Description
During this brief illustration with regards to coming from good stock career histories and new venture.
Coming from Good Stock:
Career Histories and New Venture Formation
*
M. Diane Burton
Sloan School of Management
Massachusetts Institute of Technology
Cambridge, MA 02142
(617) 253-5539
[email protected]
Jesper B. Sørensen
Sloan School of Management
Massachusetts Institute of Technology
Cambridge, MA 02142
(617) 253-7945
[email protected]
Christine M. Beckman
Graduate School of Management
University of California, Irvine
Irvine, CA 92697
(949) 824-3983
[email protected]
July 2001
*
This research was supported by the Division of Research of Harvard Business School, the Stanford Graduate
School of Business, the University of Chicago Graduate School of Business, the Alfred P. Sloan Foundation, and the
Kauffman Center for Entrepreneurial Leadership. Early versions of this paper were presented at the Academy of
Management, San Diego, CA, August 1998 and at the annual meeting of the American Sociological Association,
Chicago, IL, August 1999. We would like to thank Howard Aldrich, Rakesh Khurana, Josh Lerner, Nitin Nohria,
Damon Phillips, Julio Rotemberg, Bill Sahlman, Scott Shane, Olav Sorenson, Toby Stuart, and Mike Tushman for
helpful ideas and comments on early drafts. We are indebted to Mike Hannan and Jim Baron for their role in the
Stanford Project on Emerging Companies. Stephanie Woerner provided assistance of many types. Finally, we
thank Thomas Hellmann and Manju Puri for sharing their financing data.
Coming from Good Stock: Career Histories and New Venture Formation
Abstract
We examine how the social structure of existing organizations influences the entrepreneurial process and
suggest that resources accrue to entrepreneurs based on the structural position of their prior employers.
We argue that information advantages allow individuals from entrepreneurially prominent prior firms to
identify new opportunities. Entrepreneurial prominence also provides reputation benefits that reduce the
perceived uncertainty of a new venture in the eyes of external constituents. Using a sample of Silicon
Valley start-ups, we demonstrate that entrepreneurial prominence is associated with initial strategy and
the probability of attracting external financing. New ventures with high prominence are more likely to be
innovators; furthermore, innovators with high prominence are more likely to obtain financing.
Introduction
It has often been noted that some of the most radically innovative products and technologies are
developed and commercialized not by existing companies, but rather by entrepreneurial ventures
(Schumpeter 1934; Tushman and Anderson 1986; Henderson and Clark 1990). This is a remarkable fact,
given that creating a new organization requires the mobilization of a substantial array of social and
material resources (Stinchcombe 1965). These resource mobilization tasks are simplified when
entrepreneurs choose to focus on proven, established products and technologies. By contrast,
organizations devoted to new products and technologies face severe hurdles. The entrepreneur must not
only come up with a new idea, but also overcome the skepticism of resource providers, since the
uncertainty and risk associated with any new venture is particularly heightened when the underlying
product or technology is unproven (Aldrich and Fiol 1994).
Where do innovative new ventures come from? The simple answer is that they emerge – directly and
indirectly -- from established firms (Freeman 1986). Contrary to popular myths of lone college drop-out
entrepreneurs, most new ventures are founded by people with prior employment experience (Cooper
1985; Robinson and Sexton 1994). In this sense, entrepreneurs are organizational products:
“Organizations create their own competition by providing the skills and background that provide
credibility for the entrepreneur. They provide the knowledge of opportunity by placing that person in a
position to know about unserved or badly served markets” (Freeman 1986: 39).
In this paper we argue that entrepreneurial opportunities and resources accrue to incipient entrepreneurs
as a function of the structural position of their prior employers. Much as geographical regions differ in
their rates of entrepreneurial activity, established firms differ markedly along this dimension. Some firms
are entrepreneurial hotbeds, as perhaps most famously exemplified by Fairchild Semiconductor, founded
in 1957. Fairchild spawned ten new ventures in its first eight years; moreover, most of the thirty-one
semiconductor firms founded in Silicon Valley in the 1960s could trace their lineage to Fairchild
Coming from Good Stock: Career Histories and New Venture Formation
2
(Saxenian 1994; Rogers and Larsen 1984). Examples of such “Fairchildren” include Intel, Advanced
Microdevices and LSI Logic. Other firms give rise to relatively few, if any, new ventures.
1
In this paper, we argue that one consequence of these differences in rates of entrepreneurial activity
among established firms is to influence the visibility of established firms in the entrepreneurial
community. We use the term entrepreneurial prominence to describe these differences in visibility. We
consider established firms that spawn a large number of new ventures through employee departuresto be
more entrepreneurially prominent than those that do not. Our core contention is that innovative new
ventures are more likely to emerge from established firms that are entrepreneurially prominent. We
expect this for two reasons. First, we believe that there are important informational and resource benefits
to being affiliated with a prominent firm. Second, we argue that there are substantial reputational benefits
that accrue to employees of prominent firms, and that these benefits play a crucial role in reducing the
perceived uncertainty surrounding a venture. We conceive of the entrepreneurial prominence of
established firms as a form of social capital that is transferred to employees as they leave the firm and
attempt to launch new ventures. We explore how differences in the entrepreneurial prominence of
established firms affects the characteristics of entrepreneurial behavior. Thus, while established firms
may have difficulty innovating (Sørensen and Stuart 1999) and commercializing new innovations (Hiltzik
1999), they play an important passive role in shaping the emergence of entrepreneurial ventures.
Our paper is related to and draws on a number of streams of theory and research. First, there is a long
sociological interest in the emergence of new organizations. In fact, one of the early propositions put
forward by Stinchcombe is that “the probability that a man or group of men will be motivated to start an
organization is dependent on the social structure and the position of men within it.” (1965:147).
1
Note that we do not seek to explain why some firms generate more entrepreneurial offspring than others; rather, we
take this distribution as given. See Freeman (1986) and Brittain and Freeman (1986) for a discussion of these issues.
Coming from Good Stock: Career Histories and New Venture Formation
3
Furthermore, sociologists have argued for some time that the dynamics of new venture formation depend
critically on the distribution of opportunities and resources through social structure (Light 1972; Hannan
and Freeman 1977; Aldrich and Zimmer 1986). However, most efforts to examine organizational
emergence come from the ecological tradition where there is keen attention to the broad structural
characteristics of firms, but little attempt to link individuals to this social structure. At the same time,
attempts to link fixed individual attributes, such as psychological characteristics, to entrepreneurial
activity have met with limited success (Brockhaus and Horwitz 1986; Herron and Robinson 1993). Our
paper explicitly reconnects individuals and organizations with a distinctly sociological approach that
allows us to explore whether some individuals, by virtue of their location in the social structure of
existing organizations, are better able to form highly uncertain, innovative start-ups.
Second, there is a growing interest in the economic and sociological literature on technological innovation
in the operation of geographical “spillover” effects (Jaffe, Trajtenberg and Henderson 1993; Saxenian
1994; Stuart and Sorenson 1999). These studies examine how horizontal differentiation along a
geographic dimension affects the innovative activity of established firms and the emergence of new
ventures. For example, Jaffe, Trajtenberg and Henderson (1993) find that the patenting rates of
established firms are positively influenced by the patenting activity of other firms in the same
metropolitan area. Saxenian (1994) argues that the distinctive organizational arrangements and cultures
in Silicon Valley are a source of the regions’ high levels of entrepreneurial activity. Stuart and Sorenson
(1999), studying the biotechnology industry, argue that new firms are more likely to emerge in regions
that have a high density of established biotechnology activity. Similarly, there is a long-standing interest
in understanding how horizontal differentiation among industries leads to differential rates of
entrepreneurship and innovation (e.g., Cohen and Levin 1989). Our approach differs from these
literatures in that we do not seek to understand regional or industrial differences in entrepreneurial
activity; in fact, we focus on a single region, Silicon Valley and a limited number of high-technology
industries. Instead, we investigate the consequences of vertical differentiation among firms in terms of
Coming from Good Stock: Career Histories and New Venture Formation
4
status, prominence, or visibility (Benjamin and Podolny 1999; Podolny 1993; Stuart, Hoang and Hybels
1999).
Third, students of organizations have become increasingly interested in how the movement of individuals
between organizations shapes organizational behavior and industry dynamics. Most studies of this
phenomenon start from the notion that managerial outlooks and predispositions are shaped by career
histories (Gunz and Jalland 1996), and that the movement of managers across firm boundaries is an
occasion for the diffusion of ideas and innovations (Baty, Evan and Rothermel 1971; Pfeffer and
Leblebici 1973; Boeker 1997; Sørensen 1999). Because managers in different firms have divergent
experiences, the pattern of movement between firms is an important determinant of industry dynamics.
Past studies of this process have focused on the movement of managers between existing organizations.
By contrast, we focus on those managers who leave their jobs to start new firms and thus shed light on the
effects of career trajectories on industry dynamics. Fourth, since we attach primary significance to the
career histories of entrepreneurs, our research is also related to work that examines how career histories
shape the entrepreneurial process (Brittain and Freeman 1986; Boeker 1988; Higgins and Gulati 1999;
Shane and Khurana 1999). Work histories are important determinants of the resources available to
entrepreneurs. Prior research on the effects of entrepreneur’s pre-ownership experiences tends to focus
on the types of work that entrepreneurs have performed in the course of their careers (Shane, 1999; Jones-
Evans 1996; Cooper and Dunkelberg 1986). By contrast, we emphasize where entrepreneurs worked
prior to founding the new venture; in other words, we focus on the identity of previous employers. By
doing so, we capitalize on the notion that careers situate the entrepreneur in a social structure of existing
firms that facilitates or constrains the flow of opportunities and resources. We ask how the social
structure of existing firms influences the entrepreneurial process.
This paper makes a number of contributions. First, in contrast to the studies of the relationship between
managerial mobility and organizational behavior cited above, we draw attention to the informational and
reputational benefits that may come from being associated with prominent employers. Second, by
Coming from Good Stock: Career Histories and New Venture Formation
5
focusing on where founders worked, we ask how the social structure of existing organizations influences
the entrepreneurial process. Finally, by differentiating new ventures according to their initial strategy, we
offer a more nuanced explanation of the entrepreneurial process.
In order to cast light on where innovative new ventures come from, we use a unique sample of Silicon
Valley startups and investigate the determinants of their initial strategies and financing. We focus on
explaining two characteristics of these ventures: 1) their founding strategies, specifically whether or not
they pursue an innovation strategy; and 2) the ability of the new ventures to attract external financing at
founding. We use the career histories of the founders, including the identities of past employers, to
examine how differences in the prominence of the established firms affects the strategy and financing of
new ventures.
The remainder of this paper is organized as follows. In the next section, we outline a series of arguments
linking the entrepreneurial prominence of established firms to the characteristics of new ventures. We
then discuss the characteristics of our data and the methods used to test our hypotheses. The presentation
of results is followed by a discussion section.
