SSRN Id3254580 PDF

EQUITY VALUATION
Science, Art, or Craft?
Joanne M. Hill, Dave Nadig, Matt Hougan

Frank J. Fabozzi, CFA, Sergio M. Focardi, and Caroline Jonas
With an appendix on international ETFs by Deborah Fuhr
Electronic copy available at: https://ssrn.com/abstract=3254580

EQUITY VALUATION
Science, Art, or Craft?
Frank J. Fabozzi, CFA, Sergio M. Focardi, and Caroline Jonas

Statement of Purpose
The CFA Institute Research Foundation is a

not-for-profit organization established to promote
the development and dissemination of relevant
research for investment practitioners worldwide.
Neither the Research Foundation, CFA Institute, nor the publication’s edito-
rial staff is responsible for facts and opinions presented in this publication.
This publication reflects the views of the author(s) and does not represent
the official views of the CFA Institute Research Foundation.
The CFA Institute Research Foundation and the Research Foundation logo are trademarks
owned by the CFA Institute Research Foundation. CFA®, Chartered Financial Analyst®,
AIMR-PPS®, and GIPS® are just a few of the trademarks owned by CFA Institute. To view
a list of CFA Institute trademarks and the Guide for the Use of CFA Institute Marks, please
visit our website at www.cfainstitute.org.
© 2017 The CFA Institute Research Foundation. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted,
in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior written permission of the copyright holder.
This publication is designed to provide accurate and authoritative information in regard to
the subject matter covered. It is sold with the understanding that the publisher is not engaged
in rendering legal, accounting, or other professional service. If legal advice or other expert
assistance is required, the services of a competent professional should be sought.
Cover Image Photo Credit: Jay_Zynism/iStock/Getty Images
ISBN 978-1-944960-33-9

Biographies
Frank J. Fabozzi, CFA, is professor of finance at EDHEC Business School

and a senior scientific adviser at EDHEC-Risk Institute (Nice, France). He is
editor of the Journal of Portfolio Management and a trustee for the BlackRock
closed-end fund complex. The author of numerous books on asset manage-
ment and four CFA Institute Research Foundation monographs, Professor
Fabozzi is the 2007 recipient of the C. Stewart Sheppard Award and the 2015
recipient of the James R. Vertin Award, both from CFA Institute. He was
also the 2002 inductee to the Fixed Income Analysts Society Hall of Fame.
Professor Fabozzi received his bachelor’s degree in economics and statistics
and his master’s degree in economics from the City College of New York and
his PhD in economics from the City University of New York.
Sergio M. Focardi is professor of finance at Leonard de Vinci University
(Paris-La Défense, France) and a founding partner of The Intertek Group. He
serves on the editorial board of the Journal of Portfolio Management and has co-
authored numerous articles and books, including the CFA Institute Research
Foundation monographs Investment Management: A Science to Teach or an
Art to Learn?, Investment Management after the Global Financial Crisis, and
Challenges in Quantitative Equity Management, and the award-winning books
Financial Modeling of the Equity Market: From CAPM to Cointegration and
The Mathematics of Financial Modeling and Investment Management. Professor
Focardi also co-authored Financial Econometrics: From Basics to Advanced
Modeling Techniques and Robust Portfolio Optimization and Management. He
received his degree in electronic engineering from the University of Genoa
and his PhD in finance from the University of Karlsruhe.
Caroline Jonas is a managing partner of The Intertek Group in Paris,
where she is responsible for research projects. Jonas is co-author of numer-
ous reports and books on finance and technology and of three CFA Institute
Research Foundation monographs, including Investment Management: A
Science to Teach or an Art to Learn? and Investment Management after the Global
Financial Crisis. She received her bachelor’s degree from the University of
Illinois at Urbana-Champaign.
© 2017 CFA Institute Research Foundation. All rights reserved. iii

Contents
Acknowledgments.................................................................................. vii
Foreword................................................................................................. ix
Preface. . ................................................................................................... xiii
1. Finance Theory and Equity Valuation.................................................. 1

2. Tools for Valuing Stocks: Intrinsic Value and Relative Value. . ............... 13
Net Present Value Models............................................................... 14
Relative Valuation and Market Multiples........................................ 21
Valuation Methods Compared. . ...................................................... 26
Additional Questions Concerning Valuation Methods................... 32
3. Valuing Hard-to-Value Equities............................................................ 42
Valuing IPOs. . .................................................................................. 42
Valuing Privately Held Companies. . ................................................ 54
4. Fair Value, Market Value, and Price Distortions.................................... 63
Central Bank Policies and the Market Rally..................................... 63
Corporate Buybacks and the Market Rally...................................... 67
5. New Tools for Equity Valuation............................................................ 76
Data, Tools, and Their Application.................................................. 80
Expectations and Limitations......................................................... 83
Impact of Big Data and AI in Asset Management........................... 87
6. Equity Valuation: Does It, Can It Fulfill Its Promise?.. ............................ 90
Equity Valuation in a Radically Changed Environment:
Does It Retain Its Value?. . ............................................................ 96
A New Role for Fundamental Analysis in Asset Management?. . ..... 99
References............................................................................................... 103
This publication qualifies for 2 CE credits under the guide-

lines of the CFA Institute Continuing Education Program.

Acknowledgments
The authors wish to thank those from academia and the industry who accepted
the challenge to articulate their views on equity valuation. Their perspectives
are cited and attributed throughout this monograph.
We are also grateful to the CFA Institute Research Foundation for
funding this project and, in particular, to its director of research, Laurence
B. Siegel, and its executive director, Walter (Bud) Haslett, CFA, for their
encouragement, assistance, and insightful comments.
© 2017 CFA Institute Research Foundation. All rights reserved. vii

Foreword
Where do stock prices come from? The easy answer is that they are the out-
come of supply and demand; that is, the price of any given stock is that which
causes the quantity supplied to equal the quantity demanded.
Of course, that answer, like most easy answers, is unsatisfying. What
causes the supply and demand for a stock to be what it is? There are two sets
of players in the game that we call the stock market: (1) the sell side, consisting
of issuers (seekers of capital) and their agents, called “investment bankers” or
“stockbrokers,”1 and (2) the buy side, consisting of saver-investors (providers of
capital) and their agents, called “investment managers.”
Each of these characters in the financial zoo presumably has a view on
every asset in the world, the view generally being that one should ignore the
asset. There are simply too many assets for everyone to analyze, so they delegate
the task of analyzing most assets to other investors, who—through their col-
lective wisdom, it is believed—will agree on a price that makes the asset a fair
deal; in that case, the asset is held in an index fund. (In an environment where
index funds exist, simply not holding an asset expresses a strong negative view on
that asset and is in no way equivalent to holding the asset at its index weight.)
It is only when an investor believes an asset offers a better-than-fair deal—
either because it is cheap, justifying an above-index weight, or because it is
expensive, requiring a below-index weight—that he or she becomes an active
manager with respect to that asset.
Active managers contribute to the price discovery process by shifting the
demand curve for that asset—outward, if they are buyers—in a way that an
index fund investor does not do. Active managers also affect the supply curve
by offering stocks for sale. Finally, sell-siders (corporations and their invest-
ment banker agents) may also be regarded as active managers in that they,
too, hold nonindex weights of the assets they trade and thus move the supply
and demand curves for stocks in exactly the same way that buy-siders do.
Well, in almost the same way. A corporation has only one stock to sell; an
investment manager can choose from among all the stocks offered for sale in
the world. Thus, the corporate issuer’s impact is more concentrated, and the
investment manager’s impact is more diffuse. But the underlying Economics
101 of asset price discovery is the same when viewed from either side.
1
They are also called “broker/dealers”—the word “dealer” highlighting the market-making
or principal function of the institution, in contrast to the buyer–seller matching or agent
function.
© 2017 CFA Institute Research Foundation. All rights reserved. ix

Equity Valuation
In a 2003 article, 2 Barton Waring and I described active managers as

forecasters—people who
use the information available to them, and whatever their native talents
are, to make stock-by-stock (or factor-by-factor, or market-by-market)
forecasts of pure active return based on information that they believe is
not yet impounded in the price, and then they translate these forecasts into
portfolios.
Through this activity they add alpha—or hope to. With the benefit of
15 years of additional reflection, however, I’d like to rephrase Barton’s and my
description of asset managers as forecasters to say that they are disagree-ers (if
that is a word). By making forecasts “based on information that they believe
is not … impounded in the price,” active managers are people whose fore-
casts are different enough from the consensus to justify, at least in their own
minds, betting other people’s money and their own reputations on the validity
of their disagreement.
It bears repeating that a majority of these disagreements with the con-
sensus are not justified: Most active managers are beaten by their bench-
marks over most periods and eventually eat humble pie. Yet there are active
managers—enough to be much more than a statistical accident—who do beat
the “market” (that is, their fairly chosen benchmark). Inquiring how these
successful managers make their correct nonconsensus forecasts is a worth-
while pursuit.
Frank Fabozzi, Sergio Focardi, and Caroline Jonas have, in a series of
remarkable and well-received CFA Institute Research Foundation mono-
graphs, developed a survey-based strategy for investigating important
issues in investment finance. In the past, they have turned their attention to
(1) quantitative finance, (2) quantitative equity investing, (3) the investment
management industry after the global financial crisis, and (4) the distinction
between art and science in the investment management profession.3
2
M. Barton Waring and Laurence B. Siegel, “The Dimensions of Active Management,”
Journal of Portfolio Management (Spring 2003): 35–51.
3
Frank J. Fabozzi, Sergio M. Focardi, and Caroline Jonas, Trends in Quantitative Finance
(Charlottesville, VA: CFA Institute Research Foundation, 2006); Frank J. Fabozzi, Sergio M.
Focardi, and Caroline Jonas, Challenges in Quantitative Equity Management (Charlottesville,
VA: CFA Institute Research Foundation, 2008); Frank J. Fabozzi, Sergio M. Focardi, and
Caroline Jonas, Investment Management after the Global Financial Crisis (Charlottesville,
VA: CFA Institute Research Foundation, 2010); Frank J. Fabozzi, Sergio M. Focardi, and
Caroline Jonas, Investment Management: A Science to Teach or an Art to Learn? (Charlottesville,
VA: CFA Institute Research Foundation, 2014).
x © 2017 CFA Institute Research Foundation. All rights reserved.

Foreword
In this, their fifth research monograph, the authors investigate how

analysts employed by active investment managers form their nonconsensus
beliefs—their conviction that a stock offers a better-than-fair deal either to
those who buy it (or overweight it) because it is cheap or to those who sell it
(or underweight it) because it is expensive. They interview analysts, portfolio
managers, directors of research, and chief investment officers at a variety of
firms in countries in both Europe and North America.
The authors’ insights into the “science, art, or craft” of equity analysis
reflect and amplify the diligent work of the many investment professionals
who agreed to open up their work lives in these interviews. We are delighted
to present this excellent monograph, and we hope to hear from these authors
many more times in the future.
Laurence B. Siegel
Gary P. Brinson Director of Research
CFA Institute Research Foundation
November 2017
© 2017 CFA Institute Research Foundation. All rights reserved. xi

Preface
Fundamental analysts and fundamental active managers believe that one can
determine the intrinsic (fundamental) value of a company’s stock by analyzing
the company. They argue that their ability to estimate the difference between
a stock’s fundamental price and its market price allows them to outperform
the market, thereby keeping markets at least somewhat efficient while creat-
ing value (alpha) for their clients. Much of the academic community agrees
and has developed valuation models, such as present value models based on
fundamental analysis. Such models include the widely used dividend discount
model, market-multiplier models (such as the popular price-to-earnings
models), and asset- and options-based valuation models.
Studies show, however, that on average and over time, despite all the
fundamental research, active traditional managers fail to outperform mar-
kets. This fact, plus a number of other aspects, including widely published
studies showing that, on average, high-fee funds do not perform as well as
less expensive funds,4 has resulted in what the data provider Morningstar
(Morningstar Manager Research 2017) calls “a sea-change in investor prefer-
ences” (p. 1). The Morningstar Manager Research report on annual US asset
flows shows that for the full year 2016, investors withdrew almost $264 bil-
lion from actively managed US equity funds. And the withdrawals occurred
despite a trend toward stocks versus bonds as the S&P 500 Index closed up
12% for the year. For the same period, passively managed US equity funds
saw net inflows of almost $237 billion. In their Wall Street Journal article,
Tergesen and Zweig (2016) cite Morningstar estimates that passive assets
under management since 2007 have tripled to $5.7 trillion, whereas assets in
active funds have increased by only 54%, to $23.2 trillion.
Still, according to Morningstar, 66% of US mutual fund and
exchange-traded fund assets are actively invested, albeit down from 84% just
10 years ago.
But these percentages could change soon. Representing less than 20%
of US equity holdings by retail (individual) investors only a decade ago,
passive funds’ share of US equities is expected to reach 50% in 2018–2019
(Morningstar Manager Research 2017).
While the trend to passive equity funds is particularly pronounced in the
United States, it is also present in Europe, where passive equity funds now
4
According to Morningstar, the average asset-weighted annual fee for actively managed US
stock funds is currently 0.78%, compared with 0.11% for the average passive stock fund.
© 2017 CFA Institute Research Foundation. All rights reserved. xiii

Equity Valuation
attract more net flows than active equity funds, and in the Asia-Pacific region,
where net flows to passive strategies now almost match flows to active
strategies.
This monograph addresses a number of questions these trends raise:
•• Public stocks are traded in competitive markets and subject to the law of
supply and demand: Is there really such a thing as an intrinsic (or fun-
damental) value of a stock? If yes, can we determine this value using the
tools we presently have? Or can we determine only relative values?
•• What about determining the value of hard-to-value assets, such as initial
public offerings or private equity? What is the role of “hype” (hyperbole)
or information asymmetry in determining value?
•• Assuming that fundamental active analysts/managers can estimate the
intrinsic value of a stock and spot mispricings by comparing the intrinsic
price to the market price, can they execute an advantageous trade that
delivers value to the investor? What tools and heuristics do such analysts
use, and how effective are they?
•• Do economic or other phenomena, such as quantitative easing by central
banks or corporate stock buybacks, distort market prices, taking them far
from a stock’s fundamental price?
•• What is the equilibrium between the cost and the benefit of doing funda-
mental analysis, where the benefit is alpha or extra return to the investor?
•• Do fundamental analysts/managers really play an important role in keep-
ing markets (quasi) efficient?
•• Will news sources, data sources, new tools, or new technology not yet
(widely) used allow fundamental managers to better estimate a stock’s
fundamental value?
•• Has the global investment universe changed so much that the role of the
fundamental analyst/manager is no longer central? In other words, are
there better ways to generate returns for investors than traditional value
investing?
Quotations in this book that are not attributed to a source listed in the
References are from academic and industry colleagues who provided us with
their views on equity valuation in a series of interviews during the first half
of 2017.
xiv © 2017 CFA Institute Research Foundation. All rights reserved.

1. Finance Theory and Equity Valuation
“Any valuation must be in accord with the well-established theory of finance.”

That is the first of three conditions Stephen Penman (2016, p. 3), professor of
accounting at Columbia Business School, identifies for obtaining more robust
equity valuations than we currently can obtain—something that must be of
interest to the industry. So, let’s start here and see whether commonly used
valuation practices are indeed in accord with finance theory.
The conceptual foundation of fundamental analysis is a supporting pillar
of the notion of the true or intrinsic price of stocks, and the question of equity
valuation is closely related to the question of market efficiency. Thus, a proper
analysis of valuation issues cannot be made without a minimum of theory and
macroeconomic considerations.
Sébastien Lleo, who, before joining NEOMA Business School (France)
as a finance professor, had nine years of experience in financial markets,
remarked,
The concept of fundamental (or intrinsic) value rationalizes and frames our
search for the “true” value of an asset. It is the value against which we assess
today’s price to determine whether an investment is overvalued, underval-
ued, or fairly valued. This makes it one of the most important constructs
in investment theory, and one of the most important ideas in investment
practice.
Let’s start by observing that in any attempt to evaluate the “true” value
of a firm’s stock, fundamental analysis seems to be an obvious and natural
approach: An investor considering an investment in equities will probably
start by looking at a number of firms in an attempt to estimate the stream
of future revenues that each firm might provide. In so doing, the inves-
tor will likely follow the advice of two Columbia Business School profes-
sors, Benjamin Graham and David Dodd, who wrote the textbook Security
Analysis: Principles and Techniques (1934). Initially intended as a common-
sense guide to investing, the book, first published in 1934, had by the time of
Graham’s death in 1988 become the reference text for almost all subsequent
work in fundamental analysis.
As Graham and Dodd advocate, the investor, after analyzing a firm as a
potential investment, will look at the market price to determine whether the
firm’s revenue streams are in line with its market price—that is, whether a
stock’s actual market price is in line with its “true” value.
© 2017 CFA Institute Research Foundation. All rights reserved. 1

Equity Valuation
Because of the difficulty in determining a stock’s true value, Graham rec-

ommends that investors invest with a margin of safety; that is, investors should
put money into stocks priced at a discount sufficiently large, when compared
with a reasonable estimate of what the stock is worth, to make the investment
attractive even if the estimate is a little too high. Not only single stocks but
whole sectors of the markets might be trading at a discount. For example, a
2016 study by BlackRock found that over the past decade, global value stocks,
including energy and materials, were trading at a 20% discount relative to the
broader market.
At the time of Graham and Dodd’s original work, fundamental analysis
was not yet highly formalized. It was based on educated economic common
sense. Investing would become increasingly more complicated with the intro-
duction of the notion of risk. With that notion added, the price at which a
stock is traded in the market reflects not only the company’s ability to gener-
ate cash but also the risk associated with future streams of cash flows. And
ultimately, the price of each asset will be set by the supply of and demand for
that asset.
As discussed in textbooks on the classical theory of asset pricing, finan-
cial asset prices are equal to the sum of the discounted values of expected
future cash flows. The discount rate is determined as the sum of the risk-free
rates and a risk premium; it is exogenous and cannot be determined by purely
financial considerations.
In classical finance, however, a parallel framework relates to the notion of
market efficiency and intrinsic prices. This framework has become important
for both theoretical and practical reasons. A first definition of market effi-
ciency comes from Eugene F. Fama (1970), who stated that unfettered finan-
cial markets are efficient. Campbell Harvey, a professor of finance at Duke
University’s Fuqua School of Business, commented,
This [efficiency], according to Fama, means that market prices correctly
reflect all relevant information. This statement of market inefficiency is
often misunderstood. In Fama’s most extreme scenario, efficiency [exists
when] prices correctly reflect all public and private information at any point
in time. In a second scenario, prices are assumed to reflect all publicly avail-
able information. This means that any trading strategy based on publicly
available information should not earn excess returns. However, neither of
these scenarios is particularly realistic. It is naive to think that prices cor-
rectly reflect all public or public and private information. Indeed, the state-
ments are not even testable given that we never observe “true” prices. So, a
reasonable starting point is the concept of relative efficiency. It is more likely
that prices of large-cap US equities correctly reflect information than that
2 © 2017 CFA Institute Research Foundation. All rights reserved.

they do so on the Zimbabwe stock exchange. There is likely some degree of

mispricing in the US, but the degree of mispricing is likely a lot larger in
Zimbabwe.
Harvey continued,
If we view the price of financial assets—in our case, equities—as equal to
the sum of expected values of discounted future cash flows, the entire ques-
tion of market efficiency and the existence of an intrinsic price hinges on the
ability to get as close as possible to the “true” discount rates and expected
cash flows. We will never get it exactly right—nor will we know how close
we are—however, if we are closer [at estimating] the intrinsic price than
others are, there is a possibility of making excess returns.
The job of the financial analyst, then, is to discover whether a financial
asset is mispriced (that is, under- or overpriced with respect to its intrinsic
value or a comparable asset) under the assumption that markets will eventually
correct the mispricing. The identification of and the ability to take advantage
of mispricings (or price anomalies) are believed to generate profit when asset
prices revert to their true price (or when correct relative prices are reinstated).
Although markets are driven by supply and demand, it is widely believed
that intrinsic value still plays a role in delivering better returns. Kenneth
Little, managing director of the investments group at the value investment
firm Brandes Investment Partners, remarked,
Fundamental (intrinsic) value can still be used as an effective tool to out-
perform the market, but only for investors that have a sufficiently long time
horizon. While it is difficult to define how long this time horizon needs
to be, we assume that in any period of less than five years, the stock price
returns may be driven at least as much by factors such as momentum, flow
of funds, and investor sentiment.
Matteo Bonaventura, a buy-side analyst at the Milan-based asset man-
ager Banor SIM, commented,
I think that intrinsic value does exist, but its estimation requires research,
and it therefore represents a (huge) cost. Thus, in order for this process to
turn profitable, market prices have to be far from intrinsic values, and this
is most likely when large information asymmetries exist. For example, take
the Italian stock exchange, Borsa Italiana, in the period 2012–2017. Some
less researched companies delivered an astonishing 500% total return in
this period, which means roughly a 40% annual return, compared to 10.4%
for the overall market for the same period. Analysts and portfolio manag-
ers who were able to better understand the fundamentals underlying these
companies undoubtedly benefited.

Equity Valuation
However, I think that intrinsic values are dynamic and can change with
major nonanticipated shocks in supply and demand. For example, the
recent tax advantage on long-term investments in Italian small caps drove
the stock prices up to very high valuation multiples. Thus, in my opinion,
both considerations have to be taken into account when thinking about
intrinsic value.
The €205 billion Dutch fund PGGM Investments is a long-term investor
and uses the notion of intrinsic value to construct fundamental portfolios. Jaap
van Dam, the firm’s principal director of investment strategy, commented,
Intrinsic value can be used in a very narrow sense, to measure book value.
But you can broaden this a bit in the sense that intrinsic value can be used,
for example, if you invest deeply in understanding how firms create long-
term value. This requires an understanding of the firm, its markets, strate-
gies, investor stewardship; in the long term, there are a lot of agency issues.
The market price is a question of supply and demand, but behind the market
price, one can reasonably estimate a ballpark number that represents the
firm’s value. We use intrinsic price not to “beat the market”—a term I don’t
like—but to generate better returns.
Christian Kjaer, head of global equities and volatility at the $113 billion
Danish state pension fund ATP, remarked,
In the space of global equities, we consider the uncertainties around various
valuation models to be huge. Consequently, we need to observe rather extreme
mispricing relative to the fundamental model value in order to have sufficient
conviction in the perceived mispricing. In areas where we consider our fun-
damental understanding to be stronger—for example, Danish equities—we
have significantly more confidence in perceived fundamental mispricings and
use it [this information] more actively to outperform markets.
The role of theory is to provide rules that allow a financial analyst to
make forecasts. Consider sending a satellite into orbit. Doing so would be
impossible without the theory of gravity. Theory allows a physicist to use
data to determine the satellite’s future path. In finance, however, we are in
an intermediate situation; that is, to determine the future path of an asset’s
price, we need knowledge of some basic facts, plus the ability to test whether
those facts are true. Unlike in physics, in finance (and economics), we do not
have the ability to test our theories. In equity valuation, we can avoid abstract
mathematical and statistical principles, but we need an understanding of the
theoretical underpinnings of valuation.
In free markets, the price of “things” is determined by the interaction
between supply and demand—with links, possibly complex ones, between the
characteristics of “things” and their prices. Largely dependent on objective

characteristics, competition assigns a market price to goods and services in

common use. The price of a car, for example, depends on the vehicle’s size,
power, speed, and other features. But this dependence of price on character-
istics is not linear and can become highly nonlinear for vehicles that reach
the limits of performance and/or benefit from strong marketing/advertising
campaigns.
This nonlinearity in pricing is typical of things whose price is subject to
nonobjective characteristics, such as fads or branding, where social or other
motivations, such as “herding to have,” contribute to determining the price.
In economics, these goods are often referred to as “Veblen goods,” from the
Norwegian-American economist Thorstein Veblen, who analyzed the moti-
vation for consumption in the US economy at the end of the nineteenth cen-
tury. Veblen introduced the notion of “conspicuous consumption” to identify
the type of consumption that is not related to the direct fruition of the goods
consumed but to the goods’ symbolic value. But even in this case, the law
of supply and demand holds, as it holds in art markets, where the intrinsic
value of a work of art is ill defined and its market price is the amount of
money someone is willing to pay at any given time. Alternative proposals for
pricing, such as determining the value of things with respect to the skill or
time needed to produce them, as proposed by Karl Marx, among others, run
contrary to the spirit of free markets.
Free competitive markets for goods and services should be reasonably free
from arbitrage opportunities. So the “law of one price” should hold, approxi-
mately. Exceptions are frequent, however. For example, with the introduction
of revenue maximization strategies, the price of an airline seat or hotel room
changes rapidly as a function of supply and demand, and different agents may
offer the same seat or hotel room at a different price. Violations of the law of
one price do not disrupt the market for goods and services because physical
impediments prevent fully exploiting arbitrage. To summarize, in free com-
petitive markets, goods and services do not have an intrinsic price but respect
(approximately) the law of one price.
What about financial markets? Financial assets trade in markets that are
almost perfectly competitive and free. Neoclassical finance theory assumes
that the prices of financial assets are free from true arbitrage opportunities,
though financial markets might exhibit “near arbitrage”—that is, the possi-
bility of small gains with limited risk. And these gains (and the risk) can
become large if pursued with the use of leverage. A number of papers, includ-
ing one by Shleifer and Vishny (1997), have found that financial markets
present near-arbitrage opportunities but that exploiting these opportunities
can be difficult because of the decision-making process of economic agents.

Equity Valuation
In the absence of arbitrage, the price of a financial asset is the sum of

the expected cash flows discounted with market discount rates, where the
discount rate includes a risk premium. This principle applies to all competi-
tive financial markets devoid of arbitrage opportunities. In itself, the principle
characterizes neither intrinsic price nor market efficiency.
The notion of intrinsic price requires the introduction of macroeconomic
considerations. Intrinsic price is a concept related to equilibrium in a spe-
cific macroeconomic theory. Both the notion of intrinsic price and that of
deviations from the intrinsic price can be defined only within a supply-and-
demand framework. Simply stated, the intrinsic price is the price financial
assets would have in economies where the supply and demand for investments
are in equilibrium.
As Cochrane (2001) states in the preface to his book Asset Pricing,
Asset pricing theory shares the positive versus normative tension present
in the rest of economics. Does it describe the way the world does work, or
the way the world should work? We observe the prices or returns of many
assets. We can use the theory positively, to try to understand why prices or
returns are what they are. If the world does not obey a model’s predictions,
we can decide that the model needs improvement. However, we can also
decide that the world is wrong, that some assets are “mispriced” and present
trading opportunities for the shrewd investor. (p. xiii)
Any financial asset is, by definition, a package of cash flows in the form
of deferred payments. The owner of a financial asset owns the right to receive
all cash flows associated with the contract. A bond is a contract that gives the
owner the right to receive coupons plus the final repayment of the principal;
a firm’s stock gives the owner the right to receive all future dividends plus
the proceeds from the eventual final liquidation of the firm; an option gives
the owner the right to receive a final payoff conditional on given events (in the
case of stock call options, the payoff is equal to the difference between the
market price of the underlying stock and the strike of the option); and so on
for other financial contracts.
The price of financial assets is constrained by arguments about the
absence of arbitrage. Consider the simplest case of a US government note
with a nominal (face) value of $1,000 that matures in three years without
coupons (i.e., a zero-coupon note). The theoretical price of the note is the
present value of $1,000 discounted at the risk-free rate. The market price can-
not be different from the theoretical price because this difference would cre-
ate arbitrage opportunities. The price of any other government bond or note
is also uniquely determined by the risk-free rates applicable to coupons (if any
exist) and the principal. The reason is that US government bonds or notes

are considered risk free; any other price would result in arbitrage opportuni-
ties. For example, in discrete time, assuming n periods, if the risk-free rate in
period i is ri, then the present value, PV, of a government bond with coupon C
and principal M is
C C C+M
PV = + + + . (1)
(1 + r1 ) (1 + r1 )(1 + r2 ) (1 + r1 )(1 + r2 ) (1 + rn )
Present value clearly depends on interest rates.

The supply of and demand for government bills (notes and bonds), as well
as other financial assets, affects interest rates.5 The supply of and demand for
credit, as well as decisions made by central banks regarding the base rates,
determine all risk-free rates. Interest rates can fluctuate greatly. For example,
interest rates for US T-bills were below 1% in the immediate post–World
War II period, soared to above 14% in the early 1980s, dropped to around
1% after the dot-com bubble, and then rose again, only to fall to below 1%—
in fact, almost exactly zero—following the 2008 financial crisis. Figure 1.1
represents three-month T-bill rates from 1934 to 2016. Supply and demand
determine the term structure of interest rates, which in turn determine the
price of every bill, note, and bond simultaneously.
Corporate bonds, unlike US government bonds, are subject to credit risk:
The payment of coupons and repayment of the principal are uncertain because
of the possibility of default and the percentage that can be recovered in case
of default. So, in the case of corporate bonds, a bond’s price is the sum of the
expected values of coupons and of the principal, both discounted at a rate con-
sisting of the risk-free rate plus a risk premium. Supply and demand contrib-
ute to the level of risk-free rates and determine—given the risk characteristics
of each bond—the risk premium for that bond.
With stocks, the level of uncertainty is much higher than with bonds but
the theoretical pricing framework is the same: Stock prices are the sum of
the discounted values of expected cash flows. However, in the case of stocks,
cash flows are not mandated by contract but are truly uncertain even if, in
mature industries, the flow of dividends might be somewhat predictable and
5
In modern monetary systems, the vast majority of money is created when agents take out
loans. A study by Moore (1988) found that banks are not constrained by reserves in their
lending. Contrary to the theory of the multiplier, he proposed a theory called Horizontalism
by which money is created endogenously by the banking system in granting loans and simul-
taneously crediting the client’s account. For a complete analysis of the modern process of
money generation, see McLeay, Radia, and Thomas (2014a, 2014b).

Equity Valuation
Figure 1.1. Fluctuations of Three-Month T-Bill Rates, 1934–2016
Percent
18
16
14
12
10
8
6
4
2
0
1934 48 62 76 90 2004
Source: Constructed by the authors using data obtained from Federal Reserve Economic Data
Three-Month Treasury Bill: Secondary Market Rate.
the market valuation relatively stable. In addition, stocks are subject to risks
other than solely credit risk.
The price of a firm’s stock is (essentially) determined by expectations such
as the ability of the firm to produce a steady flow of dividends and/or to com-
mand a high price in the future. In the absence of arbitrage opportunities,
the market price of any financial asset is the sum of the expected discounted
value of future cash flows. The discount rate is the risk-free rate plus a risk
premium.
Discount rates for stocks are determined by the market—that is, by the
interplay of supply and demand—and by the action of central banks and gov-
ernments. Central banks determine the rate they pay on reserves—a basic
interest rate that affects all other interest rates. Other considerations, includ-
ing fiscal policies and risk regulations, affect all rates in the economy. The
notion that assets’ market prices are equal to the sum of expected discounted
cash flows is general; it does not characterize the intrinsic price. In fact, prices
can always be represented as the sum of expected discounted cash flows with
appropriate discount rates.
To do the work that has traditionally been considered the job of an equity
analyst and for which investment professionals are still widely considered to
be paid (i.e., to “beat the market”), the profession tries to identify under- and
overpriced assets and take advantage of that knowledge to realize a profit.

