Stock Values and Stock Prices

Author(s): Nicholas Molodovsky

Source: Financial Analysts Journal , May - Jun., 1960, Vol. 16, No. 3 (May - Jun., 1960),
pp. 9-12+79-82+84-87+89-92
Published by: Taylor & Francis, Ltd.

Stable URL: https://www.jstor.org/stable/4469016

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to
Financial Analysts Journal

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
 Stock Values and Stock Prices
by Nicholas Molodovsky

PART ONE makes sense. But the multipliers must reflect changes
in earning power.
A GENERATION OR TWO AGO, few investors would have
considered putting their money into stocks insufficiently Price-Earnings Ratios and Stock Market Levels
backed by tangible assets. Their net worth was an ob- Another tenacious idea was not helpful to investors:
jective standard of value. Quite independent of the the thesis that when price-earnings ratios are historically
stock's market price, it was a sensible guide for invest-high, stocks are overpriced; and that when stocks are
ing. In poor times, the principal was likely to remain cheap, the ratios become invitingly low. To bring the
safe while providing a reasonably assured and adequate relations between stock prices and price-earnings ratios
income. And if the stock's earning power increased into sharper focus, let us look at them through the
over the years, both principal and income grew. stereoscopic viewer of a scatter diagram.
Yet, frequently, some of these asset-heavy investment Figure I pictures ratios (horizontal scale) and prices
fortresses stood ponderously still at the same levels of (vertical scale), 1871-1959. To avoid cluttering the
earning power and price, while prices and earnings of chart, we omitted century figures, but there cannot be
many equities deeply submerged in asset-accounting any confusion. The earliest years plotted are the 1870's,
water surged irresistibly upward. and the most recent years do not extend beyond the
Gradually, the effect of earnings on stock prices 1950's. The years of bull market peaks are marked by
began to impress investors, and they tried to use them triangles, those of bear market troughs by heavy dots.2
in measuring stock values. One of the early notions The absence of correlation between prices and price-
was that the fair value of a stock could be ascertained earnings ratios is obvious. The points, the triangles, and
by applying a uniform multiplier to its current earnings. the heavy dots are scattered throughout the diagram.
Evidently, ratios are not always high when prices are
The 'Rule' of Ten-Times Earnings
forming tops, or low when stocks are scraping bottoms.
Valuation of Common Stocks, in the February 1959 In the turning areas of stock prices they vary greatly,
issue of this publication, described the circumstances affording no basis for judgment whether stocks are dear
that gave birth to this belief. It mentioned also that, or cheap.
according to an authoritative financial periodical, many Nor do the ratios rise in bull markets and accompany
people still held it. Browsing through earlier issues of bear markets in their slides. They do rise in 1926-29,
the same magazine, we came across some additional but with similar consistency they decline in 1934-37.
comments: At the end of each period, they had reached their re-
spective positions by very different routes and from
"The table on this page revives a method which
opposite directions.
enjoyed wide popularity during the securities mar-
The lack of correlation between prices and ratios
ket upswing of the 1920's and the tortuous down-
becomes particularly striking when we observe the
swing of the 1930's for estimating the 'worth' of tremendous range covered by prices within the same
common stocks. In the buoyant 1920's, when an
relatively limited zone. Thus, the 13 to 17 times earn-
industrial company's shares were selling at around
ings multipliers lie half way on the scale of price-
10 times earnings, actual or closely estimated, the
earnings ratios. Within this rather narrow strip, fluctua-
price was considered by the trading fraternity as
tions in stock prices have been extraordinarily wide,
reasonable, meaning reasonably low.
and have encompassed both peaks and troughs of their
"Technically, the partisans of such a measure of cyclical movements.
value spoke of the ratio of price to earnings."' While there is no iron clad pattern in the general
When earnings remain stable, a constant capitalizer structure of the diagram, the price-to-earnings ratios
tend to cluster in the middle or "temperate" zone, and
the "tropical" zone of the highest ratios is dominated
1. For References and Notes see Appendix.
by years of economic and market recessions. The
"arctic" region of the lowest ratios has less clearly
Nicholas Molodovsky holds a master's degree in economics marked characteristics, but numerous war or rearma-
from Harvard University and a doctor's degree from the Uni-
versity of Paris. He has more than 25 years of investment
ment years are in it.
management and financial analysis experience in the United The diagram gives the impression that price-earnings
States and Europe and is the author of many publications in
economics and finance. Dr. Molodovsky is an Associate Editor The author wishes to express his gratitude to Catherine
of this Journal and a correspondent of La Vie Francaise of R. May for her invaluable assistance and to Martha Ander-
Paris. son for her superlative editorial skill.

MAY-JUNE 1960 9

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
 Figure I-Prices and Price-Earnings Ratios: 1871-1959

58

58~~~~~~~~~~~~~~~~

54

52

48 50__ _ _ _
46 A~~~~~~~~~~~~~~~.

44

4 2

40 __ __ _ __ __ _ __ _ -
38

32

2 ~30t
CL28
3 65 1 -_ _ _ _ _-
r_ __ __ _ E !_ _ _ _ _
24 1A 1 14 1
22

18 5
2 0I_ _ _ _ _ _ 0
16
t O ll I I111

14

12

I0 o6_____3

8 7 8 9 1 I 1 13 14 15
PcR ICE-EARNINGS R

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
ratios reflect underlying economic conditions and are with the end of hostilities, restrictions on the produc-
not associated with any particular level of prices. tion of consumer goods were removed, thereby destroy-
On the scale of the ratios, the distance between 1928 ing this artificial dam.
and 1929 is small. But between 1929 and 1930, when
Those Tantalizing Ratios
stock prices declined not quite as much as they had
risen in the preceding year, the ratios rose markedly, The price-to-earnings ratio has become the most
indicating the passage from economic expansion to con- popular yardstick in financial analysis because it seems
traction. Strange as it may seem, price-earnings ratios so bewitchingly simple. Authors of voluminous invest-
are highest during depressions; wars and inflations tend ment treatises often take its trustworthiness for granted,
to bring low ratios. and confidence in its meaningful nature bobs up time
and again in the torrents of written and verbal comment
A Closer View on daily stock trading.
To what extent has the validity of the diagram been Nevertheless, as our illustrations suggest, the ratios
weakened by the use of annual data? They average out are not foolproof. Their apparent simplicity may be a
and may neutralize the fluctuations in stock prices. booby trap. It camouflages complex and ever-changing
Steep rises and staggering declines sometimes take place relations of two factors that are in constant flux and
during the same year. Summarization in a single figure quite frequently unrelated.
obliterates them. In terms of Dow-Jones Industrials, Perhaps the main weakness in using the ratios for
the annual mean price for 1937 was 165; but in the last evaluating stocks or the general market stems from
six months alone it rose from 163 to 190, then col- their nature: they are a quotient. When we try to
lapsed to 112. And to take an example of a turning transform them into multipliers, we have to twist them
area containing a reversal of a bear market, the annual around. It can be done, but it's a tour de force.
mean for the Dow-Jones Rail Average was 27 in 1932;
Through continuous observation and comparison,
yet, in the first nine months, it dropped from 42 to 13,
financial analysts evolve an empirical series of "mul-
then recovered to 41. These illustrations are not unique.
tiple multipliers." Different types of stocks acquire
The legitimate doubt about the significance of annual various price-earnings ratio characteristics. This cum-
data was examined in The Analysts Journal of May bersome structure is serviceable for working purposes,
1953. From this study, too, the writer borrowed the if we eliminate such misconceptions as the 10 times
scatter diagram and some parts of this discussion. To earnings "rule" and the non-existent correlation be-
that extent he has plagiarized himself. But the diagram tween price-earnings ratios and stock prices. Just the
has been brought up-to-date and recomputed in terms same, to disentangle earnings from price and check how
of the new Standard & Poor's "500" indexes. firmly each can stand on its own feet might be worth
The look at closer range in the 1953 study, in month- while.
ly figures, not only bears out but even strengthens the
The Province of Price
conclusions based upon annual data. During the early
phase of the great bull market of 1921-29, price-earn- Price may be considered first: it stands closest to the
ings ratios began to decline as prices started rising. In tempting objective of making money.
its middle years, 1924-26, the ratios moved practically In Seven Kinds of Inflation, Richard Dana Skinner
sideways along the low level of nine times earnings. In remarked: "More zeal and energy, more fanatical hope
1927 they did shift to a higher plateau, but largely be- and more intense anguish have been expended over the
cause earnings fell. Only during the last three months of past century in efforts to 'forecast' the stock market
the market's dash to its summit did the ratios also rise than in almost any other single line of human action."3
vigorously. Many of these efforts sought guidance from the action
The bull market of 1932-37 unfolds an even more of price itself.
intriguing story. With few interruptions, price-to-earn- Today, price has lost nothing of its attraction. People
ings ratios never ceased to decline from their all-time look to it as to the morning star. Price devotees are
high reached during the depression. numerous throughout the land. One of the most articu-
A relatively recent exception to the normal absence late labors for a tight, but not little, sect with head-
of correlation between stock prices and price-to-earn- quarters in a manufacturing center 65 miles southwest
ings ratios prompted some investors to a bit hasty con- of Boston as a missile flies.
clusion. During the bull market of 1942-46, prices and According to a New Yorker profile, this consecrated
ratios were in step; the latter were low at the bear man is immured in a boarded-up room where he can
market bottom of 1942, and high at the 1946 top of catch no glimpse of the activity in the street. Nor can
stock prices. However, earnings were meanwhile para- the weather play on his spirits. In the flourescent light
lyzed. Measured by the "500" index, they were 18% flooding his office, he soon forgets whether it rains or
higher at the beginning of 1942 than at the beginning shines. And in this workshop, which is completely
of 1945. During this bull market, stock prices rose soundproof, there is neither radio nor television to blare
because of the pressure of idle money accumulated out the news of the day. In this vacuum he acts as an
from the wartime shortages of civilian goods, not in investment adviser and operates in stocks for his own
response to a cyclical expansion. They broke when, account. Faithful to a self-imposed rule not to read any

