van de ven - 1993 - a community perspective on the emergence of innovations

Journal of Engineering and Technology Management, 10 (1993) 23-51 Elsevier

23

A community perspective on the emergence of innovations

Andrew H. Van de Ven Carlson School of Management, University of Minnesota, Minneapolis, MN 55455, USA

Abstract

This paper proposes a social system framework at the interorganizational community level to examine how an industrial infrastructure emerges to facilitate and inhibit technological innovation. This infrastructure includes institutional arrangements, resource endowments, and proprietary activities that are necessary to develop and transform basic scientific knowledge into commercially viable products or services. We argue that this community infrastructure emerges through the accretion of numerous institutional, resource, and proprietary events which co-produce each other through the actions of public and private sector actors over an extended period of time. The practical implications of this perspective emphasize that innovation managers must not only be concerned with micro developments of a proprietary technical device or product within their organization, but also with the creation of a macro industrial system, which embodies the social, economic, and political infrastructure that any technological community needs to sustain its members.

Keywords. Technological innovation; Innovation process; Industrial infrastructure; Industry evolution

1. Introduction

How are new technologies developed and commercialized? What are the roles of public and private sector actors in creating new industries that support the commercialization of technological innovations? Answers to these questions are central to the current debate on international competitiveness (Green- house, 1992), particularly for those who argue that the engine for corporate revitalization and economic growth is developing and commercializing “ex- traordinary innovations” (Rosenbloom, 1985 ) . These innovations have the

Correspondence to: Prof. A.H. Van de Ven, Carlson School of Management, University of Min- nesota, 271-19th Avenue South, Minneapolis, MN 55455, USA. Andrew H. Van de Ven is Vernon Heath Professor of Organization Innovation and Change in the Carlson School of Management and Director of the Minnesota Innovation Research Program in the Strategic Management Research Center at the University of Minnesota.

0923-4748/93/$06.00 0 1993 Elsevier Science Publishers B.V. All rights reserved.

24

“transilient” capacity to transform established systems of technology and markets or to create totally new industries for economic growth (Abernathy and Clark, 1985 ).

Unfortunately, management and technology scholars have been largely ab- sent from this debate because they have not paid much attention to the generative processes by which novel forms of technologies, organizations, or institutions emerge. To be sure, many scholars have been interested in the dynamics of organizational and technological transformation, particularly those who view innovation development as an evolutionary process of variation, selection, and retention (Hannan and Freeman, 1977; Aldrich, 1979; Nelson and Winter, 1982) or as a punctuated equilibrium (Utterback and Abernathy, 1975; Tushman and Romanelli, 1985; Anderson and Tushman, 1990). But these scholars have not directly examined the variation or punctuation process itself of how novel forms of technologies, organizations, or institutions arise. In- stead, they focus on the diffusion of novel forms by examining the selection and retention of pre-existing variations in a population (Romanelli, 1991). Thus, the origins and process of novelty are not explained; they are assumed to be rare discrete events produced by random or “blind” chance (Campbell, 1974; McKelvey, 1982)) individual genius (Anderson and Tushman, 1990, p. 605)) or “whatever reason . . . They just happen . . . [Natural selection] is indif- ferent to the ultimate source of variation” (Aldrich, 1979, p. 28).

One objective of this paper is to change this indifference to the process of novelty, and to urge scholars to undertake research that contributes useful knowledge to the important global debate on corporate revitalization and international competitiveness. My position is that an understanding of the processes by which novel forms of technologies, organizations, and institutions emerge is not only central to this debate, but also critical to advancing knowledge of the process of innovation development. I argue that the origination of innovations has a dynamic history of its own which requires systematic study and explanation.

To study the origins and process of innovation, I will adopt and extend a social system framework, originally introduced by Van de Ven and Garud (1989), for understanding the roles of private and public sector actors in creating an industrial infrastructure that supports the development and commercialization of new technologies. Since new-to-the-world technologies typically transcend the boundaries of many individual firms, industries, and popula- tions (Astley, 1985), the framework adopts the interorganizational community or field as the unit of analysis, and focuses on the infrastructure necessary to develop and commercialize technological innovations. This infrastructure includes a variety of institutional arrangements, resource endowments, and proprietary functions that are necessary to develop and transform basic scientific knowledge into commercially viable products or services.

25

My basic objective is to develop a theory on the infrastructure for innovation at the macro community level that is grounded in a theory of action at micro levels of individuals and firms. We argue that the odds of a firm successfully developing an innovation are largely a function of the extent to which this infrastructure is developed at the industrial community level. While this community infrastructure facilitates and constrains entrepreneurial firms, it is the actions of individuals and firms that construct and change the community infrastructure. But this infrastructure does not emerge and change all at once by the actions of one or even a few key individuals. Instead, I argue that this infrastructure emerges through the accretion of numerous institutional, resource, and proprietary events which co-produce each other through the actions of many public and private sector actors over an extended period of time.

Section 2 of this paper elaborates this social system framework and links it to several current themes in the organizational and economic literature on technological development. Section 3 develops a set of propositions which explain how this social system emerges, and how its various components interact over time to facilitate and constrain the development of a new-to-the-world innovation.

2. Technological development as an emergent social system

The proposition that technological and institutional innovations recipro- cally co-produce each other within the system under investigation is a relatively new development in economic and organization theory (Ruttan and Hayami, 1984, p. 203). Although this proposition was central to Marx’s (1906) analysis of the dialectical relations between the forces of production (technology) and the relations of production (institutions) within the superstructure of cultural and resource endowments of a society, during the 1960s and 1970s organizational and economics scholars formulated overly one-sided theories of technical and institutional change. Those advocating a technological impera- tive perspective treated technological innovation as something that happened to the firm but was not determined within it (Abernathy and Clark, 1985, p. 3). Technological innovation was viewed as an environmental shock to which organizations or an economic system were to adapt if they were to survive (see reviews by Ruttan, 1978; Freeman, 1986; and Gerwin, 1981). A somewhat less vigorous tradition maintained that institutional rather than technical change was the dynamic source of social and economic development. This institutional determinism perspective emphasized that institutional change precedes, and is more fundamental than, technical change (North and Thomas, 1973; North, 1990; Chandler, 1990). As evident in Powell and DiMaggio (1991) and March and Olsen (1989)) this argument is reflected in the new institutionalism in organization theory, which argues that a larger institutionalized environment

26

enables and constrains organizations to invent and develop only certain types of technologies and practices.

In the meantime, historians were examining the process of innovation development, defined as the temporal sequence of events and activities that occur to create and transform basic scientific knowledge into commercially viable products or services delivered to customers. Numerous case histories demon- strate that new technologies are seldom, if ever, developed by a single firm alone in the vacuum of an institutional environment (see, e.g., Usher, 1954; Jewkes et al., 1958; Constant, 1980; Nelson, 1982; and Chandler, 1990). Many complementary innovations in technical and institutional arrangements are usually required before a particular technology is suitable for commercial ap- plication (Binswanger and Ruttan, 1978; Hughes, 1983; Rosenberg, 1983 ) . Re- search reviews by Mowery (1985 ) , Thirtle and Ruttan (1986)) Freeman ( 1986)) and Dosi (1988) show that the commercial success or failure of a technological innovation is, in great measure, a reflection of the institutional arrangements and available resource endowments which embody the social, economic, and political infrastructure that any community needs to sustain its members.

The social system framework proposed by Van de Ven and Garud (1989) articulates the components of such an industrial infrastructure for technological innovation. This framework, outlined in Table 1, adopts the interorganizational community or field as the unit of analysis, and focuses on the func-

TABLE 1

A social system framework for understanding innovation development and industry emergence

Components of community infrastructure for innovation

Institutional Arrangements

- legitimation (creation of trust ) - governance (norms, rules, regulations, laws) - technology standards

Resource Endowments

- scientific/technological research - financing and insurance arrangements - human competence pool (training and accreditation )

Proprietary Functions

- technological development functions: R&D, testing, manufacturing, marketing - innovation network/resource channel activities: appropriation of common goods (science,

financing, labor) vendor-supplier-distributor channels - market creation and consumer demand

Source: Adapted from A. Van de Ven and R. Garud, “A framework for understanding the emergence of new industries”, in: R. Rosenbloom and R. Burgelman (Eds.), Research on Technological

Innovation, Management and Policy, Vol. 4, JAI Press, Greenwich, CT, 1989, pp. 195-225.

