly be called regular. The really striking point, however, is that the share income at its nadir in 1904 was still above 5 per cent., and so better than the 4 per cent. which was credited to the bond investment.
The more interesting tests, however, are those which are taken for periods in which commodity prices were falling, so that industrial profits might be expected to have been on a much lower scale. And so we find them, though not low enough to bring the income from them below that which would have been earned by bonds. In Test No. 4, from 1880 to 1899, the total income from shares in twenty years comes out at $14,528, an average of $726.4 on an investment of $10,163, or roughly 7¼ per cent. as compared with 9 per cent. in the first test. Fluctuations are also much less lively, the highest point being $923 in 1882, and the lowest $622 in 1886. But it should be noted that in this test railroad shares were included owing to the scarcity of industrial companies in those remote days. At the time of this test, bonds could be bought to yield 505 per cent., and are estimated to have shown no change in market value during the period. Thus the yield from the shares was substantially higher on the average and materially higher in their least profitable year, while their original purchase price of $10,163 in 1880 compares with a market value of $13,616 at the end of 1896, and $18,817 at the end of 1899.
Mr. Smith considers that the results of Test No. 5, covering the period from 1866 to 1885, are by far the most significant, since this was a period which witnessed a recovery in the buying power of the dollar, which "had had no counterpart since the Napoleonic Wars and has as yet had no other. . . . Based upon the change in the purchasing power of the dollar, there has never been a period which so favoured the purchase of long term bonds as against the purchase of common stocks as the period from 1866 to 1885."
Nevertheless, even in this test, common stocks more than held their own, their advantage in aggregate income offsetting the greater capital appreciation in the case of the bonds. They even succeeded in showing a small capital increase, a fact which seems to have surprised the investigator considerably, for he says: "We find, after the greatest increase in the purchasing power of the dollar that this country has ever experienced or is ever likely to experience, that his [the imaginary investor's] holdings in 1885 have a market value of $10,936, an actual increase, where, if no other factor than the appreciating dollar were in force, a drastic decrease was to have been expected." Whether a drastic decrease in prices of industrial shares was to be expected would depend, however, rather on the view that one takes of the probability that the fall in the rate of interest which almost always accompanies a fall in commodity prices would tend to maintain the prices of shares by making money plentiful and cheap. It or some other influence certainly did so in this case, in spite of a considerable decline in the earning power of the ten selected investments. Just as in England in the spell of low prices and cheap money which culminated in 1896-97, when 23 per cent. Consols rose to 114, those which were thought to be the soundest Home Railway ordinary stocks rose to a level at which they only returned about 3 per cent. to the buyer.
With regard to the income received under this test, Mr. Smith seems to depart for a moment from the serene impartiality with which he reviews the results of his investigation. For he says on page 46: "We have seen how steady an income return was paid during the period on the diversified list of stocks, selected for Test No. 5." But in fact the annual return was, from his figures, as follows:
| 1866 | $1,093.00 |
| 1867 | 1,106.00 |
| 1868 | 976.80 |
| 1869 | 1,018.30 |
| 1870 | 925.00 |
| 1871 | 970.00 |
| 1872 | 1,085.00 |
| 1873 | 966.00 |
| 1874 | 891.50 |
| 1875 | 761.25 |
| 1876 | 687.00 |
| 1877 | 752.00 |
| 1878 | 674.50 |
| 1879 | 729.00 |
| 1880 | 812.00 |
| 1881 | 800.00 |
| 1882 | 663.00 |
| 1883 | 569.50 |
| 1884 | 519.00 |
| 1885 | 565.00 |
If this be a "steady" income it is one which may well have made its owner feel uncomfortable. For if we add the receipts of the last two years of the period together we find that they are $1,084, just nine dollars less than the income received in the first single year of the period and twenty-two dollars less than in the second. In other words the income at the end of the period was half what it had been at the beginning. On the other hand the fall in income would have been balanced by the fall in prices. But the bond-holder, whose money income would have been constant at 607.7, would have had a much fuller benefit from this increase in its buying power.
Mr. Smith's tests, however, should be examined in detail by those who wish to study this interesting problem of the relative advantages of industrial ordinary shares and prior charges and the question whether the, only comparatively, greater risk carried by the former is or is not outweighed by the possibility of indefinite expansion in income and value attached to them, if they imply ownership in a well-conducted and well-financed business.
We have seen the broad result reached by his investigations, that out of twelve tests only one ended in favour of the bond, and that was only so because their greater capital appreciation had more than offset a larger income received from the shares.