Entrepreneurial Prominence and Firm Advantage
Students of the link between career histories and entrepreneurship have examined how the accumulation
of human capital through career histories influences the formation of new ventures. In particular, scholars
have emphasized how job experiences shape the technical and managerial skills of (potential)
entrepreneurs. For example, Jones-Evans (1996) examines how the occupational backgrounds of
technological entrepreneurs affect the skills they bring to their ventures. He finds, for example, that
entrepreneurs from academic research settings have strong technical skills but low levels of managerial
competence. In a similar vein, Chandler (1996) examines how the past experiences of founders affect the
success of new ventures. He finds that new venture performance improves to the extent that there is
Coming from Good Stock: Career Histories and New Venture Formation
6
similarity between the task environment of the new venture and the task environment faced by the
entrepreneur in his or her previous job. Similarly, performance tends to improve to the extent that the
skills required in the new venture are similar to those previously developed (Chandler 1996; see also
Chandler and Jansen 1992; Cooper, Gimeno-Gascon and Woo 1994). Finally, Shane (1999) demonstrates
how differences in the past experiences of a set of entrepreneurs shapes their conceptions of the
opportunities associated with the same technological innovation.
As this brief review suggests, the main focus in past research on careers and entrepreneurship has been on
how careers shape the human capital available to entrepreneurs. However, the effects of careers on social
capital are neglected in existing research (Aldrich and Zimmer 1986). Entrepreneurial activity depends
on access to ideas and resources, and such access is differentially available to individuals occupying
different positions in the social structure (Burt 1992). One of the key determinants of an individual’s
position in social structure is her career history, in particular her affiliation with different employers.
Employers shape the personal networks of their employees, expose them to new ideas, endow them with
valuable resources and confer implicit credentials upon them. At the same time, established firms are
differentiated from each other both horizontally – by virtue of being engaged in different activities – and
vertically – by virtue of being more or less visible in different arenas. The nature of the resources
available to employees therefore typically will differ according to the structural position of the employer.
Therefore, to understand how the social structure of established firms affects entrepreneurial behavior, we
must consider the consequences of such differentiation among established firms.
We conceive of the social structure of existing firms as a set of positions hierarchically ordered according
to the prominence of their occupants. Network theorists suggest that an actor’s prominence in a social
network is a function of centrality – the extent to which the actor is extensively involved in relations with
other actors (Knoke and Burt 1983). Prominence garners both informational and reputational benefits for
the actor. In recent years, organizational sociologists have applied this notion of prominence to
understanding organizational behavior and industry dynamics (Podolny 1993; Podolny, Stuart and
Coming from Good Stock: Career Histories and New Venture Formation
7
Hannan 1996; Stuart, Hoang and Hybels 1999). In this paper we are particularly interested in the
landscape of existing firms as it relates to the generation of new firms; thus, we focus on the
entrepreneurial prominence of firms in the existing social structure. Established firms acquire
entrepreneurial prominence by virtue of their being tied to a relatively large number of new ventures.
Unlike strategic alliances (Stuart, Hoang and Hybels 1999) and many other types of interorganizational
ties, these ties generally are not created intentionally by established firms; rather they are formed by
virtue of employees leaving to found new ventures.
Entrepreneurs must be adept at executing two roles: 1) scanning the environment for opportunities and
devising strategies to take advantage of them; and 2) ensuring and managing the flow of resources —
such as capital, supplier relationships and customers — to the venture such that it may pursue its business
strategies successfully. Prior employment experience shapes the capabilities of entrepreneurs with respect
to these two roles. Our argument is that entrepreneurs benefit if they launch a venture from a prominent
position in this social structure. Employment by a prominent firm benefits the entrepreneur in two ways:
1) centrality in entrepreneurial networks makes it easier to identify entrepreneurial opportunities and to
act to exploit the opportunities; and 2) the prominence of prior employers helps to reduce the perceived
uncertainty of a new venture for external constituents.
Entrepreneurial opportunities arise when a prospective entrepreneur receives new information that, when
combined with knowledge already possessed, can be translated into something of value (Shane 1999). As
such, the potential that an opportunity will be discovered is related to both the stock of knowledge an
actor possesses and the flow of new information. This implies that human capital differences can only
partially explain the entrepreneurial process. Structural differences in access to new information must
also be considered.
Network theorists have demonstrated that the quantity and quality of the information an actor receives is a
direct function of the actors social network (e.g., Granovetter 1973; Burt 1992). Entrepreneurially
Coming from Good Stock: Career Histories and New Venture Formation
8
prominent firms, by virtue of their network centrality, will be exposed to stronger flows of new
information about technologies, emerging markets, and unmet customer needs. High quality information
will pass through prominent firms in high volume and at a fast rate; thus, employees have a higher
propensity to make the necessary information combinations and recognize opportunities. But recognizing
the opportunity is only the first step in creating a new venture. Prospective entrepreneurs must take action
to transform an opportunity into a venture. Here, we believe that nascent entrepreneurs in
entrepreneurially prominent firms vicariously benefit from the experiences of those entrepreneurs who
preceded them. As employees exit to launch new ventures, they likely deposit knowledge about the
appropriate steps and methods for building an enterprise with former colleagues and coworkers. First,
coworkers rarely immediately sever ties when colleagues change employment. Second, organizations
have memories residing in long-tenured employees about the actions and activities of former employees.
Thus, tactical knowledge of entrepreneurship – which law firms to call, which financiers to meet, where
to locate offices – becomes part of the stock and flow of information available to employees of
entrepreneurially prominent firms.
Although the information benefits of working for a prominent employer can help people to identify
entrepreneurial opportunities and allow them to take appropriate steps towards becoming an entrepreneur,
prospective founders still face substantial obstacles to launching the new venture. Entrepreneurs must
successfully mobilize the resources of wealth, power and legitimacy necessary to realize their vision
(Stinchcombe 1965; Aldrich and Zimmer 1986). Doing so requires overcoming information asymmetries
that make it difficult for external resource providers to assess the quality of a new venture or its founders
ex ante. These problems are exacerbated to the extent that a venture wishes to pursue a new, unproven
strategy. Under these conditions, external actors are likely to arrive at an estimate of the quality of the
venture by considering more easily observable attributes that are thought to be associated with the quality
of the venture (Stuart, Hoang and Hybels 1999; Podolny 1993; Spence 1974).
Coming from Good Stock: Career Histories and New Venture Formation
9
One source of information on the quality of the venture lies in the prior accomplishments of the founding
team members. Here, the career histories of entrepreneurs enter through a consideration of the
experiences and skills that have accumulated through the career. Studies suggest, for example, that
venture capitalists are particularly interested in the background experiences and managerial capabilities of
entrepreneurs (MacMillan, Siegel, and SubbaNarishma 1985; Goslin and Barge 1986; but see Hall and
Hofer 1993). Indicators of technological competence might include educational credentials and patents
held.
A second class of information on the quality of new ventures is reputational, and focuses on the identity
of the entrepreneurs themselves. Sociologists have long maintained that individual reputations are in part
constructed from the identities of the parties with whom a person associates (Blau 1964). In particular,
individual reputations benefit from association with prominent actors (Goode 1978). These reputational
advantages in turn facilitate the mobilization of resources and social action. Sociologists of science, for
example, argue that scientific careers are enhanced to the extent that young scholars are affiliated with
prominent individuals in the field (Merton 1968). Latour (1987) suggests that the reception accorded to
new ideas depends on the prominence of the scientist’s associates. Podolny and Stuart (1995) show that
other actors are drawn to innovations that are advanced by actors whose prior technological contributions
are perceived as important.
In a study of entrepreneurial ventures in biotechnology, Stuart, Hoang and Hybels (1999) demonstrate
that the prominence of a venture’s alliance partners is positively associated with its performance. In their
model, new ventures with prominent affiliates benefit from an implicit transfer of status from the
affiliates. In the eyes of third parties, association with high-status partners functions as a guarantee as to
the quality of the venture. Affiliation with prominent partners therefore gives firms an advantage in the
competition for customers, suppliers and employees.
Coming from Good Stock: Career Histories and New Venture Formation
10
We argue that the prominence of prior employers plays a similar role in reducing the perceived
uncertainty of a new venture. External actors use information on previous employers to make inferences
about the likelihood that the founders will build a successful venture. Third parties suffer from an
information asymmetry that makes it difficult to assess the true abilities of potential entrepreneurs ex
ante; this asymmetry is analogous to the one faced by employers when making employment decisions. In
this setting, founders who come from employers that are established incubators for entrepreneurial talent
benefit from this association. In other words, the prominence of previous employers may function as an
indicator of the quality of the prospective founder (Spence 1974). This supposition -- that employees of
prominent firms are, on average, of higher quality -- may indeed be correct. If more prominent firms are
also more successful, for example, they can devote greater resources to attracting and retaining skilled
personnel, and they may invest more in training. Furthermore, highly skilled individuals can be expected
to prefer employment with prominent employers, improving the pool of candidates for employment at
such firms. (Note that we are careful here not to argue that employers explicitly seek to certify their
employees; nor, that they have a reputational incentive to ensure that they hire only highly qualified
employees. Unlike alliance partners, who may put their reputations at risk when associating with a new
venture (Stuart, Hoang and Hybels 1999), we do not see the reputations of prior employers as being at
risk in the entrepreneurial ventures of their employees.)
Finally, employees of entrepreneurially prominent firms are advantaged because more is likely to be
known about them in the entrepreneurial community. Just as centrality in information networks may help
founders gain access to information and resources, it also helps diffuse information about founders and
their new ventures. The experiences and accomplishments of prospective founders will be more widely
recognized if they come from prominent employers. In this respect, affiliation with prominent employers
may reduce the information asymmetries faced by a new venture directly.
For each of these reasons, then, third parties may infer that founders from more prominent employers
possess, on average, greater skills and have a higher probability of success in their new ventures. Because
Coming from Good Stock: Career Histories and New Venture Formation
11
of this, we suspect that startup proposals from employees of prominent firms will a priori seem more
promising and hence receive more attention from external actors. Thus, external actors will have a
greater level of confidence in the ability of such founders to have success in the new venture.
Employment by prominent firms, in short, should reduce the perceived uncertainty of a new venture.
This idea is consistent with early statements by Stinchcombe (1965:146-7) regarding entrepreneurship,
where “…the patterns of trust and of mobility of resources which determines whether resources can be
moved to innovators are socially patterned.” In this paper we propose a specific source of theses social
patterns – past employers.
Hypotheses
Our arguments suggest that potential entrepreneurs secure informational and reputational benefits by
virtue of having once been employed by a prominent firm. Furthermore, these benefits will be amplified
in the face of uncertainty. In order to test these assertions, we first examine the level of uncertainty
involved in the kinds of ventures entrepreneurs start. We distinguish between ventures that pursue
innovation strategies and those that pursue other strategies. Most of the organizational strategy typologies
employed by empirical scholars allow for this distinction (e.g. Miles and Snow 1978; Porter 1980; Miller
1986). One theme across all of the typologies is the importance of differentiating firms that are exploiting
an existing market from those that are creating a new market. Following Maidique and Patch (1982), we
believe that this is an especially salient distinction for technology-based firms. In our definition, firms
pursuing an innovation strategy are seeking to win a technology race in a new niche. These firms are
attempting to gain competitive advantages by being the first to develop and exploit new, hitherto
unproven technologies. By contrast, incrementalist startups build upon existing products and
technologies, and seek to gain competitive advantage through technical enhancements, superior marketing
and customer service, and/or cost advantages. A critical difference between innovative and incrementalist
startups lies in the degree of uncertainty associated with a new venture of each type. High levels of
Coming from Good Stock: Career Histories and New Venture Formation
12
uncertainty characterize innovative startups, as the core products and technologies around which they are
built are of unknown value. The level of uncertainty for incremental ventures is correspondingly lower,
since external actors can more readily judge a venture’s promise by reference to existing firms. The
information asymmetries involved in assessing a venture’s quality are exacerbated for innovative
ventures, and founders of innovative ventures must therefore overcome a greater degree of perceived
uncertainty regarding their firm’s prospects.
