A Model of the Interaction of Strategic Behavior, Corporate Context, and the Concept of Strategy Author(s): Robert A. Burgelman Source: The Academy of Management Review, Vol. 8, No. 1 (Jan., 1983), pp. 61-70 Published by: Academy of Management Stable URL: http://www.jstor.org/stable/257168 Accessed: 16/05/2009 23:19 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=aom. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy of Management Review.

http://www.jstor.org

?Academy of Management Review 1983, Vol. 8, No. 1, 61-70

A

Model

of

the

Behavior, and

the

Interaction

of

Strategic Context,

Corporate of Concept Strategy1

ROBERT A. BURGELMAN

Stanford University Based on a review of previous landmark studies and in the light offindings of recent research on internal corporate venturing, a model of the strategic process in large, complex firms is presented under which the propositions "structurefollows strategy" and "strategyfollows structure" can both be subsumed. Current corporate strategy induces some strategic behavior but changes in corporate strategy follow other, autonomous, strategic behavior. The study of the relationships between strategy and structure in large, complex firms remains of central concern to scholars in the fields of strategic management and macro organizational behavior. Previous research has, indeed, produced apparently conflicting propositions regarding the directionality of these relationships. Depending on which body of empirical evidence is used to bolster the argument, "structure follows strategy" and "strategy follows structure" both seem to be valid propositions (Bower & Doz, 1979; Galbraith & Nathanson, 1979; Hall & Saias, 1980). The present paper contributes to the resolution of this apparent contradiction by elucidating further the conditions under which each of these propositions may be valid. The analysis presented here rests on two critical insights. First, both propositions need to be considered in terms of what they imply about the nature of the strategic process. As previous researchers have observed (Bower & Doz, 1979), the strategic process in large, complex firms consists of the strategic activities of managers from different levels in the organization. Second, these strategic activi-

ties are of two kinds. Most strategic activities are induced by the firm's current concept of corporate strategy, but also emerging are some autonomous strategic activities, that is, activities that fall outside the scope of the current concept of strategy. The consequences of this distinction for the strategic process have not previously been made the subject of systematic analysis. Autonomous strategic activities have been documented by the students of unrelated diversification through internal corporate venturing (ICV) (Biggadike, 1979; Burgelman, 1980, Fast, 1979). Fast's study of new venture divisions in large, diversified firms, for instance, has provided incidental evidence of the autonomous nature of the strategic activities involved in new venturing. As a participant in one of the firms in Fast's study observed: Top managementsaw a need for venturesand said, "Go ahead and do it." Nobody reallymanagedor directedit. So the whole companybeganto get into venturesbut therewas no cleardirectionor purpose (1979, p. 76). Biggadike's (1979) large sample study of new entries at the business level of analysis also suggests the autonomous nature of new venture activities. Even though ICV projects required the commitment of substantial amounts of resources over substantial periods of time and changed the scope of the corporate business portfolio when they were successful, there seemed to come little guidance

'Support for this paper from New York University's Graduate School of Business Administration and from the Strategic Management Program of Stanford University's Graduate School of Business is gratefully acknowledged. Michael L. Tushman (Columbia University), Eric J. Walton (New York University), L. Jay Bourgeois, David B. Jemison, and Steven C. Wheelwright (all of Stanford University) have made helpful comments on earlier drafts of this manuscript.

