Kauffman Center for Entrepreneurial Leadership Award for the Best Paper with the Most Significant Implications for Practitioners

WHAT FIRM FOUNDERS DO: A LONGITUDINAL STUDY OF THE START-UP PROCESS

Frédéric Delmar, Center for Entrepreneurship and Business Creation, Stockholm School of Economics
Scott Shane, Robert H. Smith School of Business, University of Maryland

CHAPTER MENU

ABSTRACT

What founders do to organize new ventures and how it affects the outcome of the organizing process is of considerable interest to entrepreneurship researchers. In this study, the life histories of 223 Swedish new ventures started between January and September 1998 by a random sample of firm founders is tracked. We explore the effect of planning and organizing activities on the hazard of (voluntary and involuntary) disbanding and of establishing the venture on the market during their first 30 months of life. We find that activities related to planning and to establishing legitimacy reduce the hazard for disbandment and increases the hazard for establishing the new venture.

INTRODUCTION

Entrepreneurs do not instantaneously organize new firms to exploit the opportunities that they have identified (Freeman, 1982). The organization of new firms cannot be perceived as the outcome of a single decision or a single act. Rather, entrepreneurs organize new firms through a series of actions— obtaining inputs, conducting product development, hiring employees, seeking funds, and gathering information from customers—undertaken to different degrees, in different order, and at different points in time (Gartner, 1985). Moreover, more than the activities initiated by the firm founder(s), the outcome of this process can be seen as dependent on three confounding factors:

The outcome of the organizing process is dependent on these three factors, and research needs to separate their relative contribution from each other. The successful outcome of the organizing process is due to variations in the quality of the of opportunities identified by entrepreneurs, as well as in variation in individual difference among entrepreneurs exploiting the opportunity, and in the way the exploitation of the opportunity is organized. Activities are especially important early in the life of a new venture, as it is the way firm founders signal to the market the entry of a new competitor, and to the stakeholder the possible performance of the new venture as there is no historical financial performance to rely on.

The purpose of this paper is to examine how different organizing activities affect the outcome of the organizing process of newly established ventures, and at the same time try to provide appropriate controls for differences in the quality of the opportunity and the founding team. More specifically, we argue here that the organizing process is dependent on what kinds of activities are initiated. Basically entrepreneurs can engage in two different types of activities that will more or less influence on the short term and long term outcome of the process. They can either engage in either planning activities or activities related to the organizing of the new venture (operational activities). Operational activities can in their turn be divided in legitimacy building activities, resources transformation activities, or market related activities.

The paper explores the separate and combined influences of planning activities (creating a business plan, financial projections, information about customers and competitors) with operational activities (establishing legitimacy, resources transformation, and initiate market relations) on the survival likelihood of 223 new ventures initiated by a random sample of Swedish firm founders during the first nine months of 1998. Because no general theory exists to specify the nature of planning activities and organizing activities on the development of new ventures, this research is exploratory. Different outcomes of the process are examined. The paper deals with (a) antecedents of venture disbandment, that is the founding team decision to abandon the start-up effort, (b) the different reasons leading to that decision, and (c) the outcome of becoming “up and running.” The paper develops a set of hypotheses to establish a framework for examining results and for designing future investigations. We found that especially activities related to planning and the generation of legitimacy had an important effect on the process outcome. We also found significant differences between different reasons to abandon the process.

The empirical effort is important for two reasons. First, an insight into the formation of new ventures is provided. It is an important, but a little understood aspect of the life of an organization (Aldrich, 1999; Gartner, 1985; Katz & Gartner, 1988). Second, a void in the existing literature is filled by offering a theoretically grounded explanation of the role of different activities in the organizing process. By developing an explanation for why different activities are more important than others, and testing this argument empirically, we fill an important gap in the literature.

The paper proceeds as follows: the next section develops the theoretical framework of the paper and the hypotheses. The third section describes the method employed. The fourth section presents the results. The fifth section provides a discussion.

