Business Strategy and Forecasting as Competitive Advantages

Please have a great command of English with great grammar.

Consider a business with which you are familiar, and which has at least one known, significant competitor. Write a paper that includes the following sections, organized using APA headings (not the Part letter).

Part A: Introduce the paper with the background and information about the business, and the thesis for your paper (1 paragraph).

Don't use plagiarized sources. Get Your Custom Essay on
Business Strategy and Forecasting as Competitive Advantages
Just from $13/Page
Order Essay

Part B: Explain, using Teece’s (2010) research article as a basis for your assignment:

  1. The business model.
  2. The business strategy by:

    Segmenting the market.
    Creating a value proposition for each segment.
    Describing the apparatus to deliver the value.
    Creating the preventative methods to avoid being imitated (p. 180).

  3. The business’s competitive advantage, using Teece’s definition and explanation as support (3–5 paragraphs).

Part C: Describe the main advantage your business’s main competitor has against the business.

  1. Analyze, using the Red Queen effect articles, [Derfus, Maggitti, Grimm, and Smith (2008) and Giachetti, Lampel, and La Pira (2017)], your business’s potential problem caused by its competition’s solution to the Red Queen effect.
  2. Include in your analysis one paragraph that synthetically presents each of the Red Queen articles’ main themes, research findings, and implications. Explain how the concepts of Red Queen effect have grown from 2008 to 2017 (at least 2 paragraphs).
  3. In one paragraph, explain your one recommended action as a solution to the problem caused by the competitor’s action. Describe whether or how it could increase competitiveness.

Part D: Conclusion. Summarize the main points of your paper and leave the reader with a thought to go forward with as an implication or recommendation from your ideas and analysis.

Part E:

References

.

  • Include a list of all references used.
  • Use at least 4 references: the three required readings, and one supporting your information about your selected business.
  • Use APA 6th edition on your paper and submit a paper free from errors and of high academic quality.

Refer to the scoring guide to ensure you have covered all the requirements.

References

Derfus, P. J., Maggitti, P. G., Grimm, C. M., & Smith, K. G. (2008). The red queen effect: Competitive actions and firm performance. Academy of Management Journal, 51(1), 61–80.

Giachetti, C., Lampel, J., & Li Pira, S. (2017). Red queen competitive imitation in the U.K. mobile phone industry. Academy of Management Journal, 60(5), 1882–1914.

Teece, D. J. (2010). Business models, business strategy and innovation. Long Range Planning, 43(2–3), 172–194.

Long Range Planning 43 (2010) 172e194 http://www.elsevier.com/locate/lrp

Business Models, Business
Strategy and Innovation

David J. Teece

Whenever a business enterprise is established, it either explicitly or implicitly employs
a particular business model that describes the design or architecture of the value creation,
delivery, and capture mechanisms it employs. The essence of a business model is in de-
fining the manner by which the enterprise delivers value to customers, entices customers
to pay for value, and converts those payments to profit. It thus reflects management’s
hypothesis about what customers want, how they want it, and how the enterprise can
organize to best meet those needs, get paid for doing so, and make a profit. The purpose
of this article is to understand the significance of business models and explore their
connections with business strategy, innovation management, and economic theory.
� 2009 Published by Elsevier Ltd.

Introduction
Developments in the global economy have changed the traditional balance between customer and
supplier. New communications and computing technology, and the establishment of reasonably
open global trading regimes, mean that customers have more choices, variegated customer needs
can find expression, and supply alternatives are more transparent. Businesses therefore need to
be more customer-centric, especially since technology has evolved to allow the lower cost provision
of information and customer solutions. These developments in turn require businesses to re-eval-
uate the value propositions they present to customers e in many sectors, the supply side driven
logic of the industrial era has become no longer viable.

This new environment has also amplified the need to consider not only how to address customer
needs more astutely, but also how to capture value from providing new products and services.
Without a well-developed business model, innovators will fail to either deliver e or to capture e
value from their innovations. This is particularly true of Internet companies, where the creation of
revenue streams is often most perplexing because of customer expectations that basic services
should be free.

0024-6301/$ – see front matter � 2009 Published by Elsevier Ltd.
doi:10.1016/j.lrp.2009.07.003

http://www.elsevier.com/locate/lrp

A business model articulates the logic and provides data and other evidence that demonstrates
how a business creates and delivers value to customers. It also outlines the architecture of revenues,
costs, and profits associated with the business enterprise delivering that value. The different ele-
ments that need to be determined in business model design are listed in Figure 1.

The issues related to good business model design are all interrelated, and lie at the core of the
fundamental question asked by business strategists e how does one build a sustainable competitive
advantage and turn a super normal profit? In short, a business model defines how the enterprise
creates and delivers value to customers, and then converts payments received to profits.1 To profit
from innovation, business pioneers need to excel not only at product innovation but also at busi-
ness model design, understanding business design options as well as customer needs and techno-
logical trajectories. Developing a successful business model is insufficient to assure competitive
advantage as imitation is often easy: a differentiated (and hard to imitate) e yet effective and effi-
cient e business model is more likely to yield profits. Business model innovation can itself be a path-
way to competitive advantage if the model is sufficiently differentiated and hard to replicate for
incumbents and new entrants alike.

In essence, a business model [is] a conceptual, rather than financial,

model of a business.

In essence, a business model embodies nothing less than the organizational and financial ‘archi-
tecture’ of a business.2 It is not a spread sheet or computer model, although a business model might
well become embedded in a business plan and in income statements and cash flow projections. But,
clearly, the notion refers in the first instance to a conceptual, rather than a financial, model of a busi-
ness. It makes implicit assumptions about customers, the behavior of revenues and costs, the

Figure 1. Elements of business model design

Long Range Planning, vol 43 2010 173

changing nature of user needs, and likely competitor responses. It outlines the business logic re-
quired to earn a profit (if one is available to be earned) and, once adopted, defines the way the en-
terprise ‘goes to market’. But it is not quite the same as a strategy: the distinction and the
relationship between the two will be discussed later.

Despite lineage going back to when societies began engaging in barter exchange, business models
have only been explicitly catapulted into public consciousness during the last decade or so. Driving
factors include the emerging knowledge economy, the growth of the Internet and e-commerce, the
outsourcing and offshoring of many business activities, and the restructuring of the financial ser-
vices industry around the world. In particular, the way in which companies make money nowadays
is different from the industrial era, where scale was so important and the capturing value thesis was
relatively simple i.e. the enterprise simply packed its technology and intellectual property into
a product which it sold, either as a discreet item or as a bundled package. The existence of electronic
computers that allow low cost financial statement modeling has facilitated the exploration of alter-
native assumptions about revenues and costs.

Additional impetus has come from the growth of the Internet, which has raised anew, and in
a transparent way, fundamental questions about how businesses deliver value to the customer,
and how they can capture value from delivering new information services that users often expect
to receive without charge. It has allowed individuals and businesses easy access to vast amounts
of data and information, and customer power has increased as comparison shopping has been
made easier. In some industries, such as the recording industry, Internet enabled digital downloads
compete with established channels (such as physical product sales) and, partly because of the ubiq-
uity of illegal digital downloading, the music recording industry is being challenged to completely
re-think its business models. The Internet is not just a source of easy access to digital data; it is also
a new channel of distribution and for piracy which clearly makes capturing value from Internet
transactions and flows difficult for recording companies, performers and songwriters alike. More
generally, the Internet is causing many ‘bricks and mortar’ companies to rethink their distribution
strategies e if not their whole business models.

Notwithstanding how the Internet has devastated the business models of industries like music re-
cording and news, internet companies themselves have struggled to create viable business models. In-
deed, during the dot.com boom and bust of 1998e2001, many new companies with zero or negative
profits (and unprecedentedly low revenues) sought financial capital from the public markets, which e
at least for a short while e accommodated them. Promoters managed to persuade investors that tra-
ditional revenue and profitability models no longer applied e and that the dot.com companies would
(eventually) figure out (highly) profitable business models. Few have, causing one commentator to
remark that ‘the demise of a popular but unsustainable business model now seems inevitable’.3

No matter what the sector, there are criteria that enable one to determine whether or not one has
designed a good business model. A good business model yields value propositions that are compel-
ling to customers, achieves advantageous cost and risk structures, and enables significant value cap-
ture by the business that generates and delivers products and services. ‘Designing’ a business
correctly, and figuring out, then implementing e and then refining e commercially viable archi-
tectures for revenues and for costs are critical to enterprise success. It is essential when the enter-
prise is first created; but keeping the model viable is also likely to be a continuing task. Superior
technology and products, excellent people, and good governance and leadership are unlikely to pro-
duce sustainable profitability if business model configuration is not properly adapted to the com-
petitive environment. Some preliminary criteria for business model design are suggested
throughout this article, and summarised in a later section.

The concept of a business model has no established theoretical

grounding in economics or in business studies.

174

  • Business Models, Business Strategy and Innovation
  • Business models e the theoretical foundation
    The concept of a business model lacks theoretical grounding in economics or in business studies.
    Quite simply there is no established place in economic theory for business models; and there is not
    a single scientific paper in the mainstream economics journals that analyses or discusses business
    models in the sense they are defined here. (Possible exceptions are the literature on investment
    in basic research, which economists recognize as being unsupported by private business models
    (see below), and the literature on bundling, inasmuch as it deals e indirectly e with different rev-
    enue models.) The absence of consideration of business models in economic theory probably stems
    from the ubiquity of theoretical constructs that have markets solving the problems that e in the real
    world e business models are created to solve.

    Economic theory implicitly assumes that trades take place around tangible products: intangibles
    are, at best, an afterthought. In standard approaches to competitive markets, the problem of cap-
    turing value is quite simply assumed away: inventions are often assumed to create value naturally
    and, enjoying protection of iron-clad patents, firms can capture value by simply selling output in
    established markets, which are assumed to exist for all products and inventions. Thus there are no
    puzzles about how to design a business e it is simply assumed that if value is delivered, customers
    will always pay for it. Putting so called ‘public goods’ and ‘free rider’ issues to one side, business
    models are quite simply redundant because producers/suppliers can create and capture value simply
    through disposing their output at competitive market prices. Such models clearly assume away the
    essential business design issues that are the subject of this article.

    In short, figuring out business models for a new or existing product or business is an unnecessary
    step in textbook economics, where it is not uncommon to work with theoretical constructs which
    assume fully developed spot and forward markets, strong property rights, the costless transfer of
    information, perfect arbitrage, and no innovation.4 In mainstream approaches, there is simply
    no need to worry about the value proposition to the customer, or the architecture of revenues
    and costs, or about mechanisms to capture value.5 Customers will buy if the price is less that
    the utility yielded; producers will supply if price is at or above all costs including a return to
    capital e the price system resolves everything and business design issues simply don’t arise.

    But general equilibrium models, with (one-sided) markets and perfect competition are a carica-
    ture of the real world. Intangible products are in fact ubiquitous, two-sided markets are common,
    and customers don’t just want products; they want solutions to

    their perceived needs.

    In some
    cases, markets may not even exist, so entrepreneurs may have to build organizations in order to
    perform activities for which markets are not yet ready. Accordingly, in the real world, entrepreneurs
    and managers must give close consideration to the design of business models and even to building
    businesses to execute transactions which cannot yet be performed in the market.

    Equilibrium and perfect competition are a caricature of the real

    world. customers don’t just want products; they want solutions to

    their perceived needs.

    It’s also true that business models have no place within the theoretical constructs of planned
    economies (just as in a perfectly competitive economy). While central planners do need to under-
    stand the stages in the production system, in a supply driven system e where consumers merely get
    what the system produces e business models simply aren’t necessary. There is no problem associ-
    ated with producers capturing value because value doesn’t even have to be captured; the state de-
    cides what and how to produce, and how to pay for it all.

    While business models have no place in economic theory, they likewise lack an acceptable place
    in organizational and strategic studies, and in marketing science. However, there has been some
    limited discussion and research on new organizational forms. Williamson, for instance, recognizes

    Long Range Planning, vol 43 2010 175

    that ‘the 1840s marked the beginning of a great wave of organizational change that has brought us the
    modern corporation’.6 As discussed earlier, new organizational forms can be a component of a busi-
    ness model;7 but organizational forms are not business models. Clearly, the study of business
    models is an interdisciplinary topic which has been neglected e despite their obvious importance,
    it lacks an intellectual home in the social sciences or business studies. This article aims to help rem-
    edy this deficiency.

    Examples of business models
    Business models are necessary features of market economies where there is consumer choice, trans-
    action costs, and heterogeneity amongst consumers and producers, and competition. Profit seeking
    firms in competitive environments will endeavor to meet variegated consumer wants through the
    constant invention and presentation to the consumer of new value propositions. Business models
    are often necessitated by technological innovation which creates both the need to bring discoveries
    to market and the opportunity to satisfy unrequited customer needs. At the same time, as indicated
    earlier, new business models can themselves represent a form of innovation. There are a plethora of
    business model possibilities: some will be much better adapted to customer needs and business en-
    vironments than others. Selecting, adjusting and/or improving business models is a complex art.
    Good designs are likely to be highly situational, and the design process is likely to involve iterative
    processes. New business models can both facilitate and represent innovation e as history
    demonstrates.

    Traditional industries
    A striking early American example of 19th century business model innovation was Swift and Com-
    pany’s ‘reengineering’ of the meat packing industry. Prior to the 1870s, cattle were shipped live by
    rail from the Midwestern stockyard centers like Omaha, Kansas City and Chicago to East Coast
    markets where the animals were slaughtered and the meat sold by local butchers. Gustavus Swift
    sensed that if the cattle could be slaughtered in the Midwest and shipped already dressed to distant
    markets in refrigerated freight cars, great economies in ‘production’/centralization and transporta-
    tion could be achieved, along with an improvement in the quality of the final product.

    Swift’s new business model quickly displaced business models involving a network of shippers,
    East Coast butchers and the railroads. His biggest challenge was the absence of refrigerated ware-
    houses to store the beef near point of sale, which were not part of the existing distribution system.
    Swift set about creating a nationwide web of refrigerated facilities, often in partnerships with local
    jobbers. ‘Once Swift overcame the initial consumer resistance to meat slaughtered days before in distant
    places, his products found a booming market because they were as good as freshly butchered meats and
    were substantially cheaper e Swift’s success quickly attracted imitators e By the 1890s, men like Phillip
    Armour had followed on Swift’s heels’.8

    A more recent example is containerization. Malcolm McLean, owner of a large U.S. trucking
    company, was convinced that conventional shipping was highly inefficient because shipping com-
    panies typically broke bulk at dockside, and cargo ships spent most of their time in port being
    loaded or unloaded. In 1955 he hired an engineer to design a road trailer body that could be
    detached from its chassis and stacked on ships. McLean acquired a small steamship company,
    renamed it Sea-Land Industries (it eventually became absorbed into the Maersk Line). He devel-
    oped steel frames to hold the containers, first on the top decks of tankers, and then on the world’s
    first specialized cellular containership, the Gateway City, launched in 1957. To promote the stan-
    dardization necessary to develop the industry, McLean made Sea-Land’s patents available royalty
    free to the International Standards Organization (ISO). Sea-Land began service on North Atlantic
    routes in 1966. When R. J. Reynolds bought Sea-Land for $530 million in 1969, McLean received
    $160M for his share and retired.9

    Another U. S. example of successful business model innovation is Southwest Airlines, where the
    founder surmised that most customers wanted direct flights, low costs, reliability and good

    176 Business Models, Business Strategy and Innovation

    customer service, but didn’t need ‘frills’. To achieve these goals, Southwest eschews the hub-and-
    spoke model associated with alliances, nor does it allow interlining of passengers and baggage,
    or sell tickets through travel agencies e all sales are direct. Aircraft are standardized on the Boeing
    737, allowing greater efficiency and operating flexibility. Southwest’s business model e which was
    quite distinct from those of the major carriers e followed elements of a discount airline model first
    pioneered in the U.K. by Freddie Laker. Although Laker Airways eventually failed e as did other
    early followers in the U.S. such as People’s Express e Easy Jet has implemented a similar model
    in Europe, so far successfully.

    The ‘razor-razor blade model’ is another classic (and quite generic) case of a well known business
    revenue model (which is just one component of a business model), which involves pricing razors
    inexpensively, but aggressively marking-up the consumables (razor blades). Jet engines for com-
    mercial aircraft are priced the same way e manufacturers know that engines are long lived, and
    maintenance and parts is where Rolls Royce, GE, Pratt & Whitney and others make their money.
    So engines are sold relatively inexpensively e but parts (and service) involve considerable mark-ups
    and represent an income stream that may continue for decades.

    The ’razor-razor blade model’ is a classic business revenue model … jet

    engines for commercial aircraft are priced the same way.

    In the sports apparel business, sponsorship is a key component of today’s business models. Nike,
    Adidas, Reebok, Canterbury, and others sponsor football and rugby clubs and teams, providing kit
    and sponsorship dollars as well as royalties streams from the sale of replica products. After building
    brand on the field, these companies endeavor to leverage their brand into off-field products, often
    with considerable success. On-field sponsorship is almost a sine qua non for brand authenticity.
    However, this model is readily imitated, and its viability for any particular apparel company
    depends on the sponsor’s particular abilities to leverage on-field sponsorships into off-field sales.
    Relationships with clubs, teams, and with team managers and club owners become important in
    the mix.

    Performing artists have several business models they can employ. Their revenue sources might
    include live productions, movies, sale of physical CDs through stores or online music sales through
    virtual stores such Apple’s iTunes.10 Stars might decide to use concerts as their main revenue gen-
    erator, or to spend less time performing and more in the recording studio, using concerts primarily
    to stimulate sales of recordings. In earlier days when piracy was limited, the Beatles demonstrated
    that stars could quit live performances and continue to do well on royalties from the sale of re-
    corded music. Then, in the 80s and 90s, the music video became an important source of revenue,
    and more recently, ‘soundtracks’ to video games have become a significant source of revenue for
    some artists. In short, multiple revenue streams are available, and the particular revenue model em-
    ployed can depend on the marketplace, on a star’s contextual talents and preferences, and on the
    quality of copyright protection afforded to recorded music.

    Business models must morph over time as changing markets, technologies and legal structures
    dictate and/or allow. For instance, the business model that U.S. investment banks had employed
    for almost 20 years largely disappeared in 2008. From at least the 1950s through the 1990s, the in-
    vestment banking function usually generated most of the banks’ revenues. However, for Goldman
    Sachs (arguably the industry leader) that figure had fallen to 16% by 2007, while revenues from
    trading and principal investment had grown to 68%, leading it and other investment banks to
    morph their business models into something quite different e and more risky e than traditional
    investment banking. Subprime mortgages and other problematic assets became securitized and in-
    jected into the system, encouraged by Freddie and Fannie (and by Congress) with results that sub-
    sequently hit the headlines. In September 2008, Goldman Sachs and Morgan Stanley (the last two

    Long Range Planning, vol 43 2010 177

    independent investment banks left standing in the U.S. after the takeover of Bear Sterns by JP Mor-
    gan Chase, the bankruptcy of Lehman Brothers, and Merrill Lynch’s absorption by Bank of Amer-
    ica) converted themselves into federally chartered commercial banks. By accepting government
    regulation by the FDIC, Goldman Sachs and Morgan Stanley will need to maintain lower leverage,
    and accept lower risk and lower returns. In their need for a source of stable funds, both have (albeit
    reluctantly) made significant business model changes e in short, they have been obliged to abandon
    their old models entirely.

    The information/internet industries
    As noted earlier, the information industries have always raised challenging business model issues
    because information is often difficult to price, and consumers have many ways to obtain certain
    types without paying. Figuring out how to earn revenues (i.e. capture value) from the provision
    of information to users/customers is a key (but not the only) element of business model design
    in the information sector. The rules for strategic engagement promulgated by Shapiro and Varian
    are core elements of strategy in the information services sector.11

    As traditional information providers, newspapers have employed a revenue model for decades in
    which the paper is sold quite inexpensively (usually at a nominal level, insufficient to cover costs),
    while publishers looked to advertising revenue to cover remaining costs plus provide a profit. In
    recent years, this business model has been undermined by websites like eBay and Craigslist that
    have siphoned off advertising revenues from job and real estate listings and classified ads: many
    newspapers have gone out of business.

    The Internet has enabled traditional industries like DVD rentals to adopt a more modern on-line
    posture. Netflix (http://www.netflix.com) enables customers to order DVDs on-line and have expe-
    dited delivery by the U.S. Mail as a more convenient alternative to going to a rental facility, renting
    the DVD, and returning it several days later. Monthly fees are what sustain Netflix.

    The emergence of the Internet, Napster and its clones has obliged music recording companies to
    rethink their business models, which they have been doing along several fronts. On one front, they
    are moving to greatly increase the royalty rate for Internet ‘broadcast’ of their content, while on
    another, they are moving to capture advertising revenues associated with that content. For instance,
    MySpace Music (http://music.myspace.com) enables users to listen to songs from Universal, Sony
    BMG and Warner Music, and provides free advertising-supported streaming, with easy access to
    Amazon.com for music purchases. Another example is the Nokia ‘Comes with Music’ (CWM)
    handset that comes with ‘free’, unlimited music downloads for a year, with Nokia passing on
    a fee to the recording companies.

    A recent example of an Internet business model is Flickr (www.flickr.com), which has been described
    as ‘a poster child for Web 2.0 [offering] users a way to share photos easily’.12 Flickr’s friendly and easy-to-use
    web interface and its free photo management and storage service are noted as great examples of a Web 2.0
    ‘freemium’ (free and premium) business model, characterized by Fred Wilson as:

    ‘Give your service away for free, possibly ad supported but maybe not, acquire a lot of customers
    very efficiently through word of mouth, referral networks, organic search marketing, etc., then
    offer premium priced value added services or an enhanced version of your service to your
    customer base’.

    The Flickr business model (which actually evolved from gaming to on-line photo sharing, harness-
    ing user feedback generated through blogs) essentially gives away the services that amateur photogra-
    phers want most: photo sharing, on-line storage, indexing and tagging. Shuen notes that low cost
    on-line distribution and marketing and investment are associated with ‘revenue from multiple streams,
    including value-added premium services and customer acquisition.’ Flickr’s multiple revenue stream
    business model involves collecting subscription fees, charging advertisers for contextual advertising,
    and receiving sponsorship and revenue-sharing fees from partnerships with retail chains and comple-
    mentary photo service companies. Yahoo bought Flickr in March 2005 for tens of millions of dollars.

    178 Business Models, Business Strategy and Innovation

    http://www.netflix.com

    http://music.myspace.com

    http://Amazon.com

    http://www.flickr.com

    companies can adopt business models [e.g. Freemium or multiple

    revenue stream models] pioneered in one space into another.

    A business model pioneered by one company in one space may be adopted by another company
    in another space. The ‘freemium’ model has been adopted by Adobe (for its PDF reader), Skype and
    MySpace, while Outshouts Inc (www.outshouts.com) has applied Flickr’s multiple revenue streams
    model to on-line Web videos, allowing users to personalize and disseminate videos for business or
    consumer purposes. While it is common with Internet start-ups, the multiple revenue stream ap-
    proach is by no means new. Besides theatrical releases and looking to exploit an obvious extra rev-
    enue stream e the sequel e movie studios have long sought revenues from ‘ancillary’ licensing
    (toys, T-shirts, lunchboxes, backpacks), and more recently from video games and soundtracks.

    Freemium business models are also deployed by a large number of software companies (such as
    Linux, Firefox, and Apache) who operate in the open source marketplace. The standard form (or
    ‘kernel’) of the software is licensed under an open source license and then a premium version with
    additional features and/or associated services is made available under commercial license terms.
    One theory is that ‘vendors’ get customers (often, and ideally with the IT organization bypassing
    Procurement Departments altogether e because, after all, the software is ‘free’) hooked on the
    free product, and then subsequently convert them into paying customers through the sale of com-
    plementary software and/or service. However, conversion rates to paying customers have been
    poor, and it’s not clear the model works.

    The discussion so far has focused mainly on the impact of technology on value and its delivery.
    However technology can have an equally transformative effect on the cost side of the business model.
    New ‘cloud-based’ computing models, for example, remove the need for small companies to invest up-
    front in expensive servers e instead they can buy server capacity in small slices, as needed, according to
    their monthly needs. The size of such slices continues to shrink e services such as Amazon’s EC2, for
    example, even allow customers to buy virtual server capacity for a single transaction, measured in mil-
    liseconds. This kind of innovation transforms previous ‘fixed plus variable’ cost models into entirely
    variable cost models, greatly improving efficiency and reducing early-stage capital requirements.

    Business models, strategy and sustainable competitive advantage
    A business model articulates the logic, the data, and other evidence that support a value proposition
    for the customer, and a viable structure of revenues and costs for the enterprise delivering that
    value. In short, it’s about the benefit the enterprise will deliver to customers, how it will organize
    to do so, and how it will capture a portion of the value that it delivers. A good business model will
    provide considerable value to the customer and collect (for the developer or implementor of the
    business model) a viable portion of this in revenues. But developing a successful business model
    (no matter how novel) is insufficient in and of itself to assure competitive advantage. Once imple-
    mented, the gross elements of business models are often quite transparent and (in principal) easy to
    imitate e indeed, it is usually just a matter of a few years e if not months e before an evidently
    successful new business model elicits imitative efforts. In practice, successful business models very
    often become, to some degree, ‘shared’ by multiple competitors.

    A business model is more generic than a business strategy. Coupling

    strategy and business model analysis is needed to protect competitive

    advantage resulting from new business model design.

    Long Range Planning, vol 43 2010 179

    http://www.outshouts.com

    As described, a business model is more generic than a business strategy. Coupling strategy anal-
    ysis with business model analysis is necessary in order to protect whatever competitive advantage
    results from the design and implementation of new business models. Selecting a business strategy
    is a more granular exercise than designing a business model. Coupling competitive strategy analysis
    to business model design requires segmenting the market, creating a value proposition for each seg-
    ment, setting up the apparatus to deliver that value, and then figuring out various ‘isolating mech-
    anisms’ that can be used to prevent the business model/strategy from being undermined through
    imitation by competitors or disintermediation by customers.13

    Strategy analysis is thus an essential step in designing a competitively sustainable business model.
    Unless the business model survives the filters which strategy analysis imposes, it is unlikely to be
    viable, as many business model features are easily imitated. For instance, leasing vs. owning is an
    observable characteristic of business models that competitors can replicate. The ‘newspaper revenue
    model’ e i.e. low cost for the newspaper, use of advertising (including classifieds) to help cover the
    costs of generating content e is easy to replicate, and has been implemented with little variation in
    thousands of geographically separate ‘markets’ throughout the world.

    Having a differentiated (and hard-to-imitate) e but at the same time effective and efficient e
    architecture for an enterprise’s business model is important to the establishment of competitive ad-
    vantage. The various elements need to be cospecialized to each other, and work together well as
    a system. Both Dell Inc. and Wal-Mart have demonstrated the value associated with their business
    models (while Webvan and many other dotcoms demonstrated just the opposite). Dell and Wal-
    Mart’s business models were different, superior, and required supporting processes that were
    hard for competitors to replicate (at least in the U.S. e elsewhere, new entrants could adopt key
    elements of the model and pre-empt Wal-Mart, as Steven Tindall has demonstrated so ably in
    New Zealand with ‘The Warehouse’). Both Dell and Wal-Mart have also constantly adjusted and
    improved their processes over time. Michael Dell, founder of Dell, notes:

    This belief e that by working directly with customers we could get them technology faster, provide
    a better level of service, and provide better value e was the basis of the business e the fundamental
    business system was quite powerful and delivered lots of value to our customers e we screwed up lots
    of things, but the one thing we got right was this core business model, and it masked any other
    mistakes .14

    Dell’s competitors were incumbents who had difficulty in replicating its strategy, as selling direct
    to customers would upset their existing channel partners and resellers: as a new entrant, Dell had no
    such constraints. Another critical element of Dell’s success, beyond the way it organized its value
    chain, was the choice of products it sold through its distribution system. Over time, Dell developed
    (dynamic) capabilities around deciding which products to build beside desktop and laptop com-
    puters, and has since added printers, digital projectors and computer-related electronics. Of course,
    the whole strategy depended on the availability of numerous non captive suppliers able to produce
    at very competitive prices.

    Magretta points out that the business model of discount (big box) retailing had been around long
    before Wal-Mart founder Sam Walton (in his words) ‘put good sized stores into little one-horse towns
    which everybody else was ignoring’.15 Once in place, the towns Wal-Mart had selected were too small
    to support another similar sized store, so a difficult to replicate first mover advantage had been cre-
    ated. Wal-Mart promoted national brands at deep discounts, supported by innovative and lean pur-
    chasing logistics and IT systems: these were elements of its strategy that made its business model
    difficult to imitate.

    Search engine development and the Google story is another interesting business model illus-
    tration. Early efforts in this field, including Lycos, Excite, Alta Vista, Inktomi and Yahoo, would
    find lots of information e perhaps too much e and present it to users in an unhelpful manner,
    with maybe thousands of results presented in no discernible or useful order. Alta Vista presented
    links, but without using them as aids to searching. Larry Page, one of the founders of Google,

    180 Business Models, Business Strategy and Innovation

    surmised that counting links to a website was a way of ranking its popularity (much like higher
    citation counts in scientific journals point to more important contributions to the literature),
    and decided to use the number of links to important sites as a measure of priority. Using this link
    based approach, Page and his colleagues at Google devised an Internet site ranking system e the
    PageRank algorithm e which went on to be their core product/service offering, and one which
    has proved very valuable to users. The challenge was to tune the product offering and devise
    a business model to capture value, which was not easy in a world in which consumers expected
    search to be free.

    The business model developed around Google’s product/service innovation required heavy in-
    vestment in computing power as well as in software. Google writes its own software and (remark-
    ably) builds its own computers. It takes advantage of its considerable computing power to count
    words and links, and to combine information about words and links. This allows the Google search
    engine to take more factors into account than others currently in the market. The Google revenue
    model eschewed funding from advertisers: directed search biased to favor advertisers was perceived
    by Google’s founders as degrading to the integrity of the search process and to its emerging brand.
    Accordingly, it decided that the essence of its revenue model would be sponsored links i.e. no pop
    ups or other graphics interfering with the search. In short, Page and Brin found a way to accom-
    modate advertising (thereby enabling revenue generation) without subtracting from the search ex-
    perience, and arguably enhancing it.16 However, they also adopted an integrated approach (by
    fulfilling their own software and hardware requirements) to keep control of their product/service
    offering, ensuring its delivery and its quality.

    Business model choices define the architecture of the business .

    expansion paths develop from there on out.

    Business model choices define the architecture of the business, and expansion paths develop from
    there on out. But once established, enterprises often encounter immense difficulty in changing busi-
    ness models e witness the difficulties American Express and Discover Card have experienced in try-
    ing to morph to hybrid models where they issue cards themselves while simultaneously looking to
    persuade banks as partners to act as card issuers for them. This is clearly incongruous e their main
    competitors (Visa and MasterCard, who provide network services only and don’t compete with
    banks in issuing credit cards) are not hobbled by such relationship conflicts, and are clearly likely
    to be the bank’s preferred partners. Thus American Express and Discover are unlikely to have (and
    indeed have not had) much success trying to replicate the Visa/MasterCard business model while
    still maintaining their own internal issuing and acquiring functions.17

    In short, innovating with business models will not, by itself, build enterprise-level competitive advan-
    tage. However, new business models, or refinements to existing ones, like new products themselves, often
    result in lower cost or increased value to the consumer; if not easily replicated by competitors, they can
    provide an opportunity to generate higher returns to the pioneer, at least until their novel features are
    copied. These issues are summarized in Figure 2 and explored in more detail later.

    Barriers to imitating business models
    This section attempts to distil those factors that affect the ease or otherwise of imitating business
    models. At a superficial level all business models might seem easy to imitate e certainly the basic
    idea and the business logic behind a new model is unlikely itself to enjoy intellectual property protec-
    tion. In particular, a new business model, being more general than a business method, is very unlikely
    to qualify for a patent, even if certain business methods underpinning it may be patentable. Descrip-
    tions of a business model may enjoy copyright protection, but that is unlikely to be a barrier to

    Long Range Planning, vol 43 2010 181

    Figure 2. Steps to achieve sustainable business models

    copying its basic core ‘idea’. What then is it, if anything, that is likely to impede the copycat behavior
    that can so quickly erode the business model pioneer’s advantage? Three factors would seem to be
    relevant.

    First, implementing a business model may require systems, processes and assets that are hard to
    replicate e such was the situation with potential entrants into the towns too small to sustain
    a Wall-mart competitor. Similarly, while at some level Dell Computer’s direct-to-user (consumers
    and businesses) business model is obvious (you simply disintermediate wholesalers and retailers),
    when Gateway Computers tried to implemented a similar model, their failure to achieve anywhere
    near Dell’s performance levels has been attributed to the inferior implementation of processes. Ca-
    pabilities matter. Likewise, when Netflix pioneered delivery of DVDs by mail using a subscription
    system, Blockbuster video responded with a similar offering. But Netflix held on to its lead, both
    because it was not handicapped by Blockbuster’s cannibalization concerns, and because it had pat-
    ents on the ‘ordered list’ (which it later accused Blockbuster of infringing) by which subscribers
    indicated online their movie preferences.

    Second, there may be a level of opacity (Rumelt has referred to this opacity as ‘uncertain imita-
    bility’) that makes it difficult for outsiders to understand in sufficient detail how a business model is
    implemented, or which of its elements in fact constitute the source of customer acceptability.18

    Third, even if it is transparently obvious how to replicate a pioneer’s business model, incumbents
    in the industry may be reluctant to do so if it involves cannibalizing existing sales and profits or
    upsetting other important business relationships. When incumbents are constrained in this way,
    the pioneer of a new business model may enjoy a considerable period of limited competitive re-
    sponse. Notwithstanding these constraints, competition is likely to be vigorous because other
    new entrants, similarly unconstrained by incumbency and cannibalization anxieties, will be equally
    free to enter.

    Business model learning
    The moves made by an incumbent competitor to overcome such barriers to respond to Netflix’s
    entry into DVD rentals provide an interesting illustration of Business Model learning and adjust-
    ment. To respond to Netflix’s competitive inroads into its DVD store-rental model, Blockbuster
    purchased assets from NetLearn in April 2002, including those of DVDRentalCentral.com,

    182 Business Models, Business Strategy and Innovation

    http://DVDRentalCentral.com

    a subscriber-based online DVD rental service, which it renamed FilmCaddy and operated separately
    from the rest of the Blockbuster business. In August 2004, Blockbuster shut down FilmCaddy and
    launched Blockbuster Online, its new online rental service that allowed customers to rent unlimited
    DVDs (three at a time) for a monthly fee. Its initial plan included no due dates or extending view-
    ing fees, and also gave subscribers two free in-store movie rentals each month. In November 2006, it
    launched Blockbuster Total Access, coupling its online business with its in-store capabilities to al-
    low online customers the option of returning their DVDs through the mail or exchanging them for
    free-in-store movie rentals at over 5000 Blockbuster stores.

    Clearly, most elements of the Netflix business model were relatively easy to copy, and, although
    Blockbuster was undoubtedly constrained by the cannibalization of its in-store rentals by its online
    business, these moves reflected its attempts to respond (defensively) to Netflix. Netflix had figured
    out an approach and made the investments required to establish the online market. But Blockbuster
    responded by leveraging its brand equity and its network of physical stores to try to capture value
    from a modified version of the model Netflix had created: at minimum, it was intent on minimizing
    damage to its in-store franchise. Its guiding principle in responding appeared to be to offer cus-
    tomers all the functionality of Netflix plus several distinguishing features e associated with using
    its retail store footprint e which Netflix couldn’t easily match. Blockbuster’s stores also comple-
    mented its online strategy, by offering customers a choice of how to return their rented DVDs.
    While Netflix had no retail presence with which to respond to this element of Blockbuster’s offering
    directly, it did have some limited patent protection, with two patents that provided its business
    model some protection e in particular, to its ‘ordered list’ for movie selection. While these patents
    did not cover online DVD rental per se, they did cover methods allowing users to pay a flat fee to
    have a maximum number of movies out at any one time, and to return a fixed number of movies
    within a fixed time period.

    In short, Blockbuster implemented a close facsimile of the Netflix business model (even its web-
    site was very similar, featuring stars, recommendations, box shots and the ‘dynamic queue’) and
    achieved reasonable success, undoubtedly blunting Netflix’s growth. While Blockbuster Online
    was a good defensive move, Netflix’s pioneering status and its capacity to improve its business
    model, and enforce its patents, has helped undergird its competitive advantage.

    technological innovation does not guarantee business success e new

    product development efforts should be coupled with a business

    model defining their ’go to market’ and ’capturing value’ strategies.

    Business models to capture value from technological innovation

    The profiting from innovation framework
    Figuring out how to capture value from innovation is a key element of business model design. This
    is a topic on which this author has written extensively, although the treatment hitherto was not
    couched in the language of business model design. This section is more forthright in that regard.