Doing so requires determining the intrinsic value of a stock, which in turn

calls for determining the distribution of future cash flows, the true discount
rates, and the risk premiums.
Much of the vast literature on stock valuation does not address this prob-
lem theoretically but devotes a lot of time to describing how to analyze a
balance sheet and, eventually, how to forecast earnings or dividends. As for
describing how to choose the discount rate, however, most authors simply
recommend heuristics or historical values. The capital asset pricing model and
factor models are also used.
That the concept of intrinsic value is tricky is not surprising. Even while
considering intrinsic value central to investment management, the father of
value investing, Benjamin Graham, found the concept “elusive” and cau-
tioned against taking it as “definite and determinable” (Graham and Dodd,
1934; 2008 edition, p. 64). Marc Reinganum, a former senior managing
director at State Street Global Advisors, commented on the elusive nature of
intrinsic value:
One should never confuse the concept of arbitrage (riskless profit) with
intrinsic value. Intrinsic value always depends upon assumptions, and unless
all investors always agree with these assumptions, there will never be one
true value in practice. The best any one investor can hope for is that he or
she understands the dynamics of stock prices a little bit better than most
market participants.
In fact, the concept of the intrinsic value of financial assets depends
on the possibility of defining a natural rate of return or, more correctly, a
“natural rate of interest” plus a “natural risk premium.” However, defining a
natural rate of interest or a natural rate of return without recourse to mac-
roeconomic considerations is well beyond the realm of pure finance theory.
Macroeconomic considerations might shed new light on the question of
market efficiency and the intrinsic value of stocks. The idea of a natural rate
of interest was introduced by the Swedish economist Knut Wicksell (1898),
who defined the natural rate of interest as that rate of interest that guarantees
stable asset and consumer prices. We can think of a natural rate of interest as
the equilibrium rate when market capitalization grows at the same rate as that
of the economy.
The idea of a natural rate of interest returned to the fore of economic
research after central banks accepted the notion of endogenous money gen-
eration and began to target interest rates, instead of the quantity of money,
in the 1990s (see Bindseil and König 2013). Following the 2008 financial
crisis, central banks lowered interest rates to what is sometimes called “the
effective lower bound”—that is, zero—and began “unconventional” policies,

Equity Valuation
such as quantitative easing (see McLeay, Radia, and Thomas 2014b). But, as
years of prolonged near-zero interest rates by central banks failed to produce
the desired economic growth, central bankers began to question just what the
ideal interest rate should be.
If we could identify a rate of interest at which no excess demand for invest-
ments occurs, we might reasonably assume that we could identify the intrinsic
value of assets as the price obtained by discounting cash flows with the natural
rate of interest. Of course the problem of forecasting cash flows remains.
Central banks have developed models to compute the natural rate of
interest. Lubik and Matthes (2015), from the US Federal Reserve Bank of
Richmond, compared two approaches: (1) their approach, which uses a time-
varying parameter vector autoregressive model, and (2) the Laubach–Williams
model (see Laubach and Williams 2015), which uses a state–space approach.
The idea of a natural rate of interest is not without critics, who observe
that no single rate of interest is able to guarantee stable prices and equilibrium
between investments and savings.
The question of the natural rate of interest bears on the question of the
relationship between stock returns and economic growth. Studying the
period 1872–2014, Straehl and Ibbotson (2015) found that the increase of
total payout (dividends plus share buybacks) follows the increase in per capita
GDP over very long periods. Others, including Ritter (2005), have found no
relationship (not even a negative relationship) between stock returns and eco-
nomic growth. This question will be discussed in Chapter 4.
Following the notion that the intrinsic value of a financial asset is its price
under some equilibrium condition led to a revision of the efficient market
hypothesis (EMH). LeRoy (1976) was the first to point out that the original
formulation of the EMH was a tautology.6 In response to LeRoy, Fama (1976)
introduced the idea that the true price is the price in economic equilibrium.
More recently, Pilkington (2014) posited the idea that the EMH is really a
hypothesis on an equilibrium situation of economies; that is, the EMH states
that actual prices are equal to the prices of an economy where savings and
investments are in equilibrium.
6
In his 1970 paper, Fama (p. 384) suggested that markets are efficient if prices satisfy the fol-
lowing equation:
E ( p j ,t +1 | Φt ) = E  1 + (rj ,t +1 | Φt ) p j ,t ,
 

where p j ,t +1 and rj ,t +1 are prices and returns at time t + 1 anticipated at time t and Φt is the
information set at time t. LeRoy (1976) observed that this equation is a tautology because the
expectation of prices at time t + 1 with the information set at time t is by definition equal to
prices at time t multiplied by the expectation of returns at time t + 1.

The conceptual underpinning of intrinsic value as outlined here shows

that the notion of intrinsic value is neither easy to understand nor easy to
apply. The intrinsic value of a financial asset would be the price of that finan-
cial asset in a different economy. In fact, there are a number of questions:
•• Is fundamental (or intrinsic) value more than an intellectual construct?
•• Is intrinsic value based on empirical data?
•• Can intrinsic value be determined with the tools financial analysts pres-
ently have?
•• If intrinsic value is no more than an intellectual concept or if the quest to
determine intrinsic value proves elusive, why not limit our quest to deter-
mining a stock’s value as its market (or fair) value with the use of heuris-
tics such as market multiples or rules such as what has become known as
the Buffett Rule, which states that markets are over-/undervalued as a
function of the ratio between market capitalization and GDP?7
•• Or why not simply turn to alternative methods—for example, pattern or
trend analysis—which might improve one’s chances of realizing gains?
That the “true” price of an investment can be realized only under a hypo-
thetical situation of equilibrium is not obvious. Equity analysts might find
it difficult to compute the interest rates that apply to these elusive states of
equilibrium of savings and investments. And how can we be sure that the
economy will revert to a situation of equilibrium? How long will doing
so take?
In commenting on the concept of intrinsic value, Alfred Slager, a pro-
fessor of pension fund management at the TIAS School for Business and
Society at Tilburg University and a trustee at the Dutch pension fund for
general practitioners SPH, said,
I view the concept of intrinsic value mainly as an intellectual construct.
It’s very useful for understanding the building blocks of a stock market but
rather difficult to build successful investment strategies on. Determining
the value of the underlying company is tricky enough, but with the advances
in real options theory, etc., it seems rather robust.
On the other hand, the linkage between the intrinsic value and valuation
of a stock has not really made any progress in my view, and probably never
7
The accuracy of this rule is subject to changes in the amount of equity risk in a given econ-
omy that is publicly traded. For example, in the 1980s, Germany appeared to be undervalued,
but that appearance was because the equities were held privately by families, while corporate
debt was held by banks.

Equity Valuation
will. The notion that there might be a fundamental or intrinsic value in a

stock seems to me a remainder of the idea that we could/might apply gen-
eral equilibrium models to the stock market. What we know is that there is
supply and demand, which determines a stock’s price. The residual value (or
multiple), the gap between the stock price and intrinsic value, is then more
or less determined at random.
I readily admit that’s a dismal view, and one which we do not accept as a
sector. We construct models to explain this gap, and because we use similar
models and base our buy-and-sell advice on these models, on a large scale
we might find that there exists a correlation between the valuation models
and the multiple, but that is induced by market behavior and has little to do
with fundamental analysis.
Things simplify considerably if we drop the notion of intrinsic price and
try to establish only relative pricing valuations. In other words, we might sim-
ply ask whether the ordering of market prices corresponds to the ordering
of intrinsic prices. This question is much simpler and can be answered with-
out a precise knowledge of natural rates of interest or natural risk premiums.
Relative market evaluations depend, at least in first approximation, on esti-
mates of the distribution of future cash flows. In practice, these valuations can
be done with fundamental analysis, without forecasting. The crucial question
is, Does the process of discovering the intrinsic price allow better returns?
Somehow, the market price has to be tied to the intrinsic price to understand
whether situations of nonequilibrium, such as bubbles, are developing.

2. Tools for Valuing Stocks: Intrinsic Value
and Relative Value
In the previous chapter, we explored theoretical questions related to the valu-

ation of a firm’s shares and, specifically, the notion of intrinsic value in a free
market based on supply and demand. In particular, we discussed how deter-
mining the intrinsic value of a stock implies determining the interest rates
and risk premiums in a situation of market equilibrium. In practice, however,
the notion of intrinsic price is used for relative comparisons and for evaluat-
ing situations of strong disequilibrium. In other words, evaluating whether a
stock is cheap or expensive with respect to other stocks is possible, but evalu-
ating whether a stock is cheap or expensive in absolute is not, except in situa-
tions far from conditions of equilibrium.
Let’s now take a look at the tools investment professionals use to estimate
a stock’s value. We will start with tools to discover the intrinsic value of a
firm’s stock. The CFA Institute Survey of the equity valuation practices of
its members, “Equity Valuation: A Survey of Professional Practice” (Pinto,
Robinson, and Stowe 2015), had the objective of documenting professional
practice in the selection of equity valuation approaches. The survey also asked
whether a specific tool was viewed by the analyst as widely or narrowly appli-
cable. A mean frequency of greater than 50% would suggest that the tool
was viewed as a general tool, and a mean frequency of less than 50% would
indicate that use of the tool was limited to special cases.8
The survey found (see Table 2.1) that among the absolute valuation mod-
els used to determine the intrinsic value of a firm, nearly 80% of the sur-
vey participants reported that they use a discounted present value approach.
Slightly more than 60% reported using an asset-based approach, but a lower
mean frequency (just under 37%) indicates that its use is more restricted.
Note that a market multiples approach, which we will discuss later, is the
most widely used valuation method among participants.
Jarrod Wilcox, president of Wilcox Investment, commented on the use of
various models in equity valuation, saying,
A particular valuation model may be useful for one purpose and not for
another. For example, the cyclically adjusted price–earnings ratio (CAPE)
8
A total of 1,980 practitioners in the Americas (66% of the total), Asia Pacific (12% of the
total), and Europe, the Middle East, and Africa (22% of the total) participated in the CFA
Institute Survey (Pinto et al. 2015).

Equity Valuation
Table 2.1. Most Widely Used Valuation Approaches among Respondents to the 2015
CFA Institute Study
Valuation Approaches: Global ranking.
In evaluating individual equity securities, Percentage of Cases in
which of the following approaches to Which the Respondent Uses
valuation do you use? Percentage of Each of the Approachesa
N = 1,980 Respondents (mean)
A market multiples approach 92.8 68.6
A discounted present value approach 78.8 59.5
An asset-based approach 61.4 36.8
A (real) options approach 5.0 20.7
Other approach 12.7 58.1
a
Respondents using an approach were asked for the percentage of valuation cases in which the
approach is used. Thus, this column reports conditional frequencies.
Source: CFA Institute.
model is good for long-term forecasts of aggregate US stock market returns

but not so useful for stock return forecasts in the shorter term, for cross-
sectional explanations of price, or for corporate CEOs interested in the
easiest way to enhance the value of their company. Another example is the
dividend discount model, which helps us organize our thoughts but is weak
in forecasting future returns. How do we measure future dividends today?
Net Present Value Models

Models used to compute net present value (NPV) require two key steps:
(1) the forecasting of future cash flows and (2) the estimation of discount
factors—that is, the risk-free rate plus a risk premium. Forecasts of future
cash flows are based on the fundamental analysis of a firm plus models of
future projections of cash flows and their uncertainty. As discussed in
Chapter 1, the intrinsic discount factor cannot really be determined with-
out macroeconomic considerations, including knowledge of the financial and
banking system.
In practice, however, analysts use various techniques for making a rea-
sonable estimate of a required rate of return (equivalent to the discount rate
if the market is in equilibrium). The required rate of return is the benchmark
return rate used by investors in their decision-making process. If, on the one
hand, the expected return of a stock is higher than the required rate, then the
stock is considered underpriced and is a candidate for investment; if, on the
other hand, the expected return is lower than the required return, the stock
is considered overpriced and therefore not a good candidate for investment.

2. Tools for Valuing Stocks
Discounted present value approaches use company fundamentals to try

to determine the intrinsic value of a firm, in which the value is the sum of
discounted expected future cash flows. Two basic versions of this model are
used: the dividend discount model (DDM) and the discounted cash flow
(DCF) model. In the DDM, dividends are considered the relevant cash
flows. But dividends are discretionary, which makes their forecasting prob-
lematic. In fact, one has to forecast not only how the company will perform
but also the decisions that will be made about the distribution of dividends
(versus reinvestment of profits into the company’s operations). For this reason,
many prefer DCF models, which use a different concept of the company’s
discounted future cash flows.
The idea behind a DCF model is that what is important is the cash avail-
able, regardless of whether or not it is distributed. An advantage of DCF is
its applicability to listed and unlisted companies alike. The limits of these
models are well known: A DDM can be used only in cases where a firm pays
stable (and predictable) dividends; the DCF, only in cases where a firm has
positive (and predictable) free cash flows. (“Free” cash flow equals cash flow
minus an allowance for enough reinvestment of cash to keep the company in a
steady state, by replacing equipment and so forth.)
Among the investment academics and professionals with whom we
spoke, most prefer (free) cash flow models. The insight here: The price of a
stock reflects the expectations of future cash flows. Christian Kjaer, head
of global equities and volatility at Denmark’s largest pension provider ATP
($113 billion in assets under management), commented, “All models shed
light on some aspect of ‘the truth’; they all have their pros and cons. However,
on the margin, we favor the free-cash-flow model, basically because we find
free cash flows less prone to manipulation.”
Kenneth Little, managing director of the investments group at Brandes
Investment Partners, made a similar evaluation:
We believe that most or all financial models used to determine the funda-
mental value of a firm’s stock are essentially shortcuts to a full discounted
free cash flow valuation of a firm. While all valuation models have their
respective shortfalls and are very sensitive to underlying assumptions, to the
extent they are long-term in nature, focused on free cash flow, and appro-
priately risk adjusted, these models represent a reasonable methodology for
approximating a firm’s fundamental value.
Bradford Cornell, professor of financial economics at the California
Institute of Technology, also favors the cash flow model: “Ultimately, it is
the expected future cash flow that determines value. The stock market is the
mechanism by which those expectations get reflected in price.”

Equity Valuation
Fundamental equity portfolios at PGGM Investments, the Dutch pen-

sion fund with €189 billion in assets under management, are constructed on
the basis of fundamental models—in particular, DCF models. Felix Lanters,
head of equities there, said,
We consider discounted cash flow models to be the better of models—if
there is such a thing—for gaining a fair assessment of the long-term ability
of the firm to generate profits. Our objective is not to obtain a single price
but clarity on how the valuation resets in changing assumptions. As with
any model, you need to make a lot of assumptions, but the free cash flow
model allows you to change assumptions, create scenarios, and find sen-
sitivity to fluctuations. For example, if relatively small changes in a single
factor have a big impact on valuation, this raises red flags. We use the free
cash model very broadly, with more reflection of the underlying processes as
opposed to just looking at what comes out of the model.
Matteo Bonaventura, a buy-side financial analyst at Banor SIM in Milan,
also finds the scenario-like properties of the DCF model a plus. He remarked,
I think that the most interesting aspect of DCF models is sensitivity
analysis, as it provides some sort of confidence interval for your estimates.
Moreover, it can provide some interesting insights. For example, the change
in estimated intrinsic value to changes in the cost of capital can be inter-
preted as a raw proxy of the equity duration—that is, the sensitivity of
equity prices to changes in interest rates.
Although he uses both DDM and DCF models, Bonaventura prefers the
DCF model for most companies.
For estimating the intrinsic value of an investment, Sébastien Lleo,
who has nine years of experience in financial markets and is now a finance
professor at NEOMA Business School (France), also prefers discounted
free-cash-flow (DFCF) models—namely DFCF to the firm and to equity—
and adjusted present value methodologies, methods he links to John Burr
Williams’s (1938) pioneering work on investment management. However, he
has some reservations:
In theory, this class of models provides the most accurate valuation. In
practice, however, their valuation is sensitive to the choice of inputs and
their underlying assumptions (equity risk premium, annual sales growth,
long-term growth, exit multiple, …). When the choice of input is often sub-
jective, the model output becomes subjective as well. To summarize this sit-
uation, we could rephrase the expression “garbage in, garbage out,” familiar
to optimization professionals as “subjective in, subjective out.” This means
that free-cash-flow methodologies are only truly meaningful when they are
complemented by sensitivity analysis and scenario analysis. The true virtue

of these methods is, therefore, to provide a range of possible valuations

under a variety of scenarios, not a point estimate.
Bonaventura added another critique—not of the DDM or DCF models
themselves but of their application:
Many times, there is the risk that DCF becomes a back-of-the-envelope
exercise—that is, you start from an idea of target price and you adjust num-
bers to make it happen. Moreover, many times, DCF is applied incorrectly
from a theoretical point of view—for example, when applying constant cost
of capital rates or in the application of the perpetual growth rates. Empirical
evidence is reported by Professor Fernandez.9
Slager offered a somewhat different view, saying,
I find that when I teach these models, they resonate far better in the corpo-
rate finance–related courses than in [investment] finance. With corporate
finance, they offer students insights into the drivers of the value of a corpo-
ration. For investment students, they are insights in the valuation of a cor-
poration. Personally, I find that the DDM models are more relevant for the
stock market and free-cash-flow models more related to corporate finance.
Ananth Madhavan, the global head of research, exchange-traded funds,
and index investments at BlackRock and a lecturer on financial engineer-
ing at the Haas School of Business at the University of California, Berkeley,
noted, “Many practitioners combine other approaches with DCF by using
comparable firm multiples to find a more robust estimate of terminal value
that takes into account maturity.”
The challenge in using present value models is the dependence of future cash
flows on, and the predictability of, input forecasts, such as a company’s return on
invested capital, its growth rate, and its weighted average cost of capital. The key
difficulty is in forecasting future dividends, cash flows, and the discount rate.
In theory, both the DDM and DCF models include an infinite stream
of cash flows and/or the eventual final liquidation of the firm. In practice,
however, all models make forecasts of dividends or cash flows over a finite
time horizon, typically 5 to 10 years. Therefore, every present value model, in
practice, has two components: the present value of cash flows before the time
horizon and the terminal value at the time horizon.
The terminal value at the time horizon is a particularly critical issue.
Penman (2016) writes,
Finite-horizon forecasts of dividends typically do not capture value. … In
short, dividends are zero-NPV. Thus, forecasting dividends is of no help
9
This reference is to Fernandez (2015).

Equity Valuation
in establishing value (the case of a firm that “pays no dividends” being an

extreme example). (p. 6)
The relative worth of the terminal value with respect to the total value is,
therefore, an important issue. Charles Lee (2005), professor of accounting at
the Graduate School of Business at Stanford University and cofounder of San
Francisco–based Nipun Capital, observes that the terminal value depends on
the maturity of the firm and the sector. He notes that in a mature sector (such
as the tobacco industry), over an eight-year time horizon, the terminal value
represents 56% of the total; in the skin care sector, the terminal value repre-
sents the totality of the present value; and in the high-tech sector, the first
eight years of a firm yield negative cash flows, and the terminal value might
represent 125% of the total present value.
Lee remarks that there is in fact a progressive shift from models such
as the DDM or DCF model to the residual income model (RIM), a model
based more on value creation than on cash distribution. He observes that over
the entire life of a firm, wealth creation must equal the sum of dividends paid.
Over short time horizons, however, wealth creation differs from dividends
paid. The residual income is the income generated in a given period minus the
cost of capital needed to generate that income. The RIM values a company at
a given time by computing the present value of future residual income plus
the capital of the firm.
In addition to evaluating cash flows, the other crucial element in present
value models is the discount rate. As observed in Chapter 1, determining the
intrinsic value of a stock is somehow equivalent to identifying a natural rate of
interest and a natural rate of return. Both are related to a situation of macro-
economic equilibrium where prices are stable, there is full employment, and
the money available for investment equals the money needed for investment.
The problem of determining the intrinsic value of stocks is, therefore, the
problem of determining an equilibrium economic situation. Central banks in the
biggest economies have become increasingly interested in the problem of deter-
mining the natural interest rate because, in the aftermath of the 2008 financial
crisis, the policies of central banks have been focused on setting interest rates.
Among the more popular models for determining the natural interest rate
is the Laubach–Williams model. Laubach and Williams (2015) state,
The natural rate is assumed to depend on the estimated contemporane-
ous trend growth rate of potential output and a time-varying unobserved
component that captures the effects of other unspecified influences on the
natural rate. In mathematical terms, the natural rate of interest, denoted rt*
is given by: rt* = c + g t* + zt , where g t* is the estimated trend growth rate of
potential GDP, zt is an unobserved component that is assumed to follow a

random walk process, and c is an estimated coefficient that measures the

influence of the trend growth rate on the natural rate of interest.
The model is estimated using the Kalman filter.10
This and similar models are still largely ignored in the finance literature
on the DCF and DDM models, where no mention is made of how to measure
the discount factor. Talking about models makes little sense if the critical
component of the model—the discount factor—cannot be identified. In fact,
most of the investment and finance literature simply assumes that rates read
off the US Treasury curve are the correct riskless rates for discounting.
Penman (2016) discusses this problem, writing,
Modern finance has struggled with the question of the discount rate—the
cost of capital—producing the Capital Asset Pricing Model (CAPM) and
subsequent multi-factor models. Again based on no-arbitrage, the structure
of a generalized “asset pricing model” is understood, in the form of com-
mon return factors (the risk of which cannot be diversified away) and sensi-
tivity to those factors. However, operational identification of those factors
has proved elusive, let alone sensitivities to the unidentified factors. (p. 6)
In addition to the problem of forecasting related to the use of any present
value model, there are other problems. Consider, for example, the DDM and
corporate share buybacks. For the full year 2016, Standard & Poor’s (S&P
2017) announced that firms in the S&P 500 spent $536.4 billion on buybacks,
whereas for the period 2009–2016, S&P 500 firms repurchased $2.75 trillion
in stock. Another figure comes from Birinyi Associates, which estimates that
US-listed companies spent about $6.1 trillion buying back their own shares
during the 11-year period 2005–2016.11
Buybacks have two effects. First, by increasing demand for a stock, buy-
backs increase the stock price, thereby increasing the realized risk premium.
Second, buybacks reduce the amount of cash available to pay dividends; many
researchers regard them as a substitute for dividends, an alternative way
of returning cash to the shareholder. Cornell at the California Institute of
Technology commented that, given corporate buybacks and other innova-
tions, dividend models are less useful than they once were. Philip Straehl, a
senior research consultant and portfolio manager at Morningstar Investment
Management, and Roger Ibbotson, the chairman and chief investment offi-
cer (CIO) of Zebra Capital Management and a Yale School of Management
professor, agree. They argue (Straehl and Ibbotson 2015) that the shift in
The Kalman filter is a technique for estimating hidden variables in linear systems.
10
We thank Birinyi’s Chris Costelleo for providing us with this information in an Excel
11
spreadsheet.

Equity Valuation
corporate payout policy from dividends to buybacks has caused a “secular

decrease in dividend yields, and an analogous increase in per-share growth”
(p. 25). This, they add, has led to “a structural break in the return components
of the traditional supply models such as the dividend discount model” (p. 25),
creating the need for a new supply model of stock returns.
Straehl and Ibbotson (2015) propose such a model, which they call the
“total payout” model of stock returns. Their total-payout model includes both
dividends and buybacks: It is based on computing separately the extra return
that investors enjoy as a result of buybacks. The authors regard the sum of
dividends and buybacks as the reward to investors for “participating in the
real economy” through share ownership.
Hence, we have another important question. Straehl and Ibbotson (2015)
define the “total payout of stocks” as the sum of dividends plus the cash
payout resulting from buybacks. They found that over the 142-year period
1872–2014, the total payout per share and the per capita GDP of the United
States grew at approximately the same rate, albeit with large fluctuations (we
will discuss this phenomenon more in Chapter 4). They argue that their total-
payout model allows for good forecasts of long-term stock returns.
BlackRock’s Madhavan, in contrast, believes that the effect of buybacks
on the risk premium is ambiguous:
Buybacks offer an alternative (and perhaps more tax-efficient) way of distrib-
uting cash to shareholders than paying dividends. In the frictionless, sym-
metric information, and tax-free world of Miller and Modigliani, the value
of the firm is the same in both cases, although the stock price with a buyback
is higher than when a dividend is paid. Any changes in expected returns
then reflect real-world frictions such as signaling effects. Consequently,
practitioners often condition on the type of buyback—for example, whether
it is a share repurchase program, tender offer, etc.—and the context for the
buyback—for example, related to earnings or corporate events.
Other present value methods include an asset-based approach (used by
61% of the CFA Institute Survey participants, with a mean frequency of 37%)
and a (real) options approach (used by 5% of the CFA Institute Survey par-
ticipants, with a mean frequency of 21%). These approaches are clearly more
narrowly applicable than the DDM and DCF models.
Asset-based valuation is used in various contexts—in particular, in cases
where valuation does not depend on income. One example is liquidation. In
this case, assets are valued at the market price at which they could be sold.
Still, certain intangibles, such as a brand name, can be sold on the basis of
their potential to generate future income.

Asset-based valuation is also used in other cases, such as when balance

sheets need to be rectified so that each item reflects its current value or when
a value must be assigned to various parts of a business in preparation for the
purchase or sale of some part of the business. In this last case, the valuation
of each asset item follows the same principle as the valuation of the entire
business. The intrinsic value method can be used by computing the present
value of each item based on its ability to generate cash; alternatively, relative
valuation methods can be used to value an asset by comparison with similar
assets.
Relative Valuation and Market Multiples

As discussed previously, the NPV approach suffers from some significant
conceptual difficulties—not least in defining the asset’s intrinsic price. When
the use of a discounted present value approach is deemed difficult or inappro-
priate, relative valuation methods based on heuristics—in particular, market
multiples—are the tools of choice. As the name implies, market multiples are
tools to determine the price of an asset relative to the price of a similar (com-
parable) asset. So, they establish a ranking of asset values.
Referring again to the CFA Institute Survey, we note that a market mul-
tiples approach is the mostly widely used of all valuation techniques: Almost
93% of the survey participants reported that they use multiplier models,
which are considered general valuation tools (see Table 2.1). The most widely
used market multiples (see Table 2.2) are the P/E multiple (used by 88%) and
enterprise value (EV)—in particular, EV/EBITDA (used by 77%), followed
by price-to-book (59%) and price-to-(some measure of) cash flow (57%).
Commenting on market multiples, Banor SIM’s Bonaventura said,
I find that relative valuation models are very useful in the everyday activ-
ity of stock evaluation. However, the analysis needs to be integrated with
more understanding of the multiple and of why companies are trading at
a premium or a discount. Given that two companies are similar, multiples
can be a first tool to screen for economic attractiveness. While I don’t think
that multiples are able to anticipate stock returns, in my opinion they can be
a useful tool when considered in aggregate for sectors or the whole market.
For example, if reverse engineered, they can provide quick and useful infor-
mation on the current level of the risk premium.
At Brandes Investment Partners, Little remarked,
We believe relative valuation models provide a reasonable method to deter-
mine how firms are valued relative to one another, but they do little to help
forecast future equity returns on an absolute basis. There are numerous

Equity Valuation
Table 2.2. Most Widely Used Market Multiples Approach among Respondents

to the 2015 CFA Institute Study
Market Multiples Approach: When you Percentage of Cases in
use a market multiples approach, which Which the Respondent Uses
of the following ratios do you use? Percentage of Each of the Approaches
N = 1,765 Respondents (mean)
D/P (dividend yield) or P/D 35.5 44.3

(price-to-dividend)
Enterprise value (EV) or firm value 76.7 61.1
multiples (e.g., EV-to-EBITDA,
EV-to-operating profit)
P/B (price-to-book value, price-to- 59.0 44.8
adjusted book value, book-to-market)
P/CF (price to some measure of cash flow) 57.2 54.6
P/S (price-to-sales or revenues) 40.3 45.7
P/E (price to some measure of earnings) 88.1 67.2
Other ratios 11.6 58.5
Source: CFA Institute.
historical examples of firms that were “relatively” undervalued compared

to their peers but generated poor future equity returns because the entire
peer group was overvalued. And conversely, it can be the case that a firm
may generate attractive long-term equity returns despite its value appear-
ing expensive relative to its peers. The more a market is mispriced, the less
pertinent relative valuation models would appear to be.
Eric Sorensen, president and chief executive at Boston-based PanAgora
Asset Management, commented on the firm’s approach to P/E valuation for
US stock selection:
Our processes for US stocks excluded the use of P/E ratios prior to the
2005–2007 bubble in quant equity. We consider that P/E has two flaws:
(1) one size fits all is too crude, and (2) P/E is ubiquitous in free datasets.
Advanced and actionable relative value calculations for sectors and stocks
must be specific to the businesses of the firms. P/E ratios are too abundant
and have been arbitraged away for 15 years. Our work, as examples, includes
(1) for financials, valuing the risk of the cash flows for the specific loan or
contract portfolios using default signals that adjust the discount specific
rates and (2) for biotech, valuing—with an option theoretic approach—the
probabilistic potential of the specific drugs undergoing FDA [US Food and
Drug Administration] test-phase trials using Markov chain models.

Multiples are not used at all at PGGM Investments. Lanters commented,

We consider multiples a distraction from what is important—that is to say,
real valuation. Multiples are a one-dimensional measurement. A rather silly
idea: The P or the E is low, high. It does not matter if the multiple is high if
the present value is low compared to the share price.
Pablo Fernandez (2002), a professor of financial management at the
University of Navarra’s IESE Business School, gives a more fundamental
critique of equity valuation methods. Fernandez considers that all meth-
ods other than DCF methods, though used, are not in accord with finance
theory.
As mentioned previously, relative valuations are not exempt from some
of the conceptual difficulties of absolute valuations. Here’s why: Consider the
most widely used relative asset valuation tool, the P/E. Ultimately, the stron-
gest formulation of the P/E principle states that an intrinsic, natural ratio
exists between the price of a stock and its earnings and that this ratio is uni-
versal. To allow for random fluctuations, we can state that for each asset the
following relationship holds:
P 
P =   E + ε, (2)
E
where P is the price of the stock, E represents the earnings per share, and
ε is random noise. Let’s leave unanswered for the moment the question of
the timeframe over which we compute E. If a true intrinsic P/E exists, then
Equation 2 would allow us to understand whether the stock is cheap or
expensive.
But determining a natural, intrinsic P/E is akin to determining a natural
rate of return. Sometimes the average P/E of a market is compared with a
historical average of the P/E of the same market. Figure 2.1 shows the cross-
sectional average P/E for the S&P 500 for the 146-year period 1871–2017.
As can be seen from Figure 2.1, for this 146-year period, the P/E had
a mean of 15.64, with values as low as 5.31 (December 1917) and as high
as 123.73 (May 2009, truncated in the graph). In the two most recent
decades, not only did the P/E increase, but fluctuations in the ratio also
grew. Clearly, considering the 146-year average (15.64) a natural benchmark
is problematic.
Given these difficulties and the fact that stocks in different sectors often
exhibit considerably different P/Es, in using a multiples approach, analysts
typically create small groups of similar (comparable) firms. A multiples

Equity Valuation
Figure 2.1. The Cross-Sectional P/E for the S&P 500 and Predecessor Indexes,
1871–2017
P/E
80
70
60
50
40
30
20
10
0
1871 83 95 1907 19 31 43 55 67 79 91 2003 15
Source: Constructed by the authors using data obtained from www.multpl.com/table.
valuation process is thus dependent on the selection of comparable firms (that

is, firms with similar growth patterns and returns on capital and subject to
the same macroeconomic forces) and their respective pricing.
Shiller’s cyclically adjusted P/E (CAPE) is a variant of the P/E that takes
a 10-year average of earnings as the basis for looking at current pricing levels
(see Siegel 2016). The popularity of the P/E comes from its focus on earnings.
EPS is a major component in the P/E valuation. Although EPS is generally
considered a primary driver of value—determinant of a share’s price—some
analysts believe EPS can be easily manipulated by, for example, share buy-
backs, as mentioned in our discussion of the DDM.
Fabozzi, Chen, Ma, and West (2015) started with the research question:
Given the wide mistrust in measures of earnings, why are earnings so widely
used as a financial metric—as opposed to, for example, cash flow? The authors
remark,
The stock market has an unusual fascination with earnings. A company’s
earnings, measured under specific accounting standards and tax laws, are
arbitrary at best, whereas cash flow, like the balance in a checking account,
is an actual number and subject to little interpretation. In practice, a com-
pany could file for bankruptcy while showing positive net earnings but
negative cash flows on its financial statements. (p. 511)

Using a sample of US stocks from CRSP12 for the 1970–2014 period, the
authors performed an empirical study to determine the sensitivity of prices to
a number of factors and concluded,
Stock prices are, on average, affected by short-term earnings. … We find
that cash flow pricing is used primarily to price what we classify as “nega-
tive” stocks—stocks that are generally characterized as illiquid, mispriced,
or having a shorter trading history, negative earnings, or negative market
performance. Thus, the practice appears to collide with modern finance
theories. (p. 511)
Nevertheless, the authors consider the wide use of earnings rational: The
use of earnings is part of conforming to the majority.
The CAPE model—or rather, the data used to estimate the model when
valuing US equities—was the subject of a recent critique by Jeremy Siegel
(2016), professor of finance at the Wharton School of the University of
Pennsylvania. Siegel suggests that even though CAPE is among the best
forecasting models for long-term future stock returns, the CAPE model
is “overpessimistic” (p. 41) in its return forecasts because of changes in the
way GAAP earnings used in the model are calculated.13 He advocates using
National Income and Product Account after-tax corporate profits to estimate
the model. This approach, Siegel believes, will result in higher explanatory
power and significantly higher stock return forecasts.
This idea raises a general question regarding the input data when using
multiples: Do we use trailing or forward-looking multiples? A trailing multi-
ple is a multiple based on historical data; a forward-looking multiple is a mul-
tiple computed on forecast data. Value investors, including Benjamin Graham
and Warren Buffett, prefer historical data. Janet Lowe (2010) reports that
Buffett commented, “I have no use whatsoever for projections or forecasts.
They create an illusion of apparent precision. The more meticulous they are,
the more concerned you should be. We never look at projections, but we care
very much about, and look very deeply at, track records.”
A problem with using historical data is that for a firm whose earnings
change rapidly, the measure will lag.
A problem with using future market multiples is the universal problem
with forecasts—that they may be inaccurate. In their 2002 paper, how-
ever, Liu, Nissim, and Thomas report that forward earnings measures using
12
CRSP is the Center for Research in Security Prices at the University of Chicago Booth
School of Business.
13
GAAP is a standard framework of guidelines for financial accounting used in Canada, the
United Kingdom, and the United States.