21

tions or core activities necessary to develop and commercialize a technological innovation. This interorganizational community not only includes an industry, commonly defined as the set of firms producing similar or substitute products (Porter, 1980)) but also many other public and private sector actors who perform critical functions to develop and commercialize a new technology. The industrial infrastructure includes (1) institutional arrangements to legiti- mate, regulate, and standardize a new technology, (2) public resource endowments of basic scientific knowledge, financing mechanisms, and a pool of competent labor, as well as (3) proprietary R&D, manufacturing, marketing, and distribution functions by private entrepreneurial firms to commercialize the innovation for profit.

The relevance of each of these components of the social system to the emergence of innovations will now be discussed. This will be followed in Section 3 with a set of propositions on how these components of the industrial system emerge and interact over time to facilitate and constrain the development of innovations.

2.1. Proprietary functions

The proprietary component of the system incorporates the traditional industrial economic definition of an industry (Porter, 1980), which consists of the set of firms commercializing innovations that are close substitutes for each other. The focus here is on the actions of individual entrepreneurs and firms who typically appropriate basic knowledge from the public domain and transform it into proprietary knowledge through applied R&D work in areas related to a technological innovation. If they persist in developing the innovation, they subsequently develop a line of products and gain access to the complementary assets or functions (e.g., manufacturing, marketing, distribution, etc.) necessary to establish an economically viable business.

Williamson’s (1975) and Teece’s (1987) transactions costs theory is useful for understanding how firms organize to perform these proprietary functions. They emphasize that the boundaries of the firm are an important strategic variable for innovation (Teece, 1987, p. 217). They propose that those functions or complementary assets necessary for technological development which are difficult to protect from imitators (i.e., have weak appropriability regimes), require specialized investments, or are highly uncertain to execute, should be integrated and performed within the innovating firm; while those that have a strong appropriability regime, entail non-specialized investments, and for which there are a number of supply sources, should be licensed or contracted by firms with outside suppliers and vendors.

From a system perspective, these make or buy decisions by individual firms produce the aggregate industry channels of raw materials, manufacturing, marketing and distribution flows (Stern and El-Ansery, 1982; Hakansson,

28

1988). Shifts in individual firm make-or-buy decisions over time also determine changes in industry structure. Stigler (1957) examined this issue in terms of the relationship between product life cycle and vertical integration. He argued that early in the product life cycle, vertical integration of functions within firms (i.e., make decisions) will predominate because the market will not yet have had a chance to develop. As Smith (1776) indicated, the division of labor among firms is limited by the extent of the market. When the market expands, many of the tasks and functions grow into sufficient size to make it more prof- itable to “buy” them from specialized manufacturers or suppliers, as opposed to “making” them within vertically-integrated firms.

2.2. Resource endowments

Three kinds of resources are critical to the development of every technological innovation: (a) basic scientific or technological research, (b) financing mechanisms, and (c) a pool of competent human resources (Mowery and Ro- senberg, 1979). While private entrepreneurs or firms do engage in the development of these resources, typically, public organizations (often viewed as “external” to an industry) play a major role in creating and providing these “common goods”.

(a) Basic research Basic scientific or technological research provides the foundation of knowl-

edge that underlies technological innovations and makes the commercial birth of most industries possible. This basic knowledge is very costly to produce relative to its cost of diffusion and imitation (Mansfield, 1985 ). In addition, it builds in a cumulative fashion, and its generation is inherently an indivisible activity (Metcalfe and Soete, 1983). For these reasons, Nelson (1982) and Arrow (1962) argued that the social returns to research investment exceed the private returns to entrepreneurs; a condition leading to underinvestment by the firm (from the social point of view) in research. As a consequence, a variety of studies have shown that firms rely upon outside sources of knowledge and technical inventions for the vast majority of their commercially significant new products (Rosenbloom, 1966; Mueller, 1962; Utterback, 1974; Freeman, 1986; Nelson, 1982; Stobaugh, 1985).

However, the private appropriation of basic scientific knowledge is seldom a simple process of information transfer and diffusion because much of that knowledge may be tacit or “sticky”, meaning that the knowledge is difficult to disembody from its original source and transfer to an applied problem-solving site (Von Hippel, 1990). As a consequence, proprietary firms must often engage in a variety of strategies to acquire this tacit knowledge from basic research sites, ranging from simple communications (Allen, 1977) and personnel transfers (Roberts and Hauptman, 1986)) to various licensing arrangements

29

and joint R&D ventures between private firms and basic research centers (Ouchi and Bolton, 1987; Powell, 1990). These strategies appear consistent with von Hippel’s (1990) overall proposition that the location of problem- solving activities will shift over time to the sites or locations of that “sticky” data.

(b) Financing mechanisms While public institutions (e.g., the National Science Foundation or the Na-

tional Institute of Health) tend to play the major role in financing the development of basic scientific or technological knowledge, venture capital, either within a corporation or in the market, tends to be the key financial source that supports private firms in transforming basic knowledge into proprietary and commercial applications. In addition, the commercialization of many technological innovations requires unique industry-wide financing arrangements. For example, few bio-medical innovations would be commercially viable without the health care insurance industry and the creation of third-party payment reimbursement systems. Without such a financial infrastructure for a broad array of bio-medical and health care innovations, most patients would be un- able to pay for many bio-medical devices and treatments. But since these insurance systems limit coverage to specifically-designated medical devices and treatments, the firms competing to commercialize a specific bio-medical device must cooperate to both educate and influence third-party payers to include the innovation in their payment reimbursement systems.

Cc) Human resources A pool of competent human resources is another essential resource neces-

sary for the emergence of a new industry. New technologies mean new ways of performing essential tasks, be they related to research, manufacturing or marketing. This pool of competence tends to develop in three ways. First, basic research institutes and proprietary firms recruit and train people in specific skills related to the innovation. Over time, job transfers and mobility of profes- sionals between institutes and firms diffuses their skills throughout the industry (Rappa, 1989; Garud, 1990). A second method of developing this labor market is through the establishment of educational training programs and ac- credited degrees at colleges and universities, as well the organization of industry conferences, technical committees, trade publications, and technical jour- nals, which provide an opportunity for industry participants to share and learn from each other (Nelson, 1982). Finally, the competence pool is created through “collective invention” (Allen, 1983 ) among “invisible colleges” or networks of practitioners (often scientists and engineers from throughout the world) that emerge to support further development or problem solving activities within a new technological paradigm (Hull, 1988; Dosi, 1988; Rappa, 1989). For ex-

30

ample, von Hippel(1986) observed extensive trading of proprietary know-how among informal networks of process engineers in rival (and non-rival) firms in the U.S. steel minimill industry and elsewhere.

2.3. Institutional arrangements

The ultimate authorities governing and legitimating collective action are the rules and norms of the society in which organizations function (Galaskiewicz, 1985; Scott, 1987). The political context is the place for formally institution- alizing and legitimating a social system, which permits it to operate and gain access to the resources it needs (Meyer and Rowan, 1977; Pfeffer and Salancik, 1978, p. 214). The success or failure of a new industry and firms within it depends on their abilities to achieve institutional isomorphism (DiMaggio and Powell, 1983). To achieve this, firms may either adapt to institutional requirements or attempt to build their goals and procedures directly into society as institutional rules (Meyer and Rowan, 1977). Thus, firms compete not only in the marketplace, but also in this political institutional context. Rival firms often cooperate by collectively manipulating their institutional environment to legitimize and gain access to resources necessary for collective survival (Pfeffer and Salancik, 1978; Hirsch 1975; Meyer and Rowan, 1977).

(a) Governance It is widely recognized that a variety of governmental regulations and insti-

tutional arrangements facilitate and inhibit the emergence of new technologies and industries. Mowery (1985) and Nelson (1982) for example, discuss how government funding, by broadening the industry-wide knowledge base, can encourage new industry entrance, and thereby support a more competitive environment. So also, a more permissive antitrust policy permitting certain kinds of joint research ventures among competitors, as well as requirements by the Department of Defense for licensing and second sourcing of new devices, speed the diffusion of innovation and may aid the operation of competitive forces in an industry (Teece, 1987). However, as Ouchi and Bolton (1987) discuss, institutional policies encouraging rapid knowledge diffusion, if pursued too ea- gerly, may undermine the return to the knowledge producer, and thus the in- centive to invest in information producing activities. But here another institutional mechanism has been devised; the patent system grants monopoly rights to the use of knowledge for a limited period of time. Although these institutional arrangements are often highly imperfect, research shows they often exert a profound effect on technological and industry development (Nelson, 1982 ) .