THEORETICAL DEVELOPMENT AND HYPOTHESES

Entrepreneurs organize new firms through a series of actions and they are undertaken to different degrees, in different order, and at different points in time (Gartner, 1985). Hence, entrepreneurs will engage in different patterns of activities with the purpose of establishing a new organization. This leads to two follow-up conclusions. First, by engaging in different patterns of activities, firm founders will create variation in the firm formation process. Some of these activity patterns will be more successful than others. Second, depending on the entrepreneurial opportunity and the attributes of the firm founders not all activities need to be initiated or will be initiated. In order to understand how this variation occurs, we have to first focus on the characteristics defining the evolution of the process, and second on the purpose of the different activities in which firm founders can engage in.

There are four different characteristics that dictate how the patterns of activities evolve. First, not all activities are necessary for the founding team to perform. For example, if the founding team already controls resources, there is no reason for them to seek external financial support. Second, due to the firm founders’ limited cognitive capacity, they lack the ability to undertake all organizing activities simultaneously. This means that firm founders will vary in what kind of activities they will first engage in. Third, the ability to undertake some activities is dependent on or will be enhanced by the completion of other activities. For example, the completion of a business plan leads the founding team to more easily search for external funding than otherwise. Also, the effectiveness of a marketing and promotion effort is probably enhanced if the product or service development is completed. Fourth, some activities are more important early in the history of new venture, others are more important later in the life history of the new organization. For example, early in the life of a venture, the inability to get control over needed resources may increase the probability to abandon the new venture. Whereas later, competition from established firms may only affect the outcome of process, once the new venture has created a product or service in demand by customers. Hence, these four characteristics of the process will generate variations in the firm formation process.

Moreover, activities have different purposes in the organizing process. Basically two different types of activities can be identified in the start-up process:

These organizing activities can be broken down in activities related to resources transformation (acquiring and combining the physical, human, and financial resources into a marketable product or service), activities related to market behavior (marketing and promotion, establishing a relationship with customers), or activities to generate legitimacy for the new venture.

Activities related to the establishment of a new venture have different purposes and therefore of different importance for the outcome of the organizing process depending on when they were initiated and in relation to which other activities that have already been initiated. Unfortunately the literature does not guide us in which activities are relatively more important than others early in the process of organizing a new venture. On the contrary some controversy exists about not only when to initiate certain activities, but also whether some activities should be initiated altogether.

An example of the current controversy is planning which some researchers see as a futile activity distracting firm founders from more important activities directly related to organizing (Bhidé, 1999; Carter, Gartner, & Reynolds, 1996), whereas other sees it as an important part of successful organizing process (Castrogiovanni, 1996).

In this paper, we adhere to the latter opinion and argue that planning is important for two reasons. First, it is a way for the firm founders to more effectively allocate different resources in the organizing process. Hence, planning is more effective early the venture life, as it leads to more effective organizing through deciding in which order and how other activities should be initiated. Second, planning and formal plans as business plans help the founding team to signal to external stakeholders the new venture’s possible development. This is especially important early in the venture life, as there is no financial performance to base an assessment on (Lounsbury & Glynn, 2001). Hence, planning also creates cognitive legitimacy for the new venture. This leads us to the first hypothesis:

Hypothesis 1: The earlier a new venture initiates planning activities the higher the probability for the firm to survive and to get established.

The argument that organizing activities are relatively more important has to be broken down in separate statement related to the role of different organizing activities. Depending on the theoretical approach, different types of organizing activities are proposed to be more effective to the early survival of the new venture.