MAY-JUNE 1960 11

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
financial papers, nor consult any sources of business larity. Yet they are merely levels of chance oscillations
information-unless they are at least two weeks old- of successive changes in size, transcribed into connected
he studies price sheets, ignoring corporate earnings. He shapes.4
buys and sells after scanning charts of daily stock price In industry, some definite assumptions are usually
ranges and number of shares traded. He and his fol- made, although not always publicized, in the use of
lowers are convinced that market prices and the veloc- quality control or other types of test to determine the
ity of transactions flow from the collective reservoirs of occurrence of changes in a system. For example, a
all available knowledge concerning each stock. They large number of independent small factors may be as-
claim that chart patterns reflect all impulses to buy and sumed to be at work in a manufacturing or operating
sell, and can more surely guide investors than the situation. If, in their aggregate, they form a normal or
limited and often inaccurate information they could bell-shaped distribution of the observed variable, one
themselves secure and whose significance they may be often says chance is the cause. And predictions can
unable to interpret. frequently be made how the result will continue to vary
under the same system of chance causes. When it does
Operations Research not remain within the predicted limits, a so-called
Statisticians proceed upon a different logic. They "assignable," or reasonably discoverable cause, is
have found that patterns of uncoordinated individual hypothesized and, in practice, can frequently be found.
decisions often have the characteristics of a random or If not, the start-to-look limits are widened until a cause
chance series. This process can be simulated by a tech- is found when the limits are exceeded.
nique based on a suitable manipulation of random For stock prices, however, the assumptions one can
numbers, known as "The Monte Carlo Method." make for developing a theory of fluctuations-or for
For centuries mathematicians have been improving creating a simulation-are decidedly more complex.
ways of discovering the probabilities inherent in random Unlike telephone callers, buyers of stocks are influenced
series. More recently, industry has been increasingly by the activities of other buyers. As an expert in
exploiting them. Quality control, which has wide appli- mathematical controls pointed out to us, this creates a
cation in manufacturing production, is based upon feed-back problem with its attendant instability, re-
sampling techniques rooted in the theory of chance quiring more sophisticated analysis than is successful
occurrence. When we know a series is random, we can in assuring quality or providing adequate telephone
analyze it with mathematical and statistical tools, quite service. He, nevertheless, believes that some progress
often opening the way to valuable practical conclusions. can be achieved toward constructing mathematical
Time series and probabilities are among the essential models that will suitably describe changes in stock
ingredients of operations research. prices. Among others, the approaches followed in the
For example, what has become known as Queueing theory of games might eventually yield encouraging
Theory is doubtless the outgrowth of Traffic Theory, results, although they are exceedingly difficult to set
developed at Bell Laboratories for planning installa- down mathematically, except for the simplest kinds
tions of telephone equipment. To engineer and install of competitive activity.
equipment that could meet in full all demands of all For our own part, we believe that while analysis of
users at all times would be economically wasteful. The the price action of stocks is necessary and fruitful, it
goal is to provide adequate service ensuring at a reason- can play only a supporting role. We shall discuss it
able cost practically full satisfaction of all needs with in Part Two of this study.
only a quite tolerable waiting for a few at the peak of
Capitalized Earnings
activity.
People telephone for personal reasons which are un- Of the Siamese twins arbitrarily and artificially tied
connected and independent. Nevertheless, probability in the unhappy union of price-earnings ratios, earnings
theory can often plot the fluctuations in the incidence are without much doubt the more viable and stronger.
of calls, so that adequate but not extravagant equip- When separately studied, they help us to understand
ment can be installed. The recurrence, frequency, and more clearly the nature of stock values.
sequence of calls are essentially random, yet their pat- Perhaps the principal reason the hybrid ratio ap-
terns form mathematical curves analyzed years ago in proach cannot avoid being contradictory and uncertain
connection with studies of probabilities. As long as is that, unlike its predecessor, the balance sheet method,
essentially non-random individual occurrences remain it is not anchored to any general principles of unques-
collectively random, they can be statistically studied tionable validity and significance. It is therefore in-
and their patterns projected into the future. herently helpless to produce a general theory of valua-
tion of common stocks, or a method of appraisal
The market action of stocks induced by multiple and
squarely fitting - without exception - all observable
unrelated impulses of buyers and sellers, from the four
cases within an encompassing and consistent structure.
corners of the world, motivated by heterogeneous urges,
To become safely and independently airborne, capi-
suggests that it may be a random series. If so, it can
talizers applicable to earnings should not be tied to
be subjected to mathematical analysis and forecasting.
price. Earnings themselves and their own character-
Sequences of random numbers do produce patterns
creating a superficial impression of meaningful regu- (Continued on pate 79)