(b) L.egitimntion Trust, or lack of customer uncertainty about product quality, is fundamental

to efficient operation of the market institution. Under conditions of high qual-

31

ity uncertainty, “lemons” (inferior products) often drive high quality products out of the market because of the bad reputation they create for other industry products. Consequently, customers require greater assurances to purchase a product in the event it is found after the purchase to be a lemon (Akerlof, 1970). The potential for product liability suits and litigation can significantly dampen the commercialization of an innovation. The creation of trust represents a particularly significant entry barrier for product innovations that are costly, technologically sophisticated, and whose purchase entails irreversible health or welfare situations for customers. Numerous mechanisms are often established to counteract this quality uncertainty entry barrier, including guarantees, licensing practices, industry regulations, as well as endorsements by other trusted institutions.

The costs to create and maintain these industry-wide institutional mechanisms are borne both by industry members collectively and by individual firms. They are both products of, and constraints upon, the legitimacy of individual entrepreneurs or firms to engage in the commercial introduction of a technological innovation. One of the ways in which these firms collectively create and maintain these institutional legitimating devices is through industry councils, technical committees, and trade associations (Maitland, 1982 ) . These industry associations, in turn, approach, educate, and negotiate with other institutions and governmental units to obtain endorsements and develop regulatory procedures.

(c) Technology standards One of the concrete manifestations of industry legitimation is the setting of

technical standards pertaining to component specifications, processes, and performance criteria that designs of a new technology are expected to achieve. Such technical standards are powerful institutional mechanisms for selecting dominant designs from among competing technological possibilities, and re- ducing many uncertainties of technological development by channeling the directions of resource investments and technological change. Besen and Saloner (1989, p. 178) and Tushman and Rosenkopf (1990) describe various ways in which standards develop. In some cases standards are mandated by governmental regulatory agencies. In others, voluntary standards are established co- operatively, with information being exchanged, technologies being altered, or side payments being made to achieve a consensus among firms in an industry. Finally, the setting of standards may be left to the market or the de facto standards imposed by a dominant producer.

Whatever means are used, the typical process for setting standards is influenced more by social and political dynamics than by technical considerations (David, 1987; Tushman and Rosenkopf, 1990). These socio-political dynamics vary with the relative benefits to public and private parties of promoting a standard, the extent to which interested parties (producers, consumers, regu-

32

lators) have different views about the standards that are chosen (Besen and &loner, 1989), as well as the evaluation complexity of a new technology (Tushman and Rosenkopf, 1990). Inherent to this standard setting process is the paradox of cooperation and competition. It may clearly benefit all firms to cooperate to set up industry standards, but in doing so, each firm may try to ensure that standards that suit it best get institutionalized. As Besen and Sa- loner (1989, p. 219) state, standards may be used as tools of competitive strategy, with firms either seeking incompatibility or promoting their preferred technology as the standard to gain an advantage over their rivals. An understanding of this paradox can offer valuable insights on how firms learn to cooperate to sustain themselves collectively, while at the same time compete to carve out their distinctive position in an emerging industry.

3. Propositions on innovation emergence

Van de Ven and Garud’s (1989) system framework summarized here maps a conceptual territory of the essential components of an infrastructure for innovation at the interorganizational community level of analysis. Perhaps more than anything else, it helps one get a handle on the key elements of an industrial community. While the framework demands a more encompassing perspective of innovation than has often been used in the past, it integrates an eclectic body of literature which has argued that each of the system components are necessary (not sufficient) conditions to foster the development and commercialization of technological innovations. Thus, while many of these components have been studied in varying degrees by different disciplines, they have been treated as “externalities” (Porter, 1980) to the system under investigation. But by doing so, one is not likely to study or understand how these functions are interdependent in time and space. By incorporating these different components within an overall systems framework, one can undertake a systematic research agenda aimed at understanding how various actors and functions interact over time to create an infrastructure that both facilitates and constrains technological innovation.

This research agenda could be motivated with the following overall norma- tive proposition:

Proposition 1. The odds of a firm successfully developing a technological innovation are largely a function of the extent to which the institutional arrangements, resource endowments, and proprietary components of the social system are established at the industrial community level.

Embedded in this overall proposition are a number of macro and micro issues that are central to the framework. From a macro policy viewpoint, to under-

33

TABLE 2

Event tracks for studying innovation development and industry emergence

Events occurring over time

Institutional Arrangements

- Legitimation events - Regulation events - Technology standards events

Resource Endowments - Basic research events - Financing events _ Education and training events

Proprietary Functions

- Applied R&D events - Testing and evaluation events - Manufacturing and sourcing events _ Marketing and distribution events

Research Approach

Datum: event (an incident when change occurred in each track above); Observe: date, actor, action and outcome of each event; Analyze: sequence of events in development of each track and temporal relationships between system tracks.

stand the process of innovation is to know (1) how and when different components in the system emerge and are organized over time, (2) what actors create and perform these components, and (3 ) what consequences various arrangements of this community infrastructure have on the time and cost that it takes to develop and commercialize various innovations. While I propose that this infrastructure is crucial to an individual firm’s success at innovation, seldom can or does a single entrepreneurial firm perform all the functions required to create this system. Thus, from the viewpoint of an individual entrepreneurial firm, three key decisions are necessary: (1) What functions will the entrepreneurial firm perform? (2) What other organizations should the firm link or contract with to have the other functions performed? Consequently, (3) What organizations will the firm compete with on certain functions and cooperate with on others? As these questions suggest, one way to understand the implications of the systems framework is to examine the process of innovation at two levels of analysis: (1) the system level looking at the community infrastructure as a whole and the inter-relations between its components or functions, and (2) the behavior of individual entrepreneurs and firms within the industrial system.

One way to study this proposition and related questions is to undertake lon-

34

gitudinal research which tracks the development process of a few innovations from their inception to their conclusion (implementation or termination) in their natural field settings. As historians have done, process is here defined as a developmental sequence of events (Van de Ven, 1992 ) . This means that the datum, or unit of observation is an event; and the central task of longitudinal field work is to observe and record the incidents and activities pertaining to each system component that occur over time during the emergence of an innovation. Various public and private actors are involved in these events. Events are defined as incidents when actors engage in developing or changing each of the system functions listed in Table 2. Following the procedures suggested by Van de Ven and Poole (1990) a chronological recording of these events as they occur over time can then be analyzed to determine when system functions emerged, the networks of actors involved in developing each function, and how these functions and networks of actors interacted over time to facilitate and constrain innovation development.

3.1. Emergence of the community system

While Proposition 1 argues that the community infrastructure for innovation facilitates and constrains innovation efforts of individual entrepreneurs and firms, it is the latter that construct and change the community infrastructure. Thus, we view the innovation infrastructure at the macro community level as grounded in a theory of action at more micro levels of individuals and firms. This infrastructure does not emerge and change all at once by the actions of one or even a few key individuals. Instead, detailed historical studies suggest that the process of creating an infrastructure for innovation consists of an accretion of numerous institutional, resource, and proprietary events involving many actors who transcend boundaries of many public and private sector organizations.

Usher (1954, p. 60) insisted that the history of mechanical invention is not the history of single inventors nor of random chance events. Gilfillan (1935, p. 5 ) observed “a perpetual accretion of little details . . . probably having neither beginning, completion nor definable limits” in the gradual evolution of ship- building. Constant (1980) found that advances in aircraft propulsion emerged not from flashes of disembodied inspiration but from many incremental changes and recombinations of existing technology and organizational arrangements, which add up to what might be called a technological revolution.

Moreover, there is a systemic nature to technological advances, as demon- strated in studies by Hughes (1983) of electrical power, Ruttan and Hayami (1984) of agricultural innovations, and by Kuhn (1982) and Hull (1988) of science in general. Developments in other complementary technologies, institutions, and resource endowments often explain bottlenecks and breakthroughs in the development of a given technology. Thus, as Rosenberg (1983,

35

p. 49) states, “What is really involved is a process of cumulative accretion of useful knowledge, to which many people make essential contributions, even though the prizes and recognition are usually accorded to the one actor who happens to have been on the stage at a critical moment.”