61

tiatives were not the result of an a priori clearly formulated corporate strategy on the part of top management. Rather, the corporate strategy emerged through a somewhat haphazard process. It was the result of final authorizations by top management of strategic projects that had successfully absorbed the firm's excess resources and promised to do so profitably in the future. Chandler raised the important question: "Why did the new strategy which called for a change in structure, arise in the first place?" (1962, p. 14). He refers to the major changes in the external environment that had created opportunities for the use of existing excess resources of the firm. At the conclusion of the study, Chandler refers to Penrose's (1968) work and suggests that his data supports her theoretical analysis of the growth of the firm. Penrose's analysis, however, emphasizes the internal impulse toward growth. She observes that the recognition of opportunities takes place in the mind of managers and is often independent of changes in the external environment. In fact, Penrose forebodes the concept of the "enacted environment" (Weick, 1979): In the last analysis, the "environment"rejects or confirmsthe soundnessof the judgmentsabout it, but the relevantenvironmentis not an objectivefact discoverablebeforethe events(Penrose,1968,p. 41). Implicit in the affirmation of the relevance of Penrose's analysis seems to be the recognition that corporate development was not really the result of top management taking a fresh look at the environment, then formulating a strategy, and then establishing the appropriate structural arrangements to implement the strategy. In reality, the structural rearrangements reflected efforts to consolidate the results of autonomous strategic behavior. The new strategy reflected the recognition of the importance of these strategic actions. In the final analysis, Chandler's study seems to indicate that changes in corporate strategy followed autonomous strategic behavior.

from the firms' current corporate strategy for these ICV efforts. Neither Fast nor Biggadike has attempted to conceptualize the corporate strategic process in which new ventures take shape. Yet both researchers have suggested that the research of Bower (1970) and his students could be useful to conceptualize the corporate strategic processes involved in ICV. Independent of these suggestions, Burgelman's (1980) study of ICV project development in the diversified major firm has found that Bower's model is indeed useful but needs to be extended. This study has provided systematic field data from which the category of autonomous strategic behavior has been induced. It also has provided additional insight in the corporate context processes in which ICV project development is embedded (Burgelman, 1982). These insights concerning the nature of strategic behavior and corporate context processes provide the basis for reanalyzing the landmark studies from which the two apparently contradictory propositions concerning the relationships between strategy and structure have been derived.

Structure Follows Strategy The proposition that structure follows strategy became firmly established as a result of Chandler's (1962) study of the historical development of major U.S.-based industrial firms in the period 1919-1959. Subsequent empirical research in the multinational context, and in firms situated in other countries of the Western world, generally has corroborated the structure follows strategy proposition (Galbraith & Nathanson, 1979). Relatively little can be learned about the strategic process underlying Chandler's proposition from these large sample, verificationoriented follow-up studies. The original field study, however, can be reexamined to evaluate the extent to which the original theoretical generalizations were grounded. Strategy Follows Autonomous Strategic Behavior

Heroic View of Top Management

Chandler's case materials indicate that major structural adjustments followed the experience of severe management problems after strategic initiatives had been undertaken in areas unrelated, or only marginally related, to the traditional lines of business of the firm. The case data also indicate that these strategic ini-

Chandler's case data suggest that multiple layers of management were involved in the strategic initiatives that produced the extensive diversification, and in response to which the new strategy and the new structure eventually emerged. The theoretical generalizations, however, collapse this strategic 62

can be verified in the light of the findings of another major line of research in the field of strategic management.

process into a top management activity. Even though the influence of lower levels in the determination of the content of the strategy is recognized, the major emphasis is on the role of top management. Yet, as the du Pont case materials suggest, top management's influence before the reorganization was very limited, with the real influence over strategic behavior situated at the department head level. It was only after H. Fletcher Brown wrote a penetrating analysis of this situation (Chandler, 1962) that the strategic role of the executive committee (representing the whole corporation) became firmly established. In spite of a preoccupation with the role of top management in the strategic development of the firms studied, Chandler presents data that question the relative importance of this role, ironically even with respect to the decisions to change the structural arrangements: At du Pont, GeneralMotors, and JerseyStandard, the initial awarenessof the structuralinadequacies causedby the new complexitycame from executives close to top management,but who were not themselves in a positionto make organizationalchanges. In all cases, thepresidentgave no encouragementto theproposersof change(1962,p. 308, emphasisprovided). And after reflecting on the meaning of the data, Chandler puts forward another key question: But if the stockholdersand the board becamecaptivesof the fulltimeadministrators,werenot the professional entrepreneursthemselvescaptivesof their subordinates?Were not the informationand alternatives available to the top determined,possibly quite unconsciously,by junior executivesdown the line?Mustnot then the enterpriseor the organization as a whole be consideredresponsiblefor the basic economicdecisions?If this is so, then no individual or team of individualscan be identifiedas the key decisionmakersin the privatesectorof the American economy(1962, p. 313). Chandler concludes that the case data challenge the view that the role of top management was not predominant. This conclusion, however, is based on the observation that the new structure had facilitated top management's role in strategy formulation and entrepreneurship after it had been put in place. The case data relating to the situation before the reorganization do not support the heroic view of the role of top management. Furthermore, the proposition that the new type of structural arrangement would lead to a greater role for top management in the formulation of corporate strategy

