University of Southern California Dissertations and Theses

A biogeographic theory of industrial market structure and competitive dynamics

(USC Thesis Other)

A biogeographic theory of industrial market structure and competitive dynamics

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A BIOGEOGRAPHIC THEORY OF INDUSTRIAL MARKET STRUCTURE AND COMPETITIVE DYNAMICS by Robert D. Winsor A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Business Administration) August 1989 Copyright 1989 Robert D. Winsor UMI Number: DP22663 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation Publishing UMI DP22663 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 Pi. p. UNIVERSITY OF SOUTHERN CAUFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90089 Com V ? W 7 8 Z This dissertation, written by Robert D. Winsor under the direction of h..:........ Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of re quirements for the degree of DOCTOR OF PHILOSOPHY D ean o f G raduate Studies D ate ........ * r J l rJ . ? . 9 . ........ DISSERTATION COMMITTEE -—T~ / C hairperson chA- i - ( U ... CONTENTS I. INTRODUCTION AND RATIONALE FOR THE STUDY . . . 1 II. THE BIOGEOGRAPHIC THEORY OF INDUSTRIAL MARKET STRUCTURE AND COMPETITIVE DYNAMICS......... 11 III. LIVING SYSTEMS ANALOGIES .................... 48 IV. THE SIMULATION MODEL........................ 96 j V. RESULTS FROM SIMULATION EXPERIMENT AND EMPIRICAL ANALYSIS .......................... 110‘ ! VI. CONCLUSIONS AND IMPLICATIONS ................ 140, i I .............................................................I APPENDIX 1 ........................................ 148 APPENDIX 2 153 1 WORKS CITED........................................ 170 11 LIST OF FIGURES 1. Distinct Outcomes from a Common Origin .... 14 2. The Biogeographic Theory...................... 17 3. Illustration of Competitive Niche Space . . . 20 4. Niche Position................................. 24 5. Niche Position for Large vs. Small Firms . . . 26 6. Immigration/Extinction and the Competitive Equilibrium 3 0 7. Effects of Immigration Barriers upon the Competitive Equilibrium ................. 33 8. Number of Relevant Market Dimensions and Extinction Rates 3 5 9. Market Consolidation .......................... 40 10. Effects of Consolidation on Competitive Equilibrium ..................... 41 11. The Product Life Cycle as a Function of the Market Potential..................... 69 12. The Product Life Cycle and Competition .... 70 13. The Logistic Curve............................ 79 14. Number of Competitors as a Function of the Market Potential..................... 84 15. Conceptual Development of Living Systems Analogies ............................ 95 16. Simulation Model .............................. 99 17. Simulation..................................... Ill 18. Simulation................................... . 112 19. Simulation..................................... 113 20. Simulation ................................... 114 21. Simulation..................................... 122 22. Simulation..................................... 123 23. Cigarettes 12 6 24. Automobiles................................... 127 25. T i r e s ......................................... 128 26. Home Laundry Equipment........... 129 27. Beer & Malt Liquor............................ 130 28. Beer & Malt Liquor............................ 131 29. Beer & Malt Liquor................... 132 30. Railroads..................................... 133 31. Chewing G u m ................................... 134 32. Commercial Banks .............................. 135 33. Gasoline....................................... 136 34. Pig Iron....................................... 137 35. Rayon & Acetate 13 8 36. Breakfast Cereal .............................. 139 i i i H (M LIST OF TABLES Correspondence of Analogs ..................... 16 Parameters in the Simulation Model ........... 98 i v I. INTRODUCTION AND RATIONALE FOR THE STUDY ! Although the marketing discipline has achieved a substantial degree of development over its relatively |short formal history, this development has often progressed in a relatively unstructured and asymmetrical manner. In the area of marketing strategy, for example, normative frameworks which have remained unchanged in their fundamental assumptions have evolved into increasingly intricate and sophisticated blueprints to be used in formulating business policy and direction. Yet this proliferation of normative dogmata has been accompanied by a relatively negligible amount of research designed to empirically confirm or validate the assumptions which underlie these frameworks. This situation places the field of marketing strategy in a vulnerable and precarious position, since these frameworks constitute the major supporting pillars for the discipline. Furthermore, the lack of empirical research in this area is puzzling, considering the quantity and 1 magnitude of business decisions which are based upon these frameworks and the comparatively greater volume of empirical inquiry which is invested in other areas within the marketing discipline. This paradox is even more bewildering in light of the conspicuous discrepancies r iwhich seem to exist between these strategic frameworks and evidence from economic history. An example of these discrepancies can be seen in the history of the automobile industry. Significantly, the number of competing firms in the automobile industry j achieved an absolute (and rather extreme) maximum in the year 1908. While this may appear as a rather banal fact in itself, when contrasted to the prognosis from product life cycle theory that the number of competitors in an industry is expected to peak in the maturity stage, it I illuminates a substantial incongruity between empirical evidence and marketing theory. It would appear difficult to accept the conclusion that the year 1908 represented a point which was within the bounds of the maturity stage of the automobile industry "life cycle." Nor does the product life cycle depiction of the introduction and growth periods as a relatively "peaceful coexistence of competitors" (Product Marketing 1977, p. 51) escape from illustrating a similar paradox. The pocket calculator market provides a revealing testament to 2 the inaccuracy of this portrayal. Although the five years between 1972 and 1977 in this product category would be difficult to categorize as anything but introduction or early growth (with dramatic sales gains occurring for some time after this period), it was portrayed by one author as an interval where "the competitive carnage had been awesome" (Buaron 1981, p. 24) . The same phenomenon was observed by Aaker and Day (1986, p. 409) in the CT medical scanner market, which "seemed to have unlimited growth i [potential] in 1972 and attracted nearly 30 firms. By the early 1980's most of these firms . . . had been forced out." A final, yet potent illustration of this quandary can be seen in the short and recent history of the personal computer market. In this industry, Business Week observed in 1982 that although sales were expected to grow from (the then current) 6.1 billion dollars to 21 billion dollars by 1986 (a forecast which greatly underestimated the actual rate of growth which would be realized in the industry), competition was already becoming intense. At that time it was noted that, While the sales projections look rosy, the future for small companies entering the market does not. All of a sudden, companies are bumping elbows in what only a year ago seemed a no-fail market opportunity. Now, it is clear that most companies in the market will not survive (Business Week 1982, p. 72). 3 Less than one year later, the same journal announced that although the market for personal computers was doubling yearly, it was "already taking on many characteristics of 'a mature industry," with extreme levels of competition, i I competitive shake-out, and a structure which "is being dominated by a few large suppliers, and [where] marketing and distribution skills are becoming more important that the latest technology" (Business Week 1983, p. 77). i 1 While these potential "exceptions" to the product life cycle model raise questions of an important nature, they do not in themselves fully disprove the life cycle concept. Yet not only does the marketing literature fail to provide any satisfactory explanation for these and other apparently atypical and counterintuitive "anecdotes," but the strategic planning grids and frameworks developed by consulting firms and popularized by countless textbooks, case studies, and articles are also darkened by the shadow cast from these examples. As Aaker (1984, p. 86) notes: Perhaps the single most prevalent and enduring generalization in strategic management is that one should invest in growth markets. Yet it is possible to provide numerous examples of firms that have jumped into growth situations only to endure years of painful losses and ultimately, costly, and sometimes fatal exits during traumatic shakeout phases. 4 How can it be that these two frameworks for marketing strategy--both founded upon the same assumptions of product class growth— can result in such poor predictions of market behavior? Yet these two frameworks, the product life cycle and the portfolio analysis grid, are by virtually any measure the two most accepted and taught cornerstones of marketing strategy. Since the product life cycle provides the conceptual underpinnings for the portfolio analysis models, and since it is from these underpinnings that the portfolio approaches derive their assumptions regarding growth markets, the logical source of confusion would appear to lie in the PLC.1 Yet there have been few attempts to investigate or verify the l.The assumption of non-competitive environments in markets experiencing steady growth is cast into further doubt by evidence from the economics discipline. Historically, the vast majority of corporate mergers and acquisitions have occurred during periods of economic growth and high stock prices in the business cycle, rather than in recessions or business slumps (Hay and Morris 1979; Markham 1955; Nelson 1959, 1966; Scherer 1980). Although these merger "waves" occurred aggregately across most industries within the U.S. economy, it would seem to follow that a majority of individual mergers occurred within growth markets. Other research (Gort 1969) has also demonstrated that merger rates within specific industries are positively related to the growth of these industries. If these mergers had any significant impact upon the number of competing firms within their respective industries (which would necessarily follow if these mergers were horizontal in nature), then the hypotheses derived from the product life cycle model regarding competitive exit (presumably inclusive of mergers) as a function solely of stagnating demand would appear insupportable. 5 assumptions of the product life cycle concept regarding competitive structure. As Wind (1982, p. 63) has noted, Most conventional life cycle analysis recognizes the fact that the nature and number of competitors vary at different product life cycle stages. It is often recognized [theorized?] that at the early stages of the product life cycle the product has little or no competition and at later stages in its life cycle the competition becomes tougher. It might be beneficial, therefore, to examine explicitly the competitive environment and its changes over time. In doing this, Wind suggests that, specifically, the number of competitors should be examined at each stage in time. This approach would allow the assumptions of the competitive environment across the product life cycle to be either verified or refuted. The apparent importance of this task, however, has not inspired any such effort. Most typically, doctoral dissertations consist of an exercise in empirical confirmation of existing theories via the falsification of null hypotheses. Yet the anecdotal evidence described above tends to disconfirm elements of traditional marketing theory such as the product life cycle. The conspicuous implication of this discrepancy is that either a new theory of competitive markets or a reconceptualization of existing theory may be necessary. 6 Although the product life cycle concept has been subjected over many years to a constant trickle of criticism, it has remained a popular and unchanging i ( element in the marketing discipline. In this sense, it would appear to have withstood the criticism well. But i 4 iperhaps this durability is due less to the intrinsic merit of the concept than to the lack of any alternative itheoretical structure of sufficient capacity. The goal of this paper is to provide such a structure which can be used to analyze competitive dynamics within markets, and which may serve as a foundation block for a general theory in marketing— an aspiration which has been pursued by many in the discipline, yet which has eluded most. Toward achieving this goal, the path taken by this paper will lead away from traditional conceptualizations and toward a new theory of market structure and behavior. Although this new theory may be viewed as a relatively extreme departure from current marketing thought, it is derived from a conceptual foundation which is shared in common with the product life cycle and the population ecology model in the management discipline. The remainder of this dissertation will take the following form: 7 II. The Bioqeographic Theory of Industrial Market Structure. This chapter will explain the development and i key points of the biogeographic theory as they relate to I the marketing discipline. The biogeographic theory, like I the product life cycle and population ecology models, is i based upon a living system analogy. Unlike traditional interpretations of this analogy, however, which equate organizations with living organisms, the biogeographic jtheory is constructed upon the premise that organizations are more appropriately analogous to living species. While this may initially appear as both a minor and unjustifiable substitution, it will be demonstrated to lead to a potentially more valid conceptualization of competitive behavior. One major logical consequence of this theory is the hypothesis that the number of competitors in each industry may generally be expected to peak and decline before any slowing of sales growth is evidenced in that industry. This hypothesis will be tested in a later chapter of this dissertation. Ill. Review of Alternative Theories. This chapter reviews alternative theories of market structure and competitive behavior which are based upon the logic of the living system analogy. The genealogy of these theories will be exposed, demonstrating the common conceptual 8 assumptions upon which they are constructed. In doing this, it will be revealed how one critical assumption— the conceptual equivalence between organizations and living organisms— is responsible for an elaborate succession of i conjectures which provide the basis for much of the thought in marketing and management, yet which also confer a poor fit with economic reality. IV. Research Design and Methodology. In an effort to test the validity of the proposed theory, two methodologies will be employed. First, a simulation model will be constructed from the logical relationships of the biogeographic theory. Simulations will then be run using various values for the parameters contained within the model. These values will be experimentally tested in order to establish both the internal validity as well as the external realism of the model. Second, empirical, historical data will be collected from secondary sources for 12 industries. These data will be used to test the realism of the model, and will also be compared to expectations derived from traditional explanations of market structure. V. Results. This chapter will describe the results obtained from an "average" run of the simulation, and will 9 provide an explanation and justification for the parameter values used in this run. Additionally, the empirical evidence will be presented and will be shown to be consistent with the results expected in the biogeographic jtheory and obtained in the simulation model. This evidence will also be demonstrated to be inconsistent with I the results expected under traditional marketing and management theories. VI. Summary. Conclusions. and Imp1ications for Future Research. In conclusion, the "Biogeographic Theory of Industrial Market Structure and Competitive Dynamics" will be offered as an original and unique conceptualization which is capable of providing several significant contributions to the marketing discipline. In addition, this theory embodies implicit ramifications which are potentially important for corporate strategy and national policy decisions. 10 ______ I II. THE BIOGEOGRAPHIC THEORY OF INDUSTRIAL MARKET STRUCTURE AND COMPETITIVE DYNAMICS The biogeographic theory establishes a perspective which facilitates the understanding of competitive economic behavior within differentiated markets, and among unique competitors. Unlike the field of economics, the marketing discipline has little capacity for utilizing theories based upon assumptions of "perfect competition." Yet existing explanations of market structure and competitive population growth are founded upon assumptions which are virtually synonymous with this hypothetical exemplar. Furthermore, while there has been much discussion of "niches" and "competitive advantage" in the marketing and strategy disciplines, there is little in the way of a body of theory which can explain the role of these — often rather vague— concepts. As Day and Wensley (198 3, p. 83) observe, "the challenge to incorporate an orientation toward competitive advantage in both theory and research 11 has not been widely accepted by marketing scholars.1 1 This deficiency has also been noted by Henderson (1983, p. 7) who stated that, The success of any marketing strategy depends ! on the strength of the competitive analysis on i which it is based. Yet present concepts of competitive analysis in marketing are almost useless. There is no logic or conceptual framework that serves as the basis for understanding the consequences of potential alternatives for intervention into a dynamic system. . . . The present concepts may be dysfunctional and an obstacle to understanding. Current conceptualizations of market structure are also incapable of analyzing competition from a dynamic perspective. Yet competition is rarely a static Jphenomenon, and this is one reason why normative frameworks which provide the "one right answer" are conspicuously ineffective in providing functional solutions. In general, and in their current formulation, existing theories of market structure and competitive dynamics would appear particularly incapable of providing relevant insights into the issue of industrial market structure. Many of the fundamental problems which have plagued the traditional biological and ecological analogies (including the product life cycle model) can be overcome, however, if 12 the current analog of an individual organization— the organism— is replaced with the species.2 This apparently minor shift in analogs allows us to draw on a different portion of theory from the life sciences, and so provides a dramatically fresh perspective of industry structure. When this shift is made, the i existing population growth equations used by the population ecology school in management and the product life cycle concept in marketing are completely replaced I with new equations of population growth. That is, rather than focusing upon growth in the population of organisms within a species, as universally applied in the past, we now shift our focus to the growth and evolution of populations of species within a given competitive environment, or biota. Furthermore, this substitution in analogs will vastly alter the direction of logic and result in conclusions which are far removed from current living system models (see figure 1). The reasons for this shift are not arbitrary, and will become evident as we 2.Since firms are represented by species in our model, habitats or ecological environments can be seen to be equivalent to product classes or "industries" in the economic sphere. This is because as industries form the competitive "arena" for firms, habitats or biotic environments constitute the competitive arena for the species which inhabit them. 13 i i F igure 1 D i s t i n c t O u t c o m e s -From si C o mm o n Origin Product Life Cycle S Population Ecology Existing ^ ^Conceptualizations Hew Conceptualization Bi ogeographic Theory Living Syst Analogy i I i t 14 compare the abilities of the existing and new theories to explain economic reality. The term "biogeographic" is used in the title of the theory because it illustrates the fundamental difference between our conceptualization and that of the ecological I !and biological analogies (to be discussed later), and because much of the theory will be drawn from work in biogeography. The reason for this focus upon biogeography is that, while the disciplines of biology and ecology study the viability and development of individual I organisms, biogeography involves the study of species competition, evolution, and movement. Once this foundation point of the biogeographic theory is established, the remaining elements of the theory and corresponding analogs fall into place naturally. These relationships are summarized in table 1, and are elaborated below. The major concepts of the biogeographic theory of industrial market structure are illustrated in figure 2. 15 TABLE 1 CORRESPONDENCE OF ANALOGS ANALOGY ANALOG GENERAL LIVING SYSTEMS BIOLOGICAL ECOLOGICAL EVOLUTIONARY BIOGEOGRAPHIC Organism Species Habitat/ Biota Firm Firm Firm Individual Employees ? Product ? Industry Organizational Product Form Firm Form or Brand ? National National Consumer Product Class/ Economy Economy Market Industry H r i i i Figure 2 The Biogeographic Theory Government Policy Consumer Preference National Economy f Barn fjrsi Immigration Individual Firm Attributes: Overall Market \ Attributes: Niche * Growth rate * Sire Extinction Growh rate * Firm size dist. * Firm niche dist. * Number of dimensions Nature of dimensions * Bates of Immigration/ Extinction * Resources 17 Firms * Each firm can be viewed as occupying a specific niche within its market.3 This niche is multidimensional in !nature, such that each niche comprises specific regions on I each relevant market dimension. In addition, the niche is ! the sum total these regions occupied by the firm along every market dimension.^ Market dimensions are generally common to all competing firms within that market, but each firm’s niche is generally unique. That is, dimensions are 3.Multi-market firms can be viewed as occupying multiple niches. 4. An ecological niche is often characterized in the ecology literature as a "hypervolume1 1 in n dimensional space, with each dimension corresponding to a factor relevant to the organism's survival (Hutchinson 1957). As such, the niche defines both the physical/spatial and the behavioral/functional aspects of each species' characteristics. Competitors who depend upon identical environmental resources for survival are said to occupy the same niche, and the result is often competitive instability. The concept of niche has long been used in the management and marketing disciplines, where each dimension of the niche is assumed to represent a product/market attribute or other less tangible construct. In this regard, the concept of ecological niche is highly congruous to what is often termed "sustainable competitive advantage" (Aaker 1984) , and to the concept of "differential advantage," first used by Davenport (1936, cited by Grether 1983), and later adopted by Alderson (1957, 1965) and Luck and Prell (1968). Alderson, in fact, related the concept of differential advantage to the notion of ecological niche (see Alderson 1957, p. 56, and Hunt, Muncy, and Ray 1981, p. 269). Furthermore, the concept of differential advantage has remained largely undeveloped since the 1960's (Hunt 1983). 18 elements of markets, and define particular attributes within the market, while a niche is the unique identity of ' (T a particular firm (see figure 3). In addition, each firm is generally capable of altering its own niche on at ! jleast some dimensions, and this adjustment of niche space I jcomprises the major component of marketing strategy.6 i * Should any firm’s niche become equivalent to that of any other firm in the same market, either one or both of these firms will be at a high risk of failing, or becoming extinct. Equivalence in this sense refers to a complete correlation or overlap between every region which characterizes the niche on every market dimension. 5.As an example, note the sweetness of cola drinks. The relative degree of sweetness is one dimension of the cola drink market which is relevant to all brands, while the exact sweetness of Pepsi, for instance, defines the niche space occupied by this particular brand on the sweetness dimension. 6.Note, however, that product "positioning" is only a subset of marketing strategy based upon niche theory. That is, the "position" of a product in the consumers' minds embraces only a portion of the totality of niche dimensions which any organization occupies. In this sense, two firms which apparently share the same position with regard to consumer perception may, in fact, exist in niches which are at least partially unique, due to differences in manufacturing process, distribution methods, etc. Thus, "market segments" are not completely equivalent to niches, since some niche dimensions are not incorporated into the definition of market segments or positions. 19 Figure 3 Illustration o-F Competitiue Niche Space Di» B Din C Din A I competitor X owns central position on dimensions A A B competitor Y owns central position on dimensions A A C competitor /, owns central position on dimensions B A C 20 Furthermore, the closer any firm approaches niche equivalence with any competitor, the greater the risk of extinction. Thus, the greater the similarity of regions occupied along any dimension, and/or the greater the number of dimensions with overlapping or identical regions, the greater the risks of extinction. * To reduce the risk of extinction precipitated by niche overlap, firms employ a strategy of differentiation or character displacement.7 This strategy of character displacement enables firms to reduce the amount of niche overlap by shifting the region they occupy on one or more 7.Character displacement describes the phenomenon where two potentially competitive species display a higher degree of attribute similarity in separate environments than they would if in a common environment (Brown and Wilson 1956; Emmel 1973; MacArthur and Wilson 1967; Pianka 1978). Due to the effects of character displacement, multiple species which occur "sympatrically (in the same environment), and which share a high degree of niche overlap (and are thus potentially highly competitive), evolve in a way such that their common characteristics become more divergent over time; thus reducing the intensity of competition between them. Where the same species evolve ••allopatrically" (separately— on different, but similar islands, for example), they would tend to display characteristics which were more adaptively generalized, and thus more commonly similar. In this way, sympatry precipitates a niche shift in one or more of the competing species such that niche overlaps are minimized. Additionally, character displacement becomes more difficult, and thus less likely as the number of highly similar species occupying a habitat increases (Orians and Wilson 1964). 21 dimensions away from those regions occupied by competitors. * Concomitantly, the center region of each market I i |dimension will generally be the most desirable position to i occupy (being near the aggregate "ideal point" of either consumer taste or production economies) such that access to the greatest volume of resources, or consumer dollars 'is maximized in this area.8 As a result, monopolists i will generally employ a strategy of convergence. positioning themselves in this central region, while duopolists will position themselves very close to each other and on either side of this central region. Additional competitors entering the market, also desirous of the central position, will tend to locate near, but to the outside of existing firms. This often results in a clustering of dominant firms around these central regions.9 Unfortunately, the addition of excessive 8.As Kotler (1988, p. 371) notes, "a product in the center minimizes the sum of the distances of existing preferences from the actual product." 9.The concepts here are developed from the work of Hotelling (1929) in the area of spatial economics. Hotelling noted that competitors which establish a position along a single, linear market dimension will do so at a point which maximizes their overall attractiveness to consumers. Aggregately, however, this results in an "undue tendency for competitors to imitate each other in quality of goods, in location, and in other essential ways" (Hotelling 1929, p. 41). 22 numbers of firms which favor this central position results in niche overlap, and a corresponding rise in the i ’extinction risk. That is, the threat of extinction will be highest, ceteris paribus, near the central region of each dimension, and will decrease as the distance from the center position becomes more remote. * Thus, the benefits bestowed by niche centrality (from convergence) are countered by the increased risk of j ■ extinction. These risks can only be mitigated through i character displacement, such that convergence and character displacement (or differentiation) become counteropposing strategies which serve to determine the optimal position on each dimension for each firm (see figure 4). Furthermore, the opposing nature of these strategies leads to an inherent dynamism in niche positions due to competitive struggle. * Additionally, the risk of extinction realized by any particular firm is an inverse function of that firm's size (and to a lesser extent, its age [see Betton and Dess 1984; Hawley 1981; Steindl 1965]), such that small (or new) firms face the greatest threat from extinction, 23 Potential Benefits from Different iation Strategy F igui'B 4 Niche Position Potential Benefits from Conver gence Strategy N ( Ni c h e P o s i t io n o n t hi s d i me n s i o n ) Distance From Center of Dimension 24 ceteris paribus.-1 -0 Since larger firms are less at risk of extinction, ceteris paribus, the hazards of employing a strategy of convergence are lessened (see figure 5) . Thus, larger firms will tend to occupy the central regions of most niche dimensions, while smaller and newer firms I i will be more likely to employ character displacement to position themselves more remotely. Over time, as these I firms grow and strengthen their competitive abilities, j they will tend to move toward the central dimensional i regions, such that a process of convergent regress ion gradually shapes the evolution of a market.11 10.This is the phenomenon Stinchcombe (1965) has termed the "liability of newness." 