Strategic Management Journal, Vol.

9, 41-58 (1988)

7FIRST-MOVER ADVANTAGES
\MARVIN B. LIEBERMAN and DAVID B. MONTGOMERY
Graduate School o f Business, S t a n f o r d University, S t a n f o r d , California, U.S.A.

This article surveys the theoretical and empirical literature on m e c h a n i s m that confer
advantages and disadvantages on first-mover firms. Major conceptual issues are addressed,
and recommendations are given f o r future research. Managerial implications are also
discussed.

INTRODUCTION possesses some unique resources or foresight, or

simply because of luck. Once this asymmetry is
What, exactly, are first-mover advantages? Under generated a variety of mechanisms may enable
what conditions do they arise, and by what the firm to exploit its position; these mechanisms
specific mechanisms? Do first-movers make enhance the magnitude or durability (or both)
above-average profits? And when is it in a firm’s of first-mover profits.
interest to pursue first-mover opportunities, as Our discussion is organized as follows. We
opposed to allowing rivals to make the pioneering first consider theoretical models and empirical
investments? evidence on three general categories in which
In this paper we examine these and other first-mover advantage can be attained: leadership
related questions. We categorize mechanisms that in product and process technology, preemption
confer advantages and disadvantages on first- of assets, and development of buyer switching
mover firms, and critically assess the relevant costs. We then examine potential disadvantages
theoretical and empirical literature. The recent of first-mover firms (or conversely, relative
burgeoning of theoretical work in industrial advantages enjoyed by late-mover rivals). These
economics provides a rich set of models that help include free-rider problems and a tendency
make understanding of first-mover advantages toward inertia or sluggish response by established
more precise. There is also a growing body of incumbents. The next section addresses a series
empirical literature on order-of-entry effects. Our of basic conceptual issues. These include the
aim is to begin to provide a more detailed endogenous nature of first-mover opportunities,
mapping of mechanisms and outcomes, to serve and various definitional and measurement ques-
as a guide for future research. tions. We conclude with an assessment of
We define first-mover advantages in terms of opportunities for additional research and a
the ability of pioneering firms to earn positive discussion of managerial implications.
economic profits (i.e. profits in excess of the
cost of capital). First-mover advantages arise
endogenously within a multi-stage process, as MECHANISMS LEADING TO FIRST-
illustrated in Figure 1. In the first stage some MOVER ADVANTAGES
asymmetry is generated, enabling one particular
firm to gain a head start over rivals. This first- First-mover advantages arise from three primary
mover opportunity may occur because the firm sources: (1) technological leadership, (2) preemp-
0143-2095/88/050041-18$09.00
01988 by John Wiley & Sons, Ltd.
42 M . B. Lieberman and D. B. Montgomery
i

I Profits I
Figure 1. Endogenous generation of first-mover
advantages.
tion of assets, and (3) buyer switching costs. able leadership in technology. Two basic mechan-
Within each category there are a number of isms are considered in the literature: (1) advan-
specific mechanisms.’ In this section we survey tages derived from the ‘learning’ or ‘experience’
existing theoretical and empirical literature on curve, where costs fall with cumulative output,
these three general categories of first-mover and (2) success in patent or R&D races, where
advantages. advances in product or process technology are a
The theoretical models surveyed in this section function of R&D expenditures.
assume the existence of some initial asymmetry
among competitors that can be exploited by the Learning curve
first-mover firm. This initial asymmetry is critical; In the standard learning-curve model, unit pro-
without it, first-mover advantages do not arise. duction costs fall with cumulative output. This
Ways in which this asymmetry may come about generates a sustainable cost advantage for the
are considered later in the paper. early entrant if learning can be kept proprietary
and the firm can maintain leadership in market
Technological leadership share. This argument was popularized by the
First-movers can gain advantage through sustain- Boston Consulting Group during the 1970s and
has had a considerable influence on the strategic
management field.
’ Rumelt (1987) refers to these as ‘isolating mechanisms’,
In a seminal theoretical paper, Spence (1981)
since they protect ‘entrepreneurial rents’ from imitative
competition. demonstrated that when learning can be kept
 First-mover Advantages 43
proprietary, the learning curve can generate R B D and patents
substantial barriers to entry. Fewer than a
handful of firms may be able to compete When technological advantage is largely a function
profitably.2 Despite high seller concentration, of R&D expenditures, pioneers can gain advan-
incentives for vigorous competition remain. Firms tage if technology can be patented or maintained
that do enter may initially sell below cost in an as trade secrets. This has been formalized in the
effort to accumulate greater experience, and theoretical economics literature in the form of
thereby gain a long-term cost advantage. Such R&D or patent-races where advantages are often
competition sharply reduces profits. enjoyed by the first-mover firm. Gilbert and
Empirical evidence of learning-based preemp- Newbery (1982) were the first to develop a model
tion is given by Ghemawat (1984) in the case of of preemptive patenting, in which a firm with an
DuPont’s development of an innovative process early head-start in research exploits its lead
for titanium dioxide, and by Porter (1981), to deter rivals from entering the patent-race.
who discusses Procter and Gamble’s sustained Subsequent papers by Reinganum (1983), Fuden-
advantage in disposable diapers in the U.S. berg et a f . (1983) and others show that successful
Similarly, Shaw and Shaw (1984) argue that late preemption by the leader depends on assumptions
entrants into European synthetic fiber markets regarding the stochastic nature of the R&D
failed to gain significant market shares or low process and on the inability of followers to
cost positions, and many ultimately exited. ‘leapfrog’ ahead of the incumbent. One general
Learning-based advantages are also evident in defect of the patent-race literature is the assump-
the case of Lincoln Electric Company (Fast, tion that all returns go exclusively to the winner
1975); the firm’s early market entry with superior of the race.
patented products, coupled with a managerial As an empirical matter, such patent-races seem
system promoting continued cost reduction in to be important in only a few industries, such as
an evolutionary technological environment, has pharmaceuticals. In most industries, patents
enabled the company to maintain high profit- confer only weak protection, are easy to ‘invent
ability for decades. around’, or have transitory value given the pace
Inter-firm diffusion of technology, which dimin- of technological change. For a sample of 48
ishes first-mover advantages derived from the patented product innovations in pharmaceuticals,
learning curve, is emphasized in theoretical chemicals and electrical products, Mansfield et
papers by Ghemawat and Spence (1985) and a f . (1981) found that, on average, imitators could
Lieberman (1987~).It is now generally recognized duplicate patented innovations for about 65
that diffusion .occurs rapidly in most industries, percent of the innovators cost; imitation was
and learning-based advantages are less wide- fairly rapid, with 60 percent of the patented
spread than was commonly believed in the 1970s. innovations limited within 4 years. Imitation
Mechanisms for diffusion include workforce appears relatively more costly in the pharmaceu-
mobility, research publication, informal technical tical industry, where imitators must go through
communication, ‘reverse engineering’, plant the same regulatory approval procedures as the
tours, etc. For a sample of firms in ten industries, innovating firm. Levin et al. (1984) found wide
Mansfield (1985) found that process technology inter-industry variation in the cost and time
leaks more slowly than product technology, but required for imitation. They also found inter-
competitors typically gain access to detailed industry differences in appropriability mechan-
information on both products and processes isms, with lead-time and learning curve advan-
within a year of development. Lieberman (1982, tages relatively important in many industries, and
1987b) shows that diffusion of process technology patents important in few. In a study using the
enabled late entry into a sample of 40 chemical PIMS data base, Robinson (1988) found that
product industries, despite strong learning curve pioneer firms benefit from patents or trade secrets
effects at the industry level. to a significantly greater extent than followers
(29 percent vs. 13 percent). However, he also
’ In a related setting where learning depends o n accumulated found that patents accounted for only a small
investment rather than output, Gilbert and Harris (1981)
show that a first-mover will preempt in the construction of
proportion of the perceived quality advantages
new plants over multiple generations. enjoyed by pioneers.
44 M . B. Lieberman and D. B . Montgomery

