why do firms imitate each other? do firms imitate each other? marvin b. lieberman ucla anderson...

WHY DO FIRMS IMITATE EACH OTHER?

MARVIN B. LIEBERMANUCLA Anderson School of Management

SHIGERU ASABAGakushuin University

Scholars from diverse disciplines have proposed numerous theories of business imi-tation. We organize these theories into two broad categories: (1) information-basedtheories, where firms follow others that are perceived as having superior information,and (2) rivalry-based theories, where firms imitate others to maintain competitiveparity or limit rivalry. We describe conditions under which each type of imitation ismost likely and offer guidance on identifying imitation in practice. Amplificationeffects and other performance implications of imitation are also addressed.

Imitation is a common form of behavior thatarises in a variety of business domains. Firmsimitate each other in the introduction of newproducts and processes, in the adoption of man-agerial methods and organizational forms, andin market entry and the timing of investment.Despite its frequent occurrence, imitation canhave radically different causes and implica-tions. Firms may imitate to avoid falling behindtheir rivals, or because they believe that others’actions convey information. The matching of ri-vals’ actions can intensify competition, or it canhave the opposite effect, promoting collusion.Imitation can spur productive innovation, or itcan amplify the errors of early movers. Thus,imitation can lead to large positive or negativeoutcomes for individual firms and society as awhole. Given the frequency of imitative behav-ior and the fact that societal outcomes are oftennegative, it is important that business research-ers, managers, and policy makers understandwhy imitation occurs and when it may haveharmful implications.

Business scholars from a range of disciplineshave proposed numerous theories of imitation.Although these theories share common ele-ments, they have been developed for specializedaudiences and tend to emphasize different mi-metic phenomena. Thus, the large body of re-

search on imitation remains fragmented, withfew scholars aware of related work by col-leagues in other disciplines. Our primary aim inthis article is to help develop this body of theoryby drawing together common threads.

Arguably, imitation processes are most in-teresting in environments characterized by un-certainty or ambiguity. Few decisions haveoutcomes that are fully predictable. Managerstake actions, the consequences of which de-pend on the future state of the environment. (Inthe case of new product introduction, for ex-ample, such a state would correspond to aparticular level of production cost, customerdemand, market competition, and so forth.) Ata minimum, most decisions are made underconditions of risk, where the probabilities ofenvironmental states can be estimated but theactual outcome is uncertain (Knight, 1921).Managers often face more severe forms of un-certainty: they may be unable to assign prob-abilities, they may lack information on cause-effect relationships, and they may be unableto assess the full range of possible outcomesand states (Milliken, 1987).

In this survey we emphasize the role of envi-ronmental uncertainty, which makes it difficultfor managers to predict the consequences of aparticular action or behavior. Environmental un-certainty promotes certain types of imitationand raises the likelihood of undesirable out-comes. In situations where the imitated behav-ior is tacit or complex, there may be additionaluncertainty about the methods used by a lead-ing firm to achieve superior results. This second

We thank Joel Baum, Sushil Bikhchandani, David Hirsh-leifer, Charlotte Ren, Anand Swaminathan, Tatsuo Ushijima,Lynne Zucker, two anonymous referees, and seminar partic-ipants at Carnegie Mellon University for valuable insightsand suggestions. We remain responsible for all errors andomissions.

� Academy of Management Review2006, Vol. 31, No. 2, 366–385.

366

type of uncertainty, or “causal ambiguity,” im-pedes imitation, as we discuss briefly.

Imitation of superior products, processes, andmanagerial systems is widely recognized as afundamental part of the competitive process.Many studies have documented processes of“creative destruction” (Schumpeter, 1942) and“diffusion of innovations” (e.g., Rogers, 1995)that lead to widespread adoption of superiorproducts or methods. A related body of literatureon “first mover advantages” (e.g., Lieberman &Montgomery, 1988) shows that successful pio-neers seldom can prevent entry by imitative fol-lowers. Such imitation tends to reduce the inno-vator’s profits while generating broader gains ineconomic welfare as prices and costs fall. Fur-thermore, it is widely recognized that when net-work externalities give rise to standards, firmsimitate to minimize costs (Katz & Shapiro, 1985).In the absence of uncertainty, such types of im-itation are comparatively straightforward andwell understood.

In highly uncertain environments, however,imitative behavior can be dysfunctional or evenpathological. Herd behavior can lead to specu-lative bubbles and the waste of resources induplicative investments. Recent examplesabound in the internet sector, where a financialbubble in the late 1990s left in its wake a busi-ness landscape overpopulated by imitativestart-ups and organizational forms. Dozens ofinternet “business-to-business exchanges”sprouted before deficiencies of the format be-came apparent, and consulting firms rushed toset up enterprise “incubators,” which were per-ceived during the boom as a superior new or-ganizational form. As internet commerce bur-geoned, many managers believed that theywere adopting superior methods and systems,only to discover that they had followed othersdown largely fruitless paths. These examplessuggest that imitation often provokes excessiveinvestment that is focused too narrowly on alimited number of options, with poor profit out-comes for the majority of firms. Imitation is anatural response to environmental uncertainty,but, by reducing variety, it can compound thecollective risk of firms in an industry.

In environments where change is more incre-mental, imitation can defuse rivalry and reducerisk for any given firm. Knowledge that rivalswill respond in kind lowers the incentive for anyindividual firm to act aggressively in an effort to

gain competitive advantage. In the extreme,such imitation can be anticompetitive. Govern-ment antitrust authorities recognize this possi-bility and have overturned business agreementsthat promote parallel behavior. One example isthe so-called smog case, where the major U.S.automakers agreed to share pollution controltechnology adopted by any one of the firms. TheU.S. Justice Department alleged that, by facili-tating imitation, the agreement led the compa-nies to cut back on R&D (White, 1971).

When competitors take similar actions, thereis less chance that any firm will succeed or failrelative to others. Thus, imitation helps to pre-serve the status quo among competitors thatfollow each other, even in industries wherestrong rivalry is maintained. One example isCasio and Sharp, the leaders in electronic cal-culators, which repeatedly matched each other’sincremental innovations in the 1970s. Their mar-ket shares remained balanced, even though thelead shifted back and forth. Ultimately, the twofirms emerged with nearly identical productlines (Numagami, Asaba, Shintaku, & Amikura,1992). Moreover, by strengthening each other,these two Japanese producers were able to drivemany foreign rivals from the global market. Inthis case, imitation stabilized the relative posi-tions of the leaders while raising the risk offailure for those that did not follow.

As these examples suggest, imitation can oc-cur for a variety of reasons, with dramaticallydifferent implications. Under some conditionsimitation is apt to be beneficial and should bepromoted. But in other settings imitation is morelikely to have negative implications for firmsand/or society. It is therefore important to distin-guish among types of mimetic behavior and tounderstand the potential consequences. To aidthese assessments, we review a range of rele-vant theories in strategic management, econom-ics, and organization theory. Scholars fromthese disciplines have addressed the phenome-non of imitation from complementary perspec-tives.

Throughout the paper we argue that all formsof imitation have some rational basis. Thus, wereject the view of some organization scholarsthat imitation is a purely ritualistic phenome-non. We recognize, however, that imitative be-havior can often appear irrational, particularlywhen viewed in retrospect, after uncertainty hasbeen resolved.

2006 367Lieberman and Asaba

We organize theories of business imitationinto two broad categories: (1) information-based theories, where firms follow others thatare perceived (sometimes erroneously) as hav-ing superior information, and (2) rivalry-basedtheories, where firms imitate others to main-tain competitive parity or limit rivalry. In thenext two sections of the article, we describethe information-and rivalry-based theories, re-spectively. We then address more appliedproblems of identifying these two types of im-itation. The task is made difficult by the factthat both types of imitation can arise simulta-neously, and they can be hard to distinguishfrom the nonimitative case, where firms re-spond independently but identically to thesame external shock. In the final sections ofthe paper, we consider performance implica-tions of imitation and opportunities for futureresearch.

We do not review the empirical literature onbusiness imitation in great detail, since our ob-jective is to provide a conceptual survey ratherthan a discussion of specific findings. To give abroader guide to this literature, however, we listsome prominent empirical studies and theirmain conclusions in Table 1.

INFORMATION-BASED THEORIES OFIMITATION

Information-based theories of imitation havebeen proposed in the fields of economics, insti-tutional sociology, and population ecology.These theories apply in environments wheremanagers cannot assess connections betweenactions and outcomes with great confidence.Managers may be unsure of the likelihood ofpossible outcomes, and they may have morefundamental difficulties recognizing cause-effect relationships and the full range of poten-tial consequences. In such environments of un-certainty and ambiguity, managers areparticularly likely to be receptive to informationimplicit in the actions of others. Such informa-tion, although highly imperfect, can have astrong influence on managerial perceptions andbeliefs. Moreover, in uncertain environmentsmanagers may imitate to signal others abouttheir own (or their firm’s) quality.

Economic Theories

We begin by considering economic theories ofimitation, where the information component hasbeen developed most explicitly.1 The most prom-inent economic theory of herd behavior is called“information cascades” or “social learning”(Banerjee, 1992; Bikhchandani, Hirshleifer, &Welch, 1992, 1998). Information cascades occur“when it is optimal for an individual, havingobserved the actions of those ahead of him, tofollow the behavior of the preceding individualwithout regard to his own information” (Bikh-chandani et al., 1992: 994). The model formalizesa process of Bayesian learning. Suppose eachagent has some private information about thestate of nature. The first agent behaves purelybased on this private information, but theagent’s behavior reveals the information to fol-lowers. As this revealed information accumu-lates, it may be rational for followers to ignoretheir own prior information and mimic the deci-sions of others. A typical example is a restau-rant with a long queue that becomes increas-ingly popular. Many of those waiting at the endof the line may have intended to visit other res-taurants with which they are familiar, but theyare swayed by the observation of the queue,which suggests (perhaps erroneously) that therestaurant is of high quality. Thus, agents maychoose to go against their initial signals as theydraw inferences from the observed behavior ofothers.