11.This process of convergent regression is also fueled by new competitors entering the market in remote dimensional regions— a process which in aggregate tends to reduce the benefits available to existing firms from differentiation strategies. As new competitors enter from the peripheries and then gradually move toward the median point as they grow, pressure is exerted upon the competitors which originally occupied the central regions. In addition, this process of convergent regression may occur through either outright (although usually gradual) changes in firms1 niche positions, or through the expansion of existing occupied niche regions toward the dimensional medians via brand or model proliferation. 25 Potential Benefits fro m D ifferent iation Strategy Figure 5 Niche Position for Large us.Small Benefits For Smal1 Fi BeneF i ts For Large,Fi rm D istance From Center o f D im ension Potential Benefits fro m Conver gence Strategy 26 Markets * Markets are populated by a number of firms which compete along one or more of the dimensions which characterize the market. Entry into the market by new I firms, which we will term immigration, will be a direct function of the perceived attractiveness of the market to potential entrants. Similarly, failure, or exit from the {market, which we will call extinction12 (as above) will be an inverse function of the ability for any particular firm to survive (generally by making a profit) in the market.13 12.The terms immigration and extinctions are borrowed directly from the theories of island biogeography (see MacArthur & Wilson 1967). In applying these theories to economic markets, it is important to remember that organizations, unlike species, may exert their own "free will." That is, market exit (extinction) does not necessarily transpire in an involuntary manner. Furthermore, exits from economic markets may take place in the form of mergers or acquisitions. 13.Since our theory is based upon the assumption that individual firms are analogous to living species, competition between rival firms can be compared to "interspecific" competition— or competition between rival species within an environment. This form of competition leads, over time, to a process called competitive exclusion. This interspecific competition differs significantly from the "intraspecific" competition which occurs between like organisms of the same species. Unlike density-dependent selection (the process by which intraspecific competition takes place), which is precipitated by relative exhaustion of resources, "competitive exclusion is expected to occur when competing species-populations are too similar to coexist" (Pielou 1975, p. 116) and is thus a function of insufficient competitive niche differentiation. 27 * The rate of immigration (as measured by the number of firms entering per some period) into any market will be directly related to the number of firms already in the i market until some critical number of firms is reached, and thereafter, inversely related to the number of extant 1 firms.14 Conversely, the rate of extinction (as measured by the number of firms exiting per period) from the market will be always positively related to the number of extant 14. In the early stages of an industry's history, it will often be the case that as the success of pioneer firms is noted by potential entrants, new entry begins to increase at an increasing rate. Furthermore, in addition to exposing the profit potential of the industry, the pioneer firms also pave the way in developing the technological, financial, and marketing foundations for later entrants. This "proliferation of opportunity" as described by Alderson (1957, p. 115), is based upon the notion that the successes of pioneer firms create "opportunity for the entry and survival of new firms." For this reason, early rates of entry into a new industry may be positively related to the number of existing competitors. These are also the results expected under what is known as the "demonstration effect" (Gort & Konakayama 1982; Grilliches 1957). This demonstration effect is soon offset, however, by the growing difficulty new potential entrants experience in defining a unique niche to occupy. The peak in the immigration curve occurs where this difficulty of defining a unique niche (which may be manifest in increased costs or lower profitability) exactly equals the benefits conveyed by the path-breaking and developmental efforts of the pioneer firms. After this point, the costs of differentiation and niche defense begin to increase disproportionately, resulting in a downward-sloping immigration curve. 28 firms.15' 16 The convergence of these two rates will determine the "dynamic interactive" equilibrium number of firms populating the market (see figure 6).17 15.This is because the number of competitors is limited by the amount of unexploited niche space remaining in the market. As the number of competitors in a market increases, fewer unique niche spaces remain, and thus the proportion of niche overlap escalates. This niche overlap leads to increased levels of "competitive exclusion" and thus extinction. 16.The immigration and extinction functions will be unique for each market, of course. Furthermore, these functions can be seen as determining not only the equilibrium number of competitors, but also the rate of competititve turnover within a market. 17.This model of competitive equilibrium is derived from the "species equilibrium" model developed by MacArthur and Wilson (1967) in the field of island biogeography. In this model, the rate of new species immigration onto an "island" is hypothesized to decline as the number of species already present on the island increases, since the difficulty of establishing a unique and viable competitive position on the island rises as the number of competing species grows. As Hutchinson (1959, p. 150) explains it: Early in the history of a community we may suppose many niches will be empty and invasion will proceed easily; as the community becomes more diversified, the process will be progressively more difficult. In MacArthur and Wilson's (1967) original model, the immigration curve monotonically declines. Pielou (1979), however, has suggested that the model be modified to allow the immigration curve to rise initially as the number of species increases. This modification is necessary, according to Pielou, as in some cases the first "colonists" to an island enrich the environment such that the way is "paved," making subsequent colonization temporarily easier. The extinction curve, on the other hand, is expected to climb monotonically with an increase in existing species, since niche overlap will become more severe as competitors grow in number (Hutchinson 1959; MacArthur 29 F i g u r e d» I iRate o f j E n try and Exit. Immigration/Extinct ion Sind -the C o mp e t i t i ue Equ i 1 i bi' i uin mu^pp N um ber o f C om petitors * The erection of barriers to entry around any market will result in a net shift downward in the rate of jimmigration.18 This will result in a concomitant alteration in the convergence point for the immigration and extinction rates, leading toward what is known as the I"static interactive" equilibrium point (see point S2 in I -------------------------- I !1972; Pianka 1974, 1978). 18.Note that the establishment or existence of entry barriers within an industry does not require the assumption of cost advantages (economies of scale, etc.) which are possessed by incumbent firms. As Bain (1956, p. 142) points out, "great entry barriers are more frequently attributable to product differentiation than to scale economies in production and distribution." Government regulation and policies are also a significant determinant of entry barriers in many industries (Friedman 1962; Low 1970; Porter 1980b). Mueller and Hamm (1974) have also conducted empirical research which suggests the potential effectiveness of product differentiation in deterring entry, and Orr (1974), and Comanor and Wilson (1967, 1974) have found advertising to lead to entry barriers. Furthermore, it has been noted (Harris 1979; Hay 197 6; Schmalensee 1978) that brand proliferation is often a highly effective strategy in limiting entry into a market. In this sense, existing producers may proliferate brand offerings in an attempt to enlarge their niche (either by enlarging their domain on one or more existing dimensions, or by increasing the number of occupied dimensions). This enlarged niche space will then serve to limit the amount of unexploited niche space available to newcomers. Note also that as brand proliferation increases the size or number of occupied niches, the chances of extinction for competing firms may be increased. In this way, brand proliferation becomes both a defensive and an offensive strategy. 31 figure 7).19 This static equilibrium point will be at a substantially lower number of competitors than that of dynamic interactive equilibrium, and thus a significantly < lower number of firms can be expected to populate the market after barriers have been established. * Conversely, the rate of extinction will be lower in markets with a great number of relevant dimensions, !ceteris paribus, such that the extinction curve is shifted downward for heterogeneous markets; resulting in a higher 19.As Pielou (1975, p. 146) describes this process: Now visualize what would happen if an island in dynamic interactive equilibrium (hence with S-l species) were suddenly to be enclosed by a barrier that prevented the arrival of any propagules from outside. The number of species would presumably fall to a new steady value, say S2 that may be called the static interactive equilibrium level. . . . Thus we assume that for only some of the S-l species present at dynamic equilibrium is E [the probability of extinction] > 0. The remaining S2 species have E values so close to zero that the probability of their loss from the island in the foreseeable future is negligible. 32 i I I Figure 7 Effects of1 Immigration Barriers upon -the Competitive Equilibrium Rate o f Immigration/ Extinction I e : N um ber o f C om petitors 33 equilibrium level of competitors (see figure 8).20 Alternatively, an increase in the heterogeneity of a market may reduce the competitive intensity among any i !given number of competitors. * In general then, the level of competition (and the number of competitors) in any market will be jointly impacted by the number of relevant dimensions in the market and the existence (and height) of entry barriers. In most cases, competition will be greatest in those markets with no barriers to entry and few relevant dimensions, and lowest in those with many relevant dimensions and high barriers. 2 0.This is because a large number of market dimensions allow competitors to avoid niche crowding and overlap, since the total niche space will be distributed across a greater number of independent attributes. This relationship between number of dimensions and competitive severity is congruous to the area/diversity relationship observed by MacArthur and Wilson (1967). These authors demonstrated that larger habitats are more likely to be able to support a higher number of species in equilibrium than smaller areas. This area/diversity relationship is due to .the positive correlation between habitat size and environmental diversity in the ecological world, however (see MacArthur and Wilson 1967), such that the diversity or number of species within an area is implicitly and positively related to the heterogeneity (or number of viable dimensions) of the environment. 34 i F igurB 8 i Rate of Immigration/ Extinction Number of Relevant Market Dimensions and Extinction Rates number flarHet Dimen sions Number of Competi 35 * Markets are subjected to random events whose origin is external to the markets themselves, and which can be characterized as stochastic in form. These events can be [precipitated by changes in consumer preference, I | governmental policy, national economic trends, technology, and other factors which are exogenous to the control of any individual firm in the industry. Furthermore, these i stochastic effects impact each market dimension uniquely, imparting a beneficial, detrimental, or neutral consequence. Since each firm generally occupies a unique niche, or combination of space along the multiplicity of market dimensions, these stochastic effects also impact each firm uniquely. * Since the niche characterizing each firm comprises multiple dimension regions (each of which may or may not be impacted by each stochastic event), the effect of each event will accrue additively to any particular firm.21 Over time, however, the impact of each stochastic event will affect each firm in a cumulative fashion, and thus the sequential effects of these events will be 21.Each event being weighted in a unique fashion. This conceptualization is obviously a simplified model of reality, since may interrelationships among the stochastic events would likely exist. 36 multiplicative. The net result of these exogenous, stochastic effects is an evolutionary process which is manifest in changes in size (and relative niche space) for each firm in the industry. That is, these effects cumulatively determine the net growth rate for each individual firm in the industry, and these individual growth rates aggregately define the rate of growth for the overall industry. * These stochastic events tend to operate in a proportionate fashion, such that large firms tend to benefit more from advantageous events than their smaller competitors. Concomitantly, large firms are impaired to a lesser extent by disadvantageous effects than are smaller firms. This results, over time, in cumulative benefits to larger firms which are significantly greater than those which accrue to smaller firms, leading to a growing disparity in size between small and large firms. This process of proportionate growth leads toward a skewed distribution of firm sizes in the long run, such that most industries will be populated by a few large firms, and relatively many smaller firms. In many industries this distribution of firm sizes will approach that of the lognormal (or some related function of a 37 highly skewed nature). 22 (Industry firm size distributions are, of course, often constrained by government policy.) 22.It has often been observed that firm sizes within each industry are nearly universally distributed in a skew pattern, and that this pattern is similar across j industries (Buzzell 1981a; Hay and Morris 1979; Simon and Bonini 1958). That is: In virtually all cases the data exhibit a similar pattern; the size distribution of firms is highly skewed with a few large firms, rather more medium-size firms, and a large ’tail' of small firms” (Hay and Morris 1979, p. 501). In general, this skew pattern has most frequently been observed as closely approximating a lognormal distribution (Hay and Morris 1979; Scherer 1980), and is suggested to result from a process of proportionate growth. This phenomenon is directly parallel to what is observed in the natural world. That is, if the size (in terms of number of organisms in the population) of species found in a particular ecosystem is plotted against the number of species in each size category, the result usually closely approximates a lognormal distribution (Krebs 1978; Preston 1948, 1962a, 1962b, 1969; Whittaker 1970, 1972). This " spec ies-abundance curve," as it is called in biogeography, demonstrates the notion that in any given biota there are numerous species of small "size," but few of relatively large "size" (Goodall 1951; Krebs 1978; MacArthur & Wilson 1967; Preston 1948, 1962a). This distribution of species also provides an important theoretical foundation for work in the area of biogeography. That is: If we suppose that the number of individuals belonging to a given species in a community results from the combined effects of a large number of mutually independent causes that are multiplicative in their effect, then the abundance of the species is a lognormal variate (Pielou 1975, p. 47). 38 * A consolidation of previously autonomous markets i j results in niche crowding and overlap, as an increased j number of competitors must now occupy common market dimensions.23 That is, consolidation is effectively the t i ' i superimposition of one or more markets' dimensions upon t . | those of an existing market (see figure 9). The result of this niche overlap will be a substantial increase in the; i ; extinction rate (from movement up the extinction! ; I curve— see figure 10) and a corresponding reduction in the i ; number of firms. This reduction will be substantial and related in magnitude to the number of markets undergoing j consolidation. A consolidation of two previously autonomous but | | equivalent markets, for example, will result in the! I "shake-out" of approximately half of the original, 23.A hypothetical example of this consolidation would be! the "combination" of two previously autonomous "islands" or markets. This could occur either through the! establishment of an economic link between these markets— such that free trade would be facilitated— or, alternatively, the economic immigration of all competitors from these two previously autonomous areas or markets into ! a third "virgin" or unexploited market equal in size to the two former areas combined. 1 40 F igure 10 Effects of Consolidation on Competitive Equilibrium fWhere C is the Aggregate Number of Competitors fron all Markets Experienclmg Consolidation! I Rate of Immigration/ Extinction E Number of Competitors 41 competitors.24 This reduction is an expectation which is unique to the biogeographic theory of population dynamics, and which has several significant implications.25 24.The effects of consolidation are well illustrated in j the field of biogeography. As several researchers have j ! noted (Hallam 1976; Preston 1962b; Simpson 1969; Webb 1 1976), the direct effects of continental drift and union ! are plainly evident in the fossil records of North and I South America. Before the Pliocene age, South America was ' an island continent, virtually isolated from North America | or any other land mass. As the isthmuses of Tehauntepec I and Panama came into existence during the late Pliocene (thus connecting North and South America), the mammalian1 species populations of the two continents were gradually brought together, resulting in "intense intermigration and; competition" (Kurten 1969, p. 183). The result of this; contact and conflict was a clear example of the process of j competitive exclusion. The consequences of this great "Pleistocene exchange" (Simpson 1969) was that of the 55 original "families" of mammals inhabiting the formerly isolated continents (27 in North America, 29 in South| America, and one in common) , only 45 have managed to: survive the effects of competitive exclusion after 5! ; million years. Additionally, Preston (1962b) calculates that this process is only halfway to equilibrium, with the end result arriving at approximately 34 families of: mammals in another 5 million years. (Note that 5 million' years is a very short period of time on a geological scale.) 25.This process of consolidation is more relevant to the "evolution" of the economic competitive arena than may be initially discernible. As the transportation and communication technologies of the United States and other countries rapidly evolved, a rather noticeable shift from i regional competition to national competition occurred. Scherer (1980, p. 67-8) provides an excellent narration of; this historical shift: During the first half of the 19th century, nation-wide concentration of manufactured goods output was undoubtedly much lower than it is now. But markets were predominantly local then. The railroads had not been built on any significant scale; wagon roads were primitive; and the waterways system was circuitous, slow, and blocked 42 in winter. As a result, competitive contact among geographically scattered manufacturers was modest, and the amount of monopoly power they possessed must have been high. As the railroads expanded their coverage . . . and as the spread of telegraph and then telephone service greatly| facilitated communications, something resembling a true national market emerged for the first time. Indeed, if we could measure monopoly power in manufacturing directly, we might well find it to have been at an all-time low between 1870 and 1880, for there was a sharp increase in concentration following 1880. This was attributable to the rapid internal growth of those’ enterprises that proved themselves fit for the j 1 competitive struggle, and even more to an enormous1 number of mergers among previously independent, firms. ! I [ The observation that the first major wave of, horizontal mergers was directly preceded by major advances in communication and transportation technology (and the' concomitant decline in the prices for these services), has also been made by other authors (George and Silberston| 1975; Markham 1955). As Markham (1955, p. 182) observed, "the first great wave of mergers followed a period of, rapid railroad building, and the wave of the 1920's came; with the rise of motor car and motor truck transportation and a new advertising medium, the home radio." The fact that regional monopolies arise as the inevitable result of inefficient transportation technologies are a common observation in the economics literature. As Low (1970, p. Ill) observes, feudal lords, although constrained by inefficient scales of production, profited because they owned the entire supply of grain sold in the village market. "The one missing link which prevented local monopolies from merging into a purely; competitive market was good transportation." Yet these massive consolidations in the history of j American commerce can be explained neither by economies ofi scale theories (Eichner 1969), nor by the existing living systems analogies employed in the management and marketing disciplines. Although some economists have persisted in attempting to explain these "shake-outs" as the result of a fall in the rate of growth in the American economy at j the end of the nineteenth century— an argument congruent J with the supply/demand foundations of the product life j cycle model, these explanations are also untenable' j Furthermore, this process of consolidation may occur on J any market dimension. * The combined effects from the five processes mentioned, above will increase competitive pressures and result in! reductions in the number of competing firms in each industry. These five sources are: 1. Stochastic Processes which serve as a source of j i I environmental volatility; j 2. Barriers to Immigration which allow the extinction I i process to continue without compensation from new entry; 3. Consolidation. which results in niche overlap; I (Eichner 1969) . The simple fact is that these j consolidations occurred during a period of increasing j growth rather than decline (Nelson 1959). The result of these theoretical inadequacies is that, until the present! point, there has been no theory which could effectively! explain this consolidation phenomenon in American economic! history. This situation led one author to the conclusion that "the point has been reached where only in-depth investigations of individual industries over time are likely to shed further light (Eichner 1969). What is intriguing is that books and articles have been written about a wide variety of industries (including many of those presented later). In each of these cases, the authors attempt to explain the shake-out of competition in terms of factors which are specific to the industry being studied. 44 4. Convergent Regression. which leads to niche crowding; 5. A decrease in Resource Availability, in terms of consumer dollars, which may result from events either external or internal to the industry. Since only one of these processes is determined by the j sales growth rate for the industry (number 5), competitive i "shake-outs" are likely to occur in every market, irrespective of the overall sales trends. This is especially true in markets which are susceptible to entry barriers or consolidation effects. The competitive results of these processes will likely be manifest even in markets which are experiencing robust growth in sales, such that j competitive conflict and risk of extinction are to be! t expected in the most abundant periods of each industry1s[ ! history, contrary to every existing theory of competitive! I dynamics.^0 ' 26.In the natural world, interspecific competition (and thus competitive exclusion) may arise in situations characterized by virtually limitless resources. That is: An apparent abundance of food or other requisites does not preclude the occurrence of competition. . . . competition and competitive displacement can occur even when the supply of food (or requisites) is abundant in relation to the animals' immediate needs (DeBach 1966, p. 186) . 45 Specifically, then, this dissertation will contain an empirical test of hypotheses derived from the alternative conceptualizations contained within the product life cycle' and population ecology models (H0), and that contained in I the biogeographic theory (Ha). These are: 1 The number of competitors within each industry is a I ! [ direct and positive function of the aggregate sales levels within that industry. Thus, the number of competitors in; I < i ! most industries will achieve an absolute maxima atj j | I approximately the same point in time where sales levels j 1 peak or plateau, while major industry "shake-outs" will j i occur only after absolute decreases in sales volume are evidenced. Orians and Collier (1963, p. 449) provide a similar; commentary regarding the relative independence between the j competitive exclusion process and the magnitude of; resource availability: The evidence in support of competitive exclusion, derived from six main sources, strongly suggests| that interspecific competition has had anj important influence on the evolution of' contemporary community structure despite the relative abundance of resources with respect to the sizes of the populations using them. 46 ! The number of competitors within each industry is a: I j function of the amount of unexploited "niche space," and! j I ! is not directly related to the level of sales volumej I I I within that industry. Thus, the number of competitors i I will often achieve an absolute maxima significantly prior I to any apparent subsidence of sales growth rates.I ! Concomitantly, competitive "shake-outs" are expected to j i occur even in markets experiencing abundant growth. j 47 ______i III. LIVING SYSTEMS ANALOGIES The biogeographic theory, like the product life cycle model, is one specific derivative of a general group of concepts borrowed from the life sciences known as living systems analogies. These analogies have commonly been employed to explain the function, operation and composition of organizations within the business environment, and on a higher level, the structure of markets. In previous versions of this analogy, the organization is generally likened to a living organism, and the theories of the biological sciences are used to "explain" the actions of organizations— either in isolation, or in relation to other organizations or to the environment. These analogies have existed for centuries, and in varying degrees of specificity. The economics discipline has actively cultivated comparisons between economic entities and biological organisms as well. Boulding (1950, p. 3) asserted that 48 microeconomics was the "study of particular economic organisms and their interaction," and developed this idea j in considerable depth, with individual workers being compared to "atoms," the industry representing a "species," and the marketing activities of buying, holding, and selling various assets as being equivalent to i the life sustaining activities of a living system. The popularity of the living systems analogies result from their simplicity, their grounding in a "hard" science, and their intuitive "correctness." As Hirshleifer (1977, p. 1-2) noted, "the fundamental organizing concepts of the dominant analytical structures employed in economics and in sociobiology are strikingly parallel." From the basic concept of the existence of similarities between economic entities and living [ I organisms, there have evolved four distinct branches of thought. These four branches represent related, but distinct developments of the living system analogy, and provide a foundation for much of business thought, both explicitly and implicitly. 49 The Systems Analogy The first of these branches of thought, and the most basic interpretation of the analogy, is called the general living systems model. This GLS, or "systems" perspective has only recently been developed into a cohesive body of thought, but has already made a considerable impact in both the physical and social sciences (see Boulding 1972; Miller 1978; Reidenbach and Oliva 1981).27 The basic concept underlying the general living systems model is that every life form, no matter how simple, is an organized system of specialized components or subsystems. These subsystems perform specialized functions which are essential for life and which work together to bring either essential matter, energy, or information (or some combination of these) into the 27. It is important to note the distinction between the general living systems model or analogy, and what is known simply as systems theory. Although both perspectives have found their way into the marketing literature, and are often confused, they are essentially dissimilar views. Systems theory is a management science or operations research conceptualization of organizational functions, and attempts to operationalize mathematical relationships between these functions within some form of orderly "architecture" (see Adler 1967; Forrester 1958, 1961; Lewis and Erickson 1969). As such, systems theory employs no form of living system analogy, and does not attempt to equate organizational and life form functions. 50 confines of the organism, process it, and dispose of or pass along other matter, energy or information. Proponents of the systems perspective suggest that it represents a comprehensive and unified body of thought which is relevant and generalizable to many, seemingly (dissimilar disciplines. Miller (1978), considered to be I the "father1 1 of systems theory, asserts that the 19 .critical subsystems which are present in cells are also evident and functionally essential in more complex biological and social structures. That is, there are seven hierarchical levels which are functionally and structurally (but not taxonomically) equivalent, each manifesting some form of all of the 19 critical subsystems. These seven levels, from micro to macro, are: cell, organ, organism, group, organization, society, and supranational system. As noted above, the central thesis of general living systems theory is that cells, organisms, and organizations are functionally equivalent, consisting of (1) mutually interdependent subsystems which (2) actuate equivalent tasks at each level. As Miller (1978, p. 1025) describes these two components of systems theory, "cells and organs of the body receive nourishment from the food which the organism obtains from its suprasystem; the employees of a firm do work and are paid from that company's profits." 