    Every new product development effort should be coupled with the development of a business
    model which defines its ‘go to market’ and ‘capturing value’ strategies. Clearly technological inno-
    vation by itself does not automatically guarantee business or economic success e far from it. This
    was a theme in the author’s earlier work on ‘Profiting from Innovation’,19 which outlined a contin-
    gent approach with respect to how to organize the production system/value chain, taking into ac-
    count the ‘appropriability regime’ and the innovator’s prior asset positioning. Notwithstanding that
    scholars have recognized that technological innovation without a commercialization strategy is as

    Long Range Planning, vol 43 2010 183

    likely to lead to the (self-) destruction of creative enterprises as it is to profitable (Schumpeterian)
    creative destruction, technological innovation is often assumed by some to lead inexorably to com-
    mercial success. It rarely does. When executives think of innovation, they all too often neglect the
    proper analysis and development of business models which can translate technical success into
    commercial success. Good business model design and implementation, coupled with careful stra-
    tegic analysis, are necessary for technological innovation to succeed commercially: otherwise,
    even creative companies will flounder. Quintessential examples of firms that succeeded at techno-
    logical innovation but failed to get the business model and the technology strategy right included
    EMI (the CAT scanner) and Xerox (the personal computer).20

    But there are a plethora of other examples too. Eli Whitney’s 1793 invention of the cotton gin greatly
    increasing the ease with which cotton could be separated from the pod e but still he died a poor man.
    Even Thomas Edison e with his portfolio of 1000+ patents and personal fame from inventing a durable
    electric light bulb, electricity as a system, motion pictures and phonographs e failed commercially on
    many fronts. For example, he abandoned the recording business after arguably failing to get its
    business model right by insisting that Edison disks be designed to work only on Edison phonographs
    (although his early phonograph also suffered from poor sound reproduction, recordings that were too
    brief, and cylinders that could only survive a few playings). In short, getting the business model and the
    technology strategy right is necessary to achieve commercial viability if sustainable competitive advan-
    tage is to be built and innovators are to profit from their innovations.

    Figuring out how to deliver value to the customer e and to capture value while doing so e are
    the key issues in designing a business model: it is not enough to do the first without the second. The
    imperfections in the market for knowhow make capturing value from its production and sale in-
    herently difficult,21 and may often necessitate a business model where knowhow is bundled into
    products and complementary assets used to realize value to the innovator. This involves some of
    the trickiest and most frustrating issues that entrepreneurs and managers must address.

    The Profiting from Innovation framework is an effort to help entrepreneurs and strategists figure
    out appropriate business model/designs and technology strategies by delineating important features
    of business model choice, and predicting the outcomes from those choices. The framework employs
    contracting theory,22 and recognizes two extreme modes (models) by which innovators can capture
    value from innovation:

    � At one end of the scale stands the integrated business model, in which an innovating firm bun-
    dles innovation and product together, and assumes the responsibility for the entire value chain
    from A to Z including design, manufacturing, and distribution. Clearly, companies that have the
    right assets already in place are well equipped to do this; but the framework also indicates when
    the internal development and commercialization strategy is a necessity.

    � The other extreme case is the outsourced (pure licensing) business approach, one that has been
    embraced by a number of companies, like Rambus (semiconductor memory) and Dolby (high
    fidelity noise reduction technology). With respect to licensing versus internal commercialisation
    by the innovator, the framework yields answers calibrated according to the strength of the ap-
    propriability/intellectual property regime. Thus one could license e and expect the licensing
    model to work e only if one had strong intellectual property rights: without them the licensee
    might well be the one who captures value, at the expense of the innovator.

    � In between there are hybrid approaches involving a mixture of the two approaches (e.g. out-
    source manufacturing; provide company owned sales and support). Hybrid approaches are
    the most common, but they also require strong selection and orchestration skills on the part
    of management.23

    [a] licensing model [will only] work [with] strong intellectual property

    rights. [otherwise] the licensee will capture value, not the innovator.

    184 Business Models, Business Strategy and Innovation

    The Profiting from Innovation framework can thus be considered as a tool to help design busi-
    ness models, and using it allows one to map business model selection to type of innovation, while
    simultaneously enabling one to figure out where intellectual property monetization through licens-
    ing is likely to be viable, and where it’s not, or where some kind of vertical integration is indicated.24

    Although, (by construction) it is silent on many issues such as market segmentation and the choice
    of product features, it nevertheless can provide insights into how a value chain ought to be
    assembled. And it can predict winners and losers from the competitive process in the context where
    a customer need is being met.

    ‘Public’ goods and the bundling and unbundling of inventions and products
    Inventors and innovators rarely enjoy strong intellectual property protection. One well studied (and
    reasonably well understood) situation where there are serious value capture problems is investment
    in basic research and the production of scientific knowledge. Basic research usually ends up in sci-
    entific publications, so it is hard e if not impossible e to secure strong intellectual property pro-
    tection for scientific knowledge. As a result, it is very difficult to charge for discoveries, even if they
    have the potential to generate high value for society, so very few firms invest in basic research. Spill-
    overs (externalities) are simply too large; profiting from discovery is simply too difficult. There is no
    easy for-profit business model for capturing value from scientific discoveries in a world where sci-
    ence wants to be open and rapid dissemination of scientific knowledge through journals, confer-
    ences and professional contacts is almost inevitable: not surprisingly, most basic research is not
    funded by business firms, but by governments.

    Investment in scientific research is an example of what economists call ‘public goods’; a circum-
    stance in which the economic activity in question generates positive externalities or ‘spill-overs’. As
    there is no good (private) business model that can support value capture, government funding and/
    or philanthropy is required and provided. Viewed in this way, the concept of the ‘business model’
    can be integrated into almost a century of economic thought about the design of institutions and
    the role of enterprise and government in civil society. Market ‘failures’ occur in the context of
    innovation when private business models for capturing value draw forth insufficient investment
    in R&D.

    Putting basic science to one side, the most common business model to capture value from in-
    ventions is to embed them in a product, rather than simply trying to sell designs or intellectual
    property. This approach allows those that invest in R&D to ameliorate (to some degree) their
    lack of intellectual property protection. The latest cell phone, digital camera or automobile doesn’t
    come with a price for the product and an unbundled price for knowhow and/or intellectual prop-
    erty: invention/technology and product are typically bundled together, although (in theory) they
    don’t need to be.

    This discussion makes it apparent that market failures (with respect to R&D investment) are
    partly a function of the ability (or lack thereof) of entrepreneurs to create viable business models
    using the mechanisms available to them. As noted, one way to try and get around market failures in
    the ‘market for inventions’ is to bundle invention(s) and complements into products. But too often,
    firms (and in particular small start-ups) under-employ the available mechanisms, just offering cus-
    tomers ‘items’ of technology such as devices or discrete technology components. Just by itself, this
    may not represent a customer solution; a business model based on simply selling an invention e or
    even an innovative component or ‘item’ e may not enable the innovator to capture a significant
    share of the value that might be generated by their innovative technology, unless it has ironclad
    patent protection and is critical to an important and already recognized application. The proper
    ‘marketing’ of new technology often requires much more.25 The bundled provision of complemen-
    tary products and services is often necessary, not just to help capture value, but to help create it in
    the first place.

    The problem is quite general. When value delivery involves employing intangible (knowhow) as-
    sets, pricing and value capture are difficult because of the nonexistence of perfect property rights,
    which means that markets can’t work well, as Coase and many others have explained.26 As

    Long Range Planning, vol 43 2010 185

    illustrated above, many Internet services are simply provided for ‘free’ as a way to build brand and
    to indirectly promote a related value added service, and we have seen how a mixture of revenue
    approaches is usually required when trying to sell on the Internet.27 But bundling, while a common
    and helpful approach, isn’t always necessary. When the innovator has a strong patent, it is some-
    times possible to capture value either by naked licensing e or even outright sale e of intellectual
    property. Different models of value capture are available where intellectual property rights exist and
    can be enforced e so designing business models often requires the skill of the intellectual property
    lawyer as well as that of the entrepreneur.

    To summarize: the traditional revenue model used by innovators to capture value from technol-
    ogy involves the consumer buying (and paying for) products that have intellectual property embed-
    ded within them e the method is so common that it is rarely noticed or reflected on.28 This works
    well, particularly if an attractive bundled solution can be offered, if there is strong intellectual prop-
    erty, or if imitability is otherwise difficult. Many scientific discoveries and inventions are poorly
    protected by intellectual property rights, and require business models that feature public funding,
    or crafty ways to otherwise capture positive spill-overs.

    business models innovation may not seem heroic [but] without it there

    may be no reward for pioneering individuals, enterprises and nations.

    Business models as innovation
    Technological innovation is lionized in most advanced societies; that is a natural and desirable re-
    flection of the values of a technologically progressive society. However, the creation of new orga-
    nizational forms (like the Skunk Works and the multidivisional organizational structure),
    organizational methods (like the moving assembly line), and in particular new business models
    are of equal e if not greater e importance to society, and to the business enterprise. While such
    innovation may seem less heroic to many citizens e even to many scientists and engineers e with-
    out it technological innovation may be bereft of reward for pioneering individuals, as well as for
    pioneering enterprises and nations.

    The capacity of a firm (or nation) to capture value will be deeply compromised unless the capacity
    exists to create new business models. As noted, even an inventor as celebrated as Thomas Edison had
    a questionable track record in terms of business model innovation, abandoning the recording business
    and also failing to get direct (rather than alternating) current adopted as the industry standard for elec-
    tricity generation and transmission. History shows that, unless they can offer compelling value prop-
    ositions to consumers/users and set up (profitable) business systems to satisfy them with the requisite
    quality at acceptable price points, the innovator will fail, even if the innovation itself is remarkable, and
    goes on to be widely adopted by society. Of course, this makes management, entrepreneurship and
    business model design and implementation as important to economic growth as is technological in-
    novation itself. Technological creativity that is not matched by business resourcefulness and creativity
    (in designing business models) may not yield value to the inventor or even to their society.

    As discussed and illustrated in many earlier examples, technological innovation often needs to be
    matched with business model innovation if the innovator is to capture value. There are of course
    exceptions e for example, small improvements in the manufacturing process (even if cumulatively
    large) will usually not require business model innovation, and value can be captured by lowering
    price and expanding the market and market share. But the more radical the innovation, and the
    more challenging the revenue architecture, the greater the changes likely to be required to tradi-
    tional business models. And, as indicated by some of the earlier examples, business model innova-
    tion may help to establish a differentiable competitive advantage. Dell didn’t bring any
    improvements to the technology of the Personal Computer e but it did combine both suppliers’
    and its own organizational/distribution system innovations to deliver compelling value to end

    186 Business Models, Business Strategy and Innovation

    users: as have Southwest Airlines, Virgin, Virgin Blue, and JetBlue in the air passenger transport
    sector.

    Sometimes the creation of new business models leads to the creation of new industries. Consider the
    payment card industry (the core of which is credit and debit cards). The card companies provide net-
    work services, associate with banks who issue the cards, and associate with acquirers who sign up mer-
    chants to accept credit cards. Early on in the life of the industry, merchants were unwilling to accept
    a payment card that few consumers carried, just as card holders didn’t want cards that merchants did
    not accept. As Evans and Schmalensee note, inventing a new business model for credit e the credit
    card e ‘required the industry’s founders to invest enormous amounts of capital and ingenuity’.29

    Companies should be seeking and considering improvements to business models e particularly
    difficult to imitate improvements that add value for customers e at all times. Changing the firm’s
    business model literally involves changing the paradigm by which it goes to market, and inertia is
    likely to be considerable. Nevertheless, it is preferable for the firm to initiate such a change itself,
    rather than have it dictated by external events, as several investment banks in the U.S. and elsewhere
    have experienced recently.

    The role of discovery, learning and adaptation
    Designing a new business model requires creativity, insight, and a good deal of customer, compet-
    itor and supplier information and intelligence. There may be a significant tacit component. An en-
    trepreneur may be able to intuit a new model but not be able to rationalize and articulate it fully; so
    experimentation and learning is likely to be required. As mentioned earlier, the evolving reality im-
    pacting customers, society, and the cost structure of the business must be understood. It is often the
    case that the right business model may not be apparent up front, and learning and adjustments will
    be necessary: new business models represent provisional solutions to user/customer needs proposed
    by represent entrepreneurs/managers. As Shirky recognizes, a business model is provisional in the
    sense that it is likely over time to be replaced by an improved model that takes advantage of further
    technological or organizational innovations. The right business model is rarely apparent early on in
    emerging industries: entrepreneurs/managers who are well positioned, who have a good but not
    perfect business model template but who can learn and adjust, are those more

    likely to succeed

    .30

    The right business model is rarely apparent early on. entrepreneurs/

    managers who are well positioned and can learn and adjust are more

    likely to succeed

    Technological change often provides the impetus for new and better ways to satisfy customer
    needs. The horse, then the railroad, the auto and the airplane have all been technological solutions
    to society’s basic transport needs that successively complemented and displaced each other, and
    formed the basis of competing business models for carrying people from one place to another.
    The Internet and the communication and computer revolution have empowered customers, and
    both allowed and required more differentiation in product service offerings. Social networking is
    also trumping the age-old ability of using advertising to get to an audience. As Peter Sealey has
    noted with respect to new movies releases, ‘the star-power opening is fading in importance and the
    marketing and releasing of movies is going into new territory where the masses are molding the opinion
    of a movie’,31 and studio executives are having to recognize these new realities and adjust their busi-
    ness models accordingly.

    In short, one needs to distil fundamental truths about customer desires, customer assessments,
    the nature and likely future behavior of costs, and the capabilities of competitors when designing
    a commercially viable business model. Traditional market research will not often be enough to

    Long Range Planning, vol 43 2010 187

    identify as yet unarticulated needs and/or emerging trends. Changes with respect to the relative
    merits of particular organizational and technological solutions to customer needs must also be
    considered.

    Consider again the question of how society will gather and distribute the news of the day. First it
    was the town crier; later the newspaper; today the Internet has become increasingly important.
    Communication costs have dropped dramatically; but now advertising revenues are shrinking
    too. Generally, when the underlying technology changes, and an established logic for satisfying con-
    sumer needs e e.g. newspapers for providing news e is overturned, the business model must
    change too. But technological change is not always a trigger e or always necessary e to reshaping
    the business model.

    Not surprisingly, the invention of new business models can originate from many potential sour-
    ces. What business models pioneers often possess e or develop e is an understanding of some
    ‘deep truth’ about the fundamental needs of consumers and how competitors are or are not satis-
    fying those needs, and of the technological and organizational possibilities (and trajectories) for im-
    provement e some of them, though, just stumble into such understandings. In almost every case,
    however, a new business model is successfully pioneered only after considerable trial and error.

    Those entrepreneurs who understand ‘deep truths’ and can figure out what customers want and
    design a better way to satisfy them (and build sustainable organizations to address these customer
    needs) are business pioneers. They may or may not use new technology, but they must understand
    customer needs, technological possibilities, and the logic of organization. Put differently, a business
    model articulates the underlying business or ‘industrial logic’ of a firm’s go-to-market strategy.
    Once articulated, it is likely that the logic will have to be tested and retested, adjusted and tuned
    as the evidence with respect to provisional assumptions becomes clarified.

    Netflix (discussed above), the largest online DVD rental service in the U.S., offers a flat-fee DVD
    movie rental service that, by 2007, was serving over 6 million subscribers from its collection of
    75,000 titles.32 Subscribers can use the website’s browse function to search for movies by genre,
    and use an extensive movie recommendation system based on other users’ ratings to add to their
    ordered list for delivery via mail. At its initial launch, the Netflix business model was based on
    a pay-per-rental service, but this initial pricing model did not succeed, and the company almost
    failed. It was clear to management Netflix had to rejig its business model and, between September
    and October 1999, it reinvented itself with a subscription model (the ‘Marque Program’). It ended
    its pay-per-rental model entirely, and evolved the monthly fee program to allow subscribers to rent
    any number of DVDs per month (although only a limited number at any one time). The model was
    supported by a system of regional distribution centers which ensured next day delivery to over 90%
    of subscribers. Clearly, it took a while to be able to ascertain the right price points and the manner
    of pricing that was most acceptable to the customer base for its new service; but as Netflix manage-
    ment figured out viewer convenience, wants and willingness to pay, it adjusted its business model
    accordingly. This ability to perceive and adapt saved Netflix and laid the foundation for its growth
    and development: by 2006 it had reached almost $1 billion in revenues.

    Selecting the right ‘architecture’ and pricing model for a business requires not just understanding
    the choices available, but also assembling the evidence needed to validate conjectures and hunches
    about costs, customers, competitors, complementors, distributors and suppliers takes detailed fact-
    specific inquiry, and a keen understanding of customer needs and customer willingness to pay, as
    well as of competitor positioning and likely competitive responses. Entrepreneurs and executives
    must make many informed guesses about the future behavior of customer and competitor, as
    well as of costs. As the evidence with respect to initial conjectures becomes available, they need
    to adjust accordingly. Being fast in learning and making the requisite adjustments to the model
    is important.

    A helpful analytic approach for management is likely to involve systematic deconstruction/un-
    packing of existing business models, and an evaluation of each element with an idea toward refine-
    ment or replacement. The elements of a business model must be designed with reference to each
    other, and to the business/customer environment and the trajectory of technological development

    188 Business Models, Business Strategy and Innovation

    in the industry. While the questions are not as crisp as one would like, and the answers are likely to
    be ambiguous, endeavoring to answer them will impose some discipline and at least help one sort
    business propositions that are likely to be viable from those that are not. For instance, business
    propositions that are no more than good ideas fall short; likewise propositions that involve captur-
    ing 1% of huge markets show a lack of understanding of differences amongst (potential) customers,
    market segments and competition. And wonderfully novel (gimmicky) product concept that meets
    the needs of but a handful of potential customers is unlikely to yield much value. Periodic review
    can increase the chances of avoiding blind spots: long-lived structural elements e choices made per-
    haps decades ago in different environments e need to be scrutinized especially thoroughly.

    A provisional business model must be evaluated against the current

    state of the business ecosystem, and against how it might evolve

    A provisional business model must be evaluated against the current state of the business ecosys-
    tem, and also against how it might evolve. Questions to consider (which are summarized in
    Figure 3) include:

    � How does the product or service bring utility to the consumer? How is it likely to be used? In-
    asmuch as innovation requires the provision of complements, are the necessary complements al-
    ready available to the consumer with the convenience and price that is desirable (or possible)?;

    � What is the ‘deep truth’ about what customers really value and how will the firm’s service/prod-
    uct offering satisfy those needs? What might the customer ‘pay’ for receiving this value?;

    � How large is the market? Is the product/service honed to support a mass market?;
    � Are there alternative offerings already in the market? How is the offering superior to them?;
    � Where is the industry in its evolution? Has a ‘dominant design’ emerged? Strategic requirements

    are likely to be different in the pre- and post-paradigmatic periods;33

    � What are the (contractual) structures needed to combine the activities that must be performed to
    deliver value to the consumer? Both lateral and vertical integration and outsourcing issues need
    to be considered. (Contract theory/transaction cost economics is a useful lens through which to
    view many of these issues. So is capability theory);

    � What will it cost to provide the product/service? How will those costs behave as volume and
    other factors change?; and

    � What is the nature of the appropriability regime? How can imitators be held at bay, and how
    should value be delivered, priced, and appropriated?

    As the author has noted in previous work, beyond specifying a realistic revenue architecture, de-
    signing a business model also involves determining the set of lateral (complementary) and vertical
    activities that must be performed and assessing whether and how they can be performed sufficiently
    cheaply to enable a profit to be earned, and who is to perform them. It involves figuring out the
    market entry strategy e while entry timing is a strategic, rather than a business model issue, it
    may depend in part on the business model employed, particularly the complements already in
    place.34

    When establishing a new business there is likely to be uncertainty with respect to all of the above.
    Disappointments are certain to arise as a new business is built, but success rates can be improved if
    the architects of the business model learn quickly, and are able to adjust within a range that still
    yields a satisfactory profit.

    Of course, once a business model is successfully established, changing technology and enhanced
    competition will require more than defenses against imitation. It is also likely that even successful
    business models will at some point need to be revamped, and possibly even abandoned. For exam-
    ple, as the value proposition associated with the traditional personal computer software licensing

    Long Range Planning, vol 43 2010 189

    Figure 3. Questions to ask about a (provisional) business model

    model (whereby periodic updates would require the purchase of new software licenses and addi-
    tional maintenance costs) has weakened for some customers, Microsoft has changed elements of
    its business model to allow renting so as to compete with cheap or free Web alternatives. According
    to one source, Microsoft is ‘overhauling not only what it makes but how to deliver and charge for it’.35

    Microsoft has apparently begun to offer its Exchange email server program for a monthly fee, as
    well as a ‘barebones’ version of Office for free, supported in part by online advertising (in fact,
    it appears now to be offering some products under the ‘freemium’ philosophy described earlier).
    The evidence is not yet in as to whether it will work well for both Microsoft and its customers.

    designing good business models is an ‘art’ .. the chances are greater if

    entrepreneurs and managers have a deep understanding of user needs

    and are good listeners and fast learners.

    Clearly, designing good business models is in part an ‘art’. The chances of good design are greater
    if entrepreneurs and managers have a deep understanding of user needs, consider multiple alterna-
    tives, analyze the value chain thoroughly so as to understand just how to deliver what the customer
    wants in a cost-effective and timely fashion, adopt a neutrality or relative efficiency perspective to
    outsourcing decisions, and are good listeners and fast learners. Useful tools include the various
    types of market research that lead to a deep understanding of the user, along with elements of
    the Profiting from Innovation framework such as the innovation cycle, appropriability regimes,
    complementary assets and intellectual property systems.

    The selection/design of business models is a key microfoundation of dynamic capabilities e the
    sensing, seizing, and reconfiguring skills that the business enterprise needs if it is to stay in synch
    with changing markets,36 and which enable it not just to stay alive, but to adapt to and itself shape
    the (changing) business environment. Dynamic capabilities help govern evolutionary fitness, and

    190 Business Models, Business Strategy and Innovation

    help shape the business environment itself. Get the business model wrong, and there is almost no
    chance of business success e get it right, and customize it for a market segment and build in non-
    imitable dimensions, and it will contribute to the firm’s competitive advantage.

    Magretta claims that business models are ‘variations on the generic value chain underlying all busi-
    nesses’. This view would seem to overlook that a business model is only partly about how to orga-
    nize the value chain e it is also about figuring out the value proposition to the customer as well as
    the value capture mechanism. A sustainable business model is as much (as the current author has
    noted) about where to position within the value chain i.e. what are the key bottleneck assets to own/
    control in order to capture value. Clearly, the industry must perform various activities in the value
    chain e but which one(s) the firm chooses to undertake is very much a business model choice.

    Recognized (but not fully developed here) is the notion that a business model cannot be assessed
    in the abstract; its suitability can only be determined against a particular business environment or
    context. Neither business strategies, business structures nor business models can be properly cali-
    brated absent assessment of the business environment; and of course the business environment it-
    self is, in part, a choice variable; i.e. firms can both select a business environment, and be selected by
    it: and they can also shape their environment.

    .the business environment itself is a choice variable: firms can select

    a business environment or be selected by it: they can also shape it

    Zott and Amit bravely endeavor to hypothesize as to the appropriate mapping of business models
    to two product market choices: cost leadership and differentiation.37 However, our state of under-
    standing as to the precise relationship between business model choice and enterprise performance is
    both highly context dependent and rather primitive. In certain contexts (e.g. market entry strategies
    for innovators) testable propositions have been advanced (including by the current author), but
    strategic studies will have to advance further as a field before mapping can be anything other
    than suggestive.

    Of course, it may very well take time to get a business model right. Pioneers, in particular, are
    often forced to make only educated guesses as to what customers want, what they will pay for,
    and the cost structures associated with various ways to organize. As the author’s ‘Profiting from
    Innovation’ paper discusses, especially in the pre-paradigmatic industry evolution phase, it is nec-
    essary to stay flexible, experiment with the product and the business model and learn, both from
    one’s own and one’s competitors’ activities, and to keep sufficient financial resources on hand to
    remain an industry participant e and hopefully the market leader e by the time the ‘dominant de-
    sign’ emerges in the market. Indeed, one hopes to be the promoter/owner of this dominant
    design e and to have the capacity to capitalize on the situation.

    Conclusion
    All businesses, either explicitly or implicitly employ a particular business model. A business model de-
    scribes the design or architecture of the value creation, delivery and capture mechanisms employed. The
    essence of a business model is that it crystallizes customer needs and ability to pay, defines the manner by
    which the business enterprise responds to and delivers value to customers, entices customers to pay for
    value, and converts those payments to profit through the proper design and operation of the various el-
    ements of the value chain. Put differently, a business model reflects management’s hypothesis about what
    customers want, how they want it and what they will pay, and how an enterprise can organize to best meet
    customer needs, and get paid well for doing so. The goal of this article has been to advance understanding
    of the considerable significance of business models and to explore their connections to business strategy,
    innovation management and economic theory.

    Long Range Planning, vol 43 2010 191

    One key conclusion of the analysis is that, to be a source of competitive advantage, a business
    model must be something more than just a good logical way of doing business. A model must
    be honed to meet particular customer needs. It must also be non-imitable in certain respects, either
    by virtue of being hard to replicate, or by being unpalatable for competitors to replicate because it
    would disturb relationships with existing customers, suppliers, or important alliance partners. A
    business model may be difficult for competitors to replicate for other reasons too. There may be
    complicated process steps or strong intellectual property protection, or organizational structures
    and arrangements may exist that will stand in the way of implementing a new business model.
    Good business model design and implementation involves assessing such internal factors as well
    as external factors concerned with customers, suppliers, and the broader business environment.

    to be a source of competitive advantage, a business model must be

    more than just a good logical way of doing business .. It must be

    honed to meet particular customer needs .

    The paucity of literature (both theoretical and practical) on the topic is remarkable, given the
    importance of business design, particularly in the context of innovation. The economics literature
    has failed to even flag the importance of the phenomenon, in part because of an implicit assump-
    tion that markets are perfect or very nearly so. The strategy and organizations literature has done
    little better. Like other interdisciplinary topics, business models are frequently mentioned but rarely
    analyzed: therefore, they are often poorly understood. Not surprisingly, it is common to see great
    technological achievements fail commercially because little, if any, attention has been given to de-
    signing a business model to take them to market properly.

    This can and should be remedied. Increased understanding of the essence of business models and
    their place in the corpus of the social and organizational sciences should help our understanding of
    a variety of subjects including market behavior, competition, innovation, strategy and competitive
    advantage. Our understanding of the nature of the firm itself, together with the role of entrepre-
    neurs and managers in the economy and in society, should also benefit from a better appreciation
    of business models and their role in entrepreneurship, innovation and business performance.

    great technological achievements commonly fail commercially

    because little attention has been given to designing a business model

    to take them to market properly. This can and should be remedied.

    Acknowledgement
    I would like to thank Charles Baden-Fuller and Ian MacMillan for their invitation to contribute to this
    Special Issue, and Michael Akemann, Sebastien Belanger, John Blair, Hank Chesbrough, Michael Katz,
    Doug Kidder, David Mitchell, Charles O’Reilly, Richard Rumelt, Alexander Stern, Leigh Teece and
    Steve Lewis as well as the principals of Living PlanIT SA for helpful insights into the issues discussed
    here. The skilful assistance of Patricia Lonergan in preparing the manuscript is gratefully acknowledged.

    References

    1. There are other (related) definitions of a business model. Amit and Zott see R. Amit and C. Zott, Value

    creation in e-business, Strategic Management Journal 22, 493e520 (2001); and C. Zott and R. Amit, The fit

    192 Business Models, Business Strategy and Innovation

    between product market strategy and business model: implications for firm performance, Strategic Man-
    agement Journal 29, 1e26 (2008) define a business model as ‘the structure, content, and governance of trans-
    action’ between the focal firm and its exchange partners (e.g. customers, vendors, complementors). For yet
    another alternate definition see Chesbrough and Rosenbloom (see following).

    2. H. Chesbrough and R. S. Rosenbloom, The role of the business model in capturing value from innova-
    tion: evidence from xerox corporation’s technology, Industrial and Corporate Change 11(3), 529e555
    (2002).

    3. The end of the free lunch e again, The Economist 390(8623) (March 21st 2009).
    4. See K. Arrow, The Limits of Organization, Norton, New York, (1974). The ArroweDebreu model of com-

    petitive equilibrium has everything priced; but, as Arrow himself notes elsewhere, ‘in a strictly technical
    and objective sense, the price system does not work. You simply cannot price certain things (p. 22) and ‘ trust
    and similar values, loyalty and truth telling e are not commodities for which trade in the open market is tech-
    nically possible or even meaningful. (p. 23). ‘ A firm. provides another major area within which price
    relations are held in partial abeyance. (p. 25).

    5. The structure-conduct-performance paradigm in the field of industrial organization is possibly an excep-
    tion. It stressed that concentrated markets were more profitable. If translated into management/strategy
    nostrums, as Michael Porter, Competitive Strategy, Free Press, (1982) did, it suggest the benefits of either
    scale or differentiation as profit drivers. While scale and differentiation may still assist as profit drivers, the
    situation in the modern economy is that in many circumstances, these nostrums can be quite misleading.

    6. O. E. Williamson, Organizational innovation: the transaction-cost approach (1983), in J. Ronen (ed.),
    Lexington Books, Lexington, MA (1983).

    7. R Miles, G. Miles, C. Snow, K. Blomquist and H. Rocha, Business Models, Organizational Forms, and
    Managerial Values, Working paper, UC Berkeley, Haas School of Business (2009). The authors note
    how new business models, new organizational forms, new management approaches, and entrepreneurship
    are the foci of different groups of scholars who rarely meet.

    8. G. Porter, The Rise of Big Business, 1860e1910, Harland Davidson, Arlington Heights, Illinois, (1973) p. 49.
    9. C. W. Ebeling, Evolution of a box: the invention of the intermodal shipping container revolutionized the

    international transportation of goods, Invention and Technology 8e9 (2009).
    10. Apple’s iTunes music store is an example of a business model innovation, and was the first legal pay-

    as-you-go method for downloading music. Time Magazine hailed it as ‘ the coolest invention for 2003’.
    11. See C. Shapiro and H. Varian, Information Rules: A Strategic Guide to the Network Economy, Harvard Busi-

    ness School Press, Boston, MA, (1999) The rules for strategic engagement that they promulgate are core
    elements of strategy in the information services sector, and here e as elsewhere e the design of business
    models to support sustainable competitive advantage must be informed by strategy analysis.

    12. A. Shuen, Web 2.0: A Strategy Guide, O’Reilly, Sebastopol, (2008) p. 2.
    13. See also J. B. Harreld, C. A. O’Reilly and M. L. Tushman, Dynamic capabilities at IBM: driving strategy

    into action, California Management Review 49(4) (2007).
    14. M. Dell, The Early Entrepreneurial Years in Starting a Business, Harvard Business School Press, (2008) In-

    deed, a critical element of Dell’s success is not just the way it has organized the value chain, but also the
    products that it decides to sell through its distribution system. The initial products were personal
    computers, but now include printers, digital projectors, and computer-related electronics.

    15. Quoted in J. Magretta, Why business models matter, Harvard Business Review 6 (2002).
    16. For an insightful treatment of the Google story, see D. A. Vise, The Google Story, Bantam Dell, New York

    (2008).
    17. See J. M. de Figueiredo and D. J. Teece, Mitigating Procurement hazards in the context of innovation,

    Industrial and Corporate Change 5(2), (1996) for an analysis of some ways to mitigate the hazards of
    competing with one’s suppliers.

    18. S. Lippman and R. Rumelt, Uncertain imitability: an analysis of interfirm differences in efficiency under
    competition, Bell Journal of Economics 13, 413e438 (1982).

    19. D. J. Teece, Profiting from technological innovation: implications for integration, collaboration, licensing
    and public policy, Research Policy 15(6), 285e305 (1986); D. J. Teece, Reflections on profiting from tech-
    nological innovation, Research Policy 35(8), 1131e1146 (2006).

    20. See D. J. Teece (1986) ibid.; and G. Pisano and D. J. Teece, How to capture value from innovation: shap-
    ing intellectual property and industry architecture, California Management Review 50(1), 278e296 (2007).

    21. D. J. Teece, The multinational enterprise: market failure and market power considerations, Sloan Manage-
    ment Review 22(3), 3e17 (1981).

    Long Range Planning, vol 43 2010 193

    22. S. Winter, The logic of appropriability: from Schumpeter to Arrow to Teece, Research Policy 35,
    1100e1106 (2006).

    23. D. J. Teece, Explicating dynamic capabilities: the nature and microfoundations of (sustainable) enterprise
    performance, Strategic Management Journal 28(13), 1319e1350 (2007).

    24. An application of the framework in the biotech industry context is discussed in G. Pisano, Science Business:
    The Promise, the Reality, and the Future of Biotech, Harvard Business School Press, (2006) which includes
    a carefully analysis of the sources of failure in the market for know how.

    25. For further development of this idea, see W. Davidow, High Technology Marketing, Free Press (1986).
    26. R. Coase, The problem of social cost, Journal of Law and Economics 3, 1e44 (1960).
    27. The proliferation of illegal digital downloads of recorded music has led recording companies to try to in-

    sist on (and sometimes achieve) so called ‘360 contracts’, so they can participate in all sources of revenue
    from their artistes’ activities e branded clothing, performances, and other public appearances e as well as
    just recorded music.

    28. D. J. Teece (1986) op. cit. at Ref 19; D. Somaya and D. J. Teece, Patents, licensing and entrepreneurship:
    effectuating innovation in multi-invention contexts in Sheshinki, Baumol and Strom (eds.), Entrepreneur-
    ship, Innovation, and the Growth of Free-Market Economies, Princeton University Press (2007).

    29. D. Evans and R. Schmalensee, Paying with Plastic, MIT Press, Cambridge, (1999) p. 3.
    30. C. Shirky, Here Comes Everybody: The Power of Organizing Without Organizations, Penguin, New York

    (2008).
    31. See Claudia Eller ‘Little Love this Summer for A-List Movie Actors’, Los Angeles Times, 29 June, 2009.
    32. http://ir.netflix.com (visited April 2007).
    33. See D. J. Teece (1986) op. cit. at Ref 19.
    34. W. Mitchell and Dual Clocks, Entry order influences on industry incumbents and newcomer market share

    and survival when specialized assets retain their value, Strategic Management Journal 12(2), 85e100
    (1991).

    35. Peter Burrows, Microsoft defends its empire, Business Week p. 28 (6 July 2009).
    36. D. J. Teece, G. Pisano and A. Shuen, Dynamic capabilities and strategic management, Strategic Manage-

    ment Journal 18(7), 509e533 (1997); D. J. Teece (2007), op. cit. at Ref 23; D. J. Teece, Dynamic Capabil-
    ities and Strategic Management: Organizing for Innovation and Growth, Oxford University Press (2009).

    37. C. Zott and R. Amit (2008), op. cit. at Ref 1.

    Biography
    David J. Teece has a Ph.D. in economics from the University of Pennsylvania. His research interests span

    industrial organization, business strategy, organizational economics, and public policy. He is the author of over

    200 published articles and books. His most recent book is Dynamic Capabilities and Strategic Management:

    Organizing for Innovation and Growth (Oxford University Press, 2009). He has four honorary doctorates and

    was the co-founder and Vice Chairman of LECG Corporation. Institute for Business Haas School of

    Business University of California, Berkeley Berkeley, California 94720. Tel: 510-642-1075; Fax: 510-642-2826;

    E-mail: davidjteee@teece.net

    194 Business Models, Business Strategy and Innovation

    http://ir.netflix.com

    http://davidjteee@teece.net

      Business Models, Business Strategy and Innovation
      Introduction
      Business models – the theoretical foundation
      Examples of business models
      Traditional industries
      The information/internet industries
      Business models, strategy and sustainable competitive advantage
      Barriers to imitating business models
      Business model learning
      Business models to capture value from technological innovation
      The profiting from innovation framework
      ‘Public’ goods and the bundling and unbundling of inventions and products
      Business models as innovation
      The role of discovery, learning and adaptation
      Conclusion
      Acknowledgements
      References

    1/23/2020

    Business Strategy and Forecasting as Competitive Advantages Scoring Guide

    https://courserooma.capella.edu/bbcswebdav/institution/BMGT/BMGT8130/200100/Scoring_Guides/u02a1_scoring_guide.html 1/1

    Business Strategy and Forecasting as Competitive Advantages Scoring Guide

    Due Date: End of Unit 2
    Percentage of Course Grade: 30%.

    CRITERIA NON-PERFORMANCE BASIC PROFICIENT DISTINGUISHED

    Describe a business
    model, using Teece
    (2010) to analyze the
    strategy of the
    organization.
    20%

    Does not
    explain the
    background of
    the business as
    related to its
    strategic
    business model,
    and/or does not
    cite Teece
    (2010).

    Provides the
    background of a
    business and its
    model, but does not
    align with Teece
    (2010), or fails to
    provide strategic
    insights.

    Describes a business
    model, using Teece
    (2010) to analyze the
    strategy of the
    organization.

    Analyzes a business
    model, using Teece (2010)
    to evaluate the strategy of
    the organization.

    Describe the
    business strategy
    using market
    segmentation, value
    proposition,
    apparatus, and
    prevention of
    imitability, and
    describe the
    competitive
    advantage of the
    organization.
    Substantively use
    Teece (2010) as
    support.
    20%

    segmentation,
    value
    proposition,
    apparatus, or
    prevention of
    imitability, and
    does not
    describe the
    competitive
    advantage of
    the
    organization.

    Discusses some of
    the business strategy
    using market
    segmentation, value
    proposition,
    apparatus, or
    prevention of
    imitability. Does not
    cite Teece (2010) for
    substantive
    purposes.

    Describes the business
    strategy using market
    segmentation, value
    proposition, apparatus,
    and prevention of
    imitability, and describes
    the competitive
    advantage of the
    organization.
    Substantively uses Teece
    (2010).

    Evaluates the business
    strategy using market
    segmentation, value
    proposition, apparatus, and
    prevention of imitability,
    and describes the
    competitive advantage of
    the organization. Uses
    Teece (2010) to support the
    evaluation.

    Analyze and
    describe Part C1 of
    the assignment,
    using one of the Red
    Queen articles as
    support.
    20%

    Does not
    incorporate Part
    C1 of the
    assignment.

    Describes Part C1 of
    the assignment but
    fails to explain or
    incorporate the Red
    Queen literature or
    leaves out some
    parts.

    Analyzes and describes
    Part C1 of the
    assignment, using one of
    the Red Queen articles
    as support.

    Thoroughly evaluates,
    analyzes, and describes
    Part C1 of the assignment,
    using both Red Queen
    articles as support.