12 THE FINANCIAL ANALYSTS JOURNAL

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
Stock Values and Stock Prices fellow chemists. Men of affairs - if they heard about
him at all-probably considered him a stodgy egghead,
(Continued from page 12)
an insufferable bore. In their practical work they
istics must govern the size of the multiplier that will could get along quite well without dull theorizing. Yet
lift earnings to the level of value. And for reasons clearer concepts led to technological advances that
detailed in the writer's previous papers, earnings, i.e., would have been unimaginable without them.
the capitalization base itself, can be portrayed only by Our beloved teacher, Frank W. Taussig, once wrote:
reliably computed trends, not by flashlight snapshots of "Economists are often twitted with being theoretical and
their fleeting current figures. out of touch with the facts of industry. Much more
Besides, current earnings are no more practical as unpractical is the attitude of the average business man,
a base for capitalizing stocks than they are meaningful who is familiar with but one small corner of the in-
theoretically. Between them and average stock prices dustrial world, contents himself with the most super-
there is no significant correlation.5 Flitting across the ficial commonplaces, and knows so little of the essen-
screen of daily events, they may trigger impressionable tial problems of economics that he is hardly aware even
or uninformed people into emotional action. But repre- of their existence."6
sentative earning power and its rate of growth alone Financial Analysts are aware of the importance of
possess theoretical and practical usefulness. basic research for industrial progress. Investors may
Changes in growth rates of projected trend lines find a basic valuation approach equally vital.
of future earnings, and the number of years during
which each rate will prevail, will determine the capi- Pioneers of Valuation Studies
talizers. The level of the trend line at the point from In his essay, "On Liberty", John Stuart Mill makes
which future projections are made will be the point of a strong plea to the individual to assert himself. The
departure of the computations. Finally, the growth majority imposes its opinions so completely, and so
rate of the trend line connecting this point with the harshly censors independent thinking, that belief tends
historical past exerts an influence also because of its to become stagnant. We must, therefore, be especially
effect on present payout. grateful to those who led the efforts to penetrate more
Valuation of Common Stocks described all this in deeply into the realities of our own field of endeavor.
detail. It is briefly repeated here to emphasize that no Valuation of Common Stocks gives credit to the pio-
element of price enters into appraisals of common neers who opened for us new avenues for methods of
stocks by this method. Value is the standard we seek stock appraisal. To the names previously mentioned,
by which to measure whether price is high, low, or in we wish to add another. A. Wilfred May, Executive
line with the standard. When we use price-earnings Editor of The Commercial and Financial Chronicle,
ratios we are prisoners of price. Even after such "mul- and faculty member of the New School, has striven for
tipliers" have been refined by being filtered through many years to convince the financial community that
multiple comparative experience, they are still deriva- investment policy must be built around a concept of
tives of price which they are supposed to measure. A value. He has favored specifically gearing the pur-
ratio remains a ratio even after having been adjusted. chase price to the probability of recovering the princi-
We are spinning inside a looping hula hoop. And when pal, together with its investment return and a profit in-
we attach ratios to fluctuating current earnings, we per- crement within a foreseeable future. Mr. May thinks
form our circling act on quicksand. that an advantageous buying price for stocks of com-
panies with stable earnings might be estimated by capi-
Importance of Adequate Concepts
talizing the expected future dividends at a rate embody-
The story of price-earnings ratios in not unlike that ing the rental value of the capital, and a proportion of
of phlogiston. Acording to chemistry textbooks, three a reserve sufficient to amortize the investment during
centuries ago a German scientist "explained" burning a period commensurate with the risk assumed. Finan-
by a chemical fluid he dubbed phlogiston, "the thing cial Analysts who have struggled with the problem of
that makes things burn". Everything flammable, he valuation are indebted to Mr. May's contribution and
claimed, contained some phlogiston. When it burned, to his tireless campaign for sound methods of apprais-
the phlogiston escaped, causing crackling, sparks, ing common stocks.
flames and smoke.
THE FACTOR OF TIME
While phlogiston presumably accounted for some of
the aspects of burning, it led to many difficulties and The most important dimension of earnings lies along
contradictions until Priestley discovered oxygen. Not the axis of time. Unless viewed in the perspective of
only could burning be then understood, but Lavoisier's their past and projected future, current earnings have
experiments soon laid the foundations of modern little economic meaning. Time is, in fact, a basic di-
chemistry. mension of finance. The concept and practice of the
False as were the ideas about what happened in interest rate the capital's share in the distribution of
burning, ores were melted into metals and fashioned wealth created by production-xpress the relation be-
into tools, ornaments, and weapons for thousands of tween present and future payments.
years before Lavoisier was born. And many years To quote F. R. Macaulay: "Investments as a class
elapsed before his teachings were accepted even by his constitute one family. They each originate in an ex-

MAY-JUNE 1960 79

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
change of present money for an expectation of future and earnings in order to find norms of relations between
money. If it were not for such an expectation they them under varying rates of earnings growth. To make
would have no exchange value. And they lose that this presentation more concrete, the growth figures are
value as the expectation dies out. The demand that given in the Appendix.8 They reveal an extraordinary
comes from the possibility of buying them and later galaxy of variations and combinations in the rates of
selling at a profit may exist for a time in a sort of eco- change of the shorter trend lines of the two factors,
nomic vacuum, but it is essentially a derived demand which blend into the uniformity of identical secular
and, in the absence of any (warranted or unwarranted) rates.
expectation of future returns, it sooner or later disap- Because we can compute theoretical dividends by
pears. means of a generalized expression of relations between
"Because the good that the common stock offers to earnings and dividends, we can find present worth
its purchaser is an expectation of future money pay- directly from earnings. We can compute capitalized
ments, the relation of its present-money price to its values of stocks independently from actual payouts,
future-money payments is as unmistakably an interest even when no dividends are paid. All stocks become
phenomenon as is the relation of the present-money comparable on the same basis and in terms of one an-
price of a bond to its future-money payments."7 other. We have an organized technique for applying
At first glance, the value of a common stock may the method of present worth to earnings. And the
not seem to be related to the future payments to be re- resulting multipliers are no longer soft and uncertain
ceived from it. Numerous stocks paying small or no but have the strength of structural steel.
dividends often command high prices. However, This is not the only approach to capital worth.
though the buyer, with an eye to profit alone, may be Valuation of Common Stocks summarizes several
indifferent to or not even aware of the fact, potential others. Yet the road to value via earnings is sound
investors' expectations of future payments are the active as long as one steers a straight course. As Dr. Julian
force supporting the price of his stock. G. Buckley correctly stated in the March-April 1960
A comic strip published by a New York evening issue of The Financial Analysts Journal, the basic fac-
paper depicted, in the spring of 1946 at the height of tor when capitalizing earnings is "the growth rate of net
the black market, the woeful experience of a lady who income". (Dr. Buckley's italics).
committed the blunder of opening a package she had The same March-April issue of our Journal also con-
purchased. The food in it was spoiled. Indignantly tained an interesting article on The Theory of "Games"
she rushed back to the store. "My dear lady," ex- in Stock Selection, by Irving Hale. Fearlessly, he
claimed the clerk, "you opened the package? Good writes: "Several new mathematical theories have
heavens! That's not for eating. It's for BUYING proved to be useful tools when used to supplement
AND SELLING." statistics in business and finance. Generally, statistics
are employed to construct theories from empirical data,
So it is too with stocks. Many purchase them for
while the new techniques, broadly classified under the
selling, never looking inside. But if the market declines
before they succeed in disposing of the package, the
value of the contents suddenly acquires a new and deci-
sive meaning.
The only financial return from a common stock
comes in the form of a payment, or of a series of pay-
ments to the holder. Dividends are the foundation on
which common stock values are erected. If expecta-
tions of future dividends fail to be realized, the stock
becomes worthless.

Earnings are the Dividends' Fund

Discovering the present worth of future dividends is *1
computing their capital worth. And this is equivalent
to finding the value of the stock producing these divi- ( 't,
dends.
This method of capitalizing income applies to divi-
dend payments. A bridge must therefore be built from
dividends to earnings for determining the proper multi-
plier applicable to the latter if they are to be the capi-
talization base. In Valuation of Common Stocks, (The
Analysts Journal, February, 1959), we treated this
problem in detail and would be glad to send a reprint
upon request. Under Finding Normal Dividends (Re-
print, pp. 12 and 13), we described how we correlated "B. C."-Before Computers
vast series of data pertaining to trend lines of dividends

80 THE FINANCIAL ANALYSTS JOURNAL

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
heading 'operations research', act in the opposite direc- indexes. Both results may still be rounded out to 5%,
tion, proceeding from the abstract towards the practi- which stands as the representative historical current re-
cal". These are the methods to which we alluded turn from common stocks.
above in discussing the interpretation of price patterns. However, by lifting the growth rates of the trend
Our methods of valuation need no recourse to higher lines of earnings and dividends to 21/2 %, while main-
mathematics. They do not call for much knowledge taining the 5% current return, we have also increased,
beyond compound interest and geometric progression. in round figures, the effective total return from divi-
But while recognizing our own limitations in the com- dends to 71/2 %. And this is the figure we must hence-
mand of mathematical tools, together with an old-fash- forth accordingly use as our discount rate for comput-
ioned reluctance to use them, we, nevertheless, welcome ing present worth.
the growing penetration of mathematics into our field. For the benefit of readers who favor precision with
Repeating the simile of a preceding article, we are not figures, as we do, we may add that if the indefinite
inclined to demand that Financial Analysts avoid tech- growth rate is assumed to be 2.5%, effective yield
niques or computers beyond those nature provides in would be 7.625%, which could be rounded to 7.5%
our fingers and toes. (compared with 2% and 7% from the old data).