Identifying opportunities for innovative ventures requires a good understanding of the future path of
technological progress and a wealth of information about technological alternatives. Entrepreneurs from
prominent firms should be at an advantage in both respects. The reputational benefits of being affiliated
with a prominent employer should make entrepreneurs more successful at reducing the perceived
uncertainty of their ventures. As a result, the informational and reputational benefits of working for a
prominent firm should translate into a superior ability to identify opportunities for innovation. Thus we
hypothesize that
Hypothesis 1: The prominence of prior employers will be positively related to whether a firm
pursues an innovative strategy.
Our second hypothesis concerns the ability of entrepreneurs to secure external financing at the time of
firm founding. In particular, we are interested in whether entrepreneurs from prominent firms are better
able to acquire resources from third parties. We suspect that the willingness of third parties to invest in a
highly uncertain venture during its infancy, before it has any track record by which to be assessed,
depends on the perceived quality of the new venture. There are three ways in which a brand new venture
can have higher perceived quality: 1) its founders have high levels of human capital; 2) it has a product
which can be independently evaluated; and 3) it has ties to prominent firms that serve as endorsements.
Coming from Good Stock: Career Histories and New Venture Formation
13
While in our analyses, we will attend to all three mechanisms, we are primarily interested in the third,
thus we hypothesize that
Hypothesis 2: The prominence of founding team’s prior employers will have a positive effect on
the probability of a new venture obtaining external financing at the time of founding.
Data and Methods
Data on new ventures
The data for this study are from the Stanford Project on Emerging Companies (SPEC). SPEC is a
stratified random sample of 173 young high-technology firms in Silicon Valley.
2
The sample is drawn
from the population of firms listed in Rich’s Everyday Sales Prospecting Guide (1994) and The
Technology Resource Guide to Greater Silicon Valley (1993/1994) and supplemented with firms from the
Silicon Valley business press that were too young to appear in published directories. SPEC is a
longitudinal study of organizational evolution with emphasis on formal systems and practices. In order to
minimize recall bias and to guarantee that the entities under consideration could potentially have the need
for formalized structures and systems, age and size criteria were used to define the population. Firms
included in the study were no older than 10 years and had at least 10 employees at the time of sampling.
At the time of sampling, the average firm was 7.3 years old and had 89 employees. The sample included
2
For details on the data collection and coding methods, see Burton 1995; Baron, Burton and Hannan 1996; Hannan,
Burton and Baron 1996. These publications describe the original sample of 100 firms for which data was gathered
in the summer of 1994. The sampling and data collection strategies were replicated in the summer of 1995 to
supplement the sample with an additional 72 firms (See Baron, Hannan and Burton 1999, 2000 for more
information).
Coming from Good Stock: Career Histories and New Venture Formation
14
firms that ranged in age from 2 to 12 years and in size from 9 to 2042 employees. The SPEC research
team conducted interviews with founders, CEOs, and senior managers responsible for human resources,
gathered survey and archival data, and compiled detailed organizational histories for each of the firms in
the study.
Bhide (1999) reports that 60% of start-ups fail within the first six years and that even those which survive
remain small. Based on these statistics, the SPEC sample is likely biased towards successful start-ups.
The nature of the sampling frame means that the firms under investigation have achieved some minimum
scale and longevity. However, it is important to note that attempts were made to include younger firms,
precisely to minimize survivor bias. Furthermore, despite this data limitation – which plagues virtually
all survey or interview-based organizational research – the SPEC sample has some noteworthy
advantages for our purposes. First and foremost, it includes firms pursuing different strategies, with
different sources of capital. Second, it is geographically constrained, which increases the probability that
multiple founders will have held positions in a given employer and thus generate variation in our measure
of prominence. For these reasons, we believe the sample is appropriate for testing our ideas.
The first dependent variable of interest in this paper is whether or not the firm was founded to pursue a
technological innovation strategy. Trained MBA and doctoral students conducted semi-structured
interviews with a founder of each of the firms asking him or her to describe the core competence of the
firm at founding. The open ended response (supplemented in some cases by early press reports, product
announcements, business plans and prospectuses) comprised the raw data that was used to categorize each
of the firms into one of four strategic archetypes: Innovators, Enhancers, Marketers and Low-Cost
Producers (see Hannan, Burton and Baron 1996). Innovators are firms that seek to gain first-mover
advantages by winning a technology race. Enhancer firms seek to produce a product similar to other
companies, but employ a general modification or enhancement to gain competitive advantage. Marketers
seek competitive advantage through superior sales, marketing or customer service. Finally, Low Cost
Producers are firms that seek cost advantages through cost efficient production techniques, relationships
Coming from Good Stock: Career Histories and New Venture Formation
15
with low cost suppliers, or economies of scale. The three latter strategies all revolve around extending
existing products or services. For the analyses presented here, we collapse the latter three categories into
one category, thereby focusing on the distinction between innovators and incrementalists. In light of
suggestions that entrepreneurs may selectively recall their company’s history (Bhide 1999), the use of a
retrospective measure of strategy may seem problematic. However, we feel confident that our measure
captures the difference between innovators and incrementalists with a high degree of accuracy. In
particular, respondents were not asked to classify their strategies themselves; rather researchers coded
strategies based on business plans, prospectuses, and articles from the business press describing the
industry. Furthermore, Hellman and Puri (1999) perform a number of post-hoc analyses of the same data,
including linking patenting activity to strategy and finding that innovators accumulate larger patent
portfolios, which increase our confidence in the measure.
The second dependent variable is whether the firm received external financing at the time of founding.
New ventures have a variety of alternative sources of capital. Some entrepreneurs self-finance the early
start-up phase using their own personal assets. Other entrepreneurs have a source of revenue or cash
flow, such as a licensing agreement or a consulting contract, that finances the venture. Still others are
able to mobilize the resources of friends and family to support the early stages of a new enterprise. Many
must seek capital from external third parties such as venture capitalists, private investors (so-called
“angels”), corporate investors, commercial or investment banks, other financial institutions such as
pension funds or insurance companies, or the government. These alternative funding sources vary in the
extent to which they are willing to associate with risky ventures and in the price at which they provide
capital. Venture capitalists and private investors anchor the end of the continuum that finances the most
uncertain enterprises at the highest price (see Roberts and Stevenson (1991) for an overview of start-up
financing). Not surprisingly, most of the start-up firms that received external funds at inception obtained
the funds from these high price sources. Thus, the task of attracting external financing was one of
persuading investors who were in the business of evaluating risky ventures.
Coming from Good Stock: Career Histories and New Venture Formation
16
Information on the financing history of each of the SPEC firms was collected via a combination of public
and proprietary databases, SEC filings and annual reports, internal company documents and a survey
instrument that was sent to the most senior finance executive at each of the firms.
3
In this paper we focus
on whether or not an entrepreneurial venture received funds from any outside investor at inception. We
do not distinguish types of investors, nor do we differentiate so-called “seed money” from first or second
round financing, nor do we account for differences in the amount of financing. Instead our variable is
simply a measure of whether the founding team had any amount of money from any third party at the
earliest moments of the firm’s existence. Of course it is interesting to note that the vast majority of
external investors in the SPEC sample were venture capitalists (71%) with some angels (13%) and
corporate investors (13%). Only one of the SPEC firms borrowed start-up funds from a commercial bank.
The invested amount for all stages of investment (for the subset of firms for which we have the data)
ranged from $10,000 to $30,000,000 with one firm receiving $100,000,000 in cash and stock as part of a
merger agreement. Excluding this outlier firm, the average investment size is just under $2.5 million, and
the amounts of initial investments are one the small end of the distribution.
We are interested in whether a founding team can persuade third parties that their venture is promising
before they have begun significant operations and have a tangible organizational track record that can be
3
The financing history data collection effort was led by Professors Thomas Hellmann and Manju Puri of the
Stanford Graduate School of Business (Hellmann and Puri 1999). Sixty-six firms (38%) responded to a finance
history survey that was addressed to the senior executive responsible for finance. Data for a large number of the
sample firms was available from commercially available databases that track the venture capital industry. 107
(62%) of the SPEC sample firms had records in the Venture One database (see Gompers and Lerner 1997 for a
discussion of this database); 95 (55%) had records in the Venture Economics database (see Lerner 1995 for a
discussion of this database). Additional information was gleaned from the founder interview transcripts as well as
archival research in the business press.
Coming from Good Stock: Career Histories and New Venture Formation
17
evaluated. Having an external third party provide financial capital is evidence that the team was
successful. Ideally, we want to capture those firms that have external funds at inception. In practice, this
is difficult to operationalize. There are ambiguities and inaccuracies inherent in both the founding dates
and the financing dates. Most scholars define the birth of a firm as the date that it was legally
incorporated. However, in constructing organizational life histories for very young firms we discovered
that many had substantial lives – ones that involved full-time employees and/or revenues from sales –
well before the founders ever approached a lawyer to incorporate them. For this reason we define the
founding date as the earliest possible date that there is any indication that a new organization exists,
including legal incorporation, having a full-time employee, or selling a product. This sets the start-up
clock to begin at the earliest possible moment. Similarly, financing dates recorded in commercial
databases and on our surveys were often recorded in terms of annual “quarters.” Even in cases where a
precise date was given, we understood this to be the date a financing deal closed, rather than the moment
when negotiations commenced or even when a “handshake” or verbal agreement was reached. Thus, in
order to accommodate these recording inaccuracies, we coded a firm as having external financing at
founding if it was received within three months of the founding date.
4
Data on Founders’ Careers
We have further augmented the SPEC data with information on the career histories of each of the
founding team members of the SPEC companies. As part of the data collection process, the SPEC
research team interviewed and surveyed a founder of each of the firms. This informant was asked to
4
We tested alternative intervals. The results when we more strictly define the date of financing are weaker, since
there are fewer positive outcomes, but in the same direction. We obtain statistically significant findings that are
substantially equivalent to those reported when we expand the financing interval to be within the first six months of
founding. We chose to report the analyses from the slightly more conservative three-month interval.
Coming from Good Stock: Career Histories and New Venture Formation
18
identify, by name, the other members of the founding team. This list of founding team members was then
verified through archival research of public documents as well as internal company records available to
the research team. Among the 173 SPEC firms, founding team size ranged from 1 to 12 (with an average
team size of 3). For each founding team member, SPEC research assistants searched a number of archival
sources, including SEC filings, company documents, newspaper articles and profiles, electronic databases
such as Lexis/Nexis, and internet archives in order to reconstruct each founder’s job history prior to
launching the new venture. For each of the 527 founders, we attempted to collect information on all jobs
held prior to the start of the new venture including the position held and the name of the employer. We
contacted the human resources department for 20% of the firms and confirmed the founders prior place of
employment. The career history data collection process generated a list of 1252 positions in 438 distinct
prior employers. In our data, the number of prior jobs held by a given founder ranges from 0 to 9. Of the
527 founders we identified at least one prior employer for 420.