Discontinuities are inherent to the numerous events in developing the institutional arrangements, resource endowments, and proprietary functions, particularly since they require the involvement of many actors from public and private organizations over an extended period of time. Individual events are often not made known to others, and various “acts of insights” pertaining to technical, resource, and institutional capabilities are often required to over- come bottlenecks. These acts or events accumulate probabilistically; they do not proceed deterministically under the stress of necessity or “progress” (Ro- senberg, 1983). They are possible for only a limited number of individuals and organizations who, by virtue of their different roles, competencies, and available resources, become exposed to conditions which bring both awareness of problems and elements of solutions within their frame of reference. Thus, Usher (1954, p. 67) stated that “emergent novelty becomes truly significant only through accumulation” of many interrelated events of technical and institutional change.

These historical studies suggest that an explanation of how innovations develop should focus on the numerous microscopic events by which institutional arrangements, resource endowments, and proprietary functions emerge over time. This system emerges as a partialljr-cumulative progression of numerous events involving many actors in the public and private sectors who invest resources and perform different roles to develop an innovation. Specifically, our basic proposition on the process of innovation emergence is the following:

Proposition 2. Technological innovations do not emerge by a few discrete events performed by a few key entrepreneurs. Instead, they emerge through accretions of many interrelated institutional, resource, and proprietary events involving many actors in the public and private sectors over an extended period of time.

This process can begin any number of ways and varies by the technology being developed. For example, it can begin with purposeful intentions and in- ventive ideas of entrepreneurs, who undertake a stream of activities to gain the resources, competence, and endorsements necessary to develop an economically viable enterprise. As they undertake these activities, the paths of independent entrepreneurs (acting out their own diverse intentions and ideas) intersect. These intersections provide occasions for interaction and for recog- nizing areas for establishing cooperative and competitive relationships.

Cooperative relationships emerge among the actors who can achieve complementary benefits by integrating their functional specializations. Competi-

36

tive relationships emerge as alternative technological paths become evident and different entrepreneurs or firms “place their bets on” and pursue alternative paths. It must be emphasized that during this initial period, applied R&D is highly uncertain and often dependent on basic science and technology. Depending upon the technological alternative chosen by an entrepreneurial individual or firm, it becomes highly dependent upon different clusters of basic research institutions (universities, laboratories, disciplines) that have been producing and directing the accumulation of basic knowledge, techniques, and experience associated with a given technological alternative.

By engaging in cooperative and competitive relationships and by interacting in the same networks, groups of entrepreneurs in the public and private sectors increasingly isolate themselves from traditional industries by virtue of their interdependencies, growing commitments to, and unique know-how of a new technology. Isolation frees an emerging system from institutional constraints of existing technologies and industries (Astley, 1985), and permits it to develop its own distinctive structural form (Rappa, 1987). Coordination among actors takes place not so much by a central plan or organizational hierarchy, nor through the price mechanism, but mostly through interactions (Mattsson, 1987) and partisan mutual adjustments among actors (Astley and Van de Ven, 1983).

As the number of organizational units and actors gains a critical mass, a complex network of cooperative and competitive relationships begins to accumulate. This network itself becomes recognized as a new “industrial sector”, and takes on the form of a hierarchical, loosely-coupled system. ’ We view this emerging system as consisting of the key entrepreneurs and firms that govern, integrate, and perform all of the functions required to transform a technological innovation into a commercially viable line of products or services delivered to customers. The structure of this system, when fully developed, consists of the institutional arrangements, resource endowments, and proprietary functions illustrated in Table 1.

‘Of course, hierarchy in an industry system is a matter of degree, and some industry systems may be only minimally, if at all, hierarchical. Hierarchy is often a consequence of institutional constraints imposed by political and governmental regulatory bodies. Hierarchy also emerges in relationships with key linking-pin organizations who either become dominant industry leaders or control access to critical resources (money, competence, technology) needed by other firms in the industry. Loose coupling promotes both flexibility and stability to the structure of an industry. Links between subsystems are only as rich or tight as is necessary to ensure the survival of the system (Aldrich and Whetten, 1981: 388). Based on Simon’s (1962) architecture of complexity, Aldrich and Whetten discuss how a loosely joined system provides short-run independence of subsystems and long-run dependence only in an aggregate way. The overall social system can be fairly stable, due to the absence of strong ties or links between elements and subsystems, but individual subsystems can be free to adapt quickly to local environmental conditions. Thus, in a complex, heterogeneous, and changing environment, a loosely joined system is highly adaptive.

37

3.2. Interactions between system components

It is generally recognized that resource endowments precede the development of proprietary and institutional system functions, because basic research (which is the search for a fundamental understanding of natural phenomena) provides the foundation of knowledge that makes possible the commercial birth of a technology (Abernathy, 1978, Rosenberg, 1983; Mowery, 1985). What is less well understood is the process by which a common pool of basic scientific or technological knowledge is appropriated and transformed by private firms into proprietary innovations that can become commercial monopolies.

Success at creating a monopoly by commercializing a new technology does not rest on a unique command of basic research, nor on the control of all the competencies and resources relevant to innovation. Instead, as Stobaugh (1985, p. 107) and Mowery (1985) discuss, it rests on orchestrating a highly uncertain journey by linking with numerous organizations and actors and appropriating the competencies and resources relevant to developing and commercializing the innovation. This journey consists of an interactive search process involving parallel developments in building: basic research, financing, and competence capabilities (the resource endowments), institutional legitimation, regulations, and standards, as well as proprietary appropriation and commercialization of new technological products for new markets.

Different search and linking patterns should be expected for innovations in different industrial sectors. As Nelson and Winter (1977, p. 51) discuss, in many sectors there are many R&D organizations, some profit oriented, some governmental, some academic, doing different things, but interacting in a syn- ergistic way. In particular, in medicine, agriculture, and several other sectors, private for-profit organizations do the bulk of R&D that leads to marketing products, but academic institutions play a major role in creating basic knowledge and data used in the more applied work.

Relatedly, most people understand that R&D is an uncertain business. Un- certainty resides at the level of an entrepreneurial firm, where the ‘best’ way to proceed is seldom apparent and the individuals involved have to be satisfied with finding a potentially promising technological path. Less often understood is that the source of much of this uncertainty confronting individual entrepreneurs and investors resides at the system or community level. As the system framework highlights, if institutional arrangements and resource endowments have not yet emerged for an innovation, proprietary entrepreneurs are exposed to high uncertainties and risks in not knowing what kinds of institutional regulations, technical standards, financing arrangements and specialized competencies will emerge for the innovation. Uncertainties are reduced as these institutional arrangements and resource endowments become established and embodied in a dominant technological design for the innovation. This leads to

38

our third proposition on interactions between proprietary, resource, and institutional components of the system framework:

Proposition 3. The time, cost, and risk incurred by proprietary firms in developing an innovation are inversely related to the developmental progress of building institutional arrangements and resource endowments for the new technology.

A concrete example of this proposition emerged from a longitudinal study in which Michael Rappa and I tracked the development of gallium arsenide integrated circuits in the U.S., Japan, and Western Europe from 1983 to 1987 (see Rappa, 1987). While far more firms and scientists were engaged in the development of this technology in the U.S. than Japan, by 1985 Japan was judged to be several years ahead of the U.S. in commercial developments and applications of the technology. One possible reason for the more rapid advancement of the technology with fewer scientists and engineers is that in Japan a system infrastructure was already well established through the Min- istry of International Trade and Industry (MITI), which encouraged firms who were competing on proprietary technical developments, to cooperate with one another and many other actors in various industry and trade committees. They were meeting to develop commercial applications for the technology, influence industrial governance policies, and create a competence pool through training programs and informal information sharing. In the U.S. no compa- rable industry infrastructure was in place in 1985. Instead, it appeared that many U.S. firms, while investing heavily in their own proprietary R&D proj- ects, were “sitting on the fence” waiting for others to build the industry infrastructure for collective advancement.

Of course, the degree of system change varies with the novelty of the innovation being developed and commercialized. Some innovations change the entire order of things, obsoleting the old ways and perhaps sending entire busi- nesses the way of the slide rule or the buggy whip. Most simply build upon what is already there, requiring modifications in existing system functions and practices. We expect that innovations of different levels of novelty will require differing degrees of change in system functions. For new technologies within established industries, some of the functions, such as governance institutions, may be already established and may change in only subtle, nearly invisible ways. That, however, does not deny their importance. It largely explains why radical new-to-the-world innovations are far more difficult to develop and commercialize than incremental innovations within established industries. Specifically, we propose the following:

39

Proposition 4. The more novel the innovation the greater the changes required in all system functions, and hence, the greater the time and chance of failure incurred in developing the innovation.