Such processes have some power to explainthe imitative behavior that contributed to theinternet bubble in the late 1990s. Consider anentrepreneur contemplating a new retail ven-ture, with an initial preference for “brick andmortar” outlets rather than internet-based sales.Observing the growing wave of entry into theinternet sector (supported by the enthusiasticforecasts of analysts, the trade press, and risingstock prices), the entrepreneur concludes thatperhaps others have superior information aboutthe prospects for internet retailing. Eventually,the observed signals grow in strength relative tothe entrepreneur’s prior belief, and the entrepre-neur decides to follow others and enter the in-

1 The economic theories are an outgrowth of earlier workon how information affects the operation of markets, forwhich George Akerlof, Michael Spence, and Joseph Stiglitzwere awarded the 2001 Nobel Prize.

368 AprilAcademy of Management Review

TABLE 1Empirical Studies of Imitative Behavior

Topic Area and Study Phenomena/Industry Key Findings

Multimarket contact to mitigatecompetition

Heggestad & Rhoades (1978) Banks Multimarket contact stabilizes dominant firms’market share.

Scott (1982, 1991) Diversification Multimarket contact is associated with higherprofits in highly concentrated industries.

Rhoades & Heggestad (1985) Banks “Mutual forbearance” is not confirmed.Barnett (1993) Telephone equipment Weak competitors can survive longer than

strong competitors, owing to forbearance.Evans & Kessides (1994) Airlines Major airlines set higher fares on routes

where average levels of multimarket contactare higher.

Baum & Korn (1996) California commuterairlines (1979–1984)

Lower likelihood of exit for firms havinggreater multimarket contact with marketincumbents.

Gimeno & Woo (1996) Airlines Multimarket contact decreases the intensity ofcompetition.

Boeker, Stephan, & Murmann (1997) California hospitals(1980–1986)

Multimarket competition lowers the rate ofexit.

Baum & Korn (1999) California commuterairlines (1979–1984)

Effects of multimarket competition vary acrosscompetitor dyads.

Investment bunching as risk reductionKnickerbocker (1973) FDI by U.S. firms Bunching behavior is more likely to occur in

moderately concentrated industries.Flowers (1976) FDI in the United

States byCanadian andEuropean firms

Bunching of entry is positively related tohome market concentration.

Caves, Porter, & Spence (1980) FDI in Canada byU.S. firms


Yu & Ito (1988) FDI by U.S. tire andtextile firms


Kogut & Chang (1991) FDI in the UnitedStates by Japanesefirms


Chen & MacMillan (1992) Action response byairlines

A firm is more likely to match a move with itsdependence on the market.

Cockburn & Henderson (1994) R&D bypharmaceuticalfirms

R&D expenditures among firms are weaklyand positively correlated.

Hennart & Park (1994) FDI in the UnitedStates by Japanesefirms

Positive relation between bunching behaviorand market concentration is not confirmed.

Yamawaki (1998) FDI in the UnitedStates by Japanesefirms


Makino & Delios (2000) FDI by Japanese autoand electroniccompanies

Strong bunching effect in timing of FDI isobserved.

Mimetic isomorphism (organizationalsociology)

Davis (1991) Adoption of thepoison pill byFortune 500 firms

Spread through director interlock contact,rather than imitation of structurallyequivalent others.

(Continued)


ternet sector. In a similar way, financial inves-tors initially skeptical of internet commerce mayhave been swayed by the “information” re-vealed by the rising stock market, the tradepress, and other sources. Such forces helpeddrive the internet bubble upward.

As more entrepreneurs and investors are per-suaded by such observations, the wave of en-

trants grows. But as Bikhchandani et al. (1992)point out, such processes are inherently fragileand subject to reversal. Just as a critical mass ofpositive actions is needed to start the cascadeupward, the emergence of a sufficient number ofnegative signals will reverse the process. Thismay characterize the collapse of the internetbubble in mid 2000, as pessimistic assessments

TABLE 1(Continued)

Topic Area and Study Phenomena/Industry Key Findings

Haunschild (1993) Acquisitions during the1981–1990 period

Managers imitate the acquisition activities ofthose other firms to which they are tied viadirectorships.

Haveman (1993) Entry in savings and loanindustry

Rate of entry has inverted U-shapedrelationship with density (competition� vs.legitimation�).

Greve (1995, 1996) Adoption of new formats byradio stations

Radio stations imitate stations of the samecorporation.

Baum & Haveman (1997) Hotel location decisions inManhattan

New hotels locate close to established hotelsthat are similar in price but different in size.

Haunschild & Miner (1997) Investment banker chosenfor acquisitions

Imitation is influenced by frequency ofobserved behavior, traits of imitated firms,and quality of outcome.

Westphal, Gulati, & Shortell (1997) Implementation of TQM Earlier adopters seek efficiency gains,whereas later adopters seek legitimacy.

Deephouse (1999) Performance and strategicsimilarity among banks

Intermediate levels of strategic similarity leadto the highest performance.

Baum, Li, & Usher (2000) Acquisition of chainorganizations (nursinghomes)

Chain organizations imitate comparableothers (similarly sized chains).

Henisz & Delios (2001) International plant location Prior decisions by others provide legitimationand information.

Garcia-Pont & Nohria (2002) Alliance formation in theautomobile industry

Firms imitate the strategic behavior of othersoccupying the same strategic niche.

Lu (2002) Entry mode choice ofJapanese firms

Later entrants tend to follow the entry mode ofearlier entrants.

Herd behavior (economics)Chang, Chaudhuri, & Jayaratne

(1997)Clustering of bank branches Branch openings follow other existing

branches.Rao, Greve, & Davis (2001) Choice of securities analysts Analysts change coverage of a firm when

peers have recently changed coverage.Kennedy (2002) Prime television

programmingTelevision networks introduce new programs

in herdlike fashion.

Studies testing among alternativetheories

Gilbert & Lieberman (1987) Capacity investment bychemical firms

Smaller firms tend to follow investmentbehavior of larger rivals.

Gimeno & Chen (1998) Market similarity in theairline industry

Firms increase market similarity with rivalshaving similar resources and higherperformance.

Asaba & Lieberman (1999) Product introduction by softdrink firms

Large firms followed for major innovations;rivals followed for minor innovations.

Gimeno, Hoskisson, Beal, & Wan(2005)

International expansion oftelecom firms

Oligopolistic firms imitate each other’s localentry moves, while local monopolists do not.


began to appear and then grew rapidly. Internetstock prices fell to a fraction of their previouslevels, and entry came to a virtual halt. Thedramatic rise and fall took place within the spanof just two or three years—much faster than therate at which concrete data emerged on thelong-term prospects for internet commerce.

In driving such a bandwagon, the actions ofsome individuals or firms may be weightedmore strongly than others. If some are perceivedas likely to have superior information, they canbecome “fashion leaders” (Bikhchandani et al.,1998). For example, small firms may followlarger rivals if they believe the latter are betterinformed. Similarly, firms that have been suc-cessful in the past are more likely to have theiractions emulated. In the case of internet retail-ing, the entry of prominent firms such as Barnes& Noble and Wal-Mart, along with the enormousstock price gains of Amazon, helped legitimizethe efforts of other retailers to quickly establisha presence on the web. Such a role for leadingfirms is elaborated in the sociological theory ofinstitutional isomorphism, discussed below.

A second economic theory of herd behavior isbased on the idea that managers ignore theirown private information and imitate the deci-sions of others in an effort to avoid a negativereputation. By imitating, managers send signalsto others about their own quality. Suppose thatthere are superior and inferior managers whohave private information about investment. Out-siders do not know which type each manager is;rather, they know only that superior managersreceive informative signals about the value ofthe investment, whereas inferior managers re-ceive purely noisy signals. Since the signalssuperior managers receive might be mislead-ing, outsiders cannot rely solely on the outcomeof the investment; they must also rely on behav-ioral similarity among managers. Therefore, inorder to be evaluated as a superior type, man-agers ignore their own information and imitateothers (Palley, 1995; Scharfstein & Stein, 1990).Such imitation serves to enhance the manager’s“status,” a point elaborated in the institutionaltheories discussed below.

This theory may help to explain the herd be-havior of analysts and institutional investors indriving the internet bubble upward. Financialactors are often evaluated on performance rela-tive to peers; those who deviate from the con-sensus and ultimately prove to be wrong are

likely to suffer a fatal loss of reputation. Duringthe rise of the internet bubble, it was widelybelieved that the leading internet analysts hadsuperior signals, which led them to be optimisticabout the future of many internet companies.Those who did not follow were often shunned fortheir failure to grasp the fundamental dimen-sions of the “new economy.” Under these circum-stances, less-informed analysts and investorsoften chose to join the crowd, pushing internetstock prices higher.2 This example shows howthe second economic theory of herd behaviorcan complement the first: information cascadeslikely contributed to the emergence of the trend,which was further sustained by reputation-based signaling on the part of analysts and in-vestors.