An institutional approach on new venture development argues that survival is best achieved by initiating activities related to the establishment of the new venture’s legitimacy. If a new venture is perceived as legitimate, i.e., is accepted by the external environment, it will survive (Carroll & Hannan, 2000; Meyer & Rowan, 1977; Suchman, 1995). An important indicator of sociopolitical legitimacy in the Swedish system (as well as many European countries) is to have the venture legally registered. Registration has to be made early in the process for a practical reason. It takes time from the moment applications have been sent in (at least in the Swedish system) until the new venture has been registered legally. As long as the firm has not been legally registered, it is severely handicapped in relation to potential stakeholders as no financial transactions or hiring of employees or selling can be done legally. More to the point, the new venture cannot proceed with other activities where financial transactions are required as long as the venture is not legally registered. This leads us to the second hypothesis:

Hypothesis 2: The earlier a new venture initiates activities to establish legitimacy the higher the probability for the firm to survive and to get established.

An approach based on the development of social ties will argue that the successful survival of new venture depends on its ability to establish social ties to crucial stakeholders, especially customers (Aldrich & Zimmer, 1986; Hansen, 1995; Johannisson, 1988; Larson & Starr, 1993). Customers are important because they represent the future source of income for the venture. The faster a new venture establishes market relations, the faster we expect the new venture to be able to generate the first sales, and in the future hopefully a positive cash-flow. Hence, a rapid establishment of market relations leads to a higher probability of generating incomes. Moreover, rapidly establishing market relations is important in order to get feedback from potential customers about the attractiveness of the offered product or service. This leads to the third hypothesis:

Hypothesis 3: The earlier a new venture initiates activities to establish market relations the higher the probability for the firm to survive and get established.

A resource-based view of new venture development sees the acquisition and management of internal resources as central to the survival of the new firm (Barney, 1991; Galunic & Rodan, 1998; Grant, 1991; Penrose, 1995). New ventures compete with established ventures and survive by offering new combinations of resources that are superior to that of established organizations. In order to successfully do that, new ventures must gain control over the needed resources, and develop routines that enables the new venture to produce on a regular basis (Nelson & Winter, 1982). This means that firm founders have to initiate contacts with suppliers, engage in product and service development, and seek funds if necessary. Hence, we formulate the fourth hypothesis as:

Hypothesis 4: The earlier a new venture initiates activities to recombine resources the higher the probability for the firm to survive and to get established.

We have hypothesized that different activities have different effects on the outcome of the organizing process. Of special interest here is the reason why new ventures are abandoned by the founding team. Arguably, there is a distinction to be made between voluntary and involuntary disbanding of the new venture. At one extreme, the founding team could voluntarily terminate the new venture for example, because loss of interest or better alternatives. At the other extreme, external stakeholders not supplying finance, competitive aggressiveness, or team conflicts may lead to the impossibility of continuing the effort and cause involuntary termination. Hence, we argue that venture disbandment can happen for different reason and that the causal antecedents vary across these different outcomes. This argument leads to the fifth and last hypothesis:

Hypothesis 5: The causal antecedents affecting venture disbandment due to voluntary reasons will differ from firm disbandment due to involuntary reasons.

METHOD

Design and Sample

During the first nine months of 1998, we contacted by telephone 35,971 adults between the ages of 16 and 70 years, who were living in Sweden. Of those contacted, 30,427 (84.6%) agreed to participate in the survey.

In order to identify those people who were in the process of starting a business during that nine-month period we asked a series of nested screening questions. To meet the sampling criteria, the respondent had to answer that (1) alone or with others, they were in the process of starting a new business, (2) the first activity to start the new venture began during the first nine months 1998, (3) the venture was independent (not part of an effort on behalf of an existing organization), and (4) the respondent was a member of the venture team (not a consultant or passive investor). If the respondent had started more than one venture during the first nine month of 1998, they were asked to discuss only their most recent venture. Of the 30,427 people surveyed, 453 respondents were members of a venture team of an independent effort to start a commercial venture in 1998.

The respondents were asked to give the month and year they first started to work actively to start a new venture. To work actively on the start-up of the new firm is defined as taking an action in the pursuit of the opportunity they have identified. In this way—contrary to Carter et al. (1996)—a difference is made between the period of time in which the founders might have thought about the venture opportunity form the point in time when they first began to exploit it.