Radical or revolutionary innovations not only represent new-to-the-world technologies, they also represent vast departures from existing industrial systems. While development and commercialization of certain radical innovations may require starting from scratch to construct an industrial de novo, more often genuinely new industries emerge by relying on metaphors and adapting institutional arrangements that are carried over from other industries. But, as the study of institutional evolution of radio broadcasting by Le- blebici et al. (1991) indicated, the more metaphors and arrangements that are borrowed from other industries, the more difficult it becomes to build an integrated infrastructure for institutional scaffolding in the new industry. This is because the components of an industrial system do not emerge independent of each other; they are highly interdependent. Many convergent and divergent events occur which become bottlenecks that delay the overall development of the system. For example, in the development and commercialization of cochlear implants, Garud and Van de Ven (1989, p. 516) identified numerous temporal interdependencies in the creation of different system functions:

Basic scientific knowledge had to first ensure safety/efficacy of the technology for use in humans before business firms would become involved. The presence of firms wanting to commercialize cochlear implants was necessary as a thrust for the creation of the FDA (Food and Drug Administration) panel for cochlear devices. FDA’s approval of cochlear devices was necessary for Medicare to extend its coverage for cochlear implants, which in turn is necessary for accessing a wider patient base.

The rate of success for entrepreneurial firms is significantly influenced by the length of time it takes for the system to become established. For an individual entrepreneur, startup funding for a venture represents an initial stock of assets that provides the entrepreneurial unit a “honeymoon” period to develop and commercialize its innovation (Fichman and Levinthal, 1988). These assets reduce the risk of terminating the innovation during its honeymoon period when setbacks arise and when initial outcomes are judged unfavorable. The likelihood of replenishing these assets is highly influenced by the duration of the change process; interest and commitment wane with time. Thus, after the honeymoon period, proprietary efforts at innovation terminate at dispro- portionately higher rates, in proportion to the time needed to develop institutional arrangements and resource endowments for the innovation.

Innovation uncertainty decreases over time as system functions emerge, which define key technical and institutional parameters for the innovation. Correspondingly, transitions from development to commercialization activi-

40

ties often entail shifts from radical to incremental and from divergent to convergent progressions in the development of system functions. Analogous patterns have been observed of innovation processes within organizations, which become more highly structured and stabilized in their patterns and less differ- entiated from other organizational arrangements when innovations are imple- mented (Tornatzky and Fleischer, 1990; Zaltman et al., 1973). Thus, at different periods of technological development, we should expect to find that different functional arrangements are needed to foster innovation, and that different components of the system become the limiting factors that serve as “bottlenecks” to sustained innovation development.

This developmental pattern often culminates in the selection of a dominant design for the technology from among competing alternatives. As Van de Ven and Garud (1992 ) observed in cochlear implants, this selection process is largely produced by a convergence in developments of institutional, resource endowments, and proprietary system functions which emerged to embody prefer- ences for the dominant design. As this dominant design emerges, there is a leveling off in further developments of system functions. Once largely established, the system infrastructure systematically channels and constrains further technological advances in the direction of the dominant design. This leads to our proposition on the temporal dynamics of system development:

Proposition 5. The very institutional mechanisms and resource endowments that initially develop to facilitate proprietary innovation development become inertial forces that constrain subsequent development in the direction of a chosen dominant design.

An examination of these propositions on the interactions between system components as they develop over time should lead one to examine if and how learning occurs between functional events, which could provide guidance as to the next paths taken by actors to develop other parts of the system. By examining the outcomes of alternative paths, one could also identify the feasible sets of paths available in the emergence of an industry. While great inter- industry differences should be expected (Mowery, 1985), only by cumulative longitudinal studies of these developmental progressions between system functions will we come to appreciate how system infrastructures emerge for innovation and entrepreneurship (Dosi, 1982).

3.3. Roles of individual entrepreneurial firms

The social system framework emphasizes that any given entrepreneurial firm is but one actor, able to perform only a limited set of roles, and dependent upon many other actors to accomplish all the functions needed for an industry to emerge and survive. As a consequence, an individual firm must make strategic

41

choices concerning the kinds of proprietary, resource endowments, and institutional functions in which it will engage, and what other actors it will link with to achieve self-interest and collective objectives. These strategic choices make clear that the ways entrepreneurial firms choose to allocate their innovation efforts are variables, and that the lines separating the firm from its innovation community are not sharply drawn, but are fluid and changing fre- quently over time. These choices and transactions evolve over time, not only as a result of individual firm behavior, but just as importantly by the interdependencies that accumulate among firms engaged in numerous components of the emerging industry.

Pragmatically, therefore, firm managers and entrepreneurs should be concerned not only with their own immediate proprietary tasks and transaction modes, but also with those of other firms in their resource distribution channel and with the overall social system. Switching involvements between different system functions and proprietary distribution channels is not inexpensive. In- fluencing one’s own existing channel may be more efficient than switching channels or creating new channels. Also, there is an ongoing tension for each industry participant to organize its own proprietary functions and distribution channels as opposed to contributing to the creation of the industry’s resources and institutional arrangements. Although the former may advance the firm’s position as a first mover in the short run, the latter provides the infrastructure that ultimately will influence the collective survival of the emerging industry. There is an important counter-intuitive implication of these decisions for individual entrepreneurs, which is captured in the following proposition:

Proposition 6. Entrepreneurial firms that run in packs will be more successful than those that go it alone to develop their innovations.

Conventional wisdom is that entrepreneurs act independently and compete to be the first into the market with their new product or service. There are many technologies and industries in which this may lead to successful monopoly profits. However, this practice may lead to unsuccessful results when the innovation involves a new technology for a new industry. Running in packs means that entrepreneurs coordinate (i.e., simultaneously cooperate and compete) with others as they develop and commercialize their innovation. Run- ning in packs is analogous to bicycle racers who cue their pace to one another and take turns breaking wind resistance until the ending sprint.

The argument for running in packs emphasizes that the interests of entrepreneurial actors with a stake in a technological innovation are both inter- twined and divergent (Ben-Ner, 1992). The actors seek both to maximize their total surplus and their respective shares in the surplus. The total surplus amounts to creating an industrial infrastructure that makes it collectively possible to develop and commercialize a new technology for a new market. This

42

draws actors together and drives them to cooperate, since no one actor has sufficient resources, competence, or legitimacy to do it alone. The goal of maximizing individual shares propels actors to compete with each other to reap monopoly profits that derive from introducing a dominant technology or product. However, enlightened actors realize that the probability of economic survival from reaping monopoly shares of an orphan technology are much lower than from gaining relatively small shares of a larger and growing new industry. This is why population ecology studies are finding that having more competitors in a new organizational niche increases the survival probability of its members up to the point where the first-order density term is positive (Han- nan and Freeman, 1989).2 Gaining legitimacy is a key problem in the early emergence of a new industry, and the growth of a critical mass of actors is often a prerequisite for legitimacy. Three corollary propositions will now be offered to elaborate this overall proposition on the self-interested and collective inter- dependence or entrepreneurial actors.

Contrary to industrial economists’ stress on competitive inter-firm relations, the social system perspective emphasizes that there are cooperative and competitive elements in each relationship (Van de Ven et al., 1974). For example, it is easy to understand that a firm needs to establish cooperative relationships with suppliers, distributors and customers in order to make its own activities meaningful. It is also easy to see that other firms who pursue competing technological routes carry out conflicting activities. However, as Matts- son (1987) discusses, there are also important elements of conflict in the relations between cooperating firms that have to do with the negotiation and administration of business transactions and adaptation processes. Between proprietary competitors, there are also elements of complementarity, not only when they cooperate to share resources or develop industry institutional functions, but also when they are complementary suppliers to the same customers.

Indeed, since firms in an emerging industry are often engaged in multiple issues simultaneously, they create a “multiplexity of ties” (Galaskiewicz, 1985, p. 296). Thus, Aldrich and Whetten (1981, p. 392) point out that it is mislead- ing to think of single relations among most firms in an industry. Common forms of multiple linkages between a given set of firms include: exchanging multiple resources, communicating with other firm representatives on industry and trade committees, sharing common pools of knowledge, acquiring personnel trained and socialized in a common pool of competence, friendship and kinship ties, and overlapping board memberships.

‘1 am indebted to Phillip Anderson, serving as editor for this special issue, for suggesting this link between population ecology study findings and my proposition on running in packs.