51 Thus, organizations or firms are viewed as functionally equivalent to living organisms, with employees being analogous to the cells of the organism. Boulding (1972, p. 80) further reinforces the organization/organism analog by comparing firms within a lgiven product industry to individual organisms which are part of a species: The theory of the general equilibrium of the prices and outputs of commodities, for instance, . . . is clearly a special case of a general system of the utmost importance .... In the simplest formulation of this system, we suppose a number n of interacting populations, each composed of the individuals belonging to a single species. The species here may be biological species, such as hummingbirds, or commodity species, such as automobiles .... These basic concepts of the general living systems model have been applied to the marketing discipline by Reidenbach and Oliva (1981, p. 36) , who proposed that the traditional 1 1 f unct ional1 1 approach of marketing be integrated into a "newer and broader systemic context." Unfortunately, while the logic, and particularly the aesthetic appeal supporting the construction of a unified body of system theory appears compelling, the theory has failed to live up to the expectations cultivated by its proponents. Even Miller (1978, p. 1025) admitted that "the fit between the conceptual system and the empirical findings . . . appears poorest at the highest two or 52 three levels." As a unifying theory, the general living systems approach appears to work well for relatively "micro" analyses of functional relationships within firms. Yet, the theory provides little insight into relationships between or among firms, and thus functions rather poorly as a tool for analyzing either market structure or competitive interaction. The Biological Analogy The second type of living system analogy, and one in which the focus is at a somewhat higher level, is the biological analogy. The biological analogy comprises a group of various models which have been developed independently by researchers in economics, organizational theory, strategic management, and marketing. The biological analogy focuses upon the similarities between individual organizations and living organisms. Unlike the general living systems model, however, which emphasizes the study of functional systems in organizations and organisms, the biological analogy examines the growth and transformation of these entities 53 over time. In this sense, the GLS provides a primarily anatomical/physiological view, whereas the biological analogy examines developmental perspectives of the |organism/organization comparison.28 The most common of the biological analogies is the j jlife cycle model. Examples of this model exist in nearly I every business subdiscipline, and include the family life I cycle (see Wells and Gubar 19 66), the retail institutional life cycle (see Davidson, Bates, and Bass 1976), the franchise life cycle (Lillis, Narayana, and Gilman 1976), the factory life cycle (see Schmenner 1983 ) , the manufacturing process life cycle (Hayes and Wheelwright 1979), the technology life cycle (Aaker 1984; Kotler 1988), the international product life cycle (see Ayal 1981; Mullor-Sebastion 1983; Onkvisit and Shaw 1983; Vernon 1966; Wells 1968, 1972), and the two which are most important to our discussion; the organizational life cycle and the product life cycle. 28.Haire (1959), and Levy and Donhowe (1962), for example, studied the quantitative relationship between the number of employees engaged in "internal" functions (e.g., accountants) and those dealing with "external" environments (e.g., sales) for different size organizations. They then compared these measurements with surface/mass equations used in biology to describe the structure of organisms. The goal of this research was to draw a parallel between organisms and industrial organizations, and demonstrate their mutual dependence upon universal physical laws as they grow in size. 54 In the first of these two most common versions of the biological analogy— the organizational life cycle— the appearance, growth, and eventual decline and disappearance of economic organizations is likened to the process of birth, growth, and death of biological organisms (Penrose 1952). This model of the life cycle originated quite early in the development of business theory,29 and is popularly used to study the development and functioning of a business firm as it passes through various "phases" within a particular environment (see Greiner 1972; Mueller i 1972, for example). I Within the confines of the organizational life cycle model, each firm is assumed to correspond to one , I individual biological organism. Following this, the j . . . I "terminological pairs such as species/industry, | mutation/innovation, evolution/progress, mutualism/ J exchange have more or less analogous denotations" (Hirshleifer 1977, p. 2). Central to the theoretical 29.Marshall (1890), for example, likened the firms of an industry to the trees of a forest. In 1914, the analogy j was presented in the following manner: Indeed the growth of a business and the volume and form which it ultimately assumes are apparently determined in somewhat the same fashion as the development of an organism in the animal or vegetable world. As there is a normal size and form for a man, so but less markedly, are there normal sizes and forms for businesses (Chapman and Ashton 1914, p. 512). 55 foundations of the organizational life cycle analogy is the understanding that the forces which determine the j ! growth and configurative dynamics of organizations are i j inherent in each of the organizations themselves. This | "genetic" perspective of the organization, then, "means i | I looking for lawful processes in organizational growth | I grounded in factors inside the firm, and for the forces jshaping it as it grows" (Haire 1959, p. 272). These !internal forces driving the development of individual ! . . . . . . . firms/organisms are constrained by limitations imposed by the environment and the availability of resources. j Problems Concerning the Biological Analogy Unfortunately, the biological analogies also possess potentially serious limitations. As noted above, the biological analogy is constructed on the premise that the corresponding analog to the individual organization is a ! I single organism, and although this would appear to be a j rational conceptualization for the model— since it is from | this assumption that the concept of "birth," "growth," and ! i eventual "death" of an organization follows— this premise j I 56 requires the forfeiture of a considerable degree of realism. A requisite component of this premise is the notion that organizational development is a function of time or age and some set of characteristics congenital to firms which are related (in the sense that they are in the same market or industry) . That is, each firm is assumed to possess some form of "genetic” component which determines its future characteristics and development. The logical result of this assumption is that firms within a particular industry, or "species," should display sufficiently similar histories (adjusting for the age of each firm), and also, implicitly, should experience equivalent futures or fates— including a predictable decline and "death." The problematic nature of this assumption is evident I in the comparison of real-world situations, however. While individual organisms of a given species, and within a given habitat generally display few and minor discernible differences in anatomy, physiology, or development,30 many markets or industries are characterized by a wide variation between organizations. These variations exist not only in immense size 30.As Pennings (1980, p. 152) notes, "biologically it |Would be absurd to refer to size differences within a population [species]." 57 differentials, but also in structure, focus, and relative success of the firms (Porter 1979b).31 As a result of these differences between biological development and business growth, the biological analogies I are capable neither of accommodating nor explaining i variations among firms, and would thus seem to offer little assistance in illuminating the issue of market t structure. 31.As one author noted: The complexities and uncertainties surrounding the life-cycle concept highlight the intrinsic disparity between the life-cycle concept as it relates to the traditional biological context and its recent application in business situations. In the biological context, there is an inevitability and predictability to life cycles that is somewhat alien to the business situation (Camillus 1989, p. 32) . This criticism is echoed by Kimberly (1980, p. 7), who notes that the biological analogy offers a poor fit to the reality of organizational development: when one goes beyond these intriguing surface similarities, questions begin to arise. . . . First, biological organisms begin to die the minute they are born. Death is an inevitable feature of biological life. The same cannot be said of organizations. . . . Death is not an inevitable feature of organizational life. Second, whereas biological organisms seem to go through relatively clear and predictable stages in development from simple to more complex, the same is not necessarily true of organizations. If there are laws the govern the development of organizations, analogous to those that apparently govern the development of organisms, they are yet to be discovered. 58 The Product Life Cycle The other major version of this life cycle model, the I product life cycle, is the "sister" to the organizational i life cycle model (especially if one confines the analogy I to single-product organizations and the brand level of j analysis) and presents a unique case to the study of markets and competitive dynamics— due to its predominance and popularity within the marketing discipline. For over three decades, the product life cycle concept32 has been used by marketing theorists, teachers, and practitioners as a descriptive device for illustrating, predicting, and analyzing firm and market behavior for product forms, classes, and brands. Furthermore, it is widely acknowledged that few concepts in marketing have been as extensively discussed or as universally accepted as the product life cycle (Buzzell 1981b; Lazer and Shaw 1986; Scheuing 1969). Smallwood (1973, p. 29), for example, describes the PLC as the marketing equivalent of the i periodic table of elements, and states that "like the 32.The references to the product life cycle which follow refer to that of the general product category, or class— often referred to in the economics discipline as the "industry"— unless otherwise specified. 59 periodic table, [the PLC] provides a framework for grouping products into families for easier predictions of reactions to various stimuli." Significantly, the product life cycle's popularity as la descriptive and pedagogical device, and the belief that it "provides insights into a product's competitive dynamics" (Kotler 1988) has precipitated a substantial •volume of normative theory based upon the concept. As a result, the PLC has become a standard and generally j accepted model which is applied in the development and supplementation of marketing strategy theory (Anderson and Zeithaml 1984; Catry and Chevalier 1974; Clifford, 1965; Cox, 1967; Doyle 1976; Fox 1973; Harrell and Taylor 1981; Howard 1983; Karnani 1984; Kotrba 1966; Savich and Thompson 1978; Scheuing 1969; Smallwood 1973; Wasson 1974). The product life cycle also serves as the underlying foundation for every product portfolio management model (Camillus 1989; Day 1977; Enis 1980; Luck and Ferrell 1985). Hofer (1975, p. 798), in fact, calls the stage of the product life cycle "the most fundamental variable in determining an appropriate business strategy," while Wasson (1974, p. 2) asserts that "successful competitive strategy must start with a sound understanding of the phenomena of product life cycles." 60 Although the product life cycle concept continues to be encumbered by several methodological and conceptual flaws, these deficiencies appear to have detracted little from the popularity of the concept among either marketing academics or practitioners. As Enis, LaGarce, and Prell (1977, p. 48) note, "the product life cycle concept will not be forgotten by marketing managers." The product life cycle concept is constructed from three major theoretical components, each originating from apparently unrelated disciplines. From the supply, or producers1 perspective, the conceptual foundation for the product life cycle is derived from the simple biological analogy equating the product with a living organism (Wind 1982). As Wasson (1974, p. 2) explains it, "products and ideas of any kind have a finite life cycle very similar to the life cycle of the animate humans who originate and consume them." Implicit in this analogy is the premise that products undergo a predicable process of development or growth, and face an inevitable and predeterminable mortality.33 That is, products, like living organisms, 33.Note that it is precisely this premise of life cycle uniformity and predictability which is responsible for the numerous attempts to model and forecast product life cycles for specific product brands (Fourt and Woodblock 1960, for example), and for entire product classes (Kovac and Dague 1972, for example). This assumption has also led to "a great deal of effort aimed at predicting the critical product life cycle prior to product introduction (in fact prior to manufacturing commitment)" (Lawrence and 61 develop, grow, and inevitably die in a process which is similar among individuals within the same species (industry). Conversely, the rationale for understanding demand, or jconsumption behavior across the PLC is acquired from the social psychological theories of diffusion and adoption of innovations (Buzzell et al. 1969; Engel, Blackwell, and Miniard, 1986; Nicholls and Roslow 1986; Polli and Cook 1969; Rink and Swan 1979; Wasson 1974; Wind 1982). I Descriptively, then, the theoretical underpinnings of the simple product life cycle are formed from two complementary parts: the biological analogy which equates a product with a living organism, and diffusion of innovation concepts which are employed to rationalize the shape and dynamics of the cycle. Finally, the interrelationships between the simple PLC (for product classes) and various other competitive and imarket variables described profusely in the literature— and the normative propositions which invariably accompany these descriptions— are based upon assumptions from traditional economic theory. With its Lawton 1981 p. 530), and an industry of market researchers engaged in the estimation of future sales of new product concepts. 62 roots firmly planted in this hypothetical-deductive tradition, the product life cycle doctrine spawns a multitude of hypotheses regarding overall market structure and normative recommendations for individual firm behavior over the history of the product category 1 l (Wind 1982 ) . These hypotheses are intuitively and I deductively derived from universally accepted economic theories regarding competitive entry and exit, market power, and competitive response as functions of the supply/demand characteristics of the product market. As a result of this foundation in classic economic theory, and in spite of the fact that there is little empirical support demonstrating the relationship of product life cycles to competition or marketing strategy (Buzzell et al. 1969; Day 1981; Wind 1982), these hypotheses have assumed the position of marketing axioms.34 These axioms have subsequently become the basic foundation blocks for normative strategies which are universally endorsed and which "represent a consensus of what most marketers would advise" (Kotler 1988, p. 3 66) . That is, "there is an implicit assumption that the life cycle stage should determine, within very broad limits, the strategy adopted (01Shaughnessy 1984, p. 149). 34.Webster's dictionary defines an "axiom" as a "proposition regarded as a self-evident truth." 63 In brief, the product life cycle concept explains the structure of the market and competitive response as the I result of the growth in product (class) sales over time— as the product progresses from "introduction" to 1 ; "growth" to "maturity" and finally enters "decline." That is, the growth of the popularity of the product over time and the consequential disparity between the level of demand for the product and its relative supply are I conceptualized as the causal factors of the structure and i competitive characteristics within the industry. Specifically, as the product is introduced, demand for the product is assumed to grow slowly at first. The number of competitors marketing the product in this "introduction" phase is assumed to be minimal, since profits are likely to be low, overall potential demand for the product is relatively unknown, and there has been insufficient time for other firms to launch their own versions. As product demand and overall sales begin to grow at a rapidly increasing rate (the "growth" stage), and as j i positive profits can be evidenced, other potential competitors are expected to rush to introduce their own versions of the product. Eventually, the "saturation" point of the market is reached, and as sales levels cease growing, production capacity (and thus product supply) overtakes market demand. At this point— the "maturity" stage of the product life cycle— competitive interaction is expected to intensify due to the pressures exerted by excess industry capacity. These first three stages of the product life cycle are i [traditionally viewed as constituting an "S-shaped" growth curve (Kotler 1965; Nicholls and Roslow 1986; Wind 1982). That is, product categories first grow slowly as the product and distribution is perfected and as consumer resistance is gradually overcome. This growth rate then accelerates swiftly, and finally slows as the number of potential new buyers dwindles.35 Eventually, as the product becomes less effective at satisfying consumer needs, and as new substitute products appear, demand begins to deteriorate. As the shrinking demand in this "decline" stage leads to oversupply within the industry, the weaker competitors are expected to "shake-out" of the market. Significantly, it should be noted that an implicit, yet critical assumption of the product life cycle model is 35.Because of these assumptions, dual asymptote models such as the logistic and Gompertz functions (which result in S-shaped growth curves) are typically used to model product growth. Wind (1982), for example, explains the traditional S-shape of the PLC as being a direct application of the logistic curve. 65 that there is an inherent, natural limit to growth for every product category (Buzzell 1972, 1981b; Fourt and Woodblock 1960). This "market potential" or "ceiling on demand" (Fox 1973) is a constraining factor in that no product/market is viewed as capable of growing indefinitely.36 36.Some versions of the PLC, in fact, use level of saturation as the vertical axis by which sales are measured (Smallwood 1973). The concept of an inherent "market potential" or saturation point as a limiting constraint for product sales is also derived from the theory of diffusion and adoption of innovations, and is an implicit and determinant factor in models of this process and in models of demand in general (such as Bain 1963; Bass 1969; Chambers, Mullick and Smith 1974; Midgley 1980; Tigert and Farivar 1981). One reason for this demand constraint is that it allows for a computationally simpler model (see Flath and Leonard 1981). In one form or another, every basic model for product growth— which generally employ one form or another of the modified exponential, the logistic, the cumulative lognormal, or the Gompertz curves (Bain 1963; Mahajan and Muller 1981; Wind 1982)— requires some form of limiting "market potential." That is, "eventually the rate of growth decreases as the proportion of adopters gets closer and closer to a maximum" (Polli and Cook 1969, p. 386). Furthermore, as sales of the product accumulate over time, the attributes of those individuals purchasing the product are expected to vary, with the degree of reluctance toward product adoption progressively increasing. Eventually, the product becomes completely diffused throughout the potential market, and only replacement sales are then made. Note also that the noncumulative adoption curve is generally described in the diffusion of innovation literature as a normal distribution (Rogers and Shoemaker 1971; Rogers 1983, for example), which cumulatively yields a logistic curve (Mahajan and Muller 1981). This is one reason why the logistic curve is conceptually relevant as a description of the growth of demand across the product life cycle. 66 As a result of these assumptions regarding market "saturation," the "resources" of the marketplace (in terms of unsold sales prospects) are assumed to be the determinant factors of the competitive characteristics of the industry (Levitt 19 65). That is, the competitive opportunities within a market are viewed as being the difference between the existing industry sales of the product and the eventual capacity or overall potential of 3 |the market in the long run (either absolutely or jcumulatively, depending upon the nature of the good and [the role ascribed to replacement sales).37 As this is ^described in one journal: If one defines market capacity as the total potential sales volume of all competitors in a market, and market volume as the total actual sales volume of all firms, then the product life cycle is the relationship between these two sets of figures. In other words, as the product life cycle advances through its various stages, more and more market capacity is converted into market volume (Product Marketing 1977, p. 46; also see Scheuing 1969, for another example of this relationship). As the product life cycle progresses towards maturity, this "usage gap" between the industry market "potential" 37.This relative level of market saturation is alternatively termed the level of market penetration, which is often defined as the "ratio of current users to [the number of current potential users" (Pessemier 1982, p. 99) . 67 and actual sales (Weber 1976a, 1976b) is expected to diminish, leading to competitive pressures (see figure 11) . The ultimate result of these assumptions regarding the product life cycle is that competitive entry and exit are believed to reflect the opportunities provided by this difference between actual sales and potential sales of the product.38 That is, strong and stable market growth and a large usage gap are considered to be the impetus for competitive entry, while stagnation of overall industry demand and the elimination of untapped market potential are assumed to be the stimuli for competitive exit ("shake-out") from the industry. In this way, the overall number of competitors within an industry is viewed as varying directly with the sales level of the product (see figure 12). This conceptualization is accepted virtually 38.Interestingly, one author (Scheuing 1969) compared the entry of new competitors into an emerging industry to the process of new product adoption described in diffusion of jinnovation theory. Competitive entry was differentiated over the life of the product life cycle, with "pioneer" [firms entering the market initially. These pioneers are then followed by "imitators," "early adopters," and finally, "late adopters." A later article (Gort and Konakayama 1982) also described the net entry of producers into the industry as a "diffusion in production." 1 F ± gure 11 T h e P r o d u c t . L i f e C y c l e ais 3L f u n c t i o n o f t h e Market Potential Markei x ______ Pofeniial ' Usage Gap S a le : Introduction Growth Meturity Decline Time [after Weber 1976a] i 1 I I 69 Sales and Number of Competitors [after Hartley Figure 12 T h e P r o d u c t : L i f e C y c l e a n d Competition Sal es \ Number Competltors Maturity Decline Introduction Growth ► Time 1976J 70 unquestioningly within the marketing community,39 and provides the basic foundation for a significant proportion of strategy development in the discipline.40'41 39.The global degree to which this view is endorsed within the marketing discipline is reflected in the fact that a imajority of articles referring to the product life cycle, [and virtually every marketing and marketing strategy |textbook describe the growth stage of the PLC as a period ,of modest competitive rivalry in which the number of Icompetitors is growing, and the maturity stage as one in which the number of competitors is "many" or at a maximal point (see appendix 1 for examples of these views). 4 0. In what appears to be the only empirical study of the relationship between competitive populations and the product life cycle, Thorelli and Burnett (1981) found that the number of competitors continues to increase throughout the positive growth section (growth through maturity) of the PLC. This tends to support the conventional wisdom that competitive shake-outs only result from an absolute decline in the sales of a product class. Unfortunately, the results of this study appear to be potentially unreliable, since (1) the authors utilized a cross- sectional analysis of PIMS data, (2) all industries were aggregated together, thus obscuring competitive dynamics unique to individual markets, and (3) the "maturity" stage of each industry was determined by date of product introduction, thus requiring the assumption that all products "age" at the same rate and reach maturity at the same age. 41. Note that if every firm in the industry possessed the same exact sales potential and ultimate market share, then the number of competitors would indeed covary directly jWith the product sales level (the product life cycle). (This is because sales volume would then be equally 'distributed across all competitors, and these competitors, being unable to either grow or shrink, would necessarily have their population size determined by the level of 'sales. Of course, this is obviously a highly unrealistic assumption to make. 71 i The Ecological Analogy The third version of the living system analogy is that of the ecological model, and comprises a group of more recent theoretical work which attempts to explain business behavior and industrial structure in terms of ecological homeostasis. The ecological model represents an enhancement of the biological analogy, and integrates an environmental perspective into this representation.42 The most popular variation of the ecological analogy is termed the population ecology model, and has developed a large following within the management and organizational 42.Although the ecological analogy is implicit in much of marketing thought, few writers in this discipline have actually developed it in any explicit fashion. This lack of attention has occurred despite Alderson's (1957, p. 101) suggestion that "the application of ecology to marketing organizations provides a new starting point for ithe study of competition." Alderson (1964) also later proposed a normative theory of marketing which was centered upon an "ecological framework." Thorelli (1967, p. 19) also employed the ecological model, noting that marketing exchange resulted from "a process of interaction (between the organization and its environment." More Irecently, Henderson (1983, 1984) has also suggested the (relevance of the ecological analogy to the marketing discipline. 72 theory disciplines.43 In this sense, it would be difficult to overestimate the influence that the population ecology model has had upon the management discipline. Not only is the population ecology perspective one of the most popular in the management area, with hundreds of books and articles written within this framework in the last decade, but many authors have literally based their entire careers upon this analogy. As a result, this perspective has transcended its status i as a simple analogy, and become a true paradigm. The ecological model retains the organization/organism correspondence, but attempts to study interrelationships among like organizations. Furthermore, the population ecology perspective incorporates an attempt to study the effectiveness of particular organizational "forms," under the assumption that individual organizations can be classified into "populations" which share a common form (Ulrich and Barney 1984). Thus, in this model, the species is viewed as analogous to organizational form, 43.All future references to the "population ecology" model or school of thought are intended to designate the model as it is applied in the management literature, rather than in the ecological literature. 73 rather than an "industry" as in the biological model.44 A key concept introduced by the population ecology jinodel is the process of "natural selection." The critical element of this concept is that viable and superior forms of organizational structure (i.e.: "bureaucratic versus 1 t inon-bureaucratic," "capital intensive versus labor i . |mtensive," and "organic versus mechanical" [Aldrich 1979, « Ip. 108]) arise within the competitive environment not through the efforts of individual firms at optimization, but rather because of an optimization process taking place 1 within the entire environment (Hannan and Freeman 1977). I |In this way, the "laws" of the environment come to be I viewed as moderators between organizational strategy and performance (Prescott 1986), or, more commonly, as the primary or sole determinants of organizational performance 44.Morgan (1986, p. 56), for example, cites the work of Mintzberg in identifying five "species" of organizations: the "machine bureaucracy," the "divisionalized form," the "professional bureaucracy," the "simple structure," and the "adhocracy." McKelvey (1982, p. 192), on the other hand, asserts that "organizational species are polythetic groups of competence-sharing populations isolated from each other because their dominant competencies are not easily learned or transmitted." In this sense, the population ecology concept of species as distinct organizational forms or groups of competencies seems noticeably related to Porter's concept of "strategic groups" as businesses within an industry which are "related" in that they follow a similar strategy (see Frazier and Howell 1983; Porter 1979, 1980a). Similarly, Winter (1971, p. 258) conceptualized economic "species" as a "collection of firms with similar decision rules." 