Several case studies have examined the role of turing locations. Here, the returns garnered by
patents in sustaining first-mover advantages. the first-mover are pure economic rents.’ A
Bresnahan (1985) discusses Xerox’s use of patents first-mover with superior information can (in
as an entry barrier. In addition to key patents principle) collect all such rents earned on non-
on the basic Xerography process, Xerox patented mobile assets such as resource deposits and real
a thicket of alternative technologies which e ~ t a t eThe
. ~ firm may also be able to appropriate
defended the firm from entry until challengers some of the rents that accrue to potentially
used anti-trust actions to force compulsory mobile assets such as employees, suppliers and
licensing. Bright (1949) argues that GE’s long- distributors. The firm can collect such rents if
term dominance of the electric lamp industry was these factors are bound to the firm by switching
initially derived from control of the basic Edison costs, so that their mobility is restricted.
patent, and later maintained through the accumu- One empirical study of first-mover advantages
lation of hundreds of minor patents on the lamp in controlling natural resources is Main (19%).
and associated equipment. Main argues that the concentration of high-grade
R&D and innovation need not be limited to nickel deposits in a single geographic area made
physical hardware; firms also make improvements it possible for the first company in the area to
in managerial systems and may invent new secure rights to virtually the entire supply, and
organizational forms. Organizational innovation thus dominate world production for decades.
is often slow to diffuse, and hence may convey
a more durable first-mover advantage than
product or process innovation (Teece, 1980). Preemption of locations in geographic and
Chandler (1977) describes managerial innovations product characteristics space
that enabled producers to exploit newly available
scale economies in manufacturing and distribution First-movers may also be able to deter entry
in the late nineteenth century. Many of these through strategies of spatial preemption. In many
firms--e.g. American Tobacco, Campbell Soup, markets there is ‘room’ for only a limited number
Quaker Oats, Procter and Gamble-still retain of profitable firms; the first-mover can often
dominant positions in their industries. select the most attractive niches and may be able
to take strategic actions that limit the amount
of space available for subsequent entrants.
heempiion of scarce assets Preemptable ‘space’ can be interpreted broadly
to include not only geographic space, but also
The first-mover firm may be able to gain shelf space and ‘product characteristics space’
advantage by preempting rivals in the acquisition (i.e. niches for product differentiation).
of scarce assets. Here, the first-mover gains The theory of spatial preemption is developed
advantage by controlling assets that already exist, in papers by Prescott and Visscher (1977),
rather than those created by the firm through Schmalensee (1978), Rao and Rutenberg (1979)
development of new technology. Such assets may and Eaton and Lipsey (1979, 1981). The basic
be physical resources or other process inputs. argument is that the first-mover can establish
Alternatively, the assets may relate to positioning positions in geographic or product space such
in ‘space’, including geographic space, product that latecomers find it unprofitable to occupy the
space, shelf space, etc. interstices. If the market is growing, new niches
are filled by incumbents before new entry

Preemption of input factors The basic argument is standard economic analysis, and can
be traced back to Ricardo’s analysis of rents captured by
landowners (first-movers) in the market for wheat in
If the first-mover firm has superior information, nineteenth-century England.
it may be able to purchase assets at market prices Note that, with complete markets, a first-mover with
below those that will prevail later in the evolution superior information need not actually own o r control such
assets to capture economic rents. Hirshleifer (1971) argues
of the market. Such assets include natural that if futures markets exist, the firm can simply assume
resource deposits and prime retailing or manufac- forward market positions that exploit its superior information.
 First-mover Advantages 45

becomes profitable.5 Entry is repelled through quality superiority over imitators and sub-
the threat of price warfare, which is more sequently developed advantages in the form of
intense when firms are positioned more closely. a broader product line. Thus, there is evidence
Incumbent commitment is provided through sunk that pioneers try to reinforce their early lead by
investment costs.'j filling product differentiation niches.
Empirical evidence suggests that successful
preemption through geographic space packing is Preemptive investment in plant and equipment
rare. In their study of the cement industry, Another way in which an established first-mover
Johnson and Parkman (1983) found no evidence can deter entry is through preemptive investment
of successful geographic preemption even though in plant and equipment. Here, the enlarged
structural characteristics of the industry suggest capacity of the incumbent serves as a commitment
that such strategies would be likely. In a study to maintain greater output following entry,
of local newspaper markets, Glazer (1985) found with price cuts threatened to make entrants
no difference in survival rates between first- and unprofitable. In these models the incumbent may
second-mover firms. One explanation for these successfully deter new entry, as in Spence (1977),
findings is that all firms in cement and newspaper Dixit (1980), Gilbert and Harris (1981) and Eaton
markets have similar technologies and entry and Ware (1987). Alternatively, preemptive'
opportunities, so preemptive competition for investment by the pioneer may simply deter the
preferred sites drives profits to zero. In other growth of smaller entrants, as in Spence (1979)
words, there were no initial asymmetries in timing and Fudenberg and Tirole (1983).
or information to be exploited. These investment tactics do not seem to be
One counter-example illustrating effective geo- particularly important in practice. Gilbert (1986)
graphic preemption is a case study of the Wal- argues that most industries lack the cost structure
Mart discount retailing firm (Ghemawat, 1986b). required for preemptive investment to prove
Wal-Mart targeted small southern towns located effective. Lieberman (1987a) shows that preemp-
in contiguous regions that competitors initially tive investment by incumbents was seldom suc-
found unprofitable to service. By coupling spatial cessful in deterring entry into chemical product
preemption at the retail level with an extremely industries. One exception was magnesium, where
efficient distribution network, the firm has been Dow Chemical maintained a near-monopoly
able to defend its position and earn sustained position for several decades, based largely on
high profits. investments (threatened or actual) in plant
Schmalensee's (1978) model of product space capacity (Lieberman, 1983).
preemption was developed in the context of a The role of scale economies is intentionally
lawsuit brought by the Federal Trade Commission deemphasized in the above-mentioned models of
against the three major U.S. breakfast cereal preemptive investment .8 When scale economies
companies. The FTC alleged that these firms had are large, first-mover advantages are typically
sustained their high profit rates through a strategy enhanced, with the limiting case being that of
of tacit collusion in preempting supermarket natural monopoly. However, outside of public
shelf space and product differentiation niches. utilities, scale economies approaching the natural
Although the lawsuit was dismissed, the cereal monopoly level are seldom observed in U.S.
firms have continued to sustain exceptionally manufacturing industries.Y In a theoretical treat-
high profit rates.'
Robinson and Fornell (1985) found that new We have also ignored the possibility that network externali-
ties may enhance the position of the first-mover firm. These
consumer product pioneers initially held product externalities arise if there are incentives for interconnection
or compatibility among users (see, for example, Farrell and
Saloner, 1986 and Katz and Shapiro, 1986).
Incumbents fill these niches in order to sustain monopoly For example, see Weiss (1976). This finding applies to
profits at nearby locations; these profits may be dissipated manufacturing operations only; greater scale economies may
if new entry occurs. arise in distribution and advertising. Also, many retailing
Judd (1985) argues that sunk costs are not sufficient; exit markets are geographically fragmented, leading to the
costs are required as well. possibility of spatial preemption of the sort described above.
' Of course, these profits may be derived from sources other Such preemption requires the presence of some scale
than spatial preemption. economies in the form of fixed costs.
46 M . B. Lieberman and D. B. Montgomery