SOME NEW FINDINGS An infinite projection at 2.5%, discounted at 7.5%,

would have the same effect on value in terms of present
In Valuation of Common Stocks, under Splicing the
worth as an infinite projection of 2%, discounted at
Indexes, we suggested that it might be desirable to re-
7%, because the changes in the two rates balance.
compute the data in terms of the new Standard & Poor's
However, projection at definite estimated rates in the
"500" Indexes, whose earnings figures were not yet
near future would be reflected by slightly lower values
available at the inception of our study. Such a recom-
because of the higher discount rate. The difference
putation has since been made and we now submit the
would not be large because the number of years for
results.
which projections could reasonably be undertaken
Annual Rates of Growth, 1871-1955 would be few compared with the indefinite projection
Prices Earnings Dividends of the remaining years at the overall "rate of igno-
rance." Nevertheless, our new universal rates, 21/2 %
Old Indexes 1.99% 2.13% 2.12%
for earnings growth and 71/2 % for the discount rate,
St. & P.'s "500" 2.20 2.40 2.36
give somewhat lower appraisals.
The 1871-1955 story is the same as told by both sets
Old Names, and New
of indexes as far as the relations between the three fac-
tors are concerned. But their average annual rate of Another innovation arising from further thinking on
growth is approximately 0.3% higher when measured this difficult subject, for one more year, has to do
by the new "500" Indexes. with terminology. Under The Role of Market Value
The addition of 1956-59 to this long period lifts the (Valuation of Common Stocks, Reprint, p. 22), we re-
annual rates of the "500" Indexes by approximately an- marked that our terms were convenient short labels, not
other 0.2% as follows:
final definitions. In preceding articles we called the
value of a stock appraised by the method of present
Annual Rates of Growth, 1871-1959 worth Theoretical Value and Intrinsic Value indiscrim-
Standard & Poor's "500" inately. We also interchanged freely Market Value,
Prices Earnings Dividends Normal Price and Price Orbit to denote the habitual
representative average pricing of a stock, or a stock
2.44% 2.58% 2.51%
average, by the market itself.
This difference is probably, at least in part, cyclical. To avoid any possible confusion between the two
The addition of the last four years to the Earnings series concepts, we intend to reserve for the latter exclusively
of the old index also raises its average rate of growth the term Price Orbit. Our original term, introduced
from 2.13% to 2.28%. Nevertheless, replacing the some seven years ago, has the advantage of completely
original data used in Valuation of Common Stocks, by avoiding the word "value" which tends to obscure the
the "500" Indexes, does produce a slight upward slant. difference between the two concepts.
And since we wish to take advantage of the most "Value" will refer solely to appraisals based on pres-
modem tools, we adjusted our representative average ent worth; and such appraisals will always be de-
long term rate of earnings and dividends growth fron scribed as the Investment Value of a stock or of an
2% to 21/2 % in all our appraisals. average. "Theoretical Value" makes the concept
This change does not affect the basic relations be- sound more abstract than it is; and "intrinsic", while
tween earnings and dividends, since their respective much liked in Wall Street, gives it the flavor of a "true"
growth rates have received an identical adjustment. The value, which creates an exceedingly dangerous mental
average yield, as measured by the arithmetic totals of picture. Practical Measure of Value ( Valuation of
all actual dividends and prices for the entire period Common Stocks, Reprint, p. 6) illustrates, by a parable
1871-1959 in terms of new "500" Indexes, is 4.9% as ranging from blondes to Lucifer, how it can make us
against 4.945% for 1871-1955 as measured by the old tumble into Satan's trap. We entertain no idle dreams

MAY-JUNE 1960 81

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
of reaching unattainable and futile objective true values. average's long-term trend line. The succession of secu-
Our more modest purpose is to line up practical ap- lar historical trend points provided the lowest of all
praisal schedules as a means of comparing the invest- possible series of average earnings points, thereby
ment values of various stocks. avoiding the danger of overvaluation.
However, this method had several drawbacks.
Back to Good Old DJIA
Though compensated by a higher capitalizer, a long
We hope that this clarification of terminology will term trend line cannot do full justice to recent earning
help our readers, as well as sharpen our own thinking. power. Moreover, doubts are justified about the valid-
Earlier we were inclined to share the more general view ity of the 4% growth rate which we applied to the long
that the Price Orbit, i.e., the historical record of the term trend line of DJIA earnings.
stock's or average's habitual central market price, which It may be argued, as we have done ourselves, that
gradually adjusts itself to its long-term trend, was a DJIA is not only composed of the mightiest industrial
sufficient expression of its value. But in The Outlook giants, but that others can be substituted for them when
for the Stock Market (Commercial and 'Financial they no longer measure up to this Average's deluxe
Chronicle, Dec. 8, 1955), we realized the need for an- specifications. Five stocks were replaced in 1959
other concept that would free the idea of value from alone. Far from being an "all stock" index, the DJIA
any admixture of price. is like a marble pool bordered with tiles of solid gold
We were fortunate then in forecasting the fluctuations and constantly replenished with streaming blue water.
of the next 18 months rather accurately. More im- However, generalizations, especially when waxing lyri-
portant, we sensed that investors would reappraise the cal, should be just as subject to audit as high bracket
level of representative postwar corporate earning income tax returns. Closer statistical examination re-
power. We suggested that the market would probably veals that in more recent years the DJIA annual rates
go through another cyclical test before investors would of earnings growth were frequently lower than those of
believe that the normal level of corporate earnings had the very broad Standard & Poor's "500" index.
shifted to higher ground and that capitalizers other than
Annual Rates of Earnings Growth
those applicable to historical average levels were called
for. The new plateau had become more evident by DJIA "500"

the time we appraised the Dow-Jones Industrial Aver- 1915-1959 4.44 3.26
age in an address reprinted in The Chronicle of October 1929-1959 10.36 7.73
30, 1958. But already in 1955 we had anticipated 1937-1959 -7.92 8.32
both the necessity for a better concept of value and 1946-1958 -5.75 7.20
generally higher prices in the market itself. 1954-1959 -1.47 2.06
To this writer's knowledge, his 1958 address was the
first attempt in financial literature to measure the invest- An annual rate of earnings growth in excess of 4%
ment value of a stock market average in capitalizing revealed when the DJIA is taken back to 1915-is ap-
the earning power by the method of present worth. parently due to its makeup, not to superiority of per-
formance.
Representative opinion in Wall Street held that the
DJIA was grossly overpriced. It was fluctuating below Before 1929 the DJIA probably did not contain
550 and financial sirens were screaming emergency enough stocks to be reliable. From 1885 through
alarms. In reaching our optimistic conclusion we used 1896 the index was composed mostly of railroads.
the simplest of various possible valuation bases-the Starting with June 1896, it consisted of 12 industrial
stocks; and from 1914 to 1928 it was composed of 20
industrials. Not until October 1928 was the index of
30 equities inaugurated.
Moreover, unlike Standard & Poor's, Dow-Jones
takes deficits into account. Negative figures are used
instead of substituting zero for them. The DJIA earn-
/
ings troughs are accordingly much lower in the depres-
sion years 1921 and 1932 than those of the "500".
This difference in the construction of the two indexes
is intensified by the unavailability of earnings data for
DJIA before 1915. Even some privately computed
figures do not extend into the past beyond the turn of
The Dvil' Tra the century and apply to dividends only. Futhermore,
as noted above, the Dow index was small in the early
days.
Our computations of the growth rate of DJIA's long
term earnings trend line were anchored in 1915, i.e.,
only a short time before the depressions of 1921 and
1932. The combination of "lower lows", with their
location near the inception of the trend line, acted much