43

Proposition 7. The greater the number of cooperative and competitive ties between firms, the more stable the inter-firm relationships and the more flexible the overall system.

A rupture in one aspect of a relationship does not sever the other ties, and the latter are often used to correct or smooth over the “severed” link. From an industry perspective, stability through redundance of functions and activities among actors minimizes the negative impact of the loss of services provided by one industry member on the performance of the total system.

Multiple ties among firms emerge over time, and often produce unintended consequences. Prior relationships and transactions among firms in the pursuit of an industry subsystem activity are remembered, and thereby become the infrastructure upon which subsequent relations are based (Van de Ven and Walker, 1984). Galaskiewicz (1985, p. 299) nicely summarizes some of these temporal dimensions:

The networks of resource exchange that already existed among organizations are the infrastructure upon which political coalitions are built. In all likelihood, these resource networks were created out of competitive struggle for survival by self-seeking and self-centered actors, who were seeking to minimize their dependencies upon one another. Now these networks are the infrastructure upon which coalitions to achieve collective goals are built. In turn, as political coalitions become institutionalized, they impinge on the struggle for dominance in the resource procurement/allocation arena.

An appreciation of the temporal dimension of inter-firm relationships also provides important insights on how competitors emerge in an industry. Gen- erally, the literature tends to assume that competitors are profit-seeking entrepreneurs who somehow recognize and seize commercial opportunities by entering lucrative markets. Based on their longitudinal study of the emergence of the cochlear implant industry, Garud and Van de Ven (1989) provide quite a different proposition to explain how industrial competitors emerged:

Proposition 8. Aborted efforts at establishing cooperative relationships turn out to become competitive relationships.

In studying the development of cochlear implants, Garud and Van de Ven (1989) observed two instances in which the efforts of the first mover to initiate cooperative relationships or joint ventures with other research clinics failed, leading to the birth of the firm’s competitors. Initial negotiations of possible relationships with a foreign university and a domestic university did not ma- terialize. Otological scientists and clinicians in each of these two universities subsequently entered into licensing arrangements with two other firms (one a

44

new company startup, the other a subsidiary of a large manufacturer), who two years later became the first-mover’s major competitors.

The proposition that aborted cooperative relationships lead to competitive relationships applies primarily to conditions when a small number of organizations (perhaps the size of an oligopoly) exist with the requisite unique competence or assets necessary for innovation development. Such conditions tend to exist during the early emergence of new-to-the-world technologies, where one can often count on one hand the pioneering firms and inventors world- wide who are pursuing the development of a parallel set of basic research ideas or technological designs. These pioneers make themselves known to one another in a variety of ways: by reporting results of their inventions through patents, professional publications, and association meetings. They thereby come to recognize opportunities for obtaining unique competencies or components needed to advance their own work. If efforts to obtain the needed resources go unconsummated because a cooperative relationship could not be established, then the negotiating parties will go their separate ways by entering into cooperative relationships with other parties in this limited set of pioneers who possess the unique competencies or resources. As Van de Ven and Ring (1992) discuss, this implies a shift to a competitive orientation between the parties who disconfirmed one-another’s initial efforts at cooperation.

Through this and related processes, key first-mover organizations emerge that have extensive and overlapping ties to different components of the emerging technological community, and play the key role of integrating the system. Because they have ties to more than one subsystem of a community, these first movers are the nodes through which a network is loosely joined (Aldrich and Whetten, 1981). They serve as communication channels between industry participants, and link third parties by transferring resources, information, or specialties within and outside of the industry. By being linked into multiple subsystem functions in the industry, these first movers accumulate a broad base of power for ascending to a dominant position in the industry, which permit them to survive at the expense of peripheral participants.

But these first movers also experience the greatest conflicts of interest in the emerging industry, since they also tend to have the greatest amount of visibility and which limits their abilities to capture significant proprietary ad- vantages. This is because their dominance serves as a model that is imitated by others and diffused throughout the industrial community. Thus, the leading firm that chooses to “go it alone” must bear significant “first mover” burdens which permit free-riding by other industry participants. In return for these burdens, first movers are generally believed to have the greatest degrees of freedom to shape industry rules, technology standards, and product percep- tions in the directions that benefit them the most (Porter, 1985).

However, these first-mover benefits do not appear to be empirically sub- stantiated for technologies with weak appropriability regimes that is, those

45

that are easy to imitate, reverse engineer, or substitute (Teece, 1987). Ander- son and Tushman (1990) found that the original breakthroughs in cement, glass, and minicomputers almost never became the dominant design except where strong patent protection existed. This research leads to the following caveat:

Proposition 9. The technological design of the first mover often turns out not to become the dominant design that ultimately yields the greatest profits.

This is because while striking out to be the first to introduce a new technology, the first mover will inevitably make mistakes. And the followers, who are ob- serving the practice of the first mover, can make adjustments in their own technologies. As a result, after the first mover has introduced the product in the market, then the second, third, and fourth movers (who have been carefully following the leader) can often and rapidly introduce a more significant, advanced, and better product or service. In short, there are strong economic mo- tives for first movers to run in packs, not alone.

Inherent in all the above relationships among firms engaged in an emerging industry is the paradox of cooperation and competition. Each firm competes to establish its distinctive position in the industry; at the same time, firms must cooperate to establish the industry infrastructure. Olson (1965, p. 10) summarizes the paradox:

If the firms in an industry are maximizing profits, the profits for the industry as a whole will be less than they might otherwise be. Almost everyone would agree that this theoretical conclusion fits the facts for markets characterized by pure competition. The important point is that this is true because, though all the firms have a common interest in a higher price for the industry’s product, it is in the interest of each firm that the other firms pay the costin terms of the necessary reduction in output-needed to obtain a higher price.

Another example that pertains to institutional arrangements is that it clearly benefits all firms to cooperate to set up industry standards. However, in doing so, each firm will try to ensure that the industry standards which suit it best get institutionalized.

One of the major reasons that has been offered for the origin of industry regulation is that it is an institutional means to address these collective action dilemmas, in which individual firms do not voluntarily act in a designated way to achieve benefits for all industry participants (Mitnick, 1980, pp. 164-165 ) . Institutional ways to guarantee such action must be devised to provide the benefits. Otherwise, individual self-interest may lead some members to “free ride” on whatever group benefits may have been obtained by others.

46

4. Conclusion

We believe the social system perspective and its associated propositions make four contributions to understanding the emergence of technological innovations. First, the social system framework provides a holistic perspective for examining the generative process by which a new-to-the-world technology is developed and commercialized, and the roles of public and private sector actors in creating an infrastructure that supports technological development. By tak- ing the interorganizational community or network as the relevant unit of analysis, the framework provides a more inclusive perspective of an industry than does the traditional industrial economics definition of a group of firms competing to produce similar or substitute products. In addition too this proprietary subsystem, the framework examines the industrial infrastructure that supports and constrains innovation. This infrastructure includes resource endowments of basic knowledge, financing mechanisms, and competent labor, as well as an institutional governance structure that legitimates, regulates, and standardizes the activities of industry members. Although an eclectic body of literature and research exists to substantiate the importance of these component functions of a community infrastructure, they have been treated as “externalities” (Porter, 1980). By incorporating these functions within a conceptual framework, one can undertake a systematic research agenda aimed at understanding how various actors and functions interact to create an infrastructure that both facilitates and constrains innovation.

Second, this research agenda has the objective of developing a theory on the infrastructure for innovation at the macro community level that is grounded in a theory of action at more micro levels of individuals and firms. We proposed that the odds of a firm successfully developing an innovation are largely a function of the extent to which this infrastructure is developed at the industrial community level. While this community infrastructure facilitates and constrains entrepreneurial firms, it is the actions of individuals and firms that construct and change the community infrastructure. But this infrastructure does not emerge and change all at once by the actions of one or even a few key individuals. Instead, we proposed that this infrastructure emerges through the accretion of numerous institutional, resource, and proprietary events which co-produce each other through the actions of many public and private sector actors over an extended period of time. This generative process has a dynamic history which in itself is important to study systematically if one is to understand how novel forms of technologies, organizations, and institutions emerge.