5
Our key independent variable, employer
prominence, requires that we have data on at least one prior employer for at least one member of the
founding team. We were unable to collect any career history or educational background information for
any of the founders at 9 of the 173 SPEC companies; thus our sample is reduced to 164 firms.
5
We confirmed that there were at least 38 additional founders who began working at the SPEC firm directly from
school and thus their number of prior employment ties was truly 0. For the remaining 69 it is difficult to ascertain
whether missing data arises because the founder had no prior jobs, or whether the experience was simply not
reported in our sources. We suspect that there is some bias toward large, established firms being mentioned in press
accounts about the individuals in our sample; employers that are less important in the eyes of the media may not be
mentioned in newspaper stories and press releases. We attempt to account for this problem in our analyses by
replicating the models using different numbers of prior jobs. At a minimum, it is important to note that since we
were unable to administer job history interviews to the founders, these data are imperfect records of the career
histories of the SPEC entrepreneurs.
Coming from Good Stock: Career Histories and New Venture Formation
19
Measuring the Prominence of Past Employers
The central predictor of interest is the prominence of each founder’s past employers. Our measure of
prominence should capture the extent to which an existing firm is visible to those engaged in
entrepreneurial activity (cf. Knoke and Burt 1983). Given the diversity of industries represented by our
firms, it is difficult to think of a single dimension along which all of the firms can be unambiguously
ranked. Asset- or revenue-based size measures may have some applicability to for-profit organizations,
but are difficult to apply to universities, for example. Measures of technological prominence based on
patents (Stuart, Hoang and Hybels 1999) may also be a plausible basis for ranking firms. Again,
however, difficulties arise with respect to cross-industry comparisons and with respect to the best way to
characterize the prominence of firms operating in multiple industries. More importantly, it is unclear
what criteria third parties in an entrepreneurial context use to assign prominence to existing firms, and
whether these criteria are consistent across industries. Ideally, we would want an independent reputation
survey of all existing firms completed for each of the years from 1982 to 1992 when sample ventures
were founded. Unfortunately, we know of no such survey.
For these reasons, we choose to arrive at the prominence of past employers inductively, based on the
observed pattern of entrepreneurial activity in our sample. We measure a firm’s prominence by the extent
to which it has been a source of entrepreneurial ventures. Firms that generate a lot of new ventures
should be more visible to other actors in the entrepreneurial community. In order to construct this
measure, we start with a binary matrix of ties between the 164 new ventures in the SPEC sample that have
some prior career data for the founders and the 438 past employers of all of the founders. A “tie” is
formed if any member of the founding team had worked for the past employer. Thus entries in the cells
i,j of this matrix are 0 if there is no founding member at start-up i who worked for the past employer j,
and 1 if there is at least one founder who worked for the past employer j. Summing across the rows
generates a count, for each start-up, of the number of prior employers represented on the founding team.
Coming from Good Stock: Career Histories and New Venture Formation
20
a x
i ij
j
?
?
Summing down the columns of this matrix generates a count, for each past employer, of the
number of startups in the SPEC sample that have emanated from that employer.
b x
j ij
i
? ?
?
1
We subtract 1 from b
j
to exclude the focal start-up; if a past employer has only generated one start-up
(each has generated at least one), it will have a prominence score of zero. For each venture in the SPEC
sample, we then generate a measure of the prominence of all of the past employers by summing the b
j
across each of the past employers represented on the founding team.
6
Using the observed entrepreneurial activity in our sample to measure the prominence of past employers
may strike some as tautological, given that we are seeking to explain entrepreneurial activity. There are
two reasons why we believe this is not so. First, we do not use this measure to predict the rate of
entrepreneurial activity, but rather as a predictorof the characteristics of the entrepreneurial activity. We
do not see a necessary connection between our measure of prominence and whether entrepreneurs pursue
innovation strategies, much less whether they are able to secure external financing at start -up. Second,
we conceive of the SPEC sample of start-ups as generating a sample of past employers of start-up
founders, where the past employers are represented proportional to the entrepreneurial activity that they
generate. Our strategy thus parallels the National Organizations Survey, which generated a sample of
organizations by asking randomly selected individuals to name their employers (Kalleberg, Knoke,
6
This measure of prominence will increase on average with the number of prioremployers recorded for a founding
team. In order to account for this, we control for the number of past employers in the models.
Coming from Good Stock: Career Histories and New Venture Formation
21
Marsden and Spaeth 1996; see also McPherson 1982). This sampling procedure – termed probability
proportional to size samples – are statistically optimal for populations where the elements vary widely in
size (Sudman 1976). We believe that replications of this procedure for new high -technology ventures in
Silicon Valley would generate similar lists of past employers.
The career history data collection process generated a list of 1252 positions in 438 distinct prior
employers. If people were described as being in self-employment (such as doctors or independent
consultants), or for some other reason the firm was not identified, the “prior firm” was coded as missing.
Missing prior firms account for 87 of the 1252 positions (6.9%). The remaining positions are infirms
that range from familiar high-technology employers in Silicon Valley – such as Hewlett-Packard, Intel
and Apple – through academic institutions – such as Stanford and Harvard – to the military and less well-
known firms. Despite the diversity of firms, there is a surprising degree of concentration in
entrepreneurial activity (see Table 1). For example, 6 prior employers dominate the list (IBM, Hewlett -
Packard, Stanford University, Apple Computer, Intel, and National Semi conductor) and 69 SPEC firms
(46% of the sample) have at least one founder who worked at one of these six firms. It is also worth
noting that, while this list captures many large Silicon Valley employers, it is not collinear with size. One
of the largest employers, Lockheed, with over 21,000 employees in 1990, does not appear on the list of
prominent firms. Furthermore, for firms such as Apple Computer (5,700 employees) and Sun
Microsystems (7,700 employees) the prominence measure appears unrelated to s ize. Apple’s prominence
score is double that of Sun’s (12 compared to 6).
7
7
Numbers of employees by firm is based on data from September 1990 and was reported in a San Jose Mercury
News article, “Largest Employers” printed Monday, January 14, 1991 on page 2C.
Coming from Good Stock: Career Histories and New Venture Formation
22
Control Variables
In addition to the prominence of past employers, we control for a number of other characteristics of the
SPEC companies and their founders (since we are studying the firms at their inception, there are few
organizational characteristics to measure). In the external financing models, we control for how far along
each company is in the entrepreneurial process by including an indicator as to whether or not they had a
completed product ready for shipment within six months of founding. Having a working product, or even
a product prototype, is one way that a firm can reducing the perceived uncertainty for external
stakeholders. We believe that this approach to reducinguncertainty will be particularly effective for firms
pursuing an innovation strategy; thus, we include an interaction term. We also control for industry for
several reasons. We want to capture differences in the need for capital (medical devices compani es on
average should require more initial capital than software companies) and the attractiveness of an industry
in the capital market. We also need to account for different baseline levels of innovativeness across
industries. Finally, industries are notequally represented in the sample. In the analyses presented in this
paper we include dummy variables for three broad industries: medical -related (including medical devices
and biotechnology), networking and telecommunications, and semiconductors. The omitted category
consists primarily of computer hardware and software companies, electronic component manufacturing
companies, and contract research and development firms.
Past research suggests that career experiences shape the propensity and ability ofindividuals to launch
entrepreneurial ventures. We control for a number of such experience-related characteristics in our
models. First, we control for the number of founders with prior entrepreneurial experiences. Second, we
include measures of the number of founders with prior senior management experience. Some evidence
indicates that venture capitalists take into account the management experience of entrepreneurs
(MacMillan, Siegel, and SubbaNarishma 1985). We also control for the number of founder s with
experience in sales/marketing or finance in order to control for the possibility that the perceived quality of
a team may be related to the presence of functional diversity.
Coming from Good Stock: Career Histories and New Venture Formation
23
Finally, we control for the general human capital of the founders by including education level in our
analyses.
8
Specifically, we measure the proportion of the founding team that has advanced degrees (i.e.,
more than a B.A.) We also consider the possibility that third parties might look to tangible measures of
accomplishment asan indicator of the quality of the founding team; thus, we collected information on all
of the patents granted to each of the founders in our sample prior to the launch of the new venture.
9
We
interpret this as a measure of the technological or innovative competence of the founding team
members.
10
Table 2 presents descriptive statistics for all variables in the models. Bivariate correlations are presented
in Appendix A. As is apparent from Table 2, almost half of the ventures in our sample pursue an
innovation strategy. Slightly over a third of the SPEC companies have external financing at the time of
founding, while approximately one-sixth have a product within the first six months. On average, 3.4
different employers are represented on each founding team, which is slightly more than the mean
8
The models that we report in this paper include only the education control variables. The findings are equivalent
when we include age as a proxy for experience; however our sample size is dramatically reduced due to the
difficulty in locating reliable birthdates for the founders.
9
Patent data for each individual was collected through the U.S. Patent Office’s web sitehttp://www.uspto.gov
10
We also collected information on the number of citations to each founder’s portfolio of patents; however, this had
no effect in our models.
Coming from Good Stock: Career Histories and New Venture Formation
24
founding team size (2.95).
11
The vast majority of prior employers (338 of 438) have a prominence score
of 0. The maximum prominence score for a prior employer is 21. Aggregating across all prior employers
for a team yields a prominence score range from 0 to 52 with the average SPEC firm earning 9.6 points.
Results
We argue that firms pursuing innovative strategies face higher levels of uncertainty. Evidence of the
higher level of uncertainty surrounding innovation strategists can be found in Table 3, which cross-
classifies the initial strategy of the ventures in our sample by whether or not the firm had external
financing at founding. While almost half of the incrementalist firms had external funding at the outset,
only 28% of those firms pursuing an innovation obtained such funds. This difference is statistically
significant. By construction, incrementalist firms are operating in known market niches where there are
already established entities. There is both an identifiable market opportunity and a means to assess – and
benchmark – the quality of the product or service being offered by the new venture. Neither is possible
for innovative firms. Thus, we interpret this table to support our claims that third parties, such as venture
capitalists, are less willing to provide initial funding to new ventures that pursue high -risk and uncertain
innovation strategies.
11
It is possible that our data collection strategymisses firms that are prominent in an entrepreneurial context but that
do not garner media attention. For these reasons, we conducted the analyses using only the immediately prior job
for each founder, using three prior jobs per founder, and using all available data. The results conform to our
hypothesized expectations; however, the prominence distribution is greatly constrained in the first case and
dramatically skewed in the latter. We report the intermediate choice, allowing up to three prior jobs for each
founder, in this paper.
Coming from Good Stock: Career Histories and New Venture Formation
25
In Table 4, we present logistic regression estimates of the determinants of a ne w venture’s strategy at the
time of founding. We focus our discussion on the fourth column of results. The parameter estimates
suggest that the past career experiences of founders have an impact on their choice of strategy. Teams
with a lot of experience in sales or finance are, as might be expected, less likely to pursue innovation
strategies. Graduate education also has a positive impact on the decision to pursue an innovationstrategy.