Equity Valuation
one-year or two-year forecasted earnings perform better than historical earn-

ings measures. Using a sample of companies trading on the New York Stock
Exchange, the American Stock Exchange, and NASDAQ , they found that
forward-looking measures are generally more accurate predictors of value
than historical measures. P/Es based on forecasts of income (net and operat-
ing) were preferred by slightly more than 81% of the participants in the CFA
Institute Survey of Professional Practice (Pinto et al. 2015), and trailing P/Es
(net and operating) by just over 13%.
Robeco Institutional Asset Management (2016), an Amsterdam-based
asset manager with €137 billion under management, says CAPE is its pre-
ferred approach to valuing stocks but has found the original Shiller CAPE to
be United States specific; it has therefore constructed a worldwide measure.
Although Robeco uses historical data, it has found that CAPE’s predictive
ability remains good compared with other valuation measures.
The debate about whether to use historical data or projected data is ulti-
mately ill conceived. By nature, every financial decision is forward looking—
that is, based on forecasts. Even if investors use historical data, they make
projections. The question is not whether to make projections but how to do so.
Analysts might forecast ratios and choose portfolios on the basis of forecasts
of ratios or portfolio returns; that is, they might choose portfolios on the basis
of current or past ratios.
Before we make additional observations about the valuation methods dis-
cussed here, we want to mention another heuristic popular in valuations—the
ratio of the percentage of total market capitalization to the GNP or GDP of
the relevant country. This ratio is often referred to as the Buffett ratio because
Warren Buffett told Fortune magazine journalist Carol Loomis (2001) that
he considers it likely “the best single measure of where valuations stand at any
given moment.” This ratio concerns aggregate equity returns, however, not
individual stocks; we will examine this issue in Chapter 4 in the discussion of
fair value and price distortions.
Valuation Methods Compared

“Beware of value illusion,” warns Steven Greiner, senior vice president at
Charles Schwab’s Equity Ratings Department and author of Ben Graham Was
a Quant: Raising the IQ of the Intelligent Investor (2011), who says, “A value
trap is a stock that looks underappreciated by the market as measured by typi-
cal relative valuation measures but in reality is low priced for good, funda-
mental business reasons and is a value illusion.” Although our intention here
is not to provide a review of the numerous texts on the issue of valuation
methods, we will mention briefly some academic studies that show a variety

of opinions on these methods’ usefulness. Let’s look first at the question of

present value models.
In “Valuation: The State of the Art,” Penman (2016) critiques standard
valuation models from three points of view: accordance with the established
theory of finance, practicality of use, and accounting methods and data used.
In introducing the subject, Penman writes,
There is a pervasive skepticism about formal valuation models, so much so
that practitioners often discard them, preferring rough-cut methods such as
pricing on the basis of comparables or simple P/E ratios. (p. 3)
As mentioned previously, although market multiples are the most widely
used valuation methods, they are considered by some to be less relevant than
present value models because they are only relative pricing measures: They
inform on the value of an asset relative to that of other assets, not the intrinsic
value of the asset. Behind their use is the belief that markets will eventually
identify and correct “mispricings.” In other words, the use of multiples allows
an analyst to forecast future price movements of a given stock. Multiples
can also be used to forecast earnings. If the analyst knows the multiple and
assumes that it will not change in the forecasting horizon, the analyst can
forecast prices by forecasting earnings.
In a note to business executives on the use of multiples, McKinsey con-
sultants Goedhart, Koller, and Wessels (2005) cite several ways by which
multiples can mislead. Their list includes (1) the difficulty in selecting truly
comparable firms within the same sector, (2) the fact that different multiples
can suggest conflicting conclusions, and (3) the idea that not all multiples are
meaningful in various contexts.
What value measures are good forecasters of performance? Wilcox
commented,
It is worth repeating that a good model of the current price as a function
of balance sheet, income statement, and cash flow statement may accu-
rately forecast convergence between future model values and prices for an
ensemble of firms but be of very little use in forecasting relative returns.
The convergence may come either because values change—poor accounting
is one reason (the future may reflect facts known but not in the accounting
for valid reasons)—or prices change, and only the latter indicates possible
mispricing and enhanced returns.
In the academic literature, a number of authors have contributed to the
fundamental question of what value measures are good forecasters of future
performance. Stanford University’s Lee (2003) offers a review of criteria
for choosing a valuation approach. He begins by observing that valuation

Equity Valuation
methods can be either relative or direct. Relative valuation methods include

all ratios, while direct methods include the balance sheet approach, the con-
tingent claims approach, and the cash forecast approach.
Lee notes that the balance sheet approach does not consider intangibles
and the contingent claims approach has many problems in estimation; he thus
focused on cash flow methods. According to Lee, these methods are similar
and yield similar results:
A forecast-cash-flow approach encompasses various derivatives of the divi-
dend discount model (DDM), such as the DCF model, the Edwards–Bell–
Ohlson (EBO) model, the economic value added (EVA) model, and the
residual income model (RIM). All of these models are essentially the same;
like different brands of camera, they work the same way. (p. 5)
Lee believes that these models are conceptually sound but observes that
they require the evaluation of three key elements: cash flows, terminal value,
and discount rates. As for discount rates, as we noted in Chapter 1, evaluat-
ing them is more than an estimation problem: It is a conceptual problem.
Estimating cash flows involves judgment. Lee notes that we cannot separate
forecasting cash flows and forecasting terminal value. The specific notion of
terminal value that we adopt is linked to the notion of cash flow that we
adopt. The two go together.
Moving to multiples, Lee observes that the choice of multiples goes
together with the choice of peers:
Multiples are simple to apply as a valuation tool, but valuations derived
from multiples can be subjective. That is not to say that a DCF model will
not produce a subjective valuation, but the key to a good multiple-based
approach is a judicious selection of peers. (p. 10)
Lee recommends the use of fundamental analysis to determine the group
of peers and their median values. Warranted multiples are those multiples
that apply to companies preselected with standard valuation tools. The key
takeaway: the multiples approach works well if the peer group is correctly
chosen.
Gray and Vogel (2012) performed a thorough analysis of various ratios:
earnings to market capitalization (E/M), EBITDA to total enterprise value
(EBITDA/TEV), free cash flow to total enterprise value (FCF/TEV),
gross profits to total enterprise value (GP/TEV), book value to market value
(B/M), and forward earnings estimates to market capitalization (FE/M)—all
expressed in yield format—for the 1971–2010 period. They found that rela-
tive to other valuation metrics, EBITDA/TEV is the best valuation metric
to use as an investment strategy. Eliminating stocks below the 10% NYSE

market–equity breakpoint, they found that for the period under study, an
annually rebalanced equal-weighted portfolio of high-EBITDA/TEV stocks
earned annual returns of 17.66%, with a 2.91% annual three-factor alpha.
Gray and Vogel concluded that this measure compares favorably with E/M;
cheap-E/M stocks earned 15.23% a year.
Actually, an equal-weighted portfolio is, in itself, a good active strategy.
DeMiguel, Garlappi, and Uppal (2009) claim that the equal-weighted port-
folio is very difficult to beat.
In the aforementioned Gray and Vogel study, the authors also observe
that value-weighted portfolios exhibit similar results, though returns are
smaller than those of equal-weighted portfolios. This result is reasonable given
that equal-weighted portfolios take advantage of the relative mean-reverting
behavior of stocks. Interestingly, they also found that using forward estimates
based on analysts’ consensus yields produced the worst performance.
Some, including McKinsey’s corporate finance practice in New York,
consider the P/E—ubiquitous as it is—distorted in its traditional form by dif-
ferences in capital structure and other nonoperating items, such as restructur-
ing charges and write-offs. They advise using EV/EBITA or EV/EBITDA,
the most widely used market multiples after P/E, according to participants
in the 2015 CFA Institute Survey (see Pinto et al. 2015). McKinsey’s Nolen
Foushee, Koller, and Mehta (2012) believe that these multiples do not suffer
from distortions that affect earnings ratios. Nevertheless, they write,
Comparisons based on enterprise-value multiples typically reveal a very
narrow range of peer-company multiples. A closer look at the US consumer-
packaged-goods industry is illustrative. From 1965 to 2010, the difference
in EV/EBITA multiples between top- and bottom-quartile companies was,
for the most part, less than four points, even though the industry is fairly
diverse, including companies that manufacture and sell everything from
household cleaners to soft drinks.
When we examined more closely matched peers at a given point in time,

we found even narrower ranges: for a sample of branded-food companies,
for example, EV/EBITA multiples ranged from 10.6 to 11.4. For medical-
device companies, the range was 8.4 to 9.7. In ranges this narrow, any
differences between true peers at a given point in time are typically unre-
markable. A company’s position in the ranking is likely to be quite variable
simply as a result of normal share-price fluctuations. (p. 3)
In two papers, Liu, Nissim, and Thomas (2002, 2007) looked at the rela-
tive performance of multiples. In their 2002 paper, they studied multiples
as variables for forecasting market prices in the US market. Examining the

Equity Valuation
valuation performance of a comprehensive list of value drivers, they found

that multiples derived from forward earnings explain stock prices remark-
ably well. Pricing errors were within 15% of stock prices for roughly half
their sample.
In their 2007 paper, Liu et al. extended their 2002 study by using forecasts
of operating cash flows, dividends, and earnings to compare the forecasting
performance of earnings and cash flow multiples in several countries. They
concluded that valuations based on earnings forecasts are remarkably accu-
rate, suggesting that earnings multiples be preferred over cash flow multiples.
In discussing the various relative valuation models, Alfred Slager, from
the TIAS School for Business and Society and the Dutch pension fund SPH,
commented that the choice of one method over another more or less depends
on the horizon, adding,
I tend to discard [the information from market multiples] for individual
stocks. Having once been a portfolio manager and tracking this informa-
tion made me realize that financial analysts also suffer from a form of peer
pressure—the forecast returns or multiples were too closely clustered.
The close clustering of multiples was commented on by McKinsey consul-
tants Nolan Foushee, Koller, and Mehta (2012, pp. 1, 4). Noting that within
mature industries and regardless of performance, multiples vary little among
true peers, they remark,
Companies may occasionally outperform their competitors, but industry-
wide trends show a convergence of growth and returns that is so striking
as to make it difficult for investors, on average, to predict which companies
will do so. … Across the economy, we have found substantial convergence
of revenue growth across companies.
They cite Wal-Mart and Starbucks as examples of companies whose mul-
tiples have fallen into line with those of their peers as growth slowed.
Figure 2.2, which is based on their article, shows the median portfolio
growth, in percentage terms, of US nonfinancial companies grouped by com-
parable revenue growth at the time of portfolio construction.
Because any valuation approach has limits (e.g., the present value approach
requires additional consideration of market factors; the comparables approach
is less meaningful in overvalued markets), using a combination of approaches
is perhaps a natural conclusion. Lleo remarked:
To a great extent, market multiples are a reflection of current supply-and-
demand conditions. This makes them valuable as a summary measure to
complement a DCF approach or as a quick rule of thumb to get an initial
estimate of how cheap or expensive an asset is. However, they do not scale

Figure 2.2. US Nonfinancial Companies Grouped by Comparable Revenue Growth

at Time of Portfolio Formation
Median Portfolio Growth (%)

35
30
25
20
15
10
–5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Years since Inception of Portfolio
>20% 15%–20% 10%–15%

5%–10% <5%
Notes: Companies were those with inflation-adjusted revenue of at least $200 million that were
publicly listed from 1963 to 2000. Companies were divided into five portfolios based on their
growth rate at the midpoint of each decile (1965, 1975, 1985, and 1995). Portfolios were then
aligned chronologically from Year 0 to Year 15, and their median growth rates were compared.
Source: Nolen Foushee et al. (2012).
easily (is the P/E ratio of a small firm comparable to the P/E ratio of a
large firm?) and are overly sensitive to the choice of a universe of compa-
rable firms (can we compare the EV-to-EBITDA ratio of two firms with
different strategies and product mix even if they appear to compete in the
same sector?).
This means that we cannot rely solely on a relative measure to value stocks.
Their main use is to complement DCF methodologies by providing addi-
tional vantage points from which we can assess the value of a corporation.
Cornell and Gokhale (2016) developed a corporate valuation model
that uses both market comparables and the DCF method. Their valuation

Equity Valuation
model—which they call an “enhanced multiple” corporate valuation model—

is based on the implied cost of equity capital. It takes into account the full
term structure of earnings forecasts but does not require the estimation of
the cost of equity capital. In empirical tests, they found that their model per-
forms significantly better than the DCF model but only slightly better than
standard market multiples methods; like the multiplier models, it requires the
identification of comparable companies.
The use of market multipliers is associated with traditional active
management but is not the exclusive preserve of traditional managers. For
example, the indexing pioneer Research Affiliates does what it terms “fun-
damental indexing,” which uses a firm’s fundamental attributes, such as sales,
earnings, book value, or a combination of these, in establishing index weights
and rebalancing an index. Clearly, fundamental indexing is no longer a pas-
sive strategy. In the 19 October 2016 Wall Street Journal article “The Hidden
Weaknesses of Index Funds” (see Jakab 2016), Robert Arnott, the founder
of Research Affiliates, says that the firm has found that weighting an index
by individual fundamental attributes such as earnings, sales, book value, or
a combination of such factors is superior to weighting an index by company
market capitalizations.
Additional Questions Concerning Valuation Methods

From the previous sections, one can reasonably conclude that fundamental
investing, though theoretically sound in practice, requires skilled judgment.
Methods based on multiples cannot be considered theoretically sound but
are robust heuristics that require a considerable amount of judgment. Let’s
address some additional questions regarding valuation.
Can Equity Valuation Models Be Trusted?
The concept of future prospects and particularly of continued growth in the
future invites the application of formulas out of higher order mathematics to
establish the present value of the favored issue. But the combination of pre-
cise formulas with highly imprecise assumptions can be used to establish,
or rather justify, practically any value one wishes, however high, for a really
outstanding issue. (Graham 1949, pp. 315–316 in revised fourth edition)
In Penman’s recent article “Valuation: The State of the Art” (2016), he
focuses on valuation as business valuation and accounting:
Investing is not a game against nature, but against other investors. Thus
one does not have to discover a value as if it exists in nature; the onus is not
on an analyst to come up with a valuation, but merely to accept or reject
the value in the market price. Accordingly, valuation models are not for

valuation; they are for challenging the market price. … But that challenge
is successful only to the extent of the quality of the accounting in the valu-
ation model. (p. 22)
Richard Bernstein, chief executive and CIO at Richard Bernstein
Advisors and formerly chief investment strategist at Merrill Lynch, offered
another angle: valuation models as a tug of war between buyers and sellers.
Bernstein commented,
Valuation is in the eye of the beholder, and there will always be a bid–ask
spread between the buyer’s and the seller’s valuation. There is a tug-of-war
between the seller of the asset and the buyer, and how high up the income
statement one values a company (i.e., sales instead of earnings) demonstrates
who is winning that tug of war. As an investor, one wants to skew the anal-
ysis as much as possible in one’s own favor. Yet, most valuation models are
based on a “pure” valuation, which typically favors the seller.
He added that, in his experience, using GAAP earnings rather than
operating earnings, EBIT, EVA, or other measures resulted in better portfo-
lio performance.
What Drives Valuations? Even assuming that we can determine the
intrinsic or at least the relative price of a firm’s stock by using valuation mod-
els, a number of important questions remain. First, what drives the valua-
tion methods? Although the idea that growth alone drives multiples is widely
believed, McKinsey consultants Goedhart, Koller, and Wessels (2005)
write, “In reality, growth rates and multiples don’t move in lockstep. Growth
increases the P/E multiple only when combined with healthy returns on
invested capital, and both can vary dramatically across companies” (p. 8).
MSF Investment Management’s institutional portfolio manager Robert
M. Almeida, Jr. (2016) writes, “Fundamentals drive cash flow, cash flow
drives profits, and profits drive stock prices” (p. 3). Citing Compustat’s EPS
data for the 1994–2015 period, which are shown in Figure 2.3, Almeida
continues,
When we look back at companies that have made money versus those that
haven’t, we see those with profits outperforming those that lose money,
which isn’t surprising. But the magnitude of the performance is significant.
Over the past 20 years companies that were profitable were up more than
650% (cumulative), while unprofitable ones were down 23%. (p. 2)
Robert Jarrow, professor of investment management at Cornell University,
suggests that market prices are set not by fundamental values but by expected
(or desired) resale values. In a recent paper on equity prices, Jarrow (2016)

Equity Valuation
Figure 2.3. Cumulative Return for Positive and Negative Earnings, 1994–2015
Cumulative Return (%)

700
600
500
400 Positive Earnings

300
200
100
0
Negative Earnings
–100
–200
1994 97 2000 03 06 09 12 15
Note: Each portfolio of positive and negative earnings companies was rebalanced monthly and
market cap weighted.
Sources: Almeida (2016); Compustat EPS data as of 31 December 2015.
addressed the problem of pricing “anomalies”—those based on, for example,

firm size, B/M, price momentum, and seasonal effects. He suggests that
asset-price bubbles are the norm, not the exception, and that most stocks have
small price bubbles representing between 1% and 25% of their value. These
bubbles represent the expected resale value of the shares.
The Macquarie Equities Research team (2013b) also points to the limi-
tations of valuation models in identifying mispricings. Stocks, they say, can
be driven by factors that are difficult to capture within a valuation model.
Among these factors, they mention stock sentiment, management quality,
earnings visibility, and leverage—as well as the need to complement mispric-
ing signals with news flows on, for example, corporate action. (See Chapter 4
for more on mispricing signals.)
Do Valuation Methods Have Predictive Power? The second key ques-
tion is: Do valuation models have predictive value?
Brian Jacobsen, chief portfolio strategist at Wells Fargo Asset
Management, with over $450 billion in assets, suggests caution in using
intrinsic value to make predictions as we do not know the time period over

which reversion to intrinsic value is likely to take place, nor do we know the
stability of intrinsic value:
Intrinsic value is supposed to serve as a type of gravitational pull for the
market price of a stock, but too few analysts try to figure out how strong
that pull is. How long will it take to have the market price converge to
intrinsic value? How does the intrinsic value change as you wait? Factoring
in the uncertainty around how long you might have to wait and the uncer-
tainty around how the intrinsic value might change while you wait should
make even the best of analysts more humble in the way they put client capi-
tal at risk.
Ang and Bekaert (2007) looked at the predictive power of the present
value model. In their widely cited paper, they report that they did find predict-
ability in stock returns but suggest refocusing the debate in three directions:
First, our results suggest that predictability is mainly a short-horizon,
not a long-horizon, phenomenon. Second, the strongest predictability
comes from the short rate and not from yield variables with price in the
denominator. … Third, there are tantalizing cross-country predictability
patterns that appear stronger than domestic predictability patterns. (p. 47)
ATP’s Kjaer commented,
In the inter-sector cross section, we do find some predictive power in rela-
tive valuation models. Obviously, stocks can be/are cheap/expensive for
a reason, but on average, we still find value in these models in the cross
section. Across time or sector, we have at present a relatively low convic-
tion in these types of models. However, in the case of extreme mispricing
relative to the model, we do have some confidence in these types of signals
across time.
As for the widely used price-to-earnings ratio in its 10-year cyclically
adjusted version (CAPE), the model was found to have little short-term pre-
dictive ability in a study by Antti Ilmanen (2016), principal of the quanti-
tative asset management firm AQR Capital Management. Using the range
of CAPE values without anticipation of information (i.e., using information
that would have been known at the time the model was applied to time the
market) and going back to 1900, Ilmanen found that, over the full period,
the use of CAPE as a predictive model only “mildly” outperformed a buy-
and-hold strategy, with all the outperformance occurring in the first half of
the sample. It would have underperformed for the last 50 years of the 20th
century. His conclusion: Neither a doubling of the market nor a historically
high valuation is a reliable sell signal; only with hindsight is CAPE of use in
the predictability of future returns and hence as a market-timing measure.

Equity Valuation
The Amsterdam-based asset manager Robeco Institutional Asset

Management (2016), whose preferred approach to equity valuation is CAPE,
agrees that the valuation method is not a timing factor for short-term returns,
but it has found that CAPE does have some (not to be overstated) predictive
power for medium- to long-term returns.
Lleo believes that relative valuation models may have some additional
value as summary measures of market conditions. In particular, he remarked
that Campbell and Shiller (1988) showed that the log of the P/E, calculated
by using a moving average of earnings over the previous 10 years, has predic-
tive value with respect to real S&P 500 prices over a 10-year horizon. Lleo
added,
The current relatively high value of the CAPE does not imply that returns
will be lower in the future, but it suggests a significant and rising probability
that the inflation-adjusted returns on the S&P 500 over the next 10 years
will be low, and possibly negative. As with anything related to probabilities,
the outcome is never certain. We can only look at how these probabilities
are changing and adjust our investment strategy to this change.
Brandes’s Little agrees. He said that the firm believes the relatively high
CAPE likely portends lower future long-term returns but added that CAPE
says little about near-term future returns. These returns, he said, are probably
driven by sentiment and supply/demand dynamics as opposed to fundamental
value:
We believe that market prices are mean reverting to some long-term aver-
age when viewed in aggregate. While individual company returns may
vary greatly, aggregate returns for the market will be more a function of
factors such as real economic growth, demographics, and productivity
improvements.
Bradford Cornell (2014) looked at the relationship between the dividend-
to-price ratio and stock returns, internationally and over time. His objective
was to determine whether the relationship is constant and, therefore, predic-
tive of future returns. He found that although the dividend-to-price ratio has
been widely reported to be effective when applied to US data, it is not neces-
sarily predictive of future returns globally or over time. Its effectiveness varied
from country to country and over time for a given country.
Can Asset Managers Take Advantage of Mispricings? The third
key question is: Supposing that valuation models allow active managers to
identify underpriced equities, can managers actually take advantage of such
price “anomalies”? In a recent interview with Daniel Ben-Ami (2016), Burton

Malkiel, emeritus professor of economics at Princeton University and the

author of A Random Walk Down Wall Street (1973), remarked,
If it were true [that active management can outperform the market by iden-
tifying underpriced assets], then why is it that when you look over the last
several years that it isn’t two-thirds of the managers that are outperformed
by the index but more like 80–90% of the managers?
What If Peer Firms or Whole Markets Are Mispriced? The fourth key
question: What do investors do if the comparables are mispriced or, more spe-
cifically, when valuations of a sector or the whole market are high? Consider,
for example, the telecommunications, media, and technology (TMT) sector
in the period 1995–2001, when valuations were detached from economic fun-
damentals. Before the bubble burst, leaving investors with a loss of $5 trillion
in the market value of TMT companies from their high of March 2000 to
their low in October 2002, value investors saw clients walk out the door in
search of higher returns. Alpha Architect blogger David Foulke (2016) notes
that in the period of July 1998 to the end of February 2000, the NASDAQ
was up 145%, while Warren Buffett’s Berkshire Hathaway was down 44%,
underperforming the NASDAQ by 189 percentage points. Lleo remarked,
Overall, the return of financial assets, such as stocks, cannot stay discon-
nected very long from economic growth. This was Warren Buffett’s central
thesis when he proposed the market-value-to-GNP ratio as a measure of
mispricing shortly after the dot-com bubble burst.
Do Prices Return to Some Historical Mean? Implicit in the above

question is the question of a market mean—for individual assets and the mar-
ket in general. With market valuations as measured by the cyclically adjusted
CAPE in September 2017 some 70% above their historical average, can we
assume that the relatively high valuation implies that returns will be low in
the future? That is, will the CAPE return to some historical average? Or,
more generally, are market valuations mean reverting to some (long-term)
average? For, if there is no mean reversion, then investment strategies based
on evaluating a stock’s intrinsic value and the difference between that intrin-
sic (or fundamental) value and its market value are of little interest.
ATP’s Kjaer noted that in cases of extreme mispricing on the low side,
ATP analysts find that in general, “expensive P/Es” across time have rela-
tively low prediction power. He added, “They do seem to revert but over very
long time horizons, and, in addition, it is a question mark as to whether and
when the mean reversion is due to the P or the E.”

Equity Valuation
PGGM Investments’ Lanters agrees that mean reversion is “definitely a

long-term phenomenon, impossible to use for timing the market. It would be
a very poor compass.” His colleague Jaap van Dam, the firm’s principal direc-
tor of investment strategy, remarked,
Personally, I am becoming more and more convinced that a time-varying
risk premium is something in market valuation that informs long-term
expected returns. If you look at CAPE, Shiller himself suggests that the
more expensive decile will generate lower future expected returns versus a
low CAPE, which indicates higher future returns. There is some informa-
tion in this, but you must look at the circumstances.
Slager commented on mean reversion and its applicability to portfolio
management, saying,
As an aggregate indicator, I find information [obtained from the market
valuation exercise] very important. Research on ALM [asset–liability man-
agement], P/E valuation, etc., suggests a simple rule of thumb: Starting
valuations matter for the midterm returns, with an inverse relationship.
Intuitively, this makes sense too. For example, the lower the CAPE, the
higher the expected risk premium should be. In that sense, there seem to be
mean-reverting elements at work.
The problem is that in a portfolio context, there needs to be some agree-
ment on the investment horizon during which mean reversion might
materialize—7–10 years, for example—and it needs to be firmly embedded
in the investment policy framework. The investment horizon of trustees and
regulators (and investment managers for that matter) tends to be shorter
than the horizon needed to exploit the mean-reverting opportunities. This
means that the governance of the investment process should be inspired
by Ulysses: Tie the investment process to the mast; do not change course
under any circumstances.
In its study “The Price Is Right,” Macquarie’s research group (Macquarie
Equities Research 2013b) adopted the Chee–Sloan–Uysal (2013) valuation-
based framework to explain stock returns.14 The main source of data for the
Macquarie group was I/B/E/S for consensus analyst EPS estimates; the
region covered was Europe. The group’s findings—which are in line with
academic research—are that company fundamentals matter more over long
investment horizons. They reported that over a one-month period, valuation
14
Chee et al. (2013) formalized the Graham–Dodd approach to building a valuation-
based framework to explain stock returns and show that returns can be decomposed into
expected returns, returns driven by unexpected cash flow news, and unexpected discount
rate news.

models explained just 3% of the returns, rising to 9%, 17%, and 25%, respec-
tively, as they increased the investment horizon to 3-, 6-, and 12-month peri-
ods. The Macquarie Equities Research group cited the study by Chee et al.
(2013) showing that company fundamentals can explain about 60% of stock
returns over a five-year period.
A recent blog post by Damodaran (2016) raised a number of issues rela-
tive to the use of mean reversion, widely considered a robust underpinning of
many investment strategies. Damodaran remarked that, even if one believes
stock returns are mean reverting, using reversion might be tricky. First, the
mean will critically depend on the time period for which it is estimated.
Second, estimating the time to reversion is difficult but critical from the point
of view of investment strategies. In addition, structural breaks in the markets
can invalidate mean reversion. Creating a test strategy to understand the abil-
ity of CAPE to predict returns, Damodaran found that CAPE is better at
predicting short-horizon returns than long-term returns. Results, however,
depend on the choices made in estimation: Different results can be obtained
if slight changes are made in the estimation parameters. Damodaran con-
cluded with a warning that in these times of economic change, one has to
be particularly attentive in using ideas such as the mean reversion of CAPE.
“Statistical significance is not cash in the bank,” he wrote.
Another skeptical view on the use of mean reversion in asset management
comes from Charles Schwab’s Greiner:
I personally subscribe to the concept that financial asset prices and metrics
like valuation generally never mean-revert. Mean reversion implies station-
ary behavior in the time series. Since market prices are nonstationary, there
really is no mean to revert to. In physics (and in finance), the behavior called
“mean reverting” is more correctly called “anti-persistent,” as in the physics
of signal processing.
What about Momentum Markets in Which Asset Prices Deviate

from Their Intrinsic Value? Valuation is typically considered the preserve
of fundamental active managers who, with their research, keep markets effi-
cient; trend analysis is considered the preserve of quants and index manag-
ers. Investors using pattern or trend analysis have been accused of creating
momentum markets responsible for pushing stock prices up (or down) relative
to their intrinsic value. Quants reply that active managers can be guilty of
herd behavior that creates momentum markets.
These views were debated in 2014 by the quantitative fund manager
Man AHL’s academic advisory board (Man Group 2014). Nicholas Barberis,
a participant in the debate and professor of finance at the Yale School of

Equity Valuation
Management,15 argued that value investing and momentum investing might

actually be more similar than they appear: “According to under-reaction theo-
ries of momentum—for example, the slow diffusion of information theory—a
stock that has been trading up is also a cheap stock: not all information about
it has been absorbed into the price” (p. 5).
Barberis added that “momentum traders are actually expediting the
incorporation of fundamental information into prices” (Man Group 2014,
p. 5), thereby helping the price-discovery process. Other participants noted
that analysts’ “stories” about fundamentals also play an important role in
moving prices, for example, in driving up prices of TMT firms before the
dot-com bubble burst in early 2000. Douglas Greening, former chief risk offi-
cer at Man AHL, asked: “What is the difference between buying really hard
into a story and momentum trading?” (Man Group 2014, p. 5).
Slager had a different take on trend analysis:
Trend analysis is an answer to our fundamental psychological need to
observe patterns and make sense of the world, the stock market, because
the other answer—returns and patterns are highly randomized—makes us
highly insecure. Trend analysis is potentially destabilizing, but with a twist.
Due to momentum effects, it tends to extrapolate downward and upward
trends. However, stock markets tend to be choosy: In my observation,
the upward-trend forecasts are followed when stock markets are up while,
on the other hand, the upward-trend (i.e., mean-reversion) forecasts are
ignored when stock markets are down. Rationally, investors should act the
other way around. So, it is not so much trend analysis that is destabilizing as
the selective interpretation of the investors in the market.
ATP’s Kjaer observed that although anecdotal evidence in the very short
term might indicate that trend analysis leads to instabilities, in the medium
term, he does not consider this to be the case. This seems to be the evaluation
of most persons who gave their opinion on the issue.
Lleo remarked that trend analysis and momentum investing have devel-
oped rapidly, parallel to the rapid development of statistical and machine
learning adopted by trading desks, saying,
When implemented by a single trader, a momentum strategy will seldom
lead to price disruption. However, the collective action of a large number
of market participants implementing similar strategies will be procycli-
cal, leading to an increased risk of price instability and market disruption.
This is especially true when these strategies are implemented via algorith-
mic trading without human intervention. As more algorithms identify the
15
For his papers, see http://som.yale.edu/nicholas-c-barberis.

same trend, the trading activity creates a feedback loop, which may lead
to an amplification and acceleration of the trend. The risk of a flash crash
becomes substantial. These phenomena are particularly dangerous because
they are nonlinear, which makes them difficult to predict.
Lleo suggested putting safeguards in place: “Computing an intrinsic
value and using this value as an anchor in a trend-following algorithm—for
example, if you are more than X% above/below the intrinsic value, then stop
following the trend—can reduce the risk of following the crowd into flash
bubbles and crashes.”
What about Investor Objectives? A general question is, What are
the objectives of the investor? Is the investor striving to beat the market or
achieve some other goal? For Slager, the objective of a pension fund is not to
beat the market—a chapter he considers time to close—but to help the fund
achieve its goals. He cited added value for active managers in working with
funds to create new metrics and strategies to aid the fund to reach its goal.
“Trustees,” he said, “no longer need to be drawn into overly technical asset-
pricing discussions but can focus, instead, on what matters—assessing where
active management can work.”