ment, Schmalensee (1981) shows that in most brand they encounter that performs the job
realistic industry settings, scale economies provide satisfactorily. Brand loyalty of this sort may be
only minor entry barriers and hence potential for particularly strong for low-cost ‘convenience
enhanced profits. goods’ where the benefits of finding a superior
brand are seldom great enough to justify the
Switching costs and buyer choice under additional search costs that must be incurred
uncertainty (Porter, 1976). In such an environment, early-
mover firms may be able to establish a reputation
Switching costs for quality that can be transferred to additional
First-mover advantages may also arise from buyer products through umbrella branding and other
switching costs. With switching costs, late entrants tactics (Wernerfelt, 1987).
must invest extra resources to attract customers Similar arguments derived from the psychology
away from the first-mover firm. Several types of literature suggest that the first product introduced
switching costs can arise. First, switching costs received disproportionate attention in the cons-
can stem from initial transactions costs or umer’s mind. Late entrants must have a truly
investments that the buyer makes in adapting to superior product, or else advertise more fre-
the seller’s product. These include the time and quently (or more creatively) than the incumbent
resources spent in qualifying a new supplier, the in order to be noticed by the consumer. In
cost of ancillary products such as software for a a laboratory study using consumer products,
new computer, and the time, disruption, and Carpenter and Nakamoto (1986) found that
financial burdens of training employees. A second the order-of-entry influences the formation of
category of switching costs arises due to supplier- consumer preferences. If the pioneer is able to
specific learning by the buyer. Over time, the achieve significant consumer trial, it can define
buyer adapts to characteristics of the product the attributes that are perceived as important
and its supplier and thus finds it costly to change within a product category. Pioneers such as Coca-
over to another brand (Wernerfelt, 1985). For Cola and Kleenex have become prototypical,
example, nurses become accustomed to the occupying a unique position in the consumer’s
intravenous solution delivery systems of a given mind. Their large market shares tend to persist
supplier and are reluctant to switch (Porter, because perceptions and preferences, once for-
1980). A third type of switching cost is contractu- med, are difficult to alter.
ral switching cost that may be intentionally More traditional marketing studies confirm the
created by the seller. Airline frequent-flyer existence of such perceptual effects. In a study
programs fit in this category (Klemperer, 1986). of two types of prescription pharmaceuticals-oral
Theoretical models of market equilibrium with diuretics and antianginals-Bond and Lean (1977)
buyer switching costs include Klemperer (1986) found that physicians ignored ‘me-too’ products,
and Wernerfelt (1986, 1988). Switching costs even if offered at lower prices and with substantial
typically enhance the value of market share marketing support. ‘(I Montgomery (1975) found
obtained early in the evolution of a new market. that a product’s newness was one of the two key
Thus they provide a rationale for pursuit of market variables necessary to gain acceptance onto
share. However, first-movers with large market supermarket shelves.
shares do not necessarily earn high profits; early These imperfect information effects should be
competition for share can dissipate profits; and greater for individual consumers than corporate
under some conditions the inertia of an incumbent buyers, since the latter’s larger purchase volume
with a large customer base can make the firm justifies greater investment in information acqui-
vulnerable to late entrants, who prove to be sition activities.l’ Using the PIMS data base,
relatively more profitable (Klemperer , 1986).
‘ “ O n e explanation of these findings is that physicians are
price-insensitive because they d o not actually pay the
Buyer choice under uncertainty prescription costs. However, the Carpenter and Nakarnoto
(1986) experiments found that more typical consumers are
A related theoretical literature (e.g. Schmalensee, also unwilling to switch to objectively similar ‘me-too’ brands,
1982) deals with the imperfect information of even at subtantially lower prices.
Moreover, switching costs in industrial markets often
buyers regarding product quality. In such a dissipate over time as buyers become more knowledgeable
context, buyers may rationally stick with the first about competing products (Cady, 1985).
 First-mover Advantages 47

Robinson (1988) and Robinson and Fornell form of information spillovers in R&D (Spence,
(1985) found that pioneers had larger market 1984; Baldwin and Childs, 1969), and learning-
shares than followers in both consumer and based productivity improvement (Ghemawat and
industrial markets, but the effect was much Spence, 1985; Lieberman, 1987~).As mentioned
greater for consumer goods-order of entry previously, empirical studies document a high
explained 18 percent of the variance in market rate of inter-firm diffusion of technology in most
share in consumer goods markets, but only 8 industries.
percent in industrial markets. For a sample of Guasch and Weiss (1980) assess free-rider
129 consumer packaged goods, Urban et al. effects operating in the labor market. They give
(1986) found a strong inverse relation between a theoretical argument that late-mover firms may
order-of-entry and market share. be able to exploit employee screening performed
Brand positions remain remarkably durable in by early entrants, and thus acquire skilled labor
many consumer markets. Ries and Trout (1986) at lower cost. This is in addition to the fact that
noted that of 25 leading brands in 1923, 20 were early entrants may invest in employee training,
still in first place some 60 years later. Davidson with benefits enjoyed by later entrants who may
(1976) found that two-thirds of the pioneers in be able to hire away the trained personnel.
18 United Kingdom grocery product categories Teece (1986a,b) argues that the magnitude of
developed since 1945 retained their market free-rider effects depends in part on ownership of
leadership through the mid-1970s. assets that are complementary or ‘co-specialized’
with the underlying innovation. For example,
EM1 developed the first C T scanner but lost in
FIRST-MOVER DISADVANTAGES the marketplace because the firm lacked a
technology infrastructure and marketing base in
The mechanisms that benefit the first-mover may the medical field. Pilkington, by comparison, was
be counterbalanced by various disadvantages. able to profit handsomely from its pioneering
These first-mover disadvantages are, in effect, float glass process due to its assets and experience
advantages enjoyed by late-mover firms. Late- in the glass industry. In other instances late-
movers may benefit from: (1) the ability to ‘free- mover firms have been successful largely because
ride’ on first-mover investments, (2) resolution they were able to exploit existing assets in areas
of technological and market uncertainty, (3) such as marketing, distribution, and customer
technological discontinuities that provide ‘gate- reputation-e.g. IBM in personal computers and
ways’ for new entry, and (4) various types of Matsushita in VCRs (Schnaars, 1986).
‘incumbent inertia’ that make it difficult for the
incumbent to adapt to environmental change.
Resolution of technological or market
These phenomena can reduce, or even completely
uncertainty
negate, the net advantage of the incumbent
derived from mechanisms considered previously. Late-movers can gain an edge through resolution
of market or technological uncertainty. ’* Werner-
felt and Karnani (1987) consider the effects of
Free-rider effects
uncertainty on the desirability of early versus late
Late-movers may be able to ‘free-ride’ on a market entry. Entry in an uncertain markzt
pioneering firm’s investments in a number of obviously involves a high degree of risk. They
areas including R&D, buyer education, and argue that early entry is more attractive when
infrastructure development. As mentioned pre- the firm can influence the way that uncertainty
viously, imitation costs are lower than innovation is resolved. For example, the firm may be able
costs in most industries. However, innovators to set industry standards in its favor. Firm size
enjoy an initial period of monopoly that is not
available to imitator firms. The ability of follower ,z A related point is that a late-mover may be to take
firms to free-ride reduces the magnitude and advantage of the first-mover’s mistakes. For example, when
durability of the pioneer’s profits, and hence its Toyota was first planning to enter the U.S. market it
incentive to make early investments. interviewed owners of Volkswagens, the leading small car at
that time. Information on what owners liked and disliked
The theoretical literature has focused largely about the Volkswagen was incorporated in the design process
on the implications of free-rider effects in the for the new Toyota.
48 M . B. Lieberman and D. B. Montgomery

may also be important-large firms may be better Honeywell, IBM and Burroughs, all of whom
equipped to wait for resolution of uncertainty, offered computer systems to meet the emerging
or to hedge by maintaining a more flexible need for electronic funds transfer (Abell, 1978).
investment portfolio.
In many new product markets, uncertainty is
Incumbent inertia
resolved through the emergence of a ‘dominant
design’. The Model T Ford and the DC-3 are Vulnerability of the first-mover is often enhanced
examples of dominant designs in the automotive by ‘incumbent inertia’. Such inertia can have
and aircraft industries. After emergence of such several root causes: (1) the firm may be locked-
a design, competition often shifts to price, thereby in to a specific set of fixed assets, (2) the firm
conveying greater advantage over firms possessing may be reluctant to cannibalize existing product
skills in low-cost manufacturing (Teece, 1986b). lines, or (3) the firm may become organizationally
inflexible. These factors inhibit the ability of the
firm to respond to environmental change or
Shifts in technology or customer needs
competitive threats.
Schumpeter (1961) conceived of technological Incumbent inertia is often a rational, profit-
progress as a process of ‘creative destruction’ in maximizing response, even though it may lead
which existing products are superseded by the to organizational decline. For example, Tang
innovations of new firms. New entrants exploit (1988) presents a model that rationalizes the
technological discontinuities to displace existing decisions of most U.S. steel producers to continue
incumbents. Empirical studies which consider investing in open-hearth furnace technology even
these technological discontinuities or ‘gateways’ after it had become clear that basic oxygen
for new entry include Yip (1982) and Bevan furnaces were superior. A firm with heavy sunk
(1974). Foster (1986) gives practical advice on costs in fixed plant or marketing channels that
how such discontinuities can be exploited by ultimately prove sub-optimal may find it rational
entrants, who might be defined as ‘first-movers’ to ‘harvest’ these investments rather than attempt
into the next technological phase. Scherer (1980) to transform itself radically. l 3 MacMillan (1983)
provides a list of innovative entrants who suggests that in the rapidly changing environment
revolutionized existing industries with new prod- of health care, old health care systems may
ucts and processes. He also cites numerous currently be harvesting from their initial invest-
examples of dominant incumbents that proved ments in locations and personnel. The appropriate
slow innovators but aggressive followers. choice between adaptation and harvesting
Since the replacement technology often appears depends on how costly it is t o convert the firm’s
while the old technology is still growing, it may existing assets to alternative uses. And, as we
be difficult for an incumbent to perceive the discuss below, organizational inertia has often
threat and ta$e adequate preventative steps.
I led firms to continue investing in their existing
C o 9 e r and Schendel (lQ76) provide several asset base well beyond the point where such
examples, such as the failure of steam locomotive investments are economically justified.
mariufacturers 10 respond to the invention of Much of the literature on cannibalization-
diesel. Foster (1986) cites American Viscose’s avoidance refers to R&D. Arrow (1962) was the
failure to recognize the potential of polyester first to lay out the theoretical argument that an
as a replacement for rayon, and Transitron’s incumbent monopolist is less likely to innovate
inattention to silicon as a substitute for germanium than a new entrant, since innovation destroys
in semiconductor fabrication. This perceptual rents on the firm’s existing products. More recent
failure is closely related to ‘incumbent inertia’ theoretical studies along these lines include
considered below. Reinganum (1983) and Ghemawat (1986a). Bres-
Customer needs are also dynamic, creating nahan (1985) argues that Xerox exhibited such
opportunities for later entrants unless the first- behavior following the expiration of its patent-
mover is alert and able to respond. Docutel, as enforced monopoly-Xerox lagged in certain
the pioneer, supplied virtually all of the automatic types of innovations and was sluggish to cut
teller machine market up to late 1974. Over the
l3 For first-movers, sunk costs are a two-edged sword: they
next 4 years its market share declined to lock the firm into a particular course of action but also
less than 10 percent under the onslaught of provide commitment value that can help deter entry.
 First-mover Advantages 49