Third, we argued that institutional arrangements, resource endowments, and proprietary functions are highly interdependent and co-produce each other over time. The framework allows us to examine the dynamic interplay in the

47

progression of the different system components over time, as we have shown elsewhere in the emergence of the cochlear implant industry (Garud and Van de Ven, 1989; Van de Ven and Garud, 1992). Indeed, the very institutional arrangements and resource endowments created to facilitate industry emergence can become inertial forces that hinder subsequent technological development and adaptation by proprietary firms. However, while institutional arrangements, resource endowments, and proprietary functions co-produce each other, they are not expected to be completely determined. When tracking a highly uncertain generative process, there should be large unexplained components of chance, noise, or error in the process.

Finally, the social system perspective emphasizes that the debate on corporate revitalization through innovation is not limited to the for-profit sector; numerous entrepreneurial actors in the public and not-for-profit sectors play crucial roles which enable a new technology to be developed and commercialized. It also enables one to examine the different roles and events of these different actors, and how their roles affect the roles of other actors as well as their joint contributions to the emergence of the entire industry. This in turn makes it possible to understand how the time, cost, and direction of proprietary innovation efforts are significantly influenced by developments in the overall infrastructure for innovation.

Acknowledgments

We gratefully appreciate useful suggestions on earlier drafts of this paper from Philip Anderson, Joseph Galaskiewicz, Elaine Romanelli, and two anon- ymous JET-Mreviewers. Support for this research has been provided (in part) by a grant to the Strategic Management Research Center at the University of Minnesota from the Program on Organization Effectiveness, Office of Naval Research, under contract No. N00014-84-K-0016.

References

Abernathy, W.J., 1918. The Productivity Dilemma: Roadblock to Innovations in the Automobile

Industry. Johns Hopkins University Press, Baltimore, MD. Abernathy, W.J. and Clark, K.B., 1985. Innovation: Mapping the winds of creative destruction.

Res. Policy, 14: 3-22.

Akerlof, G.A., 1970. ‘The market for lemons’: quality, uncertainty and the market mechanism. Q. J. Econ., 84: 488-500.

Aldrich, H., 1979. Organizations and Environments. Prentice Hall, Englewood Cliffs, NJ. Aldrich, H. and Whetten, D., 1981. Organization-sets, action sets, and networks: Making the most

of simplicity. In: PC. Nystrom and W.H. Starbuck (Eds.), Handbook of Organizational De-

sign. Oxford University Press, Oxford, pp. 385-40’7. Allen, R.C., 1983. Collective invention. J. Econ. Behav. Organ., 4 (1) (March): l-24.

48

Allen, T.J., 1977. Managing the Flow of Technology. MIT Press, Cambridge, MA. Anderson, P. and Tushman, M.L., 1990. Technological discontinuities and dominant designs: a

cyclical model of technological change. Admin. Sci. Q., 35: 604-633.

Arrow, K.J., 1962. Economic welfare and the allocation of resources for innovative activity. In: R.R. Nelson (Ed.), The Rate and Direction oflnuentioe Activity. Princeton University Press, Princeton, NJ.

Astley, W.G., 1985. The two ecologies: Population and community perspectives on organizational evolution. Admin. Sci. Q., 30: 224-241.

Astley, W.G. and Van de Ven, A.H., 1983. Central perspectives and debates in organization theory. Admin. Sci. Q., 28: 245-273.

Ben-Ner, A., 1992. Cooperation, conflict, and control in organizations. In: S. Bowles, H. Gintis and B. Gustafson (Eds.), Democracy and Markets: Participation, Accountability, and Effi-

ciency. Cambridge University Press, Cambridge. Besen, SM. and Saloner, G., 1989. The economics of telecommunications standards. In: R.W.

Crandall and K. Flamm (Eds.), Changing the Rules: Technological Change, International

Competition, and Regulation in Communication. The Brookings Institution, Washington, DC. Binswanger, H.P. and Ruttan, V.W., 1978. Induced Znnouation. Johns Hopkins University Press,

Baltimore, MD. Campbell, D.T., 1974. Evolutionary epistemology. In: P.A. Schilpp (Ed.), The Philosophy of Carl

Popper. (The Library of Living Philosophers, Vol. 14, No. I and II.) Open Court, La Salle, IL, pp. 413-463.

Chandler, Jr., A.D.,,1990. Scale and Scope: The Dynamics of Industrial Capitalism. Harvard University Press, Cambridge, MA.

Constant, E.W., 1980. The Origins of the Turbojet Reuolution. The Johns Hopkins University Press, Baltimore, MD.

David, P., 1987. Some new standards for the economics of standardization in the information age. In: P. Dasgupta and P. Stoneman (Eds.), Economic Policy and Technological Performance.

Cambridge University Press, Cambridge, pp. 206-239. DiMaggio, P.J. and Powell, W., 1983. The iron cage revisited: Institutional isomorphism and

collective rationality in organizational fields. Am. Social. Reu., 48: 147-161.

Dosi, G., 1982. Technological paradigms and technological trajectories. Res. Policy, 11: 147-162. Dosi, G., 1988. Sources, procedures, and microeconomic effects of innovation. J. Econ. Lit., 26:

1120-1171. Fichman, M.A. and Levinthal, D.A., 1988. Honeymoons and the liability of adolescence: A new

perspective on duration dependence in social and organizational relationships. Working Paper, GSIA, Carnegie Mellon University, Pittsburgh, PA.

Freeman, C., 1986. The Economics of Industrial Innovation. MIT Press, Cambridge, MA. Galaskiewicz, J., 1985. Interorganizational relations. Annu. Rev. Social., 11: 281-304.

Garud, R., 1990. Roles of researcher sub-communities in the development of a new technology: The case of cochlear implants. Working Paper Draft, Stern School of Business, New York University, New York, NY.

Garud, R. and Van de Ven, A.H., 1989. Innovation and the emergence of industries. In: A.H. Van de Ven, H. Angle and M.S. Poole (Eds.), Research on the Management of Znnouation. Harper- Collins, Ballinger Division, New York, NY.

Gerwin, D., 1981. Relationships between structure and technology. In: P. Nystrom and W. Star- buck (Eds.), Handbook of Organization Design. Oxford University Press, London, Chap. 1, pp. 3-38.

Gilfillan, S.G., 1935. The Sociology of Invention. MIT Press, Cambridge, MA. Greenhouse, S., 1992. Attention America: Snap out of it!. New York Times, (Sunday, February

9), Section 3: 1.

49

Hakansson, H., 1988. Industrial Technological Deoelopment:A Network Approach. Croom Heim, London.

Hannan, M.T. and Freeman, J., 1977. The population ecology of organizations. Am. J. Social., 82: 929-964.

Hannan, M. and Freeman, J., 1989. Organizational Ecology. Harvard University Press, Cam- bridge, MA.

Hirsch, P.M., 1975. Organizational effectiveness and the institutional environment. Admin. Sci. Q., 20: 327-344.

Hughes, T.P., 1983. Networks of Power: Electrification in Western Society, 1880-1930. Johns Hopkins University Press, Baltimore, MD.

Hull, D.L., 1988. Science as a Process: An Euolutionary Account of the Social and Conceptual Deuelopment of Science. The University of Chicago Press, Chicago, IL.

Jewkes, J., Sawers, D. and Stillerman, R., 1958. The Sources of Invention. Macmillan, New York. Kuhn, T.S., 1982. The Structure of Scientific Reuolutions. Richard Irwin, Homewood, IL. Leblebici, H., Salancik, G.R., Copay, A. and King, T., 1991. Institutional change and the trans-

formation of interorganizational fields: An organizational history of the U.S. radio broadcasting industry. Admin. Sci. Q. 36 (3): 333-363.

Maitland, I., 1982. Organizational structure and innovation: The Japanese case. In: S. Lee and G. Schwendiman(Eds.), Management by Japanese Systems. Prager, New York, NY.

Mansfield, E., 1985. How rapidly does new industrial technology leak out? J. Ind. Econ., 34 (2 ) (December): 217-223.

March, J.G. and Olsen, J.P., 1989. Rediscovering Institutions: The Organizational Basis of Poli- tics. Free Press, New York, NY.

Marx, K., 1906. Capital, Vol. I. Modern Library, New York, NY (originally published by Charles H. Kerr, Chicago).

Mattsson, L.G., 1987. Management of strategic change in a ‘Markets-as-networks’ perspective. In: A. Pettigrew, (Ed.), The Strategic Management of Change. Basil Blackwell, London.

McKelvey, B., 1982. Organizational Systematics: Taxonomy, Evolution, Classification. Univer- sity of California Press, Berkeley, CA.