As many respondents had started their new venture in prior years, we had to correct the sample to eliminate the risk of selection bias and biased parameter estimates due to left-censoring which is problem when event history models are used to model the start-up process (Blossfeld & Rohwer, 1995; Yamaguchi, 1991). The inclusion of these ventures would have been problematic because other individuals may also have started new business in earlier years, but would not consider themselves “still-in-process” of starting a new venture. Because “long-in-process” ventures would not represent the population of new firms initiated in previous years (as those that had abandoned the effort or completed the start-up process would not be included), we have limited the analyses to the cohort of new ventures that began in 1998, the point in time contemporaneous with the investigation. Of the 453 respondents that were in the process of starting an independent business, 235 had initiated their new venture in the first nine months of 1998. Out of these, 223 ventures had complete data sets needed for the analyses, and form the sample of our investigation.

After the initial interview, we contacted the respondents every six months during two years or until the venture was abandoned. Respondents that drop due to abandonment of the start-up effort or to non-participation are treated as censored from time of last contact. Response rates for each subsequent wave of data collection were 90.5% at 6 months, 91.9% at 12 months, 98.5% at 18 months, and 96.1% at 24 months.

In short, the present design represents a number of advantages relative to previous samples of new ventures. First, the sample is based on a random sample of the adult Swedish population and accurately represents the population of new ventures initiated in Sweden during the first nine months of 1998. Second, there are no left-censored new ventures in the dataset. All ventures are observed from the time of initiation until the venture is abandoned, or the observation was censored after 30 months. Third, the sample does not suffer from the possible survival bias found in samples based on archives or public records. Fourth, this dataset examines the detailed evolution of new ventures from initiation forward in time. This approach allows the examination of the earliest days of the ventures’ lives, a part of the history that is acknowledged as important, but rarely is examined.

Dependent Variables

If, as argued here, the firm formation process is an evolutionary one, then investigation of it should begin with the initiation of new ventures and proceed forward over time. In the beginning of a venture’s life, the milestones to reach are not financial performance as return on sales or profits, but to assembly and organize the necessary resources to establish the new firm on the market. Because the new venture can be abandoned any time after it has been identified, performance in the early part of the venture’s life is best represented by not being abandoned, by becoming established, or both. That is, abandonment of the new venture and the establishment of the new firm cannot be perceived as competing risks, as it is possible to establish a firm and then abandon it during the period of observation.

The establishment of the new firm is measured by asking the respondents every six months if they perceive the new venture as “being up and running.” We divide the life histories of the ventures into monthly spells that begin with when the venture is initiated and ends when the venture is operational. If the respondents considered the venture as “up and running,” the respondents are asked to tell us the month when the venture could be considered as operational. To ensure that the “establishment of a new venture” means the same for all respondents, they were asked in an open-ended format what made them consider the venture as operational. There was almost no variation in the kind of responses they gave: A firm is perceived as being functional when sales and contacts with customers are being made on a regular basis. This dependent variable is coded as a “1” in the month that the venture was established. Otherwise this variable was coded as “0.”

Abandonment of a venture is measured by asking the respondents every six months if all members of the team pursuing it have terminated the new venture. If the venture continued to be pursued by some, but not all members of the venture team, we treat the venture as continuing. We divide the life histories of the ventures into monthly spells that begin when the venture is initiated and ends when the venture is disbanded. If all members have terminated the venture, we ask the respondents to tell us the month when the venture disbanded. This dependent variable is coded as a “1” in the month that the venture was disbanded. Otherwise this variable was coded as “0.”