3. Valuing Hard-to-Value Equities
What about valuing hard-to-value assets, such as initial public offerings

(IPOs) and privately held firms? What data and tools do analysts have, and
how relevant are they in the valuation process? Let’s start by looking at IPOs.
Valuing IPOs
Clearly, valuing IPOs can be problematic. Warren Buffett once famously said
that if he were teaching a finance course, he would ask students to evaluate
an internet stock, and any student giving an answer would flunk.16 (He made
this statement at a time when no internet stock had yet made a profit.)
In May 2016, Investopedia’s John Burke noted that 72% of the IPOs
issued in 2015 were trading below the issuance price a year later and that the
average return for a 2015 IPO stock issued in the United States was –19%.
According to data from FactSet, from 1 January to 23 December 2016, while
the S&P 500 Index was up 10.8%, the First Trust US IPO exchange-traded
fund (ETF) was up only 6.6% and the Renaissance IPO ETF was actually
down 0.4% on a year-to-date basis. The problem of IPOs trading below their
offer price and/or underperforming with respect to the overall market has
led to a loss of investor appetite, which is reflected in the number of compa-
nies going public on US exchanges. According to FactSet analyst Andrew
Birstingl (2016), only 106 IPOs were issued in 2016—the lowest number
since 2009, when 64 companies went public. The amount of money collec-
tively raised by these 106 IPOs was also down—to $20.2 billion (a 38.1%
decline from 2015), the smallest annual total since 2002, when gross pro-
ceeds were $19.5 billion.
Perhaps the IPOs that most retain the media’s attention are technology
IPOs, where performance has not been stellar. According to Reuters reporter
Dan Burns (2017), globally, shares of the 25 largest technology IPOs per-
formed poorly in their first 12 months on the public market: 16 of the 25
suffered declines from their debut-day closing price, with 8 of the 10 big-
gest falling by 25%–71%. The median one-year performance of the largest
technology IPOs was –22.3%. The medium-run performance of Snap’s stocks
following the 1 March 2017 IPO will likely affect investors’ appetite for IPOs
throughout the year.
Another explanation for the recent dearth of IPOs is offered by Gao,
Ritter, and Zhu in “Where Have All the IPOs Gone?” (2013). They note that
16
To view a related video, see www.youtube.com/watch?v=nrSB1sLgWLE.

the drop in IPO offerings was especially high among small firms and hypoth-
esize that the advantages of selling out to a larger organization have increased
relative to the benefits of operating an independent firm.
Back in 1994, Ibbotson, Sindelar, and Ritter (1994) wrote, “The
market has a great deal of difficulty in valuing issuing firms appropriately”
(p. 66). They identified three anomalies still present in IPO valuations
today: (1) short-run underpricing resulting in first-day returns that average
10%–15%, (2) cycles in the volume of new issues and the magnitude of first-
day returns, and (3) long-run (five-year) underperformance. Ibbotson et al.
consider these anomalies a challenge to the efficient market hypothesis and
conclude that raising capital “is subject to the whims of the market, as well as
the fundamentals of the company” (p. 74).
So, what tools do analysts have for valuing IPOs? Essentially, the same
tools discussed in Chapter 2 that sell-side and buy-side analysts use to value
publicly traded companies, but with some additional problems.
Penman (2016) comments thus on the valuation of IPOs:
While one cannot hope to pin down “intrinsic value” with certainty, valua-
tion aims to reduce uncertainty in investing, and standard approaches that
often introduce uncertainty do not serve us well. They even lend themselves
to “playing with mirrors.” Sell-side bankers like the models; set with the
“due-diligence” task of supporting an issue price with a formal valuation,
they look for a model that can establish, or rather justify, practically any
value one wishes, however high, for a really outstanding issue. But the
investor on the buy side of that issue, or a fiduciary of other people’s money,
is cautioned: caveat emptor; beware. (p. 4)
In “Valuing IPOs,” Kim and Ritter (1999) consider the usefulness of
various approaches as benchmarks for valuing IPOs. They report that valuing
IPOs on the basis of P/E, price-to-sales (P/S), enterprise value-to-sales, and
enterprise value-to-operating cash flow ratios has some predictive value when
used with earnings forecasts and adjusted for differences in growth and prof-
itability. However, the authors found that, when used with historical account-
ing numbers, multiples are imprecise in their ability to forecast future cash
flows of IPOs. They report a similar finding for another widely used valuation
method—the discounted cash flow (DCF) method.
Commenting on the use of market multiples for valuing IPO-issuing
firms, An Yan, a professor of finance at Fordham University’s Gabelli School
of Business, said,
The use of multiples to value listed firms where you have reliable earn-
ings and information sources is OK, but using multiples to value IPOs is
problematic: It is hard to define comparable firms, firms in an early stage of

Equity Valuation
development. An issuing firm cannot be compared to a listed firm: It might

be an industry leader in a new market.
In valuing young companies with low sales volumes and negative prof-
its, nonfinancial multiples might shed some light on valuation despite uncer-
tainty about potential market size, profitability, and required investments.
McKinsey consultants Goedhart et al. (2005) write:
Nonfinancial multiples compare enterprise value to a nonoperating statistic,
such as Web site hits, unique visitors, or the number of subscribers. Such
multiples, however, should be used only when they lead to better predic-
tions than financial multiples do. If a company can’t translate visitors, page
views, or subscribers into profits and cash flow, the nonfinancial metric is
meaningless, and a multiple based on financial forecasts will prove a supe-
rior result. Also, like all multiples, nonfinancial multiples are only relative
tools; they merely measure one company’s valuation compared with anoth-
er’s. As the experience of the late 1990s showed, an entire sector can become
detached from economic fundamentals when investors rely too heavily on
relative-valuation methods. (p. 11)
For Ibbotson et al. (1994), the difficulty in using comparable firms’ mul-
tiples (and other valuation methods) for valuing IPOs opens the door for a
role for investment banks. They write, “Because using the midpoint of the
offer price range results in smaller prediction errors than using comparables,
investment bankers apparently are able to do superior fundamental analy-
sis” (p. 436). The authors suggest that because of the dynamic information
exchange between the investment bank and institutional investors, banks are
able to achieve additional accuracy before setting the final offer price.
Peter Roosenboom (2012), professor of entrepreneurial finance and pri-
vate equity at the Rotterdam School of Management, found that underwriters
typically arrive at fair-value estimates by using three valuation methods—
multiples, dividend discount models, and DCF models—and that all three
valuation methods have similar accuracy, explainability, and (positive) bias
with respect to equilibrium market value. Using reports from underwriters of
their pre-IPO valuation process on 228 IPOs on the NYSE Euronext Paris
for the years 1990–1999,17 Roosenboom also found that underwriters, in an
attempt to encourage investor participation, deliberately discount fair-value
estimates when setting a preliminary offer price. “First-day underpricing
is part of the IPO process,” he commented. “It is the indirect cost of going
Unlike US or UK law, French law makes available how underwriters value companies and
17
the methods they use at a stage prior to taking a company public. Several other European
countries have such laws.

public. Part of underpricing is due to the deliberate price discount that attracts
investors in the early stage of the IPO process.”
Yan discussed the two-step process in valuing and pricing IPOs. In the
first step, the investment bank performs due diligence—the S-1 filings and
balance sheet analysis—to arrive at a rough idea of value based on fundamen-
tals. The second step is the “road show,” during which the investment bank
tests initial investor sentiment. “This,” Yan noted, “is key.” He continued,
The price is determined by the industry perspective and the marketing
environment. This is very important in pricing an IPO. In the end, the
price is determined by the information exchange between the underwriter
and investors, not so much by fundamentals. An investment bank cannot
underprice an IPO on grounds of the fundamentals—which might be det-
rimental to the interests of the issuing firm. The underwriter must find the
match between demand and the price. It is a question of market timing.
Firms wanting to issue an IPO find a window to go public and do so when
the market is receptive. The price depends on market sentiment, demand at
the moment of going public. It is not a question of fundamental value.
The positive role of investment banks as underwriters was studied by
Bajo, Chemmanur, Simonyan, and Tehranian (2016) in their paper on under-
writer networks, investor attention, and IPOs. The authors studied how
central lead underwriters arrive at pricing through an information exchange
with their investment banking network. This information exchange allows
the underwriter to both disseminate information on the issuing firms and
simultaneously extract information from institutional investors that will
prove useful in pricing the IPO. Bajo et al. found that IPOs underwritten by
more central lead underwriters are associated with higher absolute values of
offer price revisions, higher IPO and secondary market valuations, and higher
IPO initial returns. The authors also found that IPOs underwritten by cen-
tral lead underwriters are typically covered by a larger number of financial
analysts, have large institutional investors holding shares, and (subsequently)
have greater secondary market liquidity and better returns over a period of six
months to one year after issuance.
Matteo Bonaventura (now a buy-side analyst at Banor SIM) and
Giancarlo Giudici (2016) documented the positive role of pre-IPO book-
building activity in valuing and pricing IPOs in the Italian market for the
2000–09 period.18 Noting that one of the most common techniques used in
18
As in France, Italian law makes available information about how underwriters value com-
panies and the methods they use at a stage prior to taking a company public. Book building
is the process underwriters use to assist in price discovery when seeking to raise equity for

Equity Valuation
valuing IPOs, according to prospectuses, is the DCF method, the authors

performed reverse engineering to discover the short-term profitability implied
in the offer price. Their findings revealed that although a substantially large
mean forecast error was characterized by a significant optimistic bias in the
estimation of the future profitability, compared with ex post realizations, such
errors also characterize estimations by analysts evaluating non-IPO compa-
nies. They further noted that pre-IPO book-building activity plays a large
role in reducing the forecast error and revising expectations, whereas forecast
errors are not reduced by the market price of the first day of trading.
The significant optimistic bias in the estimation of future profitabil-
ity compared with ex post realizations that Bonaventura and Giudici (2016)
observed was studied by two academics (at that time) with degrees in engi-
neering at the start of their academic training, Amiyatosh K. Purnanandam
and Bhaskaran Swaminathan.19 Both were at Cornell University’s Johnson
Graduate School of Business when they researched their paper “Are IPOs
Really Underpriced?” (Purnanandam and Swaminathan 2004). They studied
more than 2,000 relatively large-capitalization IPOs in the 1980–97 period.
Despite the well-known phenomenon of initial underpricing, they found that
at the offer price, the median IPO was significantly overvalued relative to
valuations based on industry peer price multiples, such as price-to-EBITDA,
P/S, and P/E. Depending on the peer-matching criteria, overvaluations
ranged from 14% to 50%. Cross-sectional regressions showed that the IPOs
in the study provided first-day returns but low long-run risk-adjusted returns.
Purnanandam and Swaminathan write that the overvalued IPOs in their study
had lower profitability, higher accruals, and higher analyst growth forecasts
than “undervalued” IPOs. Ex post, the projected high growth of overvalued
IPOs failed to materialize, while their profitability declined from pre-IPO
levels. The authors suggest that investor overconfidence might be at play.
Severin Zörgiebel (2016a), a researcher in finance at Goethe University
Frankfurt, studied the valuation of 2,655 US IPOs between 1994 and 2013, a
period that stretches from before the dot-com bubble and its bursting through
clients via a public offering—either an IPO or a follow-on public offering. The bids and the
number of shares that a bidder wants at the bid are collected from both institutional and retail
investors during the period the offer is open. After the bidding process is closed, the issue
price to be used by the underwriter is then determined by the demand generated from the
book-building process.
19
Purnanandam is now a professor of finance at the University of Michigan’s Ross School
of Business; Swaminathan is now a partner and the director of research at LSV Asset
Management in Chicago and an adjunct professor of finance at Northwestern University’s
Kellogg School of Management. Swaminathan’s work on valuing the Dow Jones Industrial
Average won a Graham and Dodd Award of Excellence from CFA Institute.

the 2008 market crash and beyond. His study includes both technology and
nontechnology companies, loss-making as well as profitable IPO issuers, and
already listed firms. Loss-making IPO issuers made up roughly half of all
IPOs in the United States during the period studied. Using a variety of valu-
ation methods adopted by researchers working on similar studies and valu-
ation methods from the mergers and acquisitions field (e.g., Rhodes-Kropf,
Robinson, and Viswanathan 2005), Zörgiebel found that IPOs are, in gen-
eral, valued higher than listed peer companies and that IPOs with negative
earnings were valued higher than IPOs with positive earnings. Figure 3.1
shows the discontinuity of the market-to-book (M/B) premium versus the
net income margin.
Another finding is that IPOs with negative earnings provide long-term
underperformance relative to both listed companies and IPOs with positive
earnings.
Zörgiebel suggests that factors other than higher growth expectations
might be part of this phenomenon and identifies media coverage and hetero-
geneous beliefs as playing a substantial role in IPO valuations.
Why are investors ready to accept high valuations for IPOs? Degeorge,
Derrien, and Womack (2007) explored the role of “analyst hype” (p. 1021)
Figure 3.1. Valuations of IPOs with Negative Earnings
M/B Premium (%)

120
100
80
60
40
20
–20
–0.5 –0.4 –0.3 –0.2 –0.1 0.1 0.2 0.3 0.4
Net Income Margin
Source: Zörgiebel (2016a).

Equity Valuation
in the IPO process. In a study of the French market, they found that book
building as a selling procedure was related to the perceived benefit analyst
coverage provided to the success of the issuance and (possibly) post-IPO
coverage in an attempt to ensure aftermarket liquidity. They also found that
analysts affiliated with the lead underwriter issue more (and more favorable)
recommendations for the book-built IPOs and that lead underwriters “lean
on” unaffiliated analysts to provide favorable coverage. Analysts affiliated
with lead underwriters were also found to put out positive recommendations
(“booster shots,” p. 1023) following the poor stock market performance of
recent book-built IPOs.
Clearly, IPO valuations are more subject to qualitative or behavioral ele-
ments than are shares of already listed firms. Data on the fundamentals of
an IPO business, such as cash flow, the balance sheet, and profitability, are
often unreliable and/or not audited. Such factors as demand or the “narrative”
(or marketability of the business), including assumptions about the company’s
future growth projections, play a large role.
Chemmanur and Yan (2017) document how advertising by firms going
public affects both the valuation and price revisions of the IPO as well as
long-run post-IPO stock returns. In their sample of US IPOs from 1990 to
2007, they compared IPO firms with high and low advertising intensity in
the years running up to their IPOs. They found that companies going pub-
lic with high advertising intensity prior to their IPOs (1) are valued higher
both in the IPO and in the immediate aftermarket, (2) are associated with
greater upward price revisions from the pre-IPO filing range means, and (3)
have lower long-run post-IPO stock returns. Specifically, Chemmanur and
Yan reported that among their sample firms, the profitability (EBIT/assets) of
low-advertising-intensity firms was, on average, 9.9%, whereas that of high-
advertising-intensity firms was, on average, –6.7%.20
Zörgiebel’s (2016a) study of IPOs with negative earnings identified mar-
keting campaigns by venture capitalists and underwriters as the valuation
drivers in the IPO process. In a study specific to media coverage and startup
valuations, Zörgiebel (2016b) notes that as of October 2015, Crunchbase
counted 153 venture capital–funded startups in the Unicorn Club ($1 billion
or over) with a (post-money) valuation of about $529 billion and total funding
20
Interestingly, in their working paper “Advertising, Investor Recognition, and Stock
Returns,” Chemmanur and Yan (2011) found a similar pattern for listed stock returns. They
tested Merton’s investor recognition theory by examining the impact of advertising on stock
returns and found that a higher level of advertising growth is associated with higher stock
returns in the high-advertising year and lower ex post long-run stock returns.

of $79 billion—about 10% of the funding of the NASDAQ 100 Index or

more than 40% of the funding of the German DAX30 market capitalization.
Zörgiebel (2016b) posed the research question: What factors are driving
these high valuation levels? Referring to research by Hillert, Jacobs, Müller
(2014) showing that media coverage plays a large role in investor behavior
in an environment of high uncertainty, Zörgiebel undertook an empirical
study using Thomson VentureXpert, Crunchbase, and LexisNexis data for
the 1995–2015 period. He also took into consideration technological changes,
such as internet and mobile communications, that increase and speed up
information exchange. His empirically based finding was that, not precluding
the idea that other factors might have an incremental effect, startup (espe-
cially, Unicorn) valuations are driven to a large extent by increasing media
coverage both before a venture capital–funding round and before a public
IPO. Figure 3.2 illustrates transaction value versus media coverage per day.
Figure 3.2. Average Transaction Value vs. Media Coverage per Day of 153 Venture
Capital–Funded Startups in the Unicorn Club, 1995–2015
Transaction Value, Adjusted (US$ billion) Media Coverage/Day

2,500 2.5
2,000 2.0
1,500 1.5
1,000 1.0
500 0.5
0 0
1995 97 99 2001 03 05 07 09 11 13 15
Average Transaction Value (adjusted) Media Coverage/Day
Source: Zörgiebel (2016b). Reprinted with permission.

Equity Valuation
So, is income as researched by Aggarwal, Bhagat, and Rangan (2009)

negatively correlated with the offer value and the offer value positively corre-
lated with media coverage? In general, what can be said about the long-term
performance of IPOs? As mentioned, Ibbotson et al. (1994) identify three
anomalies in IPO valuations, the third of which is long-run underperfor-
mance. Roosenboom remarked,
There has been some discussion on the issue. Initial studies show poor long-
term performance. A myth or a reality? It depends on how performance is
measured. Are size, growth opportunities taken into consideration? What
is the benchmark one is using?
One possible reason for any long-term underperformance is offered by the

theory of the divergence of opinion. According to this theory, when an IPO
is first issued, optimistic investors purchase the stock, set the price. It is
difficult to short-sell the stock in the early stages as the market is not yet
developed. As time goes on, pessimistic sellers enter the market; short-
selling begins. You then have a mix of optimistic and pessimistic buyers/
sellers who co-determine the market price. It is the problem of the timing
on going public.
Yan noted, “In the beginning [of the IPO process], it is hard to sell short;
pessimists are not yet present in the market. Then there is profit taking over
the long term. Entrepreneurs of the issuing firm exit, and there is downward
pressure on the pricing.”
In summing up the drivers of IPO valuations and market value, Zörgiebel
said,
Based on the general IPO literature and my research, there are many driv-
ers of IPO valuations, the main drivers being growth expectations, profit-
ability, ownership (venture capitalists with a good reputation, for example),
underwriter quality, market timing (IPO bull phases, for example), indus-
try/technology trends, the number of shares issued, and last but not least,
the media coverage (you might call it “hype”) before an IPO. It is hard to
say which factors are the most important, but I guess it is often a mixture
of several. In addition, these driving forces can reinforce themselves and
increase or decrease valuation levels even further.
For example, based on the theory of heterogeneous beliefs (Miller 1977),

only the most optimistic investors with the highest valuations in mind par-
ticipate in IPOs and receive shares. The more uncertainty there is around
the value of the IPO firm, the wider the range of beliefs about the true
value. As a result, extreme valuations are more likely. When the IPO firm
is loss-making, has high growth expectations, and the whole situation is

fueled by media coverage and hype around that firm, uncertainty is higher,
and valuation levels might be pushed upwards.
The impact of uncertainty on investor behavior was studied by Alok
Kumar (2009), a professor of finance at the University of Miami School of
Business Administration. Kumar found that in situations where valuation is
highly uncertain and stocks are difficult to value, people are more likely to
use heuristics (rules of thumb), thereby creating stronger behavioral biases
and bigger investment mistakes. Kumar writes, “Both stock-level and market-
wide uncertainty adversely influence [investors’] decisions” (p. 1377). These
remarks pertain equally to IPOs and private equity discussed below.
The role of psychology was mentioned as a leading factor in IPO invest-
ment decision making by Sébastien Lleo, a finance professor at NEOMA
Business School (France). Citing work by Hersh Shefrin suggesting that
behavioral biases such as framing and conflicts of interest explain a part of the
long-term underperformance of IPOs, Lleo said,
The effect of these biases is precisely to disconnect decision makers from the
type of coherent and articulated mental framework provided by DCF meth-
ods. With IPOs, as with mergers and acquisitions, the risk in the decision
process of both managers and investors is that the exciting story (“writing a
new page in corporate history,” “grabbing some financing when the market
is hot,” and so on) will eclipse down-to-earth valuation. When this occurs,
you may observe long-term underperformance for both the company and its
stock.
Cornell and Damodaran (2014) studied the role of market sentiment
in the case of Tesla Motors. Following DeLong, Shleifer, Summers, and
Waldmann (1990), Cornell and Damodaran describe market sentiment as
a “belief about future cash flows and investment risks not justified by facts
at hand” (p. 139). Tesla went public in June 2010 (just one month after
Toyota Motor Corporation announced that it was taking a 2.5% stake in the
California company), with shares priced at $17. Cornell and Damodaran stud-
ied the stock’s rise from $36.62 on 22 March 2013 to $253.00 on 26 February
2014. To value Tesla’s shares, they constructed a DCF model and ran it at
three points: before the start of the run-up in price, during the run-up, and
at the end of the run-up. They found, “The valuations all yield value estimates
that are well below the market price (at the time of the valuation), with the
price at more than two and one-half times an aggressively optimistic value
estimate,” concluding that “investor sentiment played an important role in the
run-up” (p. 150).

Equity Valuation
In an endnote, Cornell and Damodaran (2014) underscore the impor-

tance of the assumptions used in the models. They cite a 2014 Morgan Stanley
report that put Tesla’s target price at $320. That report included the DCF
model used to arrive at the target price. In this case, the assumptions were a
forecast of rapid growth and minimal new investment. Morgan Stanley was
one of four principal underwriters of Tesla Motors’ 2010 IPO. Note that at
the time of writing this monograph (mid-year 2017), Tesla stock was valued
at around $320.
Can the valuing and pricing of IPOs be improved upon? Banor SIM’s
Bonaventura remarked,
I think that IPO prices have to accommodate the principal parties involved.
Being a transaction, it is a sort of arrangement between [institutional]
investors, the issuer, and the underwriters. In this case, I think that peer-
group comparisons and DCF models are simply an artificial construct to
justify underwriters’ valuations.
Bonaventura cited recent research by Paleari, Signori, and Vismara
(2014) showing that the peers selected by underwriters are systematically
biased toward overvalued companies, while candidate peer firms that would
make a given IPO appear overvalued are systemically excluded. The authors
found that, on average, comparable firms published in official prospectuses
have valuation multiples 13%–38% higher than those obtained from match-
ing algorithms or selected by (other) sell-side analysts. They further noted
that even if IPOs are priced at a discount compared with peers selected by the
underwriter, they still sell at a premium when compared with alternatively
selected peers. Paleari et al. suggested that this bias in an underwriter’s selec-
tion of peers helps explain the poor long-run performance of the IPO.
Roosenboom’s take on the possibility of improving the valuing and pric-
ing of IPOs is as follows:
If you look at the amount of underpricing, there is not much progress;
underpricing has been steady over time and over markets. It is still the case
that investors need to be enticed. Consider book building and auctions. In
the United States, Google went public with an auction. The idea was that
auctions led to greater efficiency in the pricing, as they provided collective
information on demand. In France, for a long time, the auction process was
the most prevalent, and there was less underpricing on the first day. In the
auction process, you set a minimum price, and investors bid up. Through
the market, it was possible to understand interest in the share, the demand.
However, now the main selling mechanism just about everywhere is book
building.

Fordham University’s Yan agreed that the possibility of improving the

valuation and pricing of IPOs is small:
Nothing can be changed. It is a question of rational investors and efficient
markets. You cannot throw out theory and make a new one based on irra-
tionality: You cannot control irrationality. Could you introduce more moni-
toring to the book-building process? I think we need to understand that
markets will come back to rationality, efficiency. The big picture is: Markets
are functioning well. Investment banks will always be rational in pricing
IPOs. They have to take advantage of irrational investors. In a hot market,
everyone is overoptimistic; the underwriter cannot underprice.
One possible route to improving the valuing and pricing of IPOs might
be to encourage rational behavior, though our interviewees were skeptical
about the possibility of changing investor behavior. Lleo said,
Individually, we have the tools to de-bias and reach more consistent values
for the IPOs. But collectively, behavioral biases tend to be recurrent and
resilient. I would not be surprised if we keep on observing an underperfor-
mance among IPOs for a long time.
Zörgiebel, too, suggested that creating investor awareness might help:
When investors are aware of the major driving forces and a potential bias
in their thinking, they can act accordingly. Research in that area helps to
detect such biases and pitfalls. Financial literacy might be the best way to
improve the way IPOs are priced.
But echoing Yan’s evaluation that markets will return to rationality and
efficiency, he added,
There are a lot of research papers showing that investors are often far from
being rational when valuing or investing in stocks—especially when there
are professional and private investors active in the market with different
incentives and different levels of financial literacy. Consider the so-called
lottery stocks—stocks that bear high risks but also a certain amount of
growth expectations. They are probably overvalued, but investors hope
that these companies will turn into multi-billion-dollar companies. Some
of them do—for example, Facebook. But many will fail. Investors in
such stocks are not acting rationally; they are more or less gambling. It is
interesting to see that IPOs are, on average, overvalued at the time of the
IPO. But over time, this overvaluation decreases, and values tend towards
those of industry peer groups. Apparently, the market learns over time and
adjusts valuations towards the peer or intrinsic valuation. Uncertainty levels
decrease, more and more hard facts about the firm become available, and
valuation models can be adjusted.

Equity Valuation
Valuing Privately Held Companies

As with IPOs, many of the methodologies used in valuing privately held
companies are the same as those used in valuing public companies. In partic-
ular, these methods include peer-group multiples and DCF analysis. A recent
survey by the consultancy Grant Thornton (“Private Equity Valuations” 2015)
found that the method most frequently used to value privately held firms is
multiples (87.2%), followed by the DCF method (76.9%). In the 2015 CFA
Institute Survey (Pinto et al. 2015), 92.8% of the respondents reported using
multiples, and 78.8% used a discounted present value approach.
Applying these methods to evaluate privately held firms adds another
set of problems to the usual challenges in using multiples or DCF analysis.
Among them, Damodaran lists that (1) financial statements are likely to go
back fewer years and have less detail and more holes in them and (2) debt and
equity have no assigned market value.21
Cash flow models are the valuation method of choice for both private
equity and public equity at the Dutch pension fund PGGM Investments.
Jaap van Dam, the fund’s principal director of investment strategy, said, “The
two [public and private equity] are not so different, except that there is no
liquid market for private equity—which might be an advantage.” As an owner
of private equity, he remarked that access to information is not a problem.
Stanley Feldman, chairman and co-founder of Axiom Valuation
Solutions, which provides valuation services for private businesses, illiquid
securities, and intangible assets, commented on the use of DCF and multiples
for valuing private equity:
When valuation methods used to value public firms are applied to private
firms, several adjustments need to be made. First, consider the cost of capi-
tal. With few exceptions, analysts used some form of the capital asset pricing
model to determine the cost of equity. Typically, a base value is estimated
and a size premium is added to it. The cost of equity should also reflect
a firm-specific risk premium. Although all firms have firm-specific risk,
theory indicates that it may not be priced in the public markets if the mar-
ginal investor is fully diversified. In the case of private equity, the research
indicates that investors in private firms are not fully diversified. This means
that if the cost of capital does not include a premium for firm-specific risk,
the value, all else equal, is overstated.
Feldman continued,
Often, analysts will use an average revenue and/or EBITDA multiple for
public firms to value a private firm. For example, the revenue multiple is
21
See http://people.stern.nyu.edu/adamodar/pdfiles/ovhds/inv2E/PvtFirm.pdf.

determined by the net operating profit after tax (NOPAT) margin, the cost
of capital (typically the firm’s weighted average cost of capital or WACC),
and expected growth. The multiple varies directly with the margin and
growth and inversely with the cost of capital. Even if the private firm being
valued has a margin and growth expectation equivalent to that of the public
firm, the cost of capital for the private firm will always be higher, if for no
other reason than lack of liquidity. Hence, using an unadjusted public-firm
multiple to value a private firm will likely result in overvaluation of the pri-
vate firm.
According to Feldman, the DCF and market multiples should be used
only for mature private equity firms or private equity firms that are already
in the commercial stage of development. His criticism of the use of DCF to
value early-stage firms (firms that have not reached commercial development
or are in the very early stages of commercial development) is as follows:
In these cases, the analyst might use a DCF and apply a very high discount
rate to reflect the firm’s uncertain future, but this simply takes away the
upside potential, and while this may be a low-probability event, the value
achieved may be very large. So, even when multiplied by a low probability,
the resulting value at the valuation date may be substantive.
Feldman believes that a Monte Carlo or option technique is preferable,
though it is not generally used.
Returning to the Grant Thornton (“Private Equity Valuations” 2015)
survey on methods frequently used to value privately held firms, participants
cited more rule-of-thumb methods than simply DCF and multiples. These
methods include the use of the price of recent transactions, such as merger
and acquisition transactions (70.5%), and recent transactions involving assets
that are the same or similar to the asset in question (69.2%). Feldman notes
that if the transactions are private, they reflect a lack of liquidity (the methods
cited above produce pre-liquidity-adjusted values). In addition, using recent
transactions of comparable companies with similar attributes (e.g., industry
group, recent timing, business offerings, and capital structure) is question-
able, because transactions are rarely directly comparable; value might be tied
to metrics other than revenue. However, in some industrial sectors where
normal profitability does not vary much, there might be an industry valuation
benchmark. Examples include price per subscriber in cable television or price
per bed for nursing home operators.
A source of transaction data for North America is GF Data Resources’
searchable database of business transactions in the mid-market ($10 million to
$250 million range). Drawing from a pool of 206 private equity firms, mezza-
nine groups, and other financial sponsors, the firm’s database has information

Equity Valuation
on valuation, leverage, and specific data organized by NAICS subindustry

codes.22 GF Data Resources also provides benchmarks for meeting the Fair
Value Measurement standards as mandated in the United States by Financial
Accounting Standards Board Statement 157.23 The standard calls for the use
of “fair value measurement” in accounting for fiscal years and interim periods
and is endorsed by, among others, CFA Institute.
In another study on methods used in valuing private equity, Gompers,
Kaplan, and Mukharlyamov (2015) surveyed 79 private equity investors with
a total of $750 billion in private equity assets under management as of year-
end 2012. Their findings showed that despite the importance given to DCF
valuation methods in academic finance programs, participants in the survey
rely on internal rate of return (IRR) and multiples of invested capital (MOIC).
Valuation metrics most widely used by participants were, by far, MOIC and
gross IRR; the highest-ranking methods used were IRR (ranked 9.1) and the
earnings multiple approach (ranked 6.1). Net present value and adjusted pres-
ent value were ranked 2.8 and 0.9, respectively. A forecast horizon of five
years was typical (96% of participants), at the end of which, terminal value
was calculated. In discussing their conclusions, Gompers et al. write:
The fact that [private equity firms] do not use DCF techniques is interest-
ing. It may indicate that IRR and MOIC techniques are sufficiently robust
or effective that DCF techniques are not necessary. Alternatively, it may
indicate some practical deficiency with DCF techniques, especially in the
private equity setting where fund structures limit investment horizons and
there is considerable asymmetric information between general and limited
partners. These settings may make managing via IRR-based investment
decisions better. (p. 42)
Damodaran (2005) observes that the DCF model is subject to signifi-
cant valuation errors if the stream of cash flows is subject to optionalities. The
stock price is an option on the value of the firm. This principle, well known
and applied in the model Merton (1974) proposed for valuing a firm’s credit
22
NAICS, the North American Industry Classification System, was developed under the aus-
pices of the Office of Management and Budget. It was developed jointly by the US Economic
Classification Policy Committee, Statistics Canada, and Mexico’s Instituto Nacional
de Estadistica y Geografia and was adopted in 1997 to replace the Standard Industrial
Classification codes. Other classification systems exist: At the international level, Standard
& Poor’s and Morgan Stanley Capital International jointly developed the Global Industry
Classification Standard system widely used by financial practitioners. Other systems include
the International Standard Industrial Classification of All Economic Activities system devel-
oped by the United Nations.
23
The Financial Accounting Standards Board is United States specific; the international stan-
dards are under the International Accounting Standards Board.

risk, allows a better valuation of firms when optionalities exist because of, for
example, flexibility in the corporate strategy, as is typical with venture capital
at seed or in the startup phase. Damodaran writes, “The equity in a firm is a
residual claim, that is, equity holders lay claim to all cash flows left over after
other financial claimholders (debt, preferred stock, etc.) have been satisfied,”
concluding that “equity can thus be viewed as a call option [on the value of]
the firm, where exercising the option requires that the firm be liquidated and
the face value of the debt (which corresponds to the exercise price) paid off”
(p. 57). Application of real option theory is based on this observation (i.e.,
that the stock price is an option on the value of the firm).
Damodaran also observes, however, that applying real option theory to
the valuation of firms is challenging. It requires a number of inputs that are
not easy to estimate. For example, option valuation models typically make
the simplifying assumption that the debt of the firm is a single zero-coupon
bond. However, firms do not have only one zero-coupon bond outstand-
ing. Estimating the volatility of the underlying value adds another layer of
difficulty.
As with valuation in general, the use of more than one valuation meth-
odology is typically recommended, because doing so allows the analyst to use
one method to cross-check another. And as with IPOs, investment banks
and, more generally, equity analysts play an important role in valuing pri-
vately held firms.
Summing up the valuation of private firms—an asset class estimated to
be worth almost $2.5 trillion globally—staff at Investopedia (“Valuing” 2016)
wrote that the process “is full of assumptions, best guess estimates, and indus-
try averages.” In an industry discussion organized by Financier Worldwide
Magazine (“Q&A: Valuations” 2014), Hilco Enterprise Services’ head of
enterprise valuation and corporate finance, Jason Frank, said,
Since all parties use essentially the same and well-known valuation meth-
odologies, inputs become the most important part of the equation. If the
assumptions are accurate and are properly applied within the chosen valu-
ation methods, much of the subjectivity is eliminated and the valuation
becomes more of a science. … Regardless of which method is utilized, it is
very important to scrutinise the underlying assumptions. The quality, rel-
evancy, and accuracy of inputs separate the valuation from a “science” to an
“art.” A formula and its inputs can be manipulated to provide any result that
is desired. A true valuation will not only follow the guideline methodolo-
gies but will be realistic in its findings.
In comparing the valuation process of private equity with that of IPOs,
our interviewees noted some similarities and some differences. Lleo remarked

Equity Valuation
that, as with IPOs but less consistently, the valuation gap observed with pri-
vate equity and leveraged buyouts (LBOs)
may reflect behavioral pitfalls such as overconfidence, excessive optimism,
confirmation bias, and framing. When a bidding war erupts, as was the
case in the 1987 LBO of RJR Nabisco, or more recently when Marriott and
AnBang locked horns over Starwood, we may also observe the effect of the
winner’s curse.
Fordham University’s Yan remarked,
In valuing a privately held firm, fundamental values and operational effi-
ciency are more important than they are in valuing an IPO. On the finan-
cial side of leverage, you need to be careful. If you go by sentiment, it might
be hard to exit a deal. On the other hand, when you talk about the second
step—that is, taking a private firm public—you encounter the same prob-
lems as in the IPO space.
As with IPOs, problems in evaluating private equity include not only the
disclosure of financial information but also the importance of such factors
as supply and demand. Gompers and Lerner (2000) studied 4,000 venture
investments between 1987 and 1995 and found a “strong positive relation
between the valuation of venture capital investments and capital inflows”
(p. 283). They rejected the possibility that changes in valuation of the firms
in their sample was related to the ultimate success of the firms. Specifically,
Gompers and Lerner found that a doubling of inflows into venture funds led
to a 7%–21% increase in the value of private equity transactions. They also
found a marginal impact of a doubling in public market values, which added a
15%–35% increase in the valuation of private equity transactions.
The strong positive relationship Gompers and Lerner found between the
valuation of venture capital investments and capital inflows in their 2000
study was reported by Investopedia journalist Ryan Downie (2016) in “Is the
Private Equity Bubble Still Expanding?” Downie remarks that during the
dot-com bubble of the late 1990s and 2000, global technology-sector valua-
tions reached an average P/E 80%–300% higher than the average for equities
in other sectors; for the 2010–15 period, the average P/E for public technol-
ogy companies was 20—only 10% higher than the market average as a whole.
Commenting on the divergence of valuation between publicly and privately
held technology firms, Downie writes,
The growth in capital earmarked for private equity was not met with a simi-
lar expansion in suitable investment opportunities. This imbalance drove
valuations higher. However, publicly traded technology firms did not expe-
rience valuation expansion of the same magnitude.