Strategy Follows Structure The process oriented line of research in strategic management has taken the concepts of the "decision making" (Cohen, March, & Olsen, 1972; Cyert & March, 1963; March & Simon, 1958) and "institutional" (Selznick, 1957) orientations in organization theory, and the "incrementalist" theory in strategy making (Lindblom, 1959) as its points of departure. These theories allow for a bottom-up conception of strategy formulation in which top management's role is not necessarily critical-one in which the concepts of strategy and structure are not clearly delineated from each other (Bower, 1974), nor are operational decisions delineated from strategic decisions (Ansoff, 1965). Basically, this is Chandler's view upside down. Important empirical research has investigated the usefulness of these theoretical orientations for the understanding of the process whereby key decisions are made in complex organizations. It has extended the theory by clarifying the role of top management in these processes (Aharoni, 1966; Allison, 1971; Carter, 1971). A landmark study concerning the strategic process is Bower's (1970) carefully designed, longitudinal field study of the management of strategic capital investment projects in the "diversified major" firm. Probably the most complex type of divisionalized firm, the latter encompasses an agglomeration of widely diversified but partially related businesses grouped into major divisions whose general managers report to corporate management. Strategy Making-A

Multilayered Process

Bower's study documents the manner in which the strategic capital investment process in such firms is spread over the management hierarchy. Three major subprocesses could be discerned, each of which, in turn, comprised three major phases related to activities of managers at particular levels in the organization. At the product/market level in a division, proposals are defined in technical/economic terms. This definition process is triggered by a perceived discrepancy between strategic business objectives 63

If structureis to shape strategy,what vision shapes structureand how is that vision to be developed? Who has a say in the process?Moreresearchis needed (1979, p. 159).

and existent physical plant capacity available to attain these. Projects survive only if they receive impetus from divisional level management. This impetus process is highly political, because managers at the divisional level are aware that their career prospects depend, to a large extent, on developing a good "batting average" in supporting strategic projects. Thus managers will evaluate proposals in the light of the reward and measurement systems that determine whether it is in their interest to provide impetus for a particular project. At the corporate level, the major contribution is precisely the manipulation of the structural context within which the proposal generation takes shape. Through the manipulation of structural context, top management can influence the type of proposals that will be defined and given impetus. Whereas definition and impetus are primarily, if not exclusively, bottom-up processes, the design of the structural context is primarily, if again not exclusively, a top-down process. Thus, to the extent that capital investment proposals reflect strategic business planning, it is possible to posit that strategy making is both a bottom-up and top-down multilayered process (Bower, 1974). Because of the effects of structural context on the generation and shaping of strategic projects, it also is possible to posit that strategy follows structure. However, to the extent that the structural context reflects a given concept of corporate strategy, Bower's study actually indicates that corporate strategy induces strategic behavior.

