The EconomistNovember 14th 2020 BriefingInvestment strategies 592
1sighted managers of the endowment funds
of universities. They saw that equities had
advantages over bonds—notably those
backed by mortgages, railroads or public
utilities—which had been the preferred as-
set of long-term investors, such as insur-
ance firms.
This new church soon had two doctrinal
texts. In 1934 Graham published “Security
Analysis” (with co-author David Dodd), a
dense exposition of number-crunching
techniques for stockpickers. Another of
Graham’s books is easier to read and per-
haps more influential. “The Intelligent In-
vestor”, first published in 1949, ran in re-
vised editions right up until (and indeed
beyond) Graham’s death in 1976. The first
edition is packed with sage analysis, which
is as relevant today as it was 70 years ago.
Underpinning it all is an important dis-
tinction—between the price and value of a
stock. Price is a creature of fickle senti-
ment, of greed and fear. Intrinsic value, by
contrast, depends on a firm’s earnings
power. This in turn derives from the capital
assets on its books: its factories, machines,
office buildings and so on.
The approach leans heavily on company
accounts. The valuation of a stock should
be based on a conservative multiple of fu-
ture profits, which are themselves based
on a sober projection of recent trends. The
book value of the firm’s assets provides a
cross-check. The past might be a crude
guide to the future. But as Graham argued,
it is a “more reliable basis of valuation than
some other future plucked out of the air of
either optimism or pessimism”. As an extra
precaution, investors should seek a margin
of safety between the price paid for a stock
and its intrinsic value, to allow for any er-
rors in the reckoning. The tenets of value
investing were thus established. Be conser-
vative. Seek shares with a low price-earn-
ings or price-to-book ratio.
The enduring status of his approach
owes more to Graham as tutor than the rep-
utation he enjoyed as an investor. Graham
taught a class on stockpicking at Columbia
University. His most famous student was
Mr Buffett, who took Graham’s investment
creed, added his own twists and became
one of the world’s richest men. Yet the sto-
ries surrounding Mr Buffett’s success are as
important as the numbers, argued Aswath
Damodaran of New York University’s Stern
School of Business in a recent series of You-
Tube lectures on value investing. The bold
purchase of shares in troubled American
Express in 1964; the decision to dissolve his
partnership in 1969, because stocks were
too dear; the way he stoically sat out the
dotcom mania decades later. These stories
are part of the Buffett legend. The philoso-
phy of value investing has been burnished
by association.
It helped also that academic finance
gave a back-handed blessing to value in-vesting. An empirical study in 1992 by Eu-
gene Fama, a Nobel-prize-winning finance
theorist, and Kenneth French found that
volatility, a measure of risk, did not explain
stock returns between 1963 and 1990, as ac-
ademic theory suggested it should. Instead
they found that low price-to-book shares
earned much higher returns over the long
run than high price-to-book shares. One
school of finance, which includes these au-
thors, concluded that price-to-book might
be a proxy for risk. For another school, in-
cluding value investors, the Fama-French
result was evidence of market inefficien-
cy—and a validation of the value approach.
All this has had a lasting impact. Most
investors “almost reflexively describe
themselves as value investors, because it
sounds like the right thing to say”, says Mr
Damodaran. Why would they not? Every in-
vestor is a value investor, even if they are
not attached to book value or trailing earn-
ings as the way to select stocks. No sane
person wants to overpay for stocks. The
problem is that “value” has become a label
for a narrow kind of analysis that often
confuses means with ends. The approach
has not worked well for a while. For much
of the past decade, value stocks have lagged
behind the general market and a long way
behind “growth” stocks, their antithesis
(see chart 1). Old-style value investing looksincreasingly at odds with how the econ-
omy operates.
In Graham’s day the backbone of the
economy was tangible capital. But things
have changed. What makes companies dis-
tinctive, and therefore valuable, is not pri-
marily their ownership of physical assets.
The spread of manufacturing technology
beyond the rich world has taken care of
that. Any new design for a gadget, or gar-
ment, can be assembled to order by con-
tract manufacturers from components
made by any number of third-party fac-
tories. The value in a smartphone or a pair
of fancy athletic shoes is mostly in the de-
sign, not the production.
In service-led economies the value of a
business is increasingly in intangibles—
assets you cannot touch, see or count easi-
ly. It might be software; think of Google’s
search algorithm or Microsoft’s Windows
operating system. It might be a consumer
brand like Coca-Cola. It might be a drug
patent or a publishing copyright. A lot of
intangible wealth is even more nebulous
than that. Complex supply chains or a set of
distribution channels, neither of which is
easily replicable, are intangible assets. So
are the skills of a company’s workforce. In
some cases the most valuable asset of all is
a company’s culture: a set of routines, pri-
orities and commitments that have been
internalised by the workforce. It can’t al-
ways be written down. You cannot easily
enter a number for it into a spreadsheet.
But it can be of huge value all the same.A beancounter’s nightmare
There are three important aspects to con-
sider with respect to intangibles, says Mr
Mauboussin: their measurement, their
characteristics, and their implications for
the way companies are valued. Start with
measurement. Accounting for intangibles
is notoriously tricky. The national ac-
counts in America and elsewhere have
made a certain amount of progress in grap-
pling with the challenge. Some kinds of ex-
penditure that used to be treated as a cost of
production, such as r&dand software de-
velopment, are now treated as capital
spending in gdpfigures. The effect on mea-
sured investment rates is quite marked (see
chart 2). But intangibles’ treatment in com-
pany accounts is a bit of a mess. By their na-
ture, they have unclear boundaries. They
make accountants queasy. The more lee-
way a company has to turn day-to-day costs
into capital assets, the more scope there is
to fiddle with reported earnings. And not
every dollar of r&d or advertising spend-
ing can be ascribed to a patent or a brand.
This is why, with a few exceptions, such
spending is treated in company accounts
as a running cost, like rent or electricity.
The treatment of intangibles in mergers
makes a mockery of this. If, say, one firm
pays $2bn for another that has $1bn of tan-Not by the book
Russell 3000 stockmarketindex,totalreturns
November 1st 2010=100Source:RefinitivDatastream5004003002001000
20191817161514131220101MainValueGrowthLet’s get non-physical^2Sources: BEA; BloombergUnited States, intellectual property investment% of GDP% of non-residential
fixed investment1970 80 90 2000 10 2001234501020304050