Theories of Organizational Sociology andEcology

Organization theory gives a related explana-tion for behavioral similarity: institutional iso-morphism. DiMaggio and Powell (1983) arguethat rational actors make their organizations in-creasingly similar when they try to changethem. This process of homogenization is cap-tured by the concept of isomorphism. Isomor-phism is a constraining process that forces oneunit in a population to resemble other units thatface the same set of environmental conditions(Hawley, 1986).

Among several kinds of institutional isomor-phism, mimetic isomorphism is the processwhereby organizations model themselves onother organizations when the environment is un-certain. The modeled organization is perceivedas more legitimate or successful. Such mimeticbehavior is rational because it economizes onsearch costs to reduce the uncertainty that anorganization is facing (Cyert & March, 1963). Em-pirical studies show the operation of mimeticisomorphism in a variety of organizational do-mains. For example, Fligstein (1985) applied theconcept to explain the widespread adoption ofthe multidivisional structure, Haveman (1993)assessed the parallel diversification patterns ofCalifornia savings and loan associations, and

2 In addition, many stock analysts had conflicts of interestthat encouraged them to issue positive forecasts in order topromote their employer’s relationships with client firms.


Greve (1995, 1996) considered format changes ofradio stations.

Mimetic isomorphism can be viewed as ratio-nal imitation of a superior organization, al-though sociologists often emphasize ritualisticrather than rational motivations. For example,DiMaggio and Powell (1983) claim that the rapidproliferation of quality circles in American firmsthat modeled Japanese and European successeswas intended to enhance the legitimacy of theadopting firms. March (1981) argues that onceenough social actors adopt a certain behavior,the behavior is taken for granted or institution-alized, and, thereafter, other social actors willadopt the behavior without any thought. Institu-tionalization can be viewed as a threshold effectthat occurs once a critical mass of firms hasadopted. In this sense it bears resemblance tothe information cascades theory.

The sociological theory differs from informa-tion cascades in that, once a behavior is institu-tionalized, organizations are slow to respond tonew information. Behavior is much more dura-ble than in the economic theory, where new in-formation can lead to sudden reversals. Infor-mation cascades can be fragile, whereas thesociological theory points to the emergence of apermanent social order. Another difference isthat the sociological theory has generally beenapplied to explain the adoption of organization-al processes and innovations, whereas the eco-nomic theory’s aims are more general.3

While the economic theory of information cas-cades allows for the emergence of “fashionleaders,” organizational sociologists have actu-ally probed the issue of “who imitates whom.”Sociological studies indicate that a given firm’spropensity to be imitated increases with (1) theinformation content of its signal (where actionsby larger, more successful, or more prestigiousfirms may be seen as more informative) and (2)the focal firm’s degree of contact and communi-cation with other firms. Many studies haveshown that organizations of larger size and prof-itability are more likely to be followed (e.g.,Haunschild & Miner, 1997; Haveman, 1993).Moreover, theories of social networks (Granovet-ter, 1985; Gulati, Nohria, & Zaheer, 2000) suggest

that when organizations are linked by greaternetwork ties, they are likely to have more de-tailed information about each other, which facil-itates imitation. Along these lines, Davis (1991)and Haunschild (1993) found that imitation wasmore likely between firms with interlocking di-rectors, and Greve (1996) found that radio sta-tions were more likely to follow other stationsthat were units of the same corporation. Simu-lations by Abrahamson and Rosenkopf (1997)show how seemingly minor network structuralfeatures can affect the diffusion process.

These firm and network characteristics areseldom independent. Organizations that arecentral in a network have links with the greatestnumber of others; such organizations also tendto be larger and more prestigious. As Gulati andGargiulo point out, “The more central an orga-nization’s network position, the more likely it isto have better information” (1999: 1448).

While the above discussion emphasizes ra-tional interpretation of signals, studies flowingfrom the work of DiMaggio and Powell (1983)show that early and late movers may differ intheir motivations (e.g., Fligstein, 1985, 1991;Westphal, Gulati, & Shortell, 1997). This body ofwork suggests that early movers tend to be ra-tional, whereas late movers are often engagedin symbolic action and are merely seeking sta-tus. Such followers are not concerned about in-terpreting the signals of others; rather, by copy-ing more prestigious firms, they are seeking tosend a signal about their own legitimacy.

Viewed in the context of the reputation-basedtheory of economics, such efforts can be seen toenhance the firm’s relations with resource pro-viders if the environment is sufficiently uncer-tain. For instance, followers that entered inter-net markets during the rise of the bubble wereoften able to raise large amounts of capital,despite imitative strategies that later provedhighly flawed. Thus, status-seeking imitationcan benefit the firm and its owners, even if theimitated action is not in the firm’s best interestper se. In this sense, status-seeking imitationcan be rational behavior.

“Legitimation” is another concept of organiza-tion theory that is related to the cascade theo-ries of economics. Scholars of organizationalecology have long noted that once a new indus-try has acquired a threshold number of entrants,the firms acquire a legitimacy that facilitatestheir growth (Carroll & Hannan, 1995; Hannan &

3 Most sociological studies focus on adoption, but some,such as Davis, Diekmann, and Tinsley (1994) and Greve(1995), consider the reverse process of deinstitutionalization.


Carroll, 1992). Banks, for example, become morewilling to supply capital, and potential employ-ees can be hired more easily. This expansion inthe availability of resources, in turn, often leadsto a further wave of entry. Thus, there is athreshold effect in entry processes, similar to theeconomist’s notion of an information cascade.One difference from the economic theory is thatgrowth in the number of entrants increases le-gitimacy while also making competition moreintense. The offsetting force of competitionplaces a ceiling on the equilibrium number offirms. Limits of this sort are not normally con-sidered in the economic theory.

Interactions Between Mimetic and ExperientialLearning

The information-based theories describe pro-cesses where organizations learn by drawinginferences from the behavior of others. Otherforms of learning occur in an industry as moredetailed information emerges from the experi-ence of early movers, and as organizations as-sess their own experience (Baum, Li, & Usher,2000; Haunschild & Miner, 1997). Whether firmsemphasize one mode of learning over anotherdepends on their resources and the time theycan wait before committing to a decision. Expe-rience (or experiment) is more costly and timeconsuming than imitation, which can be viewedas a form of satisficing (Baum et al., 2000; Cyert& March, 1963; March & Simon, 1958). When firmshave adequate time and resources to exploretheir environment extensively, experientiallearning will be preferred. But in highly uncer-tain environments, where quick action is neces-sary, imitating others becomes an attractive de-cision rule. Such a rule appeals most to thosewith little prior information on which to base adecision; more knowledgeable firms may rely onwhat they know internally.

Typically, firms draw on some combination ofthese learning processes. In his studies of for-mat choice by radio stations, Greve (1996) foundthat stations were influenced by the choices ofother units within their corporation (reflecting aprocess of organizational learning), as well asby the choices of independent stations in thegeographic area. Studies of international entrydecisions show that imitation may influence thefirm’s initial decision to enter a country, afterwhich it learns from its own and others’ experi-

ence in that country (e.g., Shaver, Mitchell, &Yeung, 1997).

As a new industry or commercial areaevolves, mimetic, vicarious, and experientiallearning proceed together, often with smallerfirms mimicking the behavior of larger rivals, asthe latter gain information through investmentsin marketing and R&D. Firms may shift fromexternal to internal information sources as theybuild capabilities in a given area. Stuart (1998),for example, found that large semiconductorfirms enter fewer R&D alliances as they developbetter in-house capabilities over time.

Time lags and relative learning rates affectthe dynamics of imitation and the likelihoodthat outcomes will be inferior. If firms perceive aneed to act early in an environment where ex-periential learning is slow, mimetic processescan yield behaviors that are durable, eventhough they may ultimately prove to be highlysuboptimal. If, however, experiential learning isfast, or if firms are able to wait until outcomesare clear, the experience of early movers willresolve many uncertainties and allow followersto converge on good choices.

Followers can sometimes invest in “absorp-tive capacity” (Cohen & Levinthal, 1990) to facil-itate learning from others and to speed imple-mentation. Followers with strong absorptivecapacity may be able to delay commitment andcollect better information without compromisingtheir ability to respond. Absorptive capacity ex-tends the window for effective action, reducingthe risk that the firm will imitate too early or toolate and allowing for better decisions regardingwhether to imitate at all.

In a simulation model of innovation adop-tion, Rosenkopf and Abrahamson (1999) ex-plore the interactions among imitation, uncer-tainty, and learning lags. In their model theprofitability of early adopters is transmitted tothe remaining firms after a lag; followersplace greater weight on this information when“uncertainty” is low. The simulations showthat the longer the learning lag, the more cy-cles during which an imitative bandwagoncan build before it is potentially halted byinformation that the innovation is not profit-able.

Interestingly, the effects of the learning lag onimitation are greatest under conditions of mod-erate uncertainty. When uncertainty is low, mostfirms simply wait to learn whether the innova-


tion is profitable; when uncertainty is high, theinformation on profitability is deemed unreli-able, and, hence, it lacks sufficient weight to cutoff the imitative bandwagon. While the modelshows how learning lags can slow down or stopa growing bandwagon, the more extreme dy-namics of reversals, predicted by the informa-tion cascades theory, are ruled out in the modelby assumption.

Bikhchandani et al. (1992) make the point thatinformation cascades fail to provide “deep”learning; after the start of the cascade, the ac-tions of followers provide no additional informa-tion, since they are simply responding to theinformation revealed by the initial actors. Giventhis shallowness of beliefs, only a small amountof independent learning is needed to overturnthe cascade if the imitated behavior proves er-roneous. Subsequently, a new cascade mayarise once a sufficient number of firms discovera superior alternative. For example, one mightview the continual progression of management“fads” (e.g., “total quality management,” “re-engineering,” “employee empowerment,” etc.)as proceeding roughly in this manner, as firmsimitate organizations they believe to be betterinformed but discover, through experience, thelimits of the new managerial system.