prices on account of its large fleet of rental endogenously, through the process illustrated in
machines in the field. Brock (1975) and Ghem- Figure 1. A firm gains first-mover opportunities
awat (1986a) make similar arguments regarding through some combination of proficiency and
the innovative responses of IBM in computers luck. Various types of proficiency may be
and AT&T in PBX’s. However, Conner (1988) involved, including technological foresight, per-
shows that under a broad range of conditions the ceptive market research, or skillful product or
incumbent’s optimal strategy is to develop an process development. For example, Procter and
improved product but delay market introduction Gamble’s initial lead in disposable diapers can
until challenged by the appearance of a rival be attributed to most of these proficiency factors.
product. In addition to generating first-mover oppor-
From an organizational theory perspective, tunities, proficiency and luck also influence
Hannan and Freeman (1984) outline factors that the firm’s success in exploiting the specific
limit adaptive response by incumbents. These mechanisms discussed earlier. In the case of
include the development of organizational rou- disposable diapers, Procter and Gamble’s initial
tines and standards, internal political dynamics lead was enhanced by its skill in maintaining a
and the development of stable exchange relations proprietary learning curve in manufacturing, and
with other organizations. l 4 in preempting supermarket shelf space.
While Hannan and Freeman argue that organi- Proficiency and luck also affect profits in ways
zational inertia is often a positive outcome that are unrelated to first-mover advantages.
of selection processes, numerous dysfunctional For instance, some component of Procter and
examples of such inertia can be cited. Abernathy Gamble’s disposable diaper profits can be traced
and Wayne (1974) assess Henry Ford’s decision to the firm’s economies of shared distribution
to persist in production of the Model T , long channels and general manufacturing proficiency.
after changes in the competitive environment had The recent increase in the U.S. birth rate has
made it clear that new products were required. augmented industry profits, representing the
Bevan (1974) cites organizational blinders as a direct effect of luck.
key factor contributing to the decline of the major In the endogenous process illustrated in Figure
potato chip producer in the U.K. Historically, the 1, profits earned by first-movers are fundamen-
vast majority of chips were consumed by men in tally attributable to proficiency and luck, rather
pubs. It took the incumbent five years to realize than ‘pioneering’ per se. But as a practical matter,
that the challenger had invented a whole new it is often exceedingly difficult to distinguish
market segment-supermarket sales to women between proficiency and luck, particularly at
and children. Similarly, Jacobson and Hillkirk the stage where first-mover opportunities are
(1986) discuss Xerox’s inability to perceive the generated. Entrepreneurs often perceive ‘great
growing threat of Japanese competition in the opportunities’, many of which ultimately prove
1970s. Cooper and Schendel (1976) note that disappointing. Investors face the problem of
even when an incumbent makes a commitment distinguishing between ‘true’ and ‘false’ entre-
to change, organizational factors often sabotage preneurial vision. Even after success or failure
the effort; in their sample, 15 incumbents made is observed, disentangling the contribution of
major commitments to the new technologies but exceptional foresight or skill from that of mere
only two of them ultimately proved successful. luck is no easier.I5 We leave this difficult problem
to venture capitalists, and extremely ambitious
GENERAL CONCEPTUAL ISSUES empirical researchers.
Nevertheless, in a purely conceptual way
Endogeneity of first-mover opportunities one can usefully distinguish between these two
sources. Profits linked to first-mover opportuni-
A given firm cannot simply choose whether or
ties arising from exceptional skill or foresight can
not to pioneer; pioneering opportunities arise
be viewed as returns to superior entrepreneurship.
l4 One example of the latter is Timex’s introduction of the
disposable watch. Timex introduced the disposable watch in j5 One complication is that skill affects luck, and vice-versa.
drug stores and other mass channels, hut the Swiss watch For instance, greater skill in research increases the probability
industry was unable to follow, for fear of offending the of success in performing risky R&D. With repeated obser-
jewelry stores that were their prime mode of distribution vation the proficiency level of the firm may be revealed by
(Porter, 1980). its mean success rate.
50 M . B. Lieberman and D. B. Montgomery

Figure 1 shows that skill can also affect profits an unbiased sample earns zero economic profits.
more directly, as in situations where the firm One extension is that as the ‘riskiness’ of R&D
possesses know-how that enables it to manufac- increases, the average profit rate observed for
ture products at lower costs than its competitors. successful firms increases as well.
These returns may not always be visible as Failure is a common occurrence in practice.
company profits, as they may be captured, in For example, Mansfield (1968) found that more
part, by the employees responsible for their than one-fourth of corporate R&D projects fail
generation. to achieve their technical objectives. Davidson
Similarly, luck can affect profits directly (e.g. (1976) observed that about 70 percent of test
a factory damaged by fire) and indirectly by market brands are not expanded nationally and
influencing the quality of first-mover opportuni- hence can be considered failures. In an empirical
ties available to the firm.16 This effect of luck on study that considered the endogenous nature of
first-mover opportunities has important impli- first-mover opportunities, Boulding and Moore
cations for empirical research, since it leads to (1987) found that pioneering firms were margin-
sample selection biases that have sometimes been ally unprofitable on average.
overlooked. The fact that luck- or skill-based asymmetries
are required to generate above-normal profits is
confirmed by the results of theoretical models
Sample selection bias
where such initial asymmetries are absent, e.g.
When first-mover opportunities are generated by Glazer (1985), Fudenberg and Tirole (1985), and
processes of chance, firms that draw less attractive Gilbert and Harris (1984). Firms in these models
opportunities may choose not to enter, or they face an investment timing decision-e.g., when
may exit quickly and thus not be observed. to enter an emerging market that is initially too
Pioneers that survive this screening process would small to support even a single entrant. If all
be expected to earn above-average returns. l7 firms have identical information and investment
The role of luck is illustrated by the following opportunities, the initial entrant fails to earn
simple example. Assume that a given firm (having excess returns; competition for the first-mover
no exceptional proficiency in research) faces an position drives the associated profits to zero.
opportunity to engage in risky R&D on a Preemptive investment occurs at the earliest point
pioneering new product, with a 50 percent in time that the initial entrant can earn a non-
probability of technical success. Expected profits negative present value of profits.
are the average of two components: a gain of
P + E if the research succeeds, versus a loss of
7~ if the project fails. The firm enters the market Definitional and measurement issues
only if the project proves technically successful.
Significant profits will thus be observed contingent
What constitutes a ‘Jirst-mover’?
on entry, even though expected profits ( d 2 ) may
be barely sufficient for the firm to be induced Perhaps the most fundamental problem with the
to undertake the project. With free entry, concept of ‘first-mover’ is that of definition. If a
competition among R&D-performing firms firm enters an established market, but exploits
should drive these expected profits to near zero. some technological discontinuity or appeals to
Thus, first-mover advantages may enable suc- new demand segment, should it be classified as a
cessful pioneers to earn high returns even though first-mover? In general, how large a discontinuity
from existing practice is required to cross the
l6 Rumelt (1987) calls profits generated by luck ‘entrepreneur- threshold for definition as a pioneer?
ial rents’. His characterization differs slightly from our
approach, since he ignores the potential for first-mover profits
linked to proficiency. I * Assume that R&D projects are drawn from some random
Compounding this selection bias is the fact that if all distribution with a mean return of approximately zero, and
potential entrants perceive net disadvantages to early entry, that only successful projects with positive returns are
no entry will occur, and hence no pioneer will be observed. introduced to the market. The average profit earned
When this arises there may be an argument for public contingent on success rises with mean-preserving increases
provision, as in the case of government funding of basic in risk. For a formal analysis along these lines, see Lippman
research projects. and Rumelt (1982).
 First-mover Advantages 51