Metcalfe, J.S. and Soete, L., 1983. Notes on the evolution of technology and international competition.‘Paper presented at the’workshop on Science and Technology Policy, University of Manchester (April).

Meyer, J.W. and Rowan, B., 1977. Institutionalized organizations: Formal structure as myth and ceremony. Am. J. Social., 83(2): 340-363.

Mitnick, B.M., 1980. The Political Economy of Regulation: Creating, Designing, and Removing Regulatory Forms. Columbia University Press, New York, NY.

Mowery, D.C., 1985. Market structure and innovation: A critical survey. Presented at conference onNew Technology as Organizatio’nal Innovation at the Netherlands Institute for Advanced Studies in Humanities, Wassenaar.

Mowery, D.C. and Rosenberg, N., 1979. The influence of market demand upon innovation: A critical review of some recent empirical studies. Res. Policy, (April).

Mueller, W.F., 1962. The origins of the basic inventions underlying DuPont’s major product and process innovations, 1920-1950. In: R.R. Nelson (Ed.), The Rate and Direction of Znuentioe Actiuity. Princeton University Press, Princeton, NJ.

Nelson, R.N., 1982. Government and Technical Progress: A Cross-industry Analysis. Pergamon Press, New York, NY.

Nelson, R.N. and Winter, S.G., 1977. In search of a useful theory of innovation. Res. Policy, 6: 36- 76.

Nelson, R.N. and Winter, S.G., 1982. An Evolutionary Theory of Economic Change. Belknap Press of Harvard University Press, Cambridge, MA.

50

North, D.C., 1990. Institutions, Institutional Change and Economic Performance. Cambridge University Press, London.

North, D.C. and Thomas R.P., 1973. The Rise of the Western World. Cambridge University Press, London.

Oliver, C., 1991. Strategic responses to institutional processes. Acad. Manage. Reu., 16( 1): 145- 179.

Olson, M., 1965. The Logic of Collective Action: Public Goods and the Theory of Groups. Harvard University Press, Cambridge, MA.

Ouchi, W.G. and Bolton, M.K., 1987. The logic ofjoint research and development. Working paper, Graduate School of Management, University of Southern California-Los Angeles, Los Ange- les, CA.

Pfeffer, J. and Salancik, G., 1978. The External Control of Organizations. Harper and Row, New York, NY.

Porter, M.E., 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors. The Free Press, New York, NY.

Porter, M.E., 1985. Competitive Advantage. Free Press, New York, NY. Powell, W.W., 1990. Neither market nor hierarchy: Network forms of organization. In: L.L. Cum-

mings and B. Staw (Eds.), Research in Organizational Behauior. JAI Press, Greenwich, CT, Vol. 12, pp. 295-336.

Powell, W.W. and DiMaggio P., 1991. The Necu Znstitutionalism in Organizational Analysis. Uni- versity of Chicago Press, Chicago, IL.

Rappa, M., 1987. The structure of technological revolutions: An empirical study of the development of III-V compound semiconductor technology. Unpublished Dissertation, University of Minnesota, Carlson School of Management, Minneapolis, MN.

Rappa, M., 1989. The dynamics of R&D communities: Implications for technology strategy. Paper presented at the CORS/TIMS/ORSA Conference, May, 1989, Vancouver, Canada.

Roberts, E.B. and Hauptman, O., 1986. The process of technology transfer to the new biomedical and pharmaceutical firm. Res. Policy, 15: 107-119.

Romanelli, E., 1991. The evolution of new organizational forms. Annu. Reu. Social., 17: 79-103. Rosenberg, N., 1983. Inside the Black Box. Technology and Economics. Cambridge University

Press, Cambridge. Rosenbloom, R.S., 1966. Product innovation in a scientific age. New Ideas for Successful Mar-

keting. Proceedings of the 1966 World Congress, American Marketing Association, Chicago, 11, Chap. 23.

Rosenbloom, R.S., 1985. Managing technology for the longer term: A managerial perspective. In: K.B. Clark and C. Lorenz (Eds.), The Uneasy Alliance: Managing the Productivity Technol- ogy Dilemma. Harper and Row, Cambridge, MA, Chap. 7.

Ruttan, W.W., 1978. Induced institutional change. In: H.P. Binswanger and V.W. Ruttan (Eds.), Induced Innovation. Johns Hopkins University Press, Balitmore, MD, Chap. 12.

Ruttan, V.W. and Hayami, Y., 1984. Toward a theory of induced institutional innovation. J. Develop. Stud., 20(4): 203-223.

Scott, W.R., 1987. The adolescence of institutional theory. Admin. Soi. Q., 32: 493-511. Smith, A., 1937 (1776). The Wealth of Nations. Modern Library, New York, NY. Stern, N. and El-Ansery, A.I., 1982. Marketing Channels. Prentice Hall, Englewood Cliffs, NJ. Stigler, G.J., 1957. Perfect competition, historically contemplated. J. Polit. Econ., 65: l-16. Stobaugh, R., 1985. Creating a monopoly: Product innovation in petrochemicals. In: R. Rosen-

bloom (Ed.), Research on Technological Innovation, Management and Policy. JAI Press, New York, NY, Vol. 2, pp. 81-112.

Teece, D., 1987. Profiting from technological innovation: Implications for integration, collabora- tion, licensing, and public policy. In: D.J. Teece (Ed.), The Competitive Challenge. Harper- Collins, Ballinger Division, New York, NY.

51

Thirtle, C.G. and Ruttan, V.W., 1986. The role of demand and supply in the generation and diffusion of technical change. Bulletin No. 86-5, University of Minnesota Economic Development Center, Minneapolis, MN.

Tornatzky, L.G. and Fleischer, M., 1990. The Processes of Technological Znnouation. D.C. Heath and Company, Lexington, MA.

Tushman, M.L. and Romanelli, E., 1985. Organizational evolution: A metamorphosis model of convergence and reorientation. In: B. Staw and L. Cummings (Eds.), Research in Organiza- tional Behuuior, 7: 171-222.

Tushman, M.L. and Rosenkopf, L., 1990. On the organizational determinants of technological evolution: Towards a sociology of technology. Graduate School of Business, Columbia Univer- sity, New York, NY.

Usher, A.P., 1954. A history of mechanical inventions. Harvarduniversity Press, Cambridge, MA. Utterback, J.M., 1974. Innovation in industry and the diffusion of technology. Science, 183: 620-

626. Utterback, J.M. and Abernathy, W.J., 1975. A dynamic model of process and product innovations.

Omega, 3 (6): 639-656. Van de Ven, A.H., 1992. Suggestions for studying strategy process: A research note. Strat. Manage.

J., 13: 169-188. Van de Ven, #.H. and Garud, R. 1989. A framework for understanding the emergence of new

industries. Res. Technol. Znnoout. Manage. Policy, 4: 295-225. Van de Ven, A.H. and Garud, R., 1992. Innovation and industry development: The case of cochlear

implants. Discussion Paper, University of Minnesota, Strategic Management Research Cen- ter, Minneapolis, MN.

Van de Ven, A.H. and Poole, M.S., 1990. Methods for studying innovation development in the Minnesota Innovation Research Program. Organ. Sci., l(3) (August): 313-335.

Van de Ven, A.H. and Ring, P.S., 1992. The development of cooperative interorganizational relationships. Discussion Paper, University of Minnesota, Strategic Management Research Cen- ter, Minneapolis, MN.

Van de Ven, A.H. and Walker, G., 1984. The dynamics of interorganizational coordination. Admin. Sci. Q., 29(4): 598-621.

Van de Ven, A.H., Emmett, D. and Koenig, Jr,. R., 1974. Frameworks for interorganizational analysis. Organ. Admin. Sci. 6 ( 1) .

von Hippel, E., 1986. Cooperation between rivals: Informal know-how trading. MIT, Sloan School Working Paper #1759-86, Cambridge, MA.

von Hippel, E., 1990. The impact of ‘sticky data’ on innovation and problem-solving. Working paper #3147-go-BPS, MIT Sloan School of Management, Cambridge, MA.

Williamson, O.E., 1975. Markets and Hierarchies. Free Press, New York, NY. Zaltman, G., Duncan, R. and Holbeck, J., 1973. Innovations and Organizations. Wiley, New York,

NY.

van de ven - 1993 - a community perspective on the emergence of innovations

Documents

technology management

community infrastructure

technological community

innovation development

innovation managers

university of minnesota

political infrastructure

community perspective