Such reversals are not emphasized in the or-ganizational theories, which take the mimeticbehavior as more durable. (Abrahamson’s [1991,1996] work on management fads is a notableexception.) One reason for the difference in em-phasis is that the sociological studies focus onadoption of organizational innovations forwhich information lags are typically long andresidual uncertainty high. In many other do-mains where imitation occurs, uncertainty is re-solved more quickly or completely. This makesreversals more likely and may prevent the imi-tation entirely if firms wait to learn from theexperience of early movers.

THEORIES RELATING TO COMPETITIVERIVALRY AND RISK

A second set of theories regards imitation as aresponse designed to mitigate competitive ri-valry or risk. Firms imitate others in an effort tomaintain their relative position or to neutralizethe aggressive actions of rivals. Unlike in thetheories discussed in the previous section, firms’actions do not convey information. The theories

relating to rivalry and risk have their primaryorigin in the fields of economics and businessstrategy.

Imitation to mitigate rivalry is most commonwhen firms with comparable resource endow-ments and market positions face one another.Competition can be very intense in such cases,with prices and profits easily eroded (Peteraf,1993). To alleviate this situation, firms can pur-sue either differentiation or homogeneous strat-egies (Baum & Haveman, 1997; Deephouse, 1999;Gimeno & Chen, 1998). Firms that differentiatetheir resources and market position from thoseof competitors become insulated from the ac-tions of rivals. This reduces the likelihood ofimitation and leads to higher profits, if the dif-ferentiated position proves sufficiently attrac-tive. Pursuing a differentiation strategy, how-ever, is often difficult and risky. A firm cannot becertain that the new position or niche will besuperior. Faced with a choice, firms thereforeoften choose to pursue homogeneous strategies,where they match the behavior of rivals in aneffort to ease the intensity of competition or re-duce risk.

Homogeneous Strategies to Mitigate Rivalry

When resource homogeneity creates the po-tential for intense competition, matching behav-ior may be a way to enforce tacit collusionamong rivals. Studies of repeated games showhow “tit for tat” strategies can punish deviantbehavior, thereby maintaining cooperation (Ax-elrod, 1984). In his early work on strategicgroups, Porter suggested that firms within thesame group behave similarly because “diver-gent strategies reduce the ability of the oligopo-lists to coordinate their actions tacitly . . . reduc-ing average industry profitability” (1979: 217). Inother words, firms within the same strategicgroup may adopt similar behavior to constraincompetition and maintain tacit collusion.4

More recent work in strategy and economicsgives similar predictions. Studies on action-response dyads (Chen & MacMillan, 1992; Chen,Smith, & Grimm, 1992) suggest that matching a

4 While strategic groups may be able to sustain tacit col-lusion in this way, firms within a strategic group typicallyexperience more competition among their group membersthan with members of other strategic groups within thesame industry (Greve, 1996).


competitor’s move indicates a commitment todefend the status quo, neither giving up thecurrent position nor falling into mutually de-structive warfare. Similarly, Klemperer (1992)shows that competitors may duplicate theirproduct lines to mitigate rivalry. If firms offeridentical product ranges, each consumer canavoid the costs of dealing with multiple firms byselecting a single supplier. This segmentationof customers may make the market less compet-itive.

The hypothesis that firms adopt similar be-havior to mitigate rivalry can be also derivedfrom studies on multimarket contact (Bernheim& Whinston, 1990; Karnani & Wernerfelt, 1985;Leahy & Pavelin, 2003). Edwards (1955) was thefirst to argue that multimarket contact mightblunt the edge of competition, because “a pros-pect of advantage from vigorous competition inone market may be weighed against the dangerof retaliatory forays by the competitor in othermarkets” (from Corwin Edwards’ testimony,cited in Scherer, 1980: 340). When firms competewith each other in many markets, they can moreeasily sustain collusion, because deviations inone market can be met by aggressive responsesin many places. This is the idea of “mutual for-bearance.”5 The multimarket contact theoriessuggest two ways that competitors may imitate:(1) they may respond to a rival’s aggressivemove in one market with a similar move in an-other market; (2) they may match rivals’ entrydecisions in order to increase the degree of con-tact.

Risk Minimization

Other researchers have proposed that imita-tion stems from the desire of rivals to maintainrelative competitive position. One of the firstdocumented examples was the “bunching” offoreign direct investment (FDI), as rivals

matched each other’s entries into foreign mar-kets. Knickerbocker (1973) argued that such “fol-low-the-leader” behavior is the result of riskminimization. If rivals match each other, nonebecome better or worse off relative to each other.This strategy guarantees that their competitivecapabilities remain roughly in balance.

Motta (1994) gives a game theoretic explana-tion for this follow-the-leader behavior, andHead, Mayer, and Ries (2002) show that it can besustained only when managers are risk averse.Table 1 lists many empirical studies that pro-vide evidence of the existence of follow-the-leader behavior in foreign market entry (e.g.,Caves, Porter, & Spence, 1980; Flowers, 1976;Knickerbocker, 1973; Yamawaki, 1998; Yu & Ito,1988). Other studies in the strategic group liter-ature (e.g., Fiegenbaum & Thomas, 1995; Garcia-Pont & Nohria, 2002) show that firms are likely toimitate other group members, in an effort tomaintain competitive parity.

In domains such as FDI and other types ofmarket entry, the incentives for imitation shoulddiminish, as more firms follow and competitionintensifies within the niche. Thus, there may bea self-limiting dynamic to some rivalry-basedimitation, as in the population ecology theoriesdiscussed previously. Supporting this idea, Mar-tin, Swaminathan, and Mitchell (1998) found thelikelihood that a given Japanese automotivesupplier would enter the North American marketrose and then fell with the number of competingsuppliers that had already entered. In domainswhere this offsetting force of competition is lack-ing (e.g., imitation of organizational structures),imitation can be more widespread.

In “winner-takes-all” environments, rivalfirms may adopt similar behavior to prevent oth-ers from leading the race. For example, in R&Dcompetition, where the first inventor can obtainpatent rights to a technology so that other firmscannot use it, R&D investments among firms arepositively correlated. Such competition leads tooverinvestment (Dasgupta & Stigliz, 1980).6 Sim-ilar winner-takes-all situations can arise whenthe market has bandwagon effects or network

5 Empirical studies, however, often fail to support the mu-tual forbearance hypothesis (Heggestad & Rhoades, 1978;Rhoades & Heggestad, 1985; Scott, 1982). Bernheim andWhinston (1990) and Karnani and Wernerfelt (1985) suggestthat the ambiguous empirical results of the existing studiesare due to different effects of multimarket contact, depend-ing on the characteristics of markets and firms. Controllingcarefully for such characteristics, several recent empiricalstudies on multimarket contact support the mutual forbear-ance hypothesis (Evans & Kessides, 1994; Gimeno & Woo,1996).

6 In an empirical study of the pharmaceutical industry,Cockburn and Henderson (1994) found little evidence of thecorrelated R&D responses predicted by economic theory.They suggest that winner-takes-all situations are rare; R&Draces typically yield multiple prizes.


externalities (Katz & Shapiro, 1985; Leibenstein,1950).

DISTINGUISHING AMONG IMITATIONPROCESSES

The information- and rivalry-based theoriesdescribed above are not mutually exclusive;both types of imitation can occur simulta-neously. Firms may imitate rivals to maintaincompetitive parity and also out of the belief thatrivals may possess superior information. Never-theless, one type of imitation or the other is aptto predominate in any given context.

In this section we draw some predictionsabout the conditions under which each type ofimitation is most likely. These distinctions pro-vide guidance for researchers and managers at-tempting to identify mimetic behavior and toassess the potential consequences of imitation.In addition, we consider the problem of distin-guishing imitation from other types of isomor-phism, including the basic case where firms re-spond independently but identically to acommon environmental shock.

Using Environmental Conditions to DistinguishAmong Theories

Empirical researchers have often sought tofind evidence of imitation, taking one specifictheory or type of imitation as given. For exam-ple, in many studies in the international busi-ness literature, scholars have found evidence ofthe bunching of entry by foreign firms as theassumed consequence of interfirm rivalry. Otherresearchers in organizational ecology havecommonly found a surge of entry once a newindustry achieves “legitimacy” (cf. Carroll &Hannan, 1995; Hannan & Carroll, 1992). Whilethese presumed mechanisms of imitation maybe valid in their respective contexts, more workis needed to test alternative theories and to linktheories to the environmental conditions wherethey are most applicable.

The flowchart in Figure 1 applies three crite-ria to help distinguish between information-based and rivalry-based imitation. The first twocriteria, market overlap and resource similarity,establish whether the leader(s) and followercompete as rivals. Rivals have strong overlap inproduct lines and geographic market coverage.

Often, they have similar resources, and theymay share similar origins and history.

If the firms are not rivals, the follower can bejudged as having information-based motives forimitation. In general, information-based mo-tives are likely to be dominant when firms differin market position, size, or resources, or whenuncertainty is very high. Asymmetry limits ri-valry and raises the likelihood that some firmspossess superior information. High uncertaintyimplies that managers have weak “prior proba-bilities” about the likely success of alternativepaths and are therefore more open to externalsources of information. Furthermore, patternsmay be observed—small firms following largerfirms or general imitation of successful firms—suggesting that the imitation process is infor-mation based.