However, as new ventures can be disbanded for a number of reasons, the respondents were also asked in an open-ended format for the major reason leading to the decision to abandon the effort. A coding of the answer revealed two mutually exclusive categories of answers. A venture is either disbanded for voluntary reason related to the persons starting the business or for reason related to the nature of the opportunity and the organizing process. Examples of personal reasons are: “lost interest, got another job or did not have the time.” Examples of involuntary reasons are: “did not get money, internal conflicts in the team, competition to strong, not a god idea” . As the two categories are exclusive, they are also competing risks, and it is possible that the process leading to one form of abandonment differs from the process leading to another. Two dependent variables were constructed, one for voluntary disbandment, and one for involuntary disbandment. These dependent variables are coded as a “1” in the month that the venture was disbanded (voluntary or involuntary). Otherwise they were coded as “0.”

The new ventures that had not reached an outcome (abandoned or established) by the end of the 30 months observation period were treated as right censored. We analyze the transition to an outcome using Weibull hazard rate models with robust clustering for each venture because we found that specification best fit the underlying distribution of the data. The theoretical arguments proposed here suggest that both time-varying and time constant covariates influence the outcome of the process.

Independent and Control Variables

Independent and control covariates come from two different sources. The majority of the covariates come from the interviews conducted with the respondents. The predictor covariates are composed of different activities initiated by the founding team at various points in time to organize the venture. Examples of such activities are: writing a business plan, searching for fund, initiating marketing and promotion efforts or engage in product development.

Planning is measured as whether or not the founding teams have initiated any planning activities such as: writing a business plan, estimating financial forecasts or search for information about possible competitors and customers. Activities related to the establishment of legitimacy are measured as whether the firm as been registered with the legal authorities or not. Market activities are measured as whether or not the firm founders have established relations with customers or initiated marketing or promotion efforts. Resource combination activities are measured as whether or not the firm founders have bought any supply or inventories, established contact with suppliers, and have initiated product or service development.

These predictor covariates are complemented with a number of different control covariates that are expected to influence the evolution of the process. We control for differences in human capital, that is the size and the gender composition of the start-up team, their previous industry and start-up experience (Brüderl, Preisendörfer, & Ziegler, 1992; Carroll & Mosakowski, 1987). We also control for differences in employment size (Carroll & Hannan, 2000) and legal form (Brüderl & Schussler, 1990).

Of particular importance to understand the evolution of the entrepreneurial process is the ability to control for the nature of the entrepreneurial opportunity in order to eliminate the possibility of confounding the effects of different activities on the one hand with differences in the nature of the opportunity and the firm founders on the other hand as they tend to be correlated with each other (Shane, 2000; Shane & Venkataraman, 2000). This confounding due to intercorrelation introduces bias into efforts to examine the effect of firm organizing activities on the development of the new venture. Therefore, in order to control for the effect of the opportunity we use perceptual measures about the opportunity and industry related data. Industry data were obtained from Statistics Sweden. Based on the open-ended answers given by the respondents about the nature of their venture we asked industry experts to classify the different ventures to appropriate industries at the five digit level (the most fine grained level of coding). Two industry experts coded the ventures separately. The correct classification rate was 89.4%. Where there was no agreement, the classification of the most senior coder was retained. Thereafter, industry covariates were developed to measure differences in industry entries, exits, and population size as well as age (Aldrich & Fiol, 1994; Barron, 1999; Brüderl & Schussler, 1990; Carroll & Hannan, 2000; Romanelli, 1989; Sorensen & Stuart, 2000).

Data on the industry level was updated on a yearly basis using data for 1998, 1999 and 2000. Human capital and most of the perceived opportunity variables are time-invariant and are generated from the first survey.

RESULTS

During the first 30 months of observation, 82 (36.8%) of the 223 new ventures were disbanded. The average time of life before termination was 9.8 month (st.d. 7.0). Of the 82 disbanded ventures, 37 disbanded for voluntary reason, and 45 for involuntary reasons. The average survival time for the former group was 6.4 months (st.d. 6.0) and for the latter it was 11.2 months (st.d. 11.0). The difference is not significant. We also observed that 83 (37.2%) new ventures changed their status from “trying to start” to “up and running.” The average time for that transition was 6.4 months (st.d. 4.1), which is shorter than for venture disbanding. We can conclude from this sample that if a new venture would transfer from “trying to start” to another (either to disband or to become “up and running”), a majority of them would do so during their first year of activity. This is in line with previous results based on US data (Reynolds & White, 1997).