74 (Alchain 1950; Aldrich 1979; Astley and Fombrun 1983; Hannan and Freeman 1977). It is important to note that natural selection as "adaptation" can only occur over many generations, and is a process that enables species, rather than individual organisms, to adapt to the environment. This distinction is significant, as natural selection involves the (environment's discrimination among types of members I (organizations), rather than an individual's (organization's) choice among responses (Hannan and Freeman 1977). Accordingly, this ecological perspective i of "environmental determinism" shifts the factors causal of organizational structure, growth, ad success from the internal, "genetic" forces of the biological model to external, environmental agents. The second major contribution of the population ecology model is based upon what is known as Gause's "principle of mutual exclusion" (see Gause 1934; Hutchinson 1957). According to this principle, no two competitors which share the same niche or resource requirements can coexist continuously within the same 75 environment.45 That is, competitors which share identical modes of existence are generally at a high risk ] jof non-survival. Increasingly, the population ecology models are being 45. It is important to clarify one common source of error at this point. While the marketing, economics, and management disciplines frequently refer to both the principle of mutual exclusion and organizational and market "niches," these terms are consistently and universally misapplied under the current living system conceptualizations. That is, organisms do not fill niches. species do. The principle of mutual exclusion I states that no two species with identical niches can jcoexist simultaneously (Gause 1934; Hutchinson 1957; Ipianka 1978; Whittaker, Levin, and Root 1973). In the natural world, organisms of like species all share the same niche, whereas each species must occupy its own unique niche. Thus, the pervasive assumption within marketing and management that each organization must occupy a unique and distinct niche is irreconcilable with the traditional perspective (founded upon the biological and ecological analogies) that each firm is analogous to an individual organism, since each species would then be populated by organisms which occupied dissimilar niches. Alderson (1957), in fact, is likely to have initiated j this inappropriate usage of the niche concept in marketing. Alderson asserted that each firm must occupy its own unique niche, while likening each firm to a single living organism (see Alderson 1957, pages 52-57 and Hunt, Muncy and Ray 1981, pages 268-269 for evidence of this discrepancy.) Henderson (1983, p. 8) later discusses the role of niche theory and competitive interaction as they relate to marketing, noting that each competitor must "have a unique advantage over all others." Within this discussion, however, Henderson explicitly states that a "species" is "an industry or combination of businesses that share common evolution," and which thus "also share a common gene pool" (p. 9), thereby implying that firms are equivalent to single organisms. Yet, as noted above, the topic of niches is irrelevant to the analysis of organisms. Individual organisms which share the same species designation and thus the same gene pool must, by definition, all occupy the same exact niche and competitive characteristics. 76 used to study the issues of market structure and competitive populations within industries. In addressing these issues, the population ecology school subscribes to what is known as the "density-dependence1 1 model of population growth, which illustrates the growth of organisms of a single species in a finite environment. This model employs the Verhulst-Pearl logistic growth i jequation to model population growth over time, and in | resource constrained habitats (see Brittain and Freeman i 1980; Hannan and Freeman 1977, for examples in the management discipline; see also Pianka 1978, for an example of this logistic growth model in the field of ecology, and see Pearl 1924, for its use in population studies). This logistic equation for population growth is defined as: dN — — = rN dt (From Merrell 1981; Pianka 1978) The underlying assumption of this growth model is that the rate of growth (r) in the population of organizations (N being the population size) is constrained by the amount of unexhausted resource capacity in the environment (K being -l K - N K 77 the "carrying capacity" or ceiling at which resources are fully exploited and the intrinsic rate of growth is zero). The logic behind this conceptualization is intuitively compelling. As Pielou (1977, p. 20) observes: The growth of any population in a restricted environment must eventually be limited by a shortage of resources. Thus a stage is reached when the demands made by the existing population on these resources preclude further growth and the population is then at its "saturation level," a value determined by the "carrying capacity" of the environment. The result of this logistic growth pattern is a sigmoid or S-shaped population curve which asymptotically approaches the carrying capacity of the environment (figure 13).46 In this way, the relative supply of resources in the environment is viewed as the primary determinant of competitive numbers, and on a higher level, the overall structure of the market/industry.47 That is: 46.Note that the Gompertz equation is also employed as an alternative to the logistic (see Waltman 1983). The Gompertz shares the same assumptions as the logistic, the one difference being that the inflection point is reached earlier, thus making the Gompertz vertically asymmetrical. 47.It is important to resist being seduced by the apparent similarity of the interpretation of the word "resource" as it is used in ecological theory and its usage in the economic literature. This is one area where the population ecology school of management appears to have been misled. The authors in the management field, in fact, appear to be either confused or intentionally vague regarding the definition of this term. "Resources" in the ecological science literature are 78 Figure 13 The L o g i s t i c Curue (Population Size) 79 Organizations, like organisms in nature, depend for survival on their ability to acquire an adequate supply of the resources necessary to sustain existence. In this effort they have to face competition from other organizations, and since there is usually a resource scarcity, only the fittest survive. The nature, numbers, and distribution of organizations at any given time is dependent on resource availability . . . (Morgan 1986, p. 66). J A further implication of this logistic growth model is I that environments (markets) having large amounts of unexhausted capacity should experience high rates of * j not analogous to raw materials or industrial components used to build a finished product. Rather, ecological resources are the food, water, air, or other substances required to sustain the life of an organism. A scarcity of resources, then, is an imminent threat to the survival of the organism. Industrial raw materials do not meet this criterion, since a shortage of raw materials or inputs rarely poses a direct threat to the existence of the firm. Shortages experienced by all competitors in an industry are generally merely compensated by bidding up the price of the material and a concomitant increase in the price of the finished good. Shortages can also be dealt with by substitutions in factor inputs. These facts are especially true in markets characterized by the "perfect competition" assumptions made under the ecological analogy (discussed below). The simple fact is that raw materials do not sustain a business concern, but rather sales revenues do. Furthermore, it is not the goal of any growth oriented firm to maximize its intake of raw materials, but rather to maximize its share or intake of consumer revenue. For this reason, consumer dollars would appear a more realistic analog to ecological "resources," since a scarcity of sales or customers is a serious threat to the survival of virtually any firm. Consumer dollars then, are what ultimately determine the survival of business concerns in the long run. 80 growth in the number of organizations, while those environments whose capacity is nearly fully exploited should experience a plateauing of competitive numbers (approaching a level of stable equilibrium), and an intensification of competitive pressures (Hannan and Freeman 1977; Barnett and Carroll 1987; Tucker, Singh, Meinhard, and House 1988). The assumption is that; . . . most populations start with initially low rates of growth, which accelerate over time, reach a maximum rate, and then slowly decline until the population reaches a more or less stable size, the carrying capacity (Tucker, Singh, Meinhard, and House 1988, p. 127). As the carrying capacity of the environment is approached, the resource availability per competitor is reduced to the point where the environment is characterized as "lean." jcompetitive interaction is then expected to increase as this scarcity threatens the survival of many firms.48 The conclusion which is offered is that "lean 48.The implicit result of this assumption is that the process of natural selection among individuals within a population is viewed as being a function of resource scarcity (Aldrich 1979; Astley 1985). As one author asserts: The assumption of competitive saturation is important in this sense: for the environment to optimize, that is, to choose between competitors, the joint demand of those competitors must exceed the available supply of environmental resources . . . (Astley 1985, p. 229). 81 environments . . . promote cutthroat competitive practices" (Aldrich 1979, p. 63). In this sense, the concept of populations whose growth is limited by the finite resources of the environment is plainly Malthusian 49 m origin. 3 What becomes readily apparent is that the population ecologists' conceptualization of the size of competitive populations as being determined by environmental or market capacity— and competitive pressure as being a function of relative resource density— is virtually synonymous with that contributed by the product life cycle model. In both models, competitive intensity between organizations is viewed as a function of the size of the competitive population relative to a "saturation" point, determined by the capacity of the resource environment or market. That is, "when density is low relative to some fixed environmental carrying capacity, competition is minimal" (Tucker, Singh, Meinhard and House 1988, p. 128). This similarity in the explanations of competitive population dynamics of the PLC and ecological models can be implicitly observed by merging figure 11 and figure 12. The result of this union is a graph which shows the 49.Interestingly, Verhulst, who originally derived the logistic equation of population growth, based his model quite strictly upon the assumptions of Malthus. 82 competitive population as constrained by the long-run potential of the market (see figure 14). This, of course, is the concept represented in the logistic growth curve of figure 13. Astley (1985, p. 230) also recognized the implicit 1 i parallels between the population ecology models and the product life cycle. He states that: i Selection within a population begins to operate as available resources within a niche start to become exhausted. In industrial contexts, this occurs in the firm "shake-outs" characteristic of later stages of the product life cycle. This similarity between the population ecology and product life cycle models with regard to the treatment of competitive "populations" results from the fact that both models portray competitive numbers and levels as determined primarily by the availability of "resources" in the environment (or market). In the population ecology model, competitive levels are constrained by the unexhausted "capacity" of the environment (See Hawley 1981) , and "shake-outs" are viewed as resulting from environmental "leanness" which alters the selection process against the "least fit" organizations. In the product life cycle model, the competitive numbers are viewed as being determined by unexhausted "market potential," which is depicted as the amount by which total 83 Flgure 14 Sales and Number of Competitors Number a-F Cnmpetitors as a function of the Market Potential Httrket P otentT aT U sage Gap a] es Conpetltors Introduction Growth Maturity Decline Time 84 demand exceeds supply, while shake-outs are caused by sales declines which eliminate the weaker firms in the industry. It thus becomes obvious that the population ecology and product life cycle models are conceptually grounded upon a common foundation.50 Problems With the Ecological Analogy Unfortunately, the ecological model, while in apparent agreement with the product life cycle model, retains several disturbing inadequacies. The first of these 50.Because the biological and ecological analogies employ an asymptotic or cumulative function to model population growth, rather than using an equilibrium model as adopted in the biogeographic theory, reductions in competitive numbers (or negative rates of growth) can only occur as a result of resource shortages. This distinction is critical, since the inhibitition of entry (or "birth") will only result in a levelling of the population size in stable environments. The implications for "consolidation" are even more significant, however. If two "islands" were merged into one homogeneous region of size equal to their combined area, as in the hypothetical example discussed above, no change in the population of competitors (organisms) would be expected to occur under the density-dependent selection model of the biological and ecological analogies. This is because the size of the total land area remains unchanged, and thus the density of resources in relation to that of the number of organisms is unaltered. As a result, this perspective would suggest that consolidations have insignificant consequences. 85 inadequacies originates from the retention of the organization/organism analog, and the problematic nature of the role of natural selection within the model. In the life sciences, natural selection occurs at the species, or population level, and individual organisms which display ! inappropriate characteristics are selected out. Yet this implies the relative incapability of individual organizations to adapt to changing environmental conditions, such that survival becomes the sole measurement of "success" (Aldrich and Pfeffer 1976).I This feature denies the role of strategy or managerial control in business concerns, since "strategic choice and environmental determinism represent mutually exclusive, competing explanations of organizational adaptation" (Hrebiniak and Joyce 1985, p. 336). Furthermore, since Gause's "principle of mutual i exclusion" is intended to explain why species are differentiated with regard to resource usage and competitive abilities, it implies nothing about differentiation between organisms in the same species. As noted above, the population ecology perspective of organizational forms as being analogous to species is heralded as a substantial improvement over the industry/species analog of the biological model, as accommodations for variations in firm sizes within 86 industries can thus be made. Yet this enhancement is largely illusory. This is due to the fact that, in a competitive market, most firms will adopt the same organizational form (whichever is most effective). As Hanan and Freeman (1977, p. 94 3) assert: If two populations of organizations sustained by identical environmental resources differ in some organizational characteristic, that population with the characteristic less fit to environmental contingencies will tend to be eliminated. The stable equilibrium will then contain only one population .... Aldrich (1979, p. 112) also argues that "in a purely competitive market, organizations competing for the same resources are pushed toward adopting the same form or perishing." If all competitors in an industry have the same organizational form, however, and if all organizations which share the same form can be assumed to be equivalent in size and resource dependency (Hannan and Freeman 1977), then it would implicitly follow that all industries should be populated by firms of the same size, strategies, and resource usage rates (market share)— an expectation which is easily discredited. Furthermore, the nearly unanimous endorsement and usage of the logistic growth model by population ecologists as a tool for illustrating or estimating population growth within industries tends to reinforce 87 these inferences of competitive homogeneity, since the model requires the assumption of uniformity of organizational size and resource requirements. An illustration of this complication— which is apparently ! unrecognized by the population ecologists— is provided by Aldrich (1979, p. 64): i Environmental capacity, or what ecologists call carrying capacity, sets limits on the size of a population of organizations. Assume that organizations are of uniform size and ! environmental resources are fixed, and assume further that new organizations can be started fairly easily, such as through imitation. . . . In this case, we can use a logistic equation for growth of the population of organizations. These three assumptions— that all competitors are equivalent, that the resource level (and thus the carrying capacity of the environment) is an immutable constant, and that there is no response lag--are implicit in the logistic growth equation (Pianka 1978). What should be obvious, however, is the relative lack of realism of these three assumptions with regard to economic environments. It is also interesting to note the parallels between these required assumptions of the logistic growth model and the (equally unrealistic) assumptions of the economists' hypothetical "perfect competition." In summation, it would appear that the ecological analogy suffers from conceptual and theoretical problems 88 which are severe enough to cast doubt upon its usefulness in analyzing economic and market behavior. As one author (Young 1988, p. 21) noted, the concepts borrowed from ecological theory as they are currently used in the t management discipline, "have to be stretched beyond recognition to fit organizational phenomena." To compensate for these inadequacies, the endorsers of the ecological analogy have resorted to gleaning "theoretical propositions . . . from sociology . . . from administrative science, and from organizational politics— in short, from all sorts of disciplines but not from the biological model" (Young 1988, p. 22). As a result of the many problems of the analogy, Young recommends that the entire endeavor be terminated: Granted, not all the information for making a judgment is in at this time. But if these serious problems remain unsolved after 10 years of work, I think we have the right to say the theory has had a fair showing and has not contributed to the understanding of organizations (Young 1988, p. 23) . The Evolutionary Analogy The fourth and final living systems analogy is that of the evolutionary model. Although this school of thought 89 is still in the formative stages, due to the limited quantity, specificity, and depth of articles in the area, it has made some headway within the marketing discipline, where it apparently originated. Conceptually, the j beginnings of evolutionary thought in marketing can be i seen in the works of Alderson (1957 , 1965; later formalized by Hunt, Muncy and Ray 1981). Alderson (1957) observed that products (evidently he was referring to product classes) could be likened to animal species which evolved over time. More recently, Tellis and Crawford (1981) suggest using the concept of biological "species, genus, family, and class" to portray the "evolution" of a product as an explanation which is "complementary" to the , product life cycle concept at every level (product class, form and brand). The evolutionary model attempts to remedy the problems inherent in the biological analogy, specifically, those of the product life cycle model. In attempting this, the primary contribution of the evolutionary model is the recognition that product classes, unlike organisms, are not generally governed by intrinsic factors which lead to an inevitable "birth-growth-maturity-decline" sequence, and may thus endure for long periods — surviving many temporary upswings and downswings. 90 To accommodate this recognition of indeterminable product growth and potential, the evolutionary model assumes a Darwinian stance, and postulates that the species, rather than the organism, represents a more realistic, and thus more appropriate, analogy for the product class. This is because species, like product classes, have an indeterminate life span, and, also like products, evolve in a cumulative fashion toward greater and greater complexity and diversity (Crawford 1984; Tellis and Crawford 1981). Additionally, the evolutionary analogy focuses upon the role of processes which allow for the change and adaptation of products to their markets over time. Tellis and Crawford (1981), for example, compare the processes of genetic variation and environmental selection to that of entrepreneurial creativity and consumer choice, respectively, to demonstrate how products may become adaptively more "fit" to their markets over time. As in the product life cycle and the ecological analogies, competition which results from excess supply and/or capacity (and thus a scarcity of customer dollars or "resources") is viewed as the catalyst through which the process of selection, and thus evolution, occurs (Gross, 1968). In examining these processes, the proponents of the evolutionary analogy repeatedly emphasize the 91 similarity between the evolution of life forms and the apparent "evolution” of products and technology over time.51 Gross (1968, p. 6) provides an apt illustration of this parallelism: To dramatize the analogy, let us suppose a Martian, unfamiliar with the earth, were shown a series of 60 still photographs starting with the eohippus and a series of its descendants l million years apart— up to a modern horse. Side by side with that series of photographs, the Martian is shown a series of photographs starting with one of an ice box in 1901 and ascending, one each year, through a refrigerator-freezer in 1960— or of an automobile in 1901 through the 1960 model. I believe a rational Martian, upon an examination of the photographs, would conclude that he was observing the same kind of phenomenon in each process— the phenomenon we have called evolution. Unfortunately, and although the evolutionary analogy is effective in dealing with the fact that products can continue to exist and evolve indefinitely (a much needed modification of the PLC concept), none of the authors using the evolutionary analogy develop this concept to where it is useful in a competitive setting. A further problem is the apparent lack of consensus among authors employing the evolutionary analogy. This problem is compounded by a lack of recognition of previous work in 51.Dawkins (1986), a zoologist, has also noted the similarity between the evolution of life forms and the trends of product change over time. 92 the area. Whereas Tellis and Crawford (1981), for example, assert the applicability of the species analog to brands as well as product classes, Alderson (1957, p. 54) incorporates more of the general living systems i perspective by comparing the organization to a "system" I which, "like the human body . . . has the power of repairing and replenishing itself."52 Gross (1968, p. 2), on the other hand, developed a i model where product class is analogous to genus, product form is analogous to species, brand is analogous to a unique race (categories within species which are capable of inter-breeding), and one individual unit of a product is "analogous to an individual unique organism." More recently, Mittelstaedt (1986) developed an approach to analyzing market shares which possesses many conceptual similarities to Gross's (1968) work. Specifically, Mittelstaedt equates species with product, and units sold with individuals. Tellis and Crawford's work (1981) also describes a conceptualization which is highly similar to Gross's efforts. 52.In this sense, the true focus of Alderson's intentions in employing living systems analogies is somewhat ambiguous. As noted above, Alderson adopted concepts compatible with systems, biological, ecological, and evolutionary perspectives. Clearly, however, Alderson consistently conceptualized organizations as analogous to living organisms. 93 As a result of these problems, and while each of these discussions appear to make valid and useful contributions to marketing thought, they have received relatively little attention. Furthermore, these concepts have never precipitated further research, been integrated into an j operationalizable or comprehensive body of theory, and, j like the product life cycle, they have never been empirically tested. Summary of the Living Systems Analogy Overall, the living system analogies can be seen as i constituting a "family" of models. These models share many fundamental assumptions and conceptual components, such that logical conclusions based upon these models are often highly congruous (see figure 15). Significantly, a majority of thought in the management and marketing disciplines can been seen as resting upon this relatively narrow footing of fundamental premises. This extensively! leveraged position may not be problematic in itself, but should this foundation of presumption prove to be unsuitable, a significant amount of business theory and practice may be in jeopardy. 94 Figure 15 Ppoduoi iLife Cycle Ecologica Analogy logical analogy Z Evolutionary Analogy organ!sm Biogeograph Analogy species Cnncsptual DeuelapmEnt o-F Liuing Systems firia 1 ny ies Blogeographic Theory I 95 IV. THE SIMULATION MODEL To explore the internal validity of the biogeographic theory, a computer simulation was constructed using the constructs and relationships incorporated within the theory.53 The methodology of simulation was chosen since the complexities of the interrelationships among the parameters of the model, and the stochastic nature of some functions within the model tend to preclude any analytic solution or analysis. A simulation model assembled from discrete-event, system dynamics, and Monte-Carlo components was thus used to model a system, or "industry" in terms of the logical relationships contained within the theory. As Turban and Meredith (1985, p. 612) note, simulation can be defined as 53.The model employed in the simulation includes the relatively minor modification of linearized immigration J and extinction functions, making the model mathematically simpler. As a result, the immigration function can be seen as represented by a linearized spline function. 96 "a technique for conducting experiments with a digital computer on a model of a management system over an extended period of time." Not only does simulation allow the analysis of complex relationships within a model, but it also allows for "what if" scenario development. For our purposes, simulation offered the additional advantage of the capability of analyzing the results of this complex model over many individual years or periods. In this sense, simulation provided the only viable methodology for our purposes. The simulation program was coded using the Simscript II. 5 programming language from CACI (PC version, release 2.30). Table 2 contains a list of the parameters embodied in the simulation model. Figure 16 illustrates some of these parameters of the model. The computer model was constructed with the intention of maximizing its flexibility with regard to parameter configuration. This was accomplished by allowing the interactive input of the majority of the model's parameters immediately prior to the simulation run. This feature facilitated the overall objective of the simulation, which was to allow a comprehensive exploration of numerous scenarios. The simulation model developed herein represents an attempt to build a model which embodies most of the 97 TABLE 2 PARAMETERS IN THE SIMULATION MODEL Parameters which can be entered interactively; 1) The Slope of the Immigration Function (I) 2) The Y-Intercept of the Immigration Function (M) j 3) The Peak Point in the Immigration Function (P) 4) The Slope of the Extinction Function (E) 5) The X-Intercept of the Extinction Function (X) 6) The Number of Pioneer Firms in the Industry 7) The Period in Which Entry Barriers Begin to be Erected 8) The Length of Time Until Entry Barriers are Complete 9) The Percentage of Extinctions Which are Mergers 10) The Period in Which Consolidation Takes Place 11) The Consolidation Multiplier (The Number of Previously Autonomous Markets Undergoing Consolidation) 12) The Mean Growth Rate for the Market 13) The Standard Deviation of Growth in the Market 14) The Seed Number Used in Generating Pseudorandom' Numbers for the Monte-Carlo (Stochastic) Component 15) The Initial Size of Pioneer Firms (Units) 16) The Number of Periods the Simulation is to Run Parameters Which are Fixed Within the Model (Value): 17) The Number of Firms to Include in Concentration Ratio ; 18) The Mean Size of Immigrating Firms (Growth of Overall! Market in Units for that Period Divided by the Number' of Immigrants) ; 19) The Standard Deviation in Size of Immigrating Firms (Variable 18 Divided by Three) 20) The Seed Number For Pseudorandom Number Generation ini Stochastic Model Allocating Immigrating Firm Size (2) 1 21) Distribution of Stochastic Function Allocating Growth! (Normal) 22) Distribution of Stochastic Function Assigning Immigrant Size (Normal) j 23) Distribution of Stochastic Function Assigning Firm Becoming Extinct to Either Merger or Failure (Uniform) 24) Seed Number for Pseudorandom Number Generation in Stochastic Function Assigning Firm Becoming Extinct to Either Merger or Failure (6) \ 25) Distribution of Stochastic Function Assigning Acquired! Firm to Acquiring Firm (Uniform) j 26) Seed Number for Pseudorandom Number Generation in; Stochastic Function Assigning Acquired Firm to Acquiring Firm (7) 98 Figure 16 Simulation Model Y Rate E n try / Exit 0 Number of Firms 9 features of the biogeographic theory. This model is a simplified, yet powerful representation of an economic "industry," which operates within the functional constraints imposed by the theory. The components of the simulation model are described below. Those parameters which are underlined may be entered interactively prior to the simulation run, those which are underlined and within brackets are fixed within the model (but could be modified if necessary), and those symbols which are underlined and within parentheses are parameters which are illustrated in figure 16. Starting State 1. The Number of Pioneer Firms denotes the number of competing firms which compose the industry prior to the beginning of the simulation. In this sense, "pioneer firms" represent those firms which initially develop and introduce the product in the "introduction" stage of industry development, prior to any assurance of commercial success of the product category. 