FIGURE 1Conditions to Distinguish Between

Information-Based and Rivalry-Based Imitation


If firms do compete as rivals, both types ofimitation may coexist. Even so, the third crite-rion (degree of uncertainty) has some power todistinguish between the two motives for imita-tion. Rivalry-based motives are likely to domi-nate when uncertainty is low or when competi-tors are closely matched; such firms often havesimilar information but strong rivalry. Multimar-ket contact (not incorporated in Figure 1) furtherincreases the likelihood of rivalry-based imita-tion, since it expands the domains where imita-tion can occur and raises the probability thatfirms respond to each other in kind. Firms thatare closely matched may also be risk averse,particularly to loss of market share—a conditionthat is necessary for some types of rivalry-basedimitation.

In several empirical studies researchers haveapplied such criteria to classify imitation pro-cesses. In a study of entry by U.S. telecommuni-cations firms into foreign markets, Gimeno,Hoskisson, Beal, and Wan (2005) used differ-ences in domestic market overlap to distinguishrivalry-based imitation. They found a clusteringof foreign entries by firms that competed di-rectly with each other in regional U.S. markets,but no such pattern for local monopolist “BabyBells.” This suggests rivalry as the dominantmotive for the bunching of foreign market entry,a conclusion consistent with the assumptions ofprior FDI studies. In another study (Asaba &Lieberman, 1999), we used differences in uncer-tainty to distinguish among imitation processesrelating to new product introductions in the Jap-anese soft drink industry. We found a tendencyfor larger firms to be followed in cases of majorinnovations where uncertainty was high,whereas close rivals were followed for incre-mental product changes. This supports the ideathat information-based motives prevail underconditions of high uncertainty, but rivalry mo-tives prevail when uncertainty is low. In a studyof investment timing by chemical producers,Gilbert and Lieberman (1987) found a pattern inwhich small firms mimicked the capacity ex-pansions of large firms, whereas the latteravoided imitating each other for fear of creatingovercapacity. Such a pattern is consistent withinformation-based motives, where small pro-ducers draw on the superior ability of largefirms to forecast growth in demand.

To be sure, Figure 1 does not provide a perfectguide for distinguishing between information

and rivalry motives. When firms are direct com-petitors, the two sets of motives may be closelyintertwined. Rivals that share common technol-ogy, organization, and market orientation maybe particularly informative to each other. Evenso, differences in the degree of environmentaluncertainty provide some basis for judging therelative importance of the two motives. Amongexamples we consider, the degree of uncertaintyvaries greatly. At one extreme, internet com-merce in the late 1990s was characterized byenormous uncertainty and ambiguity. In this en-vironment, imitation was likely to have beenmostly information based, even for firms com-peting with one another. In comparison, uncer-tainty was relatively low for incremental prod-uct enhancements in the calculator industry,where rivalry was likely the primary motive forimitation. Other examples where both motivesfor imitation seem to have been strong (e.g.,Japanese FDI in North America) are intermedi-ate between these extremes.

Identical Responses to Common EnvironmentalShock

We have argued that fundamental character-istics of the industry environment and the iden-tity of initiating firms provide a basis for distin-guishing between information- and rivalry-based imitation processes. One complication isthat both types of imitation may occur simulta-neously, even though one is predominant. A fur-ther complication is that what looks like imita-tion may simply be firms’ independent responsesto a common external stimulus. For example, con-sider an economic recession that induces manyfirms to lay off part of their workforce. Such layoffdecisions are made primarily on the basis of fore-casts of future sales. To the extent that firms aresubject to the same demand fluctuations and haveaccess to the same public information about mac-roeconomic conditions, one would expect them tomake reasonably similar and simultaneous cut-backs. Regarding such behavior as imitationwould clearly be incorrect.

Some degree of imitation may neverthelessoccur in such situations, stemming from infor-mation or rivalry motives (or both). For example,firms may look to the announcements of othersas a source of information about the likely depthof the recession in their industry. Similarly, ifrivals have not yet announced layoffs, a firm


may be reluctant to act alone for fear that itcould lose competitive position. In such in-stances, once one firm announces cutbacks, oth-ers may follow suit.

Thus, we often observe the confluence of bothimitative and nonimitative responses to exter-nal shocks. This simultaneity makes clearcutidentification of imitative behavior a thornyproblem for empirical researchers attempting tocharacterize imitation processes. When data areavailable on many organizational units, thetime lags associated with imitation provide ameans of distinguishing between the two typesof response. The diffusion framework of Strangand Tuma (1993), which provides the basis ofrecent applied work by organization research-ers, allows one to separate independent re-sponses to a shock (in the intrinsic propensityvector) from contagion effects (in the suscepti-bility, proximity, and infectiousness vectors).Strang and Tuma’s event-history method allowsone to characterize contagion processes withina population of potential adopters. However, themethod may be less useful for distinguishingimitation and related processes among smallnumbers of competitors within a market orniche. In general, economists have been muchless sanguine than sociologists about the abilityto carry out empirical research on social inter-actions (e.g., Manski, 2000).

Resource and Complexity Constraints onImitative Behavior

Imitation processes are also influenced by re-source constraints that limit the scope of firms’behavior. Firms with very different resource en-dowments may be unable to behave similarly,even if they face the same environment. This isbecause strategy is constrained by the currentlevel of resources, as many scholars of the re-source-based view of the firm point out (e.g.,Collis, 1991; Teece, Pisano, & Shuen, 1997). Firmsmay be able to mimic others only when theirresource endowments are comparable. In thepetroleum industry, for example, Helfat (1997)shows that synthetic fuels became an attractiveopportunity when oil prices rose sharply in the1970s, but only those firms with requisite exper-tise in petroleum refining R&D were able to in-vest. Since firms with similar resources are oftendirect rivals, resource constraints can make itappear that rivals are responding to each other,

even though their actions are independent re-sponses to a common environmental shock (asin the case of synthetic fuels).

Complexity serves as a further constraint onimitative behavior. Firms with adequate re-sources can easily copy simple actions but notcomplex repertoires containing many elements,particularly when tacit skills are involved.Causal ambiguity about which elements aremost important leads to “uncertain imitability”(Lippman & Rumelt, 1982). If many elements andtheir interactions must be duplicated to achievesuccess, the sheer burden of the task may pre-vent imitation (Rivkin, 2000). Applied studies bySzulanski (1996) and Ounjian and Carne (1987)confirm that imitation is impeded by causal am-biguity and complexity.

PERFORMANCE IMPLICATIONS

In previous sections of this article we de-scribed the amplification effects of imitationthat make outcomes more extreme, with conse-quences that may be good or bad for both firmsand society. On the positive side, information-based imitation can speed the adoption of use-ful innovations, and rivalry-based imitation canspur firms to improve their products and ser-vices. Both types of imitation have negative im-plications if they lead firms to squander re-sources on wasteful, duplicative investments.Thus, the two modes of imitation can have sim-ilar effects, although there can be important dif-ferences as well. We first address performanceimplications that apply to both types of imita-tion, followed by more specific implications ofinformation- and rivalry-based processes.

Performance Implications of Both ImitationTypes

Imitation processes lead firms to converge oncommon choices more rapidly and in largernumbers than they would otherwise. The conse-quences, when beneficial, are reasonablystraightforward, but when negative they are of-ten dramatic. Industries may lock in to inferiorchoices or greatly overshoot the optimum levelof investment. As discussed previously, imita-tion helped to promote the early boom-bust cy-cle of internet commerce and numerous busi-ness fads. Imitative investments in optical fibercables during the internet boom led to a glut of


telecommunications capacity, culminating inspectacular failures by Global Crossing andWorldCom—the latter representing the largestcorporate bankruptcy in history (Fransman,2004). Such examples suggest that dysfunctionalimitation abounds, even though it is hard todocument definitively.

If early movers have chosen a productivepath, imitation accelerates the industry’s con-vergence on a good solution. Imitation can helpto promote network effects and common stan-dards, with broad potential benefits for firmsand consumers. In the emerging market forVCRs, for example, Japanese producers bene-fited from their early convergence on magnetictape as the storage medium. Sony, the Japanesepioneer, had correctly recognized that tape wassuperior to alternatives being pursued in the1970s, such as the videodisk developed by RCA.Sony’s Japanese rivals focused their efforts onimproving magnetic tape technology, thespeedy development of which enabled Japanesemanufacturers to dominate the global market(Rosenbloom & Cusumano, 1987).

If the wrong path is chosen, however, imita-tion can be costly for firms and for society. In theearly years of high-definition television (HDTV),Japanese electronics firms adopted analog tech-nology and heavily promoted its development.Eventually, it became clear that the analog ap-proach was inferior to digital. Despite their dom-inance in many areas of consumer electronics,the Japanese firms found themselves at a seri-ous disadvantage in world markets for HDTV,with the result that the growth of HDTV in Japanand elsewhere was hampered.

The VCR and HDTV examples illustrate thatimitation raises the odds of extreme outcomeswhen the environment is uncertain. On the onehand, if the leaders have superior informationand luck, imitation leads to quick convergenceon superior choices and is socially beneficial.Rivalry and shared learning may stimulatefirms and accelerate progress. On the otherhand, if the path that is imitated proves inferior,imitation can create an industry-wide “compe-tency trap” (Levitt & March, 1988; Miner & Haun-schild, 1995). In comparison, when firms act in-dependently, they converge more slowly, butsuch diversity avoids the worst industry out-comes and is collectively more robust.

Thus, by reducing variation in firms’ strate-gies and technological paths, imitation raises

the collective risk of an industry. When firmsimitate each other in an uncertain environment,they place identical bets on the future, therebyraising the odds of large positive or negativeoutcomes. As a result, society bears a higherrisk, even though individual firms may diminishtheir risk of falling behind rivals.