We offer no good answers to these questions; profit data are seldom obtainable.2” Hence,
available data and expedience have been the market shares and rates of company survival are
criteria used in most empirical studies. Clearly, typically used as surrogate measures. Market
adopting a loose definition causes a large fraction share and survival have both been shown to be
of entrants to be classified as pioneers.I9 In the correlated with profits and thus have some
PIMS data, for example, more than half of all validity as proxies. But the correlation of these
business units are ‘pioneers’, including multiple measures with profits is not always strong and
competitors within the same market segment causality is often ambiguous. For example,
(Buzzell and Gale, 1987). managers can often increase share at the expense
There is also the question of whether the of profits, e.g. by drastically cutting price,
criterion for first-movership is actual market or increasing promotional activity. Moreover,
entry, or the initiation of preliminary work such measures of market share can vary greatly
as R&D. The standard definition is based on depending on whether markets have been broadly
market entry, which we agree is the appropriate or narrowly defined.
criterion. Market share and survival also have some
Assuming some reasonable definition of what spurious correlations with first-movership and are
constitutes a pioneer, for empirical work there thus inherently biased. Consider, as a stylized
remains the problem of distinguishing among example, a market with two identical firms, A
later entrants. Such entrants can be classified by and B, that grow at identical rates, say 20 percent
(1) their numerical order in the sequence of per year. Assume that both firms earn zero
entry, (2) elapsed time since entry of the pioneer, economic profits. A enters 1 year ahead of B,
or (3) general categories such as early follower, making A the ‘first-mover’. In this case firm A
late follower, differentiated follower, ‘me-too’ will always maintain a market share 20 percent
follower, etc. These categories are not, in general, greater than B, even though by our definition
consistent or comparable across markets. For there are no ‘first-mover advantages’. While this
example, a firm that is third in the order of entry example is extreme, our point is that early
is likely to have been an early follower in a entrants have natural advantages in market share
market with 20 firms, but a late follower in a that do not necessarily translate into higher
market with four. profits.
Survival rates suffer from similar biases. For
Alternative measures of ‘jirst-mover advantage’: example, first-movers may be intrinsically ‘stron-
profits vs. market share vs. probability of ger’ or more proficient than later entrants.
sur v iv aI Consider first-mover advantages generated
through firm proficiency, as illustrated in Figure
Earlier we argued that economic profits are the 1. First-movers may exhibit higher survival rates,
appropriate measure of first-mover advantage. but it may be exceedingly difficult to ascertain
Profit maximization is the sole objective of whether this stems from pioneering, per se, or
stockholders in all modern theories of the firm. whether it reflects some more basic characteristic
First-mover advantages exist when the pioneering of the firm.
firm earns positive present value of profits as the
consequence of its early entry (i.e. positive profits
net of those attributable to more general types Magnitude and duration of first-mover
of firm proficiency). advantages
A serious problem confonting those engaged The term ‘first-mover advantage’ suggests that
in empirical work is the fact that disaggregate the pioneer remains more profitable than later
entrants, but careful reflection reveals that this
I y In this article we use the terms ‘first-mover’ and ‘pioneer’
interchangeably to refer to a unique firm within a given need not be the case. Pioneering firms can enjoy
market. However, a broader definition of ‘pioneer’ is often significant first-mover advantages but be less
applied, as in the PIMS questionnaire which asks whether
the business unit was ‘one of the pioneers’ at the time it
entered the market. In the extreme, if market niches are 2”The only general source of such data is the PIMS data
defined narrowly enough, virtually all firms can be classified base, which has numerous limitations. For an assessment,
as ‘pioneers’. see Anderson and Paine (1978).
52 M . B. Lieberman and D. B. Montgomery

profitable than later entrants when viewed over that future empirical research needs to be
an extended period. more precise in elucidating specific first-mover
consider, for example, a pioneer with patent mechanisms. Theoretical work in the area has
protection, whose economic profits fall to near- often suffered from the opposite prob-
zero after the patent expiration date. This pioneer lem: models have been designed to articulate
has clearly gained a first-mover advantage, even one piece of the first-mover puzzle but have
though after expiration of the patent the pioneer failed to embed it within a suitably endogenous
may be less profitable and hold a smaller market system.
share than other firms. Indeed, many pioneers Below, we develop these points further and
exploit their initial advantages, then exit or sell offer some specific suggestions for future research.
out to others.*’ We propose several directions that we regard as
Note that this raises the possibility that interesting, even though we doubt that all our
there may be both first-mover and late-mover suggestions are truly feasible.
advantages in a given market. For example, the
pioneering firm described above may earn high
Theoretical and conceptual issues
initial profits based on its patent, but ultimately
be overshadowed by a larger firm with unique In this survey we have stressed the endogenous
marketing skills that yield above-average returns. nature of first-mover advantages. Most theoretical
In this case, first-mover advantages accrue to studies have begun by assuming some initial
the pioneer, and late-mover advantages to the asymmetry, thereby sidestepping the endogeneity
marketing firm. Whether the first-mover or late- issue. The fundamental (and in our view most
mover has greater relative advantage depends on interesting) question of how first-mover oppor-
the point in time that the market is observed. tunities arise and are pursued by specific firms
Moreover, for any given firm, the question of remains almost completely unexplored. An
whether early or late entry is more advantageous important theoretical challenge is to flesh out
depends on the firm’s particular characteristics. this dynamic process, and distinguish the impact
In the above example, if one firm has unique of firm proficiency from that of luck.
R&D capabilities while the other has strong A related need is for more theoretical work
marketing skills, it is in the interest of the first linking individual firm characteristics to optimal
firm to pioneer and the second firm to enter at timing strategy. For example, what types of firms
a later date. Both may earn significant profits are best suited to pioneer, and what types are
entering in this sequence, but neither would gain best suited to follow? Models from the economics
if the (attempted) order of entry were reversed. literature typically make only the most rudimen-
tary distinctions among firms-indeed, firms are
usually assumed identical along all dimensions
ISSUES FOR FUTURE RESEARCH except timing and size. The strategic management
field has traditionally emphasized a wider set of
Given the problems discussed in the previous inter-firm differences. The implications of these
section, is ‘first-mover advantage’ a useful differences in the context of mechanisms for first-
research concept? Clearly, it does provide a mover advantage needs to be explored in greater
unifying framework for a broad class of phenom- depth. Work by Teece (1986a,b) and Wernerfelt
ena. But as a focus for empirical research, the and Karnani (1987) provides a useful start, but
concept of first-mover advantage may be too many opportunities remain.
general and definitionally elusive to be useful. In Despite its deficiencies, theoretical work on
recent years a number of cross-section empirical first-mover advantages remains far ahead of our
studies have attempted to detect and categorize understanding of mechanisms that assist late-
a wide range of ‘pioneering advantages’. While movers. There has, for example, been little
these studies provide useful groundwork, we feel conceptual work on resolution of technological
and market uncertainty, even though this is often
For example, DuPont was the world’s first commercial the macor factor affecting the timing of entry in
producer of cyclohexane, but exited when the firm was no
longer able to compete with oil companies using refinery- practice. Similarly, incumbent inertia is one area
based processes (Stobaugh, 1988). where our understanding remains weak. For
 First-mover Advantages 53