82 THE FINANCIAL ANALYSTS JOURNAL

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
as would a heavy weight placed on the sawed-off short earnings and indicated lower earnings for 1961, mark-
end of a seesaw, tipping it off balance. ing the latter as the first year of a cyclical recession.
With the single exception mentioned, they projected
Appraisal of DJIA
the next cyclical earnings trough as coming in 1962.
We abandon, without regret, the 4% growth rate for Despite the differences in their approache-s and varia-
DJIA basic earnings trend. We used it because the tions in their techniques, the projections are not too
relative shortness of the trend line itself (going back dissimilar. Between them, they form a good cross
only to World War I) seemed to reduce the obvious section of authoritative opinion representing leading
objections to appraising present and projected future banks, statistical services, as well as internationally
values from secular measures rooted in the distant past. known investment advisory and brokerage firms. Most
And we liked it because it simplified our task. Yet an of the trend lines fitted into each individual projection
identical technique for all appraisals without exception cross each other about 1965 or 1966; and they average
is undoubtedly sounder. out at an annual rate of earnings growth at close to
If we should plot the DJIA earnings since 1929, 5% for 1960-70.
when the complete list of 30 stocks came of age, super- For the purpose of appraisal, the majority and
imposing it on the "500" index by adjusting the respec- minority opinions produce identical results because the
tive scales, we would see the two indexes moving closely much more widely fluctuating earnings projected by
together. This takes a lot of wind out of the frequent the latter are also adequately represented by the same
criticisms that the DJIA is not sufficiently representat- 1960-70 trend line.
tive to be an index of the general market.
Projecting beyond 1970 an earnings trend line rising
To project future DJIA earnings, we were fortunate indefinitely at an annual rate of 21/2 %-our accepted
in securing the estimates of several well-known experts. rate of ignorance-the 1960 Investment Value for the
Some expressed their projections in other indexes. To DJIA works out at 570.
make them all comparable, they were transposed into
the DJIA. Valuation Bases

The specialists used various approaches, including In the light of the DJIA rates of per share earnings
estimates of the Gross National Product and its normal growth, tabulated on Figure II, and of a similar com-
relations to corporate profits in periods of expanding pilation for various other periods in the section com-
and contracting business, as well as representative pro- paring the DJIA with the "500" Index, a projected
portions of earnings per DJIA index share to total growth rate of 5% for 1960-70 can hardly be con-
corporation earnings. Some took into account returns sidered unduly high. Yet Figure II produces an uneasy
on the book value of an index share and the probable feeling. Once again we are facing the same situation
growth rates of undistributed earnings. Others empha- as we did a year ago in our study of General Motors
sized margins of profit, their probable relation to GNP (The Analysts Journal, May, 1959). This time, how-
during the entire period envisaged, and their probable ever, we are dealing with the general market, not with
time-leads with respect to earnings. a single stock.
One of these friendly experts, whose authority and The visual impression is strong that the projected
experience are second to none, made perhaps the most trend lines of earnings are too high, suggesting that an
painstaking projection of all. He separately considered appraisal based on them must necessarily cause drastic
such factors as civilian and defense production, and overvaluation.
worked with statistically-grounded assumptions as to We know of course that a secular trend line of earn-
population growth, productivity rates, wage rates, and ings beginning in 1871 could not be a base for com-
manufactured goods prices. He also estimated for each puting present investment value. It is merely an average
year, on the basis of past studies of a large sample of of countless other trends that have characterized differ-
industrial companies, the most important items of cost ent periods of varying durations. It courses over almost
as well as direct production labor. He projected depre- a century. During this long period earnings were alter-
ciation by special methods consistent with assumed nately under the influence of all conceivable conditions
growth of production and prices. affecting national life-foreign wars and domestic strife,
Interestingly enough, he alone, of all the economists staggering depressions and ballooning inflations as well
consulted, foresaw a deep recession in the early 1960's. as peaceful prosperity and moderate cyclical fluctua-
but also a sharp upturn from the projected 1964 trough tions. The average secular growth rate is the consum-
to "catch up" with the GNP trend line. A dotted curve mation, reflection, and description of them all, each and
on Figure II features his individual views. Beginning every one. This is why it is suitable to be a common
with 1963, it is traced against the average expression of denominator, a "rate of ignorance" by which to project
the summarized expert opinion, including his own. the unknown future, thereby placing the appraisals of
This writer did not solicit projections beyond 1970. different stocks on the same comparative basis. And it
In fact, most of the contributing economists did not is entitled to remain such until the historical rate shifts.
estimate actual earnings beyond 1961. After that Conversely, the level and the growth rate of the secular
point, they projected trend lines or rates of growth. trend are unsuited to be the capitalization bases of to-
All designated 1960 as a cyclical peak of corporate day, because they stand as much for times and condi-

84 THE FINANCIAL ANALYSTS JOURNAL

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
 FIGURE II

900r . I =
700 1 tt =
800 P--R - - IC- E PRICE RA N G E S OF DJ IA
500 Jil_

400 .I I -- 4-4-7 I _

IIl~ I A R

90 EANIG TRN LINEi0ti 3 :1 1 301

a ??0 v - -- tt t e v -t . t3

80.
8 0 3 -4- 3

40 DJ IA EARNI NGS 3 ; 3
P ER S HA R E ~ ~

30 - - -

20 ..I __

8? ANNUAL G RO0W TH RA TE S
7 : , , : - OF E A R N I N G S T R E N D L I IES
8 Y 1. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.-.~...... .

MA
S
-JN
1937
196 -194
85ft g 31
; i;0I
01%
0 : X 0g 0
1941-194 5 5
I . ~~~~~~1945- 1948 19.5%V
1948 -1955
1955- 198 0 1.8%V-
1 ~~19860-19 70 0%PRJ
2 0 FROM 193
0 70 a.5%. INDEF.
0

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
 tions quite different from our own. Today's valuation mentioned, they provide the data for the correlation
bases can be only at a level and at rates commensurate formula used to compute theoretical dividends.
with the economics of our time. The enlargement of the dimensions of the country's
economy is reflected on Figure III in rising levels of
RISING LEVELS corporate dividends and earnings. Despite all setbacks,
"through chances various, through all vicissitudes"- to
At annual intervals, from the inception of the period
borrow Aeneid's phrase from the title page of Benjamin
we study, Figure 111 traces 21-year trend lines through
Graham's The Intelligent Investor-their respective
the curves of dividends and earnings. The first year of
trend lines were being steadily lifted. Since dividends
the period being 1871, the mid-point of the first trend are relatively more stable, the rise to higher plateaus is
line is placed at 1881. Between 1871 and 1955, there even more evident in their case. We did not project
are 65 such trend lines. Their respective rates of growth future dividends, but their new level may be visualized
are shown in Note 8 of the Appendix. As previously from analogy with earnings.

FIGURE III

50.04 - a m:

To.o-

j~~~~~~~~~~~~~~
60.00
7.0

40.00-

40.0 0 .-....... .... . ..

2 0 .0 0 ........ .... .-...... ...

....-.....A

.10-

7.00 . .... .. . ............ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ .0