The propensity of firms to imitate may be cul-turally or socially influenced. Some societiesmay be more prone to imitation and, as a result,may show a wider range of performance varia-tion across industries. For example, the ten-dency to copy rivals is often considered partic-ularly strong in Japan (Asaba, 1999). Suchtendencies may contribute to that nation’s supe-rior record of performance in some economicareas but to deep weaknesses in others.

Imitation tends to be socially beneficial—andpotentially profitable—in situations where theimitators complement each other. Complemen-tarities often arise in environments with net-work externalities or agglomeration economies.For example, Baum and Haveman (1997) foundthat hoteliers tend to locate new hotels close toestablished hotels. This agglomeration attractspeople, goods, and services, and, consequently,it increases the attractiveness and reputation ofthe location, which is beneficial to society aswell as to the hotels. At the same time, however,the close location of hotels can intensify pricecompetition, making hotels less profitable.Thus, for imitating firms, the benefits of networkeffects, agglomeration economies, and otherpositive externalities can be offset by pressurefor price competition.

Information-Based Imitation

Although both types of imitation can have am-plification effects, dramatic negative outcomesare more likely with information-based imita-tion. The information cascades theory is explicitabout the potential for bubbles and sudden re-versals. Other work in organization theoryshows how lags in learning processes allowbandwagons to grow. The risk of inferior out-comes is greatest if managers perceive a needto commit before major uncertainties are re-solved. During the rise of internet commerce, forexample, widespread belief in early mover ad-vantages led to a rush of commercial efforts(Lieberman, 2005). Eventually, as more informa-tion emerged about the prospects for internet


businesses, stock prices collapsed and manyfirms failed. In retrospect, it is clear that much ofthe initial rush was unnecessary and contrib-uted to the magnitude of the collapse. Had morefirms waited until major uncertainties were re-solved, many losses could have been avoided.

The speed with which uncertainty is resolveddepends on context. New products may succeedor fail within months, and key uncertainties sur-rounding new technologies are often resolvedwithin a few years. Organizational innova-tions are characterized by longer gestationlags and more residual uncertainty. Conse-quently, one often observes dysfunctional im-itation of organizational innovations and therelated phenomenon of managerial fads.Abrahamson (1991) suggests that many man-agers misunderstand the lag structure andabandon fads too quickly, following the nextwave of imitation before most benefits of theprior wave have been realized.

Individual firms fail when they attempt to im-itate a successful leader but prove incapable ofdoing so. Smaller firms may imitate in an effortto elevate their status or legitimacy, despite alack of resources to do so successfully (Fligstein,1985, 1991). Observation of the successful actionsof others may raise aspiration levels beyondwhat can realistically be attained (Greve, 1998).Moreover, even large firms may imitate the su-perficial features of complex innovations whilefailing to replicate more subtle but essentialelements. Thus, followers fail when they lackcritical resources or when complexity, tacitness,and causal ambiguity prevent them from gain-ing a sufficient understanding of the innova-tions made by the target firm.

Rivalry-Based Imitation

The theories presented earlier suggest thatrivalry-based imitation can reduce the intensityof competition in an industry—or increase it.Here again, we have possibilities for diametri-cally opposite outcomes. For example, imitationmay lead firms to cut back on R&D, as in the“smog” case cited in the introduction, or to raiseR&D investment, as in the Japanese calculatorand VCR examples. Theory offers some basis forpredicting which outcome will prevail: collusionbecomes more likely when firms have multimar-ket contact, whereas competition is promoted inwinner-takes-all environments. Empirical stud-

ies suggest that, in most cases, rivalry-basedimitation raises the intensity of competition andlowers profitability (Barreto & Baden-Fuller, inpress; Deephouse, 1999; Ghemawat, 1991; Oda-giri, 1992). One conclusion is that intensifica-tion of competition is most common, but eithertype of response can arise, depending on as-pects of firm interaction and history that canbe subtle and difficult to observe (Kreps &Spence, 1985).

Such dichotomization between competitionand collusion may, however, be too simple. Ri-valry-based imitation often proceeds over manyrounds, where firms repeatedly match each oth-er’s moves. This process can strengthen firmsthat imitate relative to those that do not, thusmaking it a form of the “red queen” effect dis-cussed in the organizational literature (Barnett& Hansen, 1996; Barnett & Sorenson, 2002). Suchimitation leads to differential performanceamong groups of firms and can create barriersto entry. If innovation is promoted and pricesfall, the process is beneficial to consumers, butif only a few firms survive, it can lead to anincrease in market power.

The electronic calculator industry providesone example. Casio and Sharp responded toeach other by introducing many new productfeatures and cost reductions, leading to marketgrowth and gains for consumers. Similarity ofproduct and market position made each firm agood reference for the other, which facilitatedlearning. Ultimately, the accumulation of prod-uct enhancements enabled Casio and Sharp todrive out their American rivals, who had pio-neered the basic technology. Despite the lossesto American producers, it seems likely that thisprocess was socially beneficial.

A related example concerns Coke and Pepsi,which matched each other’s advertising, promo-tion, and new product moves in the U.S. softdrink market over many decades (Moriguchi &Lane, 1999). Challenging and learning from eachother, the two rivals became progressivelystronger, squeezing out smaller producers whilemaintaining high profitability. One feature ofthe soft drink industry is that it has supportedmany dimensions of multimarket contact (overproducts, regions, etc.), which may have helpedCoke and Pepsi to signal each other and preventmutually destructive warfare.


CONCLUSIONS AND IMPLICATIONS FORFUTURE RESEARCH

We have surveyed theories of business imita-tion and have shown that they fall into twobroad categories: information-based theoriesand rivalry-based theories. The two types of im-itation have different implications, althoughboth have amplification properties that makeoutcomes more extreme. Information-based im-itation can speed the adoption of superior prod-ucts and methods, or it can lead to dramaticfailures, as the internet examples we cite attest.Rivalry-based imitation can facilitate collusion,although more commonly it intensifies competi-tion. In the latter case, imitation may proceedover many rounds, strengthening firms if theyhave chosen a productive path, or leading themfurther astray if they have not.

We have suggested some ways that the twotypes of imitation can be identified and distin-guished in empirical work. We have alsopointed out that this is not an easy task. Severalvexing problems make identification difficult:firms may respond identically (but not imita-tively) to common environmental stimuli, thetwo types of imitation may coexist when firmsare rivals, and key distinguishing characteris-tics (such as the “degree of environmental un-certainty”) may be hard to assess objectively.Identification of imitation processes thereforeremains a challenge for those engaged in ap-plied research.

Despite such difficulties, more research seemswarranted, given the prevalence of business im-itation and its potential consequences. Thissurvey has reviewed studies from a range ofacademic disciplines. We see abundant oppor-tunities for cross-fertilization, particularly be-tween economists and organizational scholars.

Economists have modeled information-basedimitation in a stylized way that offers concep-tual precision at the expense of recognizing thevaried forms of learning that operate in practice.In comparison, studies from the perspective oforganization theory offer a more comprehensiveview of learning, as well as insights about therole of communication networks in shaping thepath of imitation. Organizations scholars haveshown how network structure, concurrent learn-ing processes, and information lags can haveimportant effects on imitative outcomes. Such

features might usefully be incorporated in eco-nomic models.

Economists have carried out relatively fewempirical studies of imitation. Most of thesestudies have focused on the bunching of FDI.Recently, economists have tested for the effectsof information cascades in financial markets(Bikhchandani & Sharma, 2001; Hirshleifer &Teoh, 2003). We see many opportunities for stud-ies of imitation beyond the domains of financialmarkets and FDI. In terms of statistical tools, theevent history models of organizational sociologymay be applicable in economics-oriented re-search.

For organizational researchers, the tight logicof economic reasoning can bring greater clarityto work on imitation. In particular, more explicitrecognition should be made of the informationalaspects of imitation. Indeed, we have arguedthat many insights of institutional theory can beviewed as signaling processes and rational re-sponses to revealed information.

Most empirical studies of imitation in the or-ganization literature have focused on the adop-tion of organizational innovations and practices(or market entry, in the case of studies within thesubfield of population ecology). Researchersmight consider a broader set of domains whereimitation processes arise, including new prod-uct introductions, capacity expansion, R&D, andother forms of business investment. At the veryleast, researchers should recognize that the im-itation issues addressed in the organization lit-erature are a subset of a larger class.

We also see opportunities for studies that ex-plore interactions between mimetic and experi-ential forms of learning. Firms draw inferencesfrom the observed behavior of others, from directcommunication with others, and from their ownexperience; all three types of learning are im-portant. Nevertheless, the literature on imitationand that on organizational learning haveevolved almost completely independently ofeach other. Realistic models of learning mustincorporate and integrate these alternativemodes of information acquisition.

Finally, we suggest that all researchers takemore explicit account of the costs and benefits ofimitation. This study has highlighted the ampli-fication properties that make outcomes more ex-treme. We have shown that imitation has manypotential benefits: it speeds the adoption of in-novations (beneficial, if the innovation proves


useful), it can intensify pressures for firms toimprove their products, and it can promote net-work effects and other positive externalities andcomplementarities. On the negative side, imita-tion can lead to destructive competition, overin-vestment, reduced variety, and increased risk. Insome situations, imitation can support anticom-petitive outcomes. From a policy perspective, itwould be useful to improve our understandingof the benefits and costs of imitation in specificcontexts in order to better anticipate situationswhere imitation is likely to prove detrimental.