example, we have few frameworks (with the questions regarding the relative importance of
exception of population ecology, and the model first-mover advantages. Such assessments might
by Tang, 1988) that enable us to determine when be made across several dimensions. First, it would
inertia is desirable, and when it is dysfunctional be interesting to assess the magnitude of first-
(Lambkin, 1988). mover advantages relative to profits derived from
more general forms of firm proficiency, such as
superior manufacturing or marketing skills. How
Empirical issues
much of the inter-firm variance in profit rates
In our view, an important priority for empirical can be attributed to first-mover effects, as
research is to focus more precisely on the opposed to more general proficiency differentials?
evaluation of specific first-mover mechanisms, And to what extent should managerial effort
rather than on general investigations of the merits be allocated toward searching out preemption
of pioneer versus follower strategies. Researchers opportunities, as opposed to building more
should aim to test specific models and carefully general organizational capabilities? We are
distinguish among mechanisms. For example, intrigued with the possibility of obtaining some
additional empirical research is needed on the rough quantitative evidence on this score, despite
topic of switching costs, where first-mover advan- the obvious difficulties.
tages seem to be significant. Here it may be Similar comparisons might be made among
possible to distinguish empirically between risk- specific first-mover mechanisms and the industries
based models (such as Schmalensee, 1982) and in which they operate. It would be useful to
those based on psychological framing effects (e.g. know which of the various mechanisms are most
Carpenter and Nakamoto, 1986). The initial important in practice, and in what industries they
efforts of Carpenter and Nakamoto suggest that operate most strongly. Some empirical first-
experimental methods may prove fruitful for mover studies (e.g. Robinson, 1988 and Robinson
disentangling the fundamentals of such first- and Fornell, 1985) have highlighted interindustry
mover mechanisms. differences; their results suggest that first-mover
New data are required in order to broaden effects are more powerful in consumer-goods
and deepen understanding of first-mover advan- industries than in producer-goods. It would be
tages and disadvantages and the mechanisms interesting to extend such comparisons to service
that generate them. Researchers need to wean industries where casual observation suggests that
themselves from convenient but much-used data imitation occurs rapidly. Still, this produced
bases (e.g. PIMS). Wherever possible, data consumer/service industry trichotomy is rudimen-
should be collected on non-survivors, as well as tary; a more precise mapping between industry
surviving firms, to alleviate the censored sample characteristics and first-mover mechanisms would
problem. More difficult to obtain, but even more be useful.
important, would be appropriate measures of Another set of comparisons relates to the
profit, in order to reduce the dependence of duration of first-mover advantages. Under what
empirical knowledge on market share effects. conditions are they ephemeral versus long-lived?
The endogeneity issue has been ignored in How does this vary by mechanisms and by
most empirical studies as well as in theoretical industry? And what steps can management take
work. Recent research by Boulding and Moore to enhance sustainability? The existing empirical
(1987) suggests some possible empirical literature provides some evidence on these issues,
approaches. Models with endogenous structure but our knowledge is still quite limited.
are necessary for exploring the linkage between
firm characteristics, first-mover opportunities,
and performance. Potentially, such research might IMPLICATIONS FOR MANAGERS
be able t o disentangle the relative importance of
proficiency versus luck in generating first-mover Although our focus has been primarily on research
advantage. issues, we conclude with a brief discussion of
While we are skeptical about the value of more managerial implications.
broad-brush empirical efforts, we d o think there One of the first lessons of the first-mover
are a number of interesting but unanswered literature is that pioneering carries both
54 M . B. Lieberman and D. B. Montgomery

advantages and disadvantages. The net impact Managers who have chosen to emphasize
may well be negative, as illustrated by the demise pioneering must address a number of issues.
of Bomar and Osborne, pioneers in electronic First, how can the pioneer protect itself from
calculators and portable computers respectively. imitation, to prevent later entrants from ‘free-
In a cross-sectional study, Boulding and Moore riding’ on the pioneer’s efforts? Patents are one
(1987) found pioneering to be marginally obvious way to achieve this; designs that are
unprofitable on average. On the other hand, deliberately difficult to reverse engineer are
numerous studies have found enduring market another. Pioneers can sometimes preempt key
share advantages for surviving pioneers (Bond resources, such as the most desirable retailing
and Lean, 1977; Carpenter and Nakamoto, 1986; locations or distribution networks. If the product
Urban et a f . 1986; Robinson, 1988; Robinson is one where consumer switching costs will be
and Fornell, 1985). Thus, pioneering may prove important, it may be essential that the pioneer
advantageous to some firms in some circum- induce trial by the majority of potential customers
stances, but it is not necessarily a superior before rivals have an opportunity to do so.
strategy for all entrants. First-movers must also guard against incumbent
In order for a firm to become a first-mover or inertia, which may result from complacency,
pioneer, a feasible opportunity must present arrogance, or inattention to shifts in technology
itself. The occurrence of such an opportunity or customer needs. Changes in the environment
depends on the firm’s own foresight, skill and can give potential competitors a window of
luck, and that of competitors. These factors are opportunity. Pioneers may want to consider
interdependent-or as MacMillan (1983) has put broadening the product line, thereby thickening
it, ‘Good generals make their luck by shaping their relationship with customers and blocking
the odds in their favor’. The likelihood of opportunities for competitors to enter.
being a first-mover may be enhanced, often First-movers must recognize that initial success
substantially, by managements’ actions, but does not automatically confer a franchise for
opportunities for first-movership are by no means permanent Competitive advantage. Rather, first-
controlled by the firm alone. mover advantages must be sustained by careful
Firms must decide whether to invest resources nurturing. Even companies with long-established
in search of first-mover opportunities. Moreover, leadership positions, based in part on pioneering,
when a specific first-mover opportunity arises, can dissipate their advantages. In the late
managers must decide whether and how to exploit 1970s, for example, Kleenex tissues, which had
it. Sony, for example, aggressively pursues first- established the industry standard, lost significant
mover advantages from new product innovation. market share (Carpenter, 1988). Subsequent
Its rival, Matsushita (whose nickname in research indicated that consumers’ preferences
Japanese, rnaneshita denki translates as ‘elec- had shifted to competing brands that were
tronics that have been copied’), generaily lets perceived as being softer. This preference shift
Sony and others innovate; Matsushita then takes was the result of Kleenex’s production economy
a position based on its manufacturing and choices that had gradually increased the level of
marketing capabilities. Matsushita invests in R&D recycled wood pulp in the tissues. Once Kleenex
to be ready to enter the market when it begins restored and enhanced the level of softness, it
rapid growth, but the firm will not launch new regained and even surpassed its previous market
products until others have proven the market. share (Wall Street Journal, 1987).
Choice between these two alternative approaches Alert pioneers take proactive steps to protect
depends on the firm’s specific characteristics and their pioneering advantages. Pioneers need to
skills. Firms whose entrepreneurial vision and increase capacity sufficiently, especially in high-
new-product R&D are excellent will tend to find growth markets, to avoid being overtaken by
first-movership attractive, whereas firms having aggressive followers. Tohatsu, the leading
relative skill bases in manufacturing and market- Japanese motorcycle producer in the late 1950s,
ing may not. Then too, firms having a strategy failed to do so and was overtaken by Honda
of first-movership may often be forced to follow (Abegglen and Stalk, 1985). Continual product
in related product areas in order to provide a innovation is another key to retaining competitive
more complete product line. advantage. Caterpillar, by neglecting to re-
 First-mover Advantages 55

engineer their low-end models in the early 1970s, positioning of a follower brand with respect to
gave John Deere an opportunity to enter the consumer needs was the single most important
over-100 horsepower bulldozer market using predictor of a follower’s share relative to a first-
an innovative transmission as a differentiating mover. Foster (1986) cites numerous examples
feature (HBS, 1977). Finally, quality and breadth of followers that were able to successfully
of product line may also be used to sustain differentiate themselves by exploiting technologi-
pioneer advantage. Robinson (1988) found that cal discontinuities. Once the follower has gained
pioneer quality advantages tend to diminish over a strong foothold, direct challenge of the pioneer
time, but product line breadth advantages are often becomes feasible.
more sustainable. In general, pioneers that build
adequate capacity, innovate to meet changing
technologies and customer requirements, and fill SUMMARY
up available market niches, are formidable
opponents and exceedingly difficult to overcome. This article has surveyed theoretical and empirical
For firms with a follower strategy (whether by research on mechanisms that confer advantages
design o r inevitable) the major issue is whether and disadvantages on first-mover firms. Mecha-
to attack the pioneer directly, or seek a less nisms that promote first-mover advantages include
confrontational, differentiated position. If the proprietary learning effects, patents, preemption
pioneer is small, lacks resources, or has not yet of input factors and locations, and development
achieved much market penetration or recognition, of buyer switching costs. Conversely, first-mover
he may be vulnerable to aggressive attack by a disadvantages may result from free-rider prob-
follower. Urban et al. (1986) found that followers lems, delayed resolution of uncertainty, shifts in
in consumer packaged goods were able to technology or customer needs, and various types
overcome the pioneer’s market share advantage of organizational inertia.
by substantially out-spending the pioneer on We have discussed key conceptual issues and
advertising. However, a number of studies research priorities. Future research should con-
have found that ‘me-too’ strategies tend to be sider the endogenous nature of first-mover
unsuccessful (Bond and Lean, 1977; Montgomery, opportunities and the potential for sample selec-
1975; Davidson, 1976; Carpenter and Nakamoto, tion bias. Greater emphasis should be given in
1986). Even substantial price cuts may fail to empirical analysis to economic profits as a
wrest significant market share (Bond and Lean, criterion. Perhaps most important, basic questions
1977; Carpenter and Nakamoto, 1986). In gen- still need to be addressed on how ‘first-movers’
eral, ‘me-too’ strategies seem to be effective only are defined and identified in practice.
when the innovator has not done its marketing Finally, managers must decide whether a strong
properly or intensely enough.22 emphasis on pioneering is appropriate, given
Followers may find it preferable to avoid direct their firm’s resource base. In specific situations
confrontation with a strong pioneer, at least where skill and good fortune have generated a
initially. Golden Wonder developed a totally new first-mover opportunity, managers must decide
market segment and distribution channel in the whether the firm should pursue it, and if so, how
U.K. potato chip market, thereby distinguishing best to enhance its value. And managers of
itself from the dominant incumbent (Bevan, follower firms must determine whether and how
1974). Urban et al. (1986) found that superior the first-mover advantages of the pioneer can be
subverted.
22 The Carpenter and Nakamoto (1986) experiments suggest
an elaboration of the ‘me-too’ product question. They found
that ‘me-too’ brands tend to fail when positioned close to ACKNOWLEDGEMENTS
the pioneer but are often successful when positioned close
to differentiated later entrants. A brand that mimics an
entrenched pioneer usually finds it difficult to generate trial We thank Piet Vanden AbeeleyRajiv Mark
Laly