A Model of the Interaction of Strategic Behavior, Corporate Context, and the Concept of Strategy Based on the findings of the process study of ICV, and on the insights derived from the preceding review of Chandler's and Bower's studies in the light of these findings, a new model of the strategic process in large, complex firms can be constructed. This new model sheds additional light on the important questions raised by Bower and Doz. It provides a conceptual framework from which the two major, apparently contradictory, propositions in the field of strategic management can be deduced simultaneously. Figure 1 represents this model. Variation, Enactment, and Strategic Behavior This model, inductively derived, is isomorphous to the variation-selection-retention model currently emerging as a major conceptual framework for explaining organizational survival, growth, and development (Aldrich, 1979). Its orientation also is in line with current theoretical efforts (White & Hamermesh, 1981) to integrate research done in industrial organization economics, organization theory, and business policy. The model presented here, however, integrates the business and corporate levels of analysis and applies to the class of firms that are large enough and sufficiently resource-rich to be relatively independent of the tight control of external environment selection. Such firms are able to engage in "strategic choice" (Child, 1972) and, as pointed out earlier, their strategic choice process involves substantive inputs from managers from different levels in the organization. Internally generated variation, resulting from the "enactment" (Weick, 1979) of the environment is, at the minimum, a very important source of variation in such firms (Penrose, 1968). Strategic behavior, in the model presented here, refers to such enactments. The model proposes that two generic categories of strategic behavior can be discerned in such large, complex firms: induced and autonomous. Induced strategic behavior uses the categories provided by the current concepts of strategy to identify oppor-

A Less Heroic View of Top Management Bower's study has provided the basis for further research of the strategic process in various types of organizations and concerning different classes of strategic decisions. These have further elucidated the social and political forces in and around the strategic process (Hofer, 1976). Recently, Bower and Doz have articulated a major implication of this line of research, which concerns an alternative view of the role of top management in the strategic process: Thus, in contrast to strategy formulation as the criticaldirection-settinggeneralmanagementactivity, this new processschool of researchsuggestedan alternative,that is, managingthe strategicprocess (1979, p. 158).

Yet, as the authors point out, it is not clear how the management of the corporate phase can be done: 64

Figure 1 A Model of the Interaction of Strategic Behavior, Corporate Context, and the Concept of Strategy 1 (7)

--

-

-

STRONG INFLUENCE WEAK INFLUENCE

create momentum for their further development. Middle level managers attempt to formulate broader strategies for areas of new business activity and try to convince top management to support them. This is the type of strategic behavior encountered in the study of internal corporate venturing (Burgelman, 1980; Roberts, 1980). The strategic initiatives leading up to the corporate managerial problems documented by Chandler (1962) also would seem to fall under this category. Such autonomous strategic initiatives attempt to escape the selective effects of the current structural context, and they make the current concept of corporate strategy problematical. They lead to a redefinition of the corporation's relevant environment and provide the raw material for strategic renewal. They precede changes in corporate strategy.

tunities in the "enactable environment" (Weick, 1979). Being consistent with the existing categories used in the strategic planning system of the firm, such strategic behavior generates little equivocality in the corporate context. Examples of such strategic behavior emerge around, among others, new product development projects for existing businesses, market development projects for existing products, and strategic capital investment projects for existing businesses. Such strategic behavior is shaped by the current structural context. For instance, it can be judged relatively easily in the light of current evaluation and measurement systems. This is the type of strategic behavior documented by Bower (1970). It follows corporate strategy. Hence, the feedback loop (1) in Figure 1 between concept of strategy and induced strategic behavior. During any given period of time, the bulk of strategic activity in a firm is likely to be of the induced variety. The present model, however, proposes that large, resource-rich firms are likely to possess a reservoir of entrepreneurial potential at operational levels that will express itself in autonomous strategic initiatives. Autonomous strategic behavior introduces new categories for the definition of opportunities. Entrepreneurial participants, at the product/market level, conceive new business opportunities, engage in project championing efforts to mobilize corporate resources for these new opportunities, and perform strategic forcing efforts to

Corporate Context and Selection One of the key insights of the study of ICV, reflected in the model presented here, is that the corporate context within which the strategic process takes place encompasses two distinct, selective processes: structural context determination and strategic context determination. Structural context determination is a broad envelope concept used to denote the various administrative mechanisms that corporate management can manipulate to change the perceived interests of the strategic actors in the organization. In the study of

65

Strategic context determination reflects the efforts of middle level managers to link autonomous strategic behaviors at the product/market level into the corporation's concept of strategy. To do so, the middle level managers must make sense out of these autonomous strategic initiatives and formulate workable, attractive strategies for the corresponding areas of new business development. In addition, they must engage in political activities to convince top management to rationalize, retroactively, these successful initiatives by amending the concept of strategy to accommodate the strategic initiatives. This aspect of the process underlies the proposition that strategy follows autonomous strategic behavior. Thus, strategic context intervenes between autonomous strategic behavior and concept of strategy-(5) and (8) in Figure 1. The intervening effect of structural context is limited here. In the ICV study, this influence was reflected only in the concerns of the actors to demonstrate large potential size and fast growth rate for the ICV projects. Hence, the dotted arrow (6) from structural context to strategic context in Figure 1. The degree to which middle management is successful in activating the process of strategic context determination provides guidance for further entrepreneurial initiatives at the operational level. This is represented by the dotted feed-forward loop (7) in Figure 1. It is a feed-forward loop because it guides further strategic initiatives in a particular new area before this area has become incorporated in the concept of strategy of the firm. It is represented as a dotted line because the guidance is relatively tentative and ambiguous.

ICV, it was found to encompass the choices of top management regarding the overall structural configuration, the degree of formalization of positions and relationships, the criteria for project screening, the measures of managerial performance, and the appointment of middle level managers with particular orientations toward entrepreneurial initiative. Bower (1970), of course, had identified earlier the selective nature of this important process. Structural context determination reflects the efforts of corporate management to fine-tune the selective effects of the administrative arrangements so as to keep (or bring) the strategic proposal generating process in line with the current concept of strategy. This part of the model corresponds to Chandler's propostion that structure follows strategy. Hence, the feedback loop (2) in Figure 1 between concept of strategy and structural context. Over time, this fine-tuning may make the structural context more elaborate, with more rules applied to the induced strategic behavior. As a result, the range and scope of these strategic behaviors may become narrower while their probability of failure may decrease. One major consequence of the increased selective efficiency of the structural context is that fewer of the selected strategic projects have the potential to force a significant change in the concept of strategy. Standardized, quantitative procedures for project screening, uniform categories of strategic planning unit systems, selection of higher level managers with strong corporate orientation in their decision making, all tend to reduce the variation in the strategic proposals selected by the firm and provide the basis for the proposition that strategy eventually follows structure. Thus, structural context intervenes between induced strategic behavior and the concept of strategy-(3) and (4) in Figure 1. This part of the model corresponds to Bower's findings. Chandler's proposition focused on the role of top management in bringing structure in line with new strategy. Bower focused on the effects of structure, given strategy. Both studies paid relatively little attention, at least in the conceptualization of the findings, to the role of autonomous strategic behavior in the process through which corporate strategy becomes articulated and changed. The study of ICV has focused on the latter process. This has allowed identification of the process of strategic context determination.

The Concept of Strategy and Retention From the perspective of a process study, the concept of strategy of large, complex firms can be viewed as the result of the selective effects of the corporate context on the stream of strategic behaviors at operational levels. The present model proposes that the concept of corporate strategy represents the more or less explicit articulation of the firm's theory about its past concrete achievements. This theory defines the identity of the firm at any moment in time. It provides a basis for the maintenance of this identity and for the continuity in strategic activity. It induces further strategic initiative in line with it. 66