2. The Size of Pioneer Firms defines the initial size in number of units sold in period zero for the pioneer firms 100 described above. (Period zero represents the initial state of the simulation model.) This size parameter merely indicates an arbitrary beginning size of pioneer firms and implies nothing about later firm size. In addition, all pioneer firms are assigned the same initial size. i i 3. The Number of Periods the Simulation is to Run represents the duration of the simulation. These "periods" may conceptually represent any linear unit of time, such as a year or a quarter. Each period represents i one iteration through which the simulation will run, such that the Number o f Periods defines the number of iterations performed in the simulation. In the following description, the current period of the simulation will be denoted as period t. Immigration and Extinction Functions 4. The rate of Immigration into the industry denotes the rate of entry of new firms per period. In general, this rate of immigration will be a negatively sloped, linear function of the number of existing firms in the industry, 101 defined by the Slope of the Immigration Function (I). and the rate of Immigration which would be realized if no firms existed in the industry, or the Y-Interceot of the Immigration Function XXX* In general then, where N = the Number of Firms in the Industry: i I Immigration = Y + (N * I). 5. Should the number of firms existing in the industry be ! fewer than some critical number, however, the rate of Immigration will be a positive function of the number of existing firms. This critical number is represented as the Peak in the Immigration Function i_Pl* If the number of firms in the industry is less than this number, the rate of Immigration into the industry will be determined as: Immigration = ((Y+ (I * P)) /P) * N. Thus, where the number of firms in the industry does not exceed the Peak in the Immigration Function (P) . the rate of Immigration will be a function which passes through the origin and with slope (In<p) equal to (Y + (I * P))/P. 102 5. The rate of Extinction from the industry denotes the rate of exit of firms per period. This rate of Extinction is a positive function of the number of existing firms in J the industry, determined by the Slope of the Extinction Function (E) . and the X-Intercept of the Extinction Function fX) . or the number of existing firms at which this rate of exit is zero. The X-intercept of this function is utilized rather than the conventional Y- intercept so as to allow the specification of the static- \ , interactive equilibrium point (described in the theory above) prior to running the simulation. Thus: i Extinction = E (N - X). I 6- The Period in Which Entry Barriers Begin to be Erected. which we will abbreviate as , defines the period during which barriers to immigration are initiated. These barriers result in a gradual shift downward in the immigration function, and are completed after the Length of Time Until Entry Barriers are Complete. which we will denote as c, such that absolute entry barriers exist after period t The entry barriers are erected according to a linear function, shifting the immigration function downward by an equal amount each year until rate of entry 103 is zero at every point. Thus, the slope of the immigration function for the periods between t^bj and ■t(b+c) is given as: It = I - ((t - b) / c) I, and after period t(b+C) as: Similarly, the intercept of the immigration function is: Yt = Y - ((t - b) / c) Y, between periods and t(b+cj , and Y = 0 after t(b+c). 7. The number of firms in the industry during any given period, then, is jointly determined by the immigration and extinction functions, and will be at an equilibrium point where these two functions are simultaneously equal. 104 Firm Sizes and Growth 8. Individual firm growth is a function of a stochastic process described as being normally distributed with a mean defined as the Mean Growth Rate for the Market. or mu, and a standard deviation defined as the Standard Deviation of Growth in the Market. or sigma. (When entering the values for these parameters prior to each I simulation run, the actual values entered are lOOmu and j lOOsigma, such that a mean growth rate of .1 or 10% would i be entered as simply 10.) (Since Simscnpt II.5 generates pseudorandom numbers rather than true random numbers, seeds are used to initiate the number generation sequence. Ten different Seed Numbers are available for use in Simscript II.5, and any of these may be selected in the model.) Size for any particular firm, then, will be determined as: Size^ = Size^..-^ + f(Size t-l) where f is a normally distributed function with mean = mu and standard deviation = sigma. 9. The initial size of entering or immigrating firms is determined by a stochastic process which is normally 105 distributed with a mean equal to fmu*M1, where mu is the mean growth rate for the industry (as above) and M is the size of the overall market (the sizes of all firms in the market combined), and a standard deviation equal to the absolute value of fmu/31. This function was chosen with I ] the goal of allowing entering firms to assume a size which is on average approximately equal to the overall growth j experienced by the industry in that period, divided by the I j number of entering firms. The assumption underlying this decision is that the total capacity of entering, or new firms is unlikely to exceed greatly the overall incremental growth experienced by the industry as a whole for that period, as existing firms would thus be required to shrink in size to accommodate new entrants. The standard deviation value of rmu/31 was chosen in an effort to maximize the variance while maintaining a relatively "normal" or symmetrical distribution. (Since entering firms would necessarily be required to have a size greater than zero, distributed sizes of potentially negative size values were converted to absolute value.) 10. Firms which are removed from the industry according to the extinction function are those which are of the smallest size. That is, the smallest firms in the industry are removed until the rate of extinction is 106 reached for that period. Thus, if five firms are scheduled for extinction, the smallest five firms will be removed from the market.54 11. Firms which become extinct may either disappear altogether from the market or merge with another firm. i The function which determines whether any given firm is merged or eliminated is stochastic, and is based upon a uniform distribution within the open interval (0,1). I i (Pseudo)random numbers generated with this process are compared to a number defined as the Percentage of Extinctions Which are Mergers. such that firms becoming extinct are randomly assigned to merger or elimination, with the probability of merger equal to this percentage. 54.Note that the decision to remove those firms which are the smallest was based upon (1) the relatively compelling evidence that small firms are generally at a much greater risk of failure or acquisition than their larger rivals in the same industry (see Aaronovitch and Sawyer 1975; Betton J and Dess 1984; Collins and Preston 1951; Hart and Prais 1956; Hawley 1981; Mansfield 1962; Penrose 1959; Townsend 1968; Wholey and Brittain 1986), (2) a lack of any supportable alternative criterion, and (3) a desire to maintain a conservative approach— since selecting out larger competitors would generally tend to increase the skewness of firm size distribution within the industry and thus lead to a more robust (but potentially less legitimate) model. 107 12. If a firm which is scheduled for extinction is designated as a merger, the acquiring firm is determined by a stochastic process with a uniform distribution. Furthermore, only those firms with a size which is greater j than the mean firm size within the industry are eligible i | to acquire merging firms. Thus, the size of the merging | firm is randomly apportioned to one particular acquiring firm which is larger than average. The acquiring firm's size and market share are thus enlarged by quantities equal to the size and share possessed by the acquired firm. 13. If a firm which is scheduled for extinction is designated as a candidate for elimination, rather than merger, its size is distributed proportionally to all remaining firms in the market. Thus: Size of Firm-j^-t) = Size of Firm i(t-i) + ((Size of Eliminated Firm / Size of Entire Market) * Size of Firm^t-i)) 14. The capability for analyzing the effects of a consolidation is incorporated into the simulation model. The consolidation will occur during The Period in Which Consolidation Takes Place, and may be of any magnitude. 1 0 8 The magnitude of the consolidation is determined by the Consolidation Multiplier, which represents the number of previously autonomous markets which will be consolidated. i I Each market in the consolidation is a virtual duplication i of the original market with regard to firm sizes and i : distribution, and thus overall market size. Thus a \ !consolidation of four previously autonomous markets would I be simulated by selecting a Consolidation Multiplier of 4, and the market would then consist of four times the number of original firms, with each original firm now having three duplicates. Since this consolidation serves to artificially raise the number of firms above the equilibrium point, the rate of extinction will exceed the rate of immigration, and a net reduction of firms will i proceed until equilibrium is again reached. 15. The concentration ratio of the industry is defined as the combined size of the R largest firms in the industry (where R is The Number of Firms to Include in the Concentration Ratio) divided by the size of the overall total market. Thus, where R is chosen as 3, the concentration ratio is the combined market share of the three largest firms in the market. 109 V. RESULTS FROM SIMULATION EXPERIMENT AND EMPIRICAL ANALYSIS After validating the simulation model through tests utilizing the full range of possible parametric values, the model was employed to determine experimentally the outcomes of numerous competitive and economic scenarios. These scenarios were implemented through various configurations of the model’s parameters and the results were then observed. Over 200 different configurations were tested. In general, the model performed quite well, yielding results which tended to be quite congruent with previous empirical research and case studies. The simulation trials demonstrated that, under the assumptions of the biogeographic theory, highly realistic representations of J industrial market structure could be achieved. An example of the results obtained from one simulation run is presented in figures 17 through 20. For this 110 Ill Figure 17 Simulation Firms and Sales Number Sales (Thousands) 160 500 140 - 400 120 - 100 - - 300 80 - - 200 60 40 - 100 20 - 60 40 0 50 70 30 20 Period — Total No. of Firms Entry of Firms . _Exi.t_of_.EIrms______________ : _________ zzzr_.Sales .in _ U .n i ts_ 112 Figure 18... Simulation Concentration Effects Number C/R 160 140 120 100 40 30 0 10 20 40 50 60 70 - 0.8 0.6 - 0.4 0.2 Period Total No. of Firms Exit of Firms Entry of Firms Concentration Ratio 113 r ~ Figure 19 Simulation Sizes oi Firms at Period 70 Thousands 160 140 120 100 Firm 1 Firm 2 Firm 3 Firm 4 Firm 5 Firm 6 Firm 7 Firm 8 Firm 9 Sales Per Firm 1 1 4 Figure 20 Simulation Sizes of Firms at Period 70 F irm I 138.3 / 34.7% F irm 2 88.5 / 22.2% F irm 9 9.4 / 2.3% F irm 8 12.6 / 3.2% F irm 7 13.1 / 3.3% F irm 6 13.4 / 3.4% F irm 3 56.2 / 14.1% F ir m 5 19.8 / 5.0% F irm 4 47.4 / 11.9% Sales In Thousands / Market Share particular simulation run, the (interactively chosen) parameters of the model were assigned the following values: 1. The number of Pioneer Firms was chosen as three, as this number is roughly consistent with actual anecdotal evidence in numerous industries. Many industries, especially those beginning in recent history, can be seen as being pioneered by two, three, or four firms simultaneously, each introducing their version of the product at approximately the same time. This number is not critical, however, in that numbers between one and ten will all yield results which are similar in the long run. Number of Pioneer Firms Slope of Immigration Line Slope of Extinction Line Y-Intercept of Imm. Line X-Intercept of Ext. Line Peak in Imm. Line (#firms) Period Entry Barriers Begin Time Until Barriers Complete % of Extinction as Mergers Consolidation Period Consolidation Multiplier Industry Mean Growth Rate Std. Deviation of Growth Seed Number for Growth Size of Pioneer Firms 100 15 10 40 71 5 6% 16% 2 3 -.15 . 10 40.0 5.0 60. 0 Rationale For Chosen Parameter Values 1 1 5 In general, the fewer the number of pioneer firms, the higher the early levels of industry concentration. As time progresses, however, this number becomes less important. 2. The slopes of the Immigration and Extinction functions were experimentally chosen to yield entry, exit and t :dynamic interactive equilibrium figures which would be jrealistic in "average" differentiated industries. In i industries where there are few viable niche dimensions, or where minimum optimal scale requirements are great, these slopes could be expected to be more extreme. i 3. The Y—intercept of the Immigration function was also chosen in an effort to yield a realistic outcome in immigration rates. In industries characterized as "perfect competition," and with a high number of firms at dynamic equilibrium, the Y-intercept of the immigration functions would be likely higher. 4. The X-intercept of the Extinction function was chosen with the intention of yielding a static-interactive equilibrium level which would be approximately equivalent to a situation between an oligopoly and monopolistic competition. Although the X-intercept is specifiable as 1 16 indicated, the simulation model is designed to remove only "whole" numbers of firms from the industry. This will often result in a slightly higher effective X-intercept for the extinction function. In this case, and although j the X-intercept is input as 5, the effective intercept is i (actually 9.0. This number would seem to be reasonably consistent with many actual industries, where the equilibrium number of firms after barriers to entry are erected is greater than a true oligopoly, yet fewer than perfect competition. 5. The peak in the Immigration function was chosen somewhat arbitrarily after experimentation. For immigration functions which peak earlier (at few existing firms) , the initial rate of immigration is swift and reaches an early and sharp peak. Immigration functions which peak at a greater number of firms result in a more gradual rate of entry and peak. 6. The period in which entry barriers began and the amount of time until barriers were completed were chosen experimentally and with the goal of describing a reasonably realistic scenario. Barriers which are erected very late in number of periods or are not erected at all result in the plateauing of firm numbers for that 1 1 7 industry. This plateau represents the dynamic equilibrium i level of competitors. Overall, however, 15 years would seem to be a reasonable period of time for barriers to begin to be erected in modern industries, assuming they ] will be erected at all. Similarly, 10 years would appear I i jas a reasonable length of time for the completion of these i ibarriers. J I 7. The percentage of extinctions which occur as mergers was chosen arbitrarily. Although 40 percent would seem to I jbe a high proportion of mergers, it is not unrealistic for many industries. Furthermore, the proportion of extinctions which result in merger does not strongly influence the outcome of the simulation, especially when the long-run is considered. 8. The period in which consolidation takes place was chosen as 71 for two reasons. First, this number was sufficiently large to allow the analysis of the simulation results prior to consolidation. Thus, the effects of the above parameter values may be examined prior to any confounding effect which may result from consolidation. Second, this number is a direct derivation from economic history, in that many industries which originated some time around the turn of the century began to experience 1 18 the effects of global consolidation in the 1970's. 9. The consolidation multiplier was chosen as five to reflect the effects of modern global industrial consolidation. Most industries today are being impacted by imports from between 3 and 7 (or more) foreign countries. These countries commonly include Japan, Korea, West Germany, Taiwan, Great Britain, France, Italy, and/or other nations. 10. The 6% mean rate of growth and 16% standard deviation of growth were selected following Scherer (1980). Scherer conducted a simulation using these parameters to illustrate the effects of proportionate growth. In this simulation, Scherer (1980, p. 145) chose the values of 6% growth and 16% standard deviation "to reflect the average year-to-year growth actually experienced between 1954 and 1960 by 369 companies on Fortune's list of the 500 largest industrial corporations for 1955." Other means and standard deviations of growth were experimentally tested in this simulation model, with similar results. 11. The seed number for the stochastic process allocating growth was chosen randomly. Other seed numbers, of course, yielded varying, but similar, results. (Note that 1 1 9 this integer value is only a surrogate of the actual seed number.) 12. The size of pioneer firms was chosen arbitrarily as I 100, since this size does not affect the relative outcome i ! of the simulation, and since this number is analytically I |simple. I I Simulation Results The simulation run under the above configuration of parameter values yielded results that were fairly typical across numerous experimental trials using this model. I Specifically, it should be noted that the number of firms i rose to an early peak, achieving a maximum at periods 15- 16 (at the time entry barriers begin), with immigration rates peaking in period 9 (see Figure 17) . Industry 4- firm concentration ratios realized a minimum at period 30, climbing slowly thereafter (see Figure 18) . The 4-firm concentration ratio was 0.829 in period 70, with an j industry reduced to 9 surviving firms with a market share distribution illustrated in figures 19 and 20. 12 0 As a "consolidation" was scheduled for period 71, the number of firms quintupled (as intended) in this period and steadily declined thereafter, settling at 9 firms (as above) at period 91. This consolidation concomitantly reduced the 4-firm concentration ratio to below 0.2 in ! period 71, climbing to above 0.8 by period 100. See figures 21 and 22. Empirical Analysis The results obtained in the simulation above are likely to appear untenable to most of those steeped in the traditional logic of the product life cycle concept or the population ecology model. Indeed, the relationships among the variables plotted may even appear counterintuitive. Nevertheless, a comparison of the simulation results with empirical data may prove advantageous. The plots illustrated in figures 23 through 36 represent historical data collected from twelve industries, and over periods ranging from 65 to 178 years. The sources for the data are many and diverse, and are listed in appendix 2. Although the selection of industries for study was dictated by data availability, the industries illustrated here represent a wide range of 121 122 Number Figure 21 Simulation Alter Regional Consolidation (Period 71) Sales (Thousands) 1 60 140 - 120 100 - 80 ~ 60 - 40 - 20 - I I I 111 1 1 1 1 1 1 11III I ! TTTTTT 111 n r i i r r n 11111 r r r n m n i r n r 60 70 80 40 50 90 100 30 Period Total No. of Firms Sales In Units 3000' 2500 2000 1500 1000 500 0 123 Number Figure 22 Simulation After Regional Consolidation (Period 71) C/R 160 140 120 100 80 60 40 20 I i i I I TT TT! i 1 1 30 11 i i r u n TTTTTTTTTfTTT TTT r m 11111 0 10 20 60 60 40 70 80 90 100 1.2 0.8 0.6 - 0.4 0.2 Period „Total..N o. o i_F 11 m s m Concent rat ion types. Durable and non-durable consumer goods, industrial goods, and industrial and consumer services are included. The information communicated by the individual plots— and the similarities across all of them, tend to jdisconfirm the hypothesis (H0) that the number of ! competitors within industries is directly and positively related to the sales volumes experienced within these industries. This empirical study thus brings into question the underlying assumptions and propositions of the product life cycle and the population ecology models. Concomitantly, support is rendered to the biogeographic :explanations of market structure and behavior. In every industry presented, the number of competitors rises to a maximum long before any general reduction in the overall growth rate for that industry becomes j C C * evident. 3 Even in industries with evidence of a leveling or downturn in sales growth rates, the peak in numbers of 55.It is, of course, difficult if not impossible to isolate the effects of entry barriers and stochastic growth processes from those of consolidation in examining ] competitive "shake-out." For this reason this empirical analysis can only help determine whether and when shake out occurred in each industry, rather than any single cause of these competitive reductions. Since no existing theory up to this point would be able to explain any form of shake-out in growth markets, evidence of these reductions, as found, serve to discredit existing explanations and support the biogeographic theory— regardless of the actual causes. 1 2 4 competitors has occurred many years previous to this downturn. Moreover, the distance between these maxima is substantial, generally engulfing 40 or more years. In general, these plots of empirical data conform reasonably well to the experimental results obtained in the simulation trials. Although these plots are not presumed to be representative of every industry, and although they cannot be employed to "prove" the correctness of the biogeographic theory,56 they would appear to suggest that a rather comprehensive reappraisal of existing theories is warranted, and that the biogeographic theory of competitive market structure may provide a defensible alternative. 56.There is no denying that the conclusions reached in this paper contain a degree of speculation. This cannot be helped, however, given the historical nature of the evidence presented. As Savitt (198 0, p. 57) noted with regard to historical analysis; The end product of the verification process, regardless of the outcome, will be a set of conclusions which recognize, but do not necessarily strongly endorse, alternative explanations. The precision of statistical relations which clearly link events will not be present in most cases. 1 2 5 126 Figure 23 Cigarettes # ol Firms Production 400 800 300 - - 600 200 - 400 100 200 i l i m i m i l i i r m m l u m m i l i T T T T TT T TTTTT 1870 1880 1890 1900 1920 1930 1940 1950 1960 1970 1980 1990 1910 Year Bil. of Cigarettes Number ot Companies 1 2 7 Figure 24 Automobiles # oi Firms Thousands oi Cars 10000 250 8000 200 - - 6000 150 - 4000 1 00 - - 2000 50 111 I I I II 111 11111TIT I r r n n t t t t t t t t TTT 1985 1955 1965 1945 1925 1935 1915 1905 1895 L . Domestic Production Year Number oi Producers Imports (thousands) 128 Figure 25 Tires # of Firms Shipments 300 200 - 250 1 50 - - 200 - 150 1 00 - - 100 50 - - 50 TTTtTTTTTTTTT 1955 1965 1895 1925 1935 1945 1975 1985 1905 1915 Year M illio ns.pl Tires___: ~ ~ NumbeLpf.JMrm.3 129 Figure 26 Home Laundry Equipment Washers & Driers i i # of Firms Factory Sales 12000 70 60 - 10000 50 - - 8000 40 - - 6000 30 ~ 4000 20 - 2000 10 - 1990 1970 1980 1960 1950 1940 1930 1920 Year Thou sand s_of Units N umber ot Firms 130 Figure 27 Beer & Malt Liquor Number Production 5000 200 4000 - 150 3000 - 2000 - - 50 1 000 - T i T i i f i i i i m i i i m i i i i i i i i i i i h i H i i T i m i i i i i i i ! 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 Year Millions..o.l Bbls._ it .: Breweries 131 5000 4000 3000 2000 1000 Figure 28 Beer & Malt Liquor (Pre-Prohibition) # ol Breweries Production 70 60 50 40 30 20 10 T T i T i T r i T t r r ] i m n 1 1 1 1 1 1 1 1 1 11 1 1 m 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 i f f i T i T T 1810 1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 Year _________ __ _ _____ ___Millions oi Bbls. — Number oi Breweries _________________ 132 Figure 29 Beer & Malt Liquor (Post-Prohibition) ! # oi Plants/Firms Millions ol Bbls. 200 800 - 150 600 - 100 400 - - 50 200 - 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 Year — Production (U.S.) Number of Breweries + Number ol Companies______ Imports Figure 30 Railroads # of Lines Freight Traffic 1000 1500 - - 800 - 600 1000 - - 400 500 ~ 200 TTTTTTTTT 1960 1970 1890 1900 1910 1920 1930 1940 1950 1980 1990 1880 Year H W W Bil. ol Ton-Miles — Number ol Railroads 134 Figure 31 Chewing Gum # oi Firms Production 500 80 - 400 60 - - 300 40 - - 200 20 - - 100 1990 1960 1970 1980 1950 1940 1930 1910 1920 Year M illions o l Pounds : Number oi.F.irms 135 Figure 32 Commercial Banks # ol Banks Deposits 2000 35000 30000 - - 1500 25000 - 1000 20000 - 500 15000 - Tirrm^ 1930 1940 10000 i ii r r n T m r n iT T T n rrrnTTTTr t T T rrrn t i 11111111 1920 1950 1960 1970 1980 1990 1900 1910 1890 Year Deposlts (bil. $) 'Number oi banks 136 Figure 33 Gasoline # oi Plants/Firms Production 700 3000 600 - - 2500 500 - - 2000 400 - - 1500 300 - 1000 200 500 1 00 - TTnrrifiliiiiiiiiiliiiiiiiiiliiiiiiii'il'rrmTiii t t t t t h i 1111111111 r h 111! 111111111II111 TTTTTTTT 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 Year Barrels (millions) — Refineries + Companies 137 Figure 34 Pig Iron # ol Plants Production 500 120000 1 00000 400 - 80000 300 - - 60000 200 - - 40000 1 00 20000 1810 1910 1830 1850 1890 1930 1950 1970 1870 1990 Year Thousands ol Tons Number of Plants 138 Figure 35 Rayon & Acetate t # ol Firms Production 24 22 20 - 1500 18 - 16 - - 1000 12 - 1 0 500 1910 1920 1930 1940 1950 1960 1970 1980 1990 Year Millions of Pounds — Number ol Firms 139 Figure 36 Breakfast Cereal # of Firms Sales 3000 150 ~ 2500 100 - 2000 50 - 1 500 ■| r n t 11' 1970 1 000 1910 1920 1950 1960 1980 1930 1940 1990 Year Millions of Pounds Number of Firms ( I I ! VI. CONCLUSIONS AND IMPLICATIONS I I I I This dissertation has presented an alternative theory of industrial market structure. Although this theory is grounded upon a conceptual foundation which is shared by I the product life cycle and population ecology models, a focus upon the species rather that the organism leads the biogeographic theory toward logical conclusions which differ greatly from these previous perspectives. Furthermore, these logical conclusions— while apparently counterintuitive— appear to provide a closer fit to both past empirical research, and the empirical evidence presented in this paper. Specifically, the biogeographic theory suggests that the number of competitors within each industry is a function of unexhausted "niche space," rather than a function of the sales levels within these industries. This is significant because the amount of available niche space is not directly related to industry sales levels (or 140 even potential sales levels) , and may tend to become depleted before sales growth rates abate. The implication is that the number of competitors will frequently or even typically climax long before any indications of market "maturity" or slowing rates of I I igrowth are apparent. Concomitantly, the expectation that t competitive intensity may achieve formidable levels even in substantially "abundant" markets experiencing precipitous rates of growth is also implied in this interpretation. The biogeographic theory offers additional advantages which are unique, important, and not available from traditional models. These are: 1. the accommodation and expectation of both size and competitive differences among competitors; 2. the ability to utilize the concepts of "niche" and competitive differentiation in a productive and useful fashion; 3. an equilibrium model of competitive populations which differs significantly from the traditional asymptotic growth models, and which offers a more dynamic and complex interpretation of market entry and exit; 141 4. a potential explanation for the clustering of large (firms in primary (central) regions of market dimensions i and the "niching” of smaller firms in more remote regions; 5. an explanation for the commonly observed skewed firm size distributions which is richer and more general than traditional stochastic models; 6. a demonstration of a direct relationship between barriers to entry and industry "shake-out"; 7. an explanation and new interpretation of the impact of regional and global market consolidation which has significant implications for marketing strategy and national policy. i This paper also illustrates the economic consequences which are implicit in the biogeographic theory though the construction of a simulation experiment. This simulation was modeled upon the constructs embodied within the } theory, and the results of these experiments suggested that the theory was conceptually valid. Finally, this paper presentes a substantial volume of empirical evidence of a historical nature which clearly L . 142 demonstrates the inadequacy of existing marketing and management models of competitive dynamics, and which 'offered support for the biogeographic theory of market J structure. Imolications for Marketing Strategy As noted in previous chapters, the differences between the explanations of past economic history provided by the previous living systems analogies and the biogeographic theory are formidable and important. The implications of the biogeographic theory for the future of American industries are worthy of consideration as well, however. Just as regional markets consolidated into national markets for most industries during the last 100 years, the process continues at another level. Many industries are now facing competition in the global marketing arena, as national markets consolidate.57 The competitive 57.Note that this consolidation of markets on a global scale does not require the assumption of what is currently termed a "global market." As Sheth (198 6, p. 9) observes, the emergence of global competition is not synonymous with "an emerging universality of consumer needs and wants." That is, global competition is not necessarily emerging because of a reduction in the number of viable, non- spatial niche dimensions due to a homogenization of consumer tastes, but rather is being caused by an influx of new firms which have migrated across national barriers 143 implications of these consolidations are not trivial, and would seem to be directly parallel to those provided by the earlier example of regional consolidation.58 The implications suggested by this new interpretation are even more portentous when one considers the fact that economic consolidations (both regional and national) typically involve not merely two formerly autonomous areas, but many. That is, the reduction in the number of I competitors may be significantly greater that the 50% postulated in our two-area example. A consolidation of 2 0 regions in eguilibrium, for example, would theoretically result in the "shake-out" of between 88% and 95% of formerly viable competitors (depending upon the and which must attempt to fill niches already occupied by existing firms. 