purchases, due to the dominant perceptual position of the Satterthwaite and Birger Wernerfelt for helpful
pioneer. However, a brand positioned near a differentiated d~scussionson earlier drafts. The Strategic M ~ ~ -
later entrant can make it easier for both to generate trial in
comoetition with the Dioneer. Thus. some ‘me-too’ oositions agement Program at Stanford Business
may’be superior to oihers (Carpenter, 1988). provided financial support
56 M . B. Lieberman and D. B. Montgomery

REFERENCES Cooper, A. C. and D. Schendel. ‘Strategic responses

to technological threats’, Business Horizons, 19,
February 1976, pp. 61-69.
Abegglen, J. C. and G. Stalk. Kaisha, the Japanese Davidson, J. H. ‘Why most new consumer brands fail’,
Corporation, Basic Books, New York, 1985. Harvard Business Review, 54, March-April 1976,
Abell, D. F. ‘Strategic windows’, Journal of Marketing, pp. 117-122.
42, July 1978, pp. 21-26. Dixit, A. ‘The role of investment in entry deterrence’,
Abernathy, W. J. and K. Wayne. ‘Limits of the Economic Journal, 90, 1980, pp. 95-106.
learning curve’, Harvard Business Review, 52, 1974, Eaton, B . C. and R. Lipsey. ‘The theory of market
pp. 109-119. pre-emption: the persistence of excess capacity and
Anderson, C. R. and F. T. Paine, ‘PIMS: a reexami- monopoly in growing spatial markets’, Economica,
nation’, Academy of Management Review, 3, July 46, 1979, pp. 149-58.
1978, pp. 602-612. Eaton, B. C. and R. Lipsey. ‘Capital, commitment
Arrow, K. ‘Economic welfare and the allocation of and entry equilibrium’, Bell Journal of Economics,
resources to innovation’. In R. Nelson (ed.), 12, 1981, pp. 593-604.
The Rate and Direction of Inventive Activity. Eaton, B. C. and R. Ware. ‘A theory of market
Universities-National Bureau Conference Series, structure with sequential entry’, Rand Journal of
No. 14, Arno Press, New York, 1962. Economics, 18, Spring 1987, pp. 1-16.
Baldwin, W. L. and G. L. Childs. ‘The fast second Farrell, J. and G. Saloner. ‘Installed base and
and rivalry in research and development’, Southern compatibility: innovation, product preannounce-
Economic Journal, 36, 1969, pp. 18-24. ments and predation’, American Economic Review,
Bevan, A. ‘The U.K. potato crisp industry, 1960-72: 76, December 1986, pp. 940-955.
a study of new entry competition’, Journal of Fast, N. ‘Lincoln Electric Company’. Working Paper
Industrial Economics, XXII, June 1974, pp. 281-297. Case No. 37G028, Harvard Business School, 1975.
Bond, R. S. and D. F. Lean. Sales, Promotion and Foster, R. N. Innovation: The Attacker’s Advantage.
Product Differentiation in Two Prescription Drug Summit Books, New York, 1986.
Markets. U .S. Federal Trade Commission, Washing- Fudenberg, D., R. Gilbert, J. Stiglitz and J . Tirole.
ton, DC, 1977. ‘Preemption, leapfrogging, and competition in patent
Boulding, W. and M. J. Moore. ‘Pioneering and races’, European Economic Review, 22, June 1983,
profitability: structural estimates from a nonlinear pp. 3-31.
simultaneous equations model with endogenous Fundenberg, D. and J. Tirole. ‘Capital as a commit-
pioneering’. Research Paper, Fuqua School of ment: strategic investment to deter mobility’.
Business, Duke University, May 1987. Journal of Economic Theory, 31, December 1983,
Bresnahan, T. F. ‘Post-entry competition in the plain pp. 227-250.
paper copier market’, American Economic Review, Fudenberg, D. and J. Tirole. ‘The fat-cat effect, the
75, May 1985, pp. 15-19. puppy-dog ploy, and the lean and hungry look’,
Bright, A. A. The Electric Lamp Industry, Macmillan, American Economic Review, 74, May 1984, pp.
New York, 1949. 361-366.
Brock, G. W. The U.S. Computer Industry: A Study Fundenberg, D. and J . Tirole. ‘Preemption and rent
of Market Power, Ballinger Publishing Company, equalization in the adoption of new technology’,
Cambridge, MA, 1975. Review of Economic Studies, LII, 1985, pp. 383401.
Buzzell, R. D. and B. T. Gale. The PIMS Principles: Gal-Or, E. ‘First mover and second mover advantages’,
Linking Strategy to Performance, Free Press, New International Economic Review, 26, October 1985,
York, 1987. pp. 649-653.
Cady, J. F. ‘Marketing strategies in the information Ghemawat, P. ‘Capacity expansion in the titanium
industry’. In R. D. Buzzell (ed.), Marketing in an dioxide industry’, Journal of Industrial Economics,
Electronic Age. Harvard Business School Press, XXXIII, December 1984, pp. 145-163.
Boston, 1985. Ghemawat, P. ‘Drastic product innovations under
Carpenter, G. S. ‘Market pioneering and competitive uncertainty’. Working paper, Harvard Business
positioning strategy’, Annales de Telecommuni- School, 1986a.
cations (forthcoming 1988). Ghemawat, P. ‘Wal-Mart Stores’ discount operations’.
Carpenter, G. S. and K. Nakamoto. ‘Market pioneers, Working Paper Case No. &387418, Harvard
consumer learning, and product perceptions: a Business School, 1986b.
theory of persistent competitive advantage’. Ghemawat, P. and A. M. Spence. ‘Learning curve
Research Paper, Columbia University and Univer- spillovers and market performance’, Quarterly
sity of California, November, 1986. Journal of Economics, 100, 1985, pp. 839-852.
Chandler, A. D., Jr. The Visible Hand: The Managerial Gilbert, R. J. ‘Pre-emptive competition’. In J. E.
Revolution in American Business. Belknap Press o f Stiglitz and G. F. Mathewson (eds), New Develop-
Harvard University Press, Cambridge, MA, 1977. ments in the Analysis of Market Structure. MIT
Conner, K. R. ‘Strategies for product cannibalism’, Press, Cambridge, MA, 1986.
Strategic Management Journal, 9 (Special Issue), Gilbert, R. J. and R. G. Harris. ‘Investment decisions
1988, pp. 9-26. with economies of scale and learning’, American
 First-mover Advantages 57