Table 1 shows the event history models predicting the hazard of venture disbanding (models 1a to 3a) and the hazard of establishing (models 1b to 3b). These models form the basis for testing hypotheses 1 to 4. Table 2 shows the event history models predicting the hazard of venture disbanding for voluntary reasons (models 1c to 3c) and the hazard for venture disbanding for involuntary reasons (models 1d to 3d). These models form the basis for testing hypothesis 5. Model 1 shows the effect of planning activities. Model 2 shows the effect of operationalization activities. Model 3 shows the effect planning activities and organizing activities.

We describe the results from the full models (model 3a to 3d) starting with the models for venture disbanding and venture establishment. Several of the control variables predict venture disbanding or establishment consistent with previous research. For example, we find that as the venture mature with ages, the lower is the probability of it either disbanding or getting established. Prior start-up experience reduces the risk of the venture disbanding, but has no effect on the probability of establishing the venture.

The results support hypotheses 1 and 2 but not hypotheses 3 and 4. Consistent with hypothesis 1, engaging in planning activities reduces the hazard of disbanding, and increases the hazard of establishing the venture. Consistent with hypothesis 2, engaging in activities to create sociopolitical legitimacy for the new venture, also reduces the risk for disbandment and augments the probability to establish the venture. However, engaging in market related activities or activities related to resource transformation had no or an opposite effect on the outcome of the process during the first 30 months.

A possible criticism to our results for venture disbandment is that they confound termination for voluntary and involuntary reason. Hypothesis 5 suggests that there should be a difference in causal antecedents between voluntary and involuntary termination. The results support the hypothesis. We found that activities related to planning, creating legitimacy and establishing market relations had a significant effect reducing the risk of involuntary termination, but not on voluntary termination. Also several control variables related to network resources and reason for starting the venture differed between the two modes of termination.

DISCUSSION

We have argued that the activities undertaken by the founding team influence the outcome of the organizing process. We analyzed a unique data set capturing the life histories of 223 new ventures initiated by Swedish firm founders between January and September 1998, and over a period of 30 months. Controlling for a variety of factors related to characteristics of the founding team (social capital, available network resources) and the opportunity (perceived market growth and competitive advantage, and industry characteristics), we have shown that new venture are more likely to survive and get established during their first 30 month of life if the firm founders engage in activities related to planning and creating legitimacy. We did not find any support for activities related to develop market relations or resources transformation. We also found significant difference between voluntary and involuntary termination of the new venture.

The confidence in the finding is enhanced by the chosen approach. The sample represents the population of Swedish new ventures started between January and September 1998. This allows us to generalize to the broader population of new ventures in Sweden. Furthermore, the sample is not burdened with selection bias as all new ventures are observed from the point at which work on them was initiated until they disbanded, were perceived by the founding team as established or censored after 30 months. Finally, our data do not suffer from hindsight and recall bias as it examines planning and organizing activities over the period during which they are actually taking place.

Theoretical Implications

Our results provide implications for several approaches of entrepreneurship research. First, this study provides empirical support for the behavioral approach to entrepreneurship that argues that explaining what firm founders do is useful (Aldrich, 1999; Carter et al., 1996; Katz & Gartner, 1988). For example, both Aldrich (1999) and Carter et al. (1996) have argued that the evolution of new ventures is influenced by the activities undertaken during the organizing process. Our results are consistent with that approach. We find that the rate of disbanding new ventures (and the reason for doing so), and the rate for establishing a new venture is influenced by which activities are undertaken and when they are undertaken.