58.In this way, we would hypothesize that advances in transportation technologies have, in the last century, served to eliminate the spatial monopolies and oligopolies which existed in many industries, and replace these with competitive situations which approximated monopolistic competition. Then, the effects of niche overlap precipitated the process of competitive exclusion, resulting in a shake-out of competitors in these industries. Gradually, this process would appear to be beginning anew on a global level, an event we are now witnessing. The important implication is that this cycle has (at least) two distinct components, and if we focus I simply on the consolidation phase, it may appear that the jnumber of competitors for any given region (country) is actually increasing. This, however, is merely an artifact of aggregation. 144 homogeneity of the consolidated area). Furthermore, the implications of these distinct theories would appear to be important not only to corporate strategy, but also national policy.59 Implications for Future Research I 1 Obviously, the purpose of this paper is to initiate a dialectic or debate concerning theories of competitive dynamics, rather than to offer a comprehensive or conclusive bible on this subject. In this sense, the goal of this research is to open the door to another perspective of marketing thought: a perspective which hopefully will provide a richer and more fruitful domain than existing frameworks. Because this paper represents only the first step toward establishing this perspective, 59.The biogeographic theory would seem to suggest, for example, that global consolidation has raised the number of automobile manufacturers doing business in the United States significantly above the natural "equilibrium point." If this is true, the competitive exclusion process (if unimpeded by national policies) may precipitate a global shake-out of automobile producers in the future. Given this, it would appear prudent to consider the importance of the nationality of these producers. Increasingly, it becomes apparent that there is cause for concern in this matter, as Japan's annual motor vehicle production now significantly exceeds that of the United States (Conot 1987), and Korea has become established as a major player. 145 the potential range and avenues for future research in this area would seem nearly infinite. The following areas may provide a starting point: ]1. Further empirical work which could test both i {traditional explanations and the biogeographic theory. I |This empirical research might include historical entry and exit data within individual industries, an attempt to demonstrate the effects of barriers to entry, or the measurement of competitive intensity using indicators other than numbers of competitors (such as price); 2. The development of models for new product forecasting using the concepts of the biogeographic theory, and which could be compared to traditional models; 3. Normative framework development based upon the positive components of the biogeographic theory. Conclusion In conclusion, this paper presents an alternative theory and model of market structure and competitive dynamics which appears to possess the potential to shed 146 greater light upon many problems faced by the marketing community. Whether this theory will withstand subsequent empirical investigation is unknown, yet existing theories have proven themselves incapable of enduring these jefforts intact. In any case, the biogeographic theory would seem to exhibit a robustness not evidenced in the product life cycle concept. This robustness would appear indicative of an inherent potential for establishing a j foundation for the construction of a general theory of marketing. 147 APPENDIX 1 EXAMPLES OF REFERENCES TO THE PRODUCT LIFE CYCLE AND COMPETITIVE CLIMATE 1. This is a period [growth] of high and sharply- rising profits for manufacturer, distributor, and retailer. Risks can be accepted that would be disastrous in a more competitive era, for soaring demand covers a multitude of sins involving hasty or ill-considered actions (Patton 1959, p. 8). 2. As the life cycle enters the maturity stage, the market is established and the oligopoly . . . of the growth stage changes into the free-for-all of competition. Moving into the maturity phase can be traumatic, because the peaceful coexistence of competitors now turns into a fight for market shares (Product Marketing 1977, p. 46, 51). 3. The greatest number of competitors, competitive product forms, and brands exists in the maturity stage .... Thus price competition develops along with heavy promotion of whatever unique brand features still exist (Schoell & Guiltinan 1988, p. 290). 148 4. [In the maturity stage], [t]he slowdown in the rate of sales growth creates overcapacity in the industry. This overcapacity leads to intensified competition. Competitors scramble to find and enter niches. . . . A shakeout period begins and the weaker competitors start dropping out (Kotler 1988, p. 359-60). 5. The maturity stage is reached when total customer demand begins to level off. Competition becomes extensive and affects all aspects of the marketing mix. [In the decline stage] competitors that are not efficient begin to leave the market (Park & Zaltman 1987, p. 242). 6. Designs must be frozen in this phase [growth] and meeting the increasing demand involves risk- taking which would not be realistic in a later, more competitive, lower profit phase (Cunningham 1969, p. 33). 7. In the growth stage, competition is increasing but still limited . . . In the mature stage, competition intensifies . . . Prices are generally lowered at this stage as competitors struggle for market share (Reibstein 1985, p. 3 63). 8. The consequent filling of distribution pipelines generally causes the entire industry's factory sales to rise more rapidly than store sales. This creates an exaggerated impression of profit opportunity which, in turn, attracts more competitors. . . All this in time inescapably moves the industry to the threshold of a new stage of competition. This new stage is the market maturity stage. The first sign of its advent is evidence of market saturation. . . . [In the decline stage] few companies are able to weather the competitive storm (Levitt 19 65, p. 83). 149 9. Profits decline in this stage [maturity] because of the increasing number of competitive products . . . [In the decline stage] the overcapacity caused by too many competitors converging on the original growth opportunity becomes more acute. This frequently leads to price competition and an industry shake-out with many marginal competitors going out of the industry (Doyle 1976, p. 2) . 10. All sales growth sooner or later slows down. . . . The supply of potential buyers runs out. The slowdown which signals the approaching market maturity normally uncovers some degree of overcapacity and initiates a competitive battle for permanent market position (Wasson 1974, p. 8). 11. As the maturity, period is entered, demand growth tapers. The reservoir of potential buyers, who have not been using the product, is drained low. . . . Thus market saturation is reached eventually (if not soon), and our product tops out as demand starts to diminish. [In the decline stage,] competition intensifies as sales volume tends to decline (Luck 1972, p. 11, 74). 12. As sales continue to expand, the product moves into the maturity stage. Competition is typically keen as the market reaches a saturation or leveling-off point. Severe price competition often prevails, thus reducing profits (Cravens, Hills, & Woodruff 1980). 13. As severe competition ensues during the maturity stage, the marginal firms will be weeded out. More competitors drop out during the decline, as other opportunities look more attractive (Hartley 1976, p. 2 48). 150 14. In the growth phase, demand pull provides margins that may allow relatively inefficient producers to survive. . . . Shakeout is signaled by falling margins and a reduction in the rate of demand growth (Rumelt 1979, p. 205). 15. In the maturity stage growth slows, the market becomes saturated, and the product approaches its market potential. . . In competitive turbulence a shakeout frequently occurs because of excess industry capacity. During saturation . . .price reductions become common as firms battle for market share. With sales and profits falling [in the decline stage], most firms begin an exodus from the market, usually because new products create technological or fashion obsolescence (Lazer & Shaw 1986, p. 15.8-15.9). 16. The growth stage of your product will attract imitation, and competitors will begin to enter the market with similar products. In the maturity stage of the product life cycle, sales continue to rise but eventually level off as the market becomes saturated. More and more competitors may enter the market at this stage (Fox & Wheatley 1978, p. 182). 17. During this stage [growth] the number of firms in the market increases . . .During the third, or maturity stage, . . . competition becomes quite significant, forcing marginal producers and dealers out of the market (Hasty & Will 1975, p.169-70). 18. [In the maturity stage] competitive entry into the slowing market may produce overcapacity in the industry. . . As marketing costs rise and prices fall, profits on the product erode. Weaker competitors drop out while the remaining firms compete for market share (Hise, Gillett, & Ryans 1979, p. 250). 151 19. In this last stage of the life cycle [decline], the product reaches a saturation point in the market and sales begin to decline. With declining sales, however, there is a sharp reduction in the number of competitors in the market (Gwinner et al. 1977, p. 121). 20. As a product matures, its rate of growth slows down— levels off— and profits begin to decline. Competition becomes severe. . . . This is a period in which the shake out of marginal producers gets underway with grim determination. Only the strongest firms will survive (Buskirk 1970, p. 187-8). 21. Success breeds imitation, and firms rush into the market with competitive products in search of profit during the growth stage. Industry sales continue to grow during the early portion of the maturity stage but eventually reach a plateau as the backlog of potential customers is exhausted. By this time a large number of competitors have entered the market and profits decline as competition intensifies. . . . [In the decline stage] manufacturers gradually begin to leave the industry in search of more profitable products (Boone & Kurtz 1977, p. 178-9). 152 APPENDIX 2 SOURCES AND GENERAL COMMENTS REGARDING DATA i I Automobiles SOURCES: ■ Automobiles of America. 1970,1974, and World Motor Vehicle Data. 1986— Motor Vehicle Manufacturersj Association of the United States, Inc., Detroit, Michigan I i ■ Ward 1s Automotive Yearbook. 1975, 1986— Ward's] Communications, Inc., Detroit, Michigan ■ Automotive News. 1986 Market Data Book Issue— Crain Communications, Inc., Detroit, Michigan ■ Business Statistics.* 1959, 1967, 1984 ■ Basic Statistics. 1987— Standard & Poors Corporation ■ Thomas (1977) ■ White (1971) | COMMENTS: ■ Factory Sales in Thousands of Units: units manufactured by domestic producers only (includes domestic production for export) 153 --According to the United States Bureau of the Census (Historical Statistics of the United States: Colonial Times to 1970. p. 705) : Production of Passenger cars was discontinued in February 1942 to economize resources for World War II purposes, but some vehicles remaining in factory stocks were sold under rationing orders in subsequent war years. The War Production Board authorized resumption of production as of July 1, 1945, but no new cars were actually produced until 1946. ; — Although Standard & Poors cites its data source as the Motor Vehicle Manufacturers Association (cited above), there are occasional and slight discrepancies in the data presented from these two sources. ■ Number of Firms Producing Automobiles: includes only those firms which are engaged in the production of complete passenger automobiles — Number of Firms for years 1978 and later does not j include Volkswagen of America, Honda, or Nissan (which began final assembly of limited quantities of automobiles! in the U.S. in 1978, 1982, and 1985 respectively. Note j that these firms often assemble cars from completed sub- assemblies which have been imported, and that these j efforts are targeted mainly toward escaping present or J future import restrictions. j — Note that only one successful entry into automobile j manufacturing after 1946 occurred, and this was the | reentry of Willys, which resumed production of passenger | cars in 1954 (White 1971). i ! ■ Imports into the United States of Passenger j Automobiles: number of units imported into the the United States per year 154 Commercial Banks ! SOURCES: ! j i ■ Historical Statistics of the United States; Colonial| : Times to 1957* ' ! 1 I . ■ Historical Statistics of the United States: Coloniali Times to 1970* 1 l • ■ Annual Report— United States Federal Deposit Insurance ! 1 Corporation, various years, Washington D.C. I COMMENTS: I ■ Total Deposits in Billions of Dollars: yearly levels of deposits in commercial banks in the United States — Dollars are not adjusted for inflation. The gap between the prime rate and the discount rate varies (often; increasing) as the interest rate increases. For thisj reason, and in the objective of maintaining simplicity and clarity, inflationary factors are not considered. ■ Number of Commercial Banks: the number of Banks designated "Commercial Banks" by the Federal Deposit Insurance Corporation Beer SOURCES: ■ Alcohol. Tobacco. and Firearms: Summary Statistics.* fiscal Years 1981-1982— Department of the Treasury, Bureau of Alcohol, Tobacco, and Firearms !■ Brewers Almanac. 1953, 1970, 1978— United States Brewers Association, Inc. ■ Census of Manufactures.* various years Historical Statistics of the United States; Colonial ’ imes to 1957^ Historical Statistics of the United States: Colonial imes to 1970^ ■ Business Statistics.* 1959, 1967, 1984 ■ Statistical Abstract of the United States,* various years ■ Modern Brewery Age Blue Book. 1985 i ■ Modern Brewery Acre. March 25, 1985 | i ■ Brewing Industry Survey, 1974— Research Company of ; America, New York | I ■ Basic Statistics. 1987— Standard & Poors Corporation j I ■ Anderson (1973) j ■ Bull, Friedrich, and Gottschalk (1984) ■ Elzinga (1986) ■ Elzinga (1974) ■ Pluta (1983) COMMENTS: ■ Production of Beer and Malt Liquor in Millions of Barrels: barrels of 31 Wine Gallons each — Small discrepancies among the sources listed for these figures result from some barrelage reported at the end of the fiscal year as opposed to the calender year. — Prohibition began January 16, 192 0, and ended April 7, 1933. During this period, it was unlawful for any firm to sell alcoholic beverages in the United States. Prior to this period, and as early as 1846, some states and counties had experimented with prohibition. Furthermore, j 156 some states remained dry after the 1933 repeal. Alabama, for example, remained dry until March 22, 1937, while] Kansas ended prohibition on May 1, 1937. j ; — Beer during the Prohibition period was a cereal beverage! i containing less than 1/2 of 1 percent alcohol by volume| ; ("near" beer). , 1 ’ t i i i ■ Number of Breweries Operated: number of brewing! i plants licensed to operate (excludes experimental , breweries) . ' — These figures may be slightly inflated due to the! ! practice of some brewers securing a Treasury license even1 i though the brewery may have been shut down. i I ; j ■ Number of Firms Producing Beer: actual number of j companies engaged in the production of beer and malt1 liquors some owning more than one brewery — Elzinga (1986, p. 217) notes that "no firm of any; significance in the brewing industry is a new entrant' since World War II." The only "entry" in fact during this period, in fact, was a firm which imported a beer ; concentrate brewed in Europe and then processed and, bottled the output domestically (Elzinga, 1974).; , Furthermore, the ten firm concentration ratio for the' brewing industry reached 98.7 percent in 1984 (Elzinga, 1986). Breakfast Cereals SOURCES: ■ Statistical Abstract of the United States.* various years ■ Census of Manufacturers.* various years I ■ Census of Manufacturers Concentration Ratios in ‘ Manufacturing.* various years ■ Scherer (1986) i 157 I COMMENTS: ; ■ Sales of Breakfast Cereal in Millions of Pounds: j includes both cereal intended to be prepared or cooked, j j and ready-to-eat cereal j i ■ Establishments Producing Breakfast Cereal:4" thoseJ | firms engaged in the manufacture of breakfast cereali I preparations 1 Chewing Gum SOURCES: ■ Census of Manufacturers.* various years j I ' ■ Census of Manufacturers Concentration Ratios in ! Manufacturing, various years \ | COMMENTS: I ■ Number of Firms Producing Chewing Gum: number of independent companies manufacturing chewing gum ■ Production of Chewing Gum in Millions of Pounds: does not include production figures for gum base to be sold to other producers --Figures for the years 1929 to 1937 were obtained by adding weights of the following ingredients: Sugar, Corn Syrup, Chicle, Crude Gum, and Essential Oils & Flavoring Extracts. Although this figure is thus only a rough approximation of production (based upon the assumption that inputs will be roughly equivalent to outputs), it is nonetheless satisfactory. 158 Cigarettes SOURCES: ■ Annual Report of Tobacco Statistics . * various years— U.S. Department of Agriculture j j ■ Statistical Supplement to the Survey of Current{ Business.* 1967 i ■ Census of Manufacturers.* various years j i ■ Statistical Abstract of the United States.* various j years ! ■ Historical Statistics of the United States: Coloniali Times to 1970 ■ Business Statistics.* ■ Industry Surveys; 1987— Standard & Poors ■ Business Week (1989) ■ COX (1933) ■ Miles (1982) ■ Overton (1981) ■ Johnson (1984) ■ Tennant (1950) COMMENTS: ■ Yearly Production in Billions of Cigarettes: as distinguished from cigars and other tobacco products ■ Number of Firms Producing Cigarettes: includes those firms who are engaged in the manufacture of tobacco cigarettes 1959, 1967, 1984 Food Beverages & Tobacco — Prior to 1947, the number of firms includes individual plants, several of which may be commonly owned by a single organization. After 1947, individual independent companies are counted. The difference between these two figures is approximately 10; there being 28 plants and 19 firms in 1947, and 14 plants and 6 firms in 1977. i — Note that the "Tobacco Trust" was formed very early in the history of the cigarette industry, dominating the : market in a monopolistic fashion. As Cox (193 3, p. 18-9) , notes: j Starting in 1890 as a consolidation of five ; manufacturers who among them controlled approximately nine-tenths of the country's output of cigarettes, then one of the minor branches of the [tobacco] industry . . . the American Tobacco : Co. had by 1910 acquired a monopolistic control of ; every branch of the industry except cigars. | ; — In 1911, the federal government ordered the Americanj Tobacco Co. partitioned into several smaller companies, j | four of which are now members of the "big six" remaining | i producers. i Gasoline SOURCES: ■ Business Statistics.* 1964 ■ Census of Manufacturers.* various years I ■ Historical Statistics of the United States: Colonial[ | Times~to 1970* j I , , ic • ] ■ Statistical Abstract of the United States. various| j years I i I I ■ Annual Statistical Review: U. S. Petroleum Industry j Statistics. 1946-1969, 1970— Department of Statistics, | American Petroleum Institute, Washington D.C. 160 ■ Basic Statistics. 1985— Standard & Poors Corporation ■ Basic Petroluem Data Book; Petroleum Industry Statistics— American Petroleum Institute, Washington D.C. i : ■ Entry and Exit in U.S. Petroleum Refining. 1948-1985. j July, 1986— American Petroleum Institute, Washington D. C. ( J ■ 1986 National Petroleum News Factbook Issue— American 1 Petroleum Institute, Washington D. C. ■ Petroleum Refineries Including Cracking Plants in the ' U.S.. various years— United States Bureau of Mines 1 ■ Hogan (1971) , ■ Williamson, Andreano, Daum, & Klose (1963) ■ Williamson & Daum (1959) j ■ de Chazeau & Kahn (1959) COMMENTS: ■ Output of Gasoline in Millions of Barrels: one barrel 1 of gasoline equals 42 gallons — Prior to 1916, figures represent production of gasoline ; and related products such as naphthas, benzene, etc. i — Between 1916 and 1970, figures represent "output" of j gasoline from domestic refineries, and includes "special | naphtha." Between 1952 and 1970, figures include unfinished gasoline production. I — After 1970, figures represent domestic production for j gasoline. i — Despite these apparent discrepancies in the data,i relatively insignificant differences actually exist, and! thus a high degree of continuity is present in this series. j ■ Number of Companies Producing Gasoline: number of independent companies refining motor gasoline 1 6 1 ■ Number of Refineries: represents number of production facilities which produce any type of distilled petroleum products j — Includes some refineries which have been shut down, but i which could have been are able to resume production within I 90 days. Home Laundry Equipment SOURCES: j ■ Census of Manufacturers,* various years | ■ Business Statistics.* 1959, 1967, 1984 ■ Historical Statistics of the United States: Colonial Times~bo 1957* ■ Historical Statistics of the United States: Coloniali Times~~to 1970^ j ■ Census of Manufacturers Concentration Ratios in j Manufacturing. various years j • • • * . * ■ Statistical Abstract of the United States. various; years j f ■ Survey of Current Business,* various editions ! I ■ Merchandising. various issues— Dealerscope Merchandising, Philadelphia, Pennsylvania (earlier issues entitled Merchandising Week, and Electrical Merchandising Week— Electronic Industries Association) * Merchandising1s 63rd Annual Statistical and Marketing Report. March, 1985— Dealerscope Merchandising,' Philadelphia, Pennsylvania I ■ Fairchild Fact File: Major Appliances and Electrical Housewares. 1985— Fairchild Publications, New York, New York ■ Hogan (1971) 162 1 COMMENTS: j ■ Factory Sales in Thousands of Units: represents! wholesale sales figures of washing machines, driers, and! I combination units — Includes domestic and import wholesale sales of both; washing machines (automatic and semi-automatic), and! i clothes driers (gas and electric); does not include; production for export. ; --Import shipments constitute less than 5% of totali figures. i j ■ Number of Firms Manufacturing Home Laundry Equipment: i includes only those firms producing complete units i — Figures for 1977 and 1982 include those firms, manufacturing electric irons. Pig Iron 1 SOURCES: ; i : i : ■ Census of Manufacturers.* various years j 1 * ' ! ■ Statistical Abstract of the United States, various j years ! ■ A Statement of the Arts and Manufactures of the United 1 States of America for the Year 1810. compiled by Tench I Coxe, Philadelphia, 1814— reprinted by the U. S. Treasury I Department ■ Hogan (1971) ■ Kuznets (1930) | COMMENTS: I I ■ Production of Pig Iron in Thousands of Short Tons: I one short ton is equal to 2000 pounds | ■ Number of Plants Producing Pig Iron: production I facilities which produce pig iron for either resale or as j ! a material for later use within that company ! I , --Unfortunately, the number of independent companies producing pig iron was not available. This figure; represents the number of autonomous plants. Railroads ! SOURCES: ■ Annual Report of the Interstate Commerce Commission.: various years— Interstate Commerce Commission, Washington i D.C. ; ■ Business Statistics.* 1959, 1967, 1984 ■ Historical Statistics of the United States: Colonialj ! Times to 1957* ' ! I ! . . ! ■ Historical Statistics of the United States: Colonial 1 Times to 1970* I * Statistical Abstract of the United States,* various! years ■ Railroad Ten Year Trends, various years— Association: of American Railroads, Washington D. C. I COMMENTS: ■ Billions of Ton-Miles: represents number of tons of rail freight carried one mile | — For the years prior to 1916, the number of ton miles represents those carried as of December 31 of each year. After 1916, this figure represents those carried as of | June 3 0 of that year. For the year 1916, the December 31 ■ and June 30 figures were averaged together in order to | approximate a smooth transition. j --Note that revenue from freight traffic has always ! contributed the major portion of railroad earnings, with I passenger revenue contributing a relatively minor share. j ■ Number of Railroad Lines: represents number of railroad lines doing business — This figure includes Class I lines, Class II lines, and Class III lines (which include switching and terminal | companies). According to the Association of American; ; Railroads, Class I railroads account for over 95 percent; ! of the industry's traffic. (Actual figures representing! , percentage of revenue accounted for by Class I railroads' are: 96.5% in 1911, 97.5% in 1916, 98.1% in 1926, 98.8% in1 1941, 99.1% in 1945, and 98.2 in 1969.) Considered alone,! only 18 Class I systems were in operation in 1985. j i — This figure in later years is overstated in that common ownership of several railroads by a single corporate entity is typical. i — For the years prior to 1916, "Number of Railroad Lines"! represents those in existence as of December 31 of each year. After 1916, this figure represents those in existence as of June 3 0 of that year. For the year 1916, | the December 31 and June 3 0 figures were averaged together in order to approximate a smooth transition. I — For the year 1880, only those lines currently in operation are listed. A total of 1146 lines were actually chartered, but many of these were still under construction. 1 6 5 Ravon and Acetate | SOURCES: ■ Historical Statistics of the United States: Colonial I Times to 1957* 1 I ' i j ■ Historical Statistics of the United States: Colonial ; Times~~to 1970^ i | j ■ Basic Statistics. various years— Standard & Poors1 . Statistical Service j I ■ Modern Textiles. various issues, including: ! • September, 1961; March, 1970; December, 1980; March 1981— Rayon Publishing Corporation, New York, New York | ! ■ Textile Organon (formerly Ravon Organon), various) i issues I ! ■ Markham (1952) I COMMENTS: ! ■ Production of Rayon and Acetate in Millions of Pounds:j I yarn and staple products j — Represents domestic production of cellulosic fibers,I including rayon and acetate (including triacetate,! saponified acetate, diacetate, nitrocellulose, andj cuprammonium). Includes both filament yarn and staple, fiber production, but does not include that produced forj cigarette filtration purposes. I I — Rayon was first produced in 1911, and was referred to as j "artificial silk" until 1926. I ■ Number of Firms Producing Cellulosic Fiber (Rayon and Acetate): includes firms producing either rayon, acetate, or both 1 6 6 T i r e s SOURCES: ■ Census of Manufacturers.* various years ■ Basic Statistics. 1987, Standard & Poors Corporation ■ Economic Report on the Manufacture and Distribution of Automotive Tires. March, 19 66— Staff Report to the Federal Trade Commission ■ Survey of Current Business.* various editions ■ Ward1s Automotive Yearbook. 1975 ■ Business Statistics.* 1959, 1967, 1984 ■ French (1986) COMMENTS: ■ Shipments in Thousands of Casings: includes total shipments of new passenger car, truck, and bus tires for original equipment, replacement, and export j i ■ Number of Firms Manufacturing Automotive Tires:\ represents those companies engaged in automotive tire manufacture — Note that according to the 1966 FTC staff report, no new entry into the tire manufacturing business occurred after - 1923. 1 67 I i I * A publication of the United States Government + Note that the term "number of establishments"! represents a figure uniquely defined by the United States! j Bureau of the Census. As stated in Historical Statistics . of the United States: Colonial Times to 1970 (p. 653): The reporting units in each census have been establishments rather than legal entities or companies. Conceptually, an establishment is a geographically isolated manufacturing unit maintaining independent bookkeeping records, I regardless of its managerial or financial • affiliations. An establishment may be a single plant, a group of closely located plants operated as a unit, or a group of closely located plants j operated by a single company without separate records for each. The establishment is also the I basic unit of industrial classification, being assigned to an industry on the basis of its | reported product of chief total value. Establishments owned and operated by the Federal Government are excluded from census coverage. The consequences of this form of classification is an inflation in the apparent number of "firms" doing business j in these industries. This discrepancy becomes greater as the industry ages, due to corporate mergers which are| unaccounted for by this method. The overall result ini industries affected by this accounting method is a1 conspicuous understatement of the "shake-out" of competition over the life of the industry. Thus, despite whatever evidence of "shake-out" is obvious in the breakfast cereal industry above, which accounts only for "establishments," the actual decline in competitive numbers is likely even more precipitous. 1 6 8 G e n e r a l C o m m e n ts f o r A l l S e r i e s i 1. In general, the effects of wars (WWI, WWII, Korean ! Conflict, etc.) are substantially different from those of depression periods. During wars, production capacity is . merely shifted to military goods, and may even be , expanded. Thus an apparent lack of production or demand j for the product under study during war periods posed no j real threat to most firms' existence, and for many, provided an opportunity. Depression periods, however, 1 resulted in an overall decrease in demand for products, | and thus threatened many firms survival. ; 2. Ideally, domestic "consumption" (of both domestic and| | imported goods) is the most desirable form of data for thei purposes of determining the "resources" or availability of; i consumer dollars in an industry. This measure, however, i ; was unobtainable for many markets. For most of the, industries presented, however, the production figures used [ are a reliable substitute, since production generally! I closely follows consumption, and since imports are ofj minimal significance in most of the markets. I 3. It is not immediately clear from the data whether| declines in the number of production facilities1 (breweries, furnaces, plants), causes, or is caused by a! consolidation/reduction in the number of companies, or if in fact these outcomes are jointly caused by some third factor. 169 WORKS CITED Aacker, David A. (1984), Strategic Market Management. New| York: John Wiley & Sons. j i and George S. Day (1986), "The Perils of High-j growth Markets," Strategic Management Journal. 7,j 409-21. I Aaronovitch, Sam, and Malcolm C. Sawyer (1975), "Mergers, Growth, and Concentration," Oxford Economic Papers. 