    Describe the Red
    Queen articles.
    Show some
    differences between
    the articles.
    20%

    Does not
    analyze,
    evaluate, or
    synthesize the
    Red Queen
    articles.

    Recites the
    information in the
    Red Queen articles.

    Describes the Red
    Queen articles. Shows
    some differences
    between the articles.

    Synthesizes the Red
    Queen articles thoroughly.
    Explains how the concept
    has grown in the past
    decade.

    Communicate in a
    manner expected of
    doctoral-level
    composition and
    exhibit critical
    thinking skills.
    20%

    Does not
    communicate in
    a manner
    expected of
    doctoral-level
    composition
    and does not
    exhibit critical
    thinking skills.

    Inconsistently
    communicates in a
    manner expected of
    doctoral composition
    and inconsistently
    exhibits critical
    thinking skills.

    Communicates in a
    manner expected of
    doctoral-level
    composition and exhibits
    critical thinking skills.

    Communicates in a manner
    expected of doctoral-level
    composition, follows APA
    conventions of writing with
    few to no errors, and
    exhibits exceptional critical
    thinking skills.

    THE RED QUEEN EFFECT: COMPETITIVE ACTIONS AND
    FIRM PERFORMANCE

    PAMELA J. DERFUS

    PATRICK G. MAGGITTI
    Temple University

    CURTIS M. GRIMM
    KEN G. SMITH

    University of Maryland

    We investigate the Red Queen effect as a contest of competitive moves or actions among
    rivalrous firms. The results from a multi-industry study of over 4,700 actions confirms
    the existence of Red Queen competition, whereby a firm’s actions increase perfor-
    mance but also increase the number and speed of rivals’ actions, which, in turn,
    negatively affect the initial firm’s performance. We further show that this Red Queen
    effect depends on industry context and a focal firm’s market position.

    The quest to explain performance differences
    among competing firms is a fundamental issue in
    strategic management. A number of answers to this
    complex question have been offered. According to
    the industry structure viewpoint, positioning firms
    in industries where they can take advantage of fa-
    vorable competitive forces, such as barriers to entry
    or mobility (Caves & Porter, 1977), enhances per-
    formance. The resource-based view also empha-
    sizes limiting the behavior of rivals by suggesting
    that firms acquire or develop unique, valuable, and
    rare resources that are difficult for rivals to repli-
    cate (Barney, 1986). Evolutionary theory posits per-
    formance differences among firms are a function of
    a competitive race to discover profit opportunities.
    According to this view, high performance is
    achieved by speed and innovation that keep firms
    ahead of rivals (Nelson & Winter, 1982). Our focus
    in this paper is the latter perspective, perhaps the
    least understood of the three; more specifically, we
    explore “Red Queen competition” in the context of
    actions among rivals.

    Evolutionary and ecology theories focusing on
    Red Queen competition portray how entities dy-
    namically interact and coevolve with one another.
    Introduced by the biologist van Valen (1973), the

    Red Queen effect is based on the conversation be-
    tween the Red Queen and Alice in Lewis Carroll’s
    Through the Looking Glass. In that story, Alice
    realizes that although she is running as fast as she
    can, she is not getting anywhere, relative to her
    surroundings. The Red Queen responds: “Here, you
    see, it takes all the running you can do, to keep in
    the same place. If you want to get somewhere else,
    you must run at least twice as fast as that!” (Carroll,
    1960: 345).1 Van Valen used this analogy to de-
    scribe the continuous and escalating activity and
    development of participants trying to maintain rel-
    ative fitness in a dynamic system. Since then, the-
    orists have used the notion of the Red Queen to
    explain behavior in a variety of settings ranging
    from biology to military arms races (Baumol, 2004;
    Dawkins & Krebs, 1979).

    Applied to a business context, the Red Queen can
    be seen as a contest in which each firm’s perfor-
    mance depends on the firm’s matching or exceed-
    ing the actions of rivals. In these contests, perfor-
    mance increases gained by one firm as a result of
    innovative actions tend to lead to a performance
    decrease in other firms. The only way rival firms in
    such competitive races can maintain their perfor-
    mance relative to others is by taking actions of their
    own. Each firm is forced by the others in an indus-
    try to participate in continuous and escalating ac-
    tions and development that are such that all the
    firms end up racing as fast as they can just to stand
    still relative to competitors. This self-escalating,

    This research was partially supported by the Dingman
    Center for Entrepreneurship at the Robert H. Smith
    School of Business, University of Maryland, College
    Park. The authors would also like to thank our AMJ
    associate editor, R. Duane Ireland, and the three anony-
    mous reviewers for their insightful comments and sug-
    gestions during the revision process.

    1 Through the Looking Glass was originally published
    in 1871.

    � Academy of Management Journal
    2008, Vol. 51, No. 1, 61–80.

    61

    Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express
    written permission. Users may print, download or email articles for individual use only.

    coevolving system of Red Queen competition has
    been empirically shown to affect founding rates
    (Barnett & Sorenson, 2002), failure rates (Barnett &
    Hansen, 1996), and competitiveness (Barnett &
    McKendrick, 2004). Indeed, Baumol (2004) sug-
    gested that the Red Queen effect is the most pow-
    erful mechanism driving economic development in
    capitalistic society.

    Following Barnett and McKendrick (2004), we
    build a model that captures the Red Queen process
    of competitive evolution as both a positive and a
    negative force on focal firm performance in which
    the gains by one firm must come at the expense of
    another. As Barnett and McKendrick noted, “A de-
    fining characteristic of competition is that one or-
    ganization’s solution becomes its rivals’ problem.
    The resulting increased constraints, again in turn,
    are likely to trigger responses among rivals, again
    intensifying competitive constraints on the first or-
    ganization, and so on” (2004: 540). The first pur-
    pose of this study was to explicitly model the Red
    Queen “running as fast as you can” process by
    examining the relationships among focal firm ac-
    tions, rival actions, speed of rival actions, and focal
    firm performance. In doing this, we illustrate theo-
    retically and empirically that focal firm actions ver-
    sus rival actions and speed of rival actions have
    two opposing effects on focal firm performance. We
    refer to this formulation as our baseline model. A
    second goal of our research was to expand the
    baseline model by developing and testing theory
    that begins to identify the conditions that moderate
    the Red Queen effect.

    Our examination of firm action and firm perfor-
    mance and the coevolutionary dependency of firm
    action and rival action on firm performance is fin-
    er-grained and more dynamic than prior Red Queen
    research. Specifically, we contend that firms are
    prompted to search, undertake new actions, and
    learn in an effort to improve performance. This use
    of Red Queen theory is consistent with Schumpet-
    er’s 1942 argument that a dynamic process of “cre-
    ative destruction” occurs when firms launch inno-
    vative actions to gain advantage in the marketplace,
    which is then eroded by their rivals’ competitive
    moves (Schumpeter, 1976). Thus, Schumpeter’s
    perspective figures prominently in the develop-
    ment of our Red Queen model. Additionally, our
    study applies Red Queen theory in the context of
    competitive dynamics, which focuses on the ac-
    tions and reactions of firms (Smith, Grimm, & Gan-
    non, 1992). In line with prior competitive dynam-
    ics research, we define firm actions as “externally
    directed, specific and observable competitive
    moves initiated by a firm to enhance its relative
    competitive position” (Smith, Ferrier, & Ndofor,

    2001: 321). Rival actions are defined as the exter-
    nally directed competitive moves of all rivals in the
    industry in which the focal firm’s participation is
    being studied (Young, Smith, & Grimm, 1996).

    Although our conceptualization of Red Queen
    competition is consistent with Schumpeterian and
    competitive dynamics perspectives, neither
    Schumpeter nor competitive dynamics research
    fully explains the motivating factors behind the
    competitive process. The Red Queen theory out-
    lined in this article provides a more complete pic-
    ture of this competitive process by explaining how
    firms are motivated to search, act, and learn in a
    desire to improve performance. As a consequence,
    our Red Queen theory begins to clarify vital aspects
    of the competitive process—the motivation for ac-
    tion and reaction—not fully explained in prior
    literature.

    Like Barnett and Hansen (1996), we assume that
    a firm facing competition is likely to act. Further,
    we regard the competitive interactions of firms as
    constituting a mechanism for a simple search, ac-
    tion, and learning process firms undertake to im-
    prove performance (March & Simon, 1958). In our
    model, action allows firms to “learn by doing” (Ar-
    gote, 1999; Eisenhardt & Tabrizi, 1995; Pisano,
    1994). Actions of focal firms that increase their
    performance may result in a decline in rival perfor-
    mance, thus prompting those rivals to engage in
    similar search, action, and learning processes. Our
    goal was to model this reciprocal system of focal
    firm actions and rival actions, including the speed
    at which those rival actions occur, and to show the
    system’s effect on focal firm performance. We focus
    on short-run performance, or the effects of focal
    firm action and rival action on focal firm perfor-
    mance in a given year. Though we recognize that
    Red Queen theory also posits long-term conse-
    quences of action exchanges, such as greater fitness
    for all competing firms, these long-term effects are
    beyond the scope of the present work.

    As mentioned above, in addition to examining
    how firm actions, rival actions, and rival action
    speed impact short-term performance, a second
    goal of our research was to develop and test theory
    that begins to identify the conditions that amelio-
    rate or exacerbate the Red Queen effect. Such work
    is important as it may offer insights into how firms
    can successfully adapt or evolve under Red Queen
    competitive pressures. In this research, we focus on
    a set of factors that may facilitate or impede how
    managers learn from action. We argue that a better
    understanding of these factors can help explain the
    motivation for search and action and thereby, Red
    Queen competition. We suggest that industry con-
    centration, industry growth rate, and a firm’s mar-

    62 FebruaryAcademy of Management Journal

    ket position affect the search and action process
    and hence moderate the Red Queen effect.

    RED QUEEN COMPETITION: FOCAL FIRM
    ACTIONS, RIVAL ACTIONS, RIVAL ACTION
    SPEED, AND FOCAL FIRM PERFORMANCE

    In this section, we develop baseline theory that
    explains the Red Queen effect in terms of focal firm
    actions, rival actions, and rival action speed, and
    their combined impact on focal firm performance.
    We argue that search and firm action are motivated
    by a desire to learn new ways of improving perfor-
    mance. However, we assume that the relationship
    between firm action and performance is uncertain,
    dynamic, and subject to constant change by the
    very Red Queen competition that motivates search
    and action. In our model, firms search to discover
    opportunities to act. They experiment in taking
    new actions and learn from the results of action
    about the relationship between action and perfor-
    mance. We do not assume that all action is effective
    or costless, but we do assume that the average per-
    formance benefits of action outweigh the costs.
    Otherwise, firms would not have a motivation
    to act.

    Recent theory on the Red Queen effect has spec-
    ulated about the motivation for firms’ and rivals’
    actions. As Barnett and McKendrick (2004) sug-
    gested, firm aspirations expand, and goals may
    change quickly as a result of comparisons with
    others and Red Queen evolution. For instance, a
    large retailer may increase the frequency of market-
    ing campaigns on the basis of observing a positive
    relationship between prior campaigns and perfor-
    mance. Assume this action results in increased rev-
    enue and profits for the focal firm at the expense of
    a rival’s profits. Rivals may then search for and
    learn of some way to increase their own perfor-
    mance. In this example, rivals may take new price-
    cutting actions. These rival actions may adversely
    affect the performance of the focal firm, thus moti-
    vating additional search, action, and learning for
    this retailer. The learning from action in pursuit of
    profits that drives this action–rival action process
    captures the incremental and coevolutionary na-
    ture of Red Queen competition.

    The motivation to search, act, and learn elicited
    by the Red Queen effect extends Schumpeter’s the-
    ory of creative destruction regarding the relation-
    ship between action and performance in a compet-
    itive context. Schumpeter (1934) highlighted the
    interdependent nature of a competitive market-
    place, arguing that it was the result of, and the
    reason for, continuous innovation and firm action.
    If firms stand still, competitors who introduce new

    combinations that appeal to the market erode the
    inactive firms’ positions. To avoid this erosion,
    firms must continually strive to introduce new
    products, methods, and initiatives. Success, or
    profitable performance, he argued, is more the re-
    sult of “the new commodity, the new technology,
    the new source of supply, the new type of organi-
    zation” than it is the result of control of margins,
    output, or prices (Schumpeter, 1976: 84 – 85). In
    Schumpeter’s world, an efficient but lethargic firm
    would not survive for long.

    Schumpeter also pointed out, however, that in-
    novation and action draw rival action, which he
    referred to as “creative destruction.” The acting
    firm “leads in the sense that he draws other pro-
    ducers in his branch after him” (Schumpeter, 1976:
    89). Thus, innovations and the results of new ac-
    tion are visible to competitors, often spurring rival
    actions and an ongoing cycle of creation and de-
    struction. Indeed, if rivals are to survive in the
    marketplace, they cannot afford to ignore the com-
    petitive actions of other firms; they must also act
    creatively. The innovative competitive interaction
    of firms in pursuit of profits is so fundamental that
    Schumpeter (1976) argued it was the key source of
    market expansion and economic growth.

    Researchers have empirically found that firms
    that are more active (i.e., are running faster) than
    their rivals improve their competitive positions
    (Ferrier, Smith, & Grimm, 1999) and increase their
    performance (Young, 1993; Young et al., 1996),
    while firms that are more sluggish than their rivals
    experience negative performance consequences
    (Miller & Chen, 1994). The basic argument of this
    research has been that more active firms achieve
    greater performance because they have greater as-
    piration levels, are more capable at implementing
    actions, and are perceived by rivals as more aggres-
    sive competitors than are less active firms (Smith et
    al., 2001).

    Yet neither Schumpeter (1976) nor competitive
    dynamics researchers have recognized the role that
    learning from action outcomes may play in fueling
    competition and evolution. Barnett and McKen-
    drick (2004) described how learning drives Red
    Queen competition. When performance falls below
    aspirations, managers will search, act, and learn
    until performance reaches expectations. At the
    heart of this process is a manager seeking to under-
    stand a dynamic world of action cause and effect.
    Weick noted that managers cannot “ignore the ac-
    tion because they are responsible for it” (1995:
    134 –135). Thus, Red Queen theory helps to explain
    how firms incrementally evolve by taking action
    and learning from the results of action in a desire to

    2008 63Derfus, Maggitti, Grimm, and Smith

    improve performance. Given these arguments along
    with prior research, we predict:

    Hypothesis 1a. With the number of rival ac-
    tions held constant, as the number of focal firm
    actions increases, focal firm performance
    increases.

    However, Red Queen competition can also have
    negative consequences for an active firm (Barnett &
    Hansen, 1996; Barnett & McKendrick, 2004). Bar-
    nett and McKendrick contended that one organiza-
    tion’s solution to search and action can become
    another firm’s problem. In this sense, Red Queen
    competition narrows the options for firms, escalat-
    ing rivalry and races, sometimes with limited short-
    term benefits for all. Returning to the above exam-
    ple, consider a case in which a rival’s response was
    not to cut prices but to simply imitate an initially
    acting firm’s behavior by increasing the frequency
    of its own marketing campaigns, thereby regaining
    the revenues shifted by the focal firm’s initial
    action.

    Schumpeter argued that all advantages are tem-
    porary and uncertain because firms interact and the
    “perennial gale of creative destruction” erodes past
    accomplishments (1976: 89). Successful action
    evokes reaction from rivals. Indeed, it is the dy-
    namic process of firm actions and rival actions that
    defines the market process. Firms are spurred to
    engage in a cycle of action as they continually seek
    to learn more about action-performance relation-
    ships. When a firm leads with a new product or
    service, it puts pressure on competitors’ products
    and services, perhaps to the point of rendering
    them obsolete. Those competitors must act if they
    are to stay viable. Some rivals imitate, and others
    make innovative thrusts of their own. Regardless,
    the cycle repeats again and again as rivals struggle
    for profits and market share. It is this competitive
    interaction of rivals in pursuit of profits that results
    in market progress and evolution (Schumpeter,
    1976).

    Recent research supports the coevolutionary na-
    ture of action and reaction. Specifically, in a variety
    of different studies, conducted in a variety of dif-
    ferent industries, researchers have found a positive
    correlation between firm actions and rival actions.
    Indeed, response time, often a measure of compet-
    itive intensity and rival action speed, has ranged
    from as low as 8 days in the airline industry to 2

    4

    days in computer retailing and 124 days in high-
    technology industries (Grimm & Smith, 1997).

    Thus, Red Queen theory recognizes the interde-
    pendent nature of firms. Specifically, the search,
    action, and learning process does not end with a
    focal firm’s actions (Barnett & McKendrick, 2004).

    That is, the improved performance of the firm may
    come at the expense of rivals’ performance, which,
    in turn, may prompt rivals to search, act, and learn
    to improve their own performance. In this sense,
    learning and competition are codependent. To-
    gether, they explain the incremental and relative
    process by which firms evolve as they try to im-
    prove performance.

    Thus, we predict:

    Hypothesis 1b. As the number of focal firm
    actions increases, the number of rival firm ac-
    tions and the speed of rival actions increase.

    A number of studies have focused on the perfor-
    mance consequences of industry rivalry. In a sam-
    ple of software firms, Young and colleagues (1996)
    found that as industry rivalry, measured as number
    of rival actions, increased, focal firm performance
    decreased. Chen and Miller (1994) found that
    higher levels of rival responses decreased perfor-
    mance in the airline industry, and Schomburg,
    Grimm, and Smith (1994) found a negative relation-
    ship between rivalry and profitability in the beer,
    telecommunications, and personal computer in-
    dustries. Finally, Smith and colleagues (1992)
    found that increased competitive actions were re-
    lated to lower profitability in the airline industry.

    Evolutionary scholars have also examined the
    performance consequences of Red Queen competi-
    tion. Barnett and Hansen (1996) argued that a focal
    firm’s superior performance leads a rival to search
    for new opportunities to improve its own perfor-
    mance. Assuming effective actions are found, the
    rival’s position improves at the expense of the focal
    firm. In a study of Illinois banks from 1900 to 1992,
    Barnett and Hansen (1996) found that a focal firm’s
    own competitive experience increased its chances
    of success and survival, whereas its rivals’ aggre-
    gate relative experience decreased the focal firm’s
    success. They argued that firms are constrained by
    their history, falling into competency traps where
    they respond to new developments with old ac-
    tions (Ingram, 2002; Levitt & March, 1988).

    2

    In view of this theory and research, we
    hypothesize:

    Hypothesis 1c. With the number of focal firm
    actions held constant, as the number of rival
    firm actions and the speed of rival actions in-
    crease, focal firm performance decreases.

    2 As noted, although the long-term effects of competi-
    tion on performance might be positive (Porter, 1980), we
    believe the shorter-term effects will be negative.

    64 FebruaryAcademy of Management Journal

    Hypotheses 1a–1c represent our baseline Red
    Queen prediction on the positive and negative con-
    sequences of firm and rival actions for focal firm
    performance. We next consider how this Red
    Queen relationship may be affected by industry
    conditions and market position. Figure 1 is an il-
    lustration of the hypothesized relationships.

    MODERATORS OF THE RED QUEEN EFFECT

    We draw from evolutionary theory (Nelson &
    Winter, 1982), the industry position school of com-
    petitive advantage (Porter, 1980), and action re-
    search (Smith et al., 1992) to propose how industry
    concentration, industry demand conditions, and
    market position moderate the baseline model re-
    garding the relationship between focal firm action,
    rival action, rival action speed, and firm perfor-
    mance. We theorize that these factors moderate the
    relationship between focal firm action, rival action,
    rival action speed, and focal firm performance by
    affecting the ability of a focal firm and rival firms to
    learn from search and action.

    Industry Concentration

    Industry concentration, commonly measured by
    the percentage of the market share held by the
    largest firms in an industry (the Herfindahl index),
    is an important industry characteristic. As Wald-
    man and Jensen noted, “Seller concentration
    within a particular market is regarded as a signifi-
    cant aspect of market structure because of its hy-
    pothesized relationship to market power and, ulti-
    mately, to behavior and performance” (2001: 94).
    Theory from economics suggests that a small num-
    ber of dominant firms in an industry will recognize
    their mutual dependence and tacitly coordinate
    search and action in an effort to limit competition

    and rivalry (Scherer & Ross, 1990). The implicit
    motive for this coordination is that escalation of
    competition increases costs and hurts performance.
    Conversely, as the number of firms increases,
    search, action, and potential learning increase as it
    becomes more and more difficult for this coordina-
    tion to occur (Williamson, 1965). Under these con-
    ditions, it is more challenging for a focal firm to
    find unique opportunities to act and, as a conse-
    quence, effective search, action, and learning be-
    come more costly. One implication of this argu-
    ment is that the effects of focal firm search, action,
    and learning on performance are greater in more
    concentrated industries. In such environments,
    search and action are less frequent, and so it is
    easier for firms to learn and to comprehend the
    consequences of their actions as they receive more
    attention from market participants. Moreover,
    when a market consists of just a few large firms,
    customers have limited choices (fewer competi-
    tors), and they are therefore more likely to be at-
    tracted to the new actions of dominant firms.

    For the same reasons that a firm’s performance
    gains from action are likely to be greater in concen-
    trated industries, because of the high market shares
    of firms, limited choices of customers, and ease of
    learning from the effectiveness of search and ac-
    tion, the effects of rivals’ actions on the focal firm’s
    performance are also greater. Firm actions in highly
    concentrated industries are much more likely to
    garner the attention of competitors because mutual
    awareness is very high (Bain, 1951). Thus, in highly
    concentrated industries, rival firms are more apt to
    learn of the actions of a focal firm and to respond to
    those actions to stave off the negative consequences
    of nonresponse. In addition, and perhaps more im-
    portantly, in highly concentrated industries, rival
    firms are more inclined to respond, and to respond
    quickly, to teach their competitors that breaking the

    FIGURE 1
    Hypothesized Relationships

    2008 65Derfus, Maggitti, Grimm, and Smith

    unwritten covenant of tacit collusion will be pun-
    ished severely. On the other hand, in less concen-
    trated industries, where there are more competitors
    to keep track of, actions are less likely to provoke
    responses because rivals will not be aware of
    the initial behavior. As Scherer and Ross stated,
    “As the number of sellers increases and the share
    of industry output supplied by a representative
    firm decreases, individual producers are increas-
    ingly apt to ignore the effect of their price and
    output decisions on rival reactions” (1990: 277).
    Thus, in concentrated industries, focal firm actions
    will have a greater impact on performance, evoke
    a larger number and higher speed of rival ac-
    tions, and those rival actions and their speed will
    have a greater impact on the focal firm’s per-
    formance.

    From an evolutionary perspective, Barnett and
    Hansen (1996) contended that the Red Queen effect
    would be less constraining when a firm faced a
    relatively small number of competitively different
    cohorts, as in the condition of high concentration.
    These authors described how increases in the num-
    ber of competitive relationships (i.e., lower concen-
    tration) constrained effective learning and adapta-
    tion. They noted that each constraint lowers the
    likelihood that a focal firm can carry out effective
    search, action, and learning, arguing that costs of
    search and action come to outweigh the benefits.
    Using a sample of Illinois banks, they found that
    failure rates increased with the number of compet-
    itive relationships. These findings suggest that de-
    creased concentration increases the number of
    competitive relationships to manage and that under
    low concentration, a firm’s action has less impact
    on both its own performance and rival action,
    while rival action has less impact on focal firm
    performance. Carroll and Hannan’s (1989) density
    dependence theory also supports this prediction.
    Specifically, this research showed that firms
    founded under conditions of many competitors are
    less likely to survive in the long term because of
    scarcity of resources (Carroll & Hannan, 2000).

    Given the above arguments, we propose:

    Hypothesis 2a. Industry concentration posi-
    tively moderates the relationship between a
    focal firm’s actions and its performance.

    Hypothesis 2b. Industry concentration posi-
    tively moderates the relationship between a
    focal firm’s actions and rival actions and the
    speed of rival actions.

    Hypothesis 2c. Industry concentration nega-
    tively moderates the relationship between rival

    actions, the speed of rival actions, and a focal
    firm’s performance.

    Industry Demand

    The growth rate of industry demand should also
    have an impact on Red Queen competition. Studies
    by Caves (1980) and Bothwell, Cooley, and Hall
    (1984) showed that firms in high-growth industries
    are less concerned about competing with rivals be-
    cause they are able to enhance revenues simply by
    maintaining their shares of the steadily increasing
    demand. Therefore, high industry growth leads to a
    “live-and-let-live” attitude among firms (Bradburd
    & Caves, 1982; Liebowitz, 1982). A growing market
    facilitates existing routines, and each firm can in-
    crease its share of the pie by searching for and
    carrying out actions that it knows will work with-
    out affecting rivals. Conversely, a decline in indus-
    try demand will prompt firms to search for new
    ways of generating demand, by instituting a new
    price cut or new marketing campaign, that initiate
    or escalate warfare (Caves, 1980).

    Research exploring the evolution of industries
    has examined the differing effects that the early,
    high-demand, stage and the mature, decreasing de-
    mand, stage of the industry life cycle have on com-
    petition among firms (Agarwal & Gort, 1996; Agar-
    wal, Sarkar, & Echambadi, 2002; Carroll & Hannan,
    1989). Specifically this work speculates that during
    periods of high demand growth firms take actions
    that help create that demand and thereby, benefit
    all the firms in their industry (Agarwal & Bayus,
    2002). Carroll and Hannon (1989) described such
    early-stage actions as “legitimizing actions,” as op-
    posed to later-stage “competitive” actions.

    In regard to Red Queen competition, high-growth
    environments provide fertile ground for searching
    and learning about new opportunities to act, and
    limit the negative effect such actions have on com-
    petitors. Increasing industry growth mitigates the
    Red Queen argument that a firm’s performance
    gains come at the expense of other firms (Barnett &
    Hansen, 1996). For example, irrespective of rival
    actions, a focal firm’s successful new-product
    launch is going to be even more successful when
    the number of consumers seeking such products is
    growing. Accordingly, all the firms in a high-
    growth industry will be focused on developing suc-
    cessful “initial” actions and less focused on re-
    sponding more often or faster to other firms’
    actions. Further, when rivals do act, their actions
    are more likely to increase industry growth overall
    rather than have a deleterious impact on another
    firm’s performance. Therefore, we propose:

    66 FebruaryAcademy of Management Journal

    Hypothesis 3a. Industry demand positively
    moderates the relationship between a focal
    firm’s actions and its performance.

    Hypothesis 3b. Industry demand negatively
    moderates the relationship between a focal
    firm’s actions and rival actions and the speed
    of rival actions.

    Hypothesis 3c. Industry demand positively
    moderates the relationship between rival ac-
    tions, speed of rival actions, and focal firm
    performance.

    Market Position

    We next theorize about the effect of the relative
    market position of a focal firm. Research on the Red
    Queen effect has suggested that market leaders are
    less affected by Red Queen competition than are
    other firms. For example, Barnett and McKendrick
    (2004) found that market share leaders—in their
    case, large firms—were the most likely to act to
    develop new products in the disk drive industry.
    They also found that, when exposed to compe-
    tition, large firms were less likely to fail. How-
    ever, they also noted that market share leaders
    can become isolated from competition and, when
    this happens, their survival may be threatened as
    they lose their ability to learn from search and
    action.

    Action research also suggests that market leaders
    more effectively search and act than their rivals and
    react more quickly than their rivals (Smith et al.,
    2001). Presumably, market leaders have the re-
    sources to engage in more effective search and ac-
    tion, which facilitates greater learning. In essence,
    this is how they obtain and defend their market
    positions. Ferrier, Smith, and Grimm (1999) found
    that persistent market leaders act more frequently,
    faster, and with more complexity. The actions of
    market leaders should be more positively related to
    performance than the actions of other firms in an
    industry because they have more experience and
    enjoy more efficient search and action routines. In
    a sense, they have more effectively institutional-
    ized the search, action, and learning process. Spe-
    cifically, actions of market leaders are more visible
    to customers and therefore likely to garner more
    customer attention (Smith et al., 1992). Under these
    conditions, managers can more effectively learn
    from their actions. Young et al. (1996) found that
    market leaders benefit from significant scale effects
    of action that smaller firms cannot obtain. For mar-
    ket-leading firms, the cost of search, action, and
    learning can be spread over a larger customer base.

    Although different predictions are possible re-

    garding how rivals will behave with regard to mar-
    ket leaders,3 we believe the more powerful argu-
    ment is that rivals are unlikely to act or act quickly
    against leading firms because of fear of retribution
    (Scherer & Ross, 1990). Specifically, research in
    industrial-organization (IO) economics has investi-
    gated the behavior of dominant firms with regard to
    their rivals on several fronts; it has been found that
    pricing (Gaskins, 1971; Kamien & Schwartz, 1971),
    R&D and patenting (Gilbert & Newberry, 1982),
    product proliferation (Schmalensee, 1976), adver-
    tising (Comanor & Wilson, 1967; Cubbin &
    Domberger, 1988), and capacity increase (Masson &
    Shannan, 1986; Spence, 1977) actions by dominant
    firms deter rival entry. Ferrier and colleagues
    (1999) found that market leaders that engaged in
    more frequent, speedier actions and utilized more
    complex action repertoires deterred the actions of
    challengers. Thus, we expect that rivals will be
    deterred from attacking an industry leader.

    Finally, we predict that the frequency and speed
    of rival actions, if they occur, have less negative
    impact on leaders’ performance than on nonlead-
    ers’ performance. Customers of leaders are more
    likely to remain loyal in the face of rival actions.
    Leader firms have stronger brand reputation and
    customers are less likely to defect because the
    switching costs of moving from a market leader are
    potentially higher (Scherer & Ross, 1990). In sum-
    mary, prior work shows that market leaders are
    more effective than nonleaders, and their actions
    have a stronger impact on performance than the
    actions of nonleaders. Because leaders’ actions are
    more likely to deter, rather than provoke rivals,
    they evoke fewer and slower rival actions than do
    nonleaders’ actions. And, because of customer loy-
    alty and higher switching costs, rivals’ actions and
    their speed do not detract from leader performance
    as much as they do nonleaders’ performance.
    Therefore, we predict:

    Hypothesis 4a. Market position positively mod-
    erates the relationship between a focal firm’s
    actions and its performance.

    Hypothesis 4b. Market position negatively
    moderates the relationship between a focal
    firm’s actions and rival actions and the speed
    of rival actions.

    Hypothesis 4c. Market position positively mod-
    erates the relationship between rival actions,

    3 One possible alternative explanation is that rivals are
    more likely to follow or imitate market leader actions
    because they are perceived as more legitimate.

    2008 67Derfus, Maggitti, Grimm, and Smith

    the speed of rival actions, and a focal firm’s
    performance.

    RESEARCH METHODS

    Sample

    To test the hypotheses, we developed a sample of
    all the major competitors in 11 different industries
    across a broad spectrum of the U.S. economy. One
    requirement for sample inclusion was that firms
    were competing in the same markets so that their
    specific actions and firm performance could be di-
    rectly connected to the competition and perfor-
    mance in these markets. As a result, we focused
    solely on the actions and performance of U.S. firms
    and included only those industries in which 7

    0

    percent or more of industry sales was generated by
    firms that were public, had a distinct single-busi-
    ness entity competing in the specific U.S. market,
    and reported performance relative to that U.S. mar-
    ket. Firms needed to meet these criteria so that we
    could match the actions of their single-business
    entities with their performance in only in
    that market/industry.

    Eleven industries met our criteria, including ap-
    pliance manufacturing, athletic footwear manufac-
    turing, automobile manufacturing, brewing, gen-
    eral retailing, book retailing, lumber and hardware
    retailing, long-distance telephone services, steel
    manufacturing, and grocery retailing. With repre-
    sentation from manufacturing, services, and retail-
    ing, good industry variation was achieved. On
    average, included firms accounted for 87 percent
    of U.S. industry sales in their respective market/
    industry.

    Data Collection: Competitive

    Actions

    Competitive actions are defined as specific and
    observable moves, such as new marketing cam-
    paigns or new-product introductions, initiated by a
    firm to defend or improve its relative competitive
    position (Chen, 1988; Smith et al., 1992; Young et
    al., 1996). Actions that are observable to customers,
    competitors, and other industry watchers are most
    likely to be reported in the business press (Miller &
    Chen, 1994) and thereby are available for identifi-
    cation, data collection, and analysis. We identified
    and coded observable competitive actions by con-
    ducting a structured content analysis (Jauch, Os-
    born, & Martin, 1980) of newspaper and trade mag-
    azine articles found on the Lexus-Nexus article
    index. This index allows electronic searching of
    full-text articles from thousands of newspapers and
    journals. For each of the industries chosen, at least

    one industry trade magazine was searched. Addi-
    tionally, the New York Times and the Wall Street
    Journal were searched for all industries. We iden-
    tified 76,963 article citations via keyword search-
    ing of the Lexus-Nexus database. Coders then con-
    tent-analyzed the full texts of articles that
    potentially contained reports of competitive ac-
    tions. Only the earliest report of an action was
    entered into the database. This procedure resulted
    in a database containing 4,474 actions. To verify
    the accuracy of the coding, two coders reviewed 10
    percent of the article citations for each industry.
    Action identification and action-type coding agree-
    ment were obtained for 99.25 percent of the over
    7,697 citations they read.

    Actions were collected for 58 firms; missing data
    reduced the sample to 56 firms over a six-year
    period, 1993 through 1998. The mean number of
    actions per firm was 12.58. The maximum number
    of actions per firm per year was 51; the minimum,
    0. The most common actions related to pricing, and
    the least common were geographic actions.

    Firm financial data, firm size, and industry con-
    text variables were collected from Standard &
    Poor’s Compustat database, which offers financial
    data on all companies that were publicly traded on
    North American stock markets in those years.
    These data are collected from annual reports, Secu-
    rities and Exchange Commission (SEC) filings, and
    other publicly available documents. Where neces-
    sary, we adjusted financial data to remove contri-
    butions from non-U.S. operations, thereby match-
    ing the financial figures to the actions accounted
    for in the study. These adjustments were possible
    because Compustat offers detailed geographic seg-
    ment data delineating company operations in vari-
    ous countries.

    Measures

    Focal firm total actions and rival total actions.
    Five types of focal firm and rival actions were mea-
    sured: pricing, capacity, geographic, marketing,
    and product introductions. We calculated firm total
    actions or activity by simply summing the number
    of all five actions for a focal firm in a given year. We
    then operationalized rival total actions or compet-
    itive activity by subtracting a focal firm’s total num-
    ber of actions in a given year from the total number
    of actions taken by all competitors in a
    focal industry.

    Rival action speed. As in other competitive dy-
    namics research (e.g., Ferrier et al., 1999; Young et
    al., 1996), rival firm action speed quantified the
    average length of time it took rivals to act after a
    focal firm acted. To calculate this measure, we de-

    68 FebruaryAcademy of Management Journal

    termined the number of days between each firm
    action and the first rival action and then averaged
    those scores for each focal firm for each year. Fi-
    nally, we took the reciprocal of this value to aid in
    interpretation of results. The resulting measure
    equates high rival action speed values to fast rival
    action speed and low rival action speed values to
    slow rival action speed.

    Focal firm performance. Focal firm perfor-
    mance was operationalized with accounting mea-
    sures of return on sales (ROS) and return on assets
    (ROA) in the same year as the action measure.

    Industry conditions. Industry concentration and
    industry demand were used to capture the industry
    context in which firm and rival actions took place.
    These measures served as independent or control
    variables in all regressions and were also interacted
    with firm actions, rival actions, and rival action
    speed in tests of Hypotheses 2a–2c and 3a–3c. In-
    dustry concentration was calculated as the Herfin-
    dahl measure of the market shares of the firms in
    each industry for each year. Industry demand was
    measured as industry growth, defined as the per-
    cent change in sales from the previous year to a
    focal year.

    Relative market position. Relative market posi-
    tion was measured as rank order based on market
    share for each firm in each industry for each year.
    This variable was used as an independent or a
    control variable in all regressions and was also
    interacted with firm actions, rival actions, and rival
    action speed in testing Hypotheses 4a– 4c.

    Control variables. To control for unobserved dif-
    ferences in industry factors that might influence
    market dynamics, we included industry dummies
    in all regressions.

    In this study, we had two basic regression mod-
    els. In the first, we regressed our independent ac-
    tion variables on firm performance. However, since
    we had two measures of firm performance, return
    on assets and return on sales, a model is presented
    for each. In these models, we lagged a focal firm’s
    prior year return on assets or return on sales per-
    formance, respectively, to control for the influence
    the variable might have on performance in the fol-
    lowing year and also to help control for correlated
    error terms of our longitudinal data (Young et al.,
    1996). In these regressions, we also controlled for
    firm characteristics that have been shown to influ-
    ence firm actions and performance, including size
    and slack resources (Smith et al., 1992). Specifi-
    cally, sales measured size, and the quick ratio mea-
    sured slack resources (e.g., Ferrier et al., 1999). The
    logic was that firms with more assets and resources
    are able to undertake more actions (Smith et al.,

    2001). We then repeated these regressions replac-
    ing rival action speed with rival total actions.

    When we tested the effect of focal firm actions on
    rival actions and rival action speed, we controlled
    for prior year aggregate performance utilizing
    lagged rival firm prior year return on sales. This
    lagged composite was calculated as the aggregated
    net income of all rival firms in an industry divided
    by their aggregated sales. As in our regressions on
    firm performance, we also controlled for rival size
    and rival slack resources. Rival’s size was calcu-
    lated as an aggregate average measure of the relative
    size of each firm’s pool of rivals. It is an annual sum
    of the sales of the unique group of rivals pertaining
    to each focal firm. Rival quick ratio is a composite
    average of the quick ratios of those rival firms.