The problem of valuations in public versus private markets was also the
subject of an article by Investopedia’s Trevir Nath (2016). Writing at the
beginning of 2016, Nath remarked, “The mind-boggling valuations [of pri-
vate technology companies] over the past five years are more indicative of the
markets than of the true value of the company itself.” According to Nath,
seeing private tech companies valued at 100 times revenues just before going
public is not uncommon, but, he noted, apart from Facebook, almost no com-
pany has achieved a forward revenue multiple higher than 10.
Valuations appear to be high in some other sectors also. For example, in
the midmarket, Pennsylvania-based GF Data Resources found that valua-
tions for 52 completed US middle-market private equity transactions for the
year 2016 averaged 6.9 times trailing 12-month adjusted EBITDA, a record
high in their dataset that goes back to 2003 (see GF Data Resources 2017).
Can the industry do better in valuing private equity?
Frank remarked,
It is extremely difficult to arrive at an accurate valuation of a privately held
company in today’s market. Not only has the recession wreaked havoc on a
business’s operational, financial, and strategic initiatives, but there are many
other factors that contribute to the complexity of valuing the company.
Companies are faced with severe liquidity concerns, unpredictable con-
sumer demand, and challenges to supplier relationships. These factors are
creating significant uncertainty and unpredictability regarding current per-
formance, as well as a lack of visibility for future projections, which makes
accurately valuing a private entity an extremely challenging exercise.
Privately held companies are generally plagued by a lesser quality and quan-
tity of information that can be used in an analysis. Also, a private company’s
capital structure could be more complex, with various classes of equity
and debt securities. Lastly, the final value of a closely held, private busi-
ness may differ from the value calculated using the established methods
of appraisal—the income, market, and cost approaches—because various
types of discounts or premiums to the basic valuation methodology must be
considered. (“Q&A: Valuations” 2014)
Nevertheless, Feldman made several suggestions for improving the valua-
tion process in private equity:
Best practice in valuation should reflect both theoretical developments in
finance and economics as well as peer-reviewed research. Market practice
is important, but it should conform to academic discipline where possible.
Often, this is not the case. There are many examples, but one in particular
relates to using public company multiples to determine perpetuity values in
the DCF. There are a number of things wrong with this, but chief among

Equity Valuation
them is that factors that determine the multiples at the valuation date are
not likely to be the same when the firm reaches its steady state sometime in
the future.
Specific to the use of DCF models, Feldman pointed to the need to test
whether projections of revenue and EBIT are consistent with market expecta-
tions. This testing includes the following:
1. Making sure that changes in working and net fixed capital are at appro-
priate levels: Values that are too high unnecessarily burden the cash flows
and reduce the value of the firm, while values that are too low do the
reverse.
2. Considering depreciation a real expense: The acceleration adjustment
should be added to cash flows so the full impact of the acceleration shows
up in increased value.
3. Considering the perpetuity of the growth rate: This should be no greater
than the expected nominal growth in the overall economy.
Feldman remarked that the same observations apply to the use of multi-
ples, adding that public company multiples cannot be applied directly to value
private firms.
Feldman also suggested the following:
1. The need to measure lack of liquidity: He suggests using a put-option
pricing framework (its strength: the inputs are market metrics; its draw-
back: the analyst needs to determine the derivative’s life).
2. The need to consider different classes of stock—preferred versus
common—to reflect preferences for the latter over the former.
3. The need to make adjustments to financials, primarily the officer’s com-
pensation adjustment.
A brief look at returns on private equity investments may be worthwhile.
Specific to buyout-related private equity returns, Bloomberg and asset man-
ager Hamilton Lane Advisors analyzed private equity returns of 20 firms
valued at $10 billion or more and subject to buyouts in the 2005–07 period.
Their conclusion was reported by Carey and Banerjee (2016): In more than
half the deals, investors would have fared better by placing their money in
an index fund. A study of the performance of private equity buyout funds by
L’Her, Stoyanova, Shaw, Scott, and Lai (2016) found that, using an appro-
priate risk-adjusted benchmark, buyout investment funds had no significant

outperformance compared with a public market equivalent on a dollar-

weighted basis. The authors did not, however, rule out a valuable role for buy-
out funds.
A decline in the industry’s investment returns in the United States was
documented by economists Eileen Appelbaum and Rosemary Batt (2016) at
the Center for Economic and Policy Research. They reported that although
details might vary, academic studies agree that private equity’s performance
in recent years has been about even with that of the overall stock market. This
performance is in contrast to the 1990–2009 period, when the median private
equity fund outperformed the S&P 500.
Nevertheless, in the low-interest-rate environment that has prevailed since
the 2008 market crash, institutional investors, including pension funds—
many of whose asset/liability gaps have been widening—have been increas-
ing their allocations to private equity. Stephen Nesbitt (2016) of Cliffwater,
a California-based provider of alternative advisory services to institutional
investors, estimated that in fiscal 2014, private equity firms held $277 billion
of US public pension fund assets—9% of the total. The result has been a sig-
nificant rise in capital inflows into the asset class. According to the McKinsey
Private Equity and Principal Investors Practice (2017), in 2009—just one
year after the 2008 market crash—only $185 billion was newly invested in
private equity globally; in 2016, the figure was $716 billion, bringing total
investments in private equity to $2.46 trillion.24
Given the inflows into private equity, regulators and investors are increas-
ingly focusing on valuation issues within the industry. In an attempt to pro-
tect investors, the Dodd–Frank Act in the United States and the Alternative
Investment Fund Managers Directive in the European Union are looking at
how asset managers’ valuation policies are applied to calculate a fund’s net
asset value. Consistency in valuation polices is a major concern. Valuation
guidelines have also been issued by the group International Private Equity
and Venture Capital Valuation in an attempt to “self-regulate” the industry.
As for private equity investors, according to the endowment and foun-
dation polls conducted by the Boston-based investment consultancy NEPC,
valuation emerges regularly as investors’ top concern.25 Consider, for example,
the valuation of the payment-processing specialist Square Inc. Square Inc.
was valued in the private market at $6 billion in October 2014 and offered as
24
In the McKinsey report, total investments in private equity at the end of 2016 were divided
as follows: $1,474 billion in buyouts, $524 billion in venture capital, $315 billion in growth
companies, and $151 billion in other.
25
See NEPC’s Q3 2015 and Q3 2016 survey reports and its press release about the Q3 2016 sur-
vey results (http://www.nepc.com/insights/nepcs-q3-endowments-foundations-poll-results).

Equity Valuation
an IPO on the New York Stock Exchange in November 2015 with an initial
valuation of $2.9 billion—less than half its private valuation of 13 months
earlier.26 (We note that in October 2017, Square’s market capitalization was
$12.6 billion.) In general, according to Downie (2016), the estimate is that
more than 40% of the billion-dollar technology IPOs issued between 2011
and 2015 were trading at or below their last private-round valuations as of
May 2016, despite the rise in public equity indexes over much of that period.
Ernst & Young’s 2015 private equity survey found that investors are
asking for more detailed information about the key assumptions and inputs
driving valuations. According to the survey, 69% of the participating private
equity investors said transparency in financial reporting is a major concern,
and 86% believe that the involvement of third-party valuation specialists adds
a level of consistency to the valuation process. The following year’s survey of
private equity funds and investors (Ernst & Young 2016) found that for 45%
of the participating investors, reporting is now the most important require-
ment when selecting a private equity firm; that percentage is up from 11% just
one year earlier.
With investors, consultants, academics, and regulators noticing a decline
in private equity returns and calling for more transparency in valuations,
Feldman concluded,
When the market for private firms is ebullient, like now, there is less time
spent on ensuring that valuation methods applied are appropriate or gener-
ally consistent with best practice. Alternatively, the greater the oversight,
the more likely that valuation methods applied will meet best practice
guidelines. Although purchase price accounting is typically done post trans-
action, the analysis and methods used can inform the value of a private firm.
For example, the value of the firm should line up with the value of its asset
base. Purchase price allocation tests the veracity of a firm’s value, and this is
especially true when valuing a private firm. I would add that the forecast
trajectory of a firm’s cash flows should be subject to far greater uncertainty
and that simulation techniques should be more commonly used.
See Leena Rao and Dan Primack, “Square Prices IPO at Just $9 per Share, Valued at
26
$2.9 Billion,” Fortune (19 November 2015): http://fortune.com/2015/11/18/square-prices-ipo/.

4. Fair Value, Market Value, and Price
Distortions
Central bank policies and corporate buybacks are the two major phenomena
about which equity analysts and investors now worry because of their potential
to distort market valuations relative to theoretical valuations. (Central bank
policies include low interest rates and quantitative easing.) Together, these
two factors are helping sustain a market rally in the United States where, for
example, the S&P 500 Index has more than tripled since its March 2009 low.
In Europe and Asia, where central bank policies are similar but the buyback
phenomenon is (somewhat) less present, the overall picture is not dissimilar:
The S&P Europe 350 Index is at about 2.5 times its 2009 low, and the Nikkei
250 Index is 2.7 times its 2009 low.
Central Bank Policies and the Market Rally

In an opinion piece published 26 October 2016, Martin Feldstein, profes-
sor of economics at Harvard University and president emeritus of the US
National Bureau of Economic Research, identified the possibility of asset-
price declines as one of the things that “could go wrong in America.” Equity
prices at nearly 70% above their historical average as measured by the cycli-
cally adjusted price-to-earnings ratio (CAPE) for the S&P 500 in October
2016 reflected what Feldstein called “the exceptionally easy monetary policy
that has prevailed for almost a decade.”
At the time of this writing, most central banks continue to pursue a
policy of low, even negative, interest rates. Forbes’s Bryan Rich (2016) men-
tioned Warren Buffett’s evaluation of the impact of low interest rates on
equity valuations when addressing Fortune’s Most Powerful Women Summit
in October 2105. According to Rich, Buffett said that when rates are zero and
expected to remain there forever, stocks would sell at 100 or 200 times earn-
ings, because investors would have nowhere else to earn a return.27 A decade
ago, the 10-year US T-bond yield was about 4.8%; in mid-2017, after the US
Federal Reserve started to raise interest rates, it was about 2.3%—close to
zero or possibly negative after adjustment for inflation.
27
Whether Buffett was serious is difficult to determine. A more conventional analysis would
say that with a 3% equity risk premium over a riskless asset yielding zero, the stock market
should sell at 33 times earnings.

Equity Valuation
Since the 2008 financial crisis, central banks have added liquidity to mar-
kets: $4.5 trillion in bond buying by the US Federal Reserve and another
€1.7 trillion by the European Central Bank. In Japan, in addition to making
aggressive bond purchases, the central bank has embarked on a $58-billion-
a-year stock purchase program, making the Bank of Japan (BOJ) one of the
country’s biggest stock market investors. Kitanaka and Hasegawa (2016) esti-
mated that by the end of 2017, the BOJ could become the largest shareholder
of 55 companies in the country’s Nikkei 225.
Commenting on the problem the BOJ’s moves present for other investors,
Kitanaka and Hasegawa cite a Goldman Sachs report:
While the exact amount of a company’s freely-traded shares is often dif-
ficult to pin down, Goldman Sachs Group Inc. estimated in an August 10
[2016] report that BOJ purchases could soak up the remaining free float at
firms including Comsys Holdings Corp., Nissan Chemical Industries Ltd,
and Tokyo Electron Ltd. Over the next year, if the free float in some stocks
keeps shrinking, it could become more difficult for fund managers to find
the shares they need to track benchmark indexes.
Note that the difficulty in tracking is a problem only if the index the
manager is trying to track is not free float adjusted, but most are.
Kenneth Little, managing director of the investments group at value
manager Brandes Investment Partners, said,
We do consider many market valuations to be fairly high currently. While
we are much more focused on individual companies as opposed to trying to
forecast (or explain) the valuations of markets in aggregate, we do believe
that record low interest rates have been a significant factor in pushing up
valuation levels. The lack of alternatives for individuals and institutions
seeking any sort of yield has forced investors out on the risk spectrum in
order to attempt to meet their return requirements. Equities have been a
primary beneficiary of this quest for yield and return, and this demand has
been a key factor in driving up equity market valuations.
Matteo Bonaventura, a buy-side analyst at Banor SIM, agrees that quan-
titative easing (QE) and loose monetary policy have played a key role in driv-
ing equity valuations up. He noted, “Central banks have flooded the market
with money. For this reason, investors needed to allocate money in the mar-
ket. Given the artificial low returns of government and other bonds, equities
became the target of market investors.”
Jason Hsu, co-founder and vice chairman at Research Affiliates, agrees.
He told IPE Magazine journalist Charlotte Moore (2016), “This wall of
money created a spike in asset prices,” with some segments of the markets
becoming more and more expensive, thereby fueling a momentum market.

4. Fair Value, Market Value, and Price Distortions
Alfred Slager, a professor of pension fund management at TIAS School

for Business and Society at Tilburg University and a trustee at the Dutch
pension fund for general practitioners SPH, gave the question quite a differ-
ent twist:
My analysis might have a slightly different starting point—and a longer
horizon, a bit influenced by the work of, for example, Ulrich Beck’s book
Risk Society.28 I think that the central banks have done impressive work
in uncertain times. But the public impression is slightly different. Central
banks have failed to boost the economy, while governments are struggling
to hold the social fabric together of what constitutes a society or economy.
These institutions are gradually being considered as low-trust organizations,

while corporations are gradually being considered high-trust organizations.
They have been more or less consistent in their strategy compared to gov-
ernments, have their finances in order compared to governments, and are
solving problems that governments cannot; consider environmental, social,
and governance challenges. If this continues, it is a fundamental paradigm
shift and dominates questions like valuations. So, it’s perfectly possible to
have today’s high valuations continuing as long as the governments and
public institutions do not regain that trust.
Actually, analyzing how the intervention of the central banks affects asset
prices and prices more generally is challenging. The key question is why, with
all the money apparently injected into the economy, little to no inflation has
been seen in prices for consumer goods and services. If the growth of the
aggregate measures of the money stock (as measured by M2)29 is compared
with GNP (gross national product) growth in the United States, the finding
is that for the 1957–2015 period, M2 has grown more rapidly than GNP, as
can be seen in Figure 4.1. In fact, the difference in the rates of growth of M2
and GNP accelerated as of the 1970s.
According to mainstream monetary theory, the growth of the money
stock in excess of GNP growth should have some effect on the prices of goods
and services. But in practice, inflation, as measured by the US Consumer
28
In his book Risk Society: Toward a New Modernity, Ulrich Beck characterized Western
societies as risk societies in which the environmental impact of production and distribu-
tion becomes increasingly central to social organization and conflict. Risk Society was first
published in German in 1986; the first English edition was printed in 1992 by SAGE
Publications, London.
29
The US Federal Reserve defines M2 as that measure of the stock of money that consists
of a broad set of financial assets held principally by households and includes bank and sav-
ing deposits, small-denomination time deposits, and balances in retail money-market mutual
funds.

Equity Valuation
Figure 4.1. Growth of M2 Money Stock Compared with GNP in the United States,
1957–2015
1 Jan 1957 = 0
20,000
18,000
16,000
14,000
12,000 GNP
10,000
8,000 M2
6,000
4,000
2,000
0
Jan/57 Jul/68 Jan/80 Jul/91 Jan/03
Source: Constructed by the authors using data obtained from the Federal Reserve Bank of St. Louis.
Price Index (CPI), has remained very low some seven years after the intro-
duction of QE.
Where did the money go?
Richard Werner, director of the Centre for Banking, Finance and
Sustainable Development at the University of Southampton School of
Management, proposed a possible explanation. Werner (2012) argues that
the newly created money did not reach the economy uniformly but that a
growing fraction of the money reached financial markets, contributing to
asset inflation without contributing to GDP or GNP growth or to inflation
as measured by the CPI. He thus suggested the need to divide money into
two streams:
•• money used for GDP transactions—that is, used for the “real economy”
or real circulation; “real money” (MR) and
•• money used for non-GDP transactions—that is, “financial circulation”;
financial money (MF).
Therefore, M = MR + MF.
Note, in particular, that QE, as it is being implemented, is by nature
money that reaches primarily financial markets (MF) because QE consists of

central banks buying assets from nonbank institutions. The main effect is to
stimulate asset demand and therefore a rise in asset prices.
Corporate Buybacks and the Market Rally

The other important factor that market participants say is distorting
prices is the dearth of shares now in public markets—in particular, in the
United States.
A first explanation for the dearth of shares in the United States is due to
what a study by National Bureau of Economic Research researchers (Doidge,
Karolyi, and Stulz 2015) called an “abnormally” low number of listed firms in
the United States today compared with the past and with other countries. The
study found that the number of listed firms has declined from a high of 8,025
in 1996 to 4,101 in 2012 (the number of US-listed firms was around 4,300 at
the end of 2016). According to the study’s authors, delistings accounted for
46% of the listing gap; the remaining 54% is explained by the low rate of new
listings. In contrast to what was happening in the United States, non-US list-
ings increased from 30,734 to 39,427 during the same period.
A second explanation for the dearth of (US) stocks is share buybacks.
Although the figure is now down slightly from recent highs, companies in
the S&P 500 spent $536 billion on buybacks in 2016 (S&P 2017). Birinyi
Associates estimates that US-listed firms spent about $6.1 trillion buying back
their own shares during the 11-year period 2005–2016, though a drop in both
authorized and executed buybacks was noted in 2016 compared with 2015.30
The downward trend in authorizations continued into the first quarter of 2017.
Viewed from a different angle, Federal Reserve data on fund flows com-
piled by Goldman Sachs Asset Management and reported by Bloomberg’s
Lu Wang (2016b) show that US companies were the biggest buyers of stocks
every year from 2009 to 2016. Indeed, in a recent Dr. Ed’s Blog post, Edward
Yardeni (2017) remarked that since the start of the bull market during the
first quarter of 2009 through the end of 2016, buybacks totaled $3.4 trillion;
dividends added up to $2.4 trillion. “Combined,” he noted, “they pumped
$5.7 trillion into the bull market, driving stock prices higher without much,
if any, help from households, mutual funds, institutional investors, or foreign
investors. … The bull has been on steroids from share buybacks by corporate
managers.”
Share buybacks are frequent in the United States but also take place else-
where. Using Bloomberg data, journalist Sofia Horta E. Costa (2016) reported
We thank Birinyi’s Chris Costelleo for providing us with this information in an Excel
30
spreadsheet.

Equity Valuation
that 5 million shares were bought back by STOXX Europe 600 firms in
2015 (this number compares with more than 60 million shares bought back
by S&P 500 companies in the same period). In Japan, the phenomenon has
been growing in importance since 2014, which witnessed a spike in buybacks.
Indeed, in 2016 corporate buybacks were the biggest source of equity demand
in Japan, according to Leo Lewis (2016). Lewis cited Goldman Sachs esti-
mates that put buybacks for the fiscal year ending 31 March 2017 at a record
¥6.5 trillion (approximately $60 billion). Nevertheless, buybacks in Japan,
compared with those in the United States, still represent a small proportion
of market capitalization.
An example of the effect of corporate buybacks on stock prices comes
from the Nikkei Asian Review (“Fewer Japanese Companies” 2017). In
January 2017, when NTT DOCOMO reported nine-month results without
renewing its share repurchasing program, its stock price fell by 4% within a
few days. In contrast, companies that announce share buybacks see the price
of their shares increase. When Asahi Glass Company announced buybacks
of up to ¥10 billion several months later, the company’s shares went up some
9% to a six-year high. After Aoyama Trading Company announced its sixth
straight year of buybacks, shares went up 3%.
Indeed, some equity analysts consider corporate buybacks the sole factor
driving demand for equities in today’s market. In March 2016, Bloomberg
TV reported that if the pace of withdrawals from US mutual funds and
exchange-traded funds were to continue through the month, outflows would
hit $60 billion (Wang 2016a). The result would be the biggest annual gap
between outflows and corporate buybacks ($225 billion) since the “mini”
stock market crash of 27 October 1997. At the Amsterdam-based Robeco
Institutional Asset Management (2016), buybacks are considered largely
responsible for the growth in market multiples seen in recent years.
Buybacks, which in theory are a way of sharing profits with shareholders,
increase demand for and reduce the supply of a company’s shares. Given the
law of supply and demand, buybacks tend to distort valuations in the immedi-
ate term by raising earnings per share (EPS), even when total net income is
flat. Interestingly, for fear of market manipulation, companies were largely
prohibited from buying their own shares until 1982, when the Ronald Reagan
administration began to deregulate financial markets.
In a special Reuters report on buybacks, Brettell, Gaffen, and Rohde
(2015a) cited the effect of the health insurer Humana’s $500 million share
repurchase in November 2014, which allowed the firm to surpass its $7.50
EPS target by a penny. Commenting on such buybacks, Heitor Almeida,
professor of finance at the College of Business at the University of Illinois

in Urbana-Champaign, cautioned against using EPS as a performance met-

ric. “It’s too easy to manipulate,” he told Reuters reporters (Brettell, Gaffen,
and Rohde 2015b). James Montier, a member of the asset allocation team at
GMO in London, remarked that the concentration on buybacks has as its
objective raising EPS, which he attributed to the objective of hitting share
option targets (Brettell et al. 2015b).
As mentioned in Chapter 3, Straehl and Ibbotson (2015) concluded that
the shift in corporate payout policy from dividends to buybacks has caused
a “secular decrease in dividend yields, and an analogous increase in per-
share growth,” leading to a “structural break in the return components of
the traditional supply models such as the dividend discount model” (p. 25).
Chapter 3 provides the authors’ proposed supply model of stock returns,
which they dubbed the “total (dividends plus buybacks) payout model.”
Most value managers focus on individual companies, rather than try-
ing to forecast (or explain) the valuations of markets in aggregate, but at the
level of asset allocation, investors and their asset managers have an interest in
understanding the relative valuation of the investable universe. Writing back
in 1996, four years before the dot-com bubble burst, University of California,
Berkeley, economist J. Bradford DeLong remarked,
The stakes for investors are truly enormous. If the $7 trillion US stock mar-
ket is overvalued by a third, some $2 trillion plus of the wealth Americans
now hold in stocks will vanish over the next decade as stock prices return
to fundamentals. The losers will be those who remain fully invested in the
market over the next decade. If stocks are not overvalued today, the losers
will be those who—out of fear of possible overvaluation—spend the next
decade out of the stock market, with their wealth invested in lower-return
investments in bonds and in the money market.
DeLong added,
One standard measure of “fundamentals” is average earnings over the past
10 years—an average taken over a time period long enough to smooth out
business cycle fluctuations in profitability. In a typical year, a typical stock
is priced at about fifteen times its 10-year average of earnings. Today the
typical stock sells for nearly thirty times its 10-year average of earnings. …
The argument that the stock market is overvalued—and that it will come
back to earth over the next decade, perhaps in a gradual deflation of prices
like a slowly-leaking balloon and perhaps in a crash—is simple. Stocks are
tradable pieces of paper that carry “ownership” of the earnings of American
corporations. Stock price “fundamentals” are thus roughly proportional to
the earnings of American corporations. But today the stock market is sell-
ing for roughly twice its typical earnings multiple.

Equity Valuation
Are today’s stock markets—in the United States, the S&P 500 is now
roughly 70% above the historical earnings multiple—overvalued? And if so,
is the overvaluation the result (at least in part) of central bank policies and
buybacks?
PGGM Investments’ head of equities Felix Lanters said, “While I don’t
like to use the term ‘overvalued,’ there is a gap between market values and
valuation in the models. We are having more and more trouble finding com-
panies that are undervalued; their numbers are fading away.”
Bradford Cornell, a professor of financial economics at the California
Institute of Technology, agrees that prices are high but commented, “I do
note that they have been this high or higher in the past when the factors to
which you refer [central banks’ policies and stock buybacks] were not present.”
Niels C. Jensen, partner and chief investment officer at the
United Kingdom–based Absolute Return Partners, identified two additional
factors to explain market values—in particular, US market values. To low
rates and corporate buybacks, Jensen (2015) added (1) the all-time high of
capital, which is now 42%–43% of US national income, against a historical
average of 35% and (2) demographics—the biggest equity buyers are middle-
aged, and the US great equity bull market has coincided with Baby Boomers’
middle years. Jensen contrasted the situation in the United States with that in
Japan (see Table 4.1).
Christian Kjaer, head of global equities and volatility at the Danish fund
ATP, believes that the supply–demand balance is at work. He remarked,
“Supply and demand has caused ‘risk-free’ yields as well as risk premiums in
Table 4.1. Comparison of the Ratio of Total Market Capitalization to Gross Domestic

Product for Four Major Economies and Aggregate OECD Countries
1975–1985 Peak,
Region/Country % Range 1999–2000 2015
Germany 11–24% 65% 51%

Japan 3–36% 98% 112%
United Kingdom 24–72% 179% 120% a
United States 37–53% 153% 139%
OECD members 24–49% 122% 102%
a
The data for this ratio are not available from the World Bank but come from Siblis Research
(http://siblisresearch.com/data/market-cap-to-gdp-ratios/).
Note: OECD = Organisation for Economic Co-Operation and Development.
Source: Constructed by the authors using data obtained from the World Bank (http://data.world-
bank.org/indicator/CM.MKT.LCAP.GD.ZS?locations=OE-DE-JP-US-GB).

general to compress. As a consequence, valuations are high not only in equi-

ties but across asset classes.”
Sébastien Lleo, associate professor of finance at NEOMA Business
School (France), elaborated on the interconnected factors at play in today’s
high valuations and asked whether they are justified by long-term trends:
To start with, interest rates are indeed exceptionally low. This reflects the
extraordinary measures that the world’s leading central banks implemented
in the wake of the 2007–2009 global financial crisis, but also the low infla-
tion environment that has prevailed since the early 2000s.
He continued,
The direct impact on valuation is that discount rates will be low, which
justifies higher present values of future cash flows. At the same time, low
interest rates have made money market and fixed income much less attrac-
tive, pushing cash into riskier assets and, in particular, stocks. This excess
demand, at a time when the net offer of shares is muted or negative, has
compressed the equity risk premium, leading again to lower discount rates
and higher valuations.
Behavioral factors, in particular the spectacular recovery of the financial

markets following the global financial crisis, also play a role in shaping our
expectations about the future, both in terms of cash flows and discount
rates. If valuations are high, the key question we need to ask is whether
these valuations are justified by a long-term trend in both cash flows and
discount rates or represent an excessively optimistic and psychologically
biased view of future possibilities.
A different view comes from James Montier and Philip Pilkington (2016)
of GMO’s asset allocation team in London. In an attempt to understand why
their forecasts of S&P 500 returns had been too pessimistic over the past two
decades, they found that the market’s high P/E was not due to low interest rates.
They asked what, then, might account for the high valuations. Studying data
for the 1964–2016 period, they found that since 1984, “significant amounts of
annual stock market returns … were made on FOMC [Federal Open Market
Committee] meeting days” (p. 1).31 In addition, they found that “these pre-
FOMC returns have increased over time and account for sizeable fractions of
total annual realized stock returns” (p. 2). Note that their research resulted in
the development of what they call the Monetary-Policy-Adjusted CAPE.
In the United States, the Federal Reserve controls monetary policy using three tools: open
31
market operations, setting the discount rate, and setting reserve requirements. The first of
these, open market operations, is the responsibility of the FOMC.

Equity Valuation
In a recent quarterly letter, Jeremy Grantham (2017), their colleague in

the United States and the co-founder of GMO, observed that the P/E of
S&P 500 firms is now 65%–70% higher than during the 1935–1995 period.
Grantham also noted that the margins of S&P 500 firms have risen by about
30% over their pre-1997 margins. He then analyzed the sources of the increase
in corporate profits and listed among them (1) the increased value of brands
due to globalization; (2) increased corporate power over the past 40 years;
(3) increased corporate wealth, which has been used to influence policy; (4) a
decrease in capital spending as a percentage of GDP; (5) increased monopoly
power for US corporations; and (6) lower interest rates since 1997, together
with higher leverage. Grantham argues that these changes, as well as low
interest rates, are here to stay for a long period. He warns investors,
If you are expecting a quick or explosive market decline in the S&P 500
that will return us to pre-1997 ratios (perhaps because that is the kind of
thing that happened in the past), then you should at least be prepared to be
frustrated for some considerable further time, until you can feel the process
of the real interest rate structure moving back up toward its old level. (p. 15)
Note that higher profitability per se does not explain a higher P/E.
Higher margins and a shrinking number of new investments coupled with an
increased money supply, however, have attracted an excess flow of money into
the same stocks. This excess flow produces an increase in the price per unit of
profit—that is, a higher P/E.
How Do We Evaluate Whether Markets in Aggregate Are Correctly
Valued in Relation to Our Theoretical Values? Several market measures
based on historical comparisons are typically used to judge over- or undervalu-
ation. Among the most cited methods is the CAPE (or P/E 10), which uses
10 years of earnings data instead of a single year and adjusts the historical earn-
ings for inflation. The intuition behind this measure is that the return to financial
assets such as stocks cannot stay disconnected very long from earnings growth.
In mid-2017, the S&P 500 CAPE was around 29; the historical mean
(i.e., since the 1870s) is about 16.7. In the same period, the FTSE 100 CAPE
was at 15, compared to its historical mean of 16, and the FTSE 250 at 25,
compared to its historical mean of about 20. Also in mid-2017, the Nikkei
225, at 28, was not far from its historical mean of 26.
Another frequently used measure based on P/E is the forward price-to-
earnings ratio (price divided by next year’s forecast earnings).32 According to
The forward P/E is calculated by taking into consideration analysts’ future projections for
32
S&P 500 earnings and the current price of the index.