3.00~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1

9 0 . ......N~ 0 * r
a C ) * 0

860 THEFIANIALANLYTS.OUNA

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
 Figure III demonstrates that, in order to have signifi- held that industrial history reveals very long cyclical
cance as a current valuation base, a trend line of earn- waves containing several 9-10 year major cycles, each
ings must lie higher than their long-term secular trend, divisible in turn into three minor cycles lasting roughly
and higher even than the shorter trend lines averaging 40 months.
the earnings of such recent periods as the 1930's or A three-cycle scheme implies that the cycle of cycles
1940's. But how high? And at what point, in particular, contains four to six major cycles marked off on each
may it be, in turn, extended by an indefinite projection side by severe recessions; it may be assumed that the
of the rate of ignorance? all-inclusive and highest unit of cyclical fluctuations-
Figure II produces the impression that the 1970 the very long cyclical wave-is itself bordered by par-
point of $57 per share of DJIA earnings-standing at ticularly deep depressions. As the rhythm of such long
the end of the experts' projection-may be too high for waves is believed to be 40 to 60 years, an interval of
splicing to it an indefinitely extended 21/2 % annual approximately 40 years would seem an acceptable min-
growth rate. We might investigate, therefore, the impli- imum. Such a schedule could then list 1894, 1932 and
cations of starting the indefinite projection at the 1960 1973 as marking the lows of earnings declines bordering
inception of the experts' trend line, rather than at its the long waves, or cycles of cycles.
1970 terminal point. As can be seen on Figure II, this If a depression of the size pictured on Figure III
10-year trend line's base is at $35 of DJIA earnings. should be really in the cards for the 1970's, DJIA's
To make this test, we shall use the long historical overvaluation at its recent highs would have been on
perspective of Figure III. Since DJIA does not reach the order of 45 % and could justify a compensating cor-
far enough into the past, Figure III was constructed rection down to the 260 level. Assuming that we are
in terms of the "500" indexes. It utilizes the experts' presently in the early stages of so long a journey into
1960-1970 earnings trend line followed by a hypothe- the night, we are facing a bear market lasting several
sized severe decline. By precipitating projected earn- years which would bring us back to the lows of 1952
ings from their 1970 high into the abyss of a vicious and 1953 or to the average level of 1951. In the dis-
depression which makes them decline more than 80% passionate historical perspective of Figures II and III,
and, at their low, makes them touch bottom with some the decline of a stock average to its 1951 annual mean
of the deepest earnings troughs of the preceding 100 price would not be out of line with the past. And it
years, the present earnings plateau would level out at would also close the tremendous gap on the scale of
just about the 1960 base of the experts' projection. If prices on Figure I, which is so glaring above 1954 and
we extend the 21/2 % rate of ignorance indefinitely from seems to so completely detach, on that diagram, the last
this point, the resulting 1960 appraisal of the DJIA, few years from the entire previous experience, shooting
475, would be 95 points below the valuation based on them out into space. Nature abhors void and may well
the much higher 1970 point of the experts' trend line. choose to fill it.
Shattering depressions have happened before. We
Long Cycles and Deep Depressions
cannot guarantee non-recurrence. Moreover, since a
Though unpleasant to contemplate, the possibility of 1960 investment value of 475 for DJIA is itself based
such a severe depression can be rationalized. We can on the hypothesis of a deep depression in the 1970's, its
argue that by 1970 the recent powerful forces of eco- eventual onslaught-should it materialize-would in-
nomic growth may reach or pass their peak. Outside evitably be reflected in another drastic cyclical fall of
pressures could then conceivably push the American stock prices soon after their presumably sharp recovery
economy on a downward path. The economic masses from the lows of, say, 1964. Such an expectation would
now emerging in Europe could become powerful rivals. tie in rather nicely with DJIA's 1960-70 earnings pro-
This is also likely to be true of Japan and, later, even of jection expressing the minority view of one expert as
China. Soviet Russia could dump vast quantities of noted on Figure Il. Should his prophecy sadly come
cruelly competitive goods on the free world's markets. true and, in addition, a deep depression occur at the
According to press reports, Allen W. Dulles, Director of end of the period covered by Figure III, the DJIA
the Central Intelligence Agency, testifying last winter would reach by the late sixties a mountain top that
before the Joint Economic Committee of the Congress, could well stand as a lonely peak for the generation liv-
warned: "We should frankly face up to very sobering ing on its foothills.
implications of the Soviet economic program and the
A More Hopeful View
striking progress they have made over the last decade
. . . If the Soviet industrial growth rate persists at 8% How sound are the grounds for forecasting a deep
or 9% per annum over the next decade, as is forecast, depression for the early 1970's? In the light of the ex-
the gap between our two economies by 1970 will be haustive statistical investigations by the late Wesley C.
dangerously narrowed unless our own industrial growth Mitchell and Arthur F. Burns-when they worked to-
rate is substantially increased from the present pace." gether at the National Bureau of Economic Research-
some cumulation of successive cycles does seem to exist,
These forebodings of inclement economic weather
but the relations between cyclical waves of various
building up to storms fit rather well into certain as-
lengths are irregular.
sumptions of academic thinkers. Some of the greatest
theoretical economists, specializing in business cycles, Corporate earnings have sometimes avoided severe

MAY-JUNE 1960 87

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
depressions for long intervals: 27 years elapsed between reached and exceeded by the market late in 1958. As
the troughs of 1894 and 1921. Yet only 11 years sep- this article goes to press, at the beginning of April 1960,
arated the latter from the depths of 1932; and the steep stock prices, which have been declining since the turn
decline of 1938 followed within just another six years. of the year, have not receded to the 570 level. In other
On the whole, statistical explorers of business cycles be- words, in terms of this appraisal of its investment value,
lieve that it is vain to strive for a general theory. Each the DJIA has remained overvalued for almost a year
crisis, they contend, is an episode brought about by and a half, despite the corrections that have been taking
endlessly different combinations of conditions. place. At DJIA highs of early this year and late last
As we all know, cyclical fluctuations have recently summer, this overvaluation was about 20%. If the bal-
been much milder. While the three recessions that oc- ance should be redressed, eventually bringing the DJIA
curred during the postwar period were unequal in in- to a 20% undervaluation, it would decline to about 455.
tensity and duration, none was comparable in severity
Stock Values and Stock Prices
with the earlier depressions, such as those whose earn-
ings troughs-within the period covered by Figure III PART II of this study will consider the practical ap-
-were in 1876, 1894, 1921, 1932 or 1938. plications of investment values of averages of the general
market, as well as of individual stocks. It will also trace
Many economists believe that this happy situation
their price orbits and discuss the relations- of value and
will continue. In a recent address, Arthur F. Burns is
price, the effects of cyclical fluctuations in business on
reported as stating "the business cycle is unlikely to be
stock prices, and the techniques of stock market timing.
as disturbing or troublesome to our children as it once
was to us or our fathers." He attributed the improve- In the meantime, a quotation from an article by Pro-
ment to many structural changes in the economy; among fessor Robert H. Wessel, of the University of Cincinnati,
them, to the well publicized "automatic stabilizers" and summarizes well the reasons why students of stock
to the way in which consumer incomes hold up when prices need an independent measure of value:
production is declining. The Rt. Hon. Hugh Gaitskell "It is my belief that any attempt to predict where the
expressed a similar opinion earlier this year on a TV stock market is going should begin with some sort of an
"Open End" program. As a visitor to these shores and estimate of where it now is. Although some writers try
leader of the British Labour Party, there is no reason to to sidestep this question, to me it seems unavoidable.
suspect him of either wishful thinking or capitalistic When one assesses the current position of the market he
bias. necessarily employs some sort of standards that help
Statistically, this view supports the reasonableness of him judge whether it is too high, too low, or at a reason-
an indefinite extension of the earnings trend line at a able level. Any such discussion is most meaningful
21/2 % annual growth rate from the 1970 terminal point when the writer explicitly states what those standards
of the experts' projection shown on Figure II. And it are, thereby enabling the reader to follow his analysis
still allows for an earnings decline of some 40% at the without doubt or confusion.
end of the period on Figure III. To repeat, it gives 570 .... It seems to me that both the prices of individual
as a 1960 valuation for DJIA. stocks and the levels of the market as a whole should be
We are reverting to this appraisal once more-not evaluated in terms of the earning power estimates and
because it is more optimistic, but because it is more in capitalization rates implicit in these prices or market
tune with the collective view of men well trained and levels." (Today's Stock Market Values and the Longer
experienced in our profession. By splicing the rate of Term Prospects, The Commercial and Financial Chron-
ignorance to their projection, we should attain a more icle, Feb. 18, 1960)
representative appraisal of investment value. To as-
sume a deep depression may perhaps seem more con-
servative and cautious and, therefore, a more suitable
guide for wise and prudent men. Yet it represents no
more than an analogy with the past that may not be
applicable to our times.
No appraisals of investment values are meant as
Judgment Day verdicts. They are merely expressions of
present conditions and reasonable estimates of the
future. If both change, investment values are adjusted.
Indeed, values must change inevitably. The future be-
comes clearer as time passes. Today's appraisals of in-
vestment values for 1965 or 1970 will gradually change
with each successive year.
To soothe the pessimists we point out that even our
more favorable appraisal of the investment value of
DJIA was far exceeded by actual stock prices during
the bull market that had its inception in October 1957
from a level of just under 420. The 570 level was Value is on the Ground-NOT in the Stars