    We followed the practice of prior competitive
    dynamics researchers by investigating the impact
    of actions on the same year’s performance (e.g.,
    Ferrier et al., 1999; Young et al., 1996). During the
    years of our study, for these 11 industries, the av-
    erage number of days between a focal firm action
    and a rival action is 12.3 (s.d. � 19.3) and the
    maximum number of days between actions is 149.
    This relatively short time frame provided further
    support for our same-year analysis of actions and
    performance.

    In calculating the measures in this study, it was
    important to precisely define the focal and rival
    firms. For example, in the U.S. brewing industry
    there were three sampled firms: Anheuser-Busch,
    Miller Brewing, and Adolph Coors. When An-
    heuser-Busch was the focal firm, Miller Brewing
    and Adolph Coors were the rival firms. Similarly,
    when Miller Brewing was focal, Anheuser-Busch
    and Adolph Coors were the rivals. We calculated
    firm total actions and performance measures for
    each firm individually and the rival total actions
    measure for each firm’s unique set of rivals. Since
    the unit of observation was the firm-year, changes
    in the rival set from year to year must be accounted
    for. Therefore, the rival set was officially defined as
    all the other firms competing in the focal firm’s
    industry for the year under consideration.

    RESULTS

    Table 1 reports the means and correlations
    among all variables in this study. We tested hy-
    potheses with random-effects regression models to
    ensure that error due to serial correlation in our
    panel data set was specified and analyzed (Erez,
    Bloom, & Wells, 1996). In addition, we used nega-
    tive binomial regression analysis for regression
    models in which the dependent variable was rival
    total actions, a count-type variable. This type of

    2008 69Derfus, Maggitti, Grimm, and Smith

    T
    A

    B
    L

    E
    1

    D
    es

    cr
    ip

    ti
    v
    e

    S
    ta

    ti
    st

    ic
    sa

    V
    a
    ri

    a
    b
    le

    b
    M

    e

    a
    n

    s.
    d

    .
    1

    2

    3

    4
    5

    6

    7

    8

    9

    1
    0

    1
    1

    1
    2

    1
    3

    1
    4

    1
    5

    1
    6

    1
    7

    1
    8

    1
    9

    2
    0

    2
    1

    2
    2

    2
    3

    2
    4

    2
    5

    2
    6

    2
    7

    2
    8

    2
    9

    3
    0

    3
    1

    3
    2

    3
    3

    1
    .

    R
    O

    A
    3
    .7

    4

    5
    .5

    1
    2
    .

    R
    O

    S
    4
    .5

    3

    5
    .3

    3
    .4

    5
    3
    .

    F
    ir

    m
    to

    ta
    l

    ac
    ti

    o
    n

    s
    1

    3
    .1

    8
    1

    0
    .6

    5
    .2

    2
    .1

    4
    4
    .

    R
    iv

    al
    to

    ta
    l
    ac
    ti
    o
    n

    s
    6

    3
    .5

    7
    4
    4
    .6

    9

    .1
    3


    .3

    9
    .3

    1

    5
    .

    R
    iv

    al
    ac

    ti
    o
    n

    sp
    ee

    d

    0
    .1

    8
    0
    .1

    4

    .1
    1


    .3

    2
    .4

    6
    .9

    5
    6
    .

    F
    ir

    m
    m

    ar

    k
    et

    sh
    ar

    e
    ra

    n
    k

    0
    .3

    8

    0
    .2

    7
    .0

    4

    .0
    7

    .1
    5

    .3
    4

    .3
    2

    7
    .

    I

    n
    d

    u
    st

    ry
    H

    er
    fi

    n
    d

    ah
    l

    0
    .2

    2
    0
    .1

    8
    .0

    8
    .3

    8
    .0
    9

    .5
    4


    .4

    9

    .5
    5

    8
    .

    In
    d

    u
    st

    ry
    gr

    o
    w

    th

    0
    .0

    9
    0

    .1

    1
    .0

    3

    .1
    5


    .1

    5

    .1
    9


    .2

    2

    .0
    8

    .1
    4

    9
    .

    M
    ar

    k
    et
    sh
    ar
    e
    ra
    n
    k


    fi

    rm
    ac

    ti
    o
    n

    s
    5
    .5

    5

    6
    .1

    9
    .1

    9
    .0

    1
    .7

    2
    .3

    5
    .4

    2
    .6

    7

    .2
    8


    .0

    8

    1
    0
    .

    M
    ar
    k
    et
    sh
    ar
    e
    ra
    n
    k


    ri

    v
    al

    ac
    ti
    o
    n

    s
    2

    8
    .6

    6
    3

    2
    .2

    0

    .0
    7

    .2
    5
    .2
    0
    .6

    9
    .6

    3
    .8

    5

    .6
    2


    .0
    9
    .6
    4

    1
    1
    .

    H
    er

    fi
    n

    d
    ah

    l

    fi
    rm

    ac
    ti
    o
    n

    s
    3
    .4

    7

    4
    .4

    2
    .1

    6
    .4

    0
    .7

    0

    .1
    5

    .0

    1

    .2
    1

    .6
    2

    .1

    1
    .2

    1

    .2
    6

    1
    2
    .
    H
    er
    fi
    n
    d
    ah
    l

    ri
    v
    al

    ac
    ti
    o
    n

    s
    1

    1
    .9

    4

    1
    0
    .0

    1

    .0
    7

    .

    0
    2

    .3
    8

    .3
    5

    .4
    1


    .3
    4
    .4

    6

    .1
    3

    .0
    4

    .1
    3

    .4
    6

    1
    3
    .

    In
    d
    u
    st
    ry
    gr
    o
    w

    th

    fi
    rm
    ac
    ti
    o
    n

    s
    1
    .0

    9
    1
    .5

    5
    .2
    0

    .0
    6

    .3
    8
    .0
    8
    .1
    3

    .1
    2

    .0
    0

    .5
    4

    .3
    6

    .1
    1
    .1
    3
    .0
    0

    1
    4
    .

    In
    d
    u
    st
    ry
    gr
    o
    w
    th

    ri
    v
    al
    ac
    ti
    o
    n

    s
    5
    .2

    2
    5
    .7

    9

    .0
    8

    .3
    3
    .1

    1
    .6

    0
    .5

    3
    .2

    6

    .3
    2

    .4
    2

    .2
    3

    .5
    0


    .2
    0
    .1
    3
    .5
    3
    1
    5
    .
    M
    ar
    k
    et
    sh
    ar
    e
    ra
    n
    k

    ri
    v
    al
    sp
    ee

    d
    0
    .0

    8
    0
    .0

    9

    .0
    6

    .2
    2
    .2
    8
    .6

    7
    .6

    6
    .8

    7

    .6
    1


    .1

    2
    .7

    1
    .9

    8

    .2
    1

    .1
    0
    .1
    3
    .4
    6

    1
    6
    .

    H
    er
    fi
    n
    d
    ah
    l

    ri
    v
    al
    sp
    ee
    d
    0
    .0

    4
    0
    .0

    3

    .0
    5

    .1
    0

    .5
    6

    .3
    4

    .4
    8


    .2

    8
    .4

    1

    .1
    6

    .0
    7

    .1
    1
    .6

    3
    .9

    4
    .0

    6
    .1
    1

    .0
    5

    1
    7
    .

    In
    d
    u
    st
    ry
    gr
    o
    w
    th

    ri
    v
    al
    sp
    ee
    d
    0
    .0
    1
    0
    .0
    2

    .0
    6

    .3
    0
    .2

    2
    .5

    9
    .5

    8
    .2

    4

    .2
    9

    .4
    0

    .2
    8
    .4
    6

    .1
    2
    .1
    8
    .6

    4
    .9

    7
    .4

    5
    .2
    0

    1
    8
    .

    L
    ag

    ge
    d

    R
    O

    A
    0
    .0

    4
    0
    .0

    6
    .5

    2
    .2

    2
    .0

    6

    .0
    5

    .0
    7
    .0
    0
    .0
    3

    .0
    2

    .0
    7

    .0
    6

    .0
    2

    .0
    1
    .0
    8

    .0
    9


    .0
    7

    .0
    5

    .0

    9
    1
    9
    .

    L
    ag
    ge
    d
    R
    O

    S
    0
    .0

    4
    0
    .0
    5
    .3
    1
    .6
    4
    .0
    1

    .3
    3


    .2
    9

    .0
    9
    .2
    6

    .1
    7

    .0
    7

    .2

    4
    .1

    8
    .0
    1

    .0
    9

    .3
    5

    .2
    2

    .0
    4


    .3
    5
    .5

    4
    2
    0
    .

    L
    ag
    ge
    d
    ri
    v
    al
    R
    O
    S
    0
    .0

    5
    0
    .0

    4
    .0
    5

    .1
    6

    .0
    5

    .0
    4

    .1
    5

    .0
    8
    .0
    5
    .0
    9

    .0
    4

    .0
    8

    .0
    6

    .0
    6

    .0
    1

    .0
    9
    .1
    3

    .1
    5

    .0
    9
    .1
    3

    .0
    3

    2
    1
    .

    F
    ir

    m
    sa

    le
    s

    1
    5
    ,6

    6
    3

    2
    5
    ,1

    9
    0
    .1
    5
    .1
    0
    .4
    6
    .0
    5
    .1
    4
    .0
    2
    .1
    1

    .1
    1
    .2
    8
    .0
    0
    .3
    9
    .1
    6
    .1
    4

    .0
    4
    .0
    5
    .2
    9

    .0
    4
    .0

    6
    .0

    3

    .1
    0

    2
    2
    .

    F
    ir

    m
    q
    u

    ic
    k

    ra
    ti

    o
    0
    .6

    0

    0
    .4

    3
    .1

    1
    .3

    0

    .0
    7

    .1
    4

    .0
    8

    .0
    6
    .1
    0

    .1
    2

    .1
    5

    .1
    1
    .1
    4
    .0
    9

    .0
    6

    .1
    0

    .0
    9
    .1
    1

    .1
    0
    .1
    1
    .2
    1

    .2
    5

    .1
    6

    2
    3
    .

    R
    iv

    al
    sa

    le
    s

    8
    2
    ,2

    0
    2

    9
    1
    ,6

    3
    0

    .0
    9

    .2
    2

    .3
    1

    .6
    1

    .6
    6


    .0
    5

    .2
    6

    .2
    0
    .1

    4
    .2

    0

    .0
    4

    .3
    9

    .0
    6

    .2
    4

    .2
    2

    .4
    4

    .2
    4

    .0
    8

    .1
    9

    .0
    3
    .4
    3

    .2
    2

    2
    4
    .

    In
    d
    u
    st

    ry
    q
    u

    ic
    k
    ra
    ti
    o
    0
    .6

    1
    0
    .3

    1
    .0

    3
    .3

    2

    .0
    6

    .1
    5

    .0
    8
    .0
    2
    .0
    8

    .1
    9

    .1
    0

    .0
    6
    .1

    5
    .1

    0

    .1
    2

    .1
    2

    .0
    3
    .1
    2

    .1
    2
    .0
    0
    .2
    1

    .3
    4

    .1

    4
    .7

    4
    .2

    9
    2
    5
    .

    In
    d
    u
    st

    ry
    1

    0
    .0

    6
    0
    .2

    4
    .0

    9
    .2

    7
    .0
    0

    .2
    3

    .2
    0

    .1
    6
    .2
    9

    .1
    8

    .1
    0

    .1

    8
    .1

    5

    .0
    3

    .1
    2

    .1
    9

    .1
    8
    .0
    0

    .1
    9
    .0
    6
    .2
    1

    .2
    0

    .0
    7

    .0
    1

    .1
    8


    .0

    1
    2
    6
    .

    In
    d
    u
    st

    ry
    2

    0
    .0

    7
    0
    .2

    6
    .2

    8
    .0

    7
    .1

    4

    .0
    6

    .0
    8

    .1
    2
    .2
    4
    .0
    3
    .0
    0

    .1
    3
    .2

    4
    .3

    1
    .1

    7
    .0
    1

    .1
    4
    .2
    2

    .0
    1

    .2
    6
    .1
    1
    .1
    2

    .1
    5
    .3
    6

    .2
    3

    .4
    5


    .0

    6
    2
    7
    .

    In
    d
    u
    st

    ry
    3

    0
    .1

    9
    0
    .3

    9

    .2
    0


    .1
    5

    .2
    1
    .2
    1
    .2
    1

    .5
    8


    .5

    3

    .1
    4
    .1
    4

    .4
    7


    .3
    2

    .3
    5

    .1
    0
    .1
    7
    .4
    9

    .3
    1

    .1
    7


    .1
    4

    .1
    2

    .2
    9

    .2
    6
    .2
    8

    .2
    2
    .4
    1

    .1
    2

    .1

    4
    2
    8
    .

    In
    d
    u
    st

    ry
    4

    0
    .0
    6
    0
    .2
    4

    .0
    4
    .2
    3

    .2
    4

    .3
    2

    .2
    9

    .1
    9

    .0
    8

    .1
    3

    .2
    0

    .2
    1

    .1
    5

    .2
    5


    .1
    5

    .2
    1

    .2
    2

    .2
    2

    .2
    1

    .0
    6
    .1
    7

    .2
    0

    .1
    1
    .2
    2

    .1
    9
    .3
    1

    .0
    6

    .0
    7
    .1

    3
    2
    9
    .

    In
    d
    u
    st

    ry
    5

    0
    .1

    2
    0
    .3

    2

    .1
    3

    .1
    2
    .3
    0
    .4
    2
    .5
    8

    .1
    5

    .0
    1

    .0
    5
    .0
    3
    .0
    1
    .1
    5

    .5
    9

    .1
    6
    .2
    6
    .0
    7

    .6
    8

    .2
    6

    .1
    1
    .1
    2

    .3
    0

    .2
    9
    .0
    0

    .7
    2

    .0
    0

    .0
    7

    .0
    8

    .1
    5

    .0
    7

    3
    0
    .

    In
    d
    u
    st

    ry
    6

    0
    .0
    6
    0
    .2
    3

    .0
    1
    .4
    0
    .1
    8

    .1
    7

    .1
    3

    .1
    9
    .6
    2

    .1
    3

    .0
    5

    .1
    9

    .5
    3

    .3
    5

    .1
    6

    .2
    4

    .1
    8
    .3
    5

    .2
    4

    .0
    6
    .2

    5
    .0

    3
    .1
    8
    .1
    2

    .0
    9
    .1
    6

    .0
    6

    .0
    7

    .1
    4

    .0
    7

    .0

    8
    3
    1
    .

    In
    d
    u
    st

    ry
    7

    0
    .0

    8
    0
    .2

    8
    .0
    3

    .0
    2

    .1
    4

    .2
    4

    .2
    6

    .0
    1
    .0
    7
    .1
    6

    .0
    9

    .1
    8

    .0
    6

    .1
    .0

    4

    .0
    3

    .1
    9

    .1
    2

    .0

    3
    .0

    3

    .0
    1
    .2
    9

    .0
    8

    .2
    2

    .1
    9

    .3
    5

    .0
    7

    .0
    8

    .1
    6

    .0
    8

    .0

    9
    3
    2
    .

    In
    d
    u
    st

    ry
    8

    0
    .0
    8
    0
    .2
    7
    .0
    2

    .0
    3
    .0
    1

    .1
    7

    .1
    8

    .1
    5
    .0
    2

    .0
    5

    .0
    7

    .1
    8
    .0
    1

    .0
    4
    .0
    0

    .1
    1

    .1
    9

    .0
    6

    .1
    1

    .0
    1

    .0
    3

    .0
    2
    .5
    4

    .0
    6
    .5
    1

    .0
    8

    .0
    7

    .0
    8

    .1
    5

    .0
    7

    .0
    8

    .0
    8

    .0

    9
    3
    3
    .

    In
    d
    u
    st

    ry
    9

    0
    .1

    5
    0
    .3

    6
    .1
    3

    .2
    1
    .2
    1
    .5
    6
    .4
    4
    .2
    2

    .4
    3

    .1
    4
    .3
    3
    .4
    5

    .2
    1

    .1
    4
    .0
    0
    .1
    7
    .4
    0

    .1
    6
    .1
    7
    .1
    2

    .1
    7
    .3
    5
    .0
    2

    .4
    1
    .3
    5

    .5
    2


    .1
    1

    .1
    2

    .2
    4

    .1
    1

    .1
    3

    .1
    3

    .1
    4

    .1
    3

    3
    4
    .

    In
    d
    u
    st

    ry
    1
    0

    0
    .0
    7
    0
    .2
    6

    .1
    1

    .1
    3

    .0
    5

    .1
    5

    .1
    8

    .1
    5
    .2
    2
    .1
    4

    .0
    8

    .1
    7
    .0
    6
    .1
    1
    .0
    8
    .0
    3

    .1
    8
    .0
    2
    .0
    3

    .0
    7

    .0
    4
    .0
    8

    .1
    6

    .1
    4

    .2
    2

    .2
    0

    .0
    6

    .0
    7

    .1
    3

    .0
    6

    .0
    7

    .0
    7

    .0
    8

    .0
    7

    .1
    2
    a
    n


    2
    8
    1
    .

    C
    o
    rr

    el
    at

    io
    n

    s
    ab

    o
    v
    e

    .1
    1

    ar
    e

    si
    gn

    if
    ic

    an
    t

    at
    p


    .0
    5
    .

    b
    R

    O
    A

    an
    d

    R
    O

    S
    ar

    e
    p

    er
    ce

    n
    ta

    ge
    s.

    S
    al

    es
    ar

    e
    in

    m
    il

    li
    o
    n

    s
    o
    f

    d
    o
    ll

    ar
    s.

    regression was used in these models for two rea-
    sons. First, these count data are not normally dis-
    tributed, violating a key assumption of generalized
    least squares (GLS) regression analysis (Greene,
    1993). Secondly, as is typically the case, our count
    data are overdispersed, meaning the variance of the
    event counts exceeds their means (Cameron & Tra-
    vendi, 1986). A likelihood-ratio test of overdisper-
    sion also indicated that negative binomial regres-
    sion was an appropriate choice. Negative binomial
    regression overcomes distribution problems and es-
    timates an additional parameter that corrects for
    overdispersion (Frome, Kutner, & Beauchamp,
    1973). Tables 2 and 3 report the regression results.

    Table 2 reports the results for regressions relating
    firm actions, firm performance, rival actions, and
    speed of rival actions. Table 3 reports the regres-
    sion results that examine the impact of industry
    environment and market leadership on the rela-
    tionships between firm actions, firm performance,
    rival actions, and speed of rival actions.

    Hypothesis 1a states that as firm total action in-
    creases, firm performance increases. This hypothe-
    sis is fully supported. As seen in models 1 and 3 in
    Table 2, firm actions have a positive, significant
    coefficient for both return on assets (� � 0.08, p �
    .01) and return on sales (� � 0.08, p � .01). These
    results are repeated in models 2 and 4 of Table 2, in
    which rival action speed replaces rival actions in
    the models. That is, firm actions again have a pos-

    itive, significant effect on both return on assets (� �
    0.12, p � .01) and return on sales (� � 0.11, p �
    .01).

    Hypothesis 1b states that as firm actions increase,
    rival actions and rival action speed also increase.
    This hypothesis is supported. Specifically, models
    5 and 6 in Table 2 report a significant, positive
    coefficient for the relationship between firm ac-
    tions and rival actions (� � 0.01, p � .05) and firm
    actions and rival speed (� � 0.00, p � .01),
    respectively.

    Hypothesis 1c states that, as rival actions and
    rival action speed increase, focal firm performance
    decreases. This hypothesis is also fully supported.
    As reported in Table 2, rival actions is significantly
    and negatively related to focal firm ROA in model 1
    (� � �0.05, p � .01) and ROS in model 3 (� �
    �0.03, p � .01). Similarly, as shown in models 2
    and 4, rival action speed is significantly and nega-
    tively related to focal firm’s ROA in model 1 (� �
    �17.02, p � .01) and ROS in model 2 (� � �12.93,
    p � .01).

    Hypotheses 2a–2c, 3a–3c, and 4a– 4c explore the
    boundary conditions of the first hypothesis set, Hy-
    potheses 1a–1c. Specifically, in the second and
    third sets of hypotheses we examine how various
    industry situations condition the relationship be-
    tween firm actions, rival actions, rival action speed,
    and firm performance. In the fourth set of hypoth-
    eses, we look at how these relationships differ on

    TABLE 2
    Results of Random-Effects Regression Analyses of the Main Relationshipsa

    Variables
    Model 1:

    ROA
    Model 2:

    ROA
    Model 3:

    ROS

    Model 4:

    ROS

    Model 5:
    Rival Total

    Actions

    Model 6: Rival
    Speed of
    Actions

    Lagged ROA 35.74** (4.81) 35.73** (4.82)
    Lagged ROS 38.78** (4.34) 39.93** (4.31)
    Lagged rival ROS �2.14* (1.03) �0.30** (0.10)
    Firm sales 0.29† (0.21) 0.30† (0.24) �0.14 (0.17) �0.14 (0.17)
    Quick ratio 1.84* (0.75) 1.89* (0.75) 2.03** (0.60) 2.05** (0.60)
    Rival sales �0.00* (0.00) �0.00* (0.00)
    Industry quick ratio 0.22 (0.16) 0.03 (0.02)
    Market share rank 0.93 (1.27) 0.84 (1.27) 2.95** (1.04) 2.81** (1.03) �0.13* (0.18) �0.04** (0.01)
    Herfindahl index �5.52 (5.57) �5.78 (5.59) �1.19 (4.61) �1.17 (4.63) �0.34 (0.31) �0.12* (0.07)
    Industry growth 6.12* (3.11) 5.06* (3.06) 2.83 (2.51) 2.29 (2.49) 1.19** (0.29) 0.06* (0.04)
    Firm total actions 0.08** (0.03) 0.12** (0.04) 0.08** (0.03) 0.11** (0.03) 0.01* (0.00) 0.00** (0.00)
    Rival total actions �0.05** (0.01) �0.03** (0.01)
    Rival speed of actions �17.02** (5.21) �12.93** (4.23)

    Constant 1.66 (2.54) 1.99 (2.55) 0.58 (2.13) 0.66 (2.14) 1.97** (0.58) 0.05 (0.04)
    Wald chi-square 187.20** 185.18** 394.58** 406.50** 248.93** 2,117.46**

    a Standard errors are in parentheses. Industry dummy variables were included in all regression models. Results are available upon
    request. n � 281.

    † p � .10
    * p � .05

    ** p � .01

    2008 71Derfus, Maggitti, Grimm, and Smith

    the basis of the market share of a focal firm relative
    to the other firms in the industry. Table 3 reports
    the results of our tests of Hypotheses 2a–2c, 3a–3c,
    and 4a– 4c.

    Hypothesis 2a predicts industry concentration
    positively moderates the relationship between fo-
    cal firm actions and focal firm performance. This
    hypothesis is partially supported. Specifically, and
    as predicted, the effect of the interaction of indus-
    try concentration on the relationship between focal
    firm actions and firm performance is positive and
    significantly related to firm ROS in models 3 and 4
    of Table 3 (� � 0.59, p � .01; � � 0.50, p � .01).
    There was no significant finding with respect to
    focal firm ROA. This result suggests that the posi-

    tive effect of focal firm actions on performance, in
    terms of return on sales, is higher in concentrated
    industries than in nonconcentrated industries.

    In Hypothesis 2b, we predict that industry con-
    centration positively moderates the relationship
    between focal firm actions and both rival actions
    and speed of rival action. Although there were no
    significant findings with respect to rival actions,
    the results shown in model 6 of Table 3 run counter
    to this hypothesis for speed of rival action (� �
    �0.01, p � .01). That is, in concentrated industries
    the relationship between firm actions and speed of
    rival action is weaker than it is in less concentrated
    industries. No significant findings were found to
    support or refute our Hypothesis 2c, that industry

    TABLE 3
    Results of Random-Effects Regression Analyses of Interactionsa

    Variables
    Model 1:
    ROA
    Model 2:
    ROA
    Model 3:

    ROS
    Model 4:

    ROS
    Model 5: Rival
    Total Actions

    Model 6: Rival
    Speed of
    Actions

    Lagged ROA 34.48** (5.04) 34.12** (5.09)
    Lagged ROS 42.02** (4.30) 42.23** (4.34)
    Lagged rival ROS �2.18** (0.65) �0.30** (0.10)
    Sales 0.24 (0.23) 0.25 (0.23) �0.31* (0.18) �0.33* (0.18)
    Rival sales �0.00 (0.00) �0.00 (0.00)
    Quick ratio 1.89* (0.76) 1.91* (0.77) 1.83** (0.59) 1.85** (0.59)
    Industry quick ratio 0.06 (0.17) 0.02 (0.02)
    Market share rank 5.40* (3.01) 5.53* (3.09) 5.78* (2.35) 5.75* (2.41) �0.42* (0.23) 0.01 (0.02)
    Herfindahl index �4.89 (6.08) �5.42 (5.90) �5.11 (4.84) �3.90 (4.71) 1.53* (0.88) �0.05 (0.07)
    Industry growth 0.36 (4.91) 1.20 (4.63) 2.34 (3.84) 2.97 (3.62) 3.61** (0.46) 0.14** (0.05)
    Firm total actions 0.03 (0.11) 0.04 (0.12) �0.18* (0.09) �0.14† (0.09) 0.02** (0.01) 0.01** (0.00)
    Rival total actions �0.02 (0.04) �0.04† (0.03)
    Rival speed of actions �0.06 (13.65) �11.49 (10.65)
    Herfindahl � firm total

    actions
    0.05 (0.18) 0.08 (0.18) 0.59** (0.14) 0.50** (0.14) 0.01 (0.02) �0.01** (0.00)

    Herfindahl � rival total
    actions

    �0.04 (0.09) 0.10† (0.07)

    Herfindahl � rival speed of
    actions

    �28.20 (36.17) 19.54 (28.36)

    Industry growth � firm total
    actions

    0.11 (0.28) 0.15 (0.31) �0.11 (0.22) �0.17 (0.25) �0.14** (0.02) �0.01** (0.00)

    Industry growth � rival total
    actions

    0.11 (0.10) 0.15* (0.08)

    Industry growth � rival speed
    of actions

    13.43 (34.07) 34.13 (26.78)

    Market share rank � firm
    total actions

    0.07 (0.14) 0.11 (0.15) 0.20* (0.11) 0.23* (0.12) �0.01* (0.01) �0.00** (0.00)

    Market share rank � rival
    total actions

    �0.06* (0.03) �0.05* (0.03)

    Market share rank � rival
    speed of actions

    �24.84* (13.5) �19.82* (10.44)

    Constant 2.24 (3.01) 2.14 (3.00) 1.21 (2.40) 0.81 (2.39) �0.50 (0.68) �0.00 (0.047)
    Wald chi-square 193.19** 189.51** 459.54** 453.06** 1,244.63** 2,242.03**

    a Standard errors are in parentheses. Industry dummy variables were included in all regression models. Results are available upon
    request. n � 281.
    † p � .10
    * p � .05
    ** p � .01

    72 FebruaryAcademy of Management Journal

    concentration negatively moderates the relation-
    ship between rival firm actions and focal firm
    performance.

    Although we found no support for Hypothesis
    3a, predicting that industry demand conditions
    positively moderate the relationship between focal
    firm actions and focal firm performance, our Hy-
    pothesis 3b, in which we predict that industry de-
    mand negatively moderates the relationship be-
    tween focal firm actions and rival actions and rival
    action speed, was supported. That is, models 5 and
    6 of Table 3 indicate that the interaction between
    focal firm actions and industry growth was nega-
    tive and significantly related to both rival actions
    (� � �0.14, p � .01) and the speed of rival actions
    (� � �0.01, p � .01). Thus, and as predicted, as
    industry demand increases, the effect of firm ac-
    tions on rival actions and their speed declines.

    Hypothesis 3c predicts that industry demand
    positively moderates the relationship between fo-
    cal firm performance and both rival actions and the
    speed of rival actions. We found some support for
    this hypothesis, as shown in model 3 of Table 3.
    Specifically, the significant and positive effect of
    the interaction between rival actions and industry
    growth on focal firm return on sales (� � 0.15, p �
    .05) is consistent with our hypothesis.

    The influence that focal firm market position has
    on rival actions and firm performance was explored
    in Hypotheses 4a– 4c. In Hypothesis 4a, we predict
    that market position positively moderates the rela-
    tionship between focal firm actions and focal firm
    performance. This hypothesis was partially sup-
    ported, as indicated in models 3 and 4 of Table 3, in
    which the interaction of market position and firm
    actions is positive and significantly related to re-
    turn on sales (� � 0.20, p � .05; � � 0.23, p � .05).
    This result suggests that the positive impact of a
    firm’s actions on performance is greater for firms
    that have higher market shares in an industry.

    Hypothesis 4b predicts that market position neg-
    atively moderates the relationship between focal
    firm actions and both rival actions and their speed.
    This hypothesis is supported. That is, the interac-
    tion of firm actions with higher market position is
    negatively and significantly related to both rival
    firm actions in model 5 of Table 3 (� � �0.01, p �
    .05) and the speed of rival actions in model 6 of the
    same table (� � �0.004, p � .01). Thus, the actions
    of firms with higher market shares than their rivals
    tend to not increase rival actions and rival action
    speed as much as do the actions of firms with lower
    market shares.

    Hypothesis 4c predicts market position posi-
    tively moderates the relationship between both ri-
    val firm actions and rival firm action speed and

    focal firm performance. Thus, we expected that
    rival actions and rival action speed would not af-
    fect market leaders in the same way as they would
    affect the performance of nonleaders. Results did
    not support this hypothesis and, in fact, were con-
    trary to our expectation. Specifically, Table 3
    shows that the interaction between market position
    and rival actions is negatively and significantly
    related to both focal firm return on assets in models
    1 and 2 (� � �0.06, p � .05; � � �24.84, p � .05)
    and focal firm return on sales in models 3 and 4
    (� � �0.05, p � .05; � � �19.82, p � .05). Thus,
    rival actions have a greater negative impact on
    firms with larger shares of an industry market than
    on firms with smaller market shares.

    DISCUSSION

    This study has shed light on Red Queen compe-
    tition by investigating the relationships between
    focal firm actions, rival firm actions, and focal firm
    performance in a variety of industries. In line with
    Red Queen theory, all the relationships in our base-
    line model were supported and showed that even
    though a focal firm’s actions do increase its perfor-
    mance, they also increase the number and speed of
    rivals’ actions which, at least partially, negatively
    impact the focal firms’ performance. Indeed, to
    paraphrase the Red Queen in Lewis Carroll’s
    Through the Looking Glass (1960), it is true that the
    firms studied here have “to run as fast as they can
    to stay in place, and twice as fast as that” to get
    ahead. Although portions of this baseline model
    have been tested elsewhere, we are aware of no
    previous study that has examined both the positive
    and negative effects of actions, as we do in the
    present study. Studying these effects together en-
    ables us to elucidate the positive and negative as-
    pects of action, and to clarify the relative importance
    of these aspects with regard to firm performance.
    Below we offer Figures 2 and 3 to illustrate this.

    Plotting the data from the baseline models we
    used to test Hypotheses 1a–1c, models 1– 4 in Table
    2, we graphically present the Red Queen effect in
    Figure 2. This graph illustrates the counter-balanc-
    ing effects of firm and rival actions on firm perfor-
    mance as measured by return on assets (ROA) and
    return on sales (ROS). Specifically, we see the ac-
    tual average number of firm actions and rival ac-
    tions associated with various levels of perfor-
    mance. As expected, performance gains from action
    are maximized when firm actions are high and rival
    actions are low. Perhaps less expected, the number
    of rival actions necessary to seriously, negatively
    impact firm performance is surprisingly high rela-

    2008 73Derfus, Maggitti, Grimm, and Smith

    tive to the number of firm actions necessary to
    positively increase performance.

    Figure 3 illustrates the strength of the Red Queen
    effect by presenting the net incremental effect that
    firm actions have on firm performance directly, and
    indirectly, through rival actions and rival action
    speed. The line with the steepest positive slope in
    both the ROA and ROS plots represents the direct
    effects of firm actions on performance. For ROA,
    this slope is calculated as the mean of the coeffi-
    cients for firm actions in models 1 and 2 of Table 1
    (0.08 and 0.12, respectively).

    The three lines below the “firm actions line” in
    Figure 3 reveal the strength of the Red Queen effect

    in our research. Specifically, line 2 shows how
    much the direct positive effect of actions on perfor-
    mance is reduced by the indirect negative impact
    focal firm actions have when they stimulate rival
    action. We calculated this negative impact by tak-
    ing the derivative of rival total actions with respect
    to firm total actions in model 5 multiplied by the
    coefficient on rival total actions in model 1 (– 0.05).
    Similarly, line 3 shows the direct positive effect of
    actions on performance along with the indirect neg-
    ative impact focal firm actions have by stimulating
    rival speed of actions. This negative impact was
    calculated as the derivative of rival speed of actions
    with respect to firm total actions in model 6 mul-

    FIGURE 2
    Effects of Firm Actions and Rival Actions on Firm Performance

    FIGURE 3
    Incremental Effects of Firm Actions on Firm Performance

    74 FebruaryAcademy of Management Journal

    tiplied by the coefficient on rival speed of actions
    in model 2 (–17.02). Comparing lines 2 and 3, we
    see that rival action speed has a greater negative
    effect than rival actions. Line 4 shows the cumula-
    tive negative effects of rival action and rival action
    speed.

    Importantly, even net of the negative effects that
    rival actions and rival action speed have on perfor-
    mance, the relationship between firm action and
    firm performance is still positively sloped. Even
    though a Red Queen effect is present in our re-
    search, the benefits of focal firm action outweigh
    the potentially negative consequences of rival ac-
    tion in this competitive contest overall. The same
    analysis using ROS to measure performance is also
    presented in Figure 3; the results are similar to
    those for ROA.

    The equations used to generate the plots in Fig-
    ure 3 can also be used to calculate the incremental
    impact that firm actions, rival actions, and rival
    action speed can have on firm performance. In the
    case of firm ROA, each additional firm action has
    an incrementally positive effect of increasing ROA
    by .104 percent while also causing a negative effect
    through rival actions and rival action speed that
    decreases ROA by .048 percent. Therefore. the net
    incremental increase in firm ROA from one firm
    action is .056 percent.

    Furthermore, using the results from these equa-
    tions, it is possible to explore the impact that being
    more or less active can have on firm performance.
    For example, if we define active firms as those
    taking a total number of actions one standard devi-
    ation above the mean (23.83 actions) and less active
    firms as those taking a total number of actions one
    standard deviation below the mean (2.53 actions),
    we can make comparisons based on the differential
    number of total actions between the two categories
    (21.3 actions). The positive effect of those 21.3 ac-
    tions is 2.2 or 59.2 percent of the average ROA
    (3.74%) in our sample. When we include the neg-
    ative effect of those actions that occurs through
    rival actions and rival action speed, the results are
    still substantial: the net effect on ROA is 1.2 per-
    cent, 31.9 percent of the average ROA. Similar re-
    sults are found with respect to ROS.

    This research contributes by advancing under-
    standing of Red Queen competition and the coevo-
    lutionary nature of firm search and action, and the
    relationship between focal firm action and rival
    action on focal firm performance, a key question in
    strategy. Specifically, conceiving competition as a
    contest of actions, we found support for the Red
    Queen effect by theoretically specifying, and em-
    pirically detailing, how firm actions and rival ac-
    tions have opposing effects on focal firm perfor-

    mance. As competitive dynamics is a fairly new
    stream of research (Smith et al., 1992), it has lacked
    theoretical roots that could give traction to future
    research agendas (Smith et al., 2001). The present
    study, with its focus on the Red Queen effect, sug-
    gests that evolutionary theory may offer important
    insights that can advance understanding of the dy-
    namics of competition.

    In an effort to better understand the boundaries
    of the Red Queen hypothesis, we also developed
    theory on how Red Queen evolution might depend
    upon different industry conditions and market po-
    sitions. Importantly, our results indicate that these
    factors significantly moderate the effects of firm
    actions on rival actions and their joint influence on
    performance. We speculated that these moderating
    factors block, or facilitate, the learning associated
    with search and action for both a focal firm and its
    rivals. Our study of these contextual moderating
    factors went well beyond antecedent research from
    IO economics. That is, despite the central role of
    conduct in the structure-conduct-performance par-
    adigm, IO researchers testing this theory have fo-
    cused mainly on the relationship between structure
    and performance and often assumed or not mea-
    sured the role of conduct. Additionally, their re-
    search has examined the direct effect that industry
    context and market position may have on firm per-
    formance, to the exclusion of actions or conduct; in
    contrast, our study explores the relationship be-
    tween actions and performance in the context of
    varying concentration, industry demand, and mar-
    ket position.

    With regard to focal firm performance exhibiting
    a more positive relationship to firm performance in
    highly concentrated or high-growth industries, our
    findings run somewhat counter to our predictions.
    We speculate that, in the case of high concentra-
    tion, the fact that firm action was only more posi-
    tively related to firm performance for one measure
    reflects the extent to which firms closely monitor
    each other’s actions, are very familiar with each
    other’s capabilities and developments, and are so
    highly interdependent that they create an environ-
    ment in which “surprise” actions are rare, and ri-
    vals are more likely to counter actions quickly and
    efficiently, wiping out excessive gains. In high-
    growth industries, it may be that we didn’t find a
    more positive relationship between firm action and
    firm performance because firms often act ineffi-
    ciently in an effort to keep up with the demands of
    the market. Ample demand opportunities may cre-
    ate an atmosphere in which firms’ actions are stop-
    gaps carried out to meet rising demand without
    consideration of their costs. That is, high-demand
    environments may create a situation in which firms

    2008 75Derfus, Maggitti, Grimm, and Smith

    do not have the time to investigate the least costly
    way to take action. In this way, firms may waste
    resources undertaking actions in haste or perhaps
    when they are unnecessary.