FactSet analyst John Butters (2017), in mid-2017 the S&P 500’s 12-month
forward price-to-earnings ratio was 17.6, its highest level in 13 years.
In addition to forward-looking P/E and backward-looking measures
(the trailing 12-month P/E is widely used in the industry on single assets),
other measures frequently used to understand whether markets, overall, are
cheap or dear include the historical growth of capital market earnings and
the growth of per capita GDP, or a country’s stock market capitalization as a
percentage of its GDP.
Straehl and Ibbotson (2015) essentially confirmed earlier findings by
Ibbotson and Chen (2003) that US long-run stock returns “participate” in the
real economy. Ibbotson and Chen decomposed 1926–2000 historical equity
returns using factors commonly thought to describe the aggregate equity
market and overall economic productivity (i.e., inflation, EPS, dividends per
share, price per earnings, the dividend payout ratio, book value per share,
return on equity, and per capita GDP) and found that for the 1926–2000
period, the majority of historical returns can be attributed to the supply of
these components. Straehl and Ibbotson (2015) extended the period to 1871–
2014—a 143-year period—and found that total payout (dividends and buy-
backs) per share and per capita GDP grew approximately at the same rate,
albeit with large fluctuations from time to time (p. 20).
Cornell (2010) explored the link between equity returns and economic
growth in a study that took into account both theoretical models and empiri-
cal results from growth theory. He postulates that
unless corporate profits rise as a percentage of GDP, which cannot continue
indefinitely, earnings growth is constrained by GDP growth. This dynamic
means that the same factors that determine the rate of economic growth
also place bounds on earnings growth and, thereby, the performance of
equity investments. (p. 54; italics added)
Cornell concluded that the long-run performance of equity investments
is indeed fundamentally linked to growth in earnings, which in turn depends
on growth in real GDP.
The positive relationship between the growth of total payout per share
and GDP growth found by Straehl and Ibbotson (2015) and others contrasts
with the findings of Jay R. Ritter at the University of Florida’s Warrington
College of Business. In his cross-country study using data from the World
Development Indicators for the 1900–2002 period, Ritter (2005) found a
negative relationship between real stock returns and per capita GDP.
Following these and other studies, researchers at Barra/MSCI (2010)
asked whether investors, assuming a relationship (or not) between economic

Equity Valuation
growth and stock returns, should assign a higher weight to countries experi-
encing strong GDP growth. They note,
This question is not new; “supply-side” models have been developed to
explain and forecast market returns based on macroeconomic performance.
These models are based on the theory that equity returns have their roots in
the productivity of the underlying real economy and long-term returns can-
not exceed or fall short of the growth rate of the underlying economy. (p. 1)
Using long-term Morgan Stanley Capital International equity index data
(i.e., the MSCI All Country World Index and the MSCI World Index, both
of which pertain only to large-cap and mid-cap companies) and the GDP
growth of countries included in the same indexes, the researchers empirically
tested the link between economic growth and subsequent stock returns. They
observed, “Long-term trends in the link between real GDP and equity prices
are more similar for global equities than for most individual [country] mar-
kets” (Barra/MSCI 2010, p. 5). In other words, the link is stronger between
variations in global GDP and the return of the global index across time than
it is for variations from one country to another. They offered several explana-
tions for this finding: (1) given economic integration, the link appears over
global, not national, markets; (2) a large part of economic growth comes from
new, not existing, companies, which dilutes GDP growth before it reaches
shareholders; and (3) expected economic growth might already be factored
into share prices, thereby reducing future realized returns.
O’Neill, Stupnytska, and Wrisdale (2011) at Goldman Sachs Asset
Management found significant methodological issues in previous studies that
failed to establish a link between GDP and returns (e.g., problems with data,
too long a time horizon, during which many changes occur) and posed a ques-
tion: “Why, in a world of forward-looking investors, [would we] expect to find
a contemporaneous relationship between growth and returns? In fact, there
exists extensive evidence that equity price changes tend to lead GDP growth in
a number of countries” (p. 3). They believe that an existing (though not straight-
forward) link between GDP and equity returns places a renewed emphasis on
valuation. Indeed, establishing such a link lends support to present value models.
Finally, London Business School’s Dimson, Marsh, and Staunton (2014)
revisited this question: If a link between per capita GDP and equity returns
exists, why has making money by buying stocks of countries that are improv-
ing their economic position been so difficult? For example, looking at 21 coun-
tries for the 1972–2013 period and assuming that investors put their money
in the equity markets of the fastest-growing countries, the approach would
have delivered an annual return of 14.5%; had the investors put their money

in the slowest growing countries, they would have realized an annual return of
24.6%. The authors concluded, however, that although capturing returns is not
easy, stronger (aggregate) GDP growth is “generally good for investors” (p. 29).
Another measure used to understand whether markets in the aggregate
are overvalued is the ratio between a country’s total market capitalization
(TMC) and its GDP (the TMC-to-GDP ratio). This ratio is often referred to
as the Buffett ratio because following the dot-com bubble of the late 1990s,
Warren Buffett embraced the ratio of the Wilshire 5000 Full Cap Price Index
to US GDP as “probably the best single measure of where valuations stand at
any given moment” (Loomis 2001). As of mid-2017, this ratio for US markets
stands at just above 130%. According to data from the World Bank, during
the 1975–85 period, the average for the United States was under 50%, and at
the end of the dot-com bubble in early 2000, it had reached a peak of 153%.
Among developed countries, the TMC-to-GDP ratio has historically been
highest in the United Kingdom and the United States. Table 4.1 provides a
comparison of the TMC-to-GDP ratio for major economies and aggregate
OECD member states for the 40-year period 1975–2015.
In referring to lessons to draw from the TMC-to-GDP ratio following
the dot-com stock crash, Buffett told Fortune journalist Carol Loomis (2001),
“For me, the message … is this: If the percentage relationship falls to the
70% or 80% area, buying stocks is likely to work very well for you. If the ratio
approaches 200%—as it did in 1999 and a part of 2000—you are playing
with fire.” The long-term average for the United States is 79%.
The intuition behind Buffett’s advice is that the stock of capital and
the actual economic output should move together. Actually, the question is
complex. Economies are not homogeneous. An economy’s growth rate is a
simplification that does not take into account the inequalities inside the econ-
omy. Different sectors in the same economy might grow at different rates for
prolonged periods. But as with the relationship between the growth of total
returns and the growth of per capita GDP, many value managers do not use
the TMC-to-GDP ratio.
Edward Yardeni of Yardeni Research has criticized the ratio for several
reasons, among them the fact that the TMC-to-GDP ratio does not take into
account structural changes in profit margins caused by, for example, changing
tax rates, lower interest rates, or technological innovation (Siblis Research,
n.d.). It also does not take into account institutional differences between
countries; for example, a great deal of the equity in German companies is
family owned and thus not listed on exchanges. But again, the critical fact is
that economies are complex systems with complex output. To average growth
over a whole market and a whole economy is basically impossible.

5. New Tools for Equity Valuation
“It’s now an era in which analysts capable of analyzing will need to do so thor-
oughly.” These are the words of Shinichi Tamura, a Tokyo-based strategist at
Matsui Securities Company, as reported by Allan and Ito (2016). But what
is meant by “thoroughly” in an era of big data and high-performance com-
puting, advanced analytics, predictive reasoning, machine learning/artificial
intelligence, natural-language processing, fast algorithms, and automation?
During the past decade or two, many business processes in the finan-
cial services sector have been radically changed or fully automated. Consider
algorithmic (or automated) trading. In the European Union and the United
States, automated trades are estimated to constitute 50%–80% of all equity
trades, up from about 33% in 2006. As for investment management, some
automation has already taken place. Anthony Ledford, chief scientist at the
London-based quantitative investment manager Man AHL, remarked,
The balance sheets of publicly traded companies are reported on a time-
table that is known well in advance, and they largely comprise structured
summary information in a standardized format. This makes it relatively
straightforward to monitor the balance sheet changes of any particular
company through time. Modern production systems for data capture, stor-
age, and retrieval can easily scale this process across the tens of thousands
of publicly traded companies found in global portfolios. This is not machine
learning, nor is it big data—it’s simply automation.
Much progress has been made recently in data management thanks to
the structuring or standardization of data brought about by adoption of the
eXtensible Business Reporting Language (XBRL) required by regulators
for financial reporting purposes. XBRL uses a taxonomy or list of fields that
allows one to “tag” data. Tagged data are generally referred to as metadata.
The objective of XBRL is to deliver financial data from companies directly
in a computer-readable format, thereby making financial statements and data
easy to search and comparable over time and among companies. In their
study on data and technology in the financial services industry, Singh and
Peters (2016) note among the benefits brought about by the implementation
of XBRL the availability of more (and more granular) data. They cite more
than 51 million discrete facts tagged with XBRL in the US SEC’s EDGAR
database of more than 89,000 filings by some 10,000 firms.33
EDGAR stands for Electronic Data Gathering, Analysis, and Retrieval system.
33

An example of what can be done with metadata combined with artificial

intelligence (AI) comes from the software firm Yseop. Its natural-language-
generation software relies on metadata to generate business descriptions auto-
matically. The financial data company FactSet runs Yseop software across its
15 highly connected and structured databases to generate hundreds of thou-
sands of company descriptions daily. Other firms offering software to monitor
data sources in real time, detect signals, and generate analyst reports include
the Cambridge, Massachusetts–based AI firm Kensho and the content-
summarization company Agolo.
The expectation among active managers is that the vast amounts of data,
including unstructured data now available on the web, can be put to use
to outperform markets. Sébastien Lleo, a professor of finance at NEOMA
Business School (France), commented, “One of the important objectives of
data analysis and statistics is to be able to transform raw data into informa-
tion, information into knowledge, and knowledge into insights that can be
used for decision making. Big data fits in this general pattern.”
Among the data that active managers are anxious to use are real-time
macro and micro data, such as sales and price data from points of sale, news
streams, and social network data. Commenting on these and similar data
sources, Gurvinder Brar, global head of quantitative research at Macquarie
Equities Research, said, “Depending upon the sophistication of the asset
manager, I would say that these data points are already being used, and their
importance will grow over the coming years.”
Peter Hafez, the chief data scientist with big-data analytics supplier
RavenPack, agrees. He remarked,
Over time, the use of alternative data sources will span the entire indus-
try, though not all investors will adopt the new type of data. It is already
being used by the most sophisticated quantitative hedge funds and asset
managers, a process begun around the early 2000s. In recent years, inter-
est for alternative data has exploded, and in the coming years, we will see
more and more traditional asset managers make significant investments to
become data ready. This will require an investment not only in IT [informa-
tion technology] but also in data science and quantitative research. Some
firms will build up internal capabilities in alternative data; others will look
to buy ready-made alternative data factors or indicators that can be inte-
grated directly into their investment process.
Axel Pierron, co-founder and co-managing director of the financial con-
sultancy firm Opimas, noted that data sourcing and factor data are now cru-
cial in asset management, saying,

Equity Valuation
What is new today is that with artificial intelligence, firms are trying to
bridge the gap between real-time information and information with longer
periodicity. The early 2000s saw the implementation of strategies and algo-
rithms for trading, which were all similar as all were based on using market
data. With the AI tools now available, firms are looking beyond traditional
market data, which everyone is using, trying to build (trading) strategies
with subcorrelations. As firms use a greater variety of data sources, we have
seen a greater differentiation, a greater diversification of strategies.
These differences, Pierron remarked, “reduce herd behavior and lead to
greater market stability.” Opimas has recently published reports on the use of
AI (Pierron 2017) and alternative data (Marenzi 2017) in asset management.
Much of this new big data is commercially available, however, which will
eventually reduce its value. PanAgora president and chief executive Eric H.
Sorensen commented,
First, “big” alone is not a sufficient condition for added value—“big” but
not “smart” (or causal) is potentially spurious. Second, “big” brings perverse
results if everyone uses it. Smart data may be sufficient. Smart data, among
other things, is not commercially ubiquitous, is given to reasonable invest-
ment horizons, and is rich in fundamental intuition.
Sorensen referenced his recent investment insight note (2017) relating
PanAgora’s first encounter with smart data back in the early 1970s:
A major industry of the Pacific Northwest was forest products—
Weyerhaeuser and Georgia-Pacific, to name a few. One innovative analyst
determined that during periods of rising interest rates (building slump) ver-
sus falling rates (building boom), a small (versus large) lumber inventory
separated the winners from the losers. Consequently, he hired a helicopter
pilot to routinely fly him directly over the lumber yards adjacent to lumber
mills, as well as active logging sites, to assess the potential inventory levels.
The helicopter data was proprietary and intuitively causal. Most importantly,
it worked, providing valuable company insights.
Sorensen continued,
Forty-five years later the contemporary version of helicopter data is satellite
imaging of shoppers’ parked cars. Today’s modern space technology version
is big, and it seems causal. However, it may fail if everyone has access to it.
Ubiquitous is neither proprietary nor innovative, which means it fails to be
“smart.” (p. 3)
To illustrate this fact, Sorensen cited an experiment by his colleagues at
PanAgora on parking lot data for the 2010–13 period. The back test pro-
duced a 30% cumulative long–short return. But after 2013, the return became

negative and remained so thereafter: In 2013, a vendor started selling the

data to subscribers.
Just what data will prove useful and what analytical methods will be
required to help investment professionals improve their ability to predict
the value of a firm’s stock? In “Finding Big Alpha in Big Data,” members
of BlackRock’s Scientific Active Equity Group (Savi, MacCartney, Betts,
and Shen 2015) identify what they consider of interest—namely, all data,
and in particular, data that relate to economic human behavior. Just in the
area of brokerage reports, they cite an average of 4,000 reports daily, totaling
36,000 pages—in 53 languages. They argue that because most of the data of
interest are unstructured, new tools and more powerful computing platforms
will be required. Among the tools they list are machine learning, natural-
language processing, scientific data visualization, and distributed computing.
Another list of new tools for investment professionals was proposed by
the Boston Consulting Group (Sheridan, Beardsley, Ouimet, and Baltassis
2016), whose list—somewhat expanded—includes the following:
machine-learning platforms that can mine huge quantities of structured
and unstructured data; predictive-reasoning and artificial-intelligence plat-
forms that can reveal important portfolio effects; rapid statistical analyses
that generate event studies and correlation analyses on a massive scale;
semantic analyses capable of discerning context and drawing insights from
various data types; visualization tools that create intuitive displays of infor-
mation that would be too diverse and complex to summarize in writing; and
natural-language processing engines and data aggregation platforms that
give managers convenient, timely access to different forms of data.
Armed with these advanced techniques, digitally forward asset managers
can gain a significant information advantage over peers who rely mainly
on traditional data sources and analytical practices. They can crunch
through vast quantities of data; scour video and satellite imagery to gauge
a retailer’s Black Friday prospects; extract insights from social media,
texts, and e-mail to divine market sentiment; and parse a CEO’s com-
ments during an earnings call to estimate the potential impact on the next
quarter’s results. They can discern how unexpected weather disruptions
might affect their portfolio, and even disprove long-held beliefs about how
markets work.
Macquarie’s Brar remarked,
These methodologies and tools are being used by quantitative managers,
but to a lesser extent by fundamental investors. I do, however, see a greater
desire from fundamental analysts to apply these tools within big-data space,
to tease out signals which fundamentally alter their views on a single stock.

Equity Valuation
Lleo agreed:
To some extent, finance is ready for a data-centric evolution. Data—
whether financial statements, economic releases, or market prices and
yields—are already at the heart of finance. Market data, in particular, fit the
“high-volume, high-velocity” bill, and economic data are getting ready for
an upgrade, thanks to a growing interest in nowcasting.34 Since the 1970s,
banks have been constantly upgrading their computing power to run ever
more demanding pricing and risk management algorithms.
Lleo added,
Most of this evolution has taken place in trading rooms, where the emphasis
was placed on objective pricing (as in Black–Scholes-style derivative pric-
ing) rather than subjective valuation (as in DCF [discounted cash flow]-
style equity valuation). The mathematical tools and algorithms are different,
because the objective is different: Pricing is about hedging; valuation is
about predicting. Only recently have banks started to focus on predictions
on a large scale—to predict the default risk of their individual clients. Now,
machine learning–based techniques are being used in investment manage-
ment to design smart-beta strategies, as well as “robo-advisers.” We can well
envision that machine learning will soon be used in the valuation process as
well, to help analysts and portfolio managers process higher velocity data
and higher variety data—including news coverage, CEO interviews, tweets,
supplier and client data, supply chain information, etc.—or to explore the
industries from a network perspective.
Data, Tools, and Their Application

Let’s take a look at some of the methodologies and technologies at the heart
of the change. First, machine learning.
Machine learning (computer programs that learn when exposed to new
data) is a hybrid discipline that borrows from a range of areas, including
computer science, engineering/signal processing, mathematics, and statistics. In
an explanatory video, Man AHL’s Ledford (n.d.) describes machine learning as
the use of algorithms for identifying and acting on repeatable patterns
in observed data. It provides a suite of data-driven tools for quantifying
34
Nowcasting, a contraction of now and forecasting, is the prediction of the present, the near
future, and the very recent past in economics or the creation of accurate forecasts of undis-
closed target values based on publicly observable proxy values. The term has been used for
a long time in meteorology. It has recently become popular in economics, because standard
measures used to assess the state of an economy (e.g., GDP) are determined only after a long
delay and are subject to subsequent revisions. Nowcasting models have been applied in central
banks, and the technique is used routinely to monitor the state of the economy in real time.

knowledge from ensembles of such unstructured and structured data

sources, embedding their combined information within a coherent model
that—importantly—does not require the modeler to make simplistic or
artificial assumptions, such as linearity. The algorithms are not told what to
look for but seek out the patterns for themselves, a difference with respect to
other areas of data mining.
Ledford notes the approximately 1.5 billion data ticks that Man AHL
receives daily in numbers and text and from other sources of information to
underline the greater signaling power of machine learning over the human
brain. The firm has been using machine learning–based systems in its multi-
strategy client portfolios since early 2014.
RavenPack’s Hafez remarked that the greater emphasis on machine
learning and AI is due to the explosion of potential alpha sources or factors
that require new modeling techniques that researchers are combining with
traditional econometric modeling.
The quantitative research teams at Macquarie Equities Research (2015)
apply textual analysis (also called “text mining” or “natural-language pro-
cessing”) to the transcripts of earnings conference calls for global companies.
Their objective is to obtain a quantitative measure of sentiment or tone—that
is, bullish or bearish—and to use that “soft information” to predict future
stock returns. For the analysis, they use FactSet data and FactSet’s API on
the global universe of equities, coded in standard R packages, and apply
the naive Bayes classifier.35 They prefer the naive Bayes classifier because it
is computationally inexpensive compared with other classification methods
such as support vector machines or neural networks. The latter require expen-
sive training of parameters.
Interestingly, their results show that sentiment does predict returns—and
not only in the first few days after a conference call but also over horizons of
one to three months or more. Researchers at Macquarie suggest that among
the uses for the signals gained from the textual analysis of thousands of con-
ference calls, fundamental analysts can use the signals as a screening variable
to identify those corporate voices with a bullish or bearish tone or those that
display a marked difference between the direction of the surprise and tone.
Commenting on Macquarie’s work on text analysis, Brar said, “We’ve
done extensive work on analyzing corporate communications—earnings
R packages are collections of functions, data, and compiled code in a well-defined format.
35
More than 4,500 packages are available. A Bayesian classifier is a statistical technique for
predicting class membership probabilities. It can be used to estimate the probability that
a given data point is a member of a particular class. See https://monkeylearn.com/blog/
practical-explanation-naive-bayes-classifier/.

Equity Valuation
transcripts, conference call transcripts—and find these to be a good source of

alpha for both quantitative and fundamental investors.”
Macquarie’s quantitative equities team also used text-mining techniques
in the following research:
•• The analysis of changes in the complexity of texts in 10-K reports to
predict stock returns. They found that increases in complexity in the
management discussion and analysis section of annual reports are associ-
ated with future stock underperformance (Macquarie Equities Research
2013a).
•• The analysis of 6,000 quarterly conference call transcripts for US compa-
nies. They found negligible return drift following an earnings surprise,
but changes in tone sentiment were positively related to returns, and the
drift lasted up to three months (Macquarie Equities Research 2013b).
•• The analysis of press releases for earnings announcements of Russell
3000 Index companies since 2004. They found that combining the soft
information embedded in a press release with the release of hard informa-
tion during the earnings announcement period can be used to enhance
the performance of an earnings surprise strategy (Macquarie Equities
Research 2014a).
•• The analysis of the front page of single-stock fundamental analyst reports
prepared by Macquarie. They found that analysts convey significant soft
information in their research that is not captured by recommendations,
price targets, or fundamental forecasts and that provides insights into
fundamental changes (Macquarie Equities Research 2014b).
BlackRock also uses textual analysis to capture alpha. Raffaele Savi and
Jeff Shen (2015), co-heads of the firm’s Scientific Active Equity Group, write,
The intense competition in both HFT [high-frequency trading] and long-
term investing has led us to increase our research efforts on developing sig-
nals that are capable of capturing alpha over an intermediate time frame,
measured in days and weeks. … One such strategy that has proved effective
begins with identifying clusters of seemingly unrelated stocks that share
common economic return drivers. We then trade within those clusters on a
long–short basis when temporary imbalances in investor buy or sell demand
cause prices of member stocks to decouple from each other, on the expecta-
tion that such dislocations will reverse. To find fundamentally related com-
panies that are not obviously correlated, we employ text-mining algorithms
that can interpret vast quantities of written materials, such as company
reports, regulatory filings, blogs and social media.

Hafez considers text mining, or natural-language processing (NLP), one

of the most disruptive technologies in finance during the past 10–15 years.
He observed,
Ninety percent of all the data in the world has been generated over the last
two years, and about 80% of this data comes unstructured—and a large
part of this is text based. With the introduction of NLP, suddenly this type
of content can be directly incorporated into a (systematic) investment pro-
cess with little or no human intervention. It all started with news but has
expanded into social media, blogs, earnings transcripts, analyst reports,
regulatory filings, and such.
An academic attempt to use these technologies to nowcast corporate
earnings was made by Kamp, Boley, and Gärtner (2014). They ran data on all
S&P 100 Index companies for the 2008–12 period through a fully automated
machine-learning method that relies on publicly available data only. Using
a simple linear regression based on a novel set of nonnumerical factors that
determine market prices, they found that their forecasts could outperform
forecasts made by human analysts.
Another example of harnessing computer power to estimate compa-
nies’ results comes from the Japanese equity research startup Nowcast Inc.36
Nowcast provides information on consumer prices and retail sales transac-
tions in real time and, using these and other data, provides active manag-
ers with automated estimates of the earnings of consumer goods makers. The
Tokyo-based firm presently covers 200 Japanese consumer goods companies
and has plans to expand that number to about 1,000, including companies
in China and the United States. On the macro side, Nowcast couples data
from various data sources and advanced analytics to supply real-time inflation
numbers and, using nighttime illumination images from satellites as a proxy
for real-time economic activity, plans to provide real-time GDP numbers for
some of the major world economies in the first half of 2017.
Brar said, “To me, these technologies are of value to fundamental ana-
lysts. In the past, analysts used to stand outside stores to measure shopper
traffic; now we have more systematic means to do the same.”
Expectations and Limitations

Clearly (big) data and advanced analytics have the potential to bring advan-
tages to active managers. The chief investment officer at one quant manager
remarked,
Nowcast Inc. was formed in 2015 at the University of Tokyo and was subsequently pur-
36
chased by the Japanese app maker Finatext Ltd.

Equity Valuation
The traditional active manager with 20–50 stocks will know each one
extremely well and will probably have more information than we can hope
to have. But we can analyze 5,000 stocks on a consistent basis, whereas a
traditional manager using a screen to narrow down his universe will inevi-
tably lose a lot of information.
J.R. Lowry (2016), EMEA head of State Street Global Exchange, writes
that he believes that as processes continue to be automated and more informa-
tion is moved and created online, and as advances in analytic tools allow the
accumulation and assessment of the new data, “Some of this data is going to
have investable value, and investment managers need to be able to ingest and
mine it successfully. However, identifying the right monetisable insights—
separating the signal from the noise, so to speak—is a significant challenge.”
An example of the challenge comes from what is called the “Hathaway
effect.” Reportedly, when the American actress Anne Hathaway is in the
news, Berkshire Hathaway’s stock price receives a boost. In 2015, students at
the University of Kansas School of Business (“Finance Students” 2015) were
tasked with analyzing whether trading algorithms could be misled by irrel-
evant information. Focusing on news coverage of the actress and the returns
of Berkshire Hathaway’s stocks, students found no correlation between news
and the price or volatility of the stock, but they did find a positive correlation
with trading volumes.
Marcos López de Prado, a senior managing director for Guggenheim
Partners and a research fellow at Lawrence Berkeley National Laboratory,
cautioned, “Big data, machine learning, and supercomputing require new
skills. A blind application of machine learning to large financial datasets will
surely result in false discoveries.”
Also at the forefront of applying big data and machine learning in invest-
ment management, Man AHL’s Ledford also cautioned about expectations,
saying, “There is no free lunch here: adequate identification of a machine
learning model may require extremely large datasets that are, in practice,
larger than history allows.”
Which raises the question, What are the true capabilities and limits of
these technologies? Clearly, analysts can now interact with machines using
natural languages and search large databases of texts asking questions formu-
lated in natural languages. Computers equipped with algorithms now seem
capable of replacing humans in almost every job, learning from data and dis-
covering relationships better than humans can.
But computers and their algorithms implement step-by-step procedures
whereas human mental activities, such as making connections or drawing
conclusions, are not done consciously following a step-by-step procedure.

In the twentieth century, mathematicians trying to represent all math-

ematical reasoning as a step-by-step procedure soon discovered limits to this
approach. Some limits are theoretical—for example, limits in machines’ abil-
ity to perform chains of inference—while other limits are practical. Even if
we can state some problems mathematically, they are not necessarily solvable
in a finite time by any computer. For example, many optimization problems
in finance and economics imply computations that are simply too long to be
performed.
Thus, a first limitation to the use of machine learning is simply that not
all problems can be easily stated mathematically—and not all problems that
can be stated mathematically can be computed.
Learning is a case in point. Machine learning is not what we call “learn-
ing” in daily life and in higher mental activities but is essentially a process
of minimization of a cost function. Machines learn by minimizing a cost
function that has been preprogrammed. They can learn to mimic examples as
close as possible (supervised learning) or can learn to find a structure in data
(unsupervised learning). In both cases, learning is essentially a process of the
minimization of some function. For example, a machine can learn to separate
a set of stocks into groups, maximizing the similarity in each group and the
dissimilarity between groups, and to recognize price patterns that might be
profitable.
A second limitation to the use of machine learning is the fact that some
human processes might simply be too complex to be coded into a machine. By
forcing all aspects of economic life into a machine-learnable computation, we
might actually lose the richness of our thought processes.
Another important point concerns the use of data-mining techniques.
What is data mining if not the use of computers to discover patterns in the
data? It is an instance of unsupervised learning that creates a mathematical
model of data, again (usually) minimizing some function. For example, sup-
pose we want to mine a large dataset of time series—say, 15,000 time series
of 100,000 high-frequency price data, plus 100 macroeconomic time series—
to discover hidden relationships between these data for investment decision
making. Data mining works by assuming a general model of the possible
relationships, with possibly a large number of parameters. Such a model—
in principle, a universal model—would discover many hidden relationships,
including some that are spurious (random). To avoid discovering purely ran-
dom relationships, the data-mining model is constrained.
This method was popular in the late 1980s with the arrival of the first
low-cost supercomputers, and it soon became apparent that the risk of discov-
ering spurious patterns is high. Now, in applying data-mining techniques to

Equity Valuation
much larger datasets, researchers may need to place even stronger constraints
on models so that patterns discovered are significant, not spurious, or else
take a different approach.
The level of complexity in the financial services sector presents another
problem. RavenPack’s Hafez observed,
Today, the various artificial intelligence methodologies are mostly used as
part of the data preprocessing stage to produce various analytics or insights
from unstructured content rather than in the actual investment process
itself. One of the main reasons for this is that finance is a more complex
space than most other sectors. It can generally be characterized as involv-
ing incomplete information, non-stationarity, and having fuzzy objectives.
Today, machine learning can hardly distinguish cats from dogs in a pic-
ture if one adds just a bit of “noise.” Financial markets are magnitudes more
complex.
Macquarie’s Brar commented on some of the downsides of using big data
and advanced analytics, saying,
One example: If we only focus on the outputs and don’t understand the
inputs (i.e., the data), the quality of data and how it is being applied, etc., we
will fall into the trap of GIGO [garbage in, garbage out]. Moreover, if we
don’t build a hypothesis and only interpret the outputs, then again we’re at
risk. We’ve always had (big) data, but the key is having an inquisitive mind
and knowing what questions to ask.
Another consideration comes from Ledford: “Whilst the machine-learn-
ing tools and the inferred knowledge are quantitatively expressed, as for any
model, they may just be inputs within a wider qualitative framework.”
That framework is the process of fundamental valuation and valuation
models. According to Lleo, although we are still at an early stage in under-
standing how to use these new tools, machine learning will
dramatically increase the speed at which this [valuation] analysis is per-
formed, broaden the range of information used to forecast cash flows and
estimate the cost of capital, increase the breadth and depth of scenarios
considered, and upgrade the visualization of the results, thanks to interac-
tive drill-down reports.
Among other effects these tools will have on the valuation and portfo-
lio construction process, Lleo mentioned that machine learning will do the
following:
•• Bring greater coherence to the valuation of firms within an industry,
country, and internationally, with more powerful learning algorithms to
limit discrepancies in valuation between firms and countries. This benefit

is important because investment has broadened from a domestic, national

environment to an international multiasset environment.
•• Move from a target price to a target range by broadening the spectrum
of scenarios considered in the valuation exercise, thanks to the variety of
assumptions one can consider and visualization techniques.
•• Connect valuation and portfolio management more closely than in the
past. A range of valuations gives a distribution of returns for each asset,
complete with a distribution structure that can be used directly in a port-
folio optimization model (perhaps toward a greater use of forecasted data
as opposed to historical data in portfolio optimization).
But will all investment management firms be able to profit from big
data and these tools? Brar commented, “Building such tools and applica-
tions requires huge upfront costs—people, data, management commitment—
and real benefits would be realized in the future.” Already today, one-third
(33.6%) of the €56.3 trillion in 2016 global assets managed by the Top 400
asset management firms is managed by just the top 10 firms (Kennedy 2016).
Hafez agreed that the upfront costs are high but suggested alternatives to
in-house development:
It takes a serious investment in infrastructure to build and maintain these
systems in-house. The work involved in bringing in raw content and clean-
ing it is nontrivial. It requires serious investments in data acquisition, ven-
dor management, IT infrastructure, service maintenance, natural-language
expertise, and so on. Today, even the most sophisticated quantitative hedge
funds and asset managers are outsourcing the work to various data and
analytics providers, like Thomson Reuters, Bloomberg, or RavenPack. We
already see some firms combining in-house expertise with unstructured-to-
structured content from external providers.
Impact of Big Data and AI in Asset Management

How will big data and advanced analytical techniques affect the skills needed
by the analyst, investment strategies, and the industry in general?
López de Prado manages $15 billion in internal funds for Guggenheim
Partners, where he directs a team of 30 mathematicians and computer sci-
entists who apply (exclusively) machine-learning methods. Commenting on
their approach, he said,
We apply a rigorous scientific process to the research, testing, and deploy-
ment of investment strategies. Having more people would make us less
effective—and might compromise some of our trading secrets. Five years

Equity Valuation
ago, it would have been unthinkable for such a small group of people, with
no background in finance, to manage this amount of money. Machine learn-
ing and supercomputing are changing finance. In 10 years’ time, finance
will have more to do with computer science than economics.
As for the analyst, Lleo remarked, “All of these changes will have an
impact on the skill set of financial analysts. The way in which CFA charter-
holders work in 10–15 years may be very different from the way they work
now.”
All these changes will also likely reduce the number of analysts (and other
personnel) in asset management. In a recent report on the use of data and AI
in the financial services industry, Pierron (2017) estimated that these new
tools will result in a loss of more than 110,000 jobs in asset and private wealth
management between now and 2025. However, analysts also have a market-
ing function at asset management firms: They help establish the credibility of
the firm regarding its claim to be able to generate above-market returns.
Macquarie’s Brar commented on the impact of these technologies on
investment strategies:
If these signals help gain competitive advantage—extract alpha or execu-
tion—then these innovations will be a disruptive force. The impact will be
greater in less-developed markets, but one cannot ignore sophisticated mar-
kets like the United States, in which gaining a basis point matters to overall
fund performance.
As for the impact of these technologies on the industry as a whole, Brar
added,
Investors who build good platforms and systems will gain a competi-
tive edge, whilst others will have to try to enhance their investment pro-
cess across other dimensions to maintain their competitive edge. It is
too early to say whether the cost/benefit ratio is or will be positive in the
future. However, it will be very difficult for an active manager not to be in
this space.
RavenPack’s Hafez believes that the emergence of alternative data will
affect all investment styles, active and passive; all investment strategies,
including value and momentum, among others; and all investable universes.
He remarked,
These new technologies can be expected to support both the smart-beta
strategies as well as help active investors keep their competitive edge. On
the active side, more data creates more opportunities to find new alpha
sources, and the future winners in the active (systematic) space will be those
who manage to process the most data, most efficiently. The idea of creating

hundreds of thousands, or even millions, of weak learners (or alphas) seems

to be part of the winning formula within future quant investing. Of course,
there is still the other side of the coin, with a strong hypothesis-driven
investment process, combining fundamental domain expertise with the
support of alternative data sources—that is to say, using a human overlay on
data to alleviate noise.
Hafez added that although alternative data and AI are finding value
in US large-capitalization names, the most potential will be found among
companies or markets where information is harder to come by or simply not
trusted, such as in small-cap stocks or emerging markets. As for geographical
take-up, Hafez remarked that while most alternative data available today are
for US companies, facilitating the use of AI methodologies on these firms,
Europe is catching up, though language differences mean that datasets are
more dispersed. He added that interest is high in developing the data and
analytical tools for handling information in Chinese and Japanese.
Is all the talk of big data and AI in asset management a return to the
technology hype that led to the dot-com bubble? Sorensen cited a recent study
by Citi Research (2017) on the business of big data. Referring to the study,
Sorensen said,
It is big in dollars and big in space. Much of it must be stored and processed
on the cloud. The list of providers covers considerable breadth—satellite
imagery, credit card data, and clicks, to name a few. Rumor has it that the
self-identified big-data firms total in the hundreds and reportedly generate
an aggregate $50 billion in revenue annually.
Only time will separate hype from reality.