REFERENCES

Abrahamson, E. 1991. Managerial fads and fashions: Thediffusion and rejection of innovations. Academy of Man-agement Review, 16: 586–612.

Abrahamson, E. 1996. Management fashion. Academy ofManagement Review, 21: 254–285.

Abrahamson, E., & Rosenkopf, L. 1997. Social network effectson the extent of innovation diffusion: A computer simu-lation. Organization Science, 8: 289–309.

Asaba, S., 1999. Behavioral similarity among Japanese firms.Doctoral dissertation, UCLA Anderson School of Man-agement, Los Angeles.

Asaba, S., & Lieberman, M. B. 1999. Why do firms behavesimilarly? A study on new product introductions in theJapanese soft-drink industry. Paper presented at the an-nual meeting of the Academy of Management, Chicago.

Axelrod, R. 1984. The evolution of cooperation. New York:Basic Books.

Banerjee, A. V. 1992. A simple model of herd behavior. Quar-terly Journal of Economics, 107: 797–817.

Barnett, W. P. 1993. Strategic deterrence among multipointcompetitors. Industrial and Corporate Change, 2: 249–278.

Barnett W. P., & Hansen M. 1996. The red queen in organiza-tional evolution. Strategic Management Journal, 17(Spe-cial Issue): 139–157.

Barnett W. P., & Sorenson, O. 2002. The red queen in organi-zational creation and development. Industrial and Cor-porate Change, 11: 289–325.

Barreto, I., & Baden-Fuller, C. In press. To conform or toperform: Mimetic behavior, legitimacy-based groupsand performance consequences. Journal of ManagementStudies.

Baum, J. A. C., & Haveman, H. A. 1997. Love thy neighbor?Differentiation and agglomeration in the Manhattan ho-tel industry, 1989–1990. Administrative Science Quar-terly, 42: 304–338.

Baum, J. A. C., & Korn, H. J. 1996. Competitive dynamics ofinterfirm rivalry. Academy of Management Journal, 39:255–291.

Baum, J. A. C., & Korn, H. J. 1999. Dynamics of dyadic compe-tition. Strategic Management Journal, 20: 251–278.

Baum, J. A. C., Li, S. X., & Usher, J. M. 2000. Making the nextmove: How experiential and vicarious learning shapethe locations of chains’ acquisitions. Administrative Sci-ence Quarterly, 45: 766–801.

Bernheim, B. D., & Whinston, M. D. 1990. Multimarket contactand collusive behavior. Rand Journal of Economics, 21:1–26.

Bikhchandani, S., Hirshleifer, D., & Welch, I. 1992. A theory offads, fashion, custom, and cultural change as informa-tional cascades. Journal of Political Economy, 100: 992–1026.

Bikhchandani, S., Hirshleifer, D., & Welch, I. 1998. Learningfrom the behavior of others: Conformity, fads, and infor-mational cascades. Journal of Economic Perspectives,12: 151–170.

Bikhchandani, S., & Sharma, S. 2001. Herd behavior in finan-cial markets: A review. IMF Staff Papers, 47: 279–310.

Boeker, W. J. G., Stephan, J., & Murmann, J. P. 1997. Compe-tition in a multimarket environment: The case of marketexit. Organization Science, 8: 126–142.

Carroll, G. R., & Hannan, M. T. 1995. Organizations in indus-try: Strategy, structure and selection. New York: OxfordUniversity Press.

Caves, R. E., Porter, M. E., & Spence, A. M., with Scott, J. T.1980. Competition in the open economy: A model appliedto Canada. Cambridge, MA: Cambridge UniversityPress.

Chang, A., Chaudhuri, S., & Jayaratne, J. 1997. Rational herd-ing and the spatial clustering of bank branches: Anempirical analysis. Research paper No. 9724. New York:Federal Reserve Bank.

Chen, M.-J., & MacMillan, I. C. 1992. Nonresponse and de-layed response to competitive moves: The roles of com-petitor dependence and action irreversibility. Academyof Management Journal, 35: 539–570.

Chen, M.-J., Smith, K. G., & Grimm, C. M. 1992. Action char-acteristics as predictors of competitive responses. Man-agement Science, 38: 439–455.

Cockburn, I., & Henderson, R. 1994. Racing to invest? Thedynamics of competition in ethical drug discovery. Jour-nal of Economics and Management Strategy, 3: 481–519.

Cohen, W. M., & Levinthal, D. A. 1990. Absorptive capacity: Anew perspective on learning and innovation. Adminis-trative Science Quarterly, 35: 128–152.

Collis, D. J. 1991. A resource-based analysis of global com-petition: The case of the bearings industry. StrategicManagement Journal, 12: 49–68.

Cyert, R. M., & March, J. G. 1963. A behavioral theory of thefirm. Englewood Cliffs, NJ: Prentice-Hall.

Dasgupta, P., & Stiglitz, J. E. 1980. Industrial structure and thenature of innovative activity. Economic Journal, 90: 266–293.

Davis, G. F. 1991. Agents without principles? The spread of


the poison pill through the intercorporate network. Ad-ministrative Science Quarterly, 38: 583–613.

Davis, G. F., Diekmann, K. A., & Tinsley, C. H. 1994. Thedecline and fall of the conglomerate firm in the 1980s:The deinstitutionalization of an organizational form.American Sociological Review, 59: 547–570.

Deephouse, D. L. 1999. To be different, or to be the same? It’sa question (and theory) of strategic balance. StrategicManagement Journal, 20: 147–166.

DiMaggio, P. J., & Powell, W. W. 1983. The iron cage revisted:Institutional isomorphism and collective rationality inorganization fields. American Sociological Review, 48:147–160.

Edwards, C. D. 1955. Conglomerate bigness as a source ofpower. In Business concentration and price policy: 331–352. NBER conference report. Princeton, NJ: PrincetonUniversity Press.

Evans, W. M., & Kessides, I. N. 1994. Living by the “GoldenRule”: Multimarket contact in the U.S. airline industry.Quarterly Journal of Economics, 109: 341–366.

Fiegenbaum, A., & Thomas, H. 1995. Strategic groups asreference groups: Theory, modeling and empirical ex-amination of industry and competitive strategy. Strate-gic Management Journal, 16: 461–476.

Fligstein, N. 1985. The spread of the multidivisional formamong large firms, 1919–1979. American SociologicalReview, 50: 377–391.

Fligstein, N. 1991. The structural transformation of Americanindustry: An institutional account of the causes of diver-sification in the largest firms, 1919–1979. In W. W. Powell& P. J. DiMaggio (Eds.), The new institutionalism in or-ganization analysis: 311–336. Chicago: University of Chi-cago Press.

Flowers, E. B. 1976. Oligopolistic reactions in European andCanadian direct investment in the United States. Jour-nal of International Business Studies, 7: 43–55.

Fransman, M. 2004. The telecoms boom and bust 1996–2003and the role of financial markets. Journal of Evolution-ary Economics, 14: 369–406.

Garcia-Pont, C., & Nohria, N. 2002. Local versus global mi-metism: The dynamics of alliance formation in the au-tomobile industry. Strategic Management Journal, 23:307–321.

Ghemawat, P. 1991. Commitment: The dynamics of strategy.New York: Free Press.

Gilbert, R. J., & Lieberman, M. B. 1987. Investment and coor-dination in oligopolistic industries. Rand Journal of Eco-nomics, 18: 17–33.

Gimeno, J., & Chen, M.-J. 1998. The dynamics of competitivepositioning: A pair-wise perspective. Academy of Man-agement Proceedings: 01–08.

Gimeno, J., Hoskisson, R. E., Beal, B. D., & Wan, W. P. 2005.Explaining the clustering of international expansionmoves: A critical test in the U.S. telecommunicationsindustry. Academy of Management Journal, 48: 297–319.

Gimeno, J., & Woo, C. Y. 1996. Hypercompetition in a multi-market environment: The role of strategic similarity and

multimarket contact in competitive de-escalation. Orga-nization Science, 7: 322–341.

Granovetter, M. 1985. Economic action and social structure:The problem of embeddedness. American Journal of So-ciology, 91: 481–510.

Greve, H. R. 1995. Jumping ship: The diffusion of strategyabandonment. Administrative Science Quarterly, 40:444–473.

Greve, H. R. 1996. Patterns of competition: The diffusion of amarket position in radio broadcasting. AdministrativeScience Quarterly, 41: 29–60.

Greve, H. R. 1998. Performance, aspirations and risky orga-nizational change. Administrative Science Quarterly, 43:58–86.

Gulati, R., & Gargiulo, M. Where do organizational networkscome from? American Journal of Sociology, 104: 1439–1493.

Gulati, R., Nohria, N. & Zaheer, A. 2000. Strategic networks.Strategic Management Journal, 21: 203–215.

Hannan, M. T., & Carroll, G. R. 1992. Dynamics of organiza-tional populations: Density, competition and legitima-tion. New York: Oxford University Press.

Haveman, H. A. 1993. Follow the leader: Mimetic isomor-phism and entry into new markets. Administrative Sci-ence Quarterly, 38: 593–627.

Haunschild, P. R. 1993. Interorganizational imitation: Theimpact of interlocks on corporate acquisition activity.Administrative Science Quarterly, 38: 564–592.

Haunschild, P. R., & Miner, A. S. 1997. Modes of interorgani-zational imitations: The effects of outcome salience anduncertainty. Administrative Science Quarterly, 42: 472–500.

Hawley, A. H. 1986. Human ecology: A theoretical essay.Chicago: University of Chicago Press.