Economic Review, 71, May 1981, pp. 172-177. MacMillan, I. C. ‘Preemptive strategies’, Journal of
Gilbert, R. J. and R. G. Harris. ‘Competition with Business Strategy, 4, Fall 1983, pp. 16-26.
lumpy investment’, Rand Journal of Economics, MacMillan, I . C. ‘Preemptive strategies’. In W. Guth
15, Summer 1984, pp. 197-212. (ed.), Handbook of Business Strategy, Warren,
Gilbert, R. J. and D. M. G. Newbery. ‘Preemptive Gorham & Lamont, New York, 1984.
patenting and the persistence of monopoly’, Amer- Main, 0. W. The Canadian Nickel Industry. University
ican Economic Review, 72, June 1982, pp. 514-526. of Toronto Press, Toronto, 1955.
Glazer, A. ‘The advantages of being first’, American Mansfield, E. Industrial Research and Technotogical
Economic Review, 75, June 1985, pp. 473-480. Innovation: An Econometric Analysis, W. W.
Guasch, J. L. and A. Weiss. ‘Adverse selection of Norton Company, New York, 1968.
markets and the advantages of being late’, Quarterly Mansfield, E. ‘How rapidly does new industrial
Journal of Economics, May 1980, pp. 453-466. technology leak out?’, Journal of Industrial Eco-
Hannan, M. T. and J. Freeman. ‘Structural inertia nomics, XXXIV, December 1985, pp. 217-223.
and organizational change’, American Sociological Mansfield, E., M. Schwartz, and S. Wagner. ‘Imitation
Review, 49, April 1984, pp. 149-164. costs and patents: an empirical study’, Economic
Harvard Business School. ‘Deere & Company: indus- Journal, 91, December 1981, pp. 907-918.
trial equipment operations’, case no. 577-112, 1977. Montgomery, D. B. ‘New product distribution: an
Hirshleifer, J. ‘The private and social value of analysis of supermarket buyer decsions’, Journal of
information and the reward to inventive activity’, Marketing Research, 12, August 1975, pp. 255-264.
American Economic Review, 61, 1971, pp. 561-574. Porter, M. Interbrand Choice, Strategy and Bilateral
Jacobson, G. and J. Hillkirk. Xerox: A n American Market Power. Harvard University Press, Cam-
Samurai, Macmillan, New York, 1986. bridge, MA, 1976.
Johnson, R. N. and A . Parkman. ‘Spatial monopoly, Porter, M. Competitive Strategy. Free Press, New
non-zero profits and entry deterrence: the case of York, 1980.
cement’, Review of Economics and Statistics, 65, Porter, M. ‘Strategic interaction: some lessons from
August 1983, pp. 431-438. industry histories for theory and anti-trust policy’.
Judd, K. L. ‘Credible spatial preemption’, Rand In S. Salop (ed.), Strategy, Predation and Anti-trust
Journal of Economics, 16, Summer 1985, pp. Analysis. Washington: FTC, 1981.
153-1 66. Prescott, E. and M. Visscher. ‘Sequential location
Katz, M. L. and C. Shapiro. ‘Technology adoption in among firms with foresight’, Bell Journal of
the presence of network externalities’, Journal of Economics, 8, 1977, pp. 378-393.
Political Economy, 94, August 1986, pp. 822-841. Rao, R. and D. Rutenberg. ‘Pre-empting an alert
Klemperer, P. ‘Markets with consumer switching rival: strategic timing of the first plant by analysis
costs’. PhD thesis, Graduate School of Business, of sophisticated rivalry’, Bell Journal of Economics,
Stanford University, 1986. 10, 1979, pp. 412428.
Lambkin, M. ‘Order of entry and performance in new Reinganum, J. F. ‘Uncertain innovation and the
markets’, Strategic Management Journal, 9 (Special persistence of monopoly’, American Economic
Issue) 1988, pp. 127-140. Review, 73, September 1983, pp. 741-748.
Levin, R., A. Klevorick, R. Nelson and S. G. Winter. Ries, A. and J. Trout:Marketing Warfare, McGraw-
‘Survey research on R&D appropriability and Hill, New Yvrk, 1986.
technical opportunity-Part I: Appropriability’. Robinson, W. T. ‘Sources of market pioneer advan-
Working Paper, Yale University, July 1984. tages: the case of industrial goods industries’,
Lieberman, M. ‘The learning curve, pricing and market Journal of Marketing Research, (forthcoming, Sep-
structure in the chemical processing industries’, tember 1988).
PhD thesis, Harvard University, 1982. Robinson, W. T. and C. Fornell. ‘The sources of
Lieberman, M. ‘The U.S. magnesium industry’. market pioneer advantages in consumer goods
Standford University Business Case No. S-BP-231, industries’, Journal of Marketing Research, 22,
1983. August 1985, pp. 297-304.
Lieberman, M. ‘Excess capacity as a barrier to Rumelt, R. ‘Theory, strategy and entrepreneurship’.
entry: an empirical appraisal’, Journal of Industrial In D. Teece (ed.), Strategy & Organization for
Economics, 35, June 1987a, pp. 607-627. Industrial Innovation and Renewal. Ballinger, Cam-
Lieberman, M. ‘The learning curve, barriers to entry, bridge, MA, 1987.
and competitive survival in the chemical processing Scherer, F. M. Industrial Market Structure and
industries’. Unpublished, Graduate School of Busi- Economic Performance. Rand McNally, Chicago,
ness, Stanford University, March 1987b. 1980, pp. 431, 438.
Lieberman, M. ‘The learning curve, diffusion and Schmalensee, R. ‘Entry deterrence in the ready to
competitive strategy’, Strategic Management Jour- eat breakfast cereal industry’, Bell Journal of
nal, 8, September-October 1987c, pp. 441452. Economics, 9, Autumn 1978, pp. 305-327.
Lippman, S. A. and Rumelt, R. P. ‘Uncertain Schmalensee, R. ‘On the use of economic models
imitability : an analysis of interfirm differences in in antitrust: the Realemon case’, University of
efficiency under competition’, Bell Journal of Pennsylvania Law Review, 127, April 1979, pp.
Economics, 13, Autumn 1982, pp. 418-438. 345-365.
58 M . B. Lieberman and D. B. Montgomery

Schmalensee, R. ‘Economies of scale and barriers to Thomas (ed.), The Economics of Strategic Planning.
entry’, Journal of Political Economy, 89, 1981, pp. Lexington Books, Lexington, MA, 1986a.
1228-1238. Teece, D. J. ‘Profiting from technological innovation:
Schmalensee, R. ‘Product differentiation advantages implications for integration, collaboration, licensing
of pioneering brands’, American Economic Review, and public policy’, Research Policy, 15, 1986b, pp.
72, June 1982, pp. 349-365. 285-305.
Schnaars, S. P. ‘When entering growth markets, are Urban, G . L., R. Carter, S. Gaskin and Z . Mucha.
pioneers better than poachers?’, Business Horizons, ‘Market share rewards to pioneering brands: an
29, March-April 1986, pp. 27-36. empirical analysis and strategic implications’, Man-
Schumpeter, J. A. The Theory of Economic Develop- agement Science, 6, June 1986, pp. 645459.
ment. Oxford University Press, New York, 1961. Wall Street Journal. ‘Paper tiger: how Kimberly-Clark
Shaw, R. W. and S. A. Shaw. ‘Late entry, market wraps its bottom line in disposable huggies’, 23
shares and competitive survival: the case of synthetic July 1987.
fibers’, Managerial and Decision Economics, 5, Weiss, L. W. ‘Optimal plant size and the extent of
1984, pp. 72-79. sub-optimal capacity’. In R. T. Masson and P. D.
Spence, M. ‘Entry, capacity, investment and oligopolis- Qualls (eds), Essays in Industrial Organization in
tic pricing’, Bell Journal of Economics, 8, 1977, Honor of Joe S. Bain. Ballinger Press, Cambridge,
pp. 534-544. MA, 1976.
Spence, M. ‘Investment strategy and growth in a new Wernerfelt, B. ‘Brand loyalty and user skills’, Journal
market’, Bell Journal of Economics, ‘10, 1979, pp. of Economic Behavior and Organization, 6, 1985,
1-19. pp. 381-385.
Spence, M. ‘The learning curve and competition’, Bell Wernerfelt, B. ‘A special case of dynamic pricing
Journal of Economics, 12, 1981, pp. 49-70. policy’, Management Science, 32, December 1986,
Spence, M. ‘Cost reduction, competition, and industry pp. 1562-1566.
performance’, Econometrica, 52, January 1984, pp. Wernerfelt, B. ‘Umbrella branding as a signal of new
101-121. product quality: an example of signalling by posting
Stobaugh, R. Monopolies Aren’t Forever: The Case of a bond’. Unpublished, Kellogg Graduate School of
Petrochemicals. Harvard Business School Press, Management, Northwestern University, September
Boston, 1988. 1987.
Tang, M. ‘An economic perspective on escalating Wernerfelt, B. ‘General equilibrium with real time
commitment’, Strategic Management Journal, 9 search in labor and product markets’, Journal of
(Special Issue), 1988, pp. 79-92. Political Economy, 96, forthcoming June 1988.
Teece, D. J. ‘The diffusion of an administrative Wernerfelt, B. and A. Karnani. ‘Competitive strategy
innovation’, Management Science, 26, May 1980, under uncertainty’, Strategic Management Journal,
pp. 464-470. 8, March-April 1987, pp. 187-194.
Teece, D. J. ‘Firm boundaries, technological inno- Yip, G. S. Barriers to Entry. Lexington Books, D. C.
vation, and strategic management’. In L. G. Heath and Company, Lexington, MA, 1982.