    With respect to the influence of high concentra-
    tion and demand on Red Queen competition, we
    found support for the proposition that the relation-
    ship between focal firm actions and rival actions is
    more intense in highly concentrated industries and
    less intense in high-growth industries. These re-
    sults support our contention that firms in concen-
    trated industries are much more interdependent
    than firms in less concentrated industries, while
    firms in high-growth industries are less interdepen-
    dent than those in low-growth industries. Both re-
    sults show how the industry context in which com-
    petition takes place moderates the Red Queen effect
    on firm evolution and performance.

    Market position also appears to have an influ-
    ence on Red Queen competition, or the relationship
    between focal firm actions, rival actions, and focal
    firm performance. Specifically, we predicted that
    the positive relationship between a firm’s actions
    and its performance would be stronger when the
    firm was a market leader, and our findings partially
    support this notion. Results for ROS indicate that
    firms in stronger leadership positions do receive
    greater performance benefits from action. While it
    has been suggested that large firms may become
    insulated from competitive forces and unrespon-
    sive to Red Queen competition (Barnett & McKen-
    drick, 2004), our results suggest that large firms
    with greater market share can become better com-
    petitors and enhance performance by being aggres-
    sive with their actions.

    Our predictions regarding the influence that mar-
    ket leaders’ actions have on rival actions were also
    supported. We found that the positive relationships
    between focal firm actions and rival actions, and
    rival action speed, were weaker when the focal firm
    was more of a market leader. We speculate that
    either rivals are less likely to act against leading
    firms out of fear of retribution, or market leaders
    take actions to which it is more difficult for rivals to
    respond.

    Contrary to our hypothesis, we found that rival
    actions have a greater negative impact on the per-
    formance of market leaders than on the perfor-
    mance of non–market leaders; in essence we found
    that “the larger they are, the harder they fall.” This
    result may be related to the concept of “judo strat-
    egy” as developed by Yoffie and Kwak (2001). With
    judo strategy, small rivals can effectively hurt mar-
    ket leaders, by eliciting responses that hurt the
    market leader more than the rival. When the market
    leader’s response affects all customers, it can be

    more costly for the leader than the nonleader with
    its lower market share. Interestingly, while our
    findings did not replicate Barnett and McKen-
    drick’s (2004) observation that smaller organiza-
    tions were more responsive to Red Queen compe-
    tition than larger organizations, we did observe that
    smaller firms can be more effective against their
    rivals by being aggressive with their actions.

    Overall, our findings highlight the intricacies of
    the relationship between competition and perfor-
    mance and the complexity of studying Red Queen
    competition. Though the tests of our baseline hy-
    pothesis yielded results that are completely consis-
    tent with Red Queen theory, our moderation find-
    ings revealed that the effects of search, action, and
    learning on firm performance and rival interdepen-
    dence largely depend on industry and competitive
    context. Further, while context definitely impacts
    whether Red Queen competition constrains or en-
    hances learning and performance, it does not al-
    ways do so in the ways that antecedent research
    would suggest. To better understand these relation-
    ships, more research needs to be done. As the focus
    of this research was on short-term performance,
    future research could fruitfully explore longer-term
    performance consequences of Red Queen competi-
    tion. Are the most active and aggressive firms the
    best performers in the long run? How does industry
    context influence Red Queen competition over the
    long term?

    Another potentially fruitful avenue for future re-
    search would be to focus on varying action types.
    To demonstrate how future research might evolve,
    we examined one possible characterization of ac-
    tion type, positive sum actions, which may allow
    firms to mitigate or reduce the negative aspects of
    Red Queen competition. Following Porter’s (1985)
    argument that competitors can provide strategic
    benefits by helping to develop markets and in-
    crease industry demand, some action types, namely
    geographic expansions, new marketing campaigns,
    and new-product introductions, may represent a
    “positive sum competition.” An illustration of this
    win-win dynamic can be seen when Starbucks en-
    ters a new geographic market in the retail coffee
    industry. Rather than negatively affecting the exist-
    ing competition in the new market, these competi-
    tors often witness increases in their business, as
    consumers become more comfortable with spe-
    cialty coffee and overall demand increases with the
    presence of the new Starbucks (Helliker & Leung,
    2002). Similarly, new-product introductions can
    positively increase or create demand for all com-
    petitors in an industry. For example, Sony’s intro-
    duction of the Walkman and Apple’s introduction
    of the iPod opened the door for a multitude of

    76 FebruaryAcademy of Management Journal

    imitations from competitors who garnered reve-
    nues from the expanded market.

    At least three types of moves are acknowledged
    in the literature to have the potential for positive,
    demand-expanding effects: geographic expansion,
    promotional campaigns, and new-product intro-
    ductions.4 To explore the potential of positive sum
    actions for future research, we created a measure of
    focal firm and rival positive sum actions by sum-
    ming actions categorized as geographic, marketing,
    and/or product introduction. We calculated firm
    positive sum actions by simply summing the num-
    ber of instances of all three actions for a focal firm
    in a given year. Rival positive sum actions were
    operationalized by subtracting a focal firm’s total
    number of positive sum actions in a given year from
    the total number of positive sum actions taken by
    all competitors in the industry.

    Post hoc regression of these types of actions
    yielded interesting results.5 Specifically, it appears
    that focal firm positive sum actions are signifi-
    cantly and positively related to firm performance in
    the case of return on sales. This result is consistent
    with our baseline model. Unlike in our baseline
    results, however, here there is a negative and sig-
    nificant relationship between firm positive sum ac-
    tions and both rival actions and rival action speed.
    In addition, rival positive sum actions and firm
    performance exhibited no significant relationship.

    Taken together, these findings suggest that firm
    positive sum actions incite less rivalrous action
    and slower rival action speed. Further, when rivals
    take positive sum actions, focal firm performance
    does not suffer significantly. This finding is again
    consistent with the speculation that firms take ac-
    tions that build legitimacy and benefit all players in
    an industry during the early, high-growth period of
    the industry life cycle (Agarwal & Bayus, 2002).
    These results also indicate that future research in-
    corporating action type could enhance understand-
    ing of Red Queen competition. One avenue for fu-
    ture research would be to combine action type
    effects with an examination of longer-term conse-
    quences of Red Queen competition. To clarify, the
    above analysis revealed that positive sum actions
    may mitigate the negative effects of Red Queen
    competition by reducing both the number and
    speed of rival actions. However, one could also
    conjecture that when a focal firm takes positive
    sum actions, rivals are more likely to take positive
    sum actions in response, a sequence that may lead
    to a rivalry-reducing “loop” with positive perfor-
    mance consequences over the longer term. Al-
    though a complete analysis is beyond the scope of
    our current study, it is interesting to note that there
    is a relatively high correlation between firm posi-
    tive sum actions and rival positive sum actions
    within our data set (r � .41), providing an indica-
    tion that firm positive sum actions beget rival pos-
    itive sum actions.

    This study has implications for practice. Specif-
    ically, the study of Red Queen competition pro-
    vides a number of insights regarding what actions
    managers can take and under what conditions they
    should take them in an effort to increase perfor-
    mance. Managers of competing firms in highly con-
    centrated industries or low-growth industries
    should be acutely aware of their mutual interde-
    pendence and be cautious of taking actions for fear
    of competitive reprisal. Our research also suggests
    that challenger firms can effectively hurt industry
    leaders by taking action, which is somewhat con-
    trary to conventional wisdom. Still, leading firms
    received a bigger payoff for acting than did non-
    leading firms. Although more research is required,
    the post hoc results suggest that managers can po-
    tentially avoid the negative consequences of rivalry
    by emphasizing positive sum actions such as geo-
    graphic actions and product introductions.

    Like most research, our study has limitations.
    First, although we studied a minimum of 70 per-
    cent of the business activity in each of 11 different
    industries, our sample favors large, public, single
    U.S. business firms that perhaps are in the later
    stages of the organizational life cycle. In particular,

    4 Geographic expansion, extending a firm’s reach to
    customers not previously served, can allow a firm to
    avoid head-to-head competition with existing rivals.
    Geographic expansion can be targeted where competition
    is weak or nonexistent, perhaps in the process filling a
    previously underserved geographic segment (Porter,
    1980). Promotional campaigns that generate new custom-
    ers and new demand are also consistent with a positive
    sum notion. As both parties increase their marketing
    efforts, the actual or potential customer base can be ex-
    panded, creating a situation in which the marketing ac-
    tions of the focal firm and the rival firm will both have
    positive benefits for the focal firm (Warren, 2002). Mar-
    keting campaigns can also aid firms in differentiating
    their products. Barney pointed out that product differen-
    tiation “reduces the threat of rivalry, because each firm in
    an industry attempts to carve out its own unique product
    niche” (Barney, 1997: 237). New-product introductions
    can also have positive impacts on demand. Each new-
    product introduction sets expectations and challenges
    rivals to act creatively with their own products in order
    to catch up (Schumpeter, 1934). The consequence of
    such innovation is that overall demand may increase as
    more and better new products are introduced to the mar-
    ket over time.

    5 Regression results are available upon request from
    the authors.

    2008 77Derfus, Maggitti, Grimm, and Smith

    the competitive dynamics between firms that are
    young, small, diversified, and private are not cap-
    tured by this study. We speculate that the compet-
    itive dynamics among firms in fragmented indus-
    tries and in early stages of the life cycle, where
    mutual interdependence is lower, would be quite
    different. Future research should continue investi-
    gation of these issues in these other industry con-
    texts. As is the case in competitive dynamics re-
    search, our study also captures only observable
    moves reported in the press and publications we
    examined. In addition, our use of product-market-
    based industries to define the competitive land-
    scape may not address the potentially changing
    nature of competition. For example, it is likely that
    focal firms in our sample face competitors from
    outside their industry or outside the United States
    that offer substitute products, have similar resource
    positions, or exist in geographic markets that are
    new to the focal firms (Chen, 1996; Peteraf & Ber-
    gen, 2003). Future research can overcome this lim-
    itation by identifying competitors on the basis of
    criteria other than product-market commonality.

    Additionally, although we theorize that search
    and learning take place in this Red Queen com-
    petitive context, we do not directly specify and
    measure these constructs. Future research could
    further elucidate Red Queen competition by ex-
    amining more specifically the search and learn-
    ing that take place. Future research could also
    consider additional links beyond the scope of the
    current study, such as whether and how firm
    competitive experience and performance shape
    future firm action. We can speculate that a firm’s
    prior learning and experience with action affect
    its future action and action speed. Moreover, ad-
    ditional research could examine the extent to
    which a focal firm learns vicariously from other
    firms (Baum, Li, & Usher, 2000).

    In conclusion, the most important contribution
    of this research is its theoretical and empirical
    examination of Red Queen competition, con-
    ceived of as the positive and negative conse-
    quences of firm actions on performance. Specifi-
    cally, the research shows that firm actions can
    play out as a Red Queen race among rivals: Firm
    actions are related to rival actions and rival ac-
    tion speed, and all can impact focal firm perfor-
    mance. However, we also demonstrated that the
    effects of firm action on performance are complex
    and dependent on the industry context and the
    market positions of competitors. Additional the-
    ory and research are needed to improve under-
    standing of Red Queen competition.

    REFERENCES

    Agarwal, R., & Bayus, B. 2002. The market evolution and
    take-off of new product innovations. Management
    Science, 48:1024 –1041.

    Agarwal, R., & Gort, M. 1996. The evolution of markets
    and entry, exit and survival of firms. Review of
    Economics and Statistics, 78: 489 – 498.

    Agarwal, R., Sarkar, M. B., & Echambadi, R. 2002. The
    conditional effect of time on firm survival. Academy
    of Management Journal, 45: 971–994.

    Argote, L. 1999. Organizational learning: Creating, re-
    taining and transferring knowledge. Boston: Klu-
    wer.

    Bain, J. S. 1951. Relation of profit rate to industry con-
    centration: American manufacturing, 1936 –1940.
    Quarterly Journal of Economics, 65: 293–324.

    Barnett, W. P., & Hansen, M. T. 1996. The Red Queen in
    organizational evolution. Strategic Management
    Journal, 17: 139 –157.

    Barnett, W. P., & McKendrick, D. 2004. Why are some
    organizations more competitive than others? Evi-
    dence from a changing global market. Administra-
    tive Science Quarterly, 49: 535–571.

    Barnett, W. P., & Sorenson, O. 2002. The Red Queen in
    organizational creation and development. Industrial
    and Corporate Change, 11: 289 –325.

    Barney, J. 1986. The types of competition and the theory
    of strategy: Toward an integrative framework. Acad-
    emy of Management Review, 11: 791– 800.

    Barney, J. 1997. Gaining and sustaining competitive
    advantage. Reading, MA: Addison-Wesley.

    Baum, J. A. C., Li, S. X., & Usher, J. M. 2000. Making the
    next move: How experiential and vicarious learning
    shape the locations of chains’ acquisitions. Admin-
    istrative Science Quarterly, 45: 766 – 801.

    Baumol, W. J. 2004. Red-queen games: Arms races, rule
    of law and market economies. Journal of Evolution-
    ary Economics, 14: 237–247.

    Bothwell, J. T., Cooley, T., & Hall, T. 1984. A new view of
    the market structure-performance debate. Journal of
    Industrial Economics, 32: 397– 417.

    Bradburd, R. M., & Caves, R. 1982. A closer look at the
    effect of market growth on industries profits. Re-
    view of Economics and Statistics, 64: 635– 645.

    Cameron, A., & Travendi, P. 1986. Economic models
    based on count data: Comparisons and applications
    of some estimators and tests. Journal of Applied
    Econometrics, 1: 29 –53.

    Carroll, G., & Hannan, M. 1989. Density dependence in
    the evolution of newspaper organizations. American
    Sociological Review, 54: 524 –541.

    Carroll, G., & Hannan, M. 2000. The demography of
    corporations and industries. Princeton, NJ: Prince-
    ton University Press.

    78 FebruaryAcademy of Management Journal

    Carroll, L. 1960. The annotated Alice: Alice’s adven-
    tures in wonderland and through the looking-
    glass. New York: New American Library.

    Caves, R. 1980. American industry: Structure, conduct,
    performance (6th ed.). Englewood Cliffs, NJ: Pren-
    tice-Hall.

    Caves, R., & Porter, M. E. 1977. From entry barriers to
    mobility barriers. Quarterly Journal of Economics,
    91: 241–261.

    Chen, M. J. 1988. Competitive strategic interaction: A
    study of competitive actions and responses. Un-
    published manuscript, University of Maryland, Col-
    lege Park.

    Chen, M. J. 1996. Competitor analysis and interfirm ri-
    valry: Toward a theoretical integration. Academy of
    Management Review, 21: 100 –135.

    Chen, M. J., & Miller, D. 1994. Competitive attack, retal-
    iation, and performance: An expectancy-valence
    framework. Strategic Management Journal, 15: 85–
    102.

    Comanor, W., & Wilson, T. 1967. Advertising, market
    structure, and performance. Review of Economics
    and Statistics, 49: 423– 440.

    Cubbin, J., & Domberger, S. 1988. Advertising and post-
    entry oligopoly behaviour. Journal of Industrial
    Economics, 37: 123–140.

    Dawkins, R. & Krebs, J. R. 1979. Arms races between and
    within species. Proceedings of the Royal Society of
    London, B205: 489 –511.

    Eisenhardt, K. M., & Tabrizi, B. N. 1995. Accelerating
    adaptive processes: Product innovation in the global
    computer industry. Administrative Science Quar-
    terly, 40: 84 –110.

    Erez, A., Bloom, M. C., & Wells, M. T. 1996. Using ran-
    dom rather than fixed effects models in meta-analy-
    sis: Implications for situational specificity and valid-
    ity generalization. Personnel Psychology, 49: 275.

    Ferrier, W. J., Smith, K. G., & Grimm, C. M. 1999. The role
    of competitive action in market share erosion and
    industry dethronement: A study of industry leaders
    and challengers. Academy of Management Journal,
    42: 372–383.

    Frome, E. L., Kutner, M. H., & Beauchamp, J. J. 1973.
    Regression analysis of Poisson-distributed data.
    Journal of American Statistical Association, 68:
    935–940.

    Gaskins, D. W. 1971. Dynamic limit pricing: Optimal
    limit pricing under threat of entry. Journal of Eco-
    nomic Theory, 3: 306 –322.

    Gilbert, R. J., & Newberry, D. M. 1982. Preemptive pat-
    enting and the persistence of monopoly. American
    Economic Review, 72: 514 –526.

    Graham, E. M. 1979. Technological innovation and the
    dynamics of the U.S. comparative advantage in in-
    ternational trade. In C. T. Hill & J. M. Utterback

    (Eds.), Technological innovation for a dynamic
    economy. New York: Pergamon.

    Greene, W. H. 1993. Econometric analysis (2nd ed.).
    New York: MacMillan.

    Grimm, C. M., & Smith, K. G. 1997. Strategy as action:
    Industry rivalry and coordination. Cincinnati:
    South-Western College Publishing.

    Helliker, K., & Leung, S. 2002. Counting beans: Despite
    the jitters, most coffeehouses survive Starbucks.
    Wall Street Journal, September 24, 2002: A1.

    Ingram, P. 2002. Interorganizational learning. In J. A. C.
    Baum (Ed.), Companion to organizations: 642– 664.
    Oxford, UK: Blackwell.

    Jauch, L. R., Osborn, R. N., & Martin, T. N. 1980. Struc-
    tured content analysis of cases: A complimentary
    method for organizational research. Academy of
    Management Journal, 5: 517–526.

    Kamien, M., & Schwartz, N. 1971. Limit pricing and
    uncertain entry. Econometrica, 39: 441– 454.

    Levitt, B., & March, J. G. 1988. Organizational learning. In
    W. R. Scott (Ed.), Annual review of sociology, vol.
    14: 319 –340. Palo Alto, CA: Annual Reviews.

    Liebowitz, S. J. 1982. Measuring industrial disequilibria.
    Southern Economic Journal, 49: 119 –136.

    March, J., & Simon, H. 1958. Organizations (2nd ed.).
    Cambridge, MA: Blackwell.

    Masson, R. T., & Shannan, J. 1986. Excess capacity and
    limit pricing: An empirical test. Economica, 53:
    365–378.

    Miller, D., & Chen, M. J. 1994. Sources and consequence
    of competitive inertia: A study of the U.S. airline
    industry. Administrative Science Quarterly, 39:
    1–23.

    Nelson, R. R., & Winter, S. G. 1982. An evolutionary
    theory of economic change. Cambridge, MA:
    Belknap.

    Peteraf, M. A., & Bergen, M. E. 2003. Scanning dynamic
    competitive landscapes: A market-based and re-
    source-based framework. Strategic Management
    Journal, 24: 1027–1041.

    Pisano, G. P. 1994. Knowledge, integration and the locus
    of learning: An empirical analysis of process devel-
    opment. Strategic Management Journal, 15: 85–
    100.

    Porter, M. E. 1980. Competitive strategy. New York: Free
    Press.

    Porter, M. E. 1985. Competitive advantage: Creating
    and sustaining superior performance. New York:
    Free Press.

    Scherer, F. M., & Ross, D. 1990. Industrial market struc-
    ture and economic performance. Boston: Hough-
    ton-Miller.

    Schmalensee, R. 1976. Advertising and profitability: Fur-

    2008 79Derfus, Maggitti, Grimm, and Smith

    ther implications of the null hypothesis. Journal of
    Industrial Economics, 25: 45–54.

    Schomburg, A. J., Grimm, C. M., & Smith, K. G. 1994.
    Avoiding new product warfare: The role of industry
    structure. In P. Shrivastava, A. Huff, & J. Dutton
    (Eds.), Advances in strategic management, vol. 10:
    145–174. Greenwich, CT: JAI.

    Schumpeter, J. 1934. The theory of economic develop-
    ment. Cambridge, MA: Harvard University Press.

    Schumpeter, J. A. 1976. Capitalism, socialism, and de-
    mocracy (5th ed.). London: George Allen & Unwin.

    Smith, K. G., Ferrier, W. J., & Ndofor, H. 2001. Competi-
    tive dynamics research: Critique and future direc-
    tions. In M. A. Hitt, R. E. Freeman, & J. S. Harrison
    (Eds.), Handbook of strategic management: 315–
    361. Oxford, UK: Blackwell.

    Smith, K. G., Grimm, C. M., & Gannon, M. J. 1992. Dy-
    namics of competitive strategy. London: Sage.

    Spence, A. M. 1977. Entry, capacity, investment and
    oligopolistic pricing. Bell Journal of Economics, 8:
    531–544.

    van Valen, L. 1973. A new evolutionary law. Evolution-
    ary Theory, 1: 1–30.

    Waldman, D. E., & Jensen, E. J. 2001. Industrial organi-
    zation. Boston: Addison-Wesley.

    Warren, K. 2002. Competitive strategy dynamics. New
    York: Wiley.

    Weick, K. 1995. Sensemaking in organizations. Thou-
    sand Oaks, CA: Sage.

    Williamson, O. E. 1965. A dynamic theory of interfirm
    behavior. Quarterly Journal of Economics, 79: 579 –
    607.

    Yoffie, D. B., & Kwak, M. 2001. Judo strategy. Boston:
    Harvard Business School Publishing.

    Young, G. 1993. Engaging a rival: Industry and firm
    specific predictors of rivalrous firm conduct in the
    U.S. software industry. Unpublished manuscript,
    University of Maryland, College Park.

    Young, G., Smith, K. G., & Grimm, C. M. 1996. “Austrian”
    and industrial organization perspectives on firm-
    level competitive activity and performance. Organi-
    zation Science, 7: 243–254.

    Pamela J. Derfus (pderfus@verizon.net) received her
    Ph.D. in strategic management from the Robert H. Smith
    School of Business at the University of Maryland. Her
    research interests lie in the areas of competitive dynam-
    ics, strategy implementation, and cooperation between
    firms.

    Patrick G. Maggitti (maggitti@temple.edu) is an assistant
    professor of management and entrepreneurship in the
    Fox School of Business at Temple University. His re-
    search focuses on the dynamics of competition and the
    decision making of executives, entrepreneurs, and inves-
    tors. He received his Ph.D. in strategic management from
    the Robert H. Smith School of Business at the University
    of Maryland.

    Curtis M. Grimm (cgrimm@rhsmith.umd.edu) is the
    Dean’s Professor of Supply Chain and Strategy at the
    Robert H. Smith School of Business, University of Mary-
    land. He received his Ph.D. in economics from the Uni-
    versity of California, Berkeley, with primary focus on
    industrial organization. Professor Grimm’s research has
    focused on the interface of business and public policy
    with strategic management, with a particular emphasis
    on competition and competition policy.

    Ken G. Smith (kgsmith@rhsmith.umd.edu) is the Dean’s
    Chair and a professor of business strategy at the Robert H.
    Smith School of Business, University of Maryland. He
    earned a Ph.D. in business policy from the University of
    Washington. His research interests include strategic de-
    cision making, competitive dynamics, and the manage-
    ment of knowledge and knowledge creation.

    80 FebruaryAcademy of Management Journal

    r Academy of Management Journal
    2017, Vol. 60, No. 5, 1882–1914.
    https://doi.org/10.5465/amj.2015.029

    5

    RED QUEEN COMPETITIVE IMITATION IN THE U.K. MOBILE
    PHONE INDUSTRY

    CLAUDIO GIACHETTI
    Ca’ Foscari University of Veni

    ce

    JOSEPH LAMPEL
    University of Manchester

    STEFANO LI PIRA
    University of Warwick

    This paper uses Red Queen competition theory to examine competitive imitation. We
    conceptualize imitative actions by a focal firm and its rivals along two dimensions:
    imitation scope, which describes the extent to which a firm imitates a wide range (a

    s

    opposed to a narrow range) of new product technologies introduced by rivals; and
    imitation speed, namely the pace at which it imitates these technologies. We argue that
    focal firm imitation scope and imitation speed drive performance, as well as imitation
    scope and speed decisions by rivals, which in turn influence focal firm performance. We
    also argue that the impact of this self-reinforcing Red Queen process on firms’ actions
    and performance is contingent on levels of product technology heterogeneity—defined
    as the extent to which the industry has multiple designs, resulting in product variety. We
    test our hypotheses using imitative actions by mobile phone vendors and their sales
    performance in the U.K. from 1997 to 2008.

    Once we become self-consciously aware that the
    possibilities of innovation within any one company
    are in some important ways limited, we quickly see
    that each organization is compelled by competition to
    look to imitation as one of its survival and growth
    strategies. (Levitt, 1966: 38)

    The emergence of what has often been referred to
    as the “new economy” has greatly expanded re-
    search on the power of technological innovation to
    create competitive dynamics that can reshape in-
    dustries (Baumol, 2004; Teece, 1998). While the fo-
    cus on innovation as the engine of industry evolution
    reflects both the potential gains that accrue to first
    movers (Lieberman & Montgomery, 1988), and the

    dramatic impact of disruptive technologies on the
    competitive landscape (Christensen & Bower, 1996),
    it inadvertently tends to eclipse the importance of
    imitation as an agent of change. Researchers that take
    a broader perspective see imitation as the twin pro-
    cess to innovation that, arguably like innovation,
    also plays a role in industry evolution in all contexts
    (Cohen & Levinthal, 1989; Levitt, 1966; Semadeni &
    Anderson, 2010), but takes on even greater signifi-
    cance in the rapidly changing technology-intensive
    industries that constitute the new economy. As
    Baumol (2004: 246–247) observed,

    in the new economy no firm [. . .] can afford to fall
    behind its rivals. […] If a firm fails to adopt the latest
    technology—even if the technology is created by
    others—then its rivals can easily take the lead and
    make disastrous inroads into the slower firm’s sales.

    Formulating an effective imitation strategy is
    a problem that confronts managers in any industry
    (Lieberman & Asaba, 2006), but in industries with
    rapid technological change the problem is com-
    pounded by higher levels of uncertainty about the
    market performance of new product technologies
    (Utterback & Suarez, 1993). This “technological

    We would like to thank Associate Editor Dovev Lavie
    and three anonymous reviewers for their invaluable com-
    ments and guidance during the review process, which
    helped strengthen this article. We would also like to thank
    Marco Li Calzi, Massimo Warglien, Francesco Zirpoli,
    Anna Comacchio, Juan Pablo Maicas, and Gianluca Marchi
    for their insightful comments on earlier drafts of this arti-
    cle. Finally, we thank seminar participants at Ca’ Foscari
    University of Venice, and at a 2015 Academy of Manage-
    ment Conference paper session, for their astute remarks on
    earlier versions of this article.

    188

    2

    Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express
    written permission. Users may print, download, or email articles for individual use only.

    https://doi.org/10.5465/amj.2015.0295

    uncertainty” presents managers with considerable
    challenges when deciding how far and how fast they
    should imitate their rivals, and this challenge per-
    sists when their decisions, in turn, create competi-
    tive conditions that may bring further pressure to
    imitate (Gaba & Terlaak, 2013; Rhee, Kim, & Han,
    2006). In this paper, we address the questions of the
    extent to which, and speed with which, firms should
    imitate their rivals, taking into account both the com-
    petitive dynamics that ensue as a result of innovation
    and imitation decisions, and the level of technological
    uncertainty in rapidly changing technology-intensive
    industries.

    Our analysis of imitation must begin with the
    recognition that imitation is a strategic choice that
    firms pursue when they wish to lower risks and
    costs by learning from their rivals’ actions, espe-
    cially when these actions involve pioneering new
    technologies, or launching radically innovative
    products (Ethiraj & Zhu, 2008; Lee, Smith, Grimm, &
    Schomburg, 2000). However, imitation is also
    a competitive move that can pose a threat to rivals
    that have yet to adopt the pioneering technologies,
    or introduce products with similar features. When
    some of the laggards react to the threat by also imi-
    tating, this gives rise to “competitive imitation,”
    a process in which imitation by some of the firms in
    an industry puts competitive pressure on the rest to
    also imitate. This process is consistent with the re-
    lationship between action and reaction that has
    been extensively studied by competitive dynamics
    research (Smith, Ferrier, & Ndofor, 2001). The main
    premise of competitive dynamics is that the actions
    of one firm, or group of firms, trigger reactions by
    other firms, which in turn produce a series of ac-
    tions and reactions that continue as long as firms
    seek to improve their competitive position (Derfus,
    Maggitti, Grimm, & Smith, 2008; Smith, Grimm,
    Gannon, & Chen, 1991). In technology-intensive
    industries, innovation triggers the competitive im-
    itation process. Faced with mounting evidence that
    the innovator’s new product technologies are find-
    ing a market, other firms that previously refused to
    match the innovator’s move begin to experience
    increasing pressure to imitate. Their imitative move
    serves to entrench the new technologies in the
    market even further, which in turn not only in-
    creases competitive imitation but also accelerates
    the evolution of the industry.

    The coevolutionary process by which firms act
    and react to each other has been shown by organi-
    zational scholars to influence both firm performance
    and industry structure (Nelson & Winter, 1982). Not

    unexpectedly, scholars have also noted that co-
    evolutionary processes in competitive environments
    have parallels with biological evolution. The paral-
    lels have led scholars to borrow from the work of
    evolutionary biologists, notably Van Valen’s (1973)
    work on the coevolution of dynamically interacting
    species. Of particular interest is the “Red Queen”
    effect, the allusion made by Van Valen (1973) to
    Alice’s encounter with the Red Queen in Lewis
    Carroll’s Through the Looking-Glass (Carroll, 1960),
    when he sought to explain the constant probability of
    species’ extinction regardless of the duration of their
    evolutionary history.1 Organization researchers have
    argued that what holds for biological evolution is
    in principle also the case in business contexts.
    Thus, firms can be said to engage in a “Red Queen
    competition” (Barnett & Hansen, 1996; Barnett &
    Sorenson, 2002); that is, the continuous and esca-
    lating activity of firms trying to maintain relative
    fitness in a dynamic system, such that they end up
    improving as fast as they can just to stand still rela-
    tive to competitors.

    We draw on the literature on Red Queen compe-
    tition, and research on competitive dynamics, imi-
    tation, and technology innovation, to build a model
    that captures how decisions to imitate new product
    technologies stimulate further imitation by rivals,
    and how this “competitive imitation” in turn in-
    fluences, and is influenced by, changing industry
    conditions. Our study complements the competitive
    dynamics and imitation literature in several re-
    spects. To begin with, most extant research on imi-
    tation of innovations has tended to see imitation as
    a binary variable: firms either imitate or they do
    not (e.g., Greve, 1998; Hsieh & Vermeulen, 2013;
    Makadok, 1998). In practice firms seldom imitate, or
    do not imitate, every aspect of their rivals’ offerings,
    but instead tend to imitate some of the features of
    products introduced by rivals, while retaining
    existing features (Bayus & Agarwal, 2007; Giachetti &
    Dagnino, 2017). From this it follows that managers
    face two basic questions when they consider imita-
    tion as the best next move: the first is how much to
    copy, i.e., “imitation scope” (Csaszar & Siggelkow,
    2010; Narasimhan & Turut, 2013), and the second is

    1 In reference to Carroll’s tale, when the Red Queen re-
    sponds to Alice “here, you see, it takes all the running you
    can do, to keep in the same place” (Carroll, 1960: 345), Van
    Valen noted that biological evolution features such
    change: species must constantly adapt in order to survive,
    while confronting ever-evolving rival species in an ever-
    changing environment.

    2017 1883Giachetti, Lampel, and Li Pira

    how quickly to imitate, i.e., “imitation speed” (Lee
    et al., 2000). In this paper, we argue that the scope
    and speed decisions of one firm influence the scope
    and speed decisions of rivals. Rivals’ imitation
    scope and speed decisions will then influence the
    firm’s subsequent scope and speed decisions. What
    applies to the reaction of rivals to the actions of
    a single firm is also true for the industry as a whole. If
    we step one level of analysis up to consider the entire
    group of industry rivals, we can see Red Queen
    competition from a wider perspective: the speed and
    scope decisions made by firms at different times in-
    duce each other’s speed and scope decisions. Our
    contribution in this paper is to show that Red Queen
    competition in technology-intensive industries es-
    calates the magnitude of imitation speed and scope
    choices for all competitors.

    Second, thus far, competitive dynamics studies
    have not examined how changes in the techno-
    logical environment may affect the Red Queen
    cycle. This analysis is especially important in
    technology-intensive industries where technolog-
    ical change, for example the emergence or decline
    of dominant designs, can dramatically alter com-
    petition (Chen & Turut, 2013), render obsolete
    a firm’s capabilities (Barkema, Baum, & Mannix,
    2002; Bayus & Agarwal, 2007; Utterback & Suarez,
    1993), and encourage firms to develop a new
    technology imitation strategy (Narasimhan &
    Turut, 2013). It is not difficult to see that changes
    in the technological environment that drive new
    product introductions are often central to the types
    of moves that drive Red Queen competition. For
    example, recent studies in the imitation and tech-
    nology innovation literature (Argyres, Bigelow, &
    Nickerson, 2015; Giachetti & Lanzolla, 2016;
    Madhok, Li, & Priem, 2010; Posen, Lee, & Yi, 2013)
    have shown that as industries evolve, changes in
    technologies, and subsequently their diffusion,
    can influence rates of imitation. These studies
    complement prior work by authors such as
    Utterback and Suarez (1993), who pointed out that
    as the industries mature they tend to transition
    from high to low levels of product technology
    heterogeneity—where low levels of product tech-
    nology heterogeneity correspond to the emergence
    of design dominance. Low product technology
    heterogeneity in turn leads to low technological
    uncertainty: firms find it easier to know which
    design options are more likely to yield good market
    performance, and are thus more likely to imitate.
    Our paper examines how product technology het-
    erogeneity moderates the Red Queen effect.

    Third, existing competitive dynamics studies
    have tended to examine antecedents and perfor-
    mance outcomes of action types such as pricing,
    marketing, and capacity expansion, in industries
    such as professional services, professional sports,
    and motion pictures, where technology is a minor
    competitive factor (Lampel & Shamsie, 2009; Ross &
    Sharapov, 2015; Semadeni & Anderson, 2010); or in
    industries such as airlines, where technology is
    more important but is still peripheral to the main
    factors responsible for success (Chen & Miller, 1994;
    Miller & Chen, 1994; Smith et al., 1991). In contrast,
    we chose to examine Red Queen competitive dy-
    namics in an industry where “creative destruction”
    (Schumpeter, 1942), triggered by the introduction
    of new products and technologies, is the primary
    competitive force. The mobile phone industry is
    a rapidly changing technology-intensive industry
    where continuous and swift imitation of rivals’ in-
    novation is a key prerequisite for handset vendors to
    maintain competitive parity. More specifically, our
    research site is the U.K. mobile phone industry from
    1997 to 2008, a period during which the industry
    evolved rapidly, driven by incessant rivalry among
    a dozen handset vendors to get or keep ahead of one
    another.

    The rest of our paper is structured as follows. We
    begin with an overview of Red Queen theory. We
    subsequently define and discuss imitation scope,
    imitation speed, and product technology heteroge-
    neity, and derive hypotheses about how these as-
    pects influence Red Queen competitive imitation.
    We then describe our methods and present our re-
    sults. We conclude with limitations of our study and
    suggestions for future research.

    THEORY BACKGROUND AND HYPOTHESES

    Red Queen Competitive Imitation: Focal Firm,
    Rivals, and Firm Performance

    In this section, we develop a theory that explains
    the Red Queen effect in terms of a firm’s imitation of
    new product technologies, rivals’ imitation of new
    product technologies, and their combined impact on
    the firm’s performance. To ensure that our theory
    development is consistent and clear, it is important
    to define imitation in contrast to innovation before
    we move forward. As pointed out by Semadeni and
    Anderson (2010), in markets where firms can closely
    examine their competitors’ product offerings and
    track the market performance of those offerings in
    real time, firms can choose between introducing to

    1884 OctoberAcademy of Management Journal

    the market products with new features, or confining
    their actions to copying features previously in-
    troduced by rivals. We likewise also distinguish be-
    tween introduction and copying, and define
    innovation as introducing first to the market prod-
    ucts that contain new features, and imitation as
    copying others’ innovations.

    In this paper we focus our analysis on imitative
    actionswhile controlling for innovative actions. Two
    types of imitative decisions are examined: imitation
    scope and imitation speed. Our decision is based on
    evidence provided by different but complementary
    streams of literature. On the one hand, the technol-
    ogy innovation literature has argued that a wider
    imitation scope is an indication that the firm’s
    products can stay abreast of new technologies
    (Narasimhan & Turut, 2013). Yet, at the same time,
    the literature on first-mover advantage, as well as
    competitive dynamics literature, has focused more
    closely on imitation speed, arguing that higher imi-
    tation speed is a signal the firm is one of the first
    players committed to adopting new technologies so
    as to keep up with innovators and differentiate with
    respect to laggard rivals (Lee et al., 2000; Markides &
    Geroski, 2004). Though these studies have been in-
    terested in whether imitation represents a source of
    performance differences, they have posed somewhat
    different research questions, and thus have pro-
    gressed along independent trajectories. In practice,
    when a firm faces a group of rivals who are in-
    troducing new products with a variety of features at
    different times, they cannot focus only on scope or
    speed, but must consider both the question of how
    many of the features the firm should imitate, and also
    how quickly it should proceed with imitation. In this
    article, we propose to bring together the different
    analyses of imitation and competition explored in
    these bodies of literature in order to obtain a broader
    understanding of the roles that imitation scope and
    speed play in sustaining the Red Queen competitive
    imitation.

    Because Red Queen competition describes a re-
    ciprocal back-and-forth process, firms play different
    roles in different time periods, and it is important to
    be clear and consistent about the labels we use when
    referring to firms. In Red Queen papers, the “focal
    firm” and “rivals” may switch places in the analysis
    over time. The “focal firm” is the industry player
    whose imitative moves attract attention and call for
    a response from other firms, the “rivals” at that spe-
    cific point in time. For example, we could say that at
    time t the “focal firm,” having observed new product
    technologies introduced by one or several of its

    “rivals” in t 2 1, must decide how many of these
    technologies it should imitate. In this context, “ri-
    vals” are all the other firms within the industry that
    the “focal firm” sees as competitors. “Rivals,” for
    their part, observe the focal firm’s moves, gauge the
    resulting performance, and decide on how many of
    these moves they should imitate at time t 1 1. This
    turns the rival firms into focal firms, who are now
    observing and analyzing moves recently made by
    rivals. Their actions challenge rivals, who must now
    consider their moves, and so on.