20, n. s., 136-55. I Adler, Lee (1967), "Systems Approach to Marketing,"! Harvard Business Review. (May-June), 105-14. i 1 I Alchain, Armen A. (1950), "Uncertainty, Evolution, and j Economic Theory," Journal of Political Economy. 58, ■ 211-22. | Alderson, Wroe (1957), Marketing Behavior and Executive j Action: A Functionalist Approach to Marketing Theory.! Homewood, Illinois: Richard D. Irwin, Inc. ___________ (1964), "A Normative Theory of Marketing1 Systems," in Theory in Marketing. Reavis Cox, Wroe! Alderson, and Stanley Shapiro, eds. Homewood,; Illinois: American Marketing Association. i ____________ (1965), Dynamic Marketing Behavior: Ai Functionalist Theory of Marketing. Homewood, Illinois: j Richard D. Irwin, Inc. i Aldrich, Howard E. (1979), Organizations and Environments. Englewood Cliffs, New Jersey: Prentice Hall, Inc. __________ and Jeffrey Pfeffer (1976), "Environments of Organizations," American Sociological Review. 2, 79- 105. 1 7 0 Anderson, Carl R., and Carl P. Zeithaml (1984), "Stage of the Product Life Cycle, Business Strategy, and Business Performance," Academy of Management Journal. 27, 5-24. Anderson, Will (1973), The Beer Book: An IIlustrated Guide to American Breweriana. Princeton, New Jersey: The Pyne Press. Astley, W. Graham (1985), "The Two Ecologies: Population and Community Perspectives on Organizational Evolution," Administrative Science Quarterly. 30, 224-41. ___________ and Charles J. Fombrun (1983), "Collective Strategy: Social Ecology of Organizational Environments," Academy of Management Review. 8, 576- 87. Ayal, Igal (1981), "International Product Life Cycle: A Reassessment and Product Policy Implications," Journal of Marketing, 45, 91-6. Bain, A. D. (1963 ) , "The Growth of Demand for New Commodities," Journal of the Roval Statistical Society. A126, 285-99. Bain, Joe S. (1956), Barriers to New Competition. Cambridge: Harvard University Press. Bass, Frank M. (1969), "A New Product Growth Model for Consumer Durables," Management Science. 15, 215-27. Barnett, William P. and Glenn R. Carroll (1987), "Competition and Mutualism Among Early Telephone Companies," Administrative Science Quarterly. 32, 400-21. Betton, John and Gregory G. Dess (1984), "The Application of Population Ecology Models to the Study of Organizations," Academy of Management Review. 10, 750-7. Boone, Louis E., and David L. Kurtz (1977), Contemporary Marketing, second edition. Hinsdale, Illinois: The Dryden Press. Boulding, Kenneth E. (1950), A Reconstruction of Economics. New York: Wiley. 1 7 1 __________ (1972), "Economics and General Systems," in The Relevance of General Systems Theory: Papers Presented to Ludwig von Bertalanffy on His Seventieth Birthday. Ervin Laszlo, ed. New York: George Braziller, Inc. Brittain, Jack W. and John H. Freeman (1980), "Organizational Proliferation and Density Dependent Selection," In The Organizational Life Cycle. John * Kimberly and Robert H. Miles, eds. San Fransisco: i Jossey Bass, 291-338. Brown, W. L., and E. O. Wilson (1956), "Character Displacement," Systematic Zoology. 5, 49-64. Buaron, Roberto (1981), "New-game Strategies," The McKinsev Quarterly. (Spring), 24-40. Bull, Donald, Manfred Friedrich, and Robert Gottschalk (1984), American Breweries. Trumbull, Connecticut: i Bullworks. ! I Business Week (1982), "The Coming Shakeout in Personal ^ Computers," November 22, 72-83. | i I 1 Business Week (1983), "Personal Computers: And the Winner , is IBM," October 3, 76-8. 1 i Buskirk, Richard H. (1970), Principles of Marketing: The . Management View. third edition. New York: Holt, Rinehart, and Winston, Inc. j i Buzzell, Robert D. (1972), Marketing: A Contemporary ! Analysis, second edition. New York: McGraw-Hill Book Company. ____________ (1981a), "Are There 'Natural' Market Structures?" Journal of Marketing. 45, (Winter), 42- 51. __________ (1981b), "Competitive Behavior and Product Life Cycles," in New Ideas for Successful Marketing, Wright and J. Goldstucker, eds. Chicago: American Marketing Association. __________ , Victor J. Cook, James R. Peterson, and Paul Hase (1969), Product Life Cycles. Cambridge, Massachusetts: Marketing Science Institute. 1 72 Camillus, John C. (1989), "Technology-Driven and Market- Driven Life Cycles: Implications for Multinational Corporate Strategy," in Global Strategic Management Perspectives. Jagdish N. Sheth and Golpira S. Eshghi, j eds. Cincinnati, Ohio: South-Western Publishing Co, I 31-8. Catry, Bernard, and Michel Chevalier (1974), "Market Share Strategy and the Product Life Cycle," Journal of ! Marketing. 38, 29-34. Chambers, John C., Satinder K. Mullick, and Donald D. Smith (1974), An Executive1s Guide to Forecasting. New York, John Wiley & Sons. 'Chapman, S. J. , and T. S. Ashton (1914), "The Sizes of ! Businesses, Mainly in the Textile Industries," Journal of the Roval Statistical Society. 77, 469-549. 'Clifford, Donald K. , Jr. (1965), "Managing the Product i | Life Cycle," Management Review. (June), 36-8. j j •Collins, Norman R. , and Lee E. Preston (1961), "The Size j Distribution of the Largest Industrial Firms, 1909- ! 1958," The American Economic Review, 51, 986-1011. i Comanor, William S., and Thomas A. Wilson (1967), i "Advertising, Market Structure, and Performance," i Review of Economics and Statistics. 49, 423-40. __________ and Thomas A. Wilson (1974), Advertising and Market Power. Cambridge, Massachusetts: Harvard j University Press. I ; Cox, Reavis (1933), Competition in the American Tobacco [ I Industry: 1911-1932. reprinted in 1968. New York: AMS Press, Inc. Cox, William E., Jr. (1967), "Product Life Cycles as Marketing Models," The Journal of Business, 40, 375- 83. Cravens, David W., Gerald E. Hills, and Robert B. Woodruff (1980), Marketing Decision Making: Concepts and Strategy. revised edition. Homewood, Illinois: Richard D. Irwin, Inc. Crawford, C. Merle (1984), "Business Took the Wrong Life j Cycle From Biology," Journal of Consumer Marketing. 1, I 5-11. 173 I I i 1 Cunningham, M. T. (1969), "The Application of Product Life Cycles to Corporate Strategy: Some Research Findings," British Journal of Marketing. (Spring), 32-44. Davenport, Herbert J. (193 6), The Economics of Enterprise. New York: Macmillan Publishing Company. J Davidson, William, R. , Albert D. Bates, and Stephen J. Bass (1976), "The Retail Life Cycle," Harvard Business Review. November-December, 89-96. Dawkins, Richard (1986), The Blind Watchmaker. New York: W. W. Norton & Company. Day, George S. (1981), "The Product Life Cycle: Analysis and Applications Issue," Journal of Marketing. 45, 60-7. __________ (1977), "Diagnosing the Product Portfolio," Journal of Marketing. (April), 29-38. ___________ and Robin Wensley (1983), "Marketing With a Strategic Orientation," Journal of Marketing. (Fall), 79-89. DeBach, Paul (1966), "The Competitive Displacement and Coexistence Principles," Annual Review of Entvmology. j 11, 183-212. i de Chazeau, Melvin G. , and Alfred E. Kahn (1959) , Integration and Competition in the Petroleum Industry: Petroleum Monograph Series. Volume 3.. New Haven, Connecticut: Yale University Press. Doyle, Peter (1976), "The Realities of the Product Life Cycle," Quarterly Review of Marketing. (Summer), 1-6. Eichner, Alfred S. (1969), The Emergence of Oligopoly: Sugar Refining as a Case Study. Baltimore, Ohio: The Johns Hopkins Press. Elzinga, Kenneth G. (1974), "The Restructuring of the U.S. Brewing Indstry," Modern Brewery Age. (October 14), MS 52 —63. 1 74 __________ (1986), "The Beer Industry," in The Structure of American Industry. seventh edition, Walter Adams, ed. New York: Macmillan Publishing Company. 203-38. I jEmmel, Thomas C. (1973), An Introduction to Ecology and ! Population Biology. New York: W. W. Norton & Company, j Inc. sEngel, James F. , Roger D. Blackwell, & Paul W. Miniard ' (1986), Consumer Behavior, fifth edition. New York: ! The Dryden Press. i j Enis, Ben M. (1980), "GE, PIMS, BCG, and the PLC, " I Business. May-June, 10-8. I '___________, Raymond LaGarce, and Arthur E. Prell (1977), j "Extending the Product Life Cycle," Business Horizons. j (June). iFlath, David, and E. W. Leonard (1981), "A Comparison of Two Logit Models in the Analysis of Qualitative Marketing Data," in New Product Forecasting: Models and Applications. Yoram Wind, Vijay Mahajan, and Richard N. Cardozo, eds. Lexington, Massachusetts: Lexington Books, 155-67. Forrester, Jay W. (1958), "Industrial Dynamics," Harvard Business Review. July-August, 37-66. (1961) Industrial Dynamics. Cambridge, Massachusetts: MIT Press. Frazier, Gary L., and Roy D. Howell (1983), "Business Definition and Performance," Journal of Marketing. 47, 59-67. French, Michael (1986), "Structural Change and Competition in the United States Tire Industry, 1920-1937," Business History Review, 60, (Spring), 28-54. Friedman, Milton (1962), Capitalism and Freedom. Chicago: University of Chicago Press. Fourt, Louis A., and Joseph W. Woodblock (1960), "Early Prediction of Market Success for New Grocery Products," Journal of Marketing. 25, 31-8. Fox, Harold W. (1973), "A Framework for Functional Coordination," Atlanta Economic Review. November- December, 8-11. 1 7 5 Fox, Edward J. , and Edward W. Wheatley (1978), Modern Marketing; Principles and Practice. Glenview, Illinois: Scott, Foresman and Company. Gause, G. F. ( 1934) , The Struggle for Existence. Baltimore: Williams & Wilkins, (reprinted in 1964 by Hafner Publishing Co., New York). George, Kenneth D. , and Aubrey Silberston (1975), "The Causes and Effects of Mergers," Scottish Journal of Political Economy. 22, (Jume), 179-93. Gilly, Mary C., and Ben M. Enis (1982), "Recycling the Family Life Cycle: A Proposal for Redefinition and ■ Re-interpretation," in Advances in Consumer Research. j Volume 9, Andrew Mitchell, ed. Ann Arbor, Michigan: j Association for Consumer Research. I Goodall, D. W. (1951), "Quantitative Aspects of Plant Distribution," Biological Review. 27, 194-245. Gort, Michael (1969), "An Economic Disturbance Theory of Mergers," Quarterly J ournal of Economics. 83, (November), 624-42. ___________ and Akira Konakayama (1982) , "A Model of Diffusion in the Production of an Innovation," The American Economic Review. 72, 1111-20. Greiner, Larry E. (1972), "Evolution and Revolution as ! Organizations Grow," Harvard Business Review. (July- j August), 37-46. Grether, E. T. (1983), "Regional-Spatial Analysis in Marketing," Journal of Marketing. 47, (Fall), 36-43. Griliches, Zvi (1957) , "Hybrid Corn: An Exploration in the Economics of Technological Change," Econometrica. 25, 501-22. Gross, Irwin (1968), "Toward a General Theory of Product Evolution," working paper number p-43-10 (September), Marketing Science Institute. Gwinner, Robert F. , Kenneth L. Rowe, Stephen W. Brown, John L. Schlacter, Alfred J. Hagan, Alfred H. Schmidt, Lonnie L. Ostrom, and David L. Shrock (1977) , Marketing: An Environmental Perspective. New York: West Publishing Company. 1 7 6 Hallam, A. (1976), "Continental Drift and the Fossil Record," in Continents Adrift and Continents Aground: Readings from Scientific American. San Francisco: W. H. Freeman and Company. Haire, Mason (1959), "Biological Models and Empirical Histories of the Growth of Organizations," in Modern Organization Theory. Mason Haire, ed. New York: 272-306. i iHannan, Michael T. and John Freeman (1977) , "The 1 Population Ecology of Organizations," American Journal ( of. Sociology. 82, 929-64. Harrell, Stephen G., and Elmer D. Taylor (1981), "Modeling i the Product Life Cycle for Consumer Durables," Journal j ■ of Marketing. 45, 68-75. J ,Harris, Brian F. (1979), Shared Monopoly and the Cereal , Industry: An Empirical Investigation of the Effects ‘ of the FTC1s Antitrust Proposals. East Lansing, Michigan: MSU Business Studies. Hart, P. E., and S. J. Prais (1956), "The Analysis of ; Business Concentration: A Statistical Approach," Journal of the Royal Statistical Society. A119, 150- 81. ; Hartley, Robert F. (1976), Marketing Fundamentals for | Responsive Management. New York: Dun-Donnelley ! Publishing Corporation. | J Hasty, Ronald W. , and R. Ted Will (1975), Marketing. San Francisco: Canfield Press. Hawley, Amos H. (1981), "Human Ecology," American | Behavioral Scientist. 24, 423-44. j Hay, Donald A. (1976), "Sequential Entry and Entry- i Deterring Strategies in Spatial Competition," Oxford ! Economic Papers. 28, (July), 240-57. j ____________ and Derek J. Morris (1979), Industrial i Economics. New York: Oxford University Press. Hayes, Robert H., and Stephen C. Wheelwright (1979), "Link ' Manufacturing Process and Product Life Cycles," Harvard Business Review. January-February. 1 77 Henderson, Bruce D. (1983), "The Anatomy of Competiiton," Journal of Marketing. 47, (Spring), 7-11. __________ (1984), "Competitors, the Forgotten Factor in Marketing," The Canadian Business Review. Summer, 32- 5. Hirshleifer, J. (1977), "Economics From a Biological Viewpoint," The Journal of Law and Economics. 20, (April), 1-52. Hise, Richard T., Peter L. Gillett, and John K. Ryans, Jr. (1979) , Basic Marketing: Concepts and Decisions. Cambridge, Massachusetts: Winthrop Publishers, Inc. Hofer, Charles W. (1975), "Toward a Contingency Theory of Business Strategy," Academy of Management Journal. (December), 784-810. !Hogan, William T. (1971), Economic History of the Iron and j Steel Industry in the United States: Volumes 1-5. j Lexington, Massachusetts: Lexington Books. j ^Hotelling, Harold (1929), "Stability in Competition," | j Economic Journal. 39, (March), 41-57. ; Howard, John A. (1983), "Marketing Theory of the Firm," Journal of Marketing. 47, 90-100. Hrebiniak, Lawrence G. and William F. Joyce (1985), "Organizational Adaptation: Strategic Choice and Environmental Determinism," Administrative Science Quarterly. 30, 336-49. Hunt, Shelby D. (1983), "General Theories and the Fundamental Explanada of Marketing," J ournal of Marketing. (Fall), 9-17. I I , James A. Muncy, and Nina M. Ray (1981), ! "Alderson1s General Theory of Marketing: A ' Formalization," in Review of Marketing. 1981. Ben I Enis, and Kenneth Roering, eds. Chicago: American j Marketing Association. j Hutchinson, G. Evelyn (1957), "Concluding Remarks," Cold \ Spring Harbor Symposium on Quantitative Biology. XXII. 1 Population Studies: Animal Ecology and Demography. 1 Cold Spring Harbor, New York: The Biological i Laboratory, 415-27. ' __________ (1959), "Homage to Santa Rosalia, or Why Are There So Many Kinds of Animals?" The American Naturalist. 93, (May-June), 145-59. Johnson, Paul R. (1984), The Economics of the Tobacco j Industry. New York: Prager. j Karnani, Aneel (1984), "The Value of Market Share and the j Product Life Cycle--A Game-Theoretic Model," j Management Science. 30, 696-712. ! I Kimberly, John R. (1980), "The Life Cycle Analogy and the j Study of Organizations: Introduction," in The i Organizational Life Cycle: Issues in the Creation. Transformation. and Decline of Organizations. John R. | Kimberly, Robert H. Miles, and Associates, eds. San Francisco: Jossey-Bass Publishers, 1-17. , I Kotler, Philip (1965), "Competitive Strategies for New j Product Marketing Over the Life Cycle," Management Science. 12, (December), B104-19. : ___________ (1988), Marketing Management: Analysis. Planning. Implementation. and Control. sixth « edition. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Kotrba, R. William (1966), "The Strategy Selection Chart," ! ! Journal of Marketing. 30, 22-5. jKovac, F. J., and M. F. Dague (1972), "Forecasting by Product Life Cycle Analysis," Research Management. I | (July), 66-72. Krebs, Charles J. (1978), Ecology: The Experimental Analysis of Distribution and Abundance. second edition. New York: Harper & Row, Publishers. Kurten, Bjorn (1976), "Continental Drift and Evolution," in Continents Adrift and Continents Aground: Readings from Scientific American. San Francisco: W. H. j Freeman and Company. j Kurtz, David L. and Louis E. Boone (1981), Marketing. New j York: The Dryden Press. Kuznets, Simon S. (1930), Secular Movements in Production and Prices: Their Nature and Their Bearing Upon Cyclical Fluctuations. Boston: Houghton Mifflin Company. 1 79 Lawrence, Kenneth D., and William H. Lawton (1981), "Applications of Diffusion Models: Some Empirical Results," in New Product Forecasting: Models and Applications. Yoram Wind, Vijay Mahajan, and Richard ! N. Cardozo, eds. Lexington, Massachusetts: Lexington Books, 529-41. 1 Lazer, William, and Eric H. Shaw (1986), "The Product Life! Cycle," in Handbook of Modern Marketing. second1 i edition, Victor Buell, ed. New York: McGraw-Hill! Book Company. i j Levitt, Theodore (1965), "Exploit the Product Life Cycle,"^ Harvard Business Review. (November-December), 81-94. I I Levy, Seymour, and Gordon Donhowe (1962), "Exploration of; I a Biological Model of Industrial Organization," The f Journal of Business. 35, (October), 335-42. i Lewis, Richard J. and Leo G. Erickson (1969), "Marketing^ Functions and Marketing Systems: A Synthesis," Journal of Marketing. 33, (July), 10-4. | Lillis, Charles M. , Chem L. Narayana, and John L. Gilman (1976), "Competitive Advantage Variation Over the Life; Cycle of a Franchise," Journal of Marketing. October, 77-80. Low, Richard E. (1970), Modern Economic Organization. Homewood, Illinois: Richard D. Irwin, Inc. Luck, David J. (1972), Product Pol icy and Strategy. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. __________ and Arthur E. Prell (1968), Market Strategy.j New York: Appleton-Century Crofts. __________ and O. C. Ferrell (1985), Marketing Strategy! j and Plans. second edition. Englewood Cliffs, New; j Jersey: Prentice-Hall, Inc. j i ■ MacArthur, Robert H. (1972), Geographical Ecology. New- i York: Harper and Row, Publishers. | I I j _____________and Edward O. Wilson (1967) , The Theory of j 1 Island Biogeography. Princeton, New Jersey: Princeton1 University Press. I I i i iso : Mahajan, Vijay, and Eitan Muller (1981), "Innovation Diffusion and New-Product Growth Models in Marketing," in New Product Forecasting: Models and Applications. Yorai Wind, Vijay Mahajan, and Richard N. Cardozo, eds. Lexington, Massachusetts: Lexington Books, 425-56. Mansfield, Edwin (1962), "Entry, Gibrat's Law, Innovation, and the Growth of Firms," American Economic Review. | 52, 1023-51. I Markham, Jesse W. (1952 ) , Competition in the Ravon j Industry. Cambridge, Massachusetts: Harvard J University Press. I ; __________ (1955) , "Survey of the Evidence and Findings on I I Mergers," in Business Concentration and Price Policy, j Princeton, New Jersey: Princeton University Press, ! 141-82. i i ' i ! Marshall, Alfred (1890), Principles of Economics: An 1 Indroductorv Volume. (eighth edition, published in 1953). New York: The Macmillan Company. McKelvey, Bill (1982) Organizational Svstematics. ! Berkeley, California: University of California Press. Merrell, David J. (1981) , Ecological Divers itv. Minneapolis, Minnesota: University of Minnesota Press. Midgley, David F. (1980), Toward a Theory of the Product Life Cycle: Some Testable Propositions. working paper number 95, Center for Marketing Studies, University of California, Los Angeles. Miles, Robert H. (1982), Coffin Nails and Corporate Strategies. Englewood Cliffs, New Jersey: Prentice- Hall, Inc. Miller, James Grier (1978), Living Systems. New York: McGraw-Hill Book Company. Mittelstaedt, Robert A. (1986), "Perfectly Heterogeneous Demand and the Structure of Competitive Markets," paper presented at the 11th Annual Macromarketing Conference, Boulder, Colorado, (August). Morgan, Gareth (1986), Images of Organization. Beverly Hills, California: Sage Publications. 1 8 1 Mueller, Dennis (1972), "A Life-Cycle Theory of the Firm," Journal of Industrial Economics, (July), 199-218. Mueller, Willard F., and Larry G. Hamm (1974), "Trends im ; Industrial Market Concentration: 1947-1970," The j 1 Review of Economics and Statistics, 56, 511-20. i Mullor-Sebastian, Alice (1983), "The Product Life Cycle) Theory: Emprirical Evidence," Journal of International| Business Studies, Winter, 95-105. [ i Nelson, Ralph L. (1959) , Merger Movements in American1 j Industry. 1895-1956. Princeton, New Jersey: Princetonj i University Press. 1 ' I i ' I ___________ (1966), "Business Cycle Factors in the Choice! Between Internal and External Growth," in The: Corporate Merger. William W. Alberts and Joel E. Segall, eds. Chicago: University of Chicago Press, | 52-66. Nicholls, J. A. F. , and Sydney Roslow (1986), "The 'S- Curve': An Aid to Strategic Marketing," The Journal of Consumer Marketing. 3, (Spring), 53-64. I Onkvisit, Sak, and John J. Shaw (1983), "An Examination of ! the International Product Life Cycle and its 1 Application Within Marketing," Colubmia Journal of ; World Business, 18, (Fall), 73-9. ; Orians, Gordon H., and Gerald Collier (1963), "Competition i and Blackbird Social Systems," Evolution. 17, (December), 449-59. I !___________ and Mary F. Willson (1964), "Interspecific! j Territories of Birds," Ecology. 45, (Autumn), 736-45. I ! Orr, Dale (1974), "The Determinants of Entry: A Study of i j the Canadian Manufacturing Industries," The Review of i I Economics and Statistics. 56, 58-66. ! 0 1Shaughnessy, John (1984), Competitive Marketing: A Strategic Approach. Boston: George Allen & Unwin. ' Overton, James (1981), "Diversification and International, Expansion: The Future of the American Tobacco Manufacturing Industry with Corporate Profiles of the j ' Big Six1," in The Tobacco Industry in Transition. j William Finger, ed. Lexington, Massachusetts:! Lexington Books. ] 182 Park, C. Whan, and Gerald Zaltman (1987), Marketing Management. New York: The Dryden Press. Patton, Arch (1959), "Top Management's Stake in the Product Life Cycle," Management Review, (June) 3-26. Pearl, Raymond (1924), Studies in Human Biologv. Baltimore: Williams & Wilkins Company. IPennings, Johannes M. (1980), "Environmental Influences on < I the Creation Process," in The Organizational Life ’ Cvcle: Issues in the Creation. Transformation. and ' J Decline of Organizations. John R. Kimberly, Robert H. : ' Miles, and Associates, eds. San Francisco: Jossey- 1 Bass Publishers, 134-63. jPenrose, Edith T. (1952), "Biological Analogies in the j i Theory of the Firm," American Economic Review. 42, I 804-19. I Penrose, Edith T. (1959) , The Theory of Growth of the Firm. Oxford: Blackwell. I Pessemier, Edgar A. (1982), Product Management: Strategy 1 and Organization. second edition. New York: John Wiley & Sons. ; Pianka, Eric R. (1974), "Niche Overlap and Diffuse ; Competition," Proceedings of the National Academy of Science. USA. 71, 2141-5. IPianka, Eric R. (1978), Evolutionary Ecology. second ! edition. New York: Harper & Row, Publishers. I jPxelou, E. C. (1975), Ecological Diversity. New York: j John Wiley & Sons. i __________ (1977), Mathematical Ecology. New York: John Wiley & Sons. i __________ (1979), Biogeographv. New York: John Wiley & Sons. i Pluta, Joseph E. (1983), Regional Change in the U.S. j Brewing Industry. Austin, Texas: Bureau of Business | Research, Graduate School of Business, University of * Texas at Austin. ! 183 Polli, Rolando, and Victor Cook (1969), ''Validity of the Product Life Cycle," The Journal of Business. 42, 385-400. Porter, Michael E. (1979), "The Structure Within Industries and Companies' Performance," The Review of Economics and Statistics. 61, (May), 214-77. __________ (1980a), Competitive Strategy. New York: The Free Press. ________ (1980b), "Industry Structure and Competitive Strategy: Keys to Profitability," Financial Analysts Journal/ (July-August), 3 0-41. | Prescott, John E. (1986), "Environments as Moderators of j the Relationship Between Strategy and Performance," I Academy of Management Journal. 29, 329-46. i j Preston, Frank W. (1948), "The Commonness, and Rarity of j Species," Ecology. 29, 254-83. j __________ (1962a), "The Canonical Distribution of| I Commonness and Rarity: Part I," Ecology. 43, 185-215. ' i * j ____________ (1962b), "The Canonical Distribution of ; Commonness and Rarity: Part II," Ecology. 43, 410-32. j______________(1969) , "Diversity and Stability in the' ' Biological World," in Diversity and Stability in; i Ecological Systems. Report of Brookhaven Symposium j Number 22. Upton, New York: Brookhaven Nationalj Laboratories, 1-12. I Product Marketing. (1977), "Keeping Products Alive and I Well in the Market," November, 46-53. 1 i Reibstein, David J. (1985), Marketing: Concepts . j Strategies. and Decisions. Englewood Cliffs, Newi Jersey: Prentice-Hall, Inc. [ i I Reidenbach, R. Eric, and Terrence A. Oliva (1981), "General Living Systems Theory and Marketing: a ! Framework for Analysis," Journal of Marketing. 45, j 30-7. | Rink, David R. , and John E. Swan (1979), "Product Life Cycle Research: A Literature Review," Journal of Business Research. 78, 219-42. 1 84 Rogers, Everett, M. (1983), Diffusion of Innovations. third edition. New York: The Free Press. : ___________ and F. F. Shoemaker (1971), Communication of I Innovations: A Cross-Cultural Approach. New York: I The Free Press. | Rumelt, Richard P. (1979), "Evaluation of Strategy: Theory and Models," in Strategic Management: A New View of Business Policy and Planning. Dan E. Schendelj and Charles W. Hofer, eds. Boston: Little, Brown andj Company, 196-212. j i 1 J Savich, Richard S., and Laurence A. Thompson (197 8),; "Resource Allocation Within the Product Life Cycle,"! MSU Business Topics. Autumn, 35-44. ! I i Savitt, Ronald (1980), "Historical Research in Marketing,"’ Journal of Marketing. 44, (Fall), 52-8. j ! Scherer, Frederic. M. (1980), Industrial Market Structure j and Economic Performance. second edition. Boston:| I Houghton Mifflin Company. ! __________ (1986), "The Breakfast Cereal Industry," in The Structure of American Industry. seventh edition, j Walter Adams, ed. New York: Macmillan Publishing Company, 172-202. j Scheuing, Eberhard E. (1969), "The Product Life Cycle as an Aid in Strategy Decisions," Management International Review. 9, 4-5, 111-24. Schmalensee, Richard (1978), "Entry Deterrence in the Ready-to-eat Breakfast Cereal Industry," Bell Journal of Economics. 9, (Autumn), 305-27. j i Schmenner, Roger W. (1983), "Every Factory Has a Life' Cycle," Harvard Business Review. March-April, 121-29. , Schoell, William F. , and Joseph P. Guiltinan (1988), i Marketing: Contemporary Concepts and Practices. third j edition. Boston: Allyn and Bacon, Inc. | I Sheth, Jagdish N. (1986), "Global Markets or Global Competition?" Journal of Consumer Marketing. 3, (Spring), 9-11. I t 1 8 5 Simon, Herbert A., and Charles P. Bonini (1958), "The Size Distribution of Business Firms," The American Economic Review. 48, 607-17. Simpson, George G. (1969) "The First Three Billion Years' of Community Evolution," in Diversity and Stability, | in Ecological Systems. Report of Brookhaven Symposium! I Number 22. Upton, New York: Brookhaven Nationalj j Laboratories, 162-76. | | Smallwood, John E. (1973), "The Product Life Cycle: A Key; i to Strategic Marketing Planning," MSU Business Topics.; | (Winter), 29-35. i ■ Steindl, Josef (1965), Random Processes and the Growth of! Firms: A Study of the Pareto Law. New York: Hafnerj j Publishing Company. j 1 Stinchcombe, Arthur L. (1965), "Social Structure and1 Organizations," in Handbook of Organizations. James G., March, ed. Chicago: Rand McNally & Company. Tellis, Gerard J. and C. Merle Crawford (1981), "An ■ Evolutionary Approach to Product Growth Theory,", Journal of Marketing. 45, 125-32. , Tennant, Richard B. (1950) , The American Cigarette' Industry. New Haven, Connecticut: Yale University, Press. Thomas, Robert Paul (1977), An Analysis of the Pattern of Growth of the Automobile Industry: 1895-1929. New i York: Arno Press. i i ] i Thorelli, Hans B. (1967), "Ecology in Marketing," The i i Southern Journal of Business. 4, (October), 19-25. j i Thorelli, Hans, and Stephen Burnett (1981), "The Nature of I ! Product Life Cycles for Industrial Goods Business,"i Journal of Marketing. 45, 97-108. l i Tigert, Douglas, and Behrooz Farivar (1981), "The Bass Newj Product Growth Model: A Sensitivity Analysis for a' High Technology Product," Journal of Marketing. 45, 81-90. Townsend, Harry (1968), Scale. Innovation. Merger and Monopoly: An Introduction to Industrial Economics. New York: Pergamon Press. 1 8 6 Tucker, Donald J., Jitendra V. Singh, Agnes G. Meinhard, and Robert J. House (1988) , "Ecological and Institutional Sources of Change in Organizational Populations," in Ecological Models of Organizations. Glenn R. Carroll, ed. Cambridge: Ballinger Publishing Company. Turban, Efraim, and Jack R. Meredith (1985), Fundamentals of Management Science, third edition. Plano, Texas: Business Publications, Inc. j i Ulrich, David and Jay B. Barney (1984) , "Perspectives in | Organizations: Resource Dependence, Efficiency, and I Population," Academy of Management Review. 9, 471-81. Vernon, R. (1966), "International Investment and International Trade in the Product Cycle," Quarterly Journal of Economics. 80, 190-207. Waltman, Paul (1983), Competition Models in Populationj Biology. Philadelphia, Pennsylvania: Society fori Industrial and Applied Mathematics. Wasson, Chester R. (1974), Dynamic Competitive Strategy and Product Life Cycles. St. Charles, Illinois: : Challenge Books. Webb, S. David (1976), "Mammalian Faunal Dynamics of the j Great American Interchange," Paleobiology. 2, 22 0-34. j Weber, John A. (1976a), "Planning Corporate Growth With I Inverted Product Life Cycles," Long Range Planning. (October). i Weber, John A. (1976b) , Growth Opportunity Analysis. | Reston, Virginia: Reston Publishing Company, Inc. j . Wells, W. D. , and G. Gubar (1966), "Life Cycle Concept m | Marketing Research," Journal of Marketing Research. ! (November), 335-63. j Wells, Louis T. , Jr. (1968), "A Product Life Cycle fori International Trade?" Journal of Marketing. 32, 1-6. J I ____________ (1972), The Product Life Cycle and International Trade. Boston: Harvard University. White, Lawrence J. (1971), The Automobile Industry Since I 1945. Cambridge, Massachusetts: Harvard University I Press. Whittaker, Robert H. (1970), Communities and Ecosystems. New York: Macmillan. __________ (1972), "Evolution and Measurement of Species Diversity," Taxon. 21, 213-51. \ | __________ , S. A. Levin, and R. B. Root (1973), "Niche, i Habitat, and Ecotope," American Naturalist. 107, 3 21- ; 38. I , # . Wholey, Douglas R. and Jack W. Brittaxn (1986), "Organizational Ecology: Findings and Implications," Academy of Management Review. 11, 513-33. I Williamson, Harold F., Ralph L. Andreano, Arnold R. Daum,i and Gilbert C. Klose (1963) , The American Petroleum Industry: The Age of Energy 1899-1959. Evanston, Illinois: Northwestern University Press. Williamson, Harold F., and Arnold R. Daum (1959), The American Petroleum Industry: The Age of Illumination. Evanston, Illinois: Northwestern University Press. Wind, Yoram J. (1982), Product Policy: Concepts. Methods. and Strategy. Reading, Massachusetts: Addison-Wesley1 Publishing Company. Winter, Sidney G. (1971), "Satisficing, Selection, and the Innovating Remnant," Quarterly Journal of Economics. 85, 237-61. ! Young, Ruth C. (1988), "Is Population Ecology a Useful Paradigm for the Study of Organizations?" American Journal of Sociology. 94, (July), 1-24.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Communications medium and product class: Their effect on negotiation outcome