As predicted by Hypothesis 1, the prominence of past employers has a positive impact on the propensity
to pursue risky strategies. A one-standard deviation increase from average prominence of the founding
team’s past employers increase the odds of pursuing an innovation strategy by a factor of 1.65. This
result is consistent with our claim that entrepreneurs benefit from being associated with prominent
employers. Our measure of the prominence of past employers captures the extent to which firms are at
the center of entrepreneurial activity. Thiscentrality in entrepreneurial networks can have both
informational and reputational benefits which make it more likely that employees of prominent firms will
pursue innovation strategies.
It is difficult to differentiate the information and reputation acc ounts as explanations of the prominence
effect. The information story suggests that employees of prominent firms take advantage of ideas and
innovations that they are exposed to in the course of their work. Their employer may not be aware of
these ideas, or may not be interested in pursuing them. In order to explore the role that such exposure
may play, we turned to the explanations given by founders in response to the question, “What was the
catalyst or impetus for founding the company?” Out of these open-ended responses, we coded whether
the founder indicated that the idea that formed the basis of the new venture had come from work being
done at a prior employer. Of the firms for which we have suchinterviews (N=131), 23% mentioned that
projects they had undertaken in the context of a prior employment setting as the catalyst for starting the
venture. In separate models, we included a dummy variable indicating whether such a project with a prior
employer was the impetus for the new venture. This variable had no effect on the probability of pursuing
Coming from Good Stock: Career Histories and New Venture Formation
26
an innovative strategy, and had a negligible effect on the relationship between the prominence of past
employers and venture strategy.
We also experimented with a different measure of employer prominence, in part because of our lingering
concern over whether our effects are driven by differences in the size of prior employers. Size may be
relevant since it has been shown to affect organizational innovation processes (Cohen and Levin 1989).
We do not have direct measures of employer size. Instead, we created a dummy variable indicating
whether the prior employer was listed in the Silicon Valley 100, an a nnual listing of the largest firms in
Silicon Valley produced by the San Jose Mercury News. In separate analyses (available from the
authors), we experimented with various ways of including information on the number of prior employers
listed in the Silicon Valley 100. None of these affected the propensity of firms to pursue an innovation
strategy, and the effects of entrepreneurial prominence were robust throughout the different
specifications.
In Table 5 we turn our attention to the determinants of external financing at founding. These estimates
are from logistic regression models of whether or not a venture had external financing within three
months of founding. As the cross-classification in Table 3 suggested, firms pursuing an innovation
strategy are less likely to secure external financing at start-up. In the second model, we introduce an
interaction effect between the firm’s strategy and whether or not they had a product at founding. We see
that these two variables have a complex effect on the li kelihood of external financing at founding. The
main effect of the product variable indicates that firms pursuing an incrementalist strategy are less likely
to secure external financing at founding. This may seem counterintuitive. However, it is importa nt to
note that our dependent variable primarily captures infusions of venture capital, which comes at a higher
cost than traditional sources of capital (such as bank loans). Since incrementalists are operating in
established markets, those with a productin hand have the least need for this more expensive type of
financing. In fact, they may be able to generate sufficient revenue from sales to mitigate the need for any
external financing. Turning to innovation strategists, the interaction effects sugges ts that these firms,
Coming from Good Stock: Career Histories and New Venture Formation
27
having a product makes it more likely that the firm will receive external financing at founding. Unlike
incrementalist firms, however, innovative startups are more in need of venture capital due to the
uncertainty surrounding the market for their products.
In the next two models, we include measures of the experiences and achievements of the founding team.
First, we see that prior founding experience has no effect on the odds that a new venture will receive
external financing at founding. This may be due to the fact that our measures captures only whether or
not a founder had been involved with a prior start -up, but nothing about the outcome. If the prior
founding experiences have had negative outcomes, third parties may be hesitant to invest in another
venture. Alternatively, if the prior founding experience had been successful, and the entrepreneur has
“cashed out,” his or her own personal wealth may obviate the need for external financing in the early
stages of the firm. It is also worth noting that prior founding experience is significantly correlated with
senior management experience – which has a positive effect on the odds of attracting external
stakeholders. Founding teams whose members include at least one with prior senior management
experience are more likely to secure external financing at founding. This is consistent with studies
showing that venture capitalists value the management experiences of entrepreneurs when evaluating
proposals (MacMillan, Siegel, and SubbaNarishma 1985). Neither the innovative ability of the founders,
as measured by the number of patents held, nor graduate credentials have a significant effect.
12
We see in the fifth model in Table 5 that the prominence of past employers initially has no signifi cant
effect on the odds of securing external financing at startup, suggesting no support for Hyothesis 2.
However, this model does not take into account the different levels of uncertainty associated with
innovation strategies and incrementalists. We expect entrepreneurial prominence to be especially
beneficial when the perceived uncertainty of the venture is high, such as when a firm pursues an
12
We tested for an interaction effect with the strategy of the firm; it was not significant.
Coming from Good Stock: Career Histories and New Venture Formation
28
innovative strategy. In the final model (model 6) in Table 5, the effect of employer prominence differs
for the two types of firms. Among firms pursuing an innovation strategy, employer prominence has the
expected positive and statistically significant effect on the odds of securing external financing at startup.
For innovative ventures, where the quality of theventure team is arguably of greatest importance,
employer prominence has a significant effect on the ability of the founders to secure resources from
external providers.
13
(Separate analyses (not shown) using the Silicon Valley 100 measures discussed
above had no influence on the pattern of results.) This supports our claimthat the reputational benefits of
employer prominence reduces the perceived uncertainty of new ventures and facilitates entrepreneurial
activity.
Discussion and Conclusion
The analyses presented in this paper provide evidence supporting the claim that career histories shape the
entrepreneurial process. First, functional and educational backgrounds influence initial strategic choices,
and management experience is important to external stakeholders. Entrepreneurs with advanced degrees
establish firms with innovation strategies, but entrepreneurs with sales or finance experience are less
likely to pursue an innovation strategy. Entrepreneurs with senior management experience havemore
legitimacy with external constituents and are more likely to obtain external financing. These findings are
consistent with work on human capital and the importance of career histories on the formation of new
ventures. Our work moves beyond these findings, however, to address the importance of social capital
13
Arguably, the firms with the greatest uncertainty surrounding their quality are innovation strategists without a
product at the time of founding. This suggests a three-way interaction between strategy, product at founding and
employer prominence. We tested for this interaction in a separate model, not shown here. Employer prominence
has no significant effect for innovators with a product, but does have a significant effect for innovators withou t a
product. This is consistent with our argument.
Coming from Good Stock: Career Histories and New Venture Formation
29
for entrepreneurs. Entrepreneurs setting out from prominent employers have both information and
reputation advantages over those who emanate from less prominent firms. It is important to note n ot only
what experiences and background entrepreneurs have but also where these experiences come from. The
information and reputation advantages that accrue from social capital allow entrepreneurs from prominent
firms to pursue more risky ventures, such as founding a firm dedicated to establishing a new product or
market. The reputational capital derived from being affiliated with a prominent employer also allows
entrepreneurs to reduce the perceived uncertainty of their venture, thereby facilitating the acquisition of
resources from third parties. Risky ventures (those pursuing an innovation strategy) that emerge from
prominent employers are more likely to obtain external financing.
While we believe our analyses are persuasive, they are limited in certain respects. First, our data do not
allow us to distinguish between a desire to launch a new venture pursuing an innovation strategy, and the
ability to do so. This makes it difficult to specify clearly the mechanism by which employer prominence
influences the choice of initial strategy. Specifically, we cannot confidently determine whether
individuals from prominent employers are more likely to launch innovative ventures because they are
privy to superior information, or because they benefit from t he prominence of their employers in
convincing third parties to support the venture. Distinguishing between these accounts would require a
more detailed study of proposed entrepreneurial ventures and the process by which they move from initial
concepts to nascent firms. Despite this limitation, what we do know is important: entrepreneurs from
prominent employers launch more innovative ventures, and those ventures are more likely to obtain
external financing.
Second, while our interpretation of these results emphasizes the benefits of prominent structural locations,
we are sensitive to alternative explanations that point to the possible effects of unobserved heterogeneity
among founders. It is possible that the observed effects of entrepreneurial prominence are due to
unobserved characteristics of established firms and the employees they attract. For example,
entrepreneurially prominent firms may attract employees whose personal characteristics make them
Coming from Good Stock: Career Histories and New Venture Formation
30
particularly likely both to pursue innovativeventures and to win the confidence of external investors. As
with any such claim, we cannot rule out with certainty that the findings can be attributed to unobserved
heterogeneity. However, we feel confident that we have measured and controlled for seve ral of the most
important individual-level characteristics that can most plausibly be thought to affect the outcomes we
examine. Our models include measures of the patenting activities of the founders, their educational
backgrounds, their prior work experiences and their past entrepreneurial activity. Moreover, we have no
a priori reasons to expect that the firms identified as entrepreneurially prominent in this sample should
differ systematically in their recruitment behavior. At the same time, we beli eve that an important and
promising line of future research would be to explain why firms differ in the rate at which they generate
new ventures through employee departures. The limited amount of work that has been done in this area
suggests that such variations can be traced to differences in internal promotion chances, reward levels,
technological emphases and managerial practices (Freeman 1986; Brittain and Freeman 1986). A full
understanding of how established firms shape entrepreneurial behavior must attend to both the cause and
the consequences of entrepreneurial prominence.
Finally, the diversity of firms in the SPEC data set, although useful for understanding a broad set of
organizations, has certain shortcomings. Ideally, in addition to our emplo yer prominence measure, we
would have an exogenous measure of the prominence of past employers. The broad set of industries
represented in the sample make such a measure difficult to generate. For studies of new ventures within a
single industry, measures of technological or innovative prominence may be appropriate. Stuart, Hoang
and Hybels (1999), for example, measure the prominence of alliance partners using counts of citations to
a firm’s patent portfolio. The development of exogenous measures of prominence requires confidence
about the criteria by which members of the entrepreneurial community rank existing firms. To our
knowledge, this topic is unexplored in the existing literature. Furthermore, an exogenous measurement of
prominence has its own problems. No clear dimension exists on which we could compare the prominence
Coming from Good Stock: Career Histories and New Venture Formation
31
of a biotechnology firm with the prominence of a hardware firm. As such, single industry studies may be
more appropriate places to develop exogenous measures of prominence.
We began this paper by arguing that the landscape of existing firms shapes the entrepreneurial process.
We believe our results demonstrate that patterns of entrepreneurial activity are shaped by the social
structure of existing organizations (Stinchcombe 1965; Aldrich and Zimmer 1986). Our work is,
therefore, an important complement to studies showing how the general scarcity of resources affects the
formation of new firms (Hannan and Freeman 1989). In particular, these findings stress the importance of
hierarchical differentiation in the social structure of organizational populations (Podolny 1993; Podolny,
Stuart and Hannan 1996; Stuart, Hoang and Hybels 1999). We know from this existing work that
prominence dictates future patterns of affiliation, firm sur vival, and performance. We find additional
benefits accruing from prominence: firms emanating from prominent firms are more innovative.
For network theorists, our work further confirms the importance of network position. Entrepreneurs with
prominent past employers occupy a privileged place in the social structure, and their position garners
important advantages with respect to access to resources and information. What we add to the network
literature is an examination of how the network of existing organizations impacts the new venture, and by
extension the new venture network. The new ventures that spawn from prominent employers may occupy
a more prominent position in their own network. The access to ex ternal funding immediately connects
these innovative new ventures into an exclusive network of organizations. The innovative strategies of
these firms may lead them to higher visibility in their own industries. And the fact that they emerge from
prominent others may imprint them with positional advantage from the very beginning (Stinchcombe
1965). Prominence may not only be fairly stable over time, it may transfer from one organization to the
other through entrepreneurs. Furthermore, the impact of the new venture’s lineage may have implications
far beyond founding. These possibilities offer intriguing directions for future research.
Coming from Good Stock: Career Histories and New Venture Formation
32
We noted earlier that organizational researchers have grown increasingly interested in the role managerial
careers play in shaping organizational behavior and industry dynamics. Most research in this tradition
focuses on how career histories shape individual experiences and abilities (Boeker 1997; Sørensen 199 9).
Our research emphasizes that careers have important reputational consequences as well. In this respect,
the identity of a person’s employers (and perhaps other institutional affiliations) assumes primary
significance. Organizational reputations transfer to individual reputations. Inferences about the talents
and abilities of individuals are constructed from their histories of affiliation with employers. This
parallels studies of scientific careers, which have documented that the prestige of the university a person
attended has a positive effect on the prestige of the first job (Hurlbert and Rosenfeld 1992). Our results
suggest, however, that the effects of institutional or organizational prestige extend beyond the signals
associated with educational credentials and encompass the firms and other organizations that people move
through in the course of their careers. Moreover, the effects of institutional prestige extend beyond their
impact on individual life courses. The role that hierarchical differentiation among organizations plays in
both individual career dynamics and organizational populations, and the interconnections between the two
levels of analysis, is an important arena for future research.
Future research should examine other benefits of entrepreneurial prominence. We find a link between
prominence and innovative strategies and external funding, but prominent past emp loyers may continue to
impact internal organizational decisions through means like the recruitment of personnel from prominent
firms. Ventures spawned from prominent employers may be more likely to go public successfully, or
they may be more likely to be acquired by a larger, more established firm attempting to increase their own
prominence. These various research possibilities point out how disentangling where imprinting ends and
path dependence begins offers a challenge to futureresearch.
Our research contributes to a greater understanding of what differentiates new ventures. In order to
understand the emergence of innovative new ventures, we need to know where they come from in the
network of existing organizations. Past employer prominence offers firms a significant advantage in the
Coming from Good Stock: Career Histories and New Venture Formation
33
struggle for survival and success. We tie new ventures into the existing social structure and point out that
a new venture is more than a compilation of skills and experiences, but it emerges from other
organizations with positions in the social structure. Without incorporating the existing social structure
into our understanding of new ventures, we cannot hope to understand why one venture survives and
another fails, much less why the occasional venture succeeds beyond all expectations. Despite the rapid
rate of new venture formation, the ever changing technology, and the considerable hurdles new ventures
face, the underlying stability of the social structure offers a means to understand and keep up with the
changing organizational landscape.
Coming from Good Stock: Career Histories and New Venture Formation
(34)
References
Aldrich, Howard and C. Marlene Fiol. 1994. "Fools Rush In? The Institutional Context of Industry
Creation." Academy of Management Review, 19, 4: 645-670
Aldrich, Howard, and Catherine Zimmer. 1986. “Entrepreneurship Through Social Networks.” Pp. 3-23
in Donald L. Sexton and Raymond W. Smilor (eds.), The Art and Science of Entrepreneurship,
Cambridge, MA: Ballinger.
Baron, James N., M. Diane Burton, and Michael T. Hannan. 1996. “The Road Taken: Origins and
Evolution of Employment Systems in Emerging Companies,” Industrial and Corporate Change
5(2):239-275
Baty, Gordon B., William M. Evan and Terry W. Rothermel. 1971. “Personnel Flows as
Interorganizational Relations.” Administrative Science Quarterly 16:430-443.
Benjamin, Beth A. and Joel M. Podolny. 1999. “Status, Quality, and Social Order in the California Wine
Industry.” Administrative Science Quarterly 44(3):563-589.
Bhide, Amar. 1999. The Origin and Evolution of New Businesses. New York: Oxford University Press.
Blau, Peter M. 1964. Exchange and Power in Social Life. New York: Wiley.
Boeker, Warren 1988. “Organizational Origins: Entrepreneurial and Environmental Imprinting at the
Time of Founding.” Pp. 33-51 in Glenn R. Carroll (ed.) Ecological Models of Organizations. Cambridge,
MA: Ballinger.
Boeker, Warren. 1997. “Executive Migration and Strategic Change: The Effect of Top Manager
Movement on Product Entry.” Administrative Science Quarterly 42: 213-236.
Brittain, Jack W. and John Freeman. 1986. “Entrepreneurship in the Semiconductor Industry.” Mimeo.
University of California, Berkeley.
Coming from Good Stock: Career Histories and New Venture Formation
(35)
Brockhaus, Robert H., Sr., and Pamela S. Horwitz. 1986. “The Psychology of the Entrepreneur.” Pp.
25-48 in Donald L. Sexton and Raymond W. Smilor (eds.), The Art and Science of Entrepreneurship,
Cambridge, MA: Ballinger.
Burt, Ronald S. 1992. Structural Holes. Cambridge, MA: Harvard University Press.
Burton, M. Diane. 1995. The Emergence and Evolution of Employment Systems in High Technology
Firms. Unpublished Ph.D. Dissertation, Department of Sociology, Stanford University.
Chandler, Gaylen N. 1996. “Business Similarity as a Moderator of the Relationship between Pre-
Ownership Experience and Venture Performance.” Entrepreneurship Theory and Practice 20: 51-65.
Chandler, Gaylen N., and Steven N. Hanks. 1994. “Founder Competence, the Environmen t, and Venture
Performance.” Entrepreneurship Theory and Practice 18: 77-89.
Chandler, Gaylen N., and Erik J. Jansen. 1992. “Founders Self-Assessed Competence and Venture
Performance.” Journal of Business Venturing 7: 223-236.
Cohen, Wesley M. and Richard C. Levin. 1989. “Empirical Studies of Innovation and Market Structure.”
Pp. 1060-1107 in R. Schmalensee and R.D. Willig (eds.), Handbook of Industrial Organization, Volume
III. New York: Elsevier Science Publishers.
Cooper, Arnold C. 1985. “The Role of Incubator Organizations in the Founding of Growth-Oriented
Firms.” Journal of Business Venturing 1:75-86.
Cooper, Arnold C., F. Javier Gimeno Gascon, and Carolyn Y. Woo. 1994. “Initial Human and Financial
Capital as Predictors of New Venture Performance.” Journal of Business Venturing 9: 371-396.
Cooper, Arnold C., and W.C. Dunkelberg. 1986. “Entrepreneurship and Paths to Business Ownership.”
Strategic Management Journal 7: 53-68.
Coming from Good Stock: Career Histories and New Venture Formation
(36)
Freeman, John. 1986. “Entrepreneurs as Organizational Products: Semiconductor Firms and Venture
Capital Firms.” Advances in the Study of Entrepreneurship, Innovation, and Economic Growth 1: 33-52.
Gompers, Paul A., and Joshua Lerner. 1997. "Money Chasing Deals?: The Impact of Fund Inflows on
Private Equity Valuations" Harvard Business School Working Paper.
Goode, William J. 1978. The Celebration of Heroes: Prestige as a Social Control System. Berkeley, CA:
University of California Press.
Goslin, L., and B. Barge. 1986. “Entrepreneurial Qualities Considered i n Venture Capital Support.” In
R. Ronstadt, J.A. Hornaday, R. Petersen, and K.H. Vesper (eds.), Frontiers in Entrepreneurship
Research. Wellesley, MA: Babson College.
Granovetter, Mark S. 1973. "The Strength of Weak Ties." American Journal of Sociology 78:1360-1380.
Gunz, Hugh P. and R. Michael Jalland. 1996. Managerial careers and business strategies. Academy of
Management Review, 21(3):718-756.
Hall, John, and Charles W. Hofer. 1993. “Venture Capitalists Decision Criteria in New Venture
Evaluation.” Journal of Business Venturing 8: 25-42.
Hannan, Michael T., M. Diane Burton and James N. Baron. 1996 “Inertia and Change in the Early Years:
Employment Relations in Young, High-Technology Firms.” Industrial and Corporate Change 5(2):503-
536.
Hannan, Michael T., and John Freeman. 1977. “The Population Ecology of Organizations.” American
Journal of Sociology 82: 929-964.
———. 1989. Organizational Ecology. Cambridge, MA: Harvard University Press.
Hellmann, Thomas F. and Manju Puri. 1999. “The Interaction between Product Market and Financing
Strategy: The Role of Venture Capital.” Stanford Graduate School of Business Working Paper #1561.
Coming from Good Stock: Career Histories and New Venture Formation
(37)
Herron, Lanny, and Richard B. Robinson, Jr. 1993. “A Structural Model of the Effects of
Entrepreneurial Characteristics on Venture Performance.” Journal of Business Venturing 8: 281-294.
Higgins, Monica C. and Ranjay Gulati. 1999. “The Effects of IPO Team Ties on Investment Bank
Affiliation and IPO Success.” Harvard Business School Working Paper #00-025.
Hiltzik, Michael. 1999. Dealers of Lightning: Xerox PARC and the Dawn of the Computer Age. New
York, NY: Harper Business.
Hurlbert, Jeanne S., and Rachel A. Rosenfeld. 1992. “Getting a Good Job: Rank and Institutional
Prestige in Academic Psychologists'Careers.” Sociology of Education 65:188-208.
Jaffee, Adam B., Manuel Trajtenberg and Rebecca Henderson. 1993. “Geographic Localization of
Knowledge Spillovers as Evidenced by Patent Citations.” Quarterly Journal of Economics. 108:577-
598.
Jones-Evans, Dylan. 1996. “Experience and Entrepreneurship: Technology-Based Owner-Managers in
the UK.” New Technology, Work and Employment 11: 39-54.
Kalleberg, Arne L., David Knoke, Peter V. Marsden and Joe L. Spaeth. 1996. Organizations in America:
Analyzing their Structures and Human Resource Practices. Thousand Oaks, CA: Sage.
Kaplan, Jerry. 1995. Startup: A Silicon Valley Adventure Story. New York: Houghton-Mifflin.
Knoke, David and Ronald S. Burt. 1983 "Prominence." Pp. 195-222 in Ronald S. Burt and Michael J.
Minor (eds.), Applied Network Analysis. Beverly Hills: Sage Publications.
Latour, Bruno. 1987. Science in Action. Cambridge, MA: Harvard University Press.
Lerner, Joshua. 1995. “Venture Capitalists and the Oversight of Private Firms." Journal of Finance 50:
301-18.
Coming from Good Stock: Career Histories and New Venture Formation
(38)
Light, Ivan. 1972. Ethnic Enterprise in America: Business and Welfare among Chinese, Japanese and
Blacks. Berkeley: University of California Press.
Maidique, Modesto and Peter Patch. 1982. “Corporate Strategy and Technological Policy.” In M.
Tushman and W. Moore (Eds.), Readings in the Management of Innovation: 273-285. Marshfield, MA:
Pitman.
MacMillan, Ian C., Siegel, R., SubbaNarishma, P.N. 1985. “Criteria Used by Venture Capitalists to
Evaluate New Venture Proposals.” Journal of Business Venturing 1: 119-128.
McPherson, J. Miller. 1982. “Hypernetwork Sampling: Duality and Differentiation Among Voluntary
Associations.” Social Networks 3: 225-249.
Miles, R. E. and C.C. Snow. 1978. Organizational Strategy, Structure and Process. New York:
McGraw Hill Book Co.
Miller, Danny. 1986. Configurations of strategy and structure: Towards a synthesis. Strategic
Management Journal, 7(3):233-249.
Merton, Robert C. 1968 [1973] "The Matthew Effect in Science." Science 159: 56-63.
Pfeffer, Jeffrey, and Huseyin Leblebici. 1973. “Executive Recruitment and the Development of Interfirm
Organizations.” Administrative Science Quarterly 18:449-461.
Podolny, Joel M. 1993. "A status-based model of market competition." American Journal of Sociology,
98: 829-872.
Podolny, Joel M., and Toby E. Stuart. 1995 “A role-based ecology of technological change.” American
Journal of Sociology 100: 1224-1260.
Coming from Good Stock: Career Histories and New Venture Formation
(39)
Podolny, Joel M., Toby E. Stuart, and Michael T. Hannan. 1996. “Networks, Knowledge, and Niches:
Competition in the Worldwide Semiconductor Industry, 1984 -1991.” American Journal of Sociology
102: 659-689.
Porter, Michael E. 1980. Competitive Strategy. New York: Free Press.
Roberts, Michael J. and Howard H. Stevenson. 1991. “Alternative Sources of Financing.” Pp. 171-179
in William A. Sahlman and Howard H. Stevenson (Eds.) The Entrepreneurial Venture. Boston: Harvard
Business School Press.
Robinson, Peter B. and Edwin A. Sexton (1994) “The Effect of Education and Experience on Self-
Employment Success.” Journal of Business Venturing 9:141-156.
Rogers, Everett M. and Judith K. Larsen. 1984. Silicon Valley Fever: Growth of High Technology
Culture. New York: Basic Books.
Saxenian, Annalee. 1996. Regional Advantage : Culture and Competition in Silicon Valley and Route
128. Cambridge, MA: Harvard University Press.
Schumpeter, Joseph A. 1934. The Theory of Economic Development. Cambridge, MA: Harvard.
Shane, Scott. 1999. “Prior Knowledge and the Discovery of Entrepreneuri al Opportunities.”
Organization Science. Forthcoming.
Shane, Scott and Rakesh Khurana. 1999. “Career Experiences and Firm Foundings.” Working Paper.
Sloan School of Management, Massachusetts Institute of Technology.
Sørensen, Jesper B. 1999. “Executive Migration and Interorganizational Competition.” Social Science
Research 28: 289-315.
Sørensen, Jesper B. and Toby E. Stuart. 1999. “Aging, Obsolescence and Organizational Innovation.”
Administrative Science Quarterly, forthcoming.
Coming from Good Stock: Career Histories and New Venture Formation
(40)
Spence, A. Michael. 1974. Market Signaling. Cambridge, MA: Harvard University Press.
Stinchcombe, Arthur. 1965. “Social Structure and Organizations.” Pp. 142-193 in James G. March
(ed.), Handbook of Organizations. Chicago: Rand McNally.
Stuart, Toby E., Ha Hoang and Ralph C. Hybels. 1999. “Interorganizational Endorsements and the
Performance of Entrepreneurial Ventures.” Administrative Science Quarterly 44:315-349.
Stuart, Toby E. and Olav Sorenson. 1999. “The Geography of Opportunity: Spatial Heterogeneity in
Founding Rates and the Performance of Biotechnology Firms.” University of Chicago Graduate School
of Business Working Paper.
Sudman, Seymour. 1976. Applied Sampling. New York: Academic Press.
Tushman, Michael L. and Philip Anderson. 1986. “Technological Discontinuities and Organizational
Environments.” Administrative Science Quarterly 31:439-65.
Coming from Good Stock: Career Histories and New Venture Formation
(41)
Table 1. Top 24 Entrepreneurially Prominent Firms in SPEC Sample
Firm # of SPEC Firms
IBM 21
Hewlett-Packard 16
Stanford Univ. 15
Apple 12
Intel 11
National Semiconductor 10
ROLM Corporation 9
AT&T 8
Sun Microsystems 6
UC Berkeley 5
Silicon Graphics 5
Ungermann Bass 5
AMD 4
Digital Equipment Corp 4
MIPS Computer Systems 4
NASA AMES 4
NIH 4
Xerox Corp. 4
Bridge Communications, Inc. 3
Chips and Technologies 3
Control Data Corporation 3
Texas Instruments Incorporated 3
U.S Navy and Naval Reserve 3
Harvard University 3
Note: SPEC firms can have multiple “parents.” For example, a SPEC firm may have one founder who
worked at IBM, HP, and GE and a second founder who worked at HP and Honeywell. The career
backgrounds are aggregated at the firm level; thus this SPEC firm has prior ties to four firms (IBM, HP,
GE, and Honeywell) and wouldbe listed in this table as coming from both IBM andHP.
Coming from Good Stock: Career Histories and New Venture Formation
(42)
Table 2. Descriptive Statistics
Variable Mean SD Min Max N
Innovation Strategist
0.491 0.501 0 1 159
External Financing within 3 mos.
0.370 0.484 0 1 138
Product within 6 mos.
0.185 0.389 0 1 168
Medical-related Industry
0.143 0.351 0 1 168
Networking & Telecom. Industry
0.202 0.403 0 1 168
Semiconductor Industry
0.107 0.310 0 1 168
Prior Founding Experience
0.560 0.860 0 5 168
Senior Management Experience
0.935 1.079 0 5 168
Finance or Sales Experience
1.185 1.434 0 9 168
Log Number of Patents
0.698 0.967 0 3.85 166
Advanced Degrees
0.784 0.355 0 1 151
Number of Past Employers
3.440 2.107 0 11 168
Ties to Prominent Firms
9.595 10.860 0 52 168
Coming from Good Stock: Career Histories and New Venture Formation
(43)
Table 3. Relationship between Strategy and External Financing
External Financing at Founding
Strategy: No Yes
Incremental 31 29 60
52% 48%
Innovation 51 20 71
72% 28%
Total 82 49 131
?
2
= 5.65 (1 d.f.), p< 0.02
Coming from Good Stock: Career Histories and New Venture Formation
(44)
Table 4. Logistic Regression Estimates of the Determinants of Founding Innovation Strategy (N=159)
Variable (1) (2) (3) (4)
Medical-related Industry 1.745** 1.255 1.132 1.399
(0.591) (0.641) (0.670) (0.709)
Networking and -0.420 -0.522 -0.621 -0.881
Telecom. Industry
(0.419) (0.448) (0.480) (0.523)
Semiconductor Industry 0.187 -0.215 -0.343 -0.328
(0.545) (0.632) (0.682) (0.746)
Log Number of Patents 0.418 0.453 0.372
(0.217) (0.234) (0.243)
Graduate Degrees 1.429* 1.525** 1.384*
(0.553) (0.569) (0.619)
Prior Founding Experience 0.185 0.183
(0.234) (0.248)
Senior Management 0.026 -0.135
Experience
(0.201) (0.226)
Sales or Finance -0.444** -0.556**
Experience
(0.165) (0.175)
Number of Past Employers 0.164
(0.136)
Employer Prominence 0.046*
(0.023)
Constant
N 159 144 144 144
Log-Likelihood -103.03 -86.586 -81.537 -75.453
Pseudo R-squared .065 .131 .182 .243
Coming from Good Stock: Career Histories and New Venture Formation
(45)
Table 5. Logistic Regression Estimates of the Determinants of External Financing at Start -up (N=128)
Variable (1) (2) (3) (4) (5) (6)
Medical-related -0.568 -0.572 -0.197 -0.309 -0.236 -0.066
Industry
(0.586) (0.591) (0.628) (0.654) (0.701) (0.720)
Networking and 0.581 0.404 0.525 0.373 0.587 0.300
Telecom. Industry
(0.461) (0.476) (0.498) (0.515) (0.538) (0.568)
Semiconductor -0.268 -0.235 0.289 0.125 -0.062 -0.209
Industry
(0.663) (0.674) (0.730) (0.755) (0.784) (0.811)
Innovation -0.469 -1.097 -1.488* -1.568* -1.697* -2.009*
Strategist
(0.527) (0.670) (0.746) (0.767) (0.789) (0.816)
Product -0.705 -0.993* -1.018* -1.079* -0.940 -2.286**
(0.399) (0.436) (0.479) (0.517) (0.533) (0.780)
Product * Innovation 1.891 2.252 2.695* 3.025* 3.201*
Strategy
(1.129) (1.182) (1.231) (1.284) (1.306)
Graduate Degrees 0.240 0.262 0.261 0.455
(0.629) (0.636) (0.655) (0.695)
Log Number of -0.403 -0.363 -0.338 -0.318
Patents
(0.242) (0.245) (0.249) (0.256)
Prior Founding -0.058 -0.023 -0.242
Experience
(0.285) (0.289) (0.314)
Senior Management 0.350 0.513* 0.545*
Experience
(0.214) (0.235) (0.247)
Sales or Finance -0.072 -0.009 0.025
Experience
(0.157) (0.161) (0.176)
Number of Past -0.235 -0.226
Employers
(0.144) (0.149)
Employer Prominence 0.006 -0.072
(0.022) (0.042)
Innovation Strategy 0.115*
* Prominence
(0.048)
Constant -0.145 0.068 -0.282 -0.116 -0.155 0.432
(0.357) (0.375) (0.454) (0.474) (0.521) (0.596)
N 129 129 121 121 121 121
Log-Likelihood -80.408 -79.022 -71.417 -69.964 -68.350 -64.817
Pseudo R-squared .056 .072 .1 .118 .138 .183
* p < 0.05 ** p < 0.01 (two-sided tests)
Coming from Good Stock: Career Histories and New Venture Formation
(46)
Appendix A: Bivariate Correlations
1 2 3 4 5 6 7 8 9 10 11 12 13
1 Innovation Strategist
1.00
2 External Financing
within 3 mos.