    To summarize, the baseline Red Queen competi-
    tive imitation we develop in this section works as
    follows. Focal firms that successfully imitate new
    product technologies obtain performance advantages
    (e.g., sales increases) by virtue of competitive advan-
    tage that they hold vis-à-vis rivals that imitate either
    less intensively (i.e., lower imitation scope), or more
    slowly (i.e., lower imitation speed). Higher perfor-
    mance of focal firms that imitate more intensively, or
    more rapidly, combined with performance losses ex-
    perienced byrivals, willmotivatethe lattertorespond
    by increasing their imitation scope and speed. The
    more intense and rapid the rivals’ imitative response,
    the more the focal firm experiences a threat to its
    performance, and the more it feels pressure to
    respond—by innovating or imitating.

    It is worth noting that our theory of Red Queen
    competitive imitation describes competition as the re-
    sult of a sequence of imitative actions after a set of
    new product technologies are introduced. We argue
    that focal firms imitate innovators (i.e., technology pi-
    oneers), and rivals subsequently imitate focal firms
    in an incessant race to maintain competitive parity
    (Lieberman&Asaba,2006).Morespecifically,whilethe
    rationale for the first imitations (by the quickest imita-
    tors) is “informationally based”—i.e., when making
    imitative decisions first imitators use the information
    generated by market performance of the new technol-
    ogies introduced by innovators—the rationale for
    subsequent imitations is also motivated by “competi-
    tive bandwagon” pressure (Abrahamson & Rosenkopf,
    1993)—i.e., the pressure on nonimitators when they
    face diminishing profit opportunities as more of their
    competitors imitate innovative first movers.

    The Competitive Advantage of More Active Firms:
    Learning and Repertoires of Actions

    Taking their inspiration from Joseph Schumpeter,
    specifically his concept of “creative destruction”
    (Schumpeter, 1942)—which, concisely summarized,
    arguesthat competition is adynamicmarket processin

    2017 1885Giachetti, Lampel, and Li Pira

    which entrepreneurs trigger and respond to change—
    competitive dynamics research has shown that more
    “active” firms, defined as those that take more frequent
    competitive actions than most of their industry rivals,
    are more likely to attain higher performance (Ferrier,
    Smith, & Grimm, 1999; Young, Smith, & Grimm, 1996).
    In contrast, firms that lag behind most of their industry
    rivalswhen it comestotaking competitiveactionstend
    to be at a competitive disadvantage (Miller & Chen,
    1994). There are several related factors that account for
    this relationship. First, firms that are more active are
    more likely to keep pace with change in a rapidly
    changing environment (Chen, Lin, & Michel, 2010;
    Ndofor, Sirmon, & He, 2011; Smith, Ferrier, & Ndofor,
    2001). Second, because they make more moves, these
    firms are also more likely to take actions that change
    the environment in ways that are favorable to them,
    and lessfavorabletolessactivefirms (Rindova, Ferrier,
    & Wiltbank, 2010). Finally, in dynamic environments
    in which the direction and consequences of change are
    uncertain, firms that are more active have a shorter
    learning cycle compared to firms that are less active.
    Active firms capture and put to use the knowledge
    gained from observing their rivals more quickly com-
    pared to firms that hesitate (Baum & Ingram, 1998;
    Baum, Li, & Usher, 2000; Greve, 1996).

    Learning also plays a central role in research on
    Red Queen competition. Initial Red Queen studies
    sought to show that competition and learning trigger
    one another in an ongoing, self- reinforcing process
    (Barnett & McKendrick, 2004; Barnett & Sorenson,
    2002). As Barnett and Sorenson (2002: 290) put it,
    Red Queen is a process that results when “competi-
    tion among organizations triggers internal learning
    processes; and learning increases the strength of
    competition generated by an organization.” More
    recent Red Queen research has focused to a greater
    extent on learning as a process in which rivals try to
    figure out the causal mechanism that links a reper-
    toire of competitive moves to performance (Derfus
    et al., 2008). The simplest competitive repertoire
    consists of a single move. In markets where a single
    move type is central to performance (e.g., price re-
    duction), Red Queen is confined to single-type tit-
    for-tat responses. In most markets, however, the focal
    firm’s competitive advantage (or disadvantage) re-
    sults from a combination of successful (or failed)
    competitive actions,2 and firms face choices about
    which combination of moves they should employ. If

    the repertoire of possible moves focuses primarily on
    product technologies, firms have to assess which of
    the new technologies launched by rivals should be
    imitated, and which should be avoided.

    In the remaining part of this theory section, we
    develop a set of hypotheses about our theory of Red
    Queen competitive imitation. Our argument is that
    focal firms will perform better than “less active”
    imitators if they are “more active,” both in terms of
    the number of new product technologies they imitate
    and the speed at which they are able to imitate.
    Further, we argue that product technology heteroge-
    neity may constrain focal firms’ learning capabilities,
    obstructing their ability to increase performance via
    imitative actions.

    Scope and Average Speed of a Firm’s Imitation of
    New Product Technologies and its Performance

    How much to copy: Imitation scope as a com-
    petitive response. In the specific context of new
    product technology in which we are interested,
    multiple imitation opportunities present firms with
    the strategic choice regarding how many of the
    technologies introduced by rivals they should
    imitate. This scenario is typical in technology-
    intensive industries, such as consumer electronics
    (e.g., mobile phones and personal computers),
    where firms constantly face competitive threats
    from new product technologies that expand the set
    of functionalities that are offered to consumers
    (Bayus & Agarwal, 2007). The choice that confronts
    firms as new products with new functionalities
    enter the market is how many of these functional-
    ities they should incorporate into their products.
    The choice targets what we call “imitation scope;”
    that is, the extent to which a firm (in a given period)
    imitates a wide number (as opposed to a narrow
    number) of new product technologies introduced
    by competitors.

    When looking at imitation scope, we have to bear
    in mind that consumers evaluate the desirability of
    adopting new features in the context of the entire
    bundle of functionalities offered by the product
    (O’Shaughnessy, 1989). In other words, consumers
    compare products with, and without, a given func-
    tionality before making a purchase. The inclusion of
    a functionality will not necessarily motivate them to
    make a purchase, unless the additional functionality
    adds to the value of the package as a whole. First
    movers (i.e., innovators) must make this evaluation
    without prior market data (or at best consumer re-
    search data), while imitators can use the market

    2 See Chen and Miller (2012) for an extensive review
    comparing studies on single actions versus action
    repertoires.

    1886 OctoberAcademy of Management Journal

    performance of new functionalities when making
    this decision (Carpenter & Nakamoto, 1989). The
    problem, however, is that firms have data on mul-
    tiple functionalities. Some of these functionalities
    are present in the same product, which makes it
    difficult to evaluate them separately, while other
    functionalities are spread across multiple products
    and present in a variety of combinations—creating
    an even greater evaluation challenge (Krishnan &
    Bhattacharya, 2002).

    If firms cannot analyze the sales potential of indi-
    vidual functionalities, the question that arises is
    whether they can evaluate the potential of sets of
    functionalities. Technology innovation literature that
    has examined the consumer buying behavior of
    products with multiple functionalities (Chen & Turut,
    2013) has suggested that when firms have to assess
    how consumers evaluate a set of objects—in our case,
    products that offer certain functionalities—they will
    evaluate the options they are presented by consider-
    ing both the absolute utility of each feature (e.g., text
    messaging in mobile phones), and their relative
    standing in the choice set (i.e., how valuable text
    messaging is relative to other functionalities in the
    set). The evaluation relies on reference points that are
    endogenous to the choice set (Baucells, Weber, &
    Welfens, 2011). This can be the product that the
    consumer currently owns, or some idealized combi-
    nation of functionalities in the product that the con-
    sumer wishes to purchase (Zhou, 2011). Reference
    points in a technologically mature industry where
    products perform a stable set of well-established
    functionalities are more likely to be based on price,
    since the difference between the functionalities of old
    and new products is not substantial. However, in in-
    dustries where technology is evolving rapidly, as in
    most technology-intensive industries, consumers’
    reference points are future oriented, and tend to
    change as new functionalities are introduced. As
    Chen and Turut (2013: 2748) put it:

    Context dependent preferences are especially rele-
    vant for consumers’ adoption of technology in-
    novation because the reference points of product
    attributes in consumers’ minds are likely to evolve
    over time with the advance of technology and the ar-
    rival of new products in the market; this influences
    consumers’ adoption of products with new technol-
    ogy and consequently firms’ innovation strategies.

    Introduction of new functionalities in the form of
    new product features or attributes tends to shift the
    reference point toward the innovative feature, and
    away from old features. Put differently, consumers

    will value the entire set of functionalities in a prod-
    uct more if the product includes new functionalities
    that represent the next step in the evolution of un-
    derlyingtechnologies. This shift in reference point as
    technology evolves strongly influences the compet-
    itive logic in these markets. While it creates in-
    centives to innovate new functionalities, it creates
    even stronger incentives to imitate (Narasimhan &
    Turut, 2013).

    Narasimhan and Turut (2013) provided empirical
    support for the advantages of imitation, showing that
    firms attain higher performance if they choose to im-
    itate as many pioneering features introduced by rivals
    as possible, rather than differentiate by introducing
    their own features. Their conclusions are in line with
    other empirical studies of consumer attitudes sug-
    gesting that in markets where technology is rapidly
    evolving, consumers evaluate more favorably brands
    with a reputation for staying abreast of new technol-
    ogies, while at the same time displaying a strong bias
    against brands that lack the latest technologies
    (O’Shaughnessy, 1989; Pessemier, 1978). From the
    point of view of firms that are considering how many
    of the new functionalities they should adopt in the
    new product offerings, this suggests that firms are
    more likely to gain sales if they adopt as many of the
    new features as their capabilities will allow. This
    leads to the following hypothesis:

    Hypothesis 1a. An increase in the focal firm
    scope of imitation of new product technologies
    will positively influence its performance.

    How fast to copy: Average speed of imitation as
    a competitive response. Another question firms
    must confront is how quickly to imitate rivals’ moves
    (Markides & Geroski, 2004). Similar to our discus-
    sion on imitation scope in technology-intensive in-
    dustries where firms launch products that combine
    multiple technologies, and hence present multiple
    imitation opportunities, a related decision that con-
    fronts firms is how quickly these multiple technol-
    ogies should be imitated. At the product line level,
    this choice targets what we call “average speed of
    imitation:” the average time it takes for the focal firm
    to adopt the set of new product technologies in-
    troduced by rivals.

    From a decision-making perspective, the question
    of how quickly a firm should imitate its rivals has
    been explored primarily from the perspective of first-
    mover advantage (Lieberman & Montgomery, 1988).
    The merit of moving first with a new product has
    been extensively argued and documented (Makadok,
    1998). Researchers, however, have also come to

    2017 1887Giachetti, Lampel, and Li Pira

    recognize that firms that move later can avoid many
    of the risks that confront first movers by observing,
    analyzing, and then imitating their products and
    technologies (Lieberman & Montgomery, 1998;
    Markides & Geroski, 2004). What is less certain is
    how quickly late movers have to act if they want to
    minimize risks and maximize the advantages of early
    information. Studies in the competitive dynamics
    and first-mover literature have suggested that, on the
    whole, fast imitators—i.e., firms that imitate earlier
    than others pioneering innovations—will generally
    do better than firms that are slow to imitate (Lee et al.,
    2000). The advantages of fast imitation are especially
    strong in industries where first adopters of new
    product technologies benefit from “spatial pre-
    emption”; that is, the filling of product differentia-
    tion niches before late adopters enter (Rao &
    Rutenberg, 1979; Rindova et al., 2010). Because
    spatial preemption limits the product differentiation
    opportunities available to late adopters, we expect
    rapid imitation of new product technologies to de-
    liver higher performance for imitators that move
    faster. In other words, higher average speed of imi-
    tation of new product technologies offers the focal
    firm more differentiation opportunities with respect
    to later imitators, and is likely thereafter to lead to
    higher sales volume.

    The advantages of quick imitation of new product
    technologies, however, are not confined to spatial
    preemption. Quick imitation also has a significant
    impact on consumer perception of firm reputation.
    Research has shown that consumers tend to view firms
    that quickly adopt new technologies as generally more
    innovative (Alpert & Kamins, 1995; Carpenter &
    Nakamoto, 1989; Kardes & Kalyanaram, 1992). This
    judgment creates a “halo” effect that favorably skews
    the evaluation of the firm’s product line, and hence
    contributes to sales growth. In contrast, the product
    lines of firms that are slow to adopt new technologies
    (i.e., have low average speed of imitation) are judged
    morenegativelybyconsumers.Thisnegativelyskewed
    judgment tends to depress sales growth for slow
    adopters. Therefore, in a context of multiple imitation
    opportunities, firms with a high average speed of
    imitation of new product technologies will be viewed
    as technology leaders, and hence will benefit from
    a higher reputation among customers that will en-
    hance their sales performance. Thus, we predict:

    Hypothesis 1b. An increase in the focal firm’s
    average speed of imitation of new product
    technologies will positively influence its
    performance.

    Scope and Average Speed of a Firm’s Imitation of
    new Product Technologies and the Scope and
    Average Speed of Rivals’

    Imitative Actions

    As noted earlier, Red Queen competition suggests
    that as the number of focal firm actions increases, the
    number of rival firm actions increases as well (Derfus
    etal.,2008). Thatisbecausethe greaterthe focal firm’s
    competitive activity, the more competitors are likely
    to perceive a threat to their performance, which in
    turn makes it more likely that they will respond
    (Barnett & Hansen, 1996; Barnett & McKendrick,
    2004). In other words, a focal firm’s increase in com-
    petitive activity will present rivals with a challenge
    that will increase in magnitude if the focal firm moves
    ahead with new product offerings that leave rivals
    with market spaces that are less and less valued by
    customers. This threat will force rivals to respond
    with competitive moves of their own in order to close
    the gap and maintain their position.

    Lieberman and Asaba (2006: 380) noted that
    “rivalry-based imitation often proceeds over many
    rounds, where firms repeatedly match each other’s
    moves.” Generally speaking, rivalry encourages im-
    itation, which in turn encourages more rivalry. The
    competitive dynamics literature has suggested that
    competitors that wish to maintain competitive parity
    must imitate intensively (i.e., imitation scope) and
    rapidly (i.e., imitation speed). This imitation effort
    escalates as rivals struggle for profits and market
    share. Indeed, the improved focal firm performance
    derived from intense and rapid imitation of new
    product technologies comes at the expense of rivals’
    performance, which, in turn, may prompt rivals to
    trigger aggressive imitative actions that emulate the
    focal firm’s successful imitations. This gives us the
    following hypotheses:

    Hypothesis 2a. As the scope of the focal firm’s
    imitation of new product technologies in-
    creases, the scope of rivals’ imitation of new
    product technologies will also increase.

    Hypothesis 2b. As the average speed of the focal
    firm’s imitation of new product technologies
    increases, the average speed of rivals’ imitation
    of new product technologies will also increase.

    Scope and Average Speed of Rivals’ Imitation of
    New Product Technologies and the Focal Firm’s
    Performance

    Various studies in the management and strategy
    literature have analyzed whether and how the

    1888 OctoberAcademy of Management Journal

    intensity of competitive rivalry affects industry
    members’ performance. A study by Young et al.
    (1996) showed that increases in the number of rival
    actions in a sample of software firms has a detri-
    mental effect on the focal firm’s performance. Simi-
    larly, Chen and Miller’s (1994) and Smith et al.’s
    (1991) analyses of competitive dynamics in the air-
    line industry showed that when rivals respond more
    strongly to earlier moves by the focal firm, perfor-
    mance of the latter will decrease. They suggested that
    the more actions rivals carry out, and the greater the
    speed of execution, the more the focal firm’s perfor-
    mance will be damaged.

    Likewise, in their analysis of Red Queen compe-
    tition, Derfus et al. (2008) showed that when the focal
    firm undertakes a new competitive action, both the
    number and speed of rival countermoves increase,
    leading to a decrease in focal firm performance.
    Overall, extant studies have pointed to broader and
    faster imitation by rivals as having a negative impact
    on focal firm performance. This gives us the follow-
    ing hypotheses:

    Hypothesis 3a. With the scope of the focal firm’s
    imitation of new product technologies held
    constant, as the scope of rivals’ imitation of new
    product technologies increases, focal firm per-
    formance decreases.

    Hypothesis 3b. With the average speed of the
    focal firm’s imitation of new product technolo-
    gies held constant, as the average speedof rivals’
    imitation of new product technologies in-
    creases, focal firm performance decreases.

    The Moderating Effect of Product Technology
    Heterogeneity in the Market

    Recent studies in the strategy and technology in-
    novation literature (Argyres, Bigelow & Nickerson,
    2015; Giachetti & Lanzolla, 2016; Madhok et al., 2010;
    Posen et al., 2013) have suggested that evolving in-
    dustry characteristics, in particular changes caused
    bythe introduction of newtechnologies, can affectthe
    level of uncertainty in the competitive environment.
    This in turn constrains the firms’ ability to learn from
    rivals, reducing the effectiveness of imitation as
    a competitive weapon. These findings are in line with
    previous work on the industry life cycle (e.g.,
    Utterback & Suarez, 1993), which has pointed out
    that as industries mature they tend to transition
    from high to low levels of product technology
    heterogeneity—where high levels of heterogeneity

    correspond to a situation in which there are more
    designs contending for consumer attention, and more
    product features that can be incorporated into prod-
    ucts. In other words, the level of product technology
    heterogeneity expresses the extent to which products
    launched by all competitors are equipped with simi-
    lar or different technologies. A low level of product
    technology heterogeneity is the result of a “high de-
    gree of design dominance,” while a high level of
    product technology heterogeneity is the product of
    a “low degree of design dominance.”

    Since high product technology heterogeneity entails
    a situation in which a clear dominant design has yet to
    emerge, often because several key technologies are
    vying for acceptance, firms in such an environment
    have to cope with technological uncertainty when it
    comes to deciding which technologies they should
    install in their products (Lippman & Rumelt, 1982;
    Makadok, 1998; Utterback & Suarez, 1993). One way
    for firms to deal with technological uncertainty is to
    observe the technologies that rivals imitated pre-
    viously. However, the information obtained from
    observing rivals’ imitation when technological un-
    certainty is high is more noisy, and hence a less re-
    liable guide for judging the merits of new product
    technologies (Posen & Levinthal, 2012). In rapidly
    changing competitive environments, as is the case in
    Red Queen competition, technological uncertainty
    can therefore slow down the learning process, con-
    strain decision making, and hence adversely affect
    performance. As Barkema et al. (2002: 921) pointed
    out, “organizations that learn slowly from competi-
    tors may find their innovation performance rapidly
    deteriorating.”

    3

    This leads us to argue that the extent to which
    a focal firm’s and rivals’ imitative actions affect the
    focal firm’s performance (Hypotheses 4a and 4b, and
    6a and 6b), and the extent to which the focal firm’s
    imitative actions trigger rivals’ imitative actions
    (Hypotheses 5a and 5b), depends on the level of
    product technology heterogeneity.

    Product technology heterogeneity: Focal firm’s
    scope and average speed of imitation and focal
    firm performance. As we noted earlier, high product
    technology heterogeneity increases imitative un-
    certainty. This means that focal firms are less certain

    3 As also remarked by Posen and Levinthal (2012) in
    their analysis of turbulent (i.e., rapidly changing) envi-
    ronments, “turbulence reduces the value of efforts to gen-
    erate new knowledge becausethelifespan of returns to new
    knowledge is reduced in a world in which change is more
    frequent” (594).

    2017 1889Giachetti, Lampel, and Li Pira

    about which product technologies they should imi-
    tate, and which they should ignore. It also means that
    the learning process for focal firms is more difficult,
    since in this uncertain scenario firms need time and
    resources to figure out which are the most effective
    technology adoption strategies. Thus, although, in
    general, we expect focal firms that are particularly
    “active” when imitating new product technologies
    (i.e., high imitation scope and speed) to stand a better
    chance of successfully differentiating their offerings
    when compared to imitating rivals that are less active,
    this prediction may not hold when product technol-
    ogy heterogeneity is high. When product heteroge-
    neity is high firms that adopt many new product
    technologies (i.e., high imitation scope), and do so
    more quickly than their rivals (i.e., high imitation
    speed) also run the risk of betting against the design
    that will subsequently gain wide market acceptance.
    Thedecisiontobetagainst afuturedominantdesignis
    likely to adversely affect the performance of the focal
    firm (Argyres, Bigelow, & Nickerson, 2015; Utterback
    & Suarez, 1993). In contrast, low product technology
    heterogeneity (i.e., high design dominance) reduces
    imitation risks, largely because it is easier to evaluate
    the merits of new product technologies sufficiently
    early to avoid making the wrong design decisions. We
    thus posit that:

    Hypothesis 4a. Product technology heterogene-
    ity negatively moderates the relationship be-
    tween the focal firm’s scope of imitation of new
    product technologies and its performance.

    Hypothesis 4b. Product technology heterogeneity
    negatively moderates the relationship between
    the focal firm’s average speed of imitation of new
    product technologies and its performance.

    Product technology heterogeneity: Focal firm’s
    scope and average speed of imitation and rivals’
    imitation response. Various studies on organizational
    learning have examined how rival firms use imitation
    whentheperformanceoutcomesoflearningfromother
    firms are uncertain. For example, Rhee, Kim and Han
    (2006: 504) pointed out that “decision makers con-
    fronting conflicting mimetic requirements and prac-
    tices find it difficult to make an imitation decision
    because conformity to one undermines the isomorphic
    support of other elements.” Likewise, Cameron (2005)
    showed that decision makers who face conflicting ex-
    ternal information reduce the attention paid to such
    data when updating their private information, and are
    thenlikelytomakestrategicdecisionsthatdeviatefrom
    industry norms. In essence, evidence has suggested

    that obstacles to processing observed information—
    caused by heterogeneous information—reduce imita-
    tion (Gaba & Terlaak, 2013).

    When product technology heterogeneity is high,
    rivals confront markets in which many product
    configurations compete. Under these conditions it is
    unclear which of these configurations will prevail
    and which will fail. Nor can rivals assume that the
    entire set of actions by the first imitators conveys
    information that is necessarily reliable and useful for
    their imitation decisions. Their best course of action
    is to keep their strategic options more open, and
    imitate with greater caution, in terms of both scope
    and speed. The aim of rivals at this point is to reduce
    the risk of betting too early on product features that
    may not become part of the future dominant design.
    This means that rivals, having observed the focal
    firm’s imitative actions, will imitate a limited num-
    ber of technologies, and do so at lower speed. At the
    industry level, this behavior leads to reduced prob-
    ability of overreaction to new product technologies
    that are introduced by earlier movers.

    Generally speaking, therefore, the technological
    uncertainty triggered by high product technology het-
    erogeneity mitigates the pressure for imitative band-
    wagons(Abrahamson& Rosenkopf,1993).4 Incontrast,
    when there is low product technology heterogeneity,
    i.e., high degree of design dominance, there is also
    lower technological uncertainty because the market
    features fewer product configurations. Rivals can
    therefore infer more accurately the moves that focal
    firms are likely to make, and hence calculate with
    greater certainty the consequences of their moves. This
    in turn encourages rivals to pursue imitative actions
    more aggressively (i.e., higher imitation scope and
    speed). This gives us the following hypotheses.

    Hypothesis 5a. Product technology heterogene-
    ity negatively moderates the relationship be-
    tween the scope of the focal firm’s imitation of
    new product technologies and the rivals’ scope
    of imitation of new product technologies.

    Hypothesis 5b. Product technology heterogene-
    ity negatively moderates the relationship be-
    tween the average speed of the focal firm’s
    imitation of new product technologies and the
    rivals’ average speed of imitation of new product
    technologies.

    4 In a similar vein, LiCalzi and Marchiori (2013) argued
    that in a dynamic environment it is more effective to focus
    on a relatively narrow set of strategic actions in order to
    track and adapt to environmental shocks accurately.

    1890 OctoberAcademy of Management Journal

    Product technology heterogeneity: Rivals’ scope
    and average speed of imitation and focal firm
    performance. When deriving Hypothesis 5, we ar-
    gued that high product technology heterogeneity
    reduces rivals’ propensity to respond to the focal
    firm with imitation. This is because, given the high
    technological uncertainty, rivals are likely to keep
    their options more open, and follow the focal firm’s
    actions only if they prove to be successful. In fact, by
    imitating first, the focal firm runs the risk of betting
    on a design that will not become dominant (Hy-
    pothesis 4), whereas rivals, by imitating later, avoid
    wasting resources by imitating only those new
    technologies (previously adopted by the focal firm)
    that have demonstrated greater acceptance by con-
    sumers. We can regard these rival firms as “second-
    mover” imitators that derive their advantage from the
    technological uncertainty of the market (Lieberman &
    Montgomery, 1998). To put this in perspective, rivals’
    imitative decisions of new product technologies (in
    terms of scope and speed) will benefit from high
    technological uncertainty at the expense of the focal
    firm’s performance because they are able to adjust
    their actions after observing the focal firm’s earlier
    moves. This leads to the following hypotheses:

    Hypothesis 6a. Product technology heterogene-
    ity negatively moderates the relationship be-
    tween the scope of the rivals’ imitation of new
    product technologies and the focal firm’s
    performance.

    Hypothesis 6b. Product technology heterogene-
    ity negatively moderates the relationship be-
    tween the average speed of the rivals’ imitation
    of new product technologies and the focal firm’s
    performance.

    Figure 1 depicts our research model, showing
    the hypothesized relationships as described
    above.

    METHOD

    Sample and Setting

    We test the proposed hypotheses in the specific
    context of the U.K. mobile phone industry. Our
    sample includes handset vendors that were operat-
    ing in the U.K. mobile phone industry from 1997 to
    2008. During this period, 48 new product technolo-
    gies were installed in 566 new mobile phones in-
    troduced and sold by the following firms: Nokia,
    Motorola, Samsung, LG, Ericsson, Sony, Sony-
    Ericsson, Siemens, Philips, Panasonic, Sagem, NEC,

    and Alcatel. These firms constituted almost the entire
    U.K. mobile handset industry. Mobile phones can
    be distinguished into two categories: (a) “regular
    phones,” or “feature phones,” offering mainly basic
    phone and multimedia functionalities, and (b)
    “smartphones,” namely handsets equipped with ad-
    vanced operating systems offering PC-like capabil-
    ities that are more expensive than regular phones and
    targeted at the high-end market. Smartphones con-
    stitute most of the U.K. market today, but were a small
    niche during the period under study. To maintain
    consistency, we decided to exclude smartphone de-
    vices from our sample. Information about product
    innovations introduced by the 13 mobile phone ven-
    dors in the U.K. market were collected from the spe-
    cialist industry magazines What Mobile, What
    CellPhone, and Total Mobile. We selected only
    producttechnologiesthatwereexplicitlyreviewedby
    these magazines over our study period.

    We believe that there are several reasons why the
    U.K. mobile phone industry over the 1997–2008 time
    period is a particularly suitable setting to test our
    hypotheses about Red Queen competitive imitation.
    First, the mobile phone industry, especially in de-
    veloped countries such as the U.K., has often been
    described as a fast-changing environment charac-
    terized by rapid new product technology in-
    troduction and quick technological obsolescence
    (Mintel International Group Limited, 1997–2008), all
    theoretical factors that underline the pressure that
    leads firms to aggressively adopt new technologies in
    order to remain competitive.

    Second, our observation period covers various
    stages of the industry’s evolution. From the mid-
    1990s to the end of the 2000s, the mobile phone
    diffusion rate (i.e., the number of handsets per 100
    habitants) grew from about 10% to a saturation level
    (over 100%), with the growth rate of diffusion par-
    ticularly high during the second half of the 1990s,
    and gradually diminishing over the 2000s.5 More-
    over, the progressive transition of handsets in the
    U.K. from niche to mass- market products encour-
    aged competitors to launch their most advanced
    models and technologies in the market, making the
    competitive environment particularly challenging.
    These factors indicate that over the 12-year period
    analyzed, the industry passed from the growth to the
    maturity stage of its life cycle. Because our data
    covers both growth and maturity, we are able to

    5 Data about mobile phone diffusion in the U.K. market
    were collected from Ofcom, the U.K. telecoms regulatory
    body.

    2017 1891Giachetti, Lampel, and Li Pira

    examine changes in the competitive interactions and
    learning processes that may occur as the technology
    environment evolves over time (Baum et al., 2000).
    This is in line with Derfus et al.’s (2008) recom-
    mendation that research on Red Queen effects
    should study empirical settings covering both early
    and late stages of the industry’s evolution.

    Third, mobile phone vendors in our sample are
    very large companies that extensively advertise their
    product innovations in a wide variety of media and
    marketing channels. This means that competitive
    actions related to product innovations are highly
    visible—which is an important condition to assume
    that imitative actions in the U.K. mobile phone in-
    dustry are taken deliberately.

    Fourth, the information we gathered from several
    secondary sources indicated that, at least at the Eu-
    ropean level, new product technologies in the mo-
    bile phone industry were introduced in more or less
    the same year across all European countries.6 This
    makes the U.K. a representative sample of the Euro-
    pean market.

    Fifth, smartphone devices were a small market
    category prior to the introduction of Apple’s iPhone
    and its operating system iOS in mid-2007, and the
    launch of Google’s Android operating system in
    2008. The introduction of these product innova-
    tions triggered the rapid market decline of mobile
    phones that did not use advanced operating systems
    (Giachetti & Marchi, 2017). To ensure consistency in
    our analysis, we decided to consider only mobile
    phone technologies introduced before 2008.

    New Product Technologies, Technological
    Systems, and Imitation

    Our study focuses on drivers and performance
    outcomes of new product technology imitations by
    U.K. mobile phone companies. We define a product
    technology as any hardware or software that allows
    the handset to perform a certain function. We assume
    that a “new product technology imitation” occurs
    after a new product technology is introduced for the
    first time in the U.K. market by a “technology pio-
    neer,” or “innovator.” A firm is coded as an “imita-
    tor” when it adopts for the first time in one of its new
    handset models the technology previously in-
    troduced by the pioneer. In our analysis, we want to
    consider only the imitation of new product technol-
    ogies, namely those technologies only recently in-
    troduced and not widely adopted by competitors.
    We consider a product technology to be widely
    adopted by industry members if it has been installed
    in more than 50% of all products launched in the

    FIGURE

    1

    Research Model

    Product technology
    heterogeneity in the
    market

    Rivals’ scope and
    average speed of
    imitation of new
    product technologies

    Focal firm’s scope and
    average speed of
    imitation of new
    product technologies

    Focal firm
    performance

    H5a/b (-)

    H2a/b (+)
    H4a/b (-)

    H6a/b (-)

    H3a/b (-)

    Time t Time t + 1 Time t + 2

    Temporal sequence of actions and performance outcomes

    H1a/b (+)

    6 The secondary sources from which we gathered in-
    formation about the timing of new product technologies
    introduction were: (a) the FACTIVA database, which
    searches thousands of media sources at the worldwide
    level; (b) the mobile phone vendors’ annual reports and
    newsletters; (c) various online catalogs for handsets, such
    as the GSMArena website (http://www.gsmarena.com);
    (d) books, newspapers, press releases, and business
    publications.

    1892 OctoberAcademy of Management Journal

    http://www.gsmarena.com

    market. Above this level of adoption, imitation of the
    technology is no longer motivated by direct rivalry,
    but by recognition that consumers now see these
    features as intrinsic to the basic design and thus will
    not purchase handsets that lack these features. In
    total, we observed about 600 imitative actions by
    firms that fit this criterion.

    Since technologies may evolve over time, we fol-
    low the suggestion of Giachetti and Dagnino (2017)
    and analyze new product technology imitation by
    considering both the first version of a technology
    introduced in the market, and successive improve-
    ments. A list and description of the sampled product
    technologies is presented in Appendix A.

    It is important to bear in mind that handsets com-
    pete by offering consumers functionalities that are
    made possible by product technologies. In some in-
    stances, similar functionalities may be offered by
    different product technologies. Following the work
    on complex systems of Murmann and Frenken
    (2006), we define a “technological system” as a
    group of technologies that allow the product to per-
    form functions of a certain type. For example, in
    mobile phones infrared, Bluetooth, and USB ports
    are technologies that enable connectivity between
    devices, and thus belong to the same technological
    system. We grouped the 48 technologies into seven
    technological systems: networking, high-speed data
    transfer, phone call, connectivity, messaging, dis-
    play, and technological convergence (see AppendixA,
    Table A1).

    As can be expected, we found innovation and
    imitation in all the technologies in our sample.
    However, when we examined the frequency of both,
    we also found that over the analyzed time period, the
    average number of new product technologies in-
    troduced every year—i.e., innovations—was much
    lower than the average number of imitations (see
    Figure A1 in Appendix A). This finding corroborates
    what has been noted by previous studies: imitation is
    far more pervasive than innovation. Thus, firms may
    forgo the risks of innovative moves, but they cannot
    avoid imitation without suffering erosion of their
    market position (Lee et al., 2000; Levitt, 1966). It is
    also interesting to note that the average number of
    imitations rapidly increased until 2003, but started
    decreasing from 2004, and the average number of
    innovations was relatively high until 2003, declined
    in 2004, and then leveled off from then on. The main
    reason for this decline of innovations and imitations
    was the shift in the locus of technological innovation
    to smartphone devices. The regular phone market at
    this point in time entered a period of greater emphasis

    on price competition, with consequent decline in the
    rates of innovation and imitation.

    Measures

    Dependent and independent variables. Depend-
    ing on the relationship modeled in the proposed Red
    Queen competitive imitation cycle (Figure 1), we rely
    on a different set of dependent and independent var-
    iables. We assume that the focal firm’s imitative ac-
    tions at a certain time, t, trigger rivals’ response in the
    following time, t 1 1, and both the focal firm’s imita-
    tive action and rivals’ response will affect the focal
    firm’s performance at time t 1 2, as illustrated in
    Figure 1. Setting dependent and independent vari-
    ables in a logical temporal sequence is important to
    make realistic assumptions about the fact that ac-
    tions and reactions are deliberate, and take some
    time before having an effect on performance.7 De-
    pendent and independent variables are described as
    follows.

    Consistent with the extant literature (Derfus et al.,
    2008), we defined and measured the scope of a firm’s
    imitation as the total number of new product tech-
    nologies (belonging to a specific technological sys-
    tem) imitated by the focal firm within the year t.

    We measured the average speed of focal firm’s
    imitation as the average time it takes for the focal
    firm to imitate new product technologies related to
    a specific technological system. Essentially, we
    wanted to capture the speed of imitation of those
    new product technologies used to operationalize
    the imitation scope. To do this, we first computed
    the time to imitation, in months, per each of the
    technologies imitated by the firm in year t. Second,
    we normalized this latter value by dividing it by the
    maximum imitation time for that technology in the
    sample, so as to transform the variable from count to
    ratio. Third, we computed the mean of the firm’s
    imitation timing of technologies belonging to the
    technological system i (avtimei,t). We finally oper-
    ationalized the average speed of the focal firm’s
    imitation (ASi,t), as in Equation 1. The resulting
    measure ranges from 0 to 1; the greater its value
    (i.e., closer to 1) the higher the focal firm’s imitation
    speed.

    7 Since the variable rivals’ imitative response was com-
    puted at time t 1 1 and the variable focal firm performance
    was computed at time t 1 2, our empirical analysis cap-
    tures imitative actions between 1997 and 2007, and firm
    performance from 1999 and 2008.

    2017 1893Giachetti, Lampel, and Li Pira

    ASi,t 5 1 2

    avtimei,t


    (1)

    It is worth noting at this point that a higher average
    speed of imitation does not entail higher imitation
    scope. In fact, two focal firms may have the same
    score for average imitation speed but imitate a dif-
    ferent number of technologies. Moreover, if one firm
    increases the number of technologies imitated from
    one year to another (i.e., wider scope), this might
    result in either higher or lower average speed with
    respect to the previous year (e.g., “lower speed” if the
    firm imitates a “higher number” of technologies, but
    “more slowly”).

    We operationalized the scope of a rivals’ imitation
    by subtracting the total number of imitations realized
    by the focal firm at time t 1 1 from the total number of
    imitative actions taken by all competitors at the same
    time, t 1 1 (in a focal technological system). In this
    way, we accounted only for those imitative actions
    subsequent to the focal firm’s imitative actions.

    As in other competitive dynamics research
    (e.g., Ferrier et al., 1999; Young et al., 1996), we used
    rivals’ imitation speed as a measure of the average
    length of time it took rivals to act after a new product
    technology was introduced. Following the procedure
    outlined by Derfus et al. (2008), we calculated this
    measure by taking the mean of the average speed of
    imitation of all of the focal firm’s rivals at a certain
    time, t 1 1. The resulting measure ranges from 0 to 1,
    with the higher imitation speed for values closer to 1.

    Focal firm performance was operationalized us-
    ing the number of handsets sold on a yearly basis
    (i.e., sales performance) in the U.K. This measure of
    firm performance has been widely used by mobile
    phone industry specialists such as Gartner Data-
    quest and Mintel International Group Limited. Data
    on handsets sold per vendor were collected from
    Mintel International Group Limited (1997–2008),
    Euromonitor International (2003–2008), and firms’
    archival data.