PDF

A contingent claims analysis of the equity interest in property-casualty insurance companies

PDF

A conceptual framework of distribution intensity

PDF

A quantitative model for the analysis of warranty policies

PDF

Cross-sectional tests of tax and regulatory costs of a change in depreciation methods in the railroad industry

PDF

An investigation of information requirements determination and analogical problem solving

PDF

An examination of the effect of environmental cues on audit judgments

PDF

A study of the double sulphates of some rare-earth elements with sodium, potassium, ammonium and thallium

PDF

A study of the English Canadian feature film industry 1977-1981

PDF

Career advancement in managerial hierarchies in the United States firms: A multi-theoretic model and empirical tests of the determinants of managerial promotion

PDF

An investigation of the impact of congruent managerial values on appraisal behavior

PDF

A theory of deterministic consumer choice behavior: Applying generalizability theory of measurement to consumer panel data

PDF

An empirical investigation of variables related to management consulting engagement effectiveness as perceived by small business clients and their 'Big Eight' CPA consultants

PDF

A study on the content and determinants of close exchange relationships in channels of distribution

PDF

A grounded theory account of strategic decision-making as a key mediating mechanism between organization and environment

PDF

An empirical study of executive stock options and company performance

PDF

A critical study of the hydrogenation of talloel

PDF

Credit markets under imperfect information

PDF

Crystallographic studies of metal hydrides and other transition metal complexes

PDF

A new approach to the problem of measuring the properties of micelles

Asset Metadata

Creator Winsor, Robert D (author)

Core Title A biogeographic theory of industrial market structure and competitive dynamics

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Business Administration

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag business administration, general,OAI-PMH Harvest

Language English

Advisor [illegible] (committee chair), [illegible] (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-757956

Unique identifier UC11348152

Identifier DP22663.pdf (filename),usctheses-c17-757956 (legacy record id)

Legacy Identifier DP22663.pdf

Dmrecord 757956

Document Type Dissertation

Rights Winsor, Robert D.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA