The Ecology of Entrepreneurship

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/227267198 The Ecology of Entrepreneurship Chapter · March 2006 DOI: 10.1007/0-387-23622-8_9 CITATIONS READS 40 562 2 authors, including: Glenn R. Carroll Stanford University 142 PUBLICATIONS 10,528 CITATIONS SEE PROFILE Some of the authors of this publication are also working on these related projects: "Authentic Kitsch?: Donn, Vic and the Emergence of the Tiki Bar Social Category" Glenn R Carroll and Dennis Ray Wheaton View project All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately. Available from: Glenn R. Carroll Retrieved on: 18 October 2016 GLENN R. CARROLL AND OLGA M. KHESSINA THE ECOLOGY OF ENTREPRENEURSHIP1 1. INTRODUCTION Although entrepreneurship has been defined in many different ways, a common definition ties it closely to the establishment of new organizations and their viability in early years.2 By this view, the focal entrepreneurial event is the operational inception of a new firm, and an entrepreneur is an individual who plays a central role in opening a new firm. The associated entrepreneurial process consists of resource mobilization to facilitate and sustain the opening event; it typically involves many individuals, agents and social institutions, with a variety of interests. Accordingly, an entrepreneurial area or industry is one characterized by high rates of new firm establishment. Within organizational sociology, the most active line of theory and research investigating the establishment of new organizations is commonly referred to as organizational ecology (Hannan & Freeman, 1989). As is well known, organizational ecology studies populations of organizations, focusing on how they change over time, especially through demographic processes of selective replacement–organizational founding, mortality, and growth. The major theory fragments of organizational ecology predicting levels of new organizational establishments include density dependence, structural inertia, niche width, resource partitioning and others (see Carroll & Hannan, 2000). Outsiders Pfeffer (1993) and Hargens (2000) view the perspective as exceptionally cumulative for social science. Despite the apparent relevance of organizational ecology, little theory and research in the field of entrepreneurship uses it, builds on it, or even acknowledges its existence through citations. For instance, we examined all articles in a recent year (2003) of issues of the Journal of Business Venturing, the top specialized publication outlet in entrepreneurship. Of the 43 articles published, a total of 2,225 other articles were cited in the reference lists. Of these, only 59 (roughly 2.7%) cited organizational ecology articles and books, loosely defined.3 It is not clear why entrepreneurship analysts commonly ignore the ecological body of scientific work that might inform their efforts. We speculate that at least two different reasons might be responsible.4 First, the discipline of sociology, within which organizational ecology emerged, is not particularly concerned with applications, especially those of a managerial nature. This is true of much ecology as well.5 Because of the long time lag required for basic research to affect practice, it may therefore take substantial periods6 for the practical implications of new developments in sociology to be fully understood and appreciated in an applied domain such as entrepreneurship.7 Second, organizational ecology theory typically casts predictions about new organizational establishment at the population level of analysis: these arguments predict the rates of founding expected in populations under various environmental conditions. Historically, entrepreneurship has been examined as an individual-level phenomenon, with an emphasis on rationality of action. Inferences about individual behavior from empirical analysis of aggregated data can be misleading. Moreover, the types of theoretical explanations offered for population-level processes may not be directly applicable to individuals (see Aldrich & Wiedenmayer, 1993). What this situation suggests to us is that organizational ecology likely requires some translation before its relevance and applicability become apparent to entrepreneurship theorists and researchers. Background conditions that induce or limit the propensity with which individuals and others start new organizations will almost always be relevant to the study of entrepreneurship. But both the abstract conceptualization of those conditions and the aggregate level of empirical analysis common in ecology may leave implicit the significance of theory and research for the individual level so engrained in the entrepreneurship area. In our view, translations to the individual level are not necessarily straightforward and may be open to multiple interpretations. Population Founding Founding Attempts Success in Founding Attempts Organizational Mortality Mortality in New Organizations Figure 1. Sequential processes involved in founding and success of new organizations Given the ambiguity in deriving the implications of ecological studies to the level of primary interest in the entrepreneurship area, one of our primary aims here is to ease this task by offering a conceptual framework. The framework involves viewing new venture success and failure as a process involving rates of event occurrence: a population founding rate (decomposed into two constituent rates) and an individual organizational mortality rate. Within this framework, the observable founding rate of a population is depicted as the outcome of two underlying rates, each associated with a stage in the startup process. The first is the rate at which attempts at founding are initiated; the second is the rate of success of those attempts. Success in this second process is associated with the founding event itself and is to be distinguished from the success or failure of the organization once founded, which is a separate issue involving the third rate in the framework, the organizational mortality rate. The mortality rate is associated with individual organizations in a population, given that they have been founded. As depicted in Figure 1, new potential organizations travel down a sequential process, which involves step-by-step winnowing of the number of units. Organizational ecology provides theory and research about each of these pertinent types of rates, as we review in greater depth below. Among other things, our translation consists of sorting the various ecological arguments of founding rates into the two underlying component processes. We also point to several findings from ecological studies of organizational mortality that may potentially interest entrepreneurship scholars. We then shift to examine key debates about the sources of variation in founding rates in the ecological perspective. These include differences about the roles of: (1) macro vs. task environments, (2) endogenous vs. exogenous processes, and (3) agglomeration vs. differentiation. Finally, we identify a number of emerging research areas that present interesting prospects for the future of ecological studies of foundings, and perhaps entrepreneurship as well. 2. ECOLOGICAL VITAL RATES AND ENTERPRENEURSHIP Most ecological theories of organizational foundings make predictions about the rate at which new organizations will appear in a population under specified conditions. These analyses take the organizational population as the unit of analysis for basic conceptual and methodological reasons. In particular, to avoid endogenous sampling on the outcome, the founding event must occur to some pre-existing unit, which is viewed as “at risk” to experience the event. For many biological organisms, the female members of the species (perhaps only those within a certain age range) are viewed as the risk set. Organizations, however, do not have the requisite identifiable maternal unit around which to construct such a set. Even using individuals or natural persons as the unit of risk falls short on this dimension because they have multiple identities and can combine in numerous ways in starting organizations, thus generating an uncountable number of potential units at risk (see Hannan & Freeman, 1987). In our view, a population's rate of founding can be usefully decomposed into two underlying component rates. As Delacroix & Carroll (1983: 276) note: The number of organizations founded in a given period will reflect both the level of organizational attempts and the relative success of these attempts. Increasing either of these factors will result in an increased number of foundings. That is, organizational founding consists of two separate processes, a first about the rate at which individuals are propelled to initiate actions that attempt to start new organizations, and a second about the success of those attempts, in the sense that a firm actually gets opened. In a few cases, researchers have found social or legal contexts that allow them to study these two processes more or less directly (Carroll & Hannan, 2000, Ch. 15; Sørensen & Sorenson, 2003; Ruef Aldrich & Carter, 2003; Ruef, 2004; Jovanovic, 2004). But in most cases, the only available information is on the actual founding events, which represents a culmination of the operation of both processes. Nonetheless, we maintain that the theoretical arguments that ecologists typically offer to explain the overall founding rate commonly rely more heavily on one or the other process, and ignore the second process. We believe that clarifying and classifying these arguments will go some distance in making organizational ecology accessible to entrepreneurship analysts. This is because, in conceptual terms, each of these constituent processes seems easier and more straightforward to link to entrepreneurship theory and research than the typical aggregate founding study.8 So, the first part of our review consists of sorting the various ecological arguments of founding rates into these two processes and pointing to some illustrative studies. 2.1. Rate of Founding and Its Component Parts Decomposing the population’s founding rate into its component parts presents some conceptual challenges but also serves to provide a clearer connection to theory typically used in the study of entrepreneurship. The major conceptual challenge, about which there is no consensus, entails specifying the activities associated with the event defining the initiation of an organizing attempt. Abstractly, one can imagine the process beginning at a variety of stages including idea formulation, development of a business plan, assembly of human resources, capital acquisition, and the like.9 In the sparse empirical research on this process, investigators have adopted a practical stance, using convenient unambiguous indicators that clearly precede any true founding event. For instance, in studying potential automobile producers, Carroll and Hannan (2000) use listing in an industry directory, legal incorporation or development of an automobile (usually a prototype). For potential television broadcasters, Sørensen and Sorenson (2003) use applications for television broadcasting permits. For heterogenous potential entrepreneurs, Ruef et al. (2003) survey individuals to identify those actively involved in the process of resource mobilization for a firm in which they held an ownership stake.10 Ruef (2004) uses public announcements of the planned opening of medical schools. For present purposes, the major reason for decomposing the population’s founding rate is that it allows us to understand ecological theory in light of ideas and arguments common to the entrepreneurship area. To see this, first note again that most theory and research in the entrepreneurship area is tied to the individual level of analysis. Recognize also that the most commonly held theoretical model of individual entrepreneurs found in the traditional literature assumes rational action and efficient markets. In its unreconstructed form, this view holds that entrepreneurs are profit-seeking individuals who make (basically) accurate cost-benefit calculations of the likelihood of success of their planned enterprises and then behave accordingly.11 That is, an entrepreneur is someone who sees a market opportunity when it appears, moves to take advantage of it, and then reaps ample rewards after entering the market. Decisions and actions in this view of the world are forward-looking: current behavior is explained as a result of future expectations, which (in interpretations) are often assumed to be accurate (March, 1978). March and Olsen (1989: 5-6) call the set of assumptions underlying this view “historical efficiency,” because “institutions and behavior are thought to evolve through some sort of efficient historical process” leading rapidly to a unique outcome. In terms of the set of sequential component rates of the founding process described above, the model of individual rational action under historical efficiency implies a tight coupling of the two rates, meaning that the rates only move together in consistent directions. Under strong historically efficient rationality, entrepreneurs attempt to enter markets more often when they are likely to succeed in doing so and when their new enterprises are more likely to succeed in competition with others (see related discussion in Barnett, Swanson & Sorenson, 2003). These assumptions allow analysts to move back and forth from empirical facts to theoretical interpretations with ease, and without regard to causal directionality. For instance, an observed high founding rate implies that new ventures are more likely to succeed (because future favorable outcomes were seen by the many new entrepreneurs and these projections motivated them). It also implies higher rates of initiation of organizing attempts (because more entrepreneurs saw the future conditions and tried to enter) and success at founding attempts (the entrepreneurs who tried to enter worked harder, and were more likely to secure resources from gatekeepers and others who also came to see the future favorable conditions). Ecological theory and research departs from the entrepreneurship tradition in two ways. First, ecological theory operates at the population level of analysis. Second, and more important from a conceptual viewpoint, ecology does not assume a tight coupling between the two constituent parts of the founding rate. For instance, Hannan and Carroll (1992) regard each of the population’s component founding rates as driven by environmental selection processes. In their view of the world, An empirically estimated organizational founding rate reflects both types of selection processes, and these may work at cross-purposes. High rates of founding attempts can be coupled with low rates of success in organizing and vice versa. The same observed level of founding may be the result of very different environmental forces. Moreover, rationality may characterize only one of the two portions of the founding process. That is, the apparent rationality of a founding process may reflect the rationality of selection operating on random founding attempts. (Hannan & Carroll, 1992: 199-200) Note that their argument is not just about the mathematical possibility of rates diverging but about the likelihood of social and economic environmental forces making them do so. Hannan and Carroll (1992) go on to list a few factors that might drive one rate high while carrying opposing implications for the other rate. For instance, involuntary unemployment or forced retirement likely increases the likelihood of attempting to found a new business but may not increase its odds of success. Conversely, a strong regulatory regime may decrease the rate of attempts but increase the success rate of those that do. The level of analysis difference between ecology and entrepreneurship is widely recognized. But, we think that the abandoned assumption of tight coupling across the constituent parts of the founding rate is not generally recognized–or at least its full implications are not. We attribute this situation to a lack of complete conceptualization by ecologists. In our view, in formulating predictions about population level founding rates, most ecological theories rely primarily on arguments about either the rate of initiation of attempts or the rate of success of those attempts but not both. The typical theoretical argument discusses factors or processes affecting one component rate but not the other, and then proceeds to derive a consistent population-level prediction. But, such a derivation is only valid if the other, second, rate moves in the same direction, is neutral, or, if opposing in its direction, exerts less force than the focal rate. That is, only with some degree of coupling of the component rates does the usual population-level prediction make logical sense when based on an inference from only one of the components. In fact, after reviewing empirical studies of foundings, we determined that reliance on arguments about the rate of founding attempts is the most popular mode of theorizing among organizational ecologists.12 For illustration, Table 1 provides classifications of some selected studies that adopt this approach. More abstractly, we suggest that, as currently developed, theories about foundings that focus on the following mechanisms tend to use the rate of founding attempts: legitimation processes of various kinds (Marrett, 1980; Hannan, 1984; Studer-Ellis, 1995; Hannan & Carroll, 1995) including those associated with laws and regulations (Dobbin & Dowd, 1997; Wade, Swaminathan & Saxon, 1998); freed organization-specific resources (Delacroix & Carroll, 1983), including labor (Haveman & Cohen, 1994); collective action processes that increase visibility, commitment, solidarity (Carroll & Swaminathan, 2000); and niche formation (Swaminathan & Delacroix, 1991; Swaminathan, 1995). But other ecological studies and theories about foundings do rely on arguments about the success of founding attempts. Typically, these point to market or other environmental conditions that should serve to increase the viability or fitness of a particular type of organization, whose founding rate is then expected to increase. Table 1 also provides some illustrative quotes of studies of this type. In terms of general mechanisms, we suggest that, as currently developed, theories relying on the following tend to use the rate of success: competition processes, including density (Hannan, 1984), niche width (Dobrev, Kim & Hannan, 2001), niche overlap (Baum & Singh, 1994), and experience distributions (Barnett & Sorenson, 2003); differentiation processes, including resource partitioning (Carroll, 1985); collective action processes that provide support and defense for a set of organizations (Hannan & Freeman, 1987); and trust engendered through homophily (Ruef et al., 2003). To make these theories and studies more useful to the entrepreneurship area, we suggest that analysts should attempt to complete the conceptualization across the sequence of rates and work through the implications at the individual level (for an admirable example, see Ruef, 2004).13 While tight coupling of the component rates may be the easiest and most straightforward way to accomplish this task, we personally doubt that it is the most interesting and, ultimately, the most sociologically valid. Widely held notions of bounded rationality suggest that while expectations about the future may guide individual behavior, common social situations are filled with uncertainty, ambiguity and imperfect information frequently, thereby making them tenuous at best.14 Much of the recent literature on entrepreneurship builds on these notions in attempting to construct a new model of entrepreneurship (see Ruef et al., 2003). If properly integrated, ecological theories and studies might provide fresh ideas and evidence for the enterprise. For instance, Ruef (2004) shows that entrepreneurial inertia, slowness in moving to the operational stage of organization, may induce cycles in the evolution of the organizational population. Table 1. Illustrative arguments drawn from ecological studies of foundings Arguments about the rate of attempts Arguments about the success of attempts Potential entrepreneurs react to signals, created by previous failings, foundings, and new niches. Delacroix, Swaminathan and Solt (1989) Success in building a concrete organization depends on the availability of resources, both human and material…founding rates rise when the level of resources rise… Hannan and Freeman (1987) Organizational failings in the previous year may act as a signal of a poor environment for league success to entrepreneurs so as to slow the founding rate in the subsequent year. Land et al. (1994) Increasing concentration frees peripheral resources that allow more foundings at the niche periphery. Swaminathan (1995) When ethnicity is highly salient, the niche for ethnic organizations expands inducing members to build ethic organizations. When society is polarized, it becomes dangerous to erect and strengthen ethic boundaries. West (1995) Competitive relationships can be asymmetric, i.e., expansion in one population may legitimate the other, but growth in the second may worsen the life chances of the first by eroding its resource base. Lomi (1995a) Potential founders will be more sensitive to local variations in the levels of legitimation and competition because of limits in their capacity to collect information on non-local resource conditions and because of the ambiguity involved in interpreting events in more distant sites. Lomi (1995b) Organization density initially opens opportunities for protest and organization building, but later closes them. Growing density of SMOS establish a resource and institutional space facilitating mobilization by other movements. Minkoff (1997) Most new policies create constraints and incentives, rather than dictating firm behavior, and managers conduct new business strategies taking them into account. Dobbin and Dowd (1997) Cross-movement adoption of tactics entails intermovement competition, which in turn impedes a movement’s ability to mobilize social support. Olzak and Uhrig (2001) State-level anti-alcohol prohibition frees resources… entrepreneurs in neighboring states found breweries. Breweries in the prohibition state also have an incentive to shift their production to adjacent states. Prohibition’s normative effect: entrepreneurs less likely to found new breweries, because they fear losing investments. Wade et al. (1998) Foundings occur at a higher rate in industrial clusters because in such areas entrepreneurs more likely obtain the resources necessary for launching new firms. Stuart and Sorenson (2003) Entrepreneurial action occurs within a web of social relations that both enable and constrain activity …dense local concentrations of structurally equivalent organizations increase the pool of potential entrepreneurs in a region, thereby increasing founding rates. Sorenson and Audia (2000) Political instability leads to economic recession, which in turn diminishes the willingness of entrepreneurs to commit resources and invest capital, given the uncertainty of future returns. Dobrev (2001) 2.2. Rate of Mortality (Success of New Ventures) The entrepreneurship literature concerns itself not only with founding of new organizations but also with their viability in early years. Within this tradition, ecological studies are notable because of their comprehensiveness of coverage, often including observations on all organizational members of an historical population (see Carroll & Hannan, 2000). In our view, the ecological design provides a sounder basis for inference about many questions of interest and sometimes leads to different findings. For instance, it seems to us that most scholars of entrepreneurship assume that new organizations are at a higher risk of failure than established firms (Venkataraman, 1997). But a long tradition of ecological studies of historical populations now suggests that this assumption is wrong. Specifically, two different streams of empirical research find that new firms do not display higher mortality rates than that of established organizations. One stream of research, concerned with modeling age-dependence, focuses on how survival rates of firms change as they gain experience in a focal industry. The second stream of research compares survival rates by entry mode, in particular, of new ventures with those of pre-existing entrants from other industries. 2.2.1. Age Dependence The key objective of research on age dependence in organizational mortality is to understand how and why the viability of organizations changes as their tenure in a market increases (for reviews see: Hannan, 1998; Carroll & Hannan, 2000; Carroll & Khessina, 2004). Initially, organizational ecologists took a position similar to that of students of entrepreneurship: they assumed that newly founded organizations fail at a higher rate than older firms (Carroll & Delacroix, 1982; Freeman, Carroll & Hannan, 1983). In terms of modeling, this view implied organizational mortality rates with negative (or declining) age dependence. Theoretical support for this belief originated with Stinchcombe (1965), who argued that new organizations do not possess well-developed organizational routines and employment structures and established relationships with environmental actors. As a result, new firms are capable of only suboptimal operation and are at a high risk of failure. Over time, however, as surviving new ventures develop organizational routines and establish relationships with the environment, they operate more efficiently and their mortality rates decrease. Evidence of this “liability of newness” is found in many early empirical studies in ecology (Carroll & Delacroix, 1982; Freeman, Carroll & Hannan, 1983; Singh, Tucker & House, 1986; Olzak & West, 1991). Beliefs about negative age-dependence were later amended in the light of new theoretical and empirical developments. Some scholars propose that organizations might be subject to a “liability of adolescence.” They argue that at the time of founding, firms enjoy relatively low mortality rates, because they can draw from the initial resource endowment. During the first months or years of operation, as the initial supply of resources get exhausted, the mortality rates of new firms increase. Subsequently, failure rates decrease, because in the process of aging, firms develop operational routines and environmental positions (Carroll & Huo, 1987; Brüderl & Schüssler, 1990; Fichman and Levinthal, 1991). Because mortality rates are still seen as declining over most of an organization's life, this view is often regarded as a modification or refinement of the negative age dependence pattern rather than a contradiction (Hannan, 1998). A recent set of studies challenge the very idea of negative age dependence. They find that organizations exhibit positive age dependence in mortality, meaning that older organizations fail at a higher rate than young firms. In interpreting these findings, scholars argue that as firms get older, they are more likely to deplete their resources and become misaligned with the changing environment (i.e., experience a “liability of obsolescence”), and develop dysfunctional capabilities, such as political coalitions and the like (i.e., experience a “liability of senescence”). Positive age dependence has been detected in a number of empirical studies, including especially those that control for the time-varying effects of organizational size, something that previous studies did not (Barron, West & Hannan, 1994; Hannan, 1998; Khessina, 2003). 2.2.2. Entry Mode The second stream of ecological research that provides insight into understanding the viability of new ventures examines entry mode. It explores how and why firms entering an industry as new ventures (de novo firms) and those entering by diversification away from another industry (de alio firms) differ in their survival chances. Analysts typically assume that at the time of entry into the industry, start-ups and diversifiers differ systematically in their initial resource endowments and prior experiences. These differences, in turn, ought to create variations in organizational survival (Bruderl & Schussler, 1990; Carroll, Bigelow, Seidel & Tsai, 1996; Freeman, 1990; Hannan, Carroll, Dobrev & Han, 1998; Mitchell, 1994). A basic prediction contends that de alio firms enjoy significant advantages in survival, because at the time of entry they possess more resources and greater experience than de novo firms. The initial advantage of de alio firms has been documented in numerous empirical studies, spanning industries, countries and historical periods. Specifically, it has been shown that de alio firms exhibit significantly lower mortality rates than de novo firms in the following populations: American labor unions (Hannan & Freeman, 1988), U.S. semiconductor producers (Freeman, 1990; Barnett & Freeman, 2001), in the population of new firms in West Germany (Bruderl, Preisendorfer & Ziegler, 1992), U.S. automobile manufacturers (Rao, 1994; Carroll et al., 1996), facsimile transmission service organizations in Manhattan (Baum, Korn & Kotha, 1995), British, French and German automobile producers (Hannan et al., 1998; Dobrev et al., 2001), U.S. television receiver manufacturers (Klepper & Simons, 2000), and U.S. computer companies (Swanson, 2002; Barnett et al., 2003).15 In studying the U.S. automobile industry, Carroll et al. (1996) elaborate on the basic argument that de alio firms possess survival advantages. They reason that de novo firms enjoy greater flexibility and that, over time, this fact shifts the comparative advantage to de novo firms if the environment changes fast enough. Empirical studies in the U.S. automobile industry (Carroll et al., 1996), in the British, French, and German automobile industries (Hannan et al., 1998), and in the U.S. medical equipment industry (Mitchell, 1994), show, indeed, that while mortality rates of de novo firms were initially significantly higher that those of de alio firms, the survival rates of de novo firms converges with those of de alio firms as firms' tenure in the focal industry increases. Despite the evidence about de novo and de alio firms’ differences in their organizational outcomes, questions remain about the mechanisms driving them. In addressing this deficiency, Khessina (2003) speculates about the relationships between entry mode and innovative behavior, on the one hand, and entry mode and product dynamics, on the other hand. She suggests that the greater flexibility of de novo firms should allow them to innovate at a higher rate than de alio firms, but more developed organizational competence should allow de alio firms to attain more beneficial product outcomes. Khessina's (2003) empirical analysis of product innovation rates in the worldwide optical disk drive industry during the period of 1983 to 1999 confirms much of this speculation. She finds that de novo firms did, in fact, innovate at a higher rate in their products than de alio firms.16 De alio firms, however, display an advantage at product dynamics: their products show longer market life and generate stronger competitive pressures (Khessina, 2003; Khessina & Carroll, 2002). Given these countervailing effects, it is not surprising that although de novo firms show higher failure rates than de alio firms in this industry, the difference is not statistically significant. Otherwise, this non-effect would be an odd result given the vast empirical evidence of the survival advantages of de alio firms reviewed above (but see Banbury & Mitchell, 1995, for another exception). 3. KEY DISTINCTIONS IN PROCESSES DRIVING FOUNDINGS Ecologists approach the study of organizational foundings from a variety of viewpoints and conceptual frameworks. Indeed, the difference around which we organized the review of empirical studies in the above section represents one such background conceptual variation, namely, whether theoretical arguments rest primarily on ideas about the ease of launching founding attempts or on ideas about the likelihood that such efforts will succeed in producing a founding. Another related conceptual difference lies in the degree to which ecologists emphasize the causal importance of entrepreneurs' assessments of market openings versus the facilitating role played by available resources and institutional gatekeepers. For instance, the theory of resource partitioning (Carroll, 1985) predicts that market opportunities arise in certain environmental conditions and that entrepreneurs will appear with organizations designed to exploit these opportunities; by contrast, Halliday, Powell and Granfors (1987) argue that state associations should be founded at higher rates in environments where organizational knowledge has accumulated because founding costs will be lower. In this section, we identify and discuss three additional conceptual or theoretical differences in the ways ecologists typically seek to account for variations in foundings. These are differences about the roles of: (1) macro vs. task environments, (2) endogenous vs. exogenous processes, and (3) agglomeration vs. differentiation. It is our view, that while an analyst’s position on one or more of these issues is frequently used as a point of emphasis, the underlying theoretical differences are usually not investigated or appreciated in their entirety. We think that theory and research would be improved if these differences rose to the level of debate and spurred competitive empirical tests. At the current moment, we know little about the relative importance of many of the detailed effects documented in the literature. 3.1. Macro Environment vs. Task Environment One difference in the ways ecologists analyze foundings involves the degree of specificity (or proximity) conceptualized in environmental variables. Many studies include variables measuring some aspect of the macro socio-economic environment (e.g., GNP, war-year dummies, depression-year dummies, collective violence), while others include variables that impinge more directly on the focal population, variables that organizational theorists sometimes refer to as part of the “task environment” (e.g., product market size, number of competitors, prices of supplies; prohibitions of various kinds). In the typical study, both types of variables are included but the theoretical and analytical focus lies almost exclusively with the task environment variables; the macro environmental variables are treated as controls. An interesting popular variable in this context is population age, a measure of the time elapsed since the first appearance of an organization in the population. This variable is often used in model specifications but it has been treated in various ways, as part of the macro environment and as part of the task environment. While commonly regarded as a control, population age has also been seen as reflective of some underlying process. For instance, Halliday et al. (1987) suggests that population age reflects the accumulated stock of organizational knowledge in a population. In the most developed theory, Hannan (1997) argues that population age tracks population inertia, the process by which organizational processes become structurally locked in and institutionalized. Our view is that ecologists in general have been too casual in their treatments of the macro environment, simply taking whatever data is handy and regarding macro variables as providing controls so as to avoid interpretation. The problem is that while these variables often show big effects, the effects often vary across studies and specifications; inclusion or exclusion of these variables in a model often affects the findings for the variables of primary theoretical interest. Accordingly, we suggest that research would be improved if greater attention were paid to these specifications. We also suggest it would improve the situation if researchers were expected to make substantive interpretations of these effects, which could then be compared across studies. At least then anomalies could be identified as such. Although it is a long way off, this approach might ultimately lead to a conceptual framework for theory about the macro environment, something current practice seems unlikely to generate. 3.2. Endogenous vs. Exogenous Processes Perhaps the most active broad conceptual difference in the way ecological analysts approach the study of organizational foundings concerns whether they focus on endogenous or exogenous processes. Broadly defined, in this context endogenous processes are those that involve the postures and interactions of the individual entrepreneurs and organizations, while exogenous processes are those that involve individual and organizational responses to extra-population agents and social infrastructures, including the state, professional bodies and norms. Within the endogenous approach, we would place processes of density dependence, niche overlap, organizational differentiation and resource partitioning. Density dependence theory holds that foundings decrease in high-density populations (Hannan, 1984). Whether this is because entrepreneurs avoid highly competitive settings and thus attempt fewer foundings, or because their attempts are less likely to succeed, remains unresolved. Research on niche overlap looks at individual organizational niches and identifies their points of intersection, as when, say, two banks target the same consumer market with similar products within a geographic locale. Theory predicts that areas in the population’s niche space with less overlap will experience higher numbers of foundings. Arguments about organizational differentiation use very similar logic but extend it to proximity in non-overlapping areas of the niche space. By this view, proximity in the niche space indicates competition. Empirically, studies of agglomeration often analyze, separately, one-dimensional continuous representations of niche space, say geographical location or price. Resource-partitioning theory holds that when large generalists dominate a market, they cover a central position where most environmental resources are located, thereby leaving open previously occupied space in the periphery of the market. Under such conditions, specialist organizations are expected to be founded in the less competitive periphery. This argument, like those of niche overlap theory and organizational differentiation, assumes (at least implicitly) that aggregate entrepreneurial behavior is strongly rational in the sense that highly competitive contexts will show fewer foundings than less competitive ones. None of these arguments, however, advances very specific characterizations about individual entrepreneurial behavior, its motivations or its consequences, thus making difficult the sort of conceptual links advocated above. Frequently, theorists explain their predictions by resorting to implicit rational behavior. For instance, Hannan and Freeman (1989: 132) claim that “Given a set of environmental conditions that set a carrying capacity, the more abundant the competitors, the smaller will be the potential gains from founding an organization at a given level of demand for products and services.” At other times, as with Carroll and Swaminathan’s (2000) resource partitioning analysis of the microbrewery movement, entrepreneurs appear to be passionately driven by social movement-like goals. But, conceptually speaking, there should be no doubt that most theories of competition leave plenty of room for a fully rational interpretation, whereby individual entrepreneurs attempt more foundings when conditions are favorable and these attempts are more likely to meet with success in the same conditions. For exogenous processes, broad historical changes in the normative order of a population are often well known at the time of study. Research that resorts to descriptions of these broad sweeping changes, or interpretations of such based on historical time or period effects, runs the risk of restating what seems obvious to those who know the history. (e.g., Studer-Ellis’s, 1995, description of the evolution of women’s colleges). The difficulty for empirical research lies with identifying specific environmental actions or events that drive these processes. To accomplish this task, researchers often look to governmental actions as well as those of professional bodies that lie within the institutional domain. In our view, there is no applicable general theory about how entrepreneurs react to exogenous institutional and governmental changes in the environment. These studies almost always base their predictions on a detailed understanding of how particular types of entrepreneurs and organizations are likely to react to specific institutional developments. Some of them make arguments about how newly available organizationspecific resources are likely to encourage or thwart entrepreneurs. For instance, Tucker, Singh and Meinhard (1990) show that initiation of a governmental funding program for social service in Canada (Opportunities for Youth) increases the rate of founding of voluntary social services program; and, conversely, a subsequent funds reduction program by the government (Provincial Restraint) lowers the founding rate. Likewise, Swaminathan (1995) demonstrates that passage of a law allowing farm wineries to operate under special rules leads to a higher founding rate of specialist wineries. Similarly, Ingram and Inman (1996) show that development of a tourist park around Niagara Falls (as a result of public investment) generates a higher hotel founding rate. In a study of African-American protest organizations, Minkoff (1995) finds that the availability of outside funding yields more new organizations. Wade et al. (1998) offer a more subtle analysis in their study of the effects of statelevel prohibitions on brewery foundings. They argue that prohibitions in adjacent states initially increase foundings in a focal state because consumers will seek products there; however, as the number of adjacent states with prohibitions increases, the effect reverses and dampens foundings because these developments reflect general normative changes about alcohol. Other studies of how exogenous processes affect foundings typically look at broad governmental actions and trace their effects to particular populations. For instance, Dobbin and Dowd (1997) document the ways in which federal pro-cartel and anti-trust policies affected the founding rate of railroads; they also examined the effect of public capitalization of corporations. Similarly, Studer-Ellis (1995) shows that passage of women’s suffrage laws in the U.S. increased the founding rate of women’s colleges. The best of these studies examines both specific and general institutional changes and ties them to founding processes in subtle and unexpected ways. For instance, Simons and Ingram (2002; 2004) show the complex ways that organizational foundings of the ideologically charged (utopian socialist) kibbutz organizational form are affected by the political environment of Israel–manifested in state establishment, political violence, size of the Jewish population, and the densities of politically competitive organizations (Moshav and capitalist corporations). They find that the kibbutz founding rate is enhanced by the size of the Jewish population, and lowered by the density of the ideologically oppositional capitalist organizational forms as well Moshav organizations. They also find that the rate is usually lowered when the state takes on greater authority (i.e., establishment of the state in Palestine and Israel). However, they also find that political violence raises the founding rate, as does location in a politically contested geographic region. Whether looking at specific or broad exogenous changes, the underlying model of individual entrepreneurial behavior used by institutional theorists is typically not well specified; and arguments about, or implying, individual behavior are often incomplete and inconsistent. Nonetheless, in our view, most of these arguments concern primarily actions that facilitate attempts at foundings rather than the success of those attempts. Enabling laws, public funds and resources, and norms about appropriate ways to organize all likely affect entrepreneurial perceptions about opportunity. In fact, these actions may over-stimulate such perceptions, leading to overly high numbers of organizing attempts, and thus making success at founding less likely. 3.3. Agglomeration vs. Differentiation The final conceptual distinction we consider is between those studies that posit forces of agglomeration as driving organizational foundings as opposed to those that focus on differentiation processes. When niche space is explicitly operationalized and entrepreneurs are assumed to be rational and possessing accurate market knowledge, this conceptual difference is clear and straightforward. For instance, Baum and Haveman (1996) predict hotel foundings in Manhattan by assessing geographic positions on two separate dimension of niche space: room price and hotel size. They find that hotels tend to locate near similar hotels on the price dimension, thus showing agglomeration effects (as entrepreneurs are assumed to benefit from spillovers in this situation), but avoid similar hotels on the size dimension, thus showing differentiation tendencies (as entrepreneurs are assumed to benefit from avoiding head-to-head competition on size). With such assumptions and an explicit representation of niche space, empirical analysis readily tells which force operated in producing an observed pattern of foundings. It is much more challenging, however, to specify a priori which force will operate on a given dimension and to explain why. It is also more complicated, but potentially more interesting, if we do not assume rationality or accurate market knowledge on behalf of the entrepreneurs. In other words, we need better theory about the conditions and mechanisms behind the operation of agglomeration or differentiation. In many ecological studies of foundings, the agglomeration/differentiation distinction is only implicit, perhaps because only one or the other of the processes is entailed in the main argument advanced. For instance, several empirical studies examine the role of population niche overlap and non-overlap in generating foundings (Baum & Singh, 1994; Baum & Oliver, 1996), with expectations that foundings will be more prevalent in areas of non-overlap than those of overlap. Such an approach relies on the differentiation principle and assumes that entrepreneurs are at least intendedly rational in avoiding direct competition along the dimensions contained in the overlap measure.17 Similar assumptions are implied, we suggest, in the usual predictions about foundings drawn from theories of density dependence (competition component only, Hannan, 1986), resource partitioning (Carroll, 1985), and Red Queen competition (Barnett & Sorenson, 2002), although in many of these cases the implied differentiation in the face of high competition is not elsewhere within the population but in another population. Studies that rely on an agglomeration logic also invoke a variety of specific mechanisms including economic benefits that spillover (Baum & Haveman, 1996), legitimation (Hannan & Carroll, 1992; Studer-Ellis, 1995), collective action (Carroll & Swaminathan, 2000; Swaminathan & Wade, 2001), and social networks (Sorenson & Audia, 2000). With the exception of spillover economic benefits, what is especially interesting about these mechanisms is that they do not require assumptions about entrepreneurial rationality or knowledge in order to operate. For instance, in discussing how social networks operate to induce foundings in the footwear industry, Sorenson and Audia (2000: 426) claim that, “…entrepreneurial action occurs within a web of social relations that both enable and constrain activity…dense local concentrations of structurally equivalent organizations increase the pool of entrepreneurs in a region, thereby increasing founding rates.” Here there is no implication that these entrepreneurs are individually rational or that they possess (or think they possess) any information about what will lead to market success. Instead, they are propelled into action because of the pressures and information that come to them by virtue of their location in the social structure. 4. EMERGING RESEARCH AREAS A number of new research questions concerning the founding rates of new ventures have been addressed recently by organizational ecologists. Progress has been made in understanding how entrepreneurs move from preproduction activities into production, why established firms spin-off new ventures, how new venture formation is related to niche position, what the dimensions of market niches are (with special emphasis on identity), and how social movement-like behavior arises (how niches are socially constructed). 4.1. Preproduction/Initial Organizing Although many ecologists consider the founding of a new organization to be a result of a sequential two-step process (as discussed above), very little is known about what happens in between an organizing attempt and an actual founding. Many entrepreneurs manage to mobilize initial resources, register their organizations, and even create a prototype of their product or service, but do not yet deliver their products to the market. Ecologists and others call such organizations and their organizers preproducers (Carroll & Hannan, 2000; Jovanivich, 2004) or preoperational entrants (Ruef, 2004). Recent research has made some progress in understanding factors affecting the activities and fates of preproducers. Carroll and Hannan (2000) suggest that the rate at which entrepreneurs enter into the preproduction depends on organizational density. Assuming that entrepreneurs react to changes in the environment, preproduction attempts should increase during the legitimation period of industry evolution (when resources abound, founding attempts succeed and mortality rates are low) and decrease during the competition period (with scarce resources, founding and survival prospects are dull). However, analysis of the U.S. automobile industry reveals that the rate of initiation monotonically increases with the density of either producers or preproducers (Carroll & Hannan, 2002). This pattern makes sense if preproducers require only limited resources. Another important question about preproducers concerns how they assemble human resources. Ruef et al. (2003) proposed that the membership of preproducer teams likely draws from preexisting social networks as well as ecological constraints on the availability of others. Such recruitment processes will likely reflect homophily, disproportionate similarity based on gender, family and ethnicity. In other words, entrepreneurs are likely to recruit their team members among people they have strong ties with (i.e., family, friends, and colleagues). In studying organizational founding teams sampled from the U.S. population, Ruef et al. (2003) found indeed that the composition of founding teams is primarily homophilydriven, along lines of gender, ethnicity and occupation. That is, founding teams are composed disproportionately of persons of the same gender, ethnicity and occupation. The drive toward homophily is so strong that it overcomes the necessity for (functional) occupational diversity in entrepreneurial teams. The study also finds that social networks strongly affect team composition, with team members being drawn disproportionately from persons possessing strong social ties with other team members (strangers and those with weak ties are not more likely to be included). Both findings suggest a theoretical interpretation that entrepreneurs are more likely to rely on persons they are comfortable with and trust rather than persons who might hold needed technical and other skills but differ socially. Finally, the study also finds some weaker support for the notion of ecological constraint, which holds that numerical minorities become disproportionately isolated in the team composition process because of their relative lack of availability. That is, numerical minorities are more likely to become solo entrepreneurs.18 Ruef (2004) also decomposes the entry process in analyzing the establishment rates of medical schools in the U.S. He finds that the density of operational schools shows greater impact on the entry rate than does total density (preoperational and operational schools) because of social visibility, as with Carroll and Hannan's (2000) automobile preproducers, Ruef also finds differences in founding rates by organizational form and by rates of concurrent entry. Finally, Ruef finds that organizations with longer preoperational tenures enjoy better survival chances. This last finding he interprets as entrepreneurial inertia and shows that it can induce cycles in the size of organizational population. How long does it take preproducers to move into the production stage and start shipping their products to the market? What determines the length of the transition from preproduction to production? Schoonhoven, Eisenhardt and Lyman (1990) proposed that entrepreneurs’ characteristics, as well as organizational and environmental factors affect the transition time into production. The transition takes longer in industries with a high level of technological change and in industries with intense competition. Ventures founded by entrepreneurs with either relevant industry experience or experience of starting ventures, or by former colleagues display a shorter transition time in the semiconductor industry. Contrary to expectations, abundant financial resources slowed this transition. Attributes of entrepreneurs did not have any significant effect. Schoonhoven et al. (1990) assume that all preproducers sooner or later become producers. Yet in the real world this transition may never occur. Carroll and Hannan (2000) found that in the U.S. automobile industry during 1886 to 1982, only 11% of 3,845 preproducers were able to move to the production stage. What factors affect the probability that a firm succeeds in becoming a full-fledged producer? A firm’s successful movement from preproduction into production may depend on a variety of factors. Hannan and Freeman (1989) propose that success in the transition differs across organizational forms. For example, complex organizations (which are difficult to build) likely spend more time in the preproduction stage and may never move into the production. Hannan and Freeman further suggest that the probability of successful transition may depend on a preproducer’s ability to understand the industrial environment. The more difficult it is to predict the future, the less likely is the successful transition from preproduction to production. Since the ability to forecast the future declines with the length of the forecast period, firms that spend a long time in preproduction are less likely to move into production. Carroll and Hannan (2000) find partial support for this proposition when analyzing transitions from preproduction into production in the U.S. automobile industry. They establish that the transition rate is nonmonotonically related to the time a firm spent in preproduction. Specifically, the rate of a preproducer’s transition into production is very low for the first half-year in preproduction, then increases rapidly through year six, and then declines slowly. Carroll and Hannan (2000) also find that the transition rate depends strongly on preproducer density and weakly on producer density. The transition rate increases with increasing preproducer density, at low densities, but decreases with increasing either producer or preproducer density, at high densities. Besides interest in the transition from preproduction to production, organizational ecologists also study how preproduction activities affect organizational and industrial evolution. Carroll and Hannan (2000) propose that the density of preproducers in the industry has similar effects on legitimation and competition processes, and, therefore, on organizational founding rates, as the density of producers. Specifically, at low densities, an increasing number of organizing attempts should enhance the taken-for-granted status of a population by facilitating recognition of the organizational form by society, facilitating collective action by the populations’ members, and accelerating collective learning about the form. At high densities, the preproducer density should increase competition for the limited resources as the population reaches its carrying capacity. As expected, these inverted U-shaped effects of preproducer density are found in a study of American automobile producers (Carroll & Hannan, 2000). Moreover, these effects are weaker than those of the producer density, because (it was speculated) preproducers have lower social visibility than producers. Not all new entrants go through preproduction activities before entering the industry. Some new ventures start the production as soon as they collect resources (Carroll & Hannan, 2000). Are there any differences between de novo entrants with and without preproduction experience? Carroll et al. (1996) propose that preproduction activities may help de novo firms build capabilities and accumulate resources before entering into production. Consequently, de novo firms with preproduction experience should have lower morality rates initially than those without such an experience. Firms with preproduction experience, however, are likely to be more inertial. Therefore, as time passes and the environment changes, the mortality rates of de novo firms with preproduction experience may become higher than those of de novo organizations without such an experience. Carroll et al. (1996) find empirical support for these predictions, analyzing the American automobile industry. Research on preproducers is still in its infant stage. There is a need for a better theory explaining which firms are capable of making the transition from preproduction into production and what the consequences of this transition are. Toward this end, the formal model advanced by Jovanovic (2004) to explain waiting times between the preproducer and producer stages should prove very influential. 4.2. Spin-offs A central question of entrepreneurship research asks where the entrepreneurs come from, and how their origins affect venture performance. Recent research in organizational ecology makes contributions to the understanding of this issue by studying entrepreneurial activities initiated by industry veterans (i.e., spin-offs). This research explores a number of issues, such as what firms are likely to produce spin-offs, whether spin-offs differ in their performance from other ventures, and what effects spin-offs have on the performance of their parent companies. Brittain and Freeman (1986) initiated ecological research on spin-offs. They suggested that organizational differences predict the rate at which employees leave firms to start their own businesses. When analyzing the semiconductor industry in the Silicon Valley, they found that firms more likely generate entrepreneurs when they experience an external shock, such as outside succession of chief executive or the recent acquisition of the firm by a non-semiconductor firm. Firms with blocked career mobility, measured as firms with low growth rates, show a higher rate (though the effect was weak) of spin-offs as well. Employees working for generalist firms are more likely to start ventures, allegedly because they are exposed to the variety of knowledge and skills at their parent firms useful for starting ventures. More innovative firms also tend to generate more entrepreneurs. Somewhat surprisingly, effects of the environment, including organizational density, are weak and inconclusive. Current research continues to focus on the factors that influence the rate at which firms spawn spin-offs (see Klepper, 2001, for a review). Ecologists continue to make contributions to understanding the issue of employees’ ventures by focusing not only on the precedents but also on the consequences of spin-offs for both parental companies and their progenies. For example, Philips (2002) suggests that industry veterans transfer both routines and resources from their former companies to their ventures. Since transfer of routines and resources help new ventures to survive through the first difficult years, the greater this transfer, the higher survival rate of spin-offs. Parental companies, in turn, may experience negative consequences from the lost resources and, as a result, suffer worse performance and reduced survival chances when they generate spin-offs. Philips (2002) finds support for these predictions analyzing Silicon Valley law firms (2001). Another interesting finding of his study is that firms founded by people who left failing law firms show lower survival chances than non-spin-offs. 4.3. Foundings by Niche Position within a Population The success of new venture initiation and survival may depend on the specific resource niche in which the venture is founded. Many industries consist of multiple potential resource niches, and whereas some of these niches provide a fertile ground for new ventures, others can be difficult to enter and survive in. In our view, work on resourcepartitioning theory has motivated interesting theoretical and empirical developments about the founding rates of new ventures in different resource niches (Carroll, 1985; for a review, see Carroll, Swaminathan & Dobrev, 2003). As mentioned above, resource-partitioning theory proposes that (given specific environmental and organizational conditions) when market concentration goes up, the number of new venture foundings may paradoxically increase, whereas their mortality rates decrease. This process is driven by the partitioning of the resource space, so that different types of organizations operate successfully in different resource niches and do not compete directly with each other. Large generalist organizations strive to occupy the center of the market where they can exercise economies of scale. As competition intensifies, weaker generalists exit. Winning generalists get larger but paradoxically they leave some resource space on the industry periphery unused. Peripheral niches are not very appealing to generalists because of the diseconomies of scale but are attractive enough for small specialists. Since much of the peripheral space emerges as market concentration increases, greater numbers of smaller firms can successfully enter the growing peripheral niches. Therefore, industries that experience resource partitioning provide fertile ground for founding new ventures. By entering and filling peripheral niches, new ventures may expand these niches and eventually render them quite attractive in terms of economic returns. As a result, peripheral niches that were not appealing in the past may become attractive in the present to generalist organizations. For example, the brewing industry in the 1980s was dominated by a handful of large mass producers that made few varieties of generic beer. Then small brewers came into play by producing high-quality specialty beer. Although the market for the specialty beer was initially very small, it grew large and eventually attracted the attention of mass producers (Carroll & Swaminathan, 2000). However, despite later active attempts, generalists were not able to occupy these new niches. Resource-partitioning theory describes several mechanisms explaining why even when peripheral niches become attractive, generalist producers cannot enter them, whereas entrepreneurs have a substantial advantage over established firms both in terms of entry and survival. One mechanism is authenticity of an organizational form that a firm embodies. Carroll and Swaminathan (2000) propose that in order for a firm to be approved by external observers, it has to offer products or services that are authentic to its organizational form or identity. In the study of the U.S. brewing industry, Carroll and Swaminathan identify four different organizational forms: large mass producers, microbrewers, brewpubs and contract brewers. Microbrewers and brewpubs produce and sell high-quality specialty beer, usually on the production site. Their organizations resemble 19th century craft-like production firms and their consumers respond to an antimass production sentiment. Although specialist brewers began on the market periphery, over time they built an attractive, expanding market niche. When mass producers tried to move into this niche, they failed because customers perceived their product as inauthentic and refused to buy it. It is interesting that although contract brewers are also specialists they did not hold the same form identity as microbrewers and brewpubs, because they sell specialty beer they did not make themselves. The illusory identity of a specialist firm helps contract brewers succeed better than mass producers in the midst of the anti-mass production sentiment, but they did not perform as well as either microbrewers or brewpubs. Another mechanism that may prevent generalists from occupying profitable peripheral niches is service customization. For example, in the Dutch audit industry, large generalists can provide service to any firm. However, many small client firms prefer to deal with small specialized audit organizations, because the latter are more willing to cater to their specific (changing) needs (Boone, Carroll & van Witteloostuijn, 2002). Among Silicon Valley law firms, small specialist law firms are preferred by many clients to large generalists, because specialist firms provide personalized service whereas generalists firm are not willing to be so accommodating to small clients (Jaffee, 2002). New ventures may also have a higher founding rate in markets that are partitioned based on conspicuous status consumption. In industries where environmental uncertainty is low, small specialist organizations may achieve high status, which generalists cannot replicate by the nature of being big and unfocused. For example, conspicuous status consumption is clearly observed in the California wine industry where only small specialist wineries are perceived as “boutique”, “chateau”, and “farm wineries” (Swaminathan, 1995). These wineries are associated with production of a wide variety of low-volume, high-value-added products that are appreciated by customers who consider wine consumption as a sign of status. Such customers avoid buying similar wine from mass-production wineries, because mass producers do not posses high status. When environmental uncertainty is high, however, consumers are likely to attach high status to generalist firms, as occurred in investment banking (Park & Podolny, 2000). Generalists with high status are very unlikely to enter low status niches even if profitable, because this action can damage their prestige (Podolny, 1993). Thus, in markets partitioned along a status dimension, small firms do not directly compete with large ones. Therefore, new ventures can be successfully founded in niches with an appropriate status. Resource space can be partitioned by geography. Geographic areas not attractive to generalists may provide a fertile ground for founding new firms (Lomi, 1995a, 1995b). Moreover, geographic co-location near other organizations of either similar or complementary forms may prove attractive (Sorenson & Stuart, 2001; Stuart & Sorenson, 2003). Finally, the resource space can be partitioned along technological dimensions. Recent research suggests that new ventures are more likely to be founded and to survive in certain technological niches. Khessina (2003) proposes that de novo firms do better than de alio firms in niches at the technological frontier. Such differentiation happens because de novo firms possess greater flexibility in their internal routines and their relationship with the environment that allows them to be more innovative than de alio firms. She finds empirical support for this idea while analyzing product innovations in the worldwide optical disk drive industry. Interestingly, she did not find that de novo firms have a higher failure rate than de alio firms; she attributes this finding to the importance of innovation in this industry and the superiority of de novo firms at innovation. Khessina and Carroll (2002) pursue this idea further by looking at product demography to understand the technological niches that firms compete in. It turns out that new ventures generally compete at the high end of technology, whereas preexisting firms are evenly spread out. 4.4. Identity Space as a Resource Success of founding a new venture depends not only on physical resources but also on socially constructed resources. Specifically, venture formation may be affected by the ecology of organizational form identities. Industries usually host more than one organizational form. For instance, the brewing industry consists of mass producers, microbrewers, brewpubs, and contract brewers (Carroll & Swaminathan 2000). Organizational forms can be distinguished by the identities attached to them. Identity is a system of descriptive and prescriptive codes that connote cognitive recognition and imperative standing (Pólos, Hannan & Carroll, 2002). Identity delineates the magnitude of actions that a firm may undertake without violating expectations of its audiences. For example, consumers expect specialty beer to be brewed by traditional hand crafted methods. Therefore, it is not surprising that consumers perceive specialty beer as authentic only if it is made and sold by microbrewers and brewpubs but not by mass producers and contract brewers. On the other hand, consumers who prefer generic beer are likely to buy such from mass producers. The example of the brewing industry shows that identity is an important resource and the industry is usually partitioned into different identity spaces. Entrepreneurs are more likely to succeed in their attempts to found firms when they choose an appropriate identity for their ventures. Ruef (2000) conducted a broader study examining how new organizational forms emerge in the health care sector of the U.S. Between 1965 and 1994, he counts 19 new forms emerging, complementing a pre-existing set of 29 forms. Ruef's (2000) analysis focuses on the emergence of these 19 new forms as social identities. In particular, he develops testable arguments about form emergence, drawing from institutional (regulatory acts and government funding) and ecological (density dependence, competition, symbiosis, and punctuated equilibrium) theories of organization as well as social movement analysis. Analyzing texts drawn from over one thousand journals and other publications, targeted to disparate kinds of participants in the health care sector, Ruef constructed an identity space within which forms might emerge. The dependent variable for the empirical analysis consisted of the counts of the forms appearing in each cell of the identity space. The study finds that new form emergence depends on the positions of existing form identities in the sector, and by the density and size of organizations matching those identities. The density and size of organizations possessing a form increases the rate of new form emergence up to a point, at which point the rate declines. The curvilinear relationship is interpreted as reflecting legitimation initially (in the rising phase) and the competition due to saturation (in the downward phase). Identity of an organizational form does not need to be based only on product characteristics. Baron (2004) proposes that identity can also be defined by the labor relationships prevalent in a given market. Labor market identity indicates what kind of employment relationships are considered acceptable. Baron contends that although each organizational form embodies both product-based identity and labor-based identity, one or another can be more salient to the audience. Labor market identity may dominate when employment relationships and human or social capital are of paramount importance. Therefore, entrepreneurs who think about founding a firm might focus simultaneously on both market-based and product-based identities. Issues of organizational form and identity are of paramount importance in emerging industries. New industries lack legitimation. One reason for the lack of legitimation is an undeveloped organizational form on which the industry activity is based. When the organizational form is underdeveloped, it does not generate a legitimate signal and, as a result, the industry fails to attract resources and institutional approval. Organizational ecologists propose that an organizational form is more likely to develop and acquire a legitimate status with increasing density of organizations that embody this form. Furthermore, the process of organizational form development is facilitated and accelerated when firms that enter a new industry have focused identities (McKendrick & Carroll, 2002). New ventures play an especially important role in this process. In general, de novo firms should have more focused identities than de alio firms. This is because de novo firms are typically founded with a goal of operation in the focal industry and, therefore, their activity is focused on one type of business. In contrast, de alio firms likely participate in multiple industries (or did so previously, creating a legacy) and project multiple identities associated with these industries. While de novo firms would be quite visible to an audience in the focal industry, de alio firms in the focal industry may not be perceived so readily at all (especially if de alio firms derive the bulk of their revenues from other industries). As a result, a simple increase in the numbers of de alio firms may not contribute to increasing legitimation and institutionalization of a new organizational form. In an analysis of founding and survival rates of disk array producers, McKendrick, Jaffee, Carroll and Khessina (2003) find empirical support for these arguments. They find that density of de novo firms (but not that of de alio firms) has a legitimating effect on the industry by increasing founding rates of all firms and decreasing failure rates of all firms. They also find that geographically clustered firms, especially geographically clustered de novo firms, have a beneficial effect on industry dynamics. This too is interpreted as a consequence of perceived focus in organizational identities. 4.5. Social Movement-Like Behavior Organizational ecologists sometimes explore how entrepreneurs in certain industries undertake collective action to reach and persuade audiences of their worth. Founders of new ventures undertake different collective action strategies to promote the awareness of their activities. They establish industry associations (McKendrick et al., 2003), they organize advertising campaigns in the form of competitions, races (Rao, 1994; Hannan, Carroll, Dundon & Torres, 1995), conferences and rallies (Carroll & Swaminathan, 2000) and they try to reach and educate the market in many other ways (Swaminathan & Wade, 2001). Such collectively oriented behavior of entrepreneurs often resembles that of social movements (Carroll, 1997; Swaminathan & Wade, 2001). The analogy is potentially important for theory development because social movements and their leaders often defy (or least suggest arguments alternatives to) rational action interpretations (Tilly, 1978). This means that the collective action aspects of entrepreneurship may provide some clues as to how to decouple the two component rates of the founding process discussed above without assuming rationality. Although the tendency of founders to organize collectively in new industries (or in new market segments) is well documented, no compelling theoretical model has yet been advanced to analyze the patterns of these activities in this way. The density dependence model has been applied to analyses of vital rates of social movement organizations (Olzak, 1989; Olzak & West, 1991; West, 1995; Minkoff, 1997). However, this model does not really predict what types of collective action are undertaken by organizations or entrepreneurs in different industries and why. A more explicit model of collective action by entrepreneurs has yet to be proposed despite its allure and apparent promise. Table 2. Key references among ecological studies of foundings Theoretical contributions Methodological contributions Hannan (1984). Advances theory of densitydependent legitimation and competition. Predicts foundings will show an inverted Ushape relation with organizational density. Hannan and Freeman (1987). Proposes modeling foundings as arrivals in a stochastic process. Parameterizes covariates' effects on rate of arrival using Poisson regression. Carroll (1985). Advances theory of resource partitioning. Predicts foundings of specialist organizations will increase with market concentration. Barron (1992). Identifies estimation problem created by overdispersion in count data on foundings. Proposes negative binomial model (possibly with autocorrelation) as solution. Tucker et al. (1990). Accounts for foundings of voluntary social service organizations in terms of governmental policy changes. Lomi (1995b). Models unobserved heterogeneity in spatially defined founding processes. Baum and Singh (1994). Founding rates explained as a consequence of differentiation using concepts of niche overlap. Ruef (2000). Provides methodology for inferring organizational form identities from textual sources. Hannan et al. (1995). Extends model of densitydependent legitimation and competition to allow global competition and local competition. Ruef (2002). Advances structural event analysis to model demographic composition of entrepreneurial teams. Baum and Haveman (1997). Highlights the tension between processes of agglomeration and differentiation in driving foundings. McKendrick et al. (2003). Examines a possibly emergent organizational form by choosing a less than institutionalized case to study. Dobbin and Dowd (1997). Explains railroad foundings as consequence of public policy changes. Wade et al. (1998). Analyzes unintended effects of state-level prohibitions in brewery foundings of neighboring states. Sorenson and Audia (2000). Proposes social network explanation for regional concentrations of foundings. Ruef (2000). Conceives of foundings within identity space of organizational forms in a community. Swaminathan and Wade (2001). Draws inferences about foundings using social movement theory and research. 5. CONCLUSION Entrepreneurship and ecology are both huge research areas with very large literatures of theory and research. Given a common strong interest in organizational foundings and the success of newly founded organizations, it is striking how unconnected the two literatures are at present. While possibly understandable, this disjuncture seems to us to be counterproductive. Accordingly, our review here has tried to provide motivation as well as accessible conceptual linkages to ecology for those in the entrepreneurship area. Along the way, we have offered comments on the ecological ways of doing theory and research that might be of interest to those from that tradition. 6. NOTES 1 We appreciate the helpful comments of Olav Sorenson, Bill Barnett and Mike Hannan on an earlier draft. Other common definitions associate entrepreneurship with innovativeness, with passionate motivation or with organizational change. The perspectives behind these definitions are beyond the scope of this chapter, although other chapters in this book do consider them at length. 3 We used our extensive knowledge of the organizational ecology literature to identify the articles and books used in making this calculation. 4 We reject the notion that these analysts might not be aware of organizational ecology as generally implausible. The journals for organizational theory and entrepreneurship do overlap. 5 Compared to the “entry decision” of economics, the ecological appellation of “foundings” commonly used in studies of organizational births or entries into markets suggests less volition and more detachment. 6 In the physical sciences, the time lag for basic research to affect practice has been estimated at 75 years. 7 If this proposal seems unrealistic, then consider how long industrial economics existed as a discipline before its translation by Porter (1980) created a foundation for the applied field of strategic management. 8 It is common these days to regard the micro level processes as more fundamental than the macro level. Hannan (1992) provides a compelling counter argument, however. Citing the case of Darwin and genetics, he argues that a robust macro theory is important. 9 Note that the challenge of creating a widely agreed upon conceptual definition of the activities defining this event should not be seen as unusual or especially debilitating to theory or research. Indeed, a similar lack of consensus about what exactly constitutes an organizational founding can be found in the earlier ecological literature, and remains a slightly unsettled issue today. As long as researchers define and clearly operationalize their concepts, research can continue to progress and be compared. 10 Specifically, Ruef et al. (2003) asked two questions to randomly selected interviewees: (1) Are you, alone or with others, now trying to start a business? (2) Are you, alone or with others, now starting a new business or new venture with your employer? If respondents answered affirmatively to either question, they were asked two additional questions to ascertain that (1) they were actively involved and (2) they held an ownership stake. Affirmative answers to both of these later questions were required to be considered a nascent entrepreneur. 2 11 Of course, the assumptions of strong individual rational action and historical efficiency are not held by all entrepreneurship researchers. Many subscribe to a model of bounded rationality where action is intendedly rational but blurred by uncertainty, ambiguity and lack of information. We discuss these models below as we develop our argument. 12 We found this somewhat surprising because as Hannan and Carroll (1992: 199) note, “Images of rationality in organization building seem to focus on…the causes of attempts to start organizations.” Many ecologists would not necessarily view themselves as rational action theorists. 13 We doubt that this task will be taken on with eagerness by ecologists themselves. Indeed, Hannan and Carroll (1992; 194-8) argue against the fashionable practice of motivating macro theories by building up from micro assumptions and models, contending that this increases the fragility of a macro theory. They propose instead that macro theories be designed for robustness with respect to micro assumptions, and point to Darwin’s theory of natural selection as an example. 14 The consequences of different models of entrepreneurship on organizational populations are also of great potential interest. See, for instance, Ruef's (2004) analysis of how entrepreneurial inertia might generate population cycles of boom and bust. 15 De alio firms also obtained higher market shares in the American television receiver industry (Klepper and Simons 2000). 16 Specifically, Khessina (2003) found that de novo firms introduced products with performance parameters that improved over those of these firms’ previous best products at a significantly higher rate than de alio firms. Additionally, de novo firms introduced products with performance parameters that were among the top of the industry (i.e., with performance parameters close to the technological frontier) at a significantly higher rate than de alio firms. Hence, de novo firms outperformed de alio firms in product innovation. 17 We say intendedly rational because location in the non-overlapped region of a particular dimensionalization of the niche may or may not lead to a greater chance of success once the enterprise is operating. That question would need to be demonstrated in a separate empirical study of organizational viability or mortality. 18 The empirical analysis uses a novel conceptualization strategy the investigators call “structural event analysis” pioneered by Ruef (2002). In a nutshell, this technique uses combinatorial methods to generate all possible combinations of potential group members (based on the social characteristics of theoretical interest) and to specify the expectations of their distributions under different assumptions about sorting implied by the hypotheses; it then uses Poisson regressions to account for deviations between the sample data and these expected distributions. 7. REFERENCES Aldrich, H.E. and Wiedenmayer, G. (1993). From traits to rates: An ecological perspective on organizational foundings. In J. Katz and R., Advances in Entrepreneurship, Firm Emergence, and Growth, (pp.145-195). Greenwich, CT: JAI Press. Banbury, C.M., & Mitchell, W. (1995). The effect of introducing important incremental innovations on market share and business survival. Strategic Management Journal, 16, S161-S182. Barnett, W.P., & Freeman, J. (2001). Too much of a good thing? Product proliferation and organizational failure. Organization Science, 12, 539-558. Barnett W.P., & Sorenson, O. (2002). The Red Queen in organizational creation and development. Industrial and Corporate Change, 11, 289-325. Barnett, W.P., Swanson, A.N., & Sorenson, O. (2003). Asymmetric selection among organizations. Industrial and Corporate Change, 12, 673-695. Baron, J.N. (2004). Employing identities in organizational ecology. Industrial and Corporate Change,13,3-32. Barron, D.N. (1992). The analysis of count data: Overdispersion and autocorrelation. In P.V. Marsden, Sociological Methodology 1992 (pp. 179-220). Oxford. Basil Blackwell. Barron, D.N., West, E., & Hannan, M.T. (1994). A time to grow and a time to die: Growth and mortality of credit unions in New York City, 1914-1990. American Journal of Sociology, 100, 381-421. Baum J.A.C., & Haveman, H.A. (1996). Love thy neighbor? Differentiation and agglomeration in the Manhattan hotel industry, 1898-1990. Administrative Science Quarterly, 42, 304-338. Baum, J.A.C., Korn, H.J., & Kotha, S. (1995). Dominant designs and population dynamics in telecommunications services: Founding and failure of facsimile transmission service organizations, 19651992. Social Science Research, 24, 97-135. Baum, J.A.C., & Oliver, C. (1996). Towards an institutional ecology of organizational founding. Academy of Management Journal, 39, 1378-1427. Baum, J.A.C., & Singh, J.V. (1994). “Organizational niches and the dynamics of organizational mortality.” American Journal of Sociology, 100, 346-380. Boone, C., Brocheler, V., & Carroll, G.R. (2000). Custom service: Application and tests of resource-partitioning theory among Dutch auditing firms for 1896 to 1992. Organization Studies, 21, 355-381. Boone, C., Carroll, G.R., & van Witteloostuijn, A. (2002). “Resource distributions and market partitioning: Dutch daily newspapers, 1968 to 1994. American Sociological Review, 67, 408-31. Brittain, J.W., & Freeman, J. (1986). Entrepreneurship in the semiconductor industry. Unpublished manuscript, University of California, Berkeley, CA. Bruderl, J., & Schussler, R. (1990). Organizational mortality: The liabilities of newness and adolescence. Administrative Science Quarterly, 35, 530-547. Bruderl, J., Preisendorfer, P., & Ziegler, R. (1992). Survival chances of newly founded business organizations. American Sociological Review, 57, 227-242. Carroll, G.R. (1985). Concentration and specialization: Dynamics of niche width in populations of organizations. American Journal of Sociology, 90, 1262-1283. Carroll, G.R. (1997). Long-term evolutionary change in organizational populations. Industrial and Corporate Change, 6,119-43. Carroll, G.R., Bigelow, L.S., Seidel, M.D., & Tsai, L.B. (1996). The fates of de novo and de alio producers in the American automobile industry 1885-1981. Strategic Management Journal, 17,117-137. Carroll, G.R., & Delacroix, J. (1982). Organizational mortality in the newspaper industries of Argentina and Ireland: An ecological approach. Administrative Science Quarterly, 27,169-198. Carroll, G.R., Delacroix, J., & Goodstein, J. (1988). The political environments of organizations: An ecological view. Research in Organizational Behavior, 10, 359-392. Carroll, G.R., & Hannan, M.T. (2000). The demography of corporations and industries. Princeton, NJ: Princeton University Press. Carroll, G.R., & Huo, Y.P. (1986). Organizational task and institutional environments in ecological perspective: Findings from the local newspaper industry. American Journal of Sociology, 91, 838-873. Carroll, G.R., & Khessina, O.M. (2004). Organizational and corporate demography. In D.L. Poston, Jr & M. Micklin (Eds.), Handbook of Population (forthcoming). Plenum/Kluwer. Carroll, G.R., & Swaminathan, A. (2000). Why the microbrewery movement? Organizational dynamics of resource partitioning in the U.S. brewing industry. American Journal of Sociology, 106, 