MAY-JUNE 1960 89

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
 To construct an independent standard of value, we FIGURE IV

took great pains not to inject alien elements. Not only

did we keep our standard free from price itself, which
-108
it is expected to measure, but also from any indirect
mixtures. Our capitalizers are only related to the char- |~~~~~~ ~ ~ ~ T 0 C K P R I C E _ 5 108
acteristics of the earnings themselves; our discount rate ,~~~~~~~~~~ ~ ~ ~ ~ ... .. ... .. ..... ,..".....llllllllllllllllll IIIIIIIILLIIIIIIII10

is firmly rooted in the effective return from stocks. ' ! t ' 6 ^ - _ . _ 1 H H I . _ . e ' .104

Neither of these factors can be affected by money mar-

ket changes which are as unrelated to stock values as 102

are price fluctuations. And by keeping both the dis-

100
count rate and the rate of growth for projecting future
earnings beyond the horizon of visibility fixed, we not INDEX OF' DEMAND
8 0 .... . - ------
only attain comparable investment values, but also cut
down the number of variables entering into the valua- 50 6?t~~~~-
~ ~~~~~
- -IH 9
-_.
tion equation.
Our excursions into the unknown future are less ad- 40 '

venturous than they may seem. Valuation of Common

30%
Stocks discusses at length the role and the effects of the
discount rate on appraisals of present worth. By dis-
UPSIDE VOLUME
counting future payments at our 71/2 % rate, we reduce
$1 to a dime within 32 years; and an additional decade
shrinks it to a nickel. Thus, while the mathematical I iI2 I t I I I I ?tIII I IItiIM I ! I I I I t
*500,090
formulae used for the computations must be geared to
handle infinity, the practical limits of our problems are
II: tU{ iL!2l Illil I I I I DONIEVOLUME
I 1 I I I I I
much less majestic and forbidding. But even within the o o o o*0
to 0in0 to
00 0 00oooo oo000
to oo oo o o 0 0

confines of this limited future, Financial Analysts should

CM l N t10 N - - N N N Nm ) 0)
not let their imaginations run wild. By unwarranted as-
W E E K S E N DI N G:
sumptions they may be erecting Babylonian towers of
value instead of reaching legitimate appraisals. Even
relatively small changes in the multipliers, in the dis-
61, by the National Bureau of Economic Research, in
count rate, in the estimated rates of future earnings
the fall of 1958, Geoffrey H. Moore relates that such
growth, and in the number of years of their projected
duration can cumulate to surprisingly large effects, research "began during the recession of 1953-54, when
especially when the changes are all in the same direction. tables comparing the percentage changes in a long list
of economic series during the current and preceding
Business Cycles recessions were prepared for Arthur F. Bums, then
As a transition to PART II, we should stress that in- Chairman of the Council of Economic Advisers.
vestment value is only one of the factors shaping the "In October 1957 R. J. Saulnier, present Chairman of
prices of stocks. Another vital determinant is the busi- the Council, requested the assistance of the National
ness cycle. Stock prices are among the most sensitive Bureau in preparing a similar set of tables. Tables cov-
economic time series. In their later phases, cyclical ex- ering some 70 monthly and quarterly economic series
pansions and contractions bring about substantial over- were promptly prepared. . . . The electronic computer
pricing and underpricing of stocks. program was developed with the aid of a grant from the
In its March 16, 1960 issue, The Cleveland Trust National Science Foundation. The International Busi-
Company Business Bulletin reminds us that the average ness Machines Corporation generously contributed
length of the nine complete cycles of industrial produc- machine time on the 704 computer. These resources
tion, since the end of World War I, was 42 months, were essential to the pursuit of the study. No less es-
consisting of 27 months on the upside and 15 down. sential were the intellectual and financial resources that
The upward phase of the current cycle has already have over the years been invested in the National Bu-
rounded out two years; soon it will be living on bor- reau's studies in business cycles. The present report is,
rowed time. And since stock prices usually lead busi- in the truest sense, a product of these contributions, for
ness at cyclical turns, the present picture seems to fall the data and methods used here are virtually all derived
neatly into a cyclical pattern. in one way or another from this earlier work."
Because of their wide variations in amplitude and The reward of these experiments was that in the
duration, and the great complexity of the interrelations spring of 1958, when even the strongest and best in-
of economic factors, there are no reliable ways for fore- formed institutional investors were still wallowing in
casting cyclical turning points. But electronic computers deep pessimism and dejection, the Council of Economic
could be an invaluable aid in keeping abreast of devel- Advisers was already well aware that the recession was
opments almost simultaneously with their unfolding. In unlikely to last much longer. Since so much is at stake,
Measuring Recessions, published as Occasional Paper we can safely predict that the use of these techniques

90 THE FINANCIAL ANALYSTS JOURNAL

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
will become accepted practice among large investors D. Earnings for 1932 were derived from the Cowles

and that the problem of cost will be solved, if necessary, Commission Earnings-Price Ratio in order to give
recognition to the many deficits in Earnings that
by collective action.
year. The S. & P. Earnings figure was arrived at
by using zero for all deficits. This produced an
Supply and Demand
Earnings figure of 0.41 and a Price-Earnings Ratio
A further refinement of timing may be sought in of 16.9 as compared with Earnings of 0.05 and a
Price-Earnings Ratio of 138.9 when the C(owles
direct analysis of demand for and supply of stocks.
Earnings-Price Ratio is used. The latter is more
Many methods exist. Figure IV offers an illustration.
representative of the trough of this severe bear
While presented in graphic form, Figure IV has no market.
relation to charts of price action mentioned earlier in E. Price-Earnings Ratios from 1926 to date were ar-
this paper. It attempts to measure the relative strength rived at by dividing the S. & P. Prices by the
S. & P. Earnings for the year with the exception
of buying and selling; it is not preoccupied with be-
of 1932, as explained in D above.
havior patterns.
F. Prior to 1926 the Price-Earnings ratios are the
Stock market analysis should be used with great dis- reciprocals of the Cowles Commission Earnings-
cretion. It deals with surface conditions. During a Price ratios.
period of basic economic change, they may dissolve G. In all cases, Price-Earnings Ratios are computed
without much impact on the prices of stocks. from average annual Price and the year's total
Earnings.
The curve of stock prices on Figure IV traces the
author's composite index of 70 leading equities repre- Prices, Earnings, and Price-Earnings Ratios, 1871-1959

senting a wide range of economic activity. It includes Prices Earnings P/E

utilities and rails as well as industrials. The weekly
1871 - -3.90 .34 11.6
volumes are adjusted for stock splits and for the number
1872 - -4.09 .35 11.6
of trading days when holidays occur. 1873 - -3.85 .37 10.5
The Index of Demand is the ratio of moving totals 1874 - -3.68 .37 10.0
1875 3.58 .29 12.3
of daily upside volume to the moving totals of all the
1876 - -3.21 .22 14.5
shares traded in our 70 stocks. 1877 - -2.50 .24 10.5
Recently, the Index of Demand was acting better 1878 - -2.70 .25 10.8
than the Index of Stock Prices. When the latter made 1879 - -3.36 .31 10.9
1880 - -4.33 .41 10.6
its low, the Index of Demand was already in an uptrend.
1881 - -5.66 .40 14.1
Through the week ending April 1, it was revealing as 1882 - -5.52 .40 13.8
yet no precursory weakness as it did at the end of 1959. 1883 - -5.27 .37 14.2
In PART II we shall bring Figure IV up to date and 1884 - -4.41 .29 15.4
1885 - -4.29 .25 17.0
discuss at greater length the possibilities and limitations
1886 - -5.00 .31 16.3
of market analysis. 1887 - -5.16 .34 15.3
1888 - -4.80 .24 20.0
(Part II of this presentation will be published in the 1889 - -4.92 .27 17.9
July-August issue of The Financial Analysts Journal- 1890 - -4.90 .27 18.4
Editor). 1891 - -4.74 .32 15.0
1892 - -5.17 .34 15.2
APPENDIX 1893 - -4.43 .24 18.1
1894 - -4.03 .15 27.0
References and Notes 1895 - -4.14 .22 18.5
1. The Exchange, a monthly magazine published by the 1896 - -3.87 .19 20.1
1897 - -4.03 .28 14.5
New York Stock Exchange, December, 1944.
1898 - -4.59 .32 14.5
2. Data for Figure 1. 1899 - -5.82 .45 13.0
A. Prices: Standard & Poor's "500" Prices (1941- 1900 - -5.72 .44 12.9
1943 = 10) were used from 1918 to date. Prior to 1901 - -7.64 .49 15.7
1918, Prices are those of the Cowles Commission 1902 - -8.22 .61 13.4
Index Series PEa-i (i.e. Prices of stocks for 1903 - -7.01 .52 13.5
which Earnings data are available) converted to 1904 - -6.86 .47 14.5
the 1941-1943 = 10 base. The annual figures are 1905 - -8.90 .66 13.4
arithmetic averages of monthly mean Prices. 1906 - -9.61 .76 12.6
B. Earnings from 1926 to date are those published by 1907 - -7.89 .67 11.8
Standard & Poor's for the composite "500" Index. 1908 - -7.83 .58 13.5
They are twelve months' moving totals through the 1909 - -9.79 .77 12.7
fourth quarter and thus represent Earnings for 1910 - -9.37 .73 12.8
the full calendar year specified in each case. (See 1911 - -9.24 .59 15.6
D below.) 1912 - -9.52 .70 13.6
1913 8.50 .63 13.4
C. Earnings prior to 1926 were compiled by multiply-
1914 - - 8.07 .52 15.6
ing Prices by the Cowles Commission Earnings-
1915 - - 8.39 .89 9.5
Price Ratios. These ratios are total reported Earn-
1916 - - 9.56 1.55 6.2
ings in each calendar year divided by total stock
values as represented by an average of monthly
1917 - - 8.6.3 1.30 6.6
Prices for the year. (Continued on next page)