Head, K., Mayer, T., & Ries, J. 2002. Revisiting oligopolisticreaction: Are decisions on foreign direct investmentsstrategic complements? Journal of Economics and Man-agement Strategy, 11: 453–472.

Heggestad, A. A., & Rhoades, S. A. 1978. Multimarket inter-dependence and local market competition in banking.Review of Economics and Statistics, 60: 523–532.

Henisz, W. J., & Delios, A. 2001. Uncertainty, imitation, andplant location: Japanese multinational corporations,1990–1996. Administrative Science Quarterly, 46: 443–475.

Helfat, C. 1997. Know-how and asset complementarity anddynamic capability accumulation: The case of R&D.Strategic Management Journal, 18: 339–360.

Hennart, J.-F., & Park, Y.-R. 1994. Location, governance, andstrategic determinants of Japanese manufacturing in-vestment in the United States. Strategic ManagementJournal, 15: 419–436.

Hirshleifer, D., & Teoh, S. H. 2003. Herd behavior and cascad-ing in capital markets: A review and synthesis. Euro-pean Financial Management, 9: 25–66.


Karnani, A., & Wernerfelt, B. 1985. Multiple point competition.Strategic Management Journal, 6: 87–96.

Katz, M. L., & Shapiro, C. 1985. Network externalities, com-petition, and compatibility. American Economic Review,75: 424–440.

Kennedy, R. E. 2002. Strategy fads and competitive conver-gence: An empirical test for herd behavior in prime-timetelevision programming. Journal of Industrial Econom-ics, 50: 43–56.

Klemperer, P. 1992. Equilibrium product lines: Competinghead-to-head may be less competitive. American Eco-nomic Review, 82: 740–755.

Knickerbocker, F. T. 1973. Oligopolistic reaction and multi-national enterprise. Boston: Harvard Business SchoolPress.

Knight, F. H. 1921. Risk, uncertainty, and profit. Boston:Houghton Miffin.

Kogut, B., & Chang, S. J. 1991. Technological capabilities andJapanese foreign direct investment in the United States.Review of Economics and Statistics, 73: 401–413.

Kreps, D. M., & Spence, A. M. 1985. Modelling the role ofhistory in industrial organization and competition. InG. R. Feiwel (Ed.), Issues in contemporary microeconom-ics and welfare: 340–378. London: Macmillan.

Leahy, D., & Pavelin, S. 2003. Follow-my-leader FDI and tacitcollusion. International Journal of Industrial Organiza-tion, 21: 439–453.

Leibenstein, H. 1950. Bandwagon, snob, and Veblen effects inthe theory of consumers’ demand. Quarterly Journal ofEconomics, 64: 183–207.

Levitt, B., & March, J. 1988. Organizational learning. AnnualReview of Sociology, 14: 319–340.

Lieberman, M. B. 2005. Did first-mover advantage survive thedot-com crash? Working paper, UCLA Anderson Schoolof Management, Los Angeles.

Lieberman, M. B., & Montgomery, D. B. 1988. First-moveradvantages. Strategic Management Journal, 9: 41–58.

Lippman, S. A., & Rumelt, R. P. 1982. Uncertain imitability: Ananalysis of interfirm differences in efficiency undercompetition. Bell Journal of Economics, 13: 418–438.

Lu, J. W. 2002. Intra- and inter-organizational imitative be-havior: Institutional influences on Japanese firms’ entrymode choice. Journal of International Business Studies,33: 19–37.

Makino, S., & Delios, A. 2000. Competitive reaction, imitationand bunching in foreign entry decisions. Academy ofManagement Proceedings: C1–C6.

Manski, C. F. 2000. Economic analysis of social interactions.Journal of Economic Perspectives, 14(3): 115–136.

March, J. G. 1981. Decision in organizations and theories ofchoice. In A. H. Van de Ven & W. F. Joyce (Eds.), Perspec-tives on organization design and behavior: 205–244. NewYork: Wiley.

March, J. G., & Simon, H. A. 1958. Organizations. New York:Wiley.

Martin, X., Swaminathan, A., & Mitchell, W. 1998. Organiza-

tional evolution in the interorganizational environment:Incentives and constraints on international expansionstrategy. Administrative Science Quarterly, 43: 566–601.

Milliken, F. J. 1987. Three types of perceived uncertaintyabout the environment. Academy of Management Re-view, 12: 133–143.

Miner, A. S., & Haunschild, P. R. 1995. Population level learn-ing. Research in Organizational Behavior, 17: 115–166.

Moriguchi, C., & Lane, D. 1999. A hundred-year war: Coke vs.Pepsi, 1890s–1990s. Case No. 9-799-177. Boston: HarvardBusiness School Case Services.

Motta, M. 1994. International trade and investments in avertically differentiated industry. International Journalof Industrial Organization, 12: 179–196.

Numagami, T., Asaba, S., Shintaku, J., & Amikura, H. 1992.Taiwa toshiteno Kyoso—Dentaku Sangyo niokeru KyosoKodo no Saikaishaku [Competition as dialogue—reinterpretation of competitive behavior in the electricalcalculator industry]. Soshiki Kagaku [OrganizationalScience], 26: 64–79.

Odagiri, H., 1992. Growth through competition, competitionthrough growth. Oxford: Oxford University Press.

Ounjian, M. L., & Carne, E. B. 1987. A study of the factorswhich affect technology transfers in a multilocationmultibusiness unit corporation. IEEE Transactions on En-gineering Management, EM-34: 194–201.

Palley, T. I. 1995. Safety in numbers: A model of managerialherd behavior. Journal of Economic Behavior and Orga-nization, 28: 443–450.

Peteraf, M. A. 1993. The cornerstones of competitive advan-tage: A resource-based view. Strategic ManagementJournal, 14: 179–192.

Porter, M. E. 1979. The structure within industries and com-panies’ performance. Review of Economics and Statis-tics, 61: 214–227.

Rao, H., Greve, H. R., & Davis, G. F. 2001. Fool’s gold: Socialproof in the initiation and abandonment of coverage byWall Street analysts. Administrative Science Quarterly,46: 502–526.

Rivkin, J. W. 2000. Imitation of complex strategies. Manage-ment Science, 46: 824–844.

Rogers, E. M. 1995. Diffusion of innovations (4th ed.). NewYork: Free Press.

Rosenbloom, R., & Cusumano, M. 1987. Technological pio-neering and competitive advantage: The birth of theVCR industry. California Management Review, 29(4): 51–76.

Rosenkopf, L., & Abrahamson, E. 1999. Modeling reputationand information influences in threshold models of band-wagon innovation diffusion. Computational and Mathe-matical Organizational Theory, 5: 361–384.

Rhoades, S. A., & Heggestad, A. A. 1985. Multimarket inter-dependence and performance in banking: Two tests.Antitrust Bulletin, 31: 975–995.

Scherer, F. M. 1980. Industrial market structure and economicperformance. Chicago: Rand McNally.


Scharfstein, D. S., & Stein, J. C. 1990. Herd behavior andinvestment. American Economic Review, 80: 465–479.

Schumpeter, J. 1942. Capitalism, socialism and democracy.New York: Harper.

Scott, J. T. 1982. Multimarket contact and economic perfor-mance. Review of Economics and Statistics, 64: 368–375.

Scott, J. T. 1991. Multimarket contact among diversified oli-gopolists. International Journal of Industrial Organiza-tion, 9: 225–238.

Shaver, J. M., Mitchell, W., & Yeung, B. 1997. The effect ofown-firm and other-firm experience on foreign directinvestment survival in the United States 1981–1992. Stra-tegic Management Journal, 18: 811–824.

Strang, D., & Tuma, N. B. 1993. Spatial and temporal hetero-geneity in diffusion. American Journal of Sociology, 99:614–639.

Stuart, T. 1998. Producer network positions and propensitiesto collaborate: An investigation of strategic alliance for-mations in a high-technology industry. AdministrativeScience Quarterly, 43: 668–698.

Szulanski, G. 1996. Exploring internal stickiness: Impedi-ments to the transfer of best practice within the firm.Strategic Management Journal, 17: 27–43.

Teece, D. J., Pisano, G., & Shuen, A. 1997. Dynamic capabili-ties and strategic management. Strategic ManagementJournal, 18: 509–533.

Westphal, J., Gulati, R., & Shortell, S. 1997. Customization orconformity? An institutional and network perspective onthe content and consequences of TQM adoption. Admin-istrative Science Quarterly, 42: 366–394.

White, L. 1971. The automobile industry since 1945. Cam-bridge, MA: Harvard University Press.

Yamawaki, H. 1998. Bunching in foreign direct investment:Evidence on Japanese firms in U.S. manufacturing in-dustries. CIBER Working paper No. 98-20, UCLA Ander-son School of Management, Los Angeles.

Yu, C.-M. J., & Ito, K. 1988. Oligopolistic reaction and foreigndirect investment: The case of the U.S. tire and textileindustries. Journal of International Business Studies, 19:449–460.

Marvin B. Lieberman ([email protected]) is professor of policy atthe UCLA Anderson School of Management. He received his Ph.D. in business eco-nomics from Harvard University. His current research interests involve testing theo-ries of industrial competition and strategic interaction, with particular emphasis onmarket entry, cost reduction, e-business, and strategic investment.

Shigeru Asaba ([email protected]) is professor of management at GakushuinUniversity, Tokyo. He received his Ph.D. in management from UCLA. His currentresearch focuses on the analysis of competition and collaboration among Japanesefirms and the impact of foreign entry into Japanese markets.


why do firms imitate each other? do firms imitate each other? marvin b. lieberman ucla anderson...

Documents