6. E
quity Valuation: Does It, Can It Fulfill
Its Promise?
Asset managers put time and money into their promise to deliver above-
market returns to their clients. They use equity analysts to undertake funda-
mental analysis and develop valuation models to identify mispricings. Their
investment decisions are based on the belief that under- and overpriced assets
will revert to some mean or fair value within a given period of time. The
expectation is that the manager will then deliver on its promise.
But does the manager deliver? Ever since Burton G. Malkiel, then an
economics professor (now emeritus professor) at Princeton University, penned
A Random Walk Down Wall Street in 1973, the argument has been raging as to
whether equity analysts and active management add value.
A frequently used way to address the question is to look at mutual fund
returns relative to benchmark funds. However, a caveat is in order. Mutual
fund performance available to researchers is limited to returns based on his-
torical price and dividend data. A fund’s beta and return volatility are com-
puted from historical returns. Given historical returns, computed return
volatility, the estimated beta, and the benchmark, a fund manager’s relative
performance is determined. The problem is that a fund’s return is determined
by several activities performed by the fund’s management team. Typically, for
internal evaluation purposes, a fund does not naively look at only the return
and some risk measure to assess the skills of the members of its management
team. Rather, some type of performance attribution analysis is used for inter-
nal purposes by the financial manager and the trustees.
The most basic equity performance attribution model considers the con-
tribution of three activities: security selection, sector allocation, and market
timing (i.e., adjusting beta on the basis of a market view). Assuming that the
estimated beta is correct and adjusted for comparing performance against a
benchmark, the market-adjusted relative performance would be attributable
to either security selection or sector allocation. Thus, equity valuation models
may lead to superior selection of stocks within a sector, but an inferior alloca-
tion among sectors would result in overall underperformance.
With that caveat in mind, consider the S&P Indices Versus Active
(SPIVA) Scorecards, starting with actively managed large-cap funds. As
Table 6.1 illustrates, active large-cap funds in Europe, Japan, and the
United States consistently underperform the S&P large-cap indexes in their

6. Equity Valuation
Table 6.1. Percentage of Large-Cap Funds That Under- and Outperformed Their

Market Benchmark over One-, Three-, and Five-Year Periods Ending
31 December 2016
One-Year Trailing Three- Trailing Five-
Performance Year Performance Year Performance
Benchmark
Market Index Under Out Under Out Under Out
Europe S&P Europe 350 80% 20% 74% 26% 74% 26%
Japan S&P TOPIX 150 64% 36% 60% 40% 74% 26%
USA S&P 500 66% 34% 93% 7% 88% 12%

Source: Constructed by the authors using data obtained from SPIVA (http://us.spindices.com/
SPIVA/#/).
respective markets, whether the results are looked at over a one-year, three-
year, or five-year period.
Active underperformance relative to S&P indexes is even greater in mid-
and small-cap markets, where companies typically benefit from less (sell-side)
research than do large-cap firms. Table 6.2 provides data from the United
States as an example.
Although active managers as a group fail to outperform large-, mid-, and
small-cap indexes in developed countries, can they do better in emerging-
market equities? Some, including Lukas Daadler, chief of investment solu-
tions at the Dutch asset manager Robeco Institutional Asset Management,
believe they can. Daadler told IPE Magazine journalist Christopher O’Dea
(2016) that using market multiples, including price-to-book ratios and P/Es,
Robeco estimates that emerging-market stocks are undervalued, perhaps by
as much as 25%–30% compared with the MSCI World Index. This gap may
provide an opportunity for equity analysts and active managers to add value.
Table 6.2. Percentage of US Active Funds That Under- and Outperformed Their

Size-Relevant Market Benchmark over One-, Three-, and Five-Year
Periods Ending 31 December 2016
One-Year Trailing Three- Trailing Five-
Performance Year Performance Year Performance
Size
Category Benchmark Index Under Out Under Out Under Out
Midcap S&P MidCap 400 89% 11% 94% 6% 90% 10%
Small Cap S&P SmallCap 600 86% 14% 96% 4% 97% 3%
Source: Constructed by the authors using data obtained from SPIVA (http://us.spindices.com/
SPIVA/#/).

Equity Valuation
In fact, emerging-market stocks are unlikely to be uniformly undervalued;

what is more likely is that significant differences between stocks exist. This
gap offers profit opportunities.
Not everyone agrees. In a recent interview with Daniel Ben-Ami (2016),
Malkiel said that even in the case of emerging equity markets—generally
viewed as less efficient than developed equity markets—passive managers
generally outperform active managers. The only exception he had found,
Malkiel said, were local markets in China, where insider trading is said to be
common.
Could not only individual analysts but also analysts as a group,
from different regions of the world, differ in their ability to forecast stock
prices? Comparing analyst performance in both buy and sell recommenda-
tions, Harvey, Radnor, Mohammed, and Ferreira (2013) found that Asian
analysts—in particular, Japanese analysts—outperform both European and
US analysts. Asian analysts made profitable forecasts on both buy and sell
recommendations. The authors exclude the efficiency of markets as an expla-
nation but offer no alternative explanation. Might the data, the analytics, be
part of the analyst’s skill?
Sébastien Lleo, professor of finance at NEOMA Business School
(France), remarked that ever since Jensen first looked at mutual fund per-
formance in 1968, financial markets have been grappling with two key ques-
tions: (1) What is the value of active management? and (2) Does this value
justify its cost? Lleo observed,
Equity markets in developed economies are open and competitive, and this
makes them hard to beat on a risk-adjusted basis over the long term. The
results in studies of active management performance are also sensitive to
the period under consideration. In addition, the studies generally use the
same broad market index to gauge the performance of all the funds, rather
than the fund’s stated investment benchmark. While the S&P 500 may be
the stated benchmark for some active mutual fund managers, others may
have different benchmarks, such as the Russell 2000, because they have dif-
ferent strategies and emphasis. And other managers may not have bench-
marks at all.
For example, Warren Buffett famously resisted pressure from sharehold-
ers to make the S&P 500 Berkshire Hathaway’s benchmark. Although the
performance of the S&P 500 now features in Berkshire Hathaway’s annual
report, Buffett made it clear that it was for information only and that it did
not reflect the strategy of the firm. Overall, ignoring the manager’s true
benchmark will bias the results of the studies.

6. Equity Valuation
Still, the dialectic between cost and performance has shaped the investment
industry, driving the explosion in the number of hedge funds in the 2000s,
the boom in purely passive ETFs [exchange-traded funds], and now the rise
in smart-beta ETFs and alternative-beta funds. Each of these successive
designs provides a very different set of answers to the two original questions
of the value and cost/benefit of active management. Right now, smart beta
and alternative beta have anchored the debate in the “low-cost” camp, with-
out giving up on the dream that a strategy (or small set of strategies) can
systematically beat the market on a risk-adjusted basis over the long run.
The role of behavioral biases in explaining (under)performance was
underlined by Dean Mcintyre, director of performance strategy at FactSet.
Mcintyre (2016) identified two widely recognized biases with an impact on
performance: the instinct to follow the herd and the tendency to stick to pre-
vious price targets.
Another way of understanding how much value analysts and active man-
agement add is to look at persistence in performance. Soe and Poirier (2016)
write that according to the S&P Persistence Scorecard for the United States,
one of the key measurements of successful active management lies in the
ability of a manager or a strategy to deliver above-average returns consis-
tently over multiple periods. Demonstrating the ability to outperform peers
repeatedly is the only proven way to differentiate a manager’s luck from
skill. According to the S&P Persistence Scorecard, relatively few funds can
consistently stay at the top. Out of 631 domestic equity funds that were
in the top quartile as of September 2014, only 2.85% managed to stay in
the top quartile at the end of September 2016. Furthermore, 2.46% of the
large-cap funds, 2.20% of the mid-cap funds, and 3.36% of the small-cap
funds remained in the top quartile. For the three-year period that ended in
September 2016, persistence figures for funds in the top half were equally
unfavorable. Over three consecutive 12-month periods, 18.07% of large-cap
funds, 22.95% of mid-cap funds, and 20.88% of small-cap funds main-
tained a top-half ranking.
An inverse relationship generally exists between the measurement time
horizon and the ability of top-performing funds to maintain their status.
It is worth noting that less than 1% of large-cap funds and no mid-cap or
small-cap funds managed to remain in the top quartile at the end of the
five-year measurement period. This figure paints a negative picture regard-
ing the lack of long-term persistence in mutual fund returns. Similarly, only
4.47% of large-cap funds, 3.68% of mid-cap funds, and 9.27% of small-
cap funds maintained top-half performance over five consecutive 12-month
periods. Random expectations would suggest a repeat rate of 6.25%.

Equity Valuation
We can also look at the equilibrium between the cost and the gains of
active management by considering how much information equity analysts add.
At the macro level, one might well ask: Have markets been kept (quasi)
efficient, thanks to the combined action of active managers? Up until just
10 years ago, 84% of US mutual fund and ETF assets were actively invested,
though the figure had fallen to 66% by the end of 2016.37 Is evidence available
of a change of efficiency during this period?
“No,” said Alfred Slager, professor of pension fund management at TIAS
School for Business and Society and trustee at the Dutch pension fund for
general practitioners SPH:
The argument is that with less money in active funds, there are fewer trans-
actions, more information asymmetries, and thus more active opportunities.
But we observe fewer active managers, more transaction volume, and simi-
larly disappointing active results. So, this argument does not hold. We need
to adapt our view of efficiency and financial markets.
I would suggest the application of the Market Segmentation Theory from
the fixed-income realm.38 Long-term investors are typically buy-and-hold
investors who have a different supply-and-demand schedule than the ETF,
HFT [high-frequency trading], or insurance company sectors. Depending
on the distribution of the security in the different segments, one could
hypothesize whether it has become more liquid or not and whether that
indicator of efficiency has any meaning.
Concerning how much information equity analysts add—and its value
to investors—a recent article in the Economist (“Breaking Up Is Hard to Do”
2017) had some unkind words for the sell-side research bundled in banks’
services to asset managers:
At present, banks blast their clients’ inboxes with thousands of reports, only
a fraction of which are read. The problem is that most research is not very
useful—it is hard to come up with original insights about big companies
when dozens of other researchers are trying to do the same.
37
Morningstar via the Wall Street Journal (18 October 2016).
38
According to Investopedia: “Market segmentation theory is a fundamental theory regard-
ing interest rates and yield curves, expressing the idea that there is no inherent relationship
between the levels of short-term and long-term rates. According to market segmentation
theory, the prevailing interest rates for short-term, intermediate-term and long-term bonds
should be viewed separately, as items in different markets for debt securities. The major con-
clusion drawn from market segmentation theory and applied to investing is that yield curves
are determined by supply and demand forces within each separate market, or category, of
debt security maturities, and that the yields for one category of maturities cannot be used to
predict the yields for securities within a different maturity market” (www.investopedia.com/
terms/m/market-segmentation-theory.asp#ixzz4iTuVfVej).

6. Equity Valuation
Citing estimates from Frost Consulting that sell-side equity research

budgets have almost halved in the past eight years—going from $8.2 billion
in 2008 to $4 billion in 2016—and estimates from the research firm Coalition
that research jobs are down at banks by around 10% since 2012, the article
refers to a “secular decline” in banks’ equity research operations. Some of this
decline is attributed to new European financial regulations (MiFID [Markets
in Financial Instruments Directive] 2) that will force an unbundling of banks’
services to asset managers as of January 2018.
Bradford Cornell, professor of financial economics at the California
Institute of Technology, challenged the equation of market efficiency with
the inability of active managers to outperform passive indexes. In a recent
blog post, Cornell (2016) recalled work in the early 1980s by Richard Roll
and himself (Cornell and Roll 1981) and by Sanford J. Grossman and Joseph
E. Stiglitz (1980) demonstrating that markets can never be fully efficient;
efficiency, he remarked, is not the result of some natural phenomenon but the
result of research by fundamental investors. He added,
If those fundamental investors cannot earn a fair rate of return on the
resources they put into investment research they will cut back. But as fun-
damental investors cut back and indexing becomes more common, prices
will begin to diverge from fair value making investment research more
profitable. As a result, economic theory predicts that the market must be
sufficiently inefficient to allow at least sophisticated investors to earn a fair
return on their efforts by identifying mispriced securities.
Cornell (2016) cited William Sharpe’s argument that equating market
efficiency with the inability of active managers as a group to outperform pas-
sive indexes is a mistake. Sharpe’s proof, Cornell wrote, is as follows:
Divide investors into two groups: passive investors who hold the market
index and active investors who engage in research in an attempt to beat the
market. Suppose that in a given year the return on the market index is 10%.
By definition passive investors who index the market will also earn 10%. But
that means that active investors, as a group must also earn 10%, before costs.
Given the costs of active investing, active investors as a group must always
do worse than passive investors. As Sharpe stresses, this result has nothing
to do with market efficiency—it is an arithmetic identity. Even in the most
wildly inefficient market, passive investors as a group would still outperform
active investors, as a group, taking account of costs and fees. What is true
is that if the market is highly efficient, so that few securities are mispriced,
there is likely to be little superior or inferior performance among the class of
active investors. Conversely, if the market is more inefficient, then the more
sophisticated investors, who can identify mispriced securities, will benefit at
the expense of less informed active investors.

Equity Valuation
Cornell suggested that for investors who think they have the skill to
identify mispriced securities, knowing whether the current movement toward
indexing has led to increased market inefficiency would be nice. He explained,
Ideally, there would be an index of market efficiency that investors could use
to judge how likely it would be to find mispriced securities. Unfortunately,
there is no such index and there is not likely to be one in the foreseeable
future. Asset prices are so volatile and market conditions are so variable that
a reasonable index of “inefficiency” cannot be constructed. That is why, 50
years after Eugene Fama introduced the idea of market efficiency, scholars
are still arguing about how efficient the market is. There is no evidence that
the debate is subsiding. While conceptually it follows that the move toward
passive investing will lead to greater inefficiency, whether there has been
any material change in market efficiency thus far is unknown.
Some have suggested that if active managers accounted for as little as
10% of the market, efficiency would still be assured. If so, active investors as a
group will have performance problems for some time.
Keeping the market price aligned with a firm’s intrinsic value is widely
considered a key economic role of the fundamental analyst. So, fundamental
analysts play an important role in enhancing a firm’s access to capital and its
ability to invest. Derrien and Kecskés (2013), expanding on previous stud-
ies, provided empirical evidence that a decrease in analyst coverage increases
information asymmetry. They found that companies that lose an analyst
decrease their investment and financing by up to 2% of total assets compared
with similar companies that do not lose an analyst. Perhaps not surprisingly,
they found that results were stronger for small companies and those with less
analyst coverage.
Equity Valuation in a Radically Changed Environment:

Does It Retain Its Value?
Graham and Dodd wrote their seminal book Security Analysis: Principles and
Techniques in 1934. Some 20 years later, roughly 4% of the US population
owned stocks. Another 40 years later, that figure had grown fivefold: 20%
of the US population owned stocks (including stock mutual funds), thanks
largely to the introduction of individual retirement accounts (1974) and the
first index funds (1976). The investable universe was still largely domestic
and small: 2,670 listed firms in the United States in 1975, according to the
World Bank. According to the same database, only 1,398 firms were listed in
Japan in 1975, 471 in Germany, and none in either China or India. By 2016,
the number of listed firms worldwide had exploded to 43,192, of which only

6. Equity Valuation
4,331, or 10%, were in the United States; China (3,052) and India (5,820)
together accounted for 20% of all listed firms worldwide. The number of
listed firms for Germany in 2016 was 555 and for Japan 3,504.39
As discussed in Chapter 5, new technologies are accelerating the analysis
and trading of stocks worldwide. Not only has the number of stocks greatly
increased because of emerging markets and globalization, but the invest-
able universe has also significantly expanded to include products other than
equities. Investors looking for returns have many more options than they did
only two or three decades ago. This development raises the question of how
returns will be produced in such a changed scenario. Does fundamental anal-
ysis retain its central role?
Slager answers “yes and no.” He explained,
Expanding the investment universe increases the need to analyze overall
market valuations and the differences and commonalities between them.
So, on an aggregate level, fundamental analysis helps. On an individual
level, I would not be interested in whether the best security would be picked;
“fit-for-purpose” would be fine too. By fit-for-purpose, my idea would be—
besides the fact that the company is financially viable—that governance is
in order, shareholder rights are protected, and especially ESG [environmen-
tal, social, and governance] factors have been taken into consideration. All
are crucial for long-term risk management. I sort of suspect that ESG is a
form of DNA or fingerprint of the organization and in that sense, might
have more predictive value than financials.
Kenneth Little, managing director of the investments group at Brandes
Investment Partners, said,
We believe fundamental analysis can and should retain its central role in
investment management, despite the wave of new investment vehicles and
strategies over the past few decades. A key function of capital markets is
providing for the efficient allocation of capital throughout the economy.
We believe fundamental analysis is required to determine which compa-
nies deserve (or do not deserve) capital, and their share prices should adjust
accordingly.
He added, “While the growth in popularity of index investing (and
ETFs that track the indexes) has led to tremendous growth in assets in these
vehicles, it has done little to improve the efficient allocation of capital within
the respective markets.”
For more information, see World Bank data on listed companies worldwide (http://data.
39
worldbank.org/indicator/CM.MKT.LDOM.NO?locations=IN-CN-DE-JP-US).

Equity Valuation
Matteo Bonaventura, a buy-side financial analyst at Banor SIM in Milan,

also believes that equity analysis retains its central role. He cited examples
from the Italian stock exchange, Borsa Italiana, over the 2007–17 period,
during which some underresearched companies realized returns of 500%. But
equally important, he noted, is the need for fundamental analysis at several
levels, not just the firm level: “A stock market is a mirror of the country, so
understanding at least the fundamentals of the country and of the business is
crucial.”
Bradford Cornell, a professor of financial economics at the California
Institute of Technology, believes that “fundamental valuation remains the
core of investing. It’s central to market pricing and, thereby, capital alloca-
tion.” Anyone interested in investing, he believes, should master discounted
cash flow analysis.
The discussion about the relative importance of asset allocation and
active management has changed since the publication of Brinson, Hood, and
Beebower’s (1986) paper on the determinants of portfolio performance. Their
paper was generally taken to have established the dominant role (90%) of
asset allocation in explaining equity returns. The role of active management
was reasserted 24 years later by Xiong, Ibbotson, Idzorek, and Chen (2010).
They pointed out (1) that the 1986 Brinson et al. paper did not study perfor-
mance but rather variation in performance between one fund and another
and (2) that if general market movements (which they found were responsible
for about 70% of variation in performance) were excluded, the variation in
time-series returns is explained almost equally by asset allocation and active
management (16% versus 14%).
The question remains: Is going through the selection process needed to
identify an active manager capable of (consistently) outperforming the market
worth the time and effort? A recent report from Greenwich Associates (2016)
revealed a shift of emphasis on the part of institutional investors from out-
performance relative to a benchmark to asset allocation. The report attributed
this shift to market volatility and the large-scale devaluation of equity assets
during the time period covered by the study.
Andrew Clare, professor of asset management at Cass Business School
in London and a pension fund trustee, believes that for many pension fund
trustees, searching for the best active managers is not worth the effort.
Other decisions, Clare told IPE Magazine journalist Carlo Svaluto Moreolo
(2016), particularly those relating to asset allocation, are likely to make a far
greater difference to overall investment returns: “It’s just [that] the amount of

6. Equity Valuation
difference it [active equity investing] can really make is, for many trustees,
not worth the time needed to monitor those things carefully.”
Jaap van Dam, principal director of investment strategy at PGGM
Investments, the €200 billion Dutch pension fund for health and social work-
ers, voiced a similar opinion. He said, “If you look at aggregated results of
pension funds globally, there is a small contribution from active manage-
ment to the total return, and only if it is well controlled for cost.” PGGM
is reviewing its bottom-up investment decision-making processes. “We now
want more discussion on understanding where and how value is created and
transformed into profits,” Van Dam said. “We believe in creating value by
looking at the fundamentals of value creation over the long term.”
A New Role for Fundamental Analysis in Asset

Management?
Can equity analysts and active managers find a new role for themselves?
Given past performance records, Axel Pierron, co-founder and co-
managing director of the financial consultancy firm Opima, was somewhat
pessimistic about active management’s ability to compete on alpha generation.
He said,
If you look at passive investment vehicles that are gaining market share,
such as ETFs, BlackRock has 50% of the market in Europe. Branding is
becoming very important. If you don’t have brand recognition, you might
try to focus on alpha, but it is not so easy. Regulation in Europe that
imposes benchmarking will have an impact. Without strong performance,
an asset manager will need a strong brand or to be part of a large retail
bank. Independent asset managers will find it harder to acquire and keep
assets.
Looking over the past two decades—which saw the emergence of, first,
hedge funds and, more recently, smart-beta ETFs—in answering the ques-
tions of whether active management adds value and whether it is worth the
cost, Lleo remarked,
The rise of robo-advisors may change this emphasis again by transform-
ing active management into a dynamic, client-focused process, in which
the ultimate objective in constituting a portfolio is not to beat the mar-
ket but to deliver results consistent with the goals and psychology of the
investor.
Beyond psychology, Marc Reinganum, a former senior managing direc-
tor at State Street Global Advisors, suggests,

Equity Valuation
Outcome-centric investing is on the ascent. Outcomes are future

consumption or spending streams and should be treated as future liabili-
ties. The goal of an outcome-oriented strategy should be to match or exceed
changes in the present values of the liabilities with minimum return devia-
tions between the liabilities and the assets that are funding the liabilities.
Pascal Blanqué, chief investment officer and head of institutional busi-
ness at the French asset management group Amundi, suggested that the time
has come for active managers to shift the accent from stock picking to asset
allocation. Blanqué (2016) writes:
The role of active managers is increasingly challenged, as far as stock pick-
ing is concerned; nevertheless, it maintains a highly important role in
asset allocation. Even sophisticated investors with a strong asset allocation
expertise show increasing interest in innovative asset allocation methods
and processes, with a preference for absolute-return over more traditional
benchmark approaches. As a result, we are convinced that the distinction
between active and passive management is becoming less clear-cut.
The role of active management has not disappeared; its focus has just shifted
over time. The choice of factors and weighting schemes and the search for
value through asset allocation are all active decisions that investors must
now focus on. For asset managers, the challenge is therefore to provide a
combination of a wide range of passive vehicles and of selected active exper-
tise with proven alpha, with a strong capacity to accompany clients in their
asset allocation decisions and in the efficient execution of the latter.
In a similar vein, Slager had several suggestions as to how active manag-
ers might increase their value to (institutional) investors:40
From a pension fund perspective, I see that in Europe, classic active strate-
gies are being eschewed, and passive is on the rise. At the same time, in
discussions with trustees, the picture emerges that active strategies have an
added value, just not in the somewhat antiquated “let’s beat the benchmark”
form. Some ideas that have emerged from recent discussions include using
active managers to work on developing a new set of metrics. What sort of
strategy would best aid the pension funds’ goals? Which risk factors would
one add, compared to the total portfolio?
In other words, linking to the integral objectives, not sticking to isolated

carve-outs in the investment portfolio, should shape future investment
strategies. Other ideas include replacing the classical, highly bureaucratic
“Request for Proposal” with the opportunity for active managers to present
40
See also Koedijk and Slager (2011).

6. Equity Valuation
truly relevant business cases for their active management strategies. Active
managers could innovate on testing strategy introduction: How could we
emulate prototyping, testing, and the introduction of a new strategy in such
a way that it filters out the mediocre strategies from the start? Moving from
the classical database backtesting to live simulation raises the hurdle but
increases the chances of designing durable investment strategies, and of a long-
term partnership. Trustees would no longer be drawn into overly technical
asset pricing discussions but could focus on what matters—will active man-
agement work?

References
Aggarwal, Rajesh, Sanjai Bhagat, and Srinivasan Rangan. 2009. “The Impact
of Fundamentals on IPO Valuation.” Financial Management, vol. 38, no. 2
(Summer): 253–284.
Allan, Gareth, and Komaki Ito. 2016. “Fintech Venture Targets Hedge Funds
with Big-Data Research.” Bloomberg (12 September; updated 13 September):
www.bloomberg.com/news/articles/2016-09-13/fintech-startup-dives-into-
big-data-for-japanese-stock-research.
Almeida, Robert M., Jr. 2016. “Decision Drivers: Stock Prices versus
GDP.” MFS White Paper Series (October): www.mfs.com/content/dam/
mfs-enterprise/pdfs/thought-leadership/us/mfse_gdp_wp.pdf.
Ang, Andrew, and Geert Bekaert. 2007. “Stock Return Predictability: Is It
There?” Review of Financial Studies, vol. 20, no. 3 (May): 651–707.
Appelbaum, Eileen, and Rosemary Batt. 2016 (updated March 2017).
“Are Lower Private Equity Returns the New Normal?” Center for
Economic and Policy Research (June): http://cepr.net/publications/reports/
are-lower-private-equity-returns-the-new-normal.
Bajo, Emanuel, Thomas J. Chemmanur, Karen Simonyan, and Hassan
Tehranian. 2016. “Underwriter Networks, Investor Attention, and Initial
Public Offerings.” Journal of Financial Economics, vol. 122, no. 2 (November):
376–408.
Barra/MSCI. 2010. “Is There a Link between GDP Growth and Equity
Returns?” (May): www.msci.com/documents/10199/a134c5d5-dca0-420d-
875d-06adb948f578.
Beck, Ulrich. 1992. Risk Society: Toward a New Modernity, English ed.
London: SAGE Publications.
Ben-Ami, Daniel. 2016. “Active Management: The Active-Passive Debate.”
IPE Magazine (January): www.ipe.com/reports/active-managemet/special-
report-active-management-the-active-passive-debate/10011319.article.
Bindseil, Ulrich, and Philipp J. König. 2013. “Basil J. Moore’s Horizontalists
and Verticalists: An Appraisal 25 Years Later.” Review of Keynesian Economics,
vol. 1, no. 4 (Winter): 383–390.

Equity Valuation
Birstingl, Andrew. 2016. “Initial Public Offerings: Q4 2016 Highlights.”

FactSet IPO Quarterly (December): https://insight.factset.com/hubfs/Insight/
Migration/IPO%20Quarterly/IPO%20Quarterly%20Q4%202016_12.29.
pdf.
BlackRock. 2016. “Global Investment Outlook: Q2 2016.” BlackRock
Investment Institute (www.blackrock.com/corporate/en-ca/literature/white-
paper/bii-global-investment-outlook-q2-2016-us.pdf).
Blanqué, Pascal. 2016. “Asset Allocation Is an Active Process.” IPE Magazine
(January): www.ipe.com/reports/special-reports/active-management/pension-
funds-what-role-for-active-management/10011322.article.
Bonaventura, Matteo, and Giancarlo Giudici. 2016. “IPO Valuation and
Profitability Expectations: Evidence from the Italian Exchange.” Eurasian
Business Review, vol. 7, no. 2: 247–266.
“Breaking Up Is Hard to Do: Banks’ Equity-Research Operations Are in
Decline.” 2017. Economist (30 March): www.economist.com/news/finance-
and-economics/21719829-unable-give-their-research-away-they-will-
struggle-find-buyers-it-banks.
Brettell, Karen, David Gaffen, and David Rohde. 2015a. “As Stock Buybacks
Reach Historic Levels, Signs That Corporate America Is Undermining
Itself.” Part I of “The Cannibalized Company: A Reuters Special Report.”
Reuters (16 November): https://www.reuters.com/investigates/special-report/
usa-buybacks-cannibalized/.
———. 2015b. “Stock Buybacks Enrich the Bosses Even When Business
Sags.” Part II of “The Cannibalized Company: A Reuters Special Report.”
Reuters (10 December): https://www.reuters.com/investigates/special-report/
usa-buybacks-pay/.
Brinson, Gary P., L. Randolph Hood, and Gilbert L. Beebower. 1986.
“Determinants of Portfolio Performance.” Financial Analysts Journal, vol. 42,
no. 4 (July/August): 39–44.
Burke, John. 2016. “4 Reasons for the IPO Market Slowdown in 2016.”
Investopedia (14 June): www.investopedia.com/articles/markets/061416/4-
reasons-ipo-market-slowdown-2016-ipo.asp#ixzz4O5vkFSle.
Burns, Dan. 2017. “Ghosts of Past Tech IPOs Could Haunt Snap’s
Performance.” Thomson Reuters (24 February).

References
Butters, John. 2017. “Highest Forward 12-Month P/E Ratio for S&P
since 2004.” FactSet Insight (17 February): https://insight.factset.com/
earningsinsight_02.17.17.
Campbell, J.Y., and R.J. Shiller. 1988. “Stock Prices, Earnings and Expected
Dividends.” Journal of Finance, vol. 43, no. 3 (July): 661–676.
Carey, David, and Devin Banerjee. 2016. “Private Equity’s Golden Age
Wasn’t So Golden after All.” Bloomberg (21 January).
Chee, Seungmin, Richard G. Sloan, and Aydin Uysal. 2013. “A Framework
for Value Investing.” Australian Journal of Management, vol. 38, no. 3
(December): 599–633.
Chemmanur, Thomas, and An Yan. 2011. “Advertising, Investor Recognition,
and Stock Returns.” Working paper (April): http://econ.shufe.edu.cn/upload/
htmleditor/Image/74319_1105160822451.pdf.
———. 2017. “Product Market Advertising, Heterogeneous Beliefs, and
the Long-Run Performance of Initial Public Offerings.” Journal of Corporate
Finance, vol. 46 (October): 1–24 (www.sciencedirect.com/science/article/pii/
S0929119917303899).
Citi Research. 2017. “Searching for Alpha: Big Data: Navigating New
Alternative Datasets” (10 March).
Cochrane, John H. 2001 (rev. ed., 2005). Asset Pricing. Princeton, NJ:
Princeton University Press.
Cornell, Bradford. 2010. “Economic Growth and Equity Investing.” Financial
Analysts Journal, vol. 66, no. 1 (January/February): 54–64.
———. 2014. “Dividend–Price Ratios and Stock Returns: International
Evidence.” Journal of Portfolio Management, vol. 40, no. 2 (Winter): 122–127.
———. 2016. “Market Efficiency and the Impact of Passive Investing.”
Brad Cornell’s Economics Blog (7 November): http://wbcornell.blogspot.
com/2016/11/market-efficiency-and-impact-of-passive.html.
Cornell, Bradford, and Aswath Damodaran. 2014. “Tesla: Anatomy of a
Run-Up.” Journal of Portfolio Management, vol. 41, no. 1 (Fall): 139–151.
Cornell, Bradford, and Rajiv Gokhale. 2016. “An ‘Enhanced Multiple’
Corporation Valuation Model: Theory and Empirical Tests.” Business
Valuation Review, vol. 35, no. 2 (Summer): 52–61.

Equity Valuation
Cornell, Bradford, and Richard Roll. 1981. “Strategies for Pairwise

Competitions in Markets and Organizations.” Bell Journal of Economics,
vol. 12, no. 1 (Spring): 201–213.
Costa, Sofia Horta E. 2016. “Market Jolts Make European Buybacks
Newest Investor Favorite.” Bloomberg (4 February; updated 5 February):
w w w.bloomberg.com/news/articles/2016-02-05/once-pro-dividend-
european-shareholders-slowly-warm-to-buybacks.
Credit Suisse. 2017. “It’s Always Darkest before Dawn; Key Secular
Headwinds Will Start to Improve in 2018–19.” Connections Series, Credit
Suisse Global Asset Managers (4 May): www.bluetractorgroup.com/
uploads/5/7/6/1/57612741/credit_suisse_equity_research_may_ 2017_
mention_of_blue_tractor.pdf.
Damodaran, Aswath. 2005. “The Promise and Peril of Real Options.” Working
paper, New York University Stern School of Business (July): www.research-
gate.net/publication/228142262_The_Promise_and_Peril_of_Real_Options.
———. 2016. “Mean Reversion: Gravitational Super Force or Dangerous
Delusion?” Seeking Alpha (1 September): https://seekingalpha.com/
article/4003590-mean-reversion-gravitational-super-force-dangerous-delusion.
Degeorge, François, François Derrien, and Kent. L. Womack. 2007. “Analyst
Hype in IPOs: Explaining the Popularity of Bookbuilding.” Review of
Financial Studies, vol. 20, no. 4 (July): 1021–1058.
DeLong, J. Bradford. 1996. “Is the Stock Market Too High?” Slate (22
December): www.slate.com/articles/news_and_politics/the_gist/1996/12/
is_the_stock_market_too_high.html.
DeLong, J. Bradford, Andrei Shleifer, Lawrence H. Summers, and Robert
J. Waldmann. 1990. “Noise Trader Risk in Financial Markets.” Journal of
Political Economy, vol. 98, no. 4 (August): 703–738.
DeMiguel, Victor, Lorenzo Garlappi, and Raman Uppal. 2009. “Optimal
versus Naive Diversification: How Inefficient Is the 1/N Portfolio Strategy?”
Review of Financial Studies, vol. 22, no. 5 (May): 1915–1953.
Derrien, François, and Ambrus Kecskés. 2013. “The Real Effects of Financial
Shocks: Evidence from Exogenous Changes in Analyst Coverage.” Journal of
Finance, vol. 68, no. 4 (August): 1407–1440.

References
Dimson, Elroy, Paul Marsh, and Mike Staunton. 2014. “The Growth Puzzle.”
In Credit Suisse Global Investment Returns Yearbook 2014 (February): 17–29
(http://doc.xueqiu.com/14cdbae48e74653fe7546fe0.pdf).
Doidge, Craig, G. Andrew Karolyi, and René M. Stulz. 2015. “The U.S.
Listing Gap.” NBER Working Paper 21181 (May): www.nber.org/papers/
w21181.
Downie, Ryan. 2016. “Is the Private Equity Bubble Still Expanding?”
Investopedia (15 June): www.investopedia.com/articles/markets/061516/
private-equity-bubble-still-expanding-gs.asp#ixzz4O5wFX3QU.
Ernst & Young. 2015. “Positioning to Win: 2015 Global Private Equity Survey”
(www.ey.com/gl/en/industries/financial-services/fso-insights-global-private-
equity-survey-2015).
———. 2016. “Disruption Causes Seismic Shift for Private Equity:
2016 Global Private Equity Fund and Investor Survey” (www.ey.com/
gl/en/industries/private-equity/ey-2016-global-private-equity-fund-and-
investor-survey).
Fabozzi, Frank J., K.C. Chen, K.C. Ma, and Jessica West. 2015. “In Search
of Cash-Flow Pricing.” Journal of Financial Research, vol. 38, no. 4 (Winter):
511–527.
Fama, Eugene F. 1970. “Efficient Capital Markets: A Review of Theory and
Empirical Work.” Journal of Finance, vol. 25, no. 2 (May): 383–417.
———. 1976. “Reply.” Journal of Finance, vol. 31, no. 1 (March): 143–145.
Feldstein, Martin. 2016. “What Could Go Wrong in America?” Project Syndicate
(26 October): www.project-syndicate.org/commentary/asset-price-risk-in-
america-by-martin-feldstein-2016-10.
Fernandez, Pablo. 2002 (rev. 2007). “Company Valuation Methods: The Most
Common Errors in Valuation.” Working Paper 449, IESE Business School,
University of Navarra.
Fernandez, Pablo. 2015. “119 Common Errors in Company Valuations.”
Working Paper 714, IESE Business School, University of Navarra.
“Fewer Japanese Companies Planning Stock Buybacks: Declining Trend
Could Affect Supply–Demand Balance.” 2017. Nikkei Asian Review (9
February): http://asia.nikkei.com/Markets/Tokyo-Market/Fewer-Japanese-
companies-planning-stock-buybacks.