Second, our results suggest that the relative timing of undertaken activities also matters to the evolution of new firms. New ventures come into existence through a set of organizing sub-processes initiated after that the founding team committed themselves to exploitation of an opportunity. This is in opposition to many researchers that have argued that timing and the relative order in which activities related to the organizing process are undertaken does not influence the probability that the new venture will survive or be established (Carroll & Hannan, 2000; Carter et al., 1996). We have shown that the earlier planning and legitimacy creating activities are initiated the higher is the probability for the new venture to survive and get established.

Third, the results show that the “just-do-it” bias adopted by the entrepreneurship literature is inconsistent with empirical evidence about new venture formation. We have shown that planning activities undertaken early in the organizing process reduces the hazard for disbanding and increases the probability to establish the new firm. This is important because some entrepreneurship researchers have argued that firm founders are better off taking action without planning first (Bhidé, 1999; Carter et al., 1996). We suggest that planning is an important precursor to entrepreneurial action, and that entrepreneurship researchers would do well to consider this relationship in theory development.

Methodological Implications

This study also provides useful methodological contributions to research on venture organizing. We have shown how the effects timing of firm organizing activities on new venture development can be examined by comparing when different types of activities are undertaken. We believe that the use of a methodology to test timing (and ordering) will help researchers to examine more subtle questions about the evolution of new ventures that we have not been able to address previously. In particular, we believe that the examination of timing and order of activities will help us to better understand the processes that undertaken over time and have either short-term or long-term effects.

We also believe that the adopted method in this study represents an example of how researchers can examine the evolution of new ventures in ways that overcome many of the problems of existing research in the field. By studying the process of new venture creation longitudinally, we have shown how researchers can avoid the unrealistic assumption that static characteristics present at the firm founding explain the formation process. We have also shown that prospective data collection overcomes the hindsight and recall bias associated to retrospective data collection. Furthermore, by using different dependent variables measuring the outcome of the process we have tested the robustness of the results.

Limitations

This paper has an important limitation: Only the timing of the activities is measured. There is no information about the quality of the activities and with which perseverance they were initiated. This limitation was necessitated by the complexity of measuring activities longitudinally and the absence of prior evidence on the subject. Nevertheless, the results presented in this study suggest the importance of examining the timing of firm organizing activities. As new venture development is probably influenced by both the order and the quality of planning and organizing activities, future research would advance this approach by the examination of both order and quality.

CONCLUSION

In this study we have tracked the first 30 months of life of 223 Swedish new ventures initiated between January and September 1998 by a random sample of firm founders. We have explored the effect of planning and organizing activities on the development of the venture. We find that activities related to planning and creating legitimacy reduces the probability of disbanding and increases the probability of establishing the venture. We also found significant difference between voluntary and involuntary venture disbandment.

ACKNOWLEDGEMENTS

The research design owes an intellectual debt to the Panel Study of Entrepreneurial Dynamics undertaken by the Entrepreneurial Research Consortium, a temporary voluntary association of 30+ U.S. and non-U.S. universities. The Knut and Alice Wallenberg Foundation, the Swedish Foundation For Small Business Research, and the Swedish National Board for Industrial and Technical Development financed the study. We also thank ESBRI who was Delmar’s employer during the data collection. We thank Alice de Koning for helpful comments on an earlier draft.

CONTACT: Frédéric Delmar, Center for Entrepreneurship and Business Creation, Stockholm School of Economics, P.O. Box 6501, SE-113 83 Stockholm, Sweden; (T) +46-8-736 9356; (F) +46-8- 31 28 85; (E) frederic.delmar@hhs.se

REFERENCES

Aldrich, H. (1999). Organizations Evolving. London: Sage Publications.

Aldrich, H., & Zimmer, C. (1986). Entrepreneurship through social networks. In D. L. Sexton & R. W. Smilor (Eds.), The Art and Science of Entrepreneurship (pp. 3–24). Cambridge, Massachusetts: Ballinger Publishing Company.

Aldrich, H. A., & Fiol, C. M. (1994). Fools rush in? The institutional context of industry creation. Academy of Management Review, 19(4), 645–670.

Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17, 99–119.

Barron, D. N. (1999). The structuring of organizational populations. American Sociological Review, 64(June), 421–445.

Bhidé, A. V. (1999). The Origins and Evolution of New Businesses. New York: Oxford University Press Inc.

Blossfeld, H.-P., & Rohwer, G. (1995). Techniques of Event History Analysis: New Approaches to Causal Analysis. Mahwah, New Jersey: Lawrence Erlbaum Associates.

Brüderl, J., Preisendörfer, P., & Ziegler, R. (1992). Survival Chances of newly founded business organizations. American Sociological Review, 57(April), 227–242.

Brüderl, J., & Schussler, R. (1990). Organizational Mortality: The liabilities of newness and adolescence. Administrative Science Quarterly, 35(3), 530–548.

Carroll, G. R., & Hannan, M. T. (2000). The Demography of Corporations and Industries. Princeton, New Jersey: Princeton University Press.

Carroll, G. R., & Mosakowski, E. (1987). The career dynamics of self-employment. Administrative Science Quarterly, 32, 570–589.

Carter, N. M., Gartner, W. B., & Reynolds, P. D. (1996). Exploring start-up event sequences. Journal of Business Venturing, 11(3), 151–166.

Castrogiovanni, G. J. (1996). Pre-startup planning and the survival of new small businesses: Theoretical linkages. Journal of Management, 22(6), 801–822.

Freeman, J. (1982). Organizational life cycles and natural selection processes. Research in Organizational Behavior, 4(1–32).

Galunic, D. C., & Rodan, S. (1998). Resource recombinations in the firm: Knowledge structures and the potential for Schumpeterian innovation. Strategic Management Journal, 19, 1193–1201.

Gartner, W. B. (1985). A conceptual framework for describing the phenomenon of new venture creation. Academy of Management Review, 10(4), 696–706.

Grant, R. M. (1991). The resource-based theory of competitive advantage: Implications for strategy formulation. California Management Review(Spring), 114–135.

Hansen, E. L. (1995). Entrepreneurial network and new organization growth. Entrepreneurship Theory and Practice, 19(4), 7–19.

Johannisson, B. (1988). Business formation—a network approach. Scandinavian Journal of Management, 4, 83–99.

Katz, J., & Gartner, W. B. (1988). Properties of emerging organizations. Academy of Management Review, 13(3), 429–441.

Larson, A., & Starr, J. A. (1993). A network model of organizational formation. Entrepreneurship Theory & Practice, 17(2), 5–15.

Lounsbury, M., & Glynn, M. A. (2001). Cultural entrepreneurship: Stories, legitimacy, and the acquisition of resources. Strategic Management Journal, 22, 545–564.

Meyer, J. W., & Rowan, B. (1977). Institutional organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363.

Nelson, R., & Winter, S. (1982). An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap Press.

Penrose, E. (1995). The Theory of the Growth of the Firm. Oxford: Oxford University Press.

Reynolds, P. D., & White, S. B. (1997). The Entrepreneurial Process: Economic Growth, Men, Women, and Minorities. Westport, Connecticut: Quorum Books.

Romanelli, E. (1989). Environments and strategies of organization start-up: Effects on early survival. Administrative Science Quarterly, 34, 369–387.

Shane, S. (2000). Prior knowledge and the discovery of entrepreneurial opportunities. Organization Science, 11(4), 448–469.

Shane, S., & Venkataraman, S. (2000). The promise of entrepreneurship as a field of research. Academy of Management Review, 25(1), 217–266.

Sorensen, J., & Stuart, T. E. (2000). Aging, obsolescence, and organizational innovation. Administrative Science Quarterly, 45, 81–112.

Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571–610.

Yamaguchi, K. (1991). Event History Analysis (Vol. 28). Newbury Park: Sage Publications.

2002 Babson College. All Rights Reserved. Last Updated March 2003.