    We operationalized the measure of product tech-
    nology heterogeneity using the Shannon entropy
    index (Shannon, 1948). This entropy measure is
    suitable for our research setting because it captures
    the extent to which products differ in terms of tech-
    nologies that belong to a given technological system.
    A uniform distribution of the type of technologies
    products are equipped with reflects a situation in
    which firms produce a wide variety of designs, while
    a skewed distribution represents a situation in which
    there are minor differences between firms’ choice of
    design. As such, theindex can be used as an indicator
    of technological heterogeneity (Frenken, Saviotti, &

    Trommetter, 1999), a situation in which products
    offered by industry rivals widely differ in terms of the
    technologies they are equipped with. The Shannon
    entropy value of a technological system is given by
    Equation 2:

    Hi,t 5 2 +
    S

    k 5 1
    ln

    pk,t


    3 pk,t (2)

    Where Hi,t is the level of product technology het-
    erogeneity within technological system i at year t, pk,t
    is the percentage of products (introduced in year t)
    equipped with technology k (therefore 0 # pk # 1),
    and S is the number of technologies introduced and
    related to technological system i.

    The Shannon entropy index (Hi,t) is equal to zero
    when all products introduced at time t in the market
    are equipped with the same set of technologies re-
    lated to technological system i. This means that there
    is a dominant design in terms of the set of technolo-
    gies related to i. In this extreme case, pk,t would be
    equal to 1, which implies that the entropy of the
    product population equals zero:

    Hi,t 5 2lnð1Þ 3 1 5 0 (3)
    Entropy is positive otherwise, and the larger its

    value, the larger the variety in the population. Spe-
    cifically, the larger the value of Hi,t, (a) the higher the
    number of technologies in the technological system,
    and (b) the lower the diffusion of these technologies
    among existing products.

    Control variables. We also included various
    control variables (those related to the focal firm and
    at the industry level are computed at year t, those
    related to rivals are computed at year t 1 1), poten-
    tially affecting all firms’ action and performance:

    Although we are analyzing competitive dynamics
    that are triggered by imitative efforts, we had to
    control for imitation that occurs as a response to in-
    novations introduced into the technological system,
    or what we call innovation scope. This is in line with
    first-mover advantage literature, which has sug-
    gested that innovators’ monopoly profits will attract
    imitative entrants (Lieberman & Montgomery, 1988;
    Markides & Geroski, 2004). This variable was mea-
    sured as a count of new product technologies in-
    troduced by the focal firm in year t.

    Similar to how we measured the focal firm’s in-
    novation scope, we measured rivals’ innovation
    scope as a count of the new product technologies
    introduced by rivals in the year t 1 1.

    Studies of the Red Queen effect have argued that
    a firm’s relative size can influence its performance,

    1894 OctoberAcademy of Management Journal

    and rivals’ responseto its actions (Derfus et al., 2008).
    Relative market position was measured with a
    dummy variable that set the value as 1 if the level of
    sales of the firm in the year t was above the industry
    median, and 0 otherwise.

    Mobile phone vendors may follow different strat-
    egies depending on the time of year in which they
    introduce the largest number of new product models.
    We controlled for this strategic decision with a set of
    dummy variables that equaled 1 during the quarter
    when the firm introduced the largest number of new
    product models during the year t, and 0 otherwise.

    Researchinindustrialorganizationandstrategyhas
    shown that industry concentration can influence the
    intensity of competition (Derfus et al., 2008). In an
    industrywith highbarriersto entry, suchasthemobile
    phone industry, a higher level of industry concentra-
    tion usually results in a lower level of competition
    intensity, because rivals with the largest market share
    are more likely to collude on their marketing strategies
    (Waldman&Jensen,2012;Wiggins&Ruefli,2005).We
    therefore controlled for industry concentration by us-
    ing the cumulative market share of the four largest
    U.K. handset vendors as a measure.

    A three-year standard deviation of the U.K. gross
    domestic product (GDP volatility) was used to ac-
    count for the country’s macroeconomic uncertainty
    (Haddow, Hare, Hooley, & Shakir, 2013).

    RESULTS

    Hypotheses Testing

    Table 1 reports the variables’ descriptive statistics,
    while Tables 2–3 report results of the regression
    analysis. We tested the hypotheses with three re-
    gression models: (1) a robust fixed-effects regression
    when the dependent variable was the focal firm
    performance (Table 2); (2) a robust fixed-effects re-
    gression when the dependent variable was rivals’
    average speed of imitation (Table 3); (3) a robust
    fixed-effects Poisson regression when the dependent
    variable was rivals’ imitation scope, a count-type
    variable (Cameron & Trivedi, 2009) (Table 3). A
    Hausman test suggested that the use of fixed-effects
    was preferable over random-effects. Since not all
    technologies were adopted by all sampled firms, and
    not all firms were active in the U.K. market over the
    entire time period analyzed, we ended up with a
    566-observation unbalanced panel.

    Models 1–3 in Table 2 report the results for re-
    gressions relating focal firm imitation scope and
    speed, and rivals imitation scope and speed, to firm

    performance(Hypotheses1,3,4,and6).Theregression
    results that examine the impact of focal firm imitation
    scope and speed on rivals’ imitation scope and speed,
    respectively, are presented in Table 3, Models 4–9
    (Hypotheses 2 and 5). We calculated variance inflation
    factors (VIFs) to determine whether there was multi-
    collinearity in the analyses. The average VIF scores
    were all below 1.4, and no individual VIF was greater
    than 2.08, thereby all were lower than the recom-
    mended threshold of 10 (Chatterjee & Hadi, 2006).8

    Before we turn to a discussion of the coefficients of
    independent variables and moderators related to the
    presented hypotheses, we briefly examine the co-
    efficients of the control variables in the full Models 3
    (Table 2), 6, and 9 (Table 3). We found the impact of
    innovation scope on focal firm performance, as
    shown in Model 3 (Table 2), in terms of both focal
    firm innovation scope (b 5 0.00, p . .1) and rivals’
    innovation scope (b 5 20.00, p . .1), not to be sig-
    nificant. With regard to the impact of innovation
    scope on imitative actions, as shown in Models 6 and
    9 (Table 3), we found that the only significant re-
    lationship is that between rivals’ innovation scope
    and rivals’ imitation scope (Model 6: b 5 0.26,
    p , .01), showing that rivals that innovate more are
    also those that imitate more.9 We also found that the
    control variable relative market position has a sig-
    nificant effect only on focal firm performance as
    shown in Model 3 (b 5 0.27, p , .01). As for industry-
    level controls, industry concentration has a negative
    and significant effect on rivals’ average speed of
    imitation (Model 9: b 5 20.33, p , .01), while GDP
    volatility has a positive effect on focal firm perfor-
    mance (Model 3: b 5 0.05, p , .01) and a negative

    8 As can be observed in the correlation matrix presented
    in Table 1, the greatest correlation coefficient is that be-
    tween focal firm’s imitation scope and speed (r 5 0.63;
    p , .01), two key independent variables in our regression
    model. In the regression models, the maximum VIFs for
    these two variables were 2.08 and 1.79, respectively.

    9 The positive association between firm’s imitation
    scope and innovation scope can also be observed in the
    correlation matrix presented in Table 1; the correlation
    coefficients between rivals’ imitation scope and rivals’
    innovation scope (r 5 0.35, p , .01) and between focal
    firm’s imitation scope and focal firm’s innovation scope
    (r 5 0.08, p , .1) are both positive and significant. We
    believe the explanation for this is that firms with greater
    resources and the capabilities needed to imitate several
    technologies also have greater resources and capabilities to
    introduce several technologies that are new to the market,
    and vice versa.

    2017 1895Giachetti, Lampel, and Li Pira

    T
    A
    B
    L
    E
    1

    D
    es
    cr
    ip
    ti
    v
    e
    S
    t

    a
    ti

    st
    ic
    s

    M
    ea

    n
    S
    D

    1
    2

    3

    4

    5

    6

    7

    8

    9

    1
    0

    1
    1

    1

    F
    o
    ca

    l
    fi
    rm

    ’s
    p

    er
    fo

    rm

    an
    ce

    (t
    1

    2)
    a

    1
    7
    9
    7
    .2
    1

    2
    2
    7
    5
    .3
    1

    1
    .0
    0

    2
    F
    o
    ca
    l
    fi
    rm

    ’s
    im

    it
    at
    i

    o
    n

    sc
    o
    p
    e

    (t
    )

    0
    .5
    1

    0
    .8
    2

    0
    .0
    8
    1

    1
    .0
    0

    3
    F
    o
    ca
    l
    fi
    rm

    ’s
    av

    er
    ag

    e
    sp

    ee
    d
    o
    f

    im

    it
    at

    io
    n

    (t
    )

    0
    .1
    5

    0
    .2
    6

    0
    .1
    7
    *
    *

    0
    .6
    3
    *
    *

    1
    .0
    0

    4
    R
    iv
    al
    s’
    im

    it
    at
    io
    n
    sc
    o
    p
    e

    (t
    1

    1
    )

    4
    .7
    2

    4
    .4
    1

    2

    0
    .0
    6

    0
    .3
    0

    *
    *

    0
    .2
    0
    *
    *

    1
    .0
    0

    5
    R
    iv
    al
    s’
    av

    er
    ag
    e
    sp
    ee
    d
    o
    f

    im
    it

    at
    io
    n
    (t
    1

    1
    )

    0
    .3
    5

    0
    .2
    4

    2
    0
    .1
    7
    *
    *

    2

    0
    .0
    7

    1

    0
    .0
    4

    0
    .0
    9
    *

    1
    .0
    0

    6
    P

    ro
    d

    u
    ct

    te
    ch

    n
    o

    lo
    gy

    h
    et
    er
    o
    ge

    n
    ei
    ty

    (t
    )

    1
    .4
    6

    0
    .8
    7

    0
    .1
    4
    *
    *

    0
    .4
    4
    *
    *

    0
    .2
    0
    *
    *

    0
    .4
    1
    *
    *

    2
    0
    .2
    9
    *
    *

    1
    .0
    0

    7
    F
    o
    ca
    l
    fi
    rm

    ’s
    in
    n
    o
    v
    at
    io
    n

    sc
    o
    p
    e
    (t
    )

    0
    .0
    7
    0
    .3
    0
    0
    .0
    8
    1
    0
    .0
    8
    1
    0
    .0
    9
    *

    0
    .1
    1
    *
    *

    0
    .0
    7
    0
    .0
    6
    1
    .0
    0

    8
    R
    iv
    al
    s’
    in
    n
    o
    v
    at
    io
    n
    sc
    o
    p
    e

    (t
    1
    1
    )

    0
    .5
    7

    0
    .8
    0

    2

    0
    .1
    0

    *

    2

    0
    .0
    5

    2

    0
    .0
    3

    0
    .3
    5
    *
    *

    0
    .1
    9
    *
    *

    2
    0
    .0
    9
    *

    2
    0
    .0
    0

    1
    .0
    0

    9
    F
    o
    ca
    l
    fi
    rm

    ’s
    re
    la
    ti
    v
    e

    m
    ar
    k
    et

    p
    o
    si
    ti
    o
    n
    (t
    )

    0
    .5
    0

    0
    .5
    0

    0
    .6
    1
    *
    *

    0
    .0
    7
    0
    .1
    7
    *
    *

    2

    0
    .0
    2

    2
    0
    .0
    5

    0
    .0
    3
    0
    .0
    5

    2
    0
    .0
    3

    1
    .0
    0
    1
    0

    I

    n
    d

    u
    st
    ry

    co
    n
    ce
    n
    tr
    at
    io
    n
    (t
    )

    0
    .7
    8

    0
    .0
    5
    0
    .0
    8
    1
    0
    .0
    2

    2
    0
    .0
    4

    0
    .0
    1

    2
    0
    .2
    5
    *
    *

    0
    .1
    0
    *

    2
    0
    .0
    1

    2
    0
    .0
    0
    0
    .0
    8
    1
    1
    .0
    0
    1
    1

    G
    D
    P
    v
    o
    la
    ti
    li
    ty

    (t
    )b

    1
    0
    5
    7
    3
    .8
    5

    2
    6
    8
    3
    .7
    7

    2
    0
    .0
    8
    1

    2
    0
    .2
    2
    *
    *

    2
    0
    .0
    9
    *

    2

    0
    .1
    2
    *
    *

    0
    .1
    2
    *
    *

    2
    0
    .3
    5
    *
    *

    2
    0
    .0
    2

    0
    .0
    6
    2
    0
    .0
    5

    2
    0
    .4
    9
    *
    *

    1
    .0
    0

    N
    o
    te
    :n

    5
    5
    6
    6

    a
    U
    n
    it
    s
    so
    ld

    ar
    e
    ex

    p
    re
    ss
    ed

    in
    th
    o
    u
    sa
    n
    d
    s.

    b
    G
    D
    P
    v
    o
    la
    ti
    li
    ty

    is
    co

    m
    p
    u

    te
    d

    o
    n
    G
    D
    P
    v
    al
    u
    es

    in
    m
    il
    li
    o
    n
    s
    o
    f
    p
    o
    u
    n
    d
    s.

    1
    p
    ,

    0
    .1
    0

    *
    p
    ,

    0
    .0
    5

    *
    *
    p
    ,

    0
    .0
    1

    1896 OctoberAcademy of Management Journal

    effect on rivals’ imitative actions (Model 6: b 5
    20.07, p , .05; Model 9: b 5 20.23, p , .01).

    We now turn our attention to the hypotheses tests.
    Hypotheses 1a and 1b state that focal firm imitation
    scope and average speed of imitation both have
    a positive effect on its performance. As shown in
    Model 3 (Table 2), while the sign and significance of

    focal firm average speed of imitation is in line with
    our prediction (b 5 0.07, p , .05), focal firm imita-
    tion scope is significant with the opposite sign (b 5
    20.08, p , .05). Therefore, Hypothesis 1b is sup-
    ported while Hypothesis 1a is not.

    Hypothesis 2a states that as the scope of the firm’s
    imitation of new product technologies increases, the

    TABLE 2
    Robust Fixed-effects Regression Analysis: Focal Firm and Rivals’ Imitative Actions on the Focal Firm Performance

    Model 1 Model 2 Model 3

    Hypothesis

    Focal firm’s

    performance (t 1 2)

    Focal firm’s

    performance (t 1 2)
    Focal firm’s
    performance (t 1 2)

    Constant 20.10** 20.05* 20.06**
    (–5.23) (–2.47) (–2.76)

    Independent variables
    Focal firm’s imitation scope (t) 1a 20.041 20.08*

    (–1.86) (–2.43)
    Focal firm’s average speed of imitation (t) 1b 0.05* 0.07*

    (2.00) (2.50)
    Rivals’ imitation scope (t 1 1) 3a 20.05** 20.04*

    (–2.76) (–2.39)
    Rivals’ average speed of imitation (t 1 1) 3b 20.041 20.051

    (–1.98) (–1.95)
    Product technology heterogeneity (t) 0.12* 0.12**

    (2.62) (2.99)
    Interactions
    Focal firm’s imitation scope 3 Product
    technology heterogeneity

    4a 0.05*
    (2.03)

    Focal firm’s average speed of imitation 3
    Product technology heterogeneity

    4b 0.00
    (0.18)

    Rivals’ imitation scope 3 Product
    technology heterogeneity

    6a 20.04*
    (–2.31)

    Rivals’ average speed of imitation 3
    Product technology heterogeneity

    6b 20.00
    (–0.12)

    Controls
    Focal firm’s innovation scope (t) 20.00 0.00 0.00

    (–0.28) (0.17) (0.16)
    Rivals’ innovation scope (t 1 1) 20.04* 0.00 20.00

    (–2.26) (0.08) (–0.04)
    Relative market position (t) 0.30** 0.28** 0.27**

    (5.04) (5.40) (5.32)
    Industry concentration (t) 20.01 20.01 20.02

    (–0.55) (–0.56) (–0.92)
    GDP volatility (t) 0.04** 0.06** 0.05**

    (2.70) (2.78) (2.65)
    2nd quarter year t (largest new product
    launch)

    0.20** 0.15** 0.15**
    (4.96) (3.84) (3.86)

    3rd quarter year t (largest new product
    launch)

    0.12* 0.09* 0.10*
    (2.61) (2.05) (2.17)

    4th quarter year t (largest new product
    launch)

    0.03 0.01 0.01
    (0.69) (0.15) (0.36)

    n 566 566 566
    Within R-squared 0.24 0.30 0.31

    Notes: Estimates are based on standardized variables; t-statistics in parentheses.

    1p , 0.10
    *p , 0.05

    **p , 0.01

    2017 1897Giachetti, Lampel, and Li Pira

    scope of rivals’ imitation of new product technolo-
    gies will also increase. Hypothesis 2b states that as
    the average speed of the firm’s imitation of new

    product technologies increases, the average speed of
    rivals’ imitation of new product technologies will
    also increase. As can be observed from Table 3, in

    TABLE 3
    Robust Fixed-effects Regression Analysis: Focal Firm Imitative Actions on Rivals’ Imitative Actions

    Robust fixed-effects Poisson Robust fixed effects

    Model 4 Model 5 Model 6 Model 7 Model 8 Model 9

    Hypothesis

    Rivals’
    imitation

    scope(t 1 1)

    Rivals’
    imitation
    scope(t 1 1)
    Rivals’
    imitation
    scope(t 1 1)

    Rivals’ average
    speed of

    imitation(t1 1)

    Rivals’ average
    speed of

    imitation (t11)

    Rivals’ average
    speed of

    imitation(t11)

    Constant 0.25** 0.05 0.05
    (4.52) (0.79) (0.88)

    Independent
    variables
    Focal firm’s
    imitation scope (t)

    2a 0.081 0.15** 0.04 0.05
    (1.76) (3.11) (0.87) (0.93)

    Focal firm’s average
    speed of imitation (t)

    2b 0.06 0.02 0.02 0.02
    (1.61) (0.67) (0.52) (0.42)

    Product technology
    heterogeneity (t)

    0.23** 0.23** 20.62** 20.61**
    (4.30) (4.53) (–14.78) (–14.19)

    Interactions
    Focal firm’s imitation
    scope 3 Product
    technology
    heterogeneity

    5a –0.09*
    (–2.38)

    Focal firm’s average
    speed of imitation 3

    Product technology
    heterogeneity

    5b –0.02
    (–0.72)

    Controls
    Focal firm’s
    innovation scope (t)

    0.02 0.02 0.02 0.05 0.02 0.02
    (0.52) (0.63) (0.64) (1.18) (0.44) (0.45)

    Rivals’ innovation
    scope (t 1 1)

    0.18** 0.26** 0.26** 0.09* 20.03 20.02
    (5.26) (6.81) (7.15) (2.03) (–0.56) (–0.54)

    Relative market
    position (t)

    20.05 20.07 20.05 20.07 20.01 20.01
    (–1.11) (–1.46) (–1.13) (–0.88) (–0.22) (–0.20)

    Industry
    concentration (t)

    20.07 0.00 20.00 20.28** 20.33** 20.33**
    (–1.40) (0.07) (–0.10) (–7.53) (–10.15) (–10.14)

    GDP volatility (t) 20.17** 20.061 20.07* 20.07 20.23** 20.23**
    (–6.44) (–1.94) (–2.16) (–1.45) (–5.60) (–5.60)

    2nd quarter year t
    (largest new
    product launch)

    20.03 20.12 20.14 20.32** 20.09 20.09
    (–0.31) (–1.24) (–1.42) (–3.06) (–0.85) (–0.87)

    3rd quarter year t
    (largest new
    product launch)

    20.04 20.06 20.08 20.27* 20.14 20.14
    (–0.39) (–0.58) (–0.81) (–2.24) (–1.30) (–1.32)

    4th quarter year t
    (largest new
    product launch)

    0.11 0.06 0.04 20.19 1 20.03 20.04
    (1.25) (0.63) (0.38) (–1.71) (–0.32) (–0.34)

    n 566 566 566 566 566 566
    Within R-squared 0.09 0.23 0.23
    Wald x2 86.25 113.00 132.65

    Notes: Estimates are based on standardized variables; in parentheses are reported t-statistics for robust fixed-effects and z-statistics for robust
    fixed-effects Poisson; coefficients in bold are those related to the tested hypotheses.

    1p , 0.10
    *p , 0.05
    **p , 0.01

    1898 OctoberAcademy of Management Journal

    Model 6 the relationship between the scope of the
    focal firm’s imitation and the scope of rivals’ imita-
    tion is positive and significant (b 5 0.15, p , .01),
    while in Model 9 the relationship between the av-
    erage speed of the focal firm’s imitation and the av-
    erage speed of rivals’ imitation is positive but not
    significant (b 5 0.02, p . .1). Therefore, Hypothesis
    2a is supported, while Hypothesis 2b is not.

    Hypothesis 3a states that with the scope of the fo-
    cal firm’s imitation of new product technologies held
    constant, as the scope of rivals’ imitation of new
    product technologies increases, focal firm perfor-
    mance decreases. Hypothesis 3b leads us to expect
    that holding the average speed of the focal firm’s
    imitation of new product technologies constant, we
    can observe decreasing focal firm performance as the
    average speed of rivals’ imitation of new product
    technologies increases. As seen in Model 3 (Table 2),
    the coefficient of scope and average speed of rivals’
    imitation are both negative and significant (b 5
    20.04, p , .05; b 5 20.05, p , .1), thus supporting
    both Hypotheses 3a and 3b.

    Hypothesis 4a states that product technology het-
    erogeneity negatively moderates the relationship
    between the focal firm’s scope of imitation of new
    product technologies and its performance. Hypoth-
    esis 4b states that product technology heterogeneity
    negatively moderates the relationship between the
    focal firm’s average speed of imitation of new prod-
    uct technologies and its performance. As shown in
    Model 3, the coefficient of the interaction between
    focal firm imitation scope and product technology
    heterogeneity is positive and significant (b 5 0.05,
    p , .05), and the coefficient ofthe interaction between
    focal firm imitation speed and product technology
    heterogeneity is not significant (b 5 0.00, p . .1).
    Hypotheses 4a and 4b are thus not supported.

    Hypothesis 5a predicts that product technology
    heterogeneity negatively moderates the relationship
    between the scope of the firm’s imitation of new
    product technologies and the rivals’ scope of imita-
    tion of new product technologies. Hypothesis 5b
    predicts that product technology heterogeneity neg-
    atively moderates the relationship between the av-
    erage speed of the firm’s imitation of new product
    technologies and the rivals’ average speed of imita-
    tion of new product technologies. As shown in
    Model 6, the coefficient of the interaction between
    the focal firm’s imitation scope and product tech-
    nology heterogeneity is negative and significant, as
    expected (b 5 20.09, p , .05), while in Model 9
    the coefficient of the interaction between the focal
    firm’s imitation speed and product technology

    heterogeneity is negative but not significant (b 5
    20.02, p . .1). Hypothesis 5a is thus supported while
    Hypothesis 5b is not.

    With Hypothesis 6a, we predict that product
    technology heterogeneity negatively moderates
    the relationship between the scope of the rivals’
    imitation of new product technologies and focal
    firm performance. With Hypothesis 6b, we predict
    that product technology heterogeneity negatively
    moderates the relationship between the average
    speed of the rivals’ imitation of new product
    technologies and focal firm performance. As
    shown in Model 3, the coefficient of the interaction
    between rivals’ imitation scope and product tech-
    nology heterogeneity is negative and significant, as
    expected (b 5 20.04, p , .05), while the coefficient
    of the interaction between rivals’ imitation speed
    and product technology heterogeneity is negative
    but not significant (b 5 20.00, p . .1). Hypothesis
    6a is thus supported, whereas Hypothesis 6b is not.

    Table 4 offers a summary of the predicted hy-
    potheses and those that were supported by the em-
    pirical analysis. As can be observed, the Red Queen
    competitive imitation cycle (Hypotheses 1–3) is
    supported for at least one type of imitative action in
    all time frames. In the discussion section, we will
    present the plots of interaction effects and extend the
    interpretation of these findings.

    Robustness Tests

    We tested the robustness of our findings in several
    ways. First, we examined an alternative explana-
    tion to the one advanced in Hypotheses 1a and 1b.
    Specifically, if imitation scope rises, new product
    development costs will escalate, which in turn will
    lead to negative performance consequences. By the
    same token, as firms increase their imitation speed
    to catch up with their rivals, they have less time to
    adequately assess market response, and this in turn
    is likely to have negative performance conse-
    quences. Under both scenarios, we should expect
    an inverted U-shaped relationship between both
    types of imitative action and firm performance. To
    test these alternative predictions, we repeated the
    regression analysis by adding the squared term of
    both focal firm imitation scope and average speed of
    imitation. We did not find the squared terms to be
    significant.

    Second, we examined the robustness of our results
    in light of the fact that the dependent variables—
    average speed of imitation and imitation scope—are
    left–and right-censored, respectively. Average speed

    2017 1899Giachetti, Lampel, and Li Pira

    of imitation is left-censored because it is a ratio that
    cannot be less than zero, and may take the value of
    zero both when a firm has minimum average imita-
    tion speed and when the firm is not imitating any
    technology. Imitation scope is right-censored be-
    cause the number of technological attributes avail-
    able to be imitated has an upper limit. We therefore
    tested the full Models 6 and 9 (Table 3) using alter-
    native models that took into account the censored
    nature of both dependent variables. More specifi-
    cally, we repeated Models 6 and 9 using a Tobit
    fixed-effects regression based on the Honoré (1992)
    estimator with an absolute error loss function. This
    estimator was chosen because there is no conditional
    fixed-effects Tobit model, and the unconditional
    fixed-effects Tobit model is biased (Honoré, 1992).
    As shown in Table 5, Models 10 and 11, even with
    this alternative technique, results are consistent with
    those presented in Table 3.

    Third, since the regression equations in Models 6
    and 9 rely on the same set of independent variables,
    in order to account for potential correlations of the
    random error components of the two equations, we
    ran Models 6 and 9 using the seemingly unrelated
    regressions technique (Zellner, 1962). This method
    involves estimating separate equations for rivals’
    speed and scope of imitation while recognizing re-
    lationships across the two actions. As shown in
    Table 5, Models 12 and 13, results are consistent with
    those in Models 6 and 9, with the exception of

    Hypothesis 5a (which presents the expected sign, but
    is not significant).

    DISCUSSION

    Implications

    This study aims to expand our understanding of
    competitive dynamics in technology-intensive in-
    dustries with the lens of Red Queen competition. We
    do this by bringing together relevant research from
    competitive dynamics, imitation, and technology in-
    novation literature. The more recent Red Queen lit-
    erature has analyzed the conditions under which
    competitive actions increase firm performance and
    trigger rivals’ response (Derfus et al., 2008), but has
    not paid sufficient attention to (a) the analysis of Red
    Queencompetitivedynamicsintechnology-intensive
    industries, (b) the role of different types of imitative
    actions in sustaining and triggering the Red Queen
    cycle, and (c) how changes in the technological en-
    vironment moderate the Red Queen cycle. To address
    these gaps, we have developed a model of Red Queen
    competition in which the scope and speed of imita-
    tion of new product technologies is the result of
    competitive threats by rivals’ imitative actions. The
    competitive race predicted by our theory of Red
    Queen competitive imitation implies that firms
    struggle to (a) learn which technologies are, and will
    be, successful in the market, (b) imitate new product

    TABLE 4
    Predicted Hypotheses and Obtained Findings

    Obtained findingsa

    Hypotheses Predicted relationship
    Imitation scope
    (Hypotheses a)

    Average imitation
    speed (Hypotheses b)

    1 Positive effect of focal firm’s imitative actions on its
    performance

    Negativeb Positive

    2 Positive effect of focal firm’s imitative actions on
    rivals’ imitative actions

    Positive Not significant

    3 Negative effect of rivals’ imitative actions on focal
    firm’s performance

    Negative Negative

    4 Negative moderating effect of product technology
    heterogeneity on the relationship between focal
    firm’s imitative actions and its performance

    Positive Not significant

    5 Negative moderating effect of product technology
    heterogeneity on the relationship between focal
    firm’s imitative actions and rivals’ imitative actions

    Negative Not significant

    6 Negative moderating effect of product technology
    heterogeneity on the relationship between rivals’
    imitative actions and focal firm’s performance

    Negative Not significant

    a Relationships supported by the empirical analysis are in bold.
    b Positive for high levels of product technology heterogeneity (Figure 2).

    1900 OctoberAcademy of Management Journal

    technologies to maintain competitive parity with ri-
    vals, and thus (c) adapt to the evolving technological
    environment. The analysis of this self-reinforcing
    competitive mechanism enables us to shed light on
    the positive and negative aspects of different types of
    imitative action, and to clarify the relative importance
    of these aspects with regard to firm performance.

    Our first result shows that focal firms’ average
    speed of imitation positively affects their sales per-
    formance (Hypothesis 1b) while, contrary to our
    prediction, focal firms’ imitation scope has a detri-
    mental effect on performance (Hypotheses 1a). Re-
    sults of speed are consistent with previous findings
    of the competitive dynamics literature (D’Aveni,

    TABLE 5
    Tobit Fixed-effects Regression and Seemingly Unrelated Regression Analysis: Focal Firm Imitative Actions on Rivals’

    Imitative Actions

    Tobit Fixed effects Seemingly Unrelated Regressiona

    Model 10 Model 11 Model 12 Model 13

    Rivals’ imitation
    scope (t 1 1)

    Rivals’ average speed of
    imitation (t 1 1)

    Rivals’ imitation
    scope (t 1 1)
    Rivals’ average speed of
    imitation (t 1 1)

    Constant 20.13 20.31
    (–0.47) (–1.05)

    Independent variables
    Focal firm’s imitation scope (t) H2a 0.38* 0.14 0.15* 0.05

    (1.96) (0.98) (2.54) (0.94)
    Focal firm’s average speed of
    imitation (t)

    H2b 0.06 –0.15 0.04 0.02
    (0.54) (–1.18) (0.80) (0.36)

    Product technology
    heterogeneity (t)

    0.78** 21.02** 0.33** 20.61**
    (4.35) (–6.55) (5.35) (–9.58)

    Interactions
    Focal firm’s imitation scope 3
    Product technology
    heterogeneity

    H5a –0.25* –0.05
    (–2.27) (–1.04)

    Focal firm’s average speed of
    imitation 3 Product
    technology heterogeneity

    H5b –0.06 –0.03
    (–0.89) (–0.72)

    Controls
    Focal firm’s innovation scope (t) 0.01 20.03 0.02 0.02

    (0.08) (–0.33) (0.58) (0.57)
    Rivals’ innovation scope (t 1 1) 0.61** 20.02 0.32** 20.02

    (8.37) (–0.21) (8.44) (–0.64)
    Relative market position (t) 20.191 0.04 20.09 20.01

    (–1.92) (0.38) (–1.44) (–0.20)
    Industry concentration (t) 20.181 20.57** 20.02 20.33**

    (–1.76) (–3.96) (–0.50) (–7.32)
    GDP volatility (t) 20.07 20.43** 20.05 20.23**

    (–0.73) (–3.52) (–1.19) (–5.60)
    2nd quarter year t (largest new
    product launch)

    20.53** 20.04 20.13 20.09
    (–2.86) (–0.23) (–1.39) (–0.98)

    3rd quarter year t (largest new
    product launch)

    0.09 20.19 20.05 20.14
    (0.35) (–1.07) (–0.49) (–1.42)

    4th quarter year t (largest new
    product launch)

    0.24 20.321 20.05 20.04
    (1.05) (–1.66) (–0.49) (–0.35)

    n 566 566 566 566
    R-squared 0.42 0.37
    x
    2 220.05 146.93 417.94 331.59

    Notes: Estimates are based on standardized variables; z-statistics in parentheses; coefficients in bold are those related to the tested
    hypotheses.

    aFirm dummies were included but not reported.
    1p , 0.10
    *p , 0.05

    **p , 0.01

    2017 1901Giachetti, Lampel, and Li Pira

    1994; Lee et al., 2000). By contrast, the negative effect
    of scope on focal firm performance is apparently
    counterintuitive. We will consider this result again
    later in this section, since the imitation scope–
    performance relationship turns out to be positive
    when considering the moderating effect of product
    technology heterogeneity. In line with the Red
    Queen argument, we also found that focal firm imi-
    tation scope triggers rivals’ imitation scope (Hy-
    pothesis 2a), but focal firm speed of imitation does
    not trigger rivals’ rapid imitation (Hypothesis 2b).
    There are two possible interpretations of these re-
    sults: (a) it is possible that rivals perceive scope as
    more of a threat to their competitive positions com-
    pared to speed, and thus are more likely to invest
    resources matching scope rather than speed; or
    (b) rivals may not be able to move as quickly as the
    focal firms that were the earliest, if not the first, to
    make the imitative moves. Either way, whether rivals
    choose to focus resources on scope over speed, or
    cannot marshal the resources to respond quickly, ri-
    vals definitely respond to scope moves, implying that
    scope is an important strategic issue in technology-
    intensive industries. These results contrast in part
    with those studies in the competitive dynamics liter-
    ature that have described response speed as the main
    strategic issue firms focus resources on when coun-
    termoveing against rivals (Derfus et al., 2008;
    Markides & Geroski, 2004). Finally, to close the Red
    Queen cycle, we found that rivals’ imitation scope
    and speed have a negative effect on the focal firm’s
    performance (Hypotheses 3a and 3b).

    To illustrate our results, it may be useful to give an
    example. Nokia’s pioneering of digital technologies
    such as infrared, games, an email client, and WAP
    (Wireless Application Protocol) during the 1990s
    elicited various reactions from rivals. Siemens was
    among the first to imitate (high imitation speed) all of
    the technologies mentioned earlier (high imitation
    scope). This reinforced Siemens’ product portfolio
    competitiveness, and increased its sales perfor-
    mance relative to slower imitators such as NEC,
    Philips, and Sagem. Nevertheless, Siemens enjoyed
    a temporary competitive advantage that lasted until
    Nokia’s innovations were adopted by other handset
    vendors. Subsequently, at the beginning of the
    2000s, some vendors pioneered new product tech-
    nologies, such as Bluetooth, MMS (Multimedia
    Messaging Service) and photo camera, and a new
    series of imitative actions commenced, with firms
    such as Sony-Ericsson and Samsung installing this
    set of features in their new lines of phones more
    quickly compared to Siemens. Although in the first

    time period (i.e., during the 1990s) Siemens was
    able to match the scope and speed requirements,
    and in turn enjoyed a temporary competitive ad-
    vantage, in the second time period (i.e., the begin-
    ning of the 2000s) it did not possess the imitative
    capabilities to stay aligned with rivals, and strug-
    gled to catch up. To paraphrase Lewis Carroll
    (1960), Siemens realized that although it was run-
    ning as fast as it could, it was not getting anywhere
    relative to its rivals. Interestingly, the escalating
    pressure to imitate in order to retain market position
    not only increased “competitive imitation” among
    handset vendors, but also accelerated the techno-
    logical evolution of the industry. In fact, looking
    back it is remarkable how quickly the industry
    moved in a few years from basic handsets capable of
    providing only phone calls in the mid-1990s, to
    multi-tasking devices that integrate nearly all types
    of portable technologies (Figure A1).

    In order to get a clearer picture of the boundaries of
    Red Queen competition in a technology-intensive
    industry, we also examined the extent to which
    Red Queen evolution may depend upon a specific
    industry condition—in our case, the level of product
    technology heterogeneity in the market. We found
    product technology heterogeneity to have a signifi-
    cant moderating effect in all time frames of the pro-
    posed Red Queen competitive imitation cycle, for at
    least one type of imitative action (Table 4). First,
    contrary to our prediction in Hypothesis 4a, our re-
    sults indicate that product technology heterogeneity
    significantly and positively moderates the effect of
    focal firm imitation scope on focal firm performance.
    This result, combined with the negative direct effect
    of firm imitation scope on its performance, is repre-
    sented in Figure 2. More specifically, when we plot
    the data of the significant interaction (i.e., scope of
    focal firm imitation 3 product technology hetero-
    geneity), we observe that: (a) the effect of the focal
    firm’s imitation scope on its performance is positive
    for high levels of product technology heterogeneity,
    while it is negative for low levels of product tech-
    nology heterogeneity, and (b) performance gains
    from the focal firm’s imitation scope are maximized
    when this scope is large, and product technology
    heterogeneity is high. The overall picture shows that
    imitation scope may indeed have a positive effect on
    firm performance, as predicted in Hypothesis 1a, but
    this occurs only for high levels of product technology
    heterogeneity. Ex post, an explanation for this result
    could be that when product technology heterogene-
    ity is high, focal firms have to imitate as many new
    technologies as they can in order to increase the

    1902 OctoberAcademy of Management Journal

    probability of launching new product models that
    converge with the product configuration that will
    become dominant.

    Moreover, as predicted in our theory, we found
    that product technology heterogeneity negatively
    moderates the relationship between the focal firm’s
    imitation scope and rivals’ imitation scope. This is
    because when product technology heterogeneity is
    high, the focal firm’s and rivals’ learning process
    is constrained. Rivals that react to the focal firm’s
    moves are more likely to be conservative when it
    comes to the number of new technologies imitated,
    preferring to wait until the technological uncertainty
    decreases. Overall, these findings are consistent with
    observations by other studies in the Red Queen lit-
    erature, namely that learning from competitive ex-
    perience will be less effective if firms encounter
    a series of environmental shocks that render their
    learning capability obsolete (Barkema et al., 2002;
    Derfus et al., 2008). Bearing in mind that product
    technology heterogeneity changes, which in turn
    influences the pace of technological change, we be-
    lieve that our results also contribute to research on
    how technological changes in technology-intensive
    industries may influence the way firms compete
    (Agarwal, Sarkar, & Echambadi, 2002; Utterback &
    Suarez, 1993), as well as their ability to preserve their
    performance vis-à-vis rivals (Bayus & Agarwal,
    2007).