Equity Valuation
“Finance Students Investigate the ‘Hathaway Effect.’” 2015. KU (blog),

University of Kansas School of Business (25 February): https://blog.business.
ku.edu/category/anne-hathaway/.
Foulke, David. 2016. “Value Investing Got Crushed During the Internet
Bubble—Here’s Why.” Alpha Architect (blog; 10 October): https://alphaarchitect.
com/2016/10/10/value-investing-got-crushed-during-the-internet-bubble-heres-
why/.
Gao, Xiaohui, Jay R. Ritter, and Zhongyan Zhu. 2013. “Where Have All
the IPOs Gone?” Journal of Financial and Quantitative Analysis, vol. 48, no. 6
(December): 1663–1692.
GF Data Resources. 2017. “Have Mid-Market Values Peaked?” (21 February):
https://www.gfdata.com/news/articles/have-mid-market-values-peaked/.
Goedhart, Marc, Timothy Koller, and David Wessels. 2005. “The Right
Role for Multiples in Valuation.” McKinsey on Finance: Perspectives on
Corporate Finance and Strategy, no. 15 (Spring): 7–11 (www.mckinsey.
com/business-functions/strategy-and-corporate-f inance/our-insights/
the-right-role-for-multiples-in-valuation).
Gompers, Paul, Steven N. Kaplan, and Vladimir Mukharlyamov. 2015.
“What Do Private Equity Firms Say They Do?” NBER Working Paper
21133 (April): http://www.nber.org/papers/w21133.
Gompers, Paul, and Josh Lerner. 2000. “Money Chasing Deals? The Impact
of Fund Inflows on Private Equity Valuations.” Journal of Financial Economics,
vol. 55, no. 1 (January): 281–325.
Graham, Benjamin. 1949 (4th ed., 1973). The Intelligent Investor. New York:
Harper & Row.
Graham, Benjamin, and David Dodd. 1934 (1st ed.; 6th ed., 2008). Security
Analysis: Principles and Techniques. New York: McGraw-Hill.
Grantham, Jeremy. 2017. “This Time Seems Very, Very Different.” Part 2 of
“Not with a Bang but a Whimper—A Thought Experiment.” GMO Quarterly
Letter (https://seekingalpha.com/article/4068324-time-seems-different-part-
2-bang-whimper-thought-experiment).
Gray, Wesley R., and Jack Vogel. 2012. “Analyzing Valuation Measures:
A Performance Horse Race over the Past 40 Years.” Journal of Portfolio
Management, vol. 39, no. 1 (Fall): 112–121.

References
Greenwich Associates. 2016. “Is There a Future for Active Management?

How Active Managers Will Thrive in a Maturing Industry” (December).
Greiner, Steven P. 2011. Ben Graham Was a Quant: Raising the IQ of the
Intelligent Investor. Hoboken, NJ: Wiley.
Grossman, Sanford J., and Joseph E. Stiglitz. 1980. “On the Impossibility of
Informationally Efficient Markets.” American Economic Review, vol. 70, no. 3
(June): 393–408.
Harvey, Campbell R., Sam Radnor, Khalil Mohammed, and William
Ferreira. 2013. “Where Are the World’s Best Analysts?” Working paper
(25 November).
Hillert, Alexander, Heiko Jacobs, and Sebastian Müller. 2014. “Media
Makes Momentum.” Review of Financial Studies, vol. 27, no. 12 (December):
3467–3501.
Ibbotson, Roger G., and Peng Chen. 2003. “Long-Run Stock Returns:
Participating in the Real Economy.” Financial Analysts Journal, vol. 59, no.1
(January/February): 88–98.
Ibbotson, Roger G., Jody L. Sindelar, and Jay R. Ritter. 1994. “The Market’s
Problems with the Pricing of Initial Public Offerings.” Journal of Applied
Corporate Finance, vol. 7, no. 1 (Spring): 66–74.
Ilmanen, Antti. 2016. “A Historical Perspective on Time-Varying
Expected Returns.” Chapter 2 in Financial Market History: Reflections on
the Past for Investors Today, edited by David Chambers and Elroy Dimson.
Charlottesville, VA: CFA Institute Research Foundation (www.cfapubs.org/
toc/rf/2016/2016/3).
Jacobius, Arleen. 2016. “Private Equity Valuations a Concern for Endowments,
Foundations but Are Not Affecting Allocations.” Pensions & Investments
(24 October): www.pionline.com/article/20161024/ONLINE/161029952/.
Jakab, Spencer. 2016. “The Hidden Weaknesses of Index Funds.” Wall
Street Journal (18 October): www.wsj.com/articles/the-hidden-weaknesses-
of-index-funds-1476799335.
Jarrow, Robert. 2016. “Asset Price Bubbles and the Land of Oz.” Journal of
Portfolio Management, vol. 42, no. 2 (Winter): 37–42.
Jenkinson, Tim, Miguel Sousa, and Rüdiger Stucke. 2013. “How Fair Are
the Valuations of Private Equity Funds?” Working paper (February).

Equity Valuation
Jensen, Michael C. 1968. “The Performance of Mutual Funds in the Period

1945–1964.” Journal of Finance, vol. 23, no. 2 (May): 389–416.
Jensen, Niels C. 2015. “The ‘Perfect Storm.’” Absolute Return Letter, Absolute
Return Partners (April): www.arpinvestments.com/arl/the-perfect-storm.
Kamp, Michael, Mario Boley, and Thomas Gärtner. 2014. “Beating Human
Analysts in Nowcasting Corporate Earnings by Using Publicly Available
Stock Prices and Correlation Features.” Proceedings of the 2014 International
Conference on Data Mining (28 April): 641–649 (http://epubs.siam.org/doi/abs/
10.1137/1.9781611973440.74).
Kennedy, Liam. 2016. “Top 400 Asset Managers 2016: Global Assets Now
€56.3trn.” IPE Magazine ( June): www.ipe.com/reports/special-reports/
top-400-asset-managers/top-400-asset-managers-2016-global-assets-now-
563trn/10013542.fullarticle (for the full list, see www.ipe.com/Uploads/d/
t/n/Top-400-list-2016.pdf).
Kim, Moonchul, and Jay R. Ritter. 1999. “Valuing IPOs.” Journal of Financial
Economics, vol. 53, no. 3 (September): 409–437.
Kitanaka, Anna, and Toshiro Hasegawa. 2016. “BlackRock Sides with
Japan’s Central Bank over ETF Buying.” Bloomberg (14 September;
updated 15 September): www.bloomberg.com/news/articles/2016-09-14/
blackrock-sides-with-boj-in-debate-over-tokyo-whale-s-etf-buying.
Koedijk, Kees, and Alfred Slager. 2011. Investment Beliefs: A Positive Approach
to Institutional Investing. Basingstoke, UK: Palgrave Macmillan.
Kumar, Alok. 2009. “Hard-to-Value Stocks, Behavioral Biases, and Informed
Trading.” Journal of Financial and Quantitative Analysis, vol. 44, no. 6
(December): 1375–1401.
Laubach, Thomas, and John C. Williams. 2015. “Measuring the Natural
Rate of Interest Redux.” Federal Reserve Bank of San Francisco Working
Paper 2015-16 (October): www.frbsf.org/economic-research/publications/
working-papers/wp2015-16.pdf.
Ledford, Anthony. n.d. “AHL Explains Machine Learning.” Chapter 3 in
the AHL Explains series Maths: The Power to Do Incredible Things. Oxford,
UK: Man AHL (www.ahl.com/insights/machine-learning.INGS).
Lee, Charles M.C. 2003. “Choosing the Right Valuation Approach.” AIMR
Conference Proceedings, vol. 2003, no. 2 (April): 4–14 (http://www.cfapubs.
org/doi/abs/10.2469/cp.v2003.n2.3271).

References
LeRoy, Stephen F. 1976. “Efficient Capital Markets: Comment.” Journal of

Finance, vol. 31, no. 1 (March): 139–141.
Lewis, Leo. 2016. “Corporate Japan Acquires a Buyback Habit.” Financial
Times (21 December): www.ft.com/content/23076ada-c768-11e6-9043-
7e34c07b46ef?mhq5j=e6.
L’Her, Jean-François, Rossitsa Stoyanova, Kathryn Shaw, William Scott,
and Charissa Lai. 2016. “A Bottom-Up Approach to the Risk-Adjusted
Performance of the Buyout Fund Market.” Financial Analysts Journal, vol. 72,
no. 4 (July/August): 36–48.
Liu, Jing, Doron Nissim, and Jacob Thomas. 2002. “Equity Valuation Using
Multiples.” Journal of Accounting Research, vol. 40, no. 1 (March): 135–172
(www0.gsb.columbia.edu/mygsb/faculty/research/pubfiles/318/Equity_
Valuation_Using_Multiples.pdf).
———. 2007. “Is Cash Flow King in Valuations?” Financial Analysts Journal,
vol. 63, no. 2 (March/April): 56–68.
Loomis, Carol. 2001. “Warren Buffett on the Stock Market.” Fortune
(10 December): http://archive.fortune.com/magazines/fortune/fortune_
archive/2001/12/10/314691/index.htm.
Lowe, Janet. 2010. The Triumph of Value Investing: Smart Money Tactics for the
Postrecession Era. New York: Penguin.
Lowry, J.R. 2016. “Top 400: Disruptive Technology—The Four ‘Ds’ of
Disruption.” IPE Magazine (June): www.ipe.com/reports/special-reports/
top-400-asset-managers/top-400-disruptive-technology-the-four-ds-of-
disruption/10013553.fullarticle.
Lubik, Thomas A., and Christian Matthes. 2015. “Calculating the Natural
Rate of Interest: A Comparison of Two Alternative Approaches.” Federal
Reserve Bank of Richmond Economic Brief (October): https://www.
richmondfed.org/-/media/richmondfedorg/publications/research/economic_
brief/2015/pdf/eb_15-10.pdf.
Macquarie Equities Research. 2013a. “Camouflaged in Complexity: Using
Textual Analysis to Extract Signals from 10-K Reports.” Quantamentals
(6 February).
———. 2013b. “The Price Is Right.” Quantamentals (29 October): https://
www.ravenpack.com/research/quantamentals-the-price-is-right/.
———. 2014a. “A Surprising Tone.” Quantamentals (10 July).

Equity Valuation
———. 2014b. “How Are You Really Feeling?” Quantamentals (12 November).
———. 2015. “I Just Called to Say I’m Bullish.” Quantamentals (20 April).
Malkiel, Burton G. 1973 (10th ed., 2012). A Random Walk Down Wall Street:
The Time-Tested Strategy for Successful Investing. New York: Norton.
Man Group. 2014. “Is Momentum Behavioural?” AHL/MSS Academic
Advisory Board group discussion (March): www.man.com/is-momentum-
behavioural.
Marenzi, Octavio. 2017. “Alternative Data—The New Frontier in Asset
Management.” Opimas (31 March): www.opimas.com/research/217/detail/.
Mariathasan, Joseph. 2016. “Non-Traditional Investment: Quant versus
Traditional.” IPE Magazine (January): www.ipe.com/reports/special-reports/
active-management/non-traditional-investment-quant-versus-traditional/
10011326.article.
Mcintyre, Dean. 2016. “Management Performance: Good Behavior or
Good Luck?” FactSet Insight (22 November): https://insight.factset.com/
manager-performance-good-behavior-or-good-luck.
McKinsey Private Equity and Principal Investors Practice. 2017.
“McKinsey Global Private Markets Review: A Routinely Exceptional Year.”
McKinsey & Company (February): www.mckinsey.com/~/media/McKinsey/
Industries/Private%20Equity%20and%20Principal%20Investors/Our%20Insights/
A%20routinely%20exceptional%20year%20for%20private%20equity/
McKinsey-Global-Private-Markets-Review-February-2017.ashx.
McLeay, Michael, Amar Radia, and Ryland Thomas. 2014a. “Money in the
Modern Economy: An Introduction.” Quarterly Bulletin, Monetary Analysis
Directorate of the Bank of England (Q1 2014): www.bankofengland.co.uk/
publications/Documents/quarterlybulletin/2014/qb14q101.pdf.
———. 2014b. “Money Creation in the Modern Economy.” Quarterly
Bulletin, Monetary Analysis Directorate of the Bank of England (Q1 2014):
www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2014/
qb14q102.pdf.
Merton, Robert C. 1974. “On the Pricing of Corporate Debt: The Risk
Structure of Interest Rates.” Journal of Finance, vol. 29, no. 2 (May): 449–470.
Miller, Edward M. 1977. “Risk, Uncertainty, and Divergence of Opinion.”
Journal of Finance, vol. 32, no. 4 (September): 1151–1168.

References
Montier, James, and Philip Pilkington. 2016. “The Stock Market as Monetary
Policy Junkie: Quantifying the Fed’s Impact on the S&P 500.” GMO white
paper (March).
Moore, Basil. 1988. Horizontalists and Verticalists: The Macroeconomics of Credit
Money. Cambridge, UK: Cambridge University Press.
Moore, Charlotte. 2016. “Time to Become More Active?” IPE Magazine
(January): www.ipe.com/reports/special-reports/active-management/time-to-
become-more-active/10011325.article.
Moreolo, Carlo Svaluto. 2016. “Pension Funds: What Role for Active
Management?” IPE Magazine (January): www.ipe.com/reports/special-reports/
active-management/pension-funds-what-role-for-active-management/
10011322.article.
Morningstar Manager Research. 2017. “Morningstar Direct Asset Flows
Commentary: United States.” Morningstar (11 January): https://corporate.
morningstar.com/US/documents/AssetFlows/AssetFlowsJan2017.pdf.
Nath, Trevir. 2016. “Public vs. Private Tech Valuations: What’s Driving
the Divide?” Investopedia (5 February): www.investopedia.com/articles/
investing/020516/public-vs-private-tech-valuations-whats-driving-divide.asp#
ixzz4O5wesPJO.
Nesbitt, Stephen L. 2016. “An Examination of State Pension Performance: 2006
to 2015” (6 September): www.cliffwater.com/Cliffwater%20Research%20-%
20An%20Examination%20of%20State%20Pension%20Performance%20
2006-2015.pdf.
Nolen Foushee, Susan, Tim Koller, and Anand Mehta. 2012. “Why Bad
Multiples Happen to Good Companies.” McKinsey Quarterly (May): www.
mckinsey.it/idee/why-bad-multiples-happen-to-good-companies.
O’Dea, Christopher. 2016. “Valuations: Emerging from the Wings.” IPE
Magazine (October): www.ipe.com/investment/investing-in/global-equities/
valuations-emerging-from-the-wings/10015442.article.
O’Neill, Jim, Anna Stupnytska, and James Wrisdale. 2011. “Linking GDP
Growth and Equity Returns.” Monthly Insights, Goldman Sachs Asset
Management (May): http://s3.amazonaws.com/zanran_storage/www2.gold-
mansachs.com/ContentPages/2509459477.pdf.

Equity Valuation
Paleari, Stefano, Andrea Signori, and Silvio Vismara. 2014. “How Do

Underwriters Select Peers When Valuing IPOs?” Financial Management,
vol. 43, no. 4 (Winter): 731–755.
Penman, Stephen. 2016. “Valuation: The State of the Art.” Schmalenbach
Business Review, vol. 17, no. 1 (April): 3–23.
Pierron, Axel. 2017. “Artificial Intelligence in Capital Markets: The Next
Operational Revolution.” Opimas (1 March): www.opimas.com/research/210/
detail/.
Pilkington, Philip. 2014. “Endogenous Money and the Natural Rate of
Interest: The Reemergence of Liquidity Preference and Animal Spirits in
the Post-Keynesian Theory of Capital Markets.” Working Paper 817, Levy
Economics Institute of Bard College (September): www.levyinstitute.org/
pubs/wp_817.pdf.
Pinto, Jerald E., Thomas R. Robinson, and John D. Stowe. 2015. “Equity
Valuation: A Survey of Professional Practice.” Working paper, CFA Institute
(September): https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2657717.
“Private Equity Valuations: Best Practices and Pitfalls.” 2015. Grant Thornton
survey (www.grantthornton.com/~/media/content-page-files/private-equity/
pdfs/2015/150331-PE-Valuations-whitepaper-150408-FINALB.ashx).
Purnanandam, Amiyatosh K., and Bhaskaran Swaminathan. 2004. “Are
IPOs Really Underpriced?” Review of Financial Studies, vol. 17, no. 3 (July):
811–848.
“Q&A: Valuations for the Private Equity Industry.” 2014. Financier Worldwide
Magazine, Special Report: Private Equity (September): www.financierworld-
wide.com/qa-valuations-for-the-private-equity-industry/#.WNJBQPk1_IU.
Rhodes-Kropf, M., David T. Robinson, and S. Viswanathan. 2005.
“Valuation Waves and Merger Activity: The Empirical Evidence.” Journal of
Financial Economics, vol. 77, no. 3 (September): 561–603.
Rich, Bryan. 2016. “What Warren Buffett Thinks about Stock Valuations.”
Forbes (26 May): www.forbes.com/sites/bryanrich/2016/05/26/what-warren-
buffett-thinks-about-stock-valuations/#6623dcdf1f4b.
Ritter, Jay R. 2005. “Economic Growth and Equity Returns.” Pacific-Basin
Finance Journal, vol. 13, no. 5 (November): 489–503 (https://site.warrington.
uf l.edu/ritter/files/2015/04/Economic-growth-and-equity-returns-2005.
pdf).

References
Robeco Institutional Asset Management. 2016. “5-Year Expected Returns,

2018–2022” (www.robeco.com/en/themes/expected-returns/index.html).
Roosenboom, Peter. 2012. “Valuing and Pricing IPOs.” Journal of Banking &
Finance, vol. 36, no. 6 (June): 1653–1664.
Savi, Raffaele, Bill MacCartney, Bradley J. Betts, and Jeff Shen. 2015.
“Finding Big Alpha in Big Data: The Evolution of Active Investing.”
BlackRock white paper (September): www.blackrock.com/institutions/en-
axj/literature/market-commentary/finding-big-apha-in-big-data-axj.pdf.
Savi, Raffaele, and Jeff Shen. 2015. “Constant Change, Consistent Alpha:
The Innovation Challenge for Active Investors.” BlackRock white paper
(October): www.blackrock.com/institutions/en-us/literature/whitepaper/blk-
constant-change-consistent-alpha.pdf.
Sheridan, Ben, Brent Beardsley, Martin Ouimet, and Elias Baltassis. 2016.
“How Asset Managers Can Succeed with Advanced Analytics.” Boston
Consulting Group (25 July): www.bcgperspectives.com/content/articles/
f inancial-institutions-technology-digital-asset-managers-can-succeed-
advanced-analytics/.
Shleifer, Andrei, and Robert W. Vishny. 1997. “The Limits of Arbitrage.”
Journal of Finance, vol. 52, no. 1 (March): 35–55.
Siblis Research. n.d. “Global Market Cap to GNI/GDP Ratios for 28
Countries” (http://siblisresearch.com/data/market-cap-to-gdp-ratios/).
Siegel, Jeremy J. 2016. “The Shiller CAPE Ratio: A New Look.” Financial
Analysts Journal, vol. 72, no. 3 (May/June): 41–50.
Singh, Mohini, and Sandra Peters. 2016. Data and Technology: Transforming
the Financial Information Landscape. Charlottesville, VA: CFA Institute
(June): www.cfapubs.org/doi/pdf/10.2469/ccb.v2016.n7.1.
Slager, Alfred. 2017. “The Trustee Perspective: Active Manager Still Has a
Place.” IPE Magazine (3 February): www.ipe.com/reports/special-reports/
active-management/the-trustee-perspective-active-manager-still-has-a-
place/10017401.article.
Soe, Aye M., and Ryan Poirier. 2016. “Does Past Performance Matter?
The Persistence Scorecard.” S&P Dow Jones Research (December): https://
us.spindices.com/documents/spiva/persistence-scorecard-december-2016.pdf.

Equity Valuation
Sorensen, Eric H. 2017. “Investment Insight: Smart Data, Big Beta and the
Evolving Land of Quant.” PanAgora (May): https://publishing.dealogic.com/
Nomura/PanAgora.pdf.
Sorensen, Eric, and David Williamson. 1985. “Some Evidence on the Value
of Dividend Discount Models.” Financial Analysts Journal, vol. 41, no. 6
(November/December): 60–69.
S&P. 2017. “S&P 500 Buybacks Total $135.3 Billion for Q4 2016, Decline
for Full-Year 2016.” Standard & Poor’s (22 March): http://us.spindices.com/
documents/index-news-and-announcements/20170322-sp-500-buybacks-
q4-2016-pr.pdf.
Straehl, Philip U., and Roger G. Ibbotson. 2015. “The Supply of Stock
Returns: Adding Back Buybacks.” Morningstar working paper (17
December): http://corporate1.morningstar.com/WorkArea/DownloadAsset.
aspx?id=13346.
Tergesen, Anne, and Jason Zweig. 2016. “The Dying Business of
Picking Stocks.” Wall Street Journal (17 October): www.wsj.com/articles/
the-dying-business-of-picking-stocks-1476714749.
“Valuing Private Companies.” 2016. Investopedia (16 November): www.
investopedia.com/articles/fundamental-analysis/11/valuing-private-compa-
nies.asp#ixzz4c2kLLaTK.
Wang, Lu. 2016a. “There’s Only One Buyer Keeping S&P 500’s Bull
Market Alive.” Bloomberg (14 March): www.bloomberg.com/news/
articles/2016-03-14/there-s-only-one-buyer-keeping-the-s-p-500-s-bull-
market-alive.
———. 2016b. “Bull Market Losing Big Ally as Buybacks Fall Most since
2009.” Bloomberg (16 May): www.bloomberg.com/news/articles/2016-05-16/
bull-market-losing-biggest-ally-as-buybacks-fall-most-since-2009.
Werner, Richard. 2012. “The Quantity Theory of Credit and Some of Its
Applications.” Working paper, Robinson College, University of Cambridge.
Wicksell, Knut. 1898. Geldzins und Güterpreise (first English ed.: Interest and
Prices, 1936). London: Macmillan (available as a PDF file or ebook from the
Ludwig von Mises Institute, https://mises.org/library/interest-and-prices).
Williams, John Burr. 1938. The Theory of Investment Value. Cambridge, MA:
Harvard University Press.

References
World Bank. 2015. “Market Capitalization of Listed Domestic Companies

(% of GDP).” World Federation of Exchanges database (http://data.world-
bank.org/indicator/CM.MKT.LCAP.GD.ZS?end=2015&locations=US&sta
rt=1975&view=chart).
Xiong, James X., Roger G. Ibbotson, Thomas M. Idzorek, and Peng Chen.
2010. “The Equal Importance of Asset Allocation and Active Management.”
Financial Analysts Journal, vol. 66, no. 2 (March/April): 22–30.
Yardeni, Edward. 2017. “Bull by the Tail.” Dr. Ed’s Blog (5 April): http://blog.
yardeni.com/2017/04/bull-by-tail.html.
Zörgiebel, Severin Johannes. 2016a. “Valuation of IPOs with Negative
Earnings.” Working paper, Goethe University Frankfurt (June).
———. 2016b. “The Rise of the Unicorns: How Media Affects Start-Up
Valuations.” Working paper, Goethe University Frankfurt (July): https://
papers.ssrn.com/sol3/papers.cfm?abstract_id=2808458.

The CFA Institute
Research Foundation
Board of Trustees
2017–2018
Chair JT Grier, CFA* Fred Lebel, CFA
Joachim Klement, CFA Virginia Retirement HFS Hedge Fund
Fidante Partners System Selection S.A.
Ted Aronson, CFA Joanne Hill Mauro Miranda, CFA

AJO, LP ProShares CFA Society Brazil
Jeffery V. Bailey, CFA* George R. Hoguet, CFA Sophie Palmer, CFA
Tonka Bay, MN Chesham Investments, Jarislowsky Fraser
LLC
Heather Brilliant, CFA Paul Smith, CFA
First State Investments Jason Hsu, PhD CFA Institute
Rayliant Global Advisors
Bill Fung, PhD
Aventura, FL Vikram Kuriyan, PhD, CFA
Indian School of
Diane Garnick Business
TIAA
*Emeritus
Officers and Directors

Executive Director Secretary
Bud Haslett, CFA Jessica Critzer
CFA Institute CFA Institute
Gary P. Brinson Director of Research Treasurer
Laurence B. Siegel Kim Maynard
Blue Moon Communications CFA Institute
Research Foundation Review Board

William J. Bernstein Paul D. Kaplan, CFA Krishna Ramaswamy
Efficient Frontier Morningstar, Inc. University of
Advisors Pennsylvania
Robert E. Kiernan III
Elroy Dimson Advanced Portfolio Andrew Rudd
London Business School Management Advisor Software, Inc.
Stephen Figlewski Andrew W. Lo Stephen Sexauer

New York University Massachusetts Institute Allianz Global Investors
of Technology Solutions
William N. Goetzmann
Yale School of Alan Marcus Lee R. Thomas
Management Boston College Pacific Investment
Management Company
Elizabeth R. Hilpman Paul O’Connell
Barlow Partners, Inc. FDO Partners

Named Endowments
The CFA Institute Research Foundation acknowledges with sincere gratitude the
generous contributions of the Named Endowment participants listed below.
Gifts of at least US$100,000 qualify donors for membership in the Named Endow-
ment category, which recognizes in perpetuity the commitment toward unbiased,
practitioner-oriented, relevant research that these firms and individuals have ex-
pressed through their generous support of the CFA Institute Research Foundation.
Ameritech Miller Anderson & Sherrerd, LLP

Anonymous Nikko Securities Co., Ltd.
Robert D. Arnott Nippon Life Insurance Company of
Theodore R. Aronson, CFA Japan
Asahi Mutual Life Insurance Company Nomura Securities Co., Ltd.
Batterymarch Financial Payden & Rygel
Management Provident National Bank
Boston Company Frank K. Reilly, CFA
Boston Partners Asset Management, Salomon Brothers
L.P. Sassoon Holdings Pte. Ltd.
Gary P. Brinson, CFA Scudder Stevens & Clark
Brinson Partners, Inc. Security Analysts Association
Capital Group International, Inc. of Japan
Concord Capital Management Shaw Data Securities, Inc.
Dai-Ichi Life Insurance Company Sit Investment Associates, Inc.
Daiwa Securities Standish, Ayer & Wood, Inc.
Mr. and Mrs. Jeffrey Diermeier State Farm Insurance Company
Gifford Fong Associates Sumitomo Life America, Inc.
Investment Counsel Association T. Rowe Price Associates, Inc.
of America, Inc. Templeton Investment Counsel Inc.
Jacobs Levy Equity Management Frank Trainer, CFA
John A. Gunn, CFA Travelers Insurance Co.
John B. Neff, CFA USF&G Companies
Jon L. Hagler Foundation Yamaichi Securities Co., Ltd.
Long-Term Credit Bank of Japan, Ltd.
Lynch, Jones & Ryan, LLC
Meiji Mutual Life Insurance
Company
Senior Research Fellows
Financial Services Analyst Association
For more on upcoming Research Foundation
publications and webcasts, please visit
www.cfainstitute.org/learning/foundation.
Research Foundation monographs
are online at www.cfapubs.org.

RESEARCH FOUNDATION
CONTRIBUTION FORM
✓Yes, I want the Research Foundation to continue to fund innovative
research that advances the investment management profession. Please
accept my tax-deductible contribution at the following level:
Thought Leadership Circle..................... US$1,000,000 or more
Named Endowment....................... US$100,000 to US$999,999
Research Fellow............................... US$10,000 to US$99,999
Contributing Donor. . ............................US$1,000 to US$9,999
Friend........................................................... Up to US$999
I would like to donate US$______________________.
My check is enclosed (payable to the CFA Institute Research Foundation).

I would like to donate appreciated securities (send me information).
Please charge my donation to my credit card.
VISA MC Amex Diners
Card Number
—— /——
Expiration Date
_______________________________
Name on card P L E A S E P R I N T
Corporate Card
Personal Card __________________________________
Signature
This is a pledge. Please bill me for my donation of US$______________

I would like recognition of my donation to be:
Individual donation Corporate donation Different individual
____________________________________________________
PLEASE PRINT NAME OR COMPANY NAME AS YOU WOULD LIKE IT TO APPEAR
PLEASE PRINT Mr. Mrs. Ms. MEMBER NUMBER_______________________________
Last Name (Family Name) First (Given Name) Middle Initial
Title
Address
City State/Province Country ZIP/Postal Code
Please mail this completed form with your contribution to:

The CFA Institute Research Foundation • P.O. Box 2082
Charlottesville, VA 22902-2082 USA
For more on the CFA Institute Research Foundation, please visit

www.cfainstitute.org/learning/foundation/Pages/index.aspx.

ISBN 978-1-944960-33-9
Available online at www.cfapubs.org 9 781944 960339

SSRN Id3254580 PDF

Uploaded by

Copyright:

Available Formats

SSRN Id3254580 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SSRN Id3254580 PDF

Uploaded by

Copyright:

Available Formats

EQUITY VALUATION

Science, Art, or Craft?

Joanne M. Hill, Dave Nadig, Matt Hougan

Electronic copy available at: https://ssrn.com/abstract=3254580

Frank J. Fabozzi, CFA, Sergio M. Focardi, and Caroline Jonas

Electronic copy available at: https://ssrn.com/abstract=3254580

The CFA Institute Research Foundation is a

Electronic copy available at: https://ssrn.com/abstract=3254580

Frank J. Fabozzi, CFA, is professor of finance at EDHEC Business School

© 2017 CFA Institute Research Foundation. All rights reserved.  iii

Electronic copy available at: https://ssrn.com/abstract=3254580

1. Finance Theory and Equity Valuation.................................................. 1

This publication qualifies for 2 CE credits under the guide-

Electronic copy available at: https://ssrn.com/abstract=3254580

© 2017 CFA Institute Research Foundation. All rights reserved.  vii

Electronic copy available at: https://ssrn.com/abstract=3254580

© 2017 CFA Institute Research Foundation. All rights reserved.  ix

Electronic copy available at: https://ssrn.com/abstract=3254580

In a 2003 article, 2 Barton Waring and I described active managers as

x © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

In this, their fifth research monograph, the authors investigate how

© 2017 CFA Institute Research Foundation. All rights reserved.  xi

Electronic copy available at: https://ssrn.com/abstract=3254580

© 2017 CFA Institute Research Foundation. All rights reserved.  xiii

Electronic copy available at: https://ssrn.com/abstract=3254580

xiv © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

“Any valuation must be in accord with the well-established theory of finance.”

© 2017 CFA Institute Research Foundation. All rights reserved.  1

Electronic copy available at: https://ssrn.com/abstract=3254580

Because of the difficulty in determining a stock’s true value, Graham rec-

2 © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

they do so on the Zimbabwe stock exchange. There is likely some degree of

© 2017 CFA Institute Research Foundation. All rights reserved.  3

Electronic copy available at: https://ssrn.com/abstract=3254580

4 © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

characteristics, competition assigns a market price to goods and services in

© 2017 CFA Institute Research Foundation. All rights reserved.  5

Electronic copy available at: https://ssrn.com/abstract=3254580

In the absence of arbitrage, the price of a financial asset is the sum of

6 © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

Present value clearly depends on interest rates.

© 2017 CFA Institute Research Foundation. All rights reserved.  7

Electronic copy available at: https://ssrn.com/abstract=3254580

Figure 1.1. Fluctuations of Three-Month T-Bill Rates, 1934–2016

8 © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

Doing so requires determining the intrinsic value of a stock, which in turn

© 2017 CFA Institute Research Foundation. All rights reserved.  9

Electronic copy available at: https://ssrn.com/abstract=3254580

10 © 2017 CFA Institute Research Foundation. All rights reserved.

Electronic copy available at: https://ssrn.com/abstract=3254580

The conceptual underpinning of intrinsic value as outlined here shows

© 2017 CFA Institute Research Foundation. All rights reserved.  11

Electronic copy available at: https://ssrn.com/abstract=3254580

© 2017 CFA Institute Research Foundation. All rights reserved. iii

© 2017 CFA Institute Research Foundation. All rights reserved. vii

© 2017 CFA Institute Research Foundation. All rights reserved. ix

x © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. xi

© 2017 CFA Institute Research Foundation. All rights reserved. xiii

xiv © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 1

2 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 3

4 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 5

6 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 7

8 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 9

10 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 11

12 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 13

14 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 15

16 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 17

18 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 19

20 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 21

Table 2.2. Most Widely Used Market Multiples Approach among Respondents

22 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 23

24 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 25

26 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 27

28 © 2017 CFA Institute Research Foundation. All rights reserved.

© 2017 CFA Institute Research Foundation. All rights reserved. 29

30 © 2017 CFA Institute Research Foundation. All rights reserved.

Figure 2.2. US Nonfinancial Companies Grouped by Comparable Revenue Growth

© 2017 CFA Institute Research Foundation. All rights reserved. 31

32 © 2017 CFA Institute Research Foundation. All rights reserved.