The conclusion at which he arrives is that these tests are not in themselves conclusive, but "cumulatively they tend to show that well-diversified lists of common stocks selected on simple and broad principles of diversification respond to some underlying factor which gives them a margin of advantage over high grade bonds for long term investment."
This underlying factor he finds in the reserve fund policy—in the fact that the "directors of conservatively managed corporations over a period of years will never aim to declare all the company's net earnings in dividends. They will turn back a part of such earnings to surplus account, and invest this increasing surplus in productive operation. Such a policy successfully carried out is in fact a practical demonstration of the principle of compound interest" (p. 77).
Nevertheless, before we accept Mr. Smith's conclusion that there is a margin of advantage in favour of common stocks and shares, certain features about his tests have to be noted.
First, that they all refer to the United States, a country which has, ever since the end of its Civil War in 1865, enjoyed amazing growth and prosperity, tempered by short-lived panics and setbacks. This prosperity was due to the activity of a highly intelligent and enterprising people in opening up, with the assistance of a great stream of immigrants and of capital from the Old World, a country of enormous potential resources which at the beginning of the period had hardly begun to be tapped.
Second, that the industries whose securities have been used as tests have thus been especially favoured by the circumstances under which the population for which they catered has grown in numbers and in wealth with the assistance of foreign credit and capital, and have also in some cases been granted artificial advantages, at the expense of the consuming public, by the high tariff policy of the United States.
Third—and perhaps most important—eight out of the twelve test periods chosen ended in 1922, and so covered a time in which American industry was making extra-special and abnormal profits out of the European belligerents during the war and then during the after-war boom. It was not only the directly war industries—the growers of food and materials and the makers of munitions—that reaped these profits, but all the industries of the country benefited by the stream of wealth which the war poured into America, converting her from a debtor country into one which holds all the world in fee.
For these reasons I venture to think that we should add a few words to the conclusion quoted on p. 182 and say that diversified lists of common shares in the industries of a country that is enjoying exceptional growth and prosperity give them a margin of advantage over high grade bonds.
With this modification the conclusion represents much more exactly what Mr. Smith has really proved. How far does it help us in trying to secure the ideal which we have sketched for the investor of securing from his investments an increasing income and growing capital value?
Shall we be justified by it in selling all the gilt-edged securities that we hold, putting the proceeds into ordinary stocks and shares representing ownership in industry, commerce, transport and agriculture, and expecting the Board of our companies to do the trick for us by steady additions to reserves which will put compound interest on our side?
It should be noted that Mr. Smith suggests (p. 104) that the annual excess income from stocks should be treated "as a genuine reserve against the supposed risk of holding common stocks." He proposes to grant the investor only the right to spend the amount that he would have received from an investment in bonds, and to make him reinvest the annual surplus income from stocks each year as he receives it. But how many ordinary investors—could be trusted to carry out this policy, or even, if the necessary self-control were present, to make all the calculations and adjustments that would be required?
But apart from this after-thought, we have to remember that the industrial risk which is inseparable from an investment in any particular industry cannot be eliminated by the process of diversification. It is lessened by diversification, but in these times there is nearly always a good deal of sympathy between one industry and another.
Moreover, excellent as the effects of the reserve fund policy are when successfully carried out, its success, and the power of the directors to carry it out at all, depend on the efficiency with which the industry is conducted on its technical side. If profits are not earned there is nothing to put to reserve; and if the sums put to reserve are invested in expansions of the business that do not pay, they bring no advantage to shareholders, whose dividends have been diminished to provide them. Ordinary shares in a company that is profitably conducted and continually puts part of its profit into expansion are as good an investment as can be found, but profitable operation in any particular industry is evidently a factor that can be relied on with much less confidence when we try to peer into the future than, for example, the total taxable capacity of a great and wealthy people.
With these reservations it seemed to be desirable to make a test of our own, applying the reserve fund policy as a practical guide in the selection of industrial ordinary securities. So I asked my friend and former colleague, Gilbert Layton of the Economist, who spends much of his time on company accounts, to see what would have happened to an investor who had, in 1910, invested in six industrial ordinary shares or stocks, with this principle to guide him.
The imaginary investor took the six companies which had put to reserve the largest proportion of their available balance, after payment of preference dividend, during the years 1905 to 1909. The accounts tested were those of concerns with a paid-up capital of £1 million and upwards, and companies paying an ordinary dividend of less than 5 per cent. were excluded. Some such limit is evidently essential; since otherwise a company which had fared so ill that it paid nothing, and carried forward all its small earnings, would be hailed as a model of good finance. The companies are shown in the order of the proportions reserved, the Eastern Telegraph having reserved the largest and Pease & Partners the smallest percentage of the six companies.