715-762. Carroll, G.R., Dobrev, S., & Swaminathan, A. (2002). Organizational processes of resource partitioning. Research in Organizational Behavior, 24, 1-40. Delacroix, J., & Carroll, G.R. (1983). Organizational foundings: An ecological study of the newspaper industries of Argentina and Ireland. Administrative Science Quarterly, 28, 274-291. Delacroix, J., Swaminathan, A., & Solt, M.E. (1989). Density dependence versus population dynamics: An ecological study of the failing in the California wine industry. American Sociological Review, 54, 245-262. Dobbin, F., & Dowd, T. (1997). How policy shapes competition: Early railroad foundings in Massachusetts. Administrative Science Quarterly, 42, 501-529. Dobrev, S.D. (2001). Revisiting organizational legitimation: Cognitive diffusion and sociopolitical factors in the evolution of Bulgarian newspaper enterprises, 1846-1992. Organization Studies, 22, 419-444. Dobrev, S.D., Kim, T.Y., & Hannan, M.T. (2001). Dynamics of niche width and resource partitioning. American Journal of Sociology, 106, 1299-1337. Delacroix, J., & Swaminathan, A. (1991). Cosmetic, speculative, and adaptive organizational change in the wine industry: A longitudinal study. Administrative Science Quarterly, 36, 631-661. Fichman, M., & Levinthal, D.A. (1991). Honeymoons and the liability of adolescence: A new perspective on duration dependence in social and organizational relationships. Academy of Management Review, 16, 44268. Freeman, J. (1990). Ecological analysis of semiconductor firm mortality. In J.V. Singh (Ed.), Organizational Evolution: New Directions (pp. 53-77). Newbury Park, CA: Sage. Freeman, J., Carroll, G.R., & Hannan, M.T. (1983). The liability of newness: Age dependence in organizational death rates. American Sociological Review, 48, 692-710. Halliday, T.C., Powell, M.J., & Granfors, M.W. (1987). Minimalist organizations: Vital events in state bar associations, 1870-1930. American Sociological Review, 52, 456-471. Hannan, M.T. (1986). Competitive and institutional processes in organizational ecology. Technical Report 8613, Department of Sociology, Cornell University, Ithaca, NY. Hannan, M.T. (1986). Organizational population dynamics and social change. European Sociological Review 4, 95-109. Hannan, Michael T. (1992). Rationality and Robustness in Multilevel Systems. In J.S. Coleman (Ed.), Rational Action in Sociology, (pp. 120-135). Newbury Park: Sage. Hannan, M.T. (1997). Inertia, density and the structure of organizational populations: Entries in European automobile industries, 1886-1981. Organization Studies, 18, 193-228. Hannan, M.T. (1998). Rethinking age dependence in organizational mortality: Logical formalizations. American Journal of Sociology, 104, 85-123. Hannan, M.T., & Carroll, G.R. (1992). Dynamics of organizational populations: Density, legitimation, and competition. New York: Oxford University Press. Hannan, M.T., Carroll, G.R., Dobrev, S.D., & Han, J. (1998). Organizational mortality in European and American automobile industries part I: Revisiting the effects of age and size.European Sociological Review, 14, 279-302. Hannan, M.T., Carroll, G.R., Dundon, E.A., & Torres, J.C. (1995). Organizational evolution in a multinational context: Entries of automobile manufacturers in Belgium, Britain, France, Germany, and Italy. American Sociological Review, 60, 509-528. Hannan, M.T., & Freeman, J. (1987). The ecology of organizational founding: American labor unions, 18361985. American Journal of Sociology, 92, 910-943. Hannan, M.T., & Freeman, J. (1988). The ecology of organizational mortality: American labor unions, 18361985. American Journal of Sociology, 94, 25-52. Hannan, M.T., & Freeman, J. (1989). Organizational ecology. Cambridge, MA: Harvard University Press. Hargens, L.L. (2000). Using the literature: Reference networks, reference contexts and the social structure of scholarship. American Sociological Review, 65, 846-65. Haveman, H.A., & Cohen, L.E. (1994). The ecological dynamics of careers: The impact of organizational founding, dissolution, and merger on job mobility. American Journal of Sociology, 100, 104-152. Ingram, P., & Inman, C. (1996). Institutions, intergroup competition, and the evolution of hotel populations around Niagara Falls. Administrative Science Quarterly, 41, 629 -658. Ingram, P., & Simons, T. (2000). State formation, ideological competition, and the ecology of Israeli workers cooperatives, 1920-1992. Administrative Science Quarterly, 45, 25-53. Jaffee, J. (2000). The resource partitioning of a corporate legal market: The proliferation of specialist law firms in Silicon Valley, 1966-1997. Paper presented at American Sociological Association Meetings, Washington, DC. Jovanovic, B. (2004). The pre-producers. Paper presented at the Conference on Strategy and Ecology, Washington University, Olin School of Business, St. Louis, MO. Khessina, O.M. (2003). Entry mode, technological innovation and firm survival in the worldwide optical disk drive industry, 1983-1999. Unpublished Ph.D. Thesis, University of California, Berkeley, CA. Khessina, O.M., & Carroll, G.R. (2002). Product dynamics of de novo and de alio firms in the worldwide optical disk drive industry, 1983-1999. Paper presented at the Academy of Management Meetings, Denver, CO. Klepper, S. (2001). Employee startups in high-tech industries. Industrial and Corporate Change, 10, 639-674. Klepper, S., & Simons, K.L. (2000). Dominance by birthright: Entry of prior radio producers and competitive ramifications in the U.S. television receiver industry. Strategic Management Journal, 21, 997-1016. Land, K.C., Davis, W.R., & Blau, J.R. (1994). Organizing the boys of summer: The evolution of U.S. minorleague baseball, 1883-1990. American Journal of Sociology, 100, 781-813. Lomi, A. (1995a). The population and community ecology of organizational founding: Italian cooperative bans, 1936-1989. European Sociological Review, 11, 75-98. Lomi, A. (1995b). The population and community ecology of organizational founding: Location dependence and unobserved heterogeneity. Administrative Science Quarterly 40: 111-145. March, J.G. (1978). Bounded Rationality, Ambiguity, and the Engineering of Choice. Bell Journal of Economics, 9, 587-608. March, J.G. and J.P. Olsen. (1989). Rediscovering institutions: The organizational basis of politics. New York: Free Press. Marrett, C.B. (1980). Influences on the rise of new organizations: The formation of women's medical societies. Administrative Science Quarterly, 25, 185-199. McKendrick, D.G., & Carroll, G.R. (2001). On the genesis of organizational forms: Evidence from the market for disk drive arrays. Organization Science, 12, 661-82. McKendrick, D.G., Jaffee, J., Carroll, G.R., & Khessina, O.M. (2003). In the bud? Disk array producers as a (possibly) emergent organizational form. Administrative Science Quarterly, 48, 60-93 Minkoff, D.C. (1997). The sequencing of social movements. American Sociological Review, 62, 779-799. Mitchell, W. (1994). The dynamics of evolving markets: The effects of business sales and age on dissolutions and divestitures. Administrative Science Quarterly, 39, 575-602. Olzak, S. (1989). Analysis of events in studies of collective action. Annual Review of Sociology, 15, 119-141. Olzak, S., & West, E. (1991). Ethnic conflicts and the rise and fall of ethnic newspapers. American Sociological Review, 56, 458-474. Olzak, S., & Uhrig, N. (2001). The ecology of tactical overlap among new social movements in West Germany. American Sociological Review, 66, 694-718. Park, D.Y., & Podolny, J.M. (2000). The competitive dynamics of status and niche width: U.S. investment banking, 1920-1950. Industrial and Corporate Change, 9, 377-414. Pfeffer, J. (1993). Barriers to the advance of organizational science: Paradigm development as a dependent variable. Academy of Management Review, 18, 599-620. Philips, D.J. (2002). A genealogical approach to organizational life chances: The parent-progeny transfer among Silicon Valley law firms. 1946-1996. Administrative Science Quarterly, 47, 474-506. Podolny, J.M. (1993). A status-based model of market competition. American Journal of Sociology, 98, 829-72. Pólos, L., Hannan, M.T., & Carroll, G.R. (2002). Foundations of a theory of social forms. Industrial and Corporate Change, 11, 85-115. Porter, M.E. (1980). Competitive strategy: Techniques for analyzing industries and competitors. New York: Free Press. Rao, H. (1994). The social construction of reputation: Certification contests, legitimation, and the survival of organizations in the American automobile industry: 1895-1912. Strategic Management Journal, 15, 29-44. Ruef, M. (2000). The emergence of organizational forms: A community ecology approach. American Journal of Sociology, 106, 658-714. Ruef, M. (2002). A structural event approach to the analysis of group composition. Social Networks, 24, 135160. Ruef, M. (2004). Boom and bust: The effect of entrepreneurial inertia on organizational populations. Unpublished manuscript, Graduate School of Business, Stanford University, Palo Alto, CA. Ruef, M., Aldrich, H.E., & Carter, N.M. (2003). The structure of founding teams: Homophily, strong ties, and isolation among U.S. entrepreneurs. American Sociological Review, 68, 195-222. Schoonhoven, C.B., Eisenhardt, K.M., & Lyman, K. (1990). Speeding products to market: Waiting time to first product introduction in new firm. Administrative Science Quarterly, 35, 177-207. Simons, T., & Ingram, P. (2002). Enemies of the state: Competition over the supply of order and the ecology of the kibbutz, 1910-1997. Paper presented at the 25th Anniversary of Organizational Ecology Conference, Stanford University, Palo Alto, CA. Simons, T., & Ingram, P. (2004). An ecology of identity: Theory and evidence from four populations.Industrial and Corporate Change,13, 33-60. Singh, J.V., Tucker, D.J., & House, R.J. (1986). Organizational legitimacy and the liability of newness. Administrative Science Quarterly, 31, 171-193. Sørensen, J.B., & Sorenson, O. (2003). From conception to birth: Opportunity perception and resource mobilization in entrepreneurship. Advances in Strategic Management, 20, 89-117. Sorenson, O., & Audia, P.G. (2000). The social structure of entrepreneurial activity: Geographic concentration of footwear production in the U.S., 1940-1989. American Journal of Sociology, 106, 424-62. Sorenson, O., & Stuart, T.E. (2001). Syndication networks and the spatial distribution of venture capital investment. American Journal of Sociology, 106, 1546-1588 Stinchcombe, A.L. (1965). Social structure and organizations. In J.G. March (Ed.), Handbook of Organizations. Chicago: Rand McNally. Stuart, T.E., & Sorenson, O. (2003). The geography of opportunity: Spatial heterogeneity in founding rates and the performance of biotechnology firms. Research Policy, 32, 229-253. Studer-Ellis, E.M. (1995). Springboards to mortarboards: Women's college foundings in Massachusetts, New York, and Pennsylvania. Social Forces, 73, 1051-1070. Swaminathan, A. (1995). The proliferation of specialist organizations in the American wine industry, 19411990. Administrative Science Quarterly, 40, 653-680. Swaminathan, A. (1998). Entry into new market segments in mature industries: Endogenous and exogenous segmentation in the U.S. brewing industry. Strategic Management Journal, 19, 389-404. Swaninathan, A., & Delacroix, J. (1991). Differentiation within an organizational population: Additional evidence from the wine industry. Academy of Management Journal, 34, 679-692. Swaminathan, A., & Wade, J.B. (2001). Social movement theory and the evolution of new organizational forms. In C.B. Schoonhoven & E. Romanelli (Eds.), The Entrepreneurship Dynamic: Origins of Entrepreneurship and the Evolution of Industries (pp. 286-313). Palo Alto, CA: Stanford University Press. Swanson, A.N. (2002). Form coherence and the fates of de alio and de novo organizations in the U.S. digital computer industry: 1951-1994. Unpublished Ph.D. dissertation, Stanford University, Palo Alto, CA. Tucker, D.J., Singh, J.V., & Meinhard, A.G. (1990). Organizational form, population dynamics, and institutional change: The founding patterns of voluntary organizations. Academy of Management Journal, 33, 151-178. Tilly, C. (1978). From mobilization to revolution. Reading, MA: Addison-Wesley. Venkataraman, S. (1997). The distinctive domain of entrepreneurship research. Advances in Entrepreneurship, Firm Emergence and Growth, 3, 119-138. Wade, J.B., Swaminathan, A., & Saxon, M.S. (1998). Normative and resource flow consequences of local regulations in American brewing industry, 1845-1918. Administrative Science Quarterly, 43, 905-935. West, E. (1995). Organization building in the wake of ethnic conflict: A comparison of three ethnic groups. Social Forces, 4,1333-1363.