divisionalized firm emerged as a new generic type. Once the concept of strategy of the firm has been established through top management's ratification of successful autonomous strategic behavior, structures can be designed and refined to select and shape strategic proposals compatible with this concept of strategy. Bower's study has documented the latter processes. Structural design, however, does not work like a well-calibrated sieve. Autonomous strategic activities continue to escape the selective effects of the structural context by mere chance or because alert actors are able to circumvent, or play to their advantage, the selective mechanisms. In any case, the result can be strategic activity falling outside the established strategy. In a more deterministic sense, structure may motivate or impede strategic activity in unanticipated ways (Greiner, 1972; Mintzberg, 1978). Structure and strategy thus exist in a reciprocal relationship to each other. Depending on which part of the strategic process is observed, both "structure follows strategy" and "strategy follows structure" can be correct propositions. The present paper has attempted to provide further insight in the strategic process of large, complex firms by focusing on the interaction between the corporate level process of relating structure to strategy, and the process of strategic behavior at the product/market and middle levels in the firm. The model presented here accommodates the conventional, normative proposition that corporate strategy induces strategic behavior. In addition, and perhaps more fundamentally, the model reflects the new proposition that the more dramatic changes in the corporate strategy of large, complex firms are likely to have been preceded by autonomous strategic initiatives at the operational and middle levels of the organization: strategy follows autonomous strategic behavior. The complete list of propositions embedded in the model presented in this paper are summarized in Table 1. It is hoped that these will stimulate further theoretical and empirical research in the field of strategic management. The present paper focuses the attention of practitioners of strategic management on the dilemmas that result from the opposing tendencies in large, complex firms toward stability and change. Coherence, continuity, and stability in corporate strategy require the institutionalization of strategic behavior through strategic planning systems. Corporate en-

Corporate level managers in large, diversified major firms tend to rise through the ranks, having earned their reputation as head of one or more of the operating divisions. By the time they reach the top management level they have developed a highly reliable frame of reference to evaluate business strategies and resource allocation proposals pertaining to the main lines of business of the corporation. Top managers, basically, are strategies-in-action whose fundamental strategic premises are unlikely to change (Kissinger, 1979). It therefore is not surprising that corporate management focuses on the manipulation of the structural context to keep strategic behavior in line with the current concept of strategy. In the operating system of the firm, this fosters predictability and integration of strategic activity: strategy-making takes on a "planning" mode (Mintzberg, 1973). To the extent that the current concept of strategy is deeply ingrained in corporate management, its capacity to deal with the substantive issues pertaining to new technological and market developments can be expected to be low. Rather than activating the process of strategic context determination, top management is likely to rely also on the manipulation of the structural context to bring autonomous behavior under control. Ironically, from this analytical perspective, the establishment of a new venture division constitutes a manipulation of the structural context to reduce the variability in the operating divisions rather than the implementation of a strategy of unrelated diversification. Also, from this perspective, Fast's (1979) finding that the position of a new venture division in the corporate context is precarious and Burgelman's (1980) observation of wide oscillations in new venture activity are not surprising. Nor is the finding that the activation of the strategic context requires great conceptual and political skills on the part of middle level managers.

Conclusions and Implications The widening of the scope of a corporation's business portfolio as a result of successful autonomous strategic activity puts strain on its administrative machinery. Periods of unrelated diversification thus are likely to be followed by periods of consolidation. Chandler's study has documented such cycles during the period 1919-1959, out of which the 67

Table 1 Propositions Concerning the Interaction of Strategic Behavior, Corporate Context, and the Concept of Strategy

both induced and autonomous strategic behavior. They attempt to strike the kind of balance discussed in the previous paragraph. "Prospectors" emphasize autonomous strategic behavior. They, however, face the problem of maintaining coherence and continuity in their corporate strategy. "Defenders" emphasize induced strategic behavior based on a very clear concept of corporate strategy. Such firms face the long run danger of a lack of creativity and renewal in their corporate strategy. Finally, "reactors" have neither a clear corporate strategy to induce strategic behavior nor the entrepreneurial capabilities related to autonomous strategic behavior. They find themselves in a dangerously unstable situation. The model of the strategic process presented here seems also relevant for the emerging theory of organizational learning. The concept of strategy of a corporation and the corresponding structural arrangements impound the learning of the firm over time. The concept of strategy provides a more or less explicit, and more or less shared, frame of reference or "paradigm" (Duncan & Weiss, 1979; Jelinek, 1979) concerning the bases of the firm's past success. Not unlike the sociological notion of a paradigm (Kuhn, 1970; Masterman, 1970), it provides guidance for further strategic action in line with it. At the same time, it crystallizes the attitudinal and social factors that were selected together with the cognitive, substantive factors underlying the past success. As such, it also is likely to prescribe, often implicitly and tacitly, attitudes and managerial styles and an ideology deemed necessary for the prolongation of the firm's success. Autonomous strategic behavior, identified here as the major source of strategic renewal, thus is likely to encounter nonrational obstacles in its efforts to convince top management that changes in corporate strategy are necessary. Further research may find it useful to explore these less obvious, potentially entropic (Rifkin, 1980) consequences of a concept of corporate strategy for organizational learning. Such research also could shed more light on the factors-external and/or internal to the firm-that influence the balance between induced and autonomous strategic behavior at any given moment in time, and the evolution of this balance over time. In the same line of thought, further research also could investigate the role of acquisition and divestment as compensatory

(1) The current concept of strategy induces some but usually not all strategic activity in large, diversified firms. Therefore, at any moment in time, the totality of strategic activity of such firms is usually a mixture of induced and autonomous strategic behavior. (2) The current concept of strategy leads to the establishment of a structural context aimed at keeping strategic behavior at lower levels in line with the concept of corporate strategy. In this sense, structure follows strategy. (3) Structural context intervenes in the relationship between induced strategic behavior and concept of strategy. It operates as a selection mechanism on the stream of induced strategic behavior. In this sense, strategy follows structure. (4) Over time, structural context reduces the variation in induced strategic behavior, and may thereby prevent strategic learning on the part of the firm. This is another aspect of the strategy follows structure proposition. (5) Strategic context intervenes in the relationship between autonomous strategic behavior and concept of strategy. Through the activation of the process of strategic context determination, autonomous strategic behavior can become integrated in the concept of strategy of the firm. (6) Structural context intervenes only to a limited extent in the relationship between autonomous strategic behavior and concept of strategy. (7) The activation of the process of strategic context determination has a weak influence on maintaining the volume of autonomous strategic behavior in the firm. (8) Over time, changes in the concept of strategy are the result of the retroactive rationalization of autonomous strategic behavior. This, in turn, changes the basis for the further inducement of strategic behavior.

trepreneurship and the resulting strategic renewal of large, complex firms, on the other hand, require the interlocking autonomous strategic initiatives of individuals at operational and middle levels, and an experimentation-and-selection approach at the corporate level. Maintaining a pragmatic balance between these fundamentally different requirements presents a major challenge for top management. This is evident, for instance, in the problems of dealing with performance differences between divisions (Hamermesh, 1977) and in the need to provide strategic guidance for different types of strategic business units (SBUs) in the corporate business portfolio. The present paper suggests that such challenges may be met more readily by recognizing the different requirements of different strategic situations existing simultaneously in the organization. The distinction between autonomous and induced strategic behavior in the model presented in this paper also provides a theoretical foundation for the deduction of the categories in Miles and Snow's (1978) typology. "Analyzers" are firms high on 68

mechanisms-positive and negative, respectively -for the firm's adaptationefforts throughautonomous strategicbehavior. Finally,the presentpaperillustratesan important characteristicof field research. The conceptual frameworksinducedfrom such researchseldomex-

haust the full content and meaning of the data. Such researchallows progressthroughan iterative process: new conceptuallenses can be brought to bear on old data to generatenew insights.Through this process, the old insightscan be refinedand/or some of their additionalimplicationsrevealed.

References Fast, N. D. The rise andfall of corporate new venture divisions. Ann Arbor, Mich.: U.M.I. Research Press, 1979.

Aharoni, Y. The foreign investment decision process. Boston, Mass.: Graduate School of Business Administration, Harvard University, 1966.

Galbraith, J. R., & Nathanson, D. A. The role of organizational structure and process in strategy implementation. In D. E. Schendel & C. W. Hofer (Eds.), Strategic management. Boston, Mass.: Little Brown, 1979, 249-284.

Aldrich, H. E. Organizations and environments. Englewood Cliffs, N. J.: Prentice Hall, 1979. Allison, G. T. Essence of decision. Boston, Mass.: Little Brown, 1971.

Greiner, L. E. Evolution and revolution as organizations grow. Harvard Business Review, 1972, 50 (4), 37-46.

Ansoff, H. T. Corporate strategy. New York: McGraw-Hill, 1965.

Hall, D. J., & Saias, M. A. Strategy follows structure. Strategic Management Journal, 1980, 1, 149-163.

Biggadike, E. R. Corporate diversification: Entry, strategy, and performance. Cambridge, Mass.: Harvard University Press, 1979.

Hamermesh, R. G. Responding to divisional profit crises. Harvard Business Review, 1977, 55 (2), 124-130.

Bower, J. L. Managing the resource allocation process. Boston, Mass.: Graduate School of Business Administration, Harvard University, 1970.

Hofer, C. W. Research on strategic planning: A survey of past studies and suggestions for future efforts. Journal of Business and Economics, 1976, 28 (3), 261-286.

Bower, J. L. Planning and Control: Bottom-up or top-down. Journal of General Management, 1974, 1, 20-31.

Jelinek, M. Institutionalizing innovation. New York: Praeger, 1979.

Bower, J. L., & Doz, I. Strategy formulation: A social and political view. In D. E. Schendel & C. W. Hofer (Eds.), Strategic management. Boston, Mass.: Little Brown, 1979, 152-166.

Kissinger, H. A. White house years. Boston, Mass.: Little Brown, 1979. Kuhn, T. The structure of scientific revolutions. Chicago: Ill.: University of Chicago Press, 1970.

Burgelman, R. A. Managing innovating systems: A study of the process of internal corporate venturing. Unpublished doctoral dissertation, Columbia University, 1980.

Lindblom, C. E. The science of "muddling through." Public Administration Review, 1959, 19, 79-88. March, J. G., & Simon, H. A. Organizations. New York: Wiley, 1958.

Burgelman, R. A. A process model of internal corporate venturing in the diversified major firm. Research paper #636, Graduate School of Business, Stanford University, 1982.

Masterman, M. The nature of a paradigm. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge. Cambridge University Press, 1970, 59-89.

Carter, E. E. The behavioral theory of the firm and top level corporation decisions. Administrative Science Quarterly, 1971, 17, 413-428.

Miles, R. E., & Snow, C. C. Organizational strategy, structure, and process. New York: McGraw-Hill, 1978.

Chandler, A. D. Strategy and structure. Cambridge, Mass.: M.I.T. Press, 1962.

Mintzberg, H. Strategy-making in three modes. California Management Review, 1973, 16, 44-53.

Child, J. Organization structure, environment, and performance -The role of strategic choice. Sociology, 1972, 6, 1-22.

Mintzberg, H. Patterns of strategy formation. Management Science, 1978, 24, 934-948.

Cohen, M. D., March, J. G., & Olsen, J. P. A garbage can model of organizational choice. Administrative Science Quarterly, 1972, 17, 1-25. Cyert, R. M., & March, J. G. A behavioral theory of the firm. Englewood Cliffs, N. J.: Prentice Hall, 1963.

Penrose, E. The theory of the growth of the firm. Oxford, England: Blackwell, 1968. Rifkin, J. Entropy. New York: The Viking Press, 1980. Roberts, E. New ventures for corporate growth. Harvard Business Review, 1980, 58 (4), 134-142.

Duncan, R., & Weiss, A. Organizational learning: Implications for organizational design. In B. Shaw (Ed.), Research in organizational behavior (Vol. 1). Greenwich, Conn.: JAI Press, 1979, 75-124.

Selznick, P. Leadership in administration. New York: Harper & Row, 1957.

69

White, R. E., & Hamermesh, R. G. Toward a model of business unit performance: An integrative approach. Academy of Management Review, 1981, 6, 213-223.

Weick, K. The socialpsychology of organizing. Reading, Mass.: Addison-Wesley, 1979.

Robert A. Burgelman is Assistant Professor of Management in the Graduate School of Business, Stanford University.

70