The price of purchase is assumed to be the middle price of December, 1910, thus giving the supposed investor time to make his investigation into the accounts of the companies for the preceding five years. Here is the result of the test:—
| Company | Bought in Dec. 1910 at | Value June 30, 1925. | Increase or Decrease in Value. | Average Income per cent. 14 years. |
|---|---|---|---|---|
| Eastern Telegraph (£100) | 135 | 175½ | +30% | 6.349 (tax free from 1919) |
| Howard & Bullough (£1) | 2⅛ | 1¾[2] | +9.8% | 8.091 |
| Babcock & Wilcox (£1) | 5 | 2½[3] | +108% | 6.786 (all tax free) |
| Bradford Dyers | 1 | 2 | +100% | 15.821 |
| Fine Cot'n Spinners (£1) | 1¾ | 3[4] | +120.6% | 7.620 |
| Pease & Partners | 12 (£10) | 11/16 (£1)[5] | -31.25% | 10.340 (tax free from 1919) |
| Total | +54.86% | 9.168 |
The year by year performance of the companies was as follows:—
The Eastern Telegraph Company paid a dividend of 7 per cent. for the first four years, 8 per cent. for the next four, and 10 per cent., free of income tax, for the remaining six years.
Howard & Bullough paid 15 per cent. for the first two years, and with the second dividend distributed a bonus of 33⅓ per cent. In the following year the rate was 15⅚ per cent., but for the next four years it dropped to 10 per cent. In 1918 the dividend was 15 per cent., and there was also a capitalized bonus of 33⅓ per cent. Then followed three years of 10 per cent. on the larger capital, the dividend being increased to 17½ per cent. in 1922 and to 20 per cent. in 1923. In 1924 the ordinary capital was increased to £1,250,000 and the dividend fell to 124 per cent.
Our investor was fortunate in his holding of Babcock & Wilcox, for in the first year he received a capitalized bonus of 100 per cent. in addition to a dividend of 28 per cent. on his original purchase. Throughout the fourteen years this company paid its ordinary dividend free of income tax. In the second and third years the rate fell to 16 per cent. on the larger capital, dropped to 14 per cent. for the next year, rose to 15 per cent. for the next five, then to 16 per cent. for two, and then to 20 per cent. for 1922, when once again a capitalized bonus of 100 per cent. was distributed. In the last two years the dividend on the quadrupled capital was at the rate of 12 per cent.
The capital of Bradford Dyers' Association remained unchanged throughout the period, while the dividend increased in several stages from 6 per cent. in 1911 to 22½ per cent. in 1919. It then fell to 20 and 10 per cent. in the two following years, rose to 35 per cent. in 1922, and dropped to 25 per cent. in the last two years.
The dividend of the Fine Cotton Spinners' & Doublers' Association was at 8 per cent. in 191 and remained at that figure until 1917, when it was raised to 10 per cent. for two years. In 1919 the rate was 12 per cent., followed by 20 per cent. in 1920. Then came a drop to 10 per cent. for 1921 and 8 per cent. for 1922, followed by 12½ and 14 per cent. for the last two years respectively. In 1913 and 1919 the shareholder received bonuses of one 5 per cent. Preferred Ordinary share for every five ordinary shares held.
The Pease & Partners' dividend was 8 per cent. for the first two years, then 12 per cent. for two years, followed by 10 per cent. and 15 per cent. In 1917 and 1918 the rate was 17½ per cent., and in 1919 12½ per cent. free of tax, together with a capitalized bonus of 20 per cent. In 1920 the rate went up to 18 per cent., but fell to 14 and 5 per cent. in the two following years. In 1923 it rose to 6 per cent., and in 1924 to 8 percent. This, as may be seen from the table, is the only case in which our investor's capital has suffered depreciation over the fourteen years; it is accounted for by the fact that in 1925 the dividend fell away to 1½ per cent., owing to the severe depression in the iron and steel trades. The aggregate investment in the six companies, however, has appreciated by over 50 per cent. and our investor has received a dividend rising from 5.9 to 11.4 and averaging over 9 per cent.
The investigation is on too small a scale to prove much. But, as far as it goes, it is interesting and encouraging, especially when we consider that the period covers two cataclysms—the beginning of the war and the after-war collapse—and a spell, still continuing at the end of the test in June, 1925, of severe industrial depression.