    In line with our predictions, we also found that
    product technology heterogeneity negatively mod-
    erates the effect of rivals’ imitation scope on the

    focal firm’s performance. When products differ
    greatly in terms of the technologies they in-
    corporate, rivals’ imitative response to focal firm
    actions will disproportionately decrease the focal
    firm’s performance. The main reason for this, as we
    see it, is that rivals have an imitative advantage
    when the focal firm confronts greater uncertainty
    about the performance of new technologies. Rivals
    can observe the performance outcomes of the focal
    firm’s imitative action and then imitate (in the fol-
    lowing period) only new technologies that have
    demonstrated greater acceptance by consumers. In
    this way, rivals strengthen their competitive posi-
    tion with respect to the focal firm by investing only
    in value-enhancing technologies. Plotting the data
    from Model 3, we graphically represent the form of
    the significant interaction (i.e., scope of rivals’ im-
    itation 3 product technology heterogeneity) in
    Figures 3. Specifically, we show in Figure 3 the
    actual scope of rivals’ imitation and product tech-
    nology heterogeneity associated with various levels
    of performance. As expected, the relationship be-
    tween scope of rivals’ imitation and focal firm per-
    formance is more negative for high levels of product
    technology heterogeneity.

    It is worth noting that although not directly
    predicted in our theory, our regression analysis
    offers interesting results on the impact of product
    technology heterogeneity on firm performance and
    on rivals’ imitative actions. Model 3 (Table 2) and
    Figures 2 and 3 show that product technology
    heterogeneity has a positive effect on focal firm

    FIGURE 2
    Scope of Focal Firm’s Imitation, Product Technology Heterogeneity, and Focal Firm Performance

    3000

    2

    500

    2000

    1500

    1000

    500

    0
    0 1 2 3 0

    1
    2
    3
    4
    Product technology
    heterogeneity

    Focal firm’s
    imitation scope

    F
    o
    ca

    l
    fi

    rm
    p

    er
    fo

    rm
    a
    n

    ce

    2500–3000

    2000–

    2500

    1500–2000

    1000–1500

    500–1000

    0–500

    2017 1903Giachetti, Lampel, and Li Pira

    performance. In fact, when product technology
    heterogeneity is high, products introduced by in-
    dustry members are highly heterogeneous, and
    direct competition is likely to be relatively weak,
    since each firm in the industry attempts to carve
    out its own unique product niche. Thus, although
    we found that product technology heterogeneity
    may create uncertainty and hamper the effective-
    ness of focal firms’ imitative actions, overall firms
    tend to achieve higher performance in this sce-
    nario. As for the direct effect of product technology
    heterogeneity on rivals’ imitative actions, we
    found that the effect is positive on rivals’ imitation
    scope (Model 6, Table 3), while the effect is nega-
    tive on rivals’ average speed of imitation (Model 9,
    Table 3): higher heterogeneity in product designs
    triggers imitative responses aimed at catching the
    opportunities offered by the variety of available
    technologies, but the propensity to imitate several
    different technologies limits the rivals’ ability to
    imitate them rapidly.

    Finally, although our paper looks at Red Queen
    competition primarily from the point of view of
    key competitive moves that involve imitation of
    new product technologies, which in turn triggers
    rivals’ imitative response (Figure 1), we also want
    to take into account the possibility that rivals
    respond to the focal firm’s imitative actions
    with their own innovations. In Table 6, Models
    14–16, we report the analysis of the effect of
    a focal firm’s imitative actions on rivals’ innova-
    tion scope. Given the excess of zero counts in
    the rivals’ innovation scope dependent variable,

    a zero-inflated Poisson regression was used
    (Cameron & Trivedi, 2009). Model 16 is the full
    model, also taking into account the moderating
    effect of product technology heterogeneity. As can
    be observed from Model 16, while the focal firm’s
    average speed of imitation has no significant im-
    pact on rivals’ innovation scope (b 5 0.05, p . .1),
    the impact of the focal firm’s imitation scope is
    negative and significant (b 5 20.20, p , .1). This re-
    sult should be read together with the positive effect
    of the focal firm’s imitation scope on rivals’ imita-
    tion scope we found in Model 6: as the focal firm’s
    imitative action (scope) increases, rivals tend to re-
    spond with imitation at the expense of innovation.

    Limitations and Avenues for Future Research

    As may be expected, our study has limitations,
    some of which create opportunities for future re-
    search. First, as is the case in most empirical
    studies in competitive dynamics, our study cap-
    tures only observable strategies based on in-
    formation reported in the press and industry trade
    journals we examined. However, given the fact that
    mobile phones regularly incorporate technologies
    that originated in other product categories, such as
    digital cameras, MP3 players, and video games, it is
    likely that mobile phone vendors in our sample are
    influenced by technological decisions made by
    actors from other industries. This caveat applies to
    the United Kingdom as well as global mobile
    phone sales. Consequently, future research could

    FIGURE 3
    Rivals’ Scope of Imitation, Product Technology Heterogeneity, and Focal Firm Performance

    3000
    2500
    2000
    1500
    1000
    500

    0
    0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 180 1

    2 3
    4

    2500–3000

    2000–2500

    1500–2000
    1000–1500
    500–1000
    0–500

    Rivals’
    imitation scope

    Product technology
    heterogeneity
    F
    o
    ca
    l
    fi
    rm
    p
    er
    fo
    rm
    a
    n
    ce

    1904 OctoberAcademy of Management Journal

    examine how country and industry boundaries
    influence Red Queen competitive imitation.

    Second, our study examines an industry defined
    by a single product, the mobile phone. Red Queen
    competition in this case is likewise focused pri-
    marily on improvements to this device. Empiri-
    cally, studying an industry that is defined by
    a single product is an advantage inasmuch as it
    provides a context that allows us to examine Red
    Queen competition with greater precision. How-
    ever, this advantage is also a limitation, given the

    fact that competition in many industries, for ex-
    ample food retailing, is multi-product. It can be
    reasonably expected that product line diversity will
    produce different action–reaction dynamics than is
    the case when competition is focused on a single
    device. For example, the pressure to respond to
    a rival’s move in one segment of the market may be
    lower if the focal firm sees potential losses as minor
    relative to the performance of its entire product
    portfolio. Our results for imitation scope and speed
    may be generalizable to other industries where

    TABLE 6
    Robust Zero-inflated Poisson Regression Analysis: Focal Firm Imitative Actions on Rivals’ Innovation Scope

    Model 14 Model 15 Model 16

    Rivals’ innovation
    scope (t 1 1)
    Rivals’ innovation
    scope (t 1 1)
    Rivals’ innovation
    scope (t 1 1)

    Constant 20.13 20.75** 20.85**
    (–1.10) (–6.99) (–7.47)

    Independent variables
    Focal firm’s imitation scope (t) 20.05 20.201

    (–0.55) (–1.69)
    Focal firm’s average speed of imitation (t) 20.03 0.05

    (–0.44) (0.62)
    Rivals’ imitation scope (t 1 1) 0.51** 0.52**

    (12.71) (12.57)
    Rivals’ average speed of imitation (t 1 1) 0.22** 0.20*

    (2.58) (2.36)
    Product technology heterogeneity (t) 20.30** 20.29**

    (–2.87) (–2.78)
    Interactions
    Focal firm’s imitation scope 3 Product technology
    heterogeneity

    0.19 1

    (1.72)
    Focal firm’s average speed of imitation 3 Product
    technology heterogeneity

    20.04
    (–0.43)

    Controls
    Focal firm’s innovation scope (t) 20.01 20.03 20.03

    (–0.20) (–0.61) (–0.66)
    Relative market position (t) 0.03 0.02 0.01

    (0.51) (0.38) (0.24)
    Industry concentration (t) 0.11 0.07 0.09

    (1.31) (0.68) (0.96)
    GDP volatility (t) 0.081 0.04 0.05

    (1.73) (0.56) (0.82)
    2nd quarter year t (largest new product launch) 20.21 20.08 20.07

    (–1.36) (–0.58) (–0.47)
    3rd quarter year t (largest new product launch) 20.16 20.08 20.05

    (–0.98) (–0.51) (–0.32)
    4th quarter year t (largest new product launch) 20.18 20.06 20.03

    (–1.15) (–0.38) (–0.18)
    n 566 566 566
    Log-likelihood 2568.95 2518.29 2515.55
    Likelihood Ratio x2 7.49 180.11 187.05

    Notes: Estimates are based on standardized variables; z-statistics in parentheses.
    1p , 0.10
    *p , 0.05

    **p , 0.01

    2017 1905Giachetti, Lampel, and Li Pira

    single products drive competition, but may not be
    generalizable for multi-product industries. Further
    research is clearly needed to extend the findings of
    our study to industries where competition engages
    firms that offer consumers a wide range of products.

    Third, although wecontendthat product technology
    heterogeneity can affect the way firms learn from
    the technology adoption decisions of rivals, and un-
    dertake actions accordingly, scholars of organiza-
    tional learning have identified a variety of learning
    mechanisms—e.g., mimetic, vicarious, and experien-
    tial (Baum et al., 2000; Haunschild & Miner, 1997;
    Lieberman & Asaba, 2006)—that are not captured in
    our theory and empirical analysis. Whether firms se-
    lect one mode of learning over another depends on
    their resource endowment and the time they can wait
    before committing to a decision, with inevitable dif-
    ferent impacts on the type and effectiveness of their
    imitative actions. It would be useful for future research
    to develop appropriate measures of different learning
    modes,aswellastoprovideatheoreticalbasisforthese
    measures.

    Finally, analysis of Red Queen competition is
    usually studied through the lens of inter-firm ri-
    valry, with an interest in how firms react to each
    other’s moves (Delacour & Liarte, 2012). However,
    Derfus et al. (2008) suggested that it is also impor-
    tant to see Red Queen competition as a link be-
    tween micro and macro industry dynamics. They
    noted, for instance, that new product introduction
    moves may represent a “positive sum” game in
    which the race to introduce products with more
    features and better technologies can increase con-
    sumer demand for the industry as a whole. Para-
    doxically, therefore, Red Queen competition can
    lead to a competitive stalemate at the level of in-
    dividual firms, while at the same time producing
    greater benefits for all to share. The same can be
    said for technological change. Firms introduce
    new products and new technologies in order to
    retain their position, but in the process of doing so
    they move the industry’s technological frontier
    forward. In principle, we can therefore say that Red
    Queen competition often plays an important role
    in linking competitive interactions at the micro
    industry level with macro industry dynamics
    (Felin, Foss, & Ployhart, 2015). This linking role is
    potentially a fruitful area of Red Queen competi-
    tion research. Future research should therefore
    examine how different types of Red Queen com-
    petition impact the evolution of industries, and,
    vice versa, how the evolution of industries shapes
    Red Queen competition.

    REFERENCES

    Abrahamson, E., & Rosenkopf, L. 1993. Institutional and
    competitive bandwagons: Using mathematical mod-
    eling as a tool to explore innovation diffusion.
    Academy of Management Review, 18: 487–517.

    Agarwal, R., Sarkar, M. B., & Echambadi, R. 2002. The
    conditioning effect of time on firm survival: An in-
    dustry life cycle approach. Academy of Management
    Journal, 45: 971–994.

    Alpert, F. H., & Kamins, M. A. 1995. An empirical in-
    vestigation of consumer memory, attitude, and per-
    ceptions toward pioneer and follower brands. Journal
    of Marketing, 59: 34–45.

    Argyres, N., Bigelow, L., & Nickerson, J. A. 2015. Domi-
    nant designs, innovation shocks, and the follower’s
    dilemma. Strategic Management Journal, 36:
    216–234.

    Barkema, H. G., Baum, J. A., & Mannix, E. A. 2002. Man-
    agement challenges in a new time. Academy of
    Management Journal, 45: 916–930.

    Barnett, W. P., & Hansen, M. T. 1996. The red queen in
    organizational evolution. Strategic Management
    Journal, 17(S1): 139–157.

    Barnett, W. P., & McKendrick, D. G. 2004. Why are some
    organizations more competitive than others? Evi-
    dence from a changing global market. Administrative
    Science Quarterly, 49: 535–571.

    Barnett, W. P., & Sorenson, O. 2002. The Red Queen in
    organizational creation and development. Industrial
    and Corporate Change, 11: 289–325.

    Baucells, M., Weber, M., & Welfens, F. 2011. Reference
    point formation and updating. Management Science,
    57: 506–519.

    Baum, J. A. C., & Ingram, P. 1998. Survival-enhancing
    learning in the Manhattan hotel industry. Manage-
    ment Science, 44: 996–1016.

    Baum, J. A. C., Li, S. X., & Usher, J. M. 2000. Making the next
    move: How experiential and vicarious learning shape
    the locations of chains’ acquisitions. Administrative
    Science Quarterly, 45: 766–801.

    Baumol, W. J. 2004. Red-Queen games: Arms races, rule of
    law and market economies. Journal of Evolutionary
    Economics, 14: 237–247.

    Bayus, B. L., & Agarwal, R. 2007. The role of pre-entry
    experience, entry timing, and product technology
    strategies in explaining firm survival. Management
    Science, 53: 1887–1902.

    Cameron, C. A., & Trivedi, P. K. 2009. Microeconometrics
    using Stata. College Station, TX: Stata Press.

    Cameron, T. A. 2005. Updating subjective risks in the
    presence of conflicting information: an application to

    1906 OctoberAcademy of Management Journal

    climate change. Journal of Risk and Uncertainty, 30:
    63–97.

    Carroll, L. 1960. Through the looking glass and what
    Alice found there: The annotated Alice. New York,
    NY: Bramhall House.

    Carpenter, G. S., & Nakamoto, K. 1989. Consumer prefer-
    ence formation and pioneering advantage. JMR,
    Journal of Marketing Research, 26: 285–298.

    Chatterjee, S., & Hadi, A. S. 2006. Regression analysis by
    example. New York, NY: Wiley.

    Chen, M. J., Lin, H. C., & Michel, J. G. 2010. Navigating in
    a hypercompetitive environment: the roles of action
    aggressiveness and TMT integration. Strategic Man-
    agement Journal, 31: 1410–1430.

    Chen, M. J., & Miller, D. 1994. Competitive attack, retaliation
    and performance: an expectancy‐valence framework.
    Strategic Management Journal, 15: 85–102.

    Chen, M. J., & Miller, D. 2012. Competitive dynamics:
    Themes, trends, and a prospective research plat-
    form. The Academy of Management Annals, 6:
    135–210.

    Chen, Y., & Turut, O. 2013. Context-dependent preferences
    and innovation strategy. Management Science, 59:
    2747–2765.

    Christensen, C. M., & Bower, J. L. 1996. Customer power,
    strategic investment, and the failure of leading firms.
    Strategic Management Journal, 17: 197–218.

    Cohen, W. M., & Levinthal, D. A. 1989. Innovation and
    learning: The two faces of R&D. Economic Journal
    (Oxford), 99: 569–596.

    Csaszar, F. A., & Siggelkow, N. 2010. How much to copy?
    Determinants of effective imitation breadth. Organi-
    zation Science, 21: 661–676.

    D’Aveni, R. A. 1994. Hypercompetition: Managing the
    dynamics of strategic maneuvering. New York, NY:
    The Free Press.

    Delacour, H., & Liarte, S. 2012. The Red Queen Effect:
    Principles, synthesis and implications for strategy.
    M@n@gement, 15: 313–330.

    Derfus, P. J., Maggitti, P. G., Grimm, C. M., & Smith, K. G.
    2008. The Red Queen effect: Competitive actions and
    firm performance. Academy of Management Jour-
    nal, 51: 61–80.

    Ethiraj, S. K., & Zhu, D. H. 2008. Performance effects of
    imitative entry. Strategic Management Journal, 29:
    797–817.

    Euromonitor International (2003–2008). Mobile phones in
    the U.K. Annual directories. London, U.K.: Euro-
    monitor International Ltd.

    Felin, T., Foss, N. J., & Ployhart, R. E. 2015. The micro-
    foundations movement in strategy and organization

    theory. The Academy of Management Annals, 9:
    575–632.

    Ferrier, W. J., Smith, K. G., & Grimm, C. M. 1999. The role of
    competitive action in market share erosion and in-
    dustry dethronement: A study of industry leaders and
    challengers. Academy of Management Journal, 42:
    372–388.

    Frenken, K., Saviotti, P. P., & Trommetter, M. 1999. Va-
    riety and niche creation in aircraft, helicopters, mo-
    torcycles and microcomputers. Research Policy, 28:
    469–488.

    Gaba, V., & Terlaak,A. 2013. Decomposing uncertainty and
    its effects on imitation in firm exit decisions. Orga-
    nization Science, 24: 1847–1869.

    Giachetti, C., & Dagnino, G. B. 2017. The impact of tech-
    nological convergence on firms’ product portfolio
    strategy: An information-based imitation approach.
    R&D Management, 47: 17–35.

    Giachetti, C., & Lanzolla, G. 2016. Product technology
    imitation over the product diffusion cycle: Which
    companies and product innovations do competitors
    imitate more quickly? Long Range Planning, 46:
    250–264.

    Giachetti, C., & Marchi, G. 2017. Successive changes in
    leadership in the worldwide mobile phone in-
    dustry: The role of windows of opportunity and
    firms’ competitive action. Research Policy, 46:
    352–364.

    Greve, H. R. 1996. Patterns of competition: The diffusion of
    a market position in radio broadcasting. Administrative
    Science Quarterly, 41: 29–60.

    Greve, H. R. 1998. Managerial cognition and the mimetic
    adoption of market positions: What you see is what
    you do. Strategic Management Journal, 19:
    967–988.

    Haddow, A., Hare, C., Hooley, J., & Shakir, T. 2013. Mac-
    roeconomic uncertainty: What is it, how can we
    measure it and why does it matter? Bank of England
    quarterly bulletin, June, Q2.

    Haunschild, P. R., & Miner, A. S. 1997. Modes of in-
    terorganizational imitation: The effects of outcome
    salience and uncertainty. Administrative Science
    Quarterly, 42: 472–500.

    Honoré, B. E. 1992. Trimmed LAD and least squares es-
    timation of truncated and censored regression
    models with fixed effects. Econometrica, 60:
    533–565.

    Hsieh, K. Y., & Vermeulen, F. 2013. The structure of com-
    petition: How competition between one’s rivals in-
    fluences imitative market entry. Organization
    Science, 25: 299–319.

    Kardes, F. R., & Kalyanaram, G. 1992. Order-of-entry
    effects on consumer memory and judgment: An

    2017 1907Giachetti, Lampel, and Li Pira

    information integration perspective. Journal of
    Marketing Research, 29: 343–357.

    Krishnan, V., & Bhattacharya, S. 2002. Technology selec-
    tion and commitment in new product development:
    The role of uncertainty and design flexibility. Man-
    agement Science, 48: 313–327.

    Lampel, J., & Shamsie, J. 2009. All that running: Red Queen
    competition inthe U.S. motion picture industry. Paper
    presented at the Academy of Management conference,
    Chicago, August 2009.

    Lee, H., Smith, K., Grimm, C. M., & Schomburg, A. 2000.
    Timing, order and durability of new product advan-
    tages with imitation. Strategic Management Journal,
    21: 23–30.

    Levitt, T. 1966. Innovative imitation. Harvard Business
    Review, 44: 63–70.

    LiCalzi, M., & Marchiori, D. 2013. Pack light on the move:
    Exploitation and exploration in a dynamic environ-
    ment. In: S. Leitner & F. Wall (eds.), Artificial eco-
    nomics and self organization: 205–216. Berlin:
    Springer.

    Lieberman, M. B., & Asaba, S. 2006. Why do firms imitate
    each other? Academy of Management Review, 31:
    366–385.

    Lieberman, M. B., & Montgomery, D. B. 1988. First-mover
    advantages. Strategic Management Journal, 9(S1):
    41–58.

    Lieberman, M. B., & Montgomery, D. B. 1998. First-mover
    (dis)advantages: Retrospective and link with the
    resource-based view. Strategic Management Jour-
    nal, 19: 1111–1125.

    Lippman, S. A., &Rumelt, R. P. 1982. Uncertain imitability:
    An analysis of interfirm differences in efficiency un-
    der competition. The Bell Journal of Economics, 13:
    418–438.

    Madhok, A., Li, S., & Priem, R. L. 2010. The resource‐based
    view revisited: Comparative firm advantage, willingness‐
    based isolating mechanisms and competitive het-
    erogeneity. European Management Review, 7:
    91–100.

    Makadok, R. 1998. Can first-mover and early-mover ad-
    vantages be sustained in an industry with low barriers
    to entry/imitation? Strategic Management Journal,
    19: 683–696.

    Markides, C. C., & Geroski, P. A. 2004. Fast second: How
    smart companies bypass radical innovation to enter
    and dominate new markets. San Francisco, CA: John
    Wiley & Sons.

    Miller, D., & Chen, M. J. 1994. Sources and consequences of
    competitive inertia: A study of the US airline industry.
    Administrative Science Quarterly, 39: 1–23.

    Mintel International Group Limited (1997–2008). Mobile
    phones and network providers in the U.K. Annual
    Directories. London, U.K.: Mintel International
    Group Limited.

    Murmann, J. P., & Frenken, K. 2006. Toward a systematic
    framework for research on dominant designs, tech-
    nological innovations, and industrial change. Re-
    search Policy, 35: 925–952.

    Narasimhan, C., & Turut, O. 2013. Differentiate or imitate?
    The role of context-dependent preferences. Market-
    ing Science, 32: 393–410.

    Ndofor, H. A., Sirmon, D. G., & He, X. 2011. Firm resources,
    competitive actions and performance: investigating
    a mediated model with evidence from the in-vitro di-
    agnostics industry. Strategic Management Journal,
    32: 640–657.

    Nelson, R., & Winter, S. 1982. An evolutionary theory of
    economic change. Cambridge, MA: Harvard Univer-
    sity Press.

    O’Shaughnessy, J. 1989. Why people buy. Oxford, U.K.:
    Oxford University Press.

    Pessemier, E. A. 1978. Stochastic properties of changing
    preferences. The American Economic Review, 68:
    380–385.

    Posen, H. E., Lee, J., & Yi, S. 2013. The power of imperfect
    imitation. Strategic Management Journal, 34:
    149–164.

    Posen, H. E., & Levinthal, D. A. 2012. Chasing a moving
    target: Exploitation and exploration in dy-
    namic environments. Management Science, 58:
    587–601.

    Rao, R., & Rutenberg, D. 1979. Pre-empting an alert rival:
    strategic timing of the first plant by analysis of so-
    phisticated rivalry. The Bell Journal of Economics,
    10: 412–428.

    Rhee, M., Kim, Y. C., & Han, J. 2006. Confidence in imita-
    tion: Niche-width strategy in the U.K. automobile in-
    dustry. Management Science, 52: 501–513.

    Rindova, V., Ferrier, W. J., & Wiltbank, R. 2010. Value from
    gestalt: how sequences of competitive actions create
    advantage for firms in nascent markets. Strategic
    Management Journal, 31: 1474–1497.

    Ross, J. M., & Sharapov, D. 2015. When the leader follows:
    Avoiding dethronement through imitation. Academy
    of Management Journal, 58: 658–679.

    Schumpeter, J. 1942. Capitalism, socialism, and de-
    mocracy. New York, NY: Harper & Row.

    Semadeni, M., & Anderson, B. S. 2010. The follower’s di-
    lemma: Innovation and imitation in the professional
    services industry. Academy of Management Journal,
    53: 1175–1193.

    1908 OctoberAcademy of Management Journal

    Shannon, C. E. 1948. A mathematical theory of communi-
    cation. Bell System Technical Journal, 27, 623–656.

    Smith, K. G., Ferrier, W. J., & Ndofor, H. 2001. Competitive
    dynamicsresearch:Critiqueandfuturedirections.In:M.
    Hitt, R.E. Freeman, & J. Harrison (Eds.), Handbook of
    Strategic Management, 315–361. London: Blackwell.

    Smith, K. G., Grimm, C., Gannon, M., & Chen, M. J. 1991.
    Organizational information processing, competitive re-
    sponses and performance in the U.S. domestic airline
    industry. Academy of Management Journal, 34: 60–85.

    Teece,D.J.1998.Capturingvaluefromknowledgeassets:The
    new economy, markets for know-how, and intangible
    assets. California Management Review, 40: 55–79.

    Utterback, J., & Suarez, F. F. 1993. Innovation, competition
    and industry structure. Research Policy, 22: 1–21.

    Van Valen, L. 1973. A new evolutionary law. Evolutionary
    Theory, 1: 1–30.

    Waldman, D. E., & Jensen, E. J. 2012. Industrial organi-
    zation: Theory and practice. Upper Saddle River,
    N.J.: Prentice Hall.

    Wiggins, R. R., & Ruefli, T. W. 2005. Schumpeter’s ghost: Is
    hypercompetition making the best of times shorter?
    Strategic Management Journal, 26: 887–911.

    Young, G., Smith, K. G., & Grimm, C. M. 1996. “Austrian”
    and industrial organization perspectives on firm-level
    competitive activity and performance. Organization
    Science, 7: 243–254.

    Zellner, A. 1962. An efficient method of estimating seem-
    ingly unrelated regressions and tests for aggregation

    bias. Journal of the American Statistical Associa-

    tion, 57: 348–368.

    Zhou, J. 2011. Reference dependence and market compe-

    tition. Journal of Economics & Management Strat-

    egy, 20: 1073–1097.

    Claudio Giachetti (claudio.giachetti@unive.it) is an associate
    professor of strategy at Ca’ Foscari University of Venice (Italy),
    Department of Management. He received his PhD from Ca’
    Foscari University of Venice. His primary research interests
    concern competitive dynamics and product innovation in
    rapidlychangingtechnologicalandinstitutionalenvironments.

    Joseph Lampel (joseph.lampel@manchester.ac.uk) is
    Eddie Davies Professor of Enterprise and Innovation
    Management at Alliance Manchester Business School,
    University of Manchester (U.K.). He received his PhD
    from McGill University, Montreal. His research fo-
    cuses on the dynamics of competition, innovation de-
    cision making, and strategy formation in creative
    industries.

    Stefano Li Pira (stefano.li-pira@wbs.ac.uk) is an assistant
    professor at Warwick Business School, the University of
    Warwick (U.K.). He received his PhD from Ca’ Foscari
    University of Venice. His primary research interests con-
    cern competitive dynamics and imitation in technology
    intensive industries.

    2017 1909Giachetti, Lampel, and Li Pira

    mailto:claudio.giachetti@unive.it

    mailto:joseph.lampel@manchester.ac.uk

    mailto:stefano.li-pira@wbs.ac.uk

    APPENDIX A

    FIGURE A1
    Average Number of Innovations and Imitations by U.K. Mobile Phone Vendors (1997–2007)

    Total number of technologies

    42
    40
    38
    36
    34
    32
    30
    28
    26
    24
    22
    20
    18
    16
    14
    12
    10
    8
    6
    4
    2
    0

    7
    6
    5
    4
    3
    2
    1

    0
    1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

    Imitations

    F
    ir

    m
    ’s

    a
    v

    er
    a

    ge
    n

    u
    m

    b
    er

    o
    f

    im
    it
    a
    ti
    o
    n

    s
    a

    n
    d

    i
    n

    n
    o

    v
    a

    io
    n
    s

    Innovations Technologies per product

    A
    v

    er
    a
    ge
    n
    u
    m
    b
    er
    o
    f
    te
    ch
    n
    o

    lo
    gi

    es
    p

    er
    p

    ro
    d
    u
    ct

    ;
    T

    o
    ta

    l
    n

    u
    m
    b
    er
    o
    f
    te
    ch
    n
    o
    lo
    gi

    es
    c

    u
    rr

    en
    tl

    y
    a

    d
    o
    p

    te
    d

    Notes: Values presented in the figure are based only on “regular phones”—smartphones are excluded. The average number of innovations
    (imitations) expresses, on average, in a given year, how many new product technologies are introduced (imitated) by handset vendors. The
    average number of technologies per product refers to the average number of technologies that handset vendors installed in their phones in
    a given year. The total number of technologies refers to the total number of different technologies that were adopted in a given year by handset
    vendors.

    1910 OctoberAcademy of Management Journal

    TABLE A1
    Product Technologies Introduced in the U.K. Mobile Phone Industry from 1997–2007

    Technological
    system Types of functions offered

    List of technologies (month of
    introduction in the U.K. market) Description

    Networking Mobile phone networks use signals
    on specific frequency bands.
    Phones must be compatible with
    these bands in order to work with
    the network.

    Dual band (Feb 1998) Phone’s ability to work with two of
    thefourmajor GSM(GlobalSystem
    for Mobile Communications)
    frequency bands. An important
    feature for users who wish to use
    the same handset in different
    locations where the networks work
    on different bands. For example,
    some European dual-band phones
    do not work in the U.S., and vice
    versa.

    Tri band (Aug 1999) Phone’s ability to work with three of
    the four major GSM frequency
    bands, allowing it to work in most
    parts of the world.

    Quad band (Oct 2003) Phone’s ability to work with the four
    major GSM frequency bands (850/
    900/1800/1900 MHz), making it
    compatible with all the major GSM
    networks in the world.

    Wideband Code Division Multiple
    Access (WCDMA) (Mar 2003)

    Third-generation (3G) wireless
    standard that allows use of both
    voice and data. It has different
    frequency bands (Europe and
    Asia—2100MHz, North
    America—1900MHz and
    850MHz).

    High-speed data
    transfer

    Mobile phone networks support
    different types of data transfer,
    which allows users to access
    mobile internet, MMS and other
    advanced features like video
    streaming.

    High-Speed Circuit-Switched Data
    (HSCSD) (Nov 2000)

    System for data calls on GSM
    networks that came before packet-
    based systems such as GPRS and
    EDGE. It was never widely adopted
    outside Europe.

    General Packet Radio Service (GPRS)
    (Mar 2001)

    A packet-switching technology that
    enables data transfers through
    cellular networks. It is used for
    mobile internet, MMS and other
    data communications. Informally,
    GPRS is also called 2.5G.

    Enhanced Data rates for GSM
    Evolution (EDGE) (Feb 2004)

    Data system used on top of GSM
    networks. It provides nearly three
    times faster speeds than the
    outdated GPRS system. EDGE
    meets the requirements for a 3G
    network but is usually classified as
    2.75G.

    Universal Mobile
    Telecommunications System
    (UMTS) (Mar 2003)

    Includes high data speeds (2 Mbps),
    always-on data access, and greater
    voice capacity, enabling such
    advanced features as live video
    streaming.

    High Speed Downlink Packet Access
    (HSDPA) (Mar 2007)

    Upgrade for UMTS networks that
    doubles network capacity and
    increases download data speeds by
    five times or more.

    Phone call Phone call functionalities refer to the
    way the user can make a phone call
    (e.g., voicedialing the number),the

    Vibrate alert (Jan 1997) Can alert user to events such as an
    incoming call or an incoming
    message with a vibrate alert.

    2017 1911Giachetti, Lampel, and Li Pira

    TABLE A1
    (Continued)

    Technological
    system Types of functions offered
    List of technologies (month of
    introduction in the U.K. market) Description

    type of call (i.e., voice vs. video),
    and the type of call alert (the
    mobile phone can alert the user to
    events such as an incoming call or
    an incoming message in a number
    of ways).

    Voice Dial (Jul 1997) Allows the user to dial a number via
    a voice command.

    Polyphonic ringtones(Jan 2000) Creates realistic-sounding music by
    synthesizing several notes
    simultaneously. The more notes
    the synthesizer can play
    simultaneously, the richer the
    musical effect. Usually mobile
    phone synthesizers can reproduce
    from 4 to 72 simultaneous tones.

    True tones (Feb 2003) Audio recordings, typically in
    a common format such as MP3,
    AAC, or WMA.

    Downloadable ringtones (Feb 1998) Allows the user to load a new
    ringtone by downloading via
    a special SMS/MMS, or from the
    Internet.

    Composer (Aug 1997) Allows the user to create musical
    notes and then produce
    a customized ringtone.

    Recordable (Jan 2000) Permits sound recording—e.g., of
    someone’s voice—and then using
    it as a ringtone.

    Video Call (Mar 2003) 3G-network feature that allows two
    callers to talk to each other while at
    the same time viewing live video
    form each other’s phone.

    Connectivity Protocols for exchanging data over
    short distances from fixed and
    mobile devices, creating personal
    area networks.

    Infrared (Oct 1997) Standard for transmitting data using
    an infrared port. Uses a beam of
    infrared light to transmit
    information and so requires direct
    line of sight and operates only at
    close range.

    Bluetooth (Aug 2001) Wireless protocol for exchanging
    data over short distances from
    fixed and mobile devices, creating
    personal area networks.

    Universal Serial Bus (USB) (Sept
    2001)

    Standard for a wired connection
    between two electronic devices,
    including a mobile phone and
    a desktop computer. The
    connection is made using a cable
    that has a connector at either end.

    Messaging In addition to pure voice calls,
    messaging has been a core service
    since the beginning of GSM mobile
    telephony.

    Enterprise Messaging System (EMS)
    (Aug 1999)

    Extension of SMS (Short Message
    Service), which allows mobile
    phones to send and receive
    messages that have special text
    formatting, animations, graphics,
    sound effects, and ringtones. It is
    an intermediate technology
    between SMS and rich multimedia
    messages (MMS).

    Multimedia Messaging Service
    (MMS) (May 2002)

    Store-and-forward messaging service
    that allows subscribers to
    exchange multimedia files as
    messages (text, picture, audio,

    1912 OctoberAcademy of Management Journal

    TABLE A1
    (Continued)
    Technological
    system Types of functions offered
    List of technologies (month of
    introduction in the U.K. market) Description

    video, or a combination). In order
    to send or receiveanMMS,the user
    must have a compatible phone that
    is running over a GPRS or 3G
    network.

    SMS chat (Nov 2000) Analogous to the pervasive use of
    SMS as a type of instant messaging,
    much like chatting on a computer.
    The threaded message or
    conversation-style layout displays
    the incoming and outgoing
    messages between two
    participants in a single pane
    ordered chronologically.

    Instant Messaging (IM) (May 2002) Ability to engage in instant
    messaging services from a mobile
    handset. Mobile IM allows users to
    address messages to others using
    a dynamic address book full of
    users, with their online status
    updated constantly. Permits
    anyone participating to know
    when their “buddies” are available
    for chat. Mobile IM is viewed as
    a logical extension of the popular
    SMS service.

    E-mail (Mar 1998) Some phones provide a full e-mail
    client that can connect to a public
    or private e-mail server. There are
    different protocols used by the
    servers and some may not be
    supported by the phone’s e-mail
    client.

    Display Display is one of the most relevant
    aesthetic features of the mobile
    phone. Size, color, and physical
    interaction have a strong influence
    on the user’s experience.

    Colorscreen: 4 colors (Sep 1997), 256
    colors (Dec 2001), 4 K color (Jun
    2002), 65 K colors (Nov 2002),
    256 K colors (May 2004), 16 MK
    color (Aug 2005)

    Display is able to produce a number
    of different colors. A higher
    number results in a broader range
    of distinct colors. We identified six
    levels of color screen.

    Display shape: Display Vertical (May
    1998), Display Squared (Nov 2000)

    Mobile phone display shape that is
    convenient for the different
    function supported (messaging,
    photos, etc.). We identified two
    categories based on the display
    width/height ratio (squared
    display, vertical display).

    Touchscreen (May 1998) Display that responds to direct touch
    manipulation, either by finger,
    stylus, or both.

    Technological
    convergences

    Technologies traditionally
    originating in other industries, and
    “converging” into the mobile
    phone industry.

    Photo camera (Aug 2002)
    Videocamera (Mar 2003)

    Camera that can function as a digital
    camera, and in some cases can also
    shoot video.

    Photo resolution: 1 Mp 2 Mp (Oct
    2004), 2 Mp 3 Mp (Jun 2005), 3 Mp
    4 Mp (Sep 2006)

    Indicates the number of pixels on
    a display or in a camera sensor
    (specifically in a digital image). A
    higher resolution means more
    pixels and more pixels provide the

    2017 1913Giachetti, Lampel, and Li Pira

    TABLE A1
    (Continued)
    Technological
    system Types of functions offered
    List of technologies (month of
    introduction in the U.K. market) Description

    ability to display more visual
    information (resulting in greater
    clarity and more detail).

    Voice memo (Jan 1997) Permitsuserstorecordanotethatcan
    be heard whenever and wherever
    necessary. Some devices limit the
    duration of such memos, whereas
    others allow recording until they
    run out of memory.

    MP3 (Dec 2000) Audio storage protocol that stores
    music in a compressed format with
    very little loss in sound quality.
    MP3 files can be played using the
    music player of the mobile phone
    or set as a ringtone.

    Internet capabilities: HDML (Mar
    1998); WML (Aug 1999); HTML
    (Mar 1998); XHTML (Nov 2002)

    Various markup languages(ML) have
    been introduced to allow the
    handsetto surf theInternet.Mostof
    them allow only access to
    simplified Internet pages.

    Document viewer (Jul 2005) Program for displaying MS Word,
    Excel, and PowerPoint files.

    FM Radio (Apr 2000) Permits user to listen to most live-
    broadcast FM radio stations.
    Almost all phones with an FM
    radio tuner require a wired headset
    to be connected to the unit as it is
    used as an antenna.

    Games (Jan 1998) Many phones include simple games
    for the user to pass the time. The
    games referred to here are
    preinstalled on the phone and do
    not require a wireless connection
    to play.

    Notes: Definitions and technical descriptions of the sampled technologies were collected from both the special-interest magazines used for
    the analysis, and online catalogs such as www.gsmarena.com. Information on the month of introduction of a new technology in the U.K. market
    was collected from the special-interest magazines used for our analysis.

    1914 OctoberAcademy of Management Journal

    http://www.gsmarena.com

    Copyright of Academy of Management Journal is the property of Academy of Management
    and its content may not be copied or emailed to multiple sites or posted to a listserv without
    the copyright holder’s express written permission. However, users may print, download, or
    email articles for individual use.

    Calculator

    Calculate the price of your paper

    Total price:$26
    Our features

    We've got everything to become your favourite writing service

    Need a better grade?
    We've got you covered.

    Order your paper