MAY-JUNE 1960 91

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms
Prices, Earnings, and Price-Earnings Ratios, 1871-1959 8. Growth Rates of Earnings and Dividends, 1871 - 1959
Continued Cowles Commission Indexes Extended as Indicated in
Valuation of Common Stocks
Prices Earnings P/E
Growth Rates Are Computed foi 21-Year Trend Lines
1918 - -7.54 .99 7.6 Periods Earnings Dividends
1919 - -8.78 .93 9.4 1871-91 --- - 0.64% -0.66%
1920 - -7.98 .80 9.9 1872-92 --- - 0.40 -0.65
1921 6.86 .29 23.7 1873-93 - 0.51 -0.42
1922 8.41 .69 12.1 1874-94 - 1.17 -0.43
1923 - 8.57 .98 8.8 1875-95 - 1.17 -0.53
1924 9.05 .93 9.7 1876-96 - 1.64 -0.73
1925 11.15 1.25 8.9 1877-97 --- - 1.98 -0.91
1926 12.59 1.24 10.1 1878-98 --- - 2.06 -1.55
1927 15.34 1.11 13.8 1879-99 --- - 1.67 -2.14
1928 19.95 1.38 14.5 1880-1900 --- - 1.04 -2.23
1929 26.02 1.61 16.2 1881-01 --- - 0.02 -1.91
1930 21.03 .97 21.7 1882-02 + 1.37 -0.94
1931 13.66 .61 22.4 1883-03 + 2.45 +0.06
1932 6.93 .05 138.9 1884-04 + 3.28 + 0.91
1933 8.96 .44 20.4 1885-05 + 4.14 + 1.91
1934 9.84 .46 21.4 1886-06 + 4.89 + 2.74
1935 - - 10.60 .65 16.3 1887-07 + 5.62 + 3.51
1936 - - 15.47 1.03 15.0 1888-08 + 6.20 + 4.22
1937 15.41 1.20 12.8 1889-09 + 6.57 +4.85
1938 11.49 .54 21.3 1890-10 --- + 6.91 + 5.33
1939 12.06 .82 14.7 1891-11 --- + 6.79 +45.70
1940 11.02 1.04 10.6 1892-12 --- + 7.05 + 6.06
1941 9.82 1.13 8.7 1893-13 --- + 7.17 +6.41
1942 8.67 .91 9.5 1894-14 + 6.42 +6.58
1943 11.50 .83 13.9 1895-15 + 5.56 +6.34
1944 12.47 .84 14.8 1896-16 + 5.84 + 6.14
1945 - - 15.16 .88 17.2 1897-17 + 5.39 + 6.05
1946 - - 17.08 .99 17.3 1898-18 ---+ 4.86 + 5.47
1947 - - 15.17 1.61 9.4 1899-19 + 4.28 +4.70
1948 - - 15.53 2.29 6.8 1900-20 --- - 3.85 + 3.94
1949 - - 15.23 2.37 6.4 1901-21 --- + 1.98 +43.30
1950 - - 18.40 2.83 6.5 1902-22 --- + 1.47 +42.85
1951 - - 22.34 2.45 9.1 1903-23 --- - 1.71 +42.62
1952 - - 24.50 2.41 10.2 1904-24 --- + 1.58 +42.38
1953 - - 24.73 2.65 9.3 1905-25 --- - 1.63 +42.00
1954 - - 29.69 2.76 10.8 1906-26 -+ 2.04 + 1.89
1955 - - 40.49 3.54 11.4 1907-27 + 2.41 + 2.00
1956 - - 46.62 3.35 13.9 1908-28 -+ 2.81 + 2.32
1957 - - 44.38 3.28 13.5 1909-29 -+ 3.03 + 2.56
1958 - - 46.24 2.84 16.3 1910-30 -+ 2.84 + 2.79
1959 - - 57.32 *3.39 est. 16.9 1911-31 -+ 1.33 +42.76
1912-32 - 3.11 +42.08
*This is Standard & Poor's most recent estimate of 1959 1913-33 - 4.62 +41.10
Earnings for the "500" index. Being lower than the esti- 1914-34 - 5.80 +40.29
mate obtained at the time Figure I was prepared, the 1915-35 - 6.70 -0.56
adjusted 1959 P/E is higher than that shown on the 1916-36 --- - 6.14 -0.73
scatter diagram. 1917-37 --- - 4.65 -0.48
1918-38 --- - 3.96 -0.33
3. Op. cit., p. 93. This volume was published in 1937 by 1919-39 --- - 3.14 -0.29
the Whittlesey House, a division of the McGraw-Hill 1920-40 --- - 2.15 -0.23
Book Company, Inc. It was a pioneering work. The 1921-41 --- - 1.15 -0.15
late Mr. Skinner's influence continues to grow through 1922-42 - 1.75 -0.41
the increasing numbers of followers applying methods 1923-43 - 1.35 -0.59
originated by him for measuring relations between 1924-44 --- - 0.50 -0.69
stock market movements and money flows as reflected 1925-45 --- 0.23 -0.75
by banking and credit trends. 1926-46 -+ 1.66 -0.51
1927-47 + 3.61 + 0.03
4. See the stimulating paper contributed to the March
1928-48 + 5.72 + 0.87
1959 issue of The Journal of Finance by Harry V.
1929-49 -+ 7.89 + 1.91
Roberts of the University of Chicago.
1930-50 4- 10.60 +3.25
5. "Stock Prices and Current Earnings", The Analysts 1931-51 -+12.32 +44.50
Journal, August, 1955. 1932-52 -+12.38 +45.43
6. F. W. Taussig, Principles of Economics, (The Mac- 1933-53 ---+ 9.01 + 5.61
millan Company, New York, 3rd ed., 1926), Volume II, 1934-54 --- - + 8.04 + 5.41
p. 45. 1935-55 - - 7.48 4- 5.31
7. Frederick R. Macaulay, "Some Theoretical Problems 1936-56- - - 4 7.17 4- 5.17
Suggested by the Movements of Interest Rates, Bond 1937-57- - + 7.47 4- 5.54
Yields and Stock Prices in the U. S. since 1856." (Na- 1938 -58- - - 4 7.34 4- 5.92
tional Bureau of Economic Research, 1938), pp. 129-30. 1939-59- - - 4 6.74 4- 5.90

92 THE FINANCIAL ANALYSTS JOURNAL

This content downloaded from

65.88.89.49 on Tue, 17 Aug 2021 08:41:29 UTC
All use subject to https://about.jstor.org/terms