-0.19* 1.00
3 Product within 6 mos.
-0.25* -0.05 1.00
4 Medical-related
Industry
0.28* -0.11 -0.11 1.00
5 Networking & Telecom.
Industry
-0.14 0.18* -0.01 -0.21* 1.00
6 Semiconductor Industry
0.01 -0.03 -0.07 -0.14 -0.17* 1.00
7 Log Number of Patents
0.24* -0.14 -0.19* 0.32* -0.11 0.19* 1.00
8 Graduate Degrees
0.27* -0.04 -0.18* 0.09 -0.08 0.13 0.18* 1.00
9 Prior Founding
Experience
0.06 -0.01 -0.10 0.11 0.05 0.00 0.05 0.05 1.00
10 Senior Management
Experience
-0.03 0.15 -0.11 0.02 0.18* 0.00 -0.04 0.01 0.39* 1.00
11 Finance or Sales
Experience
-0.18* 0.03 -0.05 -0.05 0.04 -0.02 -0.01 0.00 0.19* 0.39* 1.00
12 Number of Employers
0.21* -0.07 -0.15 0.11 0.25* 0.01 0.14 0.08 0.31* 0.48* 0.36* 1.00
13 Employer Prominence
0.24* -0.05 -0.01 -0.14 0.11 0.10 0.07 0.15 0.04 0.14 0.17* 0.52* 1.00
doc_308279261.pdf
During this brief illustration with regards to coming from good stock career histories and new venture.
Coming from Good Stock:
Career Histories and New Venture Formation
*
M. Diane Burton
Sloan School of Management
Massachusetts Institute of Technology
Cambridge, MA 02142
(617) 253-5539
[email protected]
Jesper B. Sørensen
Sloan School of Management
Massachusetts Institute of Technology
Cambridge, MA 02142
(617) 253-7945
[email protected]
Christine M. Beckman
Graduate School of Management
University of California, Irvine
Irvine, CA 92697
(949) 824-3983
[email protected]
July 2001
*
This research was supported by the Division of Research of Harvard Business School, the Stanford Graduate
School of Business, the University of Chicago Graduate School of Business, the Alfred P. Sloan Foundation, and the
Kauffman Center for Entrepreneurial Leadership. Early versions of this paper were presented at the Academy of
Management, San Diego, CA, August 1998 and at the annual meeting of the American Sociological Association,
Chicago, IL, August 1999. We would like to thank Howard Aldrich, Rakesh Khurana, Josh Lerner, Nitin Nohria,
Damon Phillips, Julio Rotemberg, Bill Sahlman, Scott Shane, Olav Sorenson, Toby Stuart, and Mike Tushman for
helpful ideas and comments on early drafts. We are indebted to Mike Hannan and Jim Baron for their role in the
Stanford Project on Emerging Companies. Stephanie Woerner provided assistance of many types. Finally, we
thank Thomas Hellmann and Manju Puri for sharing their financing data.
Coming from Good Stock: Career Histories and New Venture Formation
Abstract
We examine how the social structure of existing organizations influences the entrepreneurial process and
suggest that resources accrue to entrepreneurs based on the structural position of their prior employers.
We argue that information advantages allow individuals from entrepreneurially prominent prior firms to
identify new opportunities. Entrepreneurial prominence also provides reputation benefits that reduce the
perceived uncertainty of a new venture in the eyes of external constituents. Using a sample of Silicon
Valley start-ups, we demonstrate that entrepreneurial prominence is associated with initial strategy and
the probability of attracting external financing. New ventures with high prominence are more likely to be
innovators; furthermore, innovators with high prominence are more likely to obtain financing.
Introduction
It has often been noted that some of the most radically innovative products and technologies are
developed and commercialized not by existing companies, but rather by entrepreneurial ventures
(Schumpeter 1934; Tushman and Anderson 1986; Henderson and Clark 1990). This is a remarkable fact,
given that creating a new organization requires the mobilization of a substantial array of social and
material resources (Stinchcombe 1965). These resource mobilization tasks are simplified when
entrepreneurs choose to focus on proven, established products and technologies. By contrast,
organizations devoted to new products and technologies face severe hurdles. The entrepreneur must not
only come up with a new idea, but also overcome the skepticism of resource providers, since the
uncertainty and risk associated with any new venture is particularly heightened when the underlying
product or technology is unproven (Aldrich and Fiol 1994).
Where do innovative new ventures come from? The simple answer is that they emerge – directly and
indirectly -- from established firms (Freeman 1986). Contrary to popular myths of lone college drop-out
entrepreneurs, most new ventures are founded by people with prior employment experience (Cooper
1985; Robinson and Sexton 1994). In this sense, entrepreneurs are organizational products:
“Organizations create their own competition by providing the skills and background that provide
credibility for the entrepreneur. They provide the knowledge of opportunity by placing that person in a
position to know about unserved or badly served markets” (Freeman 1986: 39).
In this paper we argue that entrepreneurial opportunities and resources accrue to incipient entrepreneurs
as a function of the structural position of their prior employers. Much as geographical regions differ in
their rates of entrepreneurial activity, established firms differ markedly along this dimension. Some firms
are entrepreneurial hotbeds, as perhaps most famously exemplified by Fairchild Semiconductor, founded
in 1957. Fairchild spawned ten new ventures in its first eight years; moreover, most of the thirty-one
semiconductor firms founded in Silicon Valley in the 1960s could trace their lineage to Fairchild
Coming from Good Stock: Career Histories and New Venture Formation
2
(Saxenian 1994; Rogers and Larsen 1984). Examples of such “Fairchildren” include Intel, Advanced
Microdevices and LSI Logic. Other firms give rise to relatively few, if any, new ventures.
1
In this paper, we argue that one consequence of these differences in rates of entrepreneurial activity
among established firms is to influence the visibility of established firms in the entrepreneurial
community. We use the term entrepreneurial prominence to describe these differences in visibility. We
consider established firms that spawn a large number of new ventures through employee departuresto be
more entrepreneurially prominent than those that do not. Our core contention is that innovative new
ventures are more likely to emerge from established firms that are entrepreneurially prominent. We
expect this for two reasons. First, we believe that there are important informational and resource benefits
to being affiliated with a prominent firm. Second, we argue that there are substantial reputational benefits
that accrue to employees of prominent firms, and that these benefits play a crucial role in reducing the
perceived uncertainty surrounding a venture. We conceive of the entrepreneurial prominence of
established firms as a form of social capital that is transferred to employees as they leave the firm and
attempt to launch new ventures. We explore how differences in the entrepreneurial prominence of
established firms affects the characteristics of entrepreneurial behavior. Thus, while established firms
may have difficulty innovating (Sørensen and Stuart 1999) and commercializing new innovations (Hiltzik
1999), they play an important passive role in shaping the emergence of entrepreneurial ventures.
Our paper is related to and draws on a number of streams of theory and research. First, there is a long
sociological interest in the emergence of new organizations. In fact, one of the early propositions put
forward by Stinchcombe is that “the probability that a man or group of men will be motivated to start an
organization is dependent on the social structure and the position of men within it.” (1965:147).
1
Note that we do not seek to explain why some firms generate more entrepreneurial offspring than others; rather, we
take this distribution as given. See Freeman (1986) and Brittain and Freeman (1986) for a discussion of these issues.
Coming from Good Stock: Career Histories and New Venture Formation
3
Furthermore, sociologists have argued for some time that the dynamics of new venture formation depend
critically on the distribution of opportunities and resources through social structure (Light 1972; Hannan
and Freeman 1977; Aldrich and Zimmer 1986). However, most efforts to examine organizational
emergence come from the ecological tradition where there is keen attention to the broad structural
characteristics of firms, but little attempt to link individuals to this social structure. At the same time,
attempts to link fixed individual attributes, such as psychological characteristics, to entrepreneurial
activity have met with limited success (Brockhaus and Horwitz 1986; Herron and Robinson 1993). Our
paper explicitly reconnects individuals and organizations with a distinctly sociological approach that
allows us to explore whether some individuals, by virtue of their location in the social structure of
existing organizations, are better able to form highly uncertain, innovative start-ups.
Second, there is a growing interest in the economic and sociological literature on technological innovation
in the operation of geographical “spillover” effects (Jaffe, Trajtenberg and Henderson 1993; Saxenian
1994; Stuart and Sorenson 1999). These studies examine how horizontal differentiation along a
geographic dimension affects the innovative activity of established firms and the emergence of new
ventures. For example, Jaffe, Trajtenberg and Henderson (1993) find that the patenting rates of
established firms are positively influenced by the patenting activity of other firms in the same
metropolitan area. Saxenian (1994) argues that the distinctive organizational arrangements and cultures
in Silicon Valley are a source of the regions’ high levels of entrepreneurial activity. Stuart and Sorenson
(1999), studying the biotechnology industry, argue that new firms are more likely to emerge in regions
that have a high density of established biotechnology activity. Similarly, there is a long-standing interest
in understanding how horizontal differentiation among industries leads to differential rates of
entrepreneurship and innovation (e.g., Cohen and Levin 1989). Our approach differs from these
literatures in that we do not seek to understand regional or industrial differences in entrepreneurial
activity; in fact, we focus on a single region, Silicon Valley and a limited number of high-technology
industries. Instead, we investigate the consequences of vertical differentiation among firms in terms of
Coming from Good Stock: Career Histories and New Venture Formation
4
status, prominence, or visibility (Benjamin and Podolny 1999; Podolny 1993; Stuart, Hoang and Hybels
1999).
Third, students of organizations have become increasingly interested in how the movement of individuals
between organizations shapes organizational behavior and industry dynamics. Most studies of this
phenomenon start from the notion that managerial outlooks and predispositions are shaped by career
histories (Gunz and Jalland 1996), and that the movement of managers across firm boundaries is an
occasion for the diffusion of ideas and innovations (Baty, Evan and Rothermel 1971; Pfeffer and
Leblebici 1973; Boeker 1997; Sørensen 1999). Because managers in different firms have divergent
experiences, the pattern of movement between firms is an important determinant of industry dynamics.
Past studies of this process have focused on the movement of managers between existing organizations.
By contrast, we focus on those managers who leave their jobs to start new firms and thus shed light on the
effects of career trajectories on industry dynamics. Fourth, since we attach primary significance to the
career histories of entrepreneurs, our research is also related to work that examines how career histories
shape the entrepreneurial process (Brittain and Freeman 1986; Boeker 1988; Higgins and Gulati 1999;
Shane and Khurana 1999). Work histories are important determinants of the resources available to
entrepreneurs. Prior research on the effects of entrepreneur’s pre-ownership experiences tends to focus
on the types of work that entrepreneurs have performed in the course of their careers (Shane, 1999; Jones-
Evans 1996; Cooper and Dunkelberg 1986). By contrast, we emphasize where entrepreneurs worked
prior to founding the new venture; in other words, we focus on the identity of previous employers. By
doing so, we capitalize on the notion that careers situate the entrepreneur in a social structure of existing
firms that facilitates or constrains the flow of opportunities and resources. We ask how the social
structure of existing firms influences the entrepreneurial process.
This paper makes a number of contributions. First, in contrast to the studies of the relationship between
managerial mobility and organizational behavior cited above, we draw attention to the informational and
reputational benefits that may come from being associated with prominent employers. Second, by
Coming from Good Stock: Career Histories and New Venture Formation
5
focusing on where founders worked, we ask how the social structure of existing organizations influences
the entrepreneurial process. Finally, by differentiating new ventures according to their initial strategy, we
offer a more nuanced explanation of the entrepreneurial process.
In order to cast light on where innovative new ventures come from, we use a unique sample of Silicon
Valley startups and investigate the determinants of their initial strategies and financing. We focus on
explaining two characteristics of these ventures: 1) their founding strategies, specifically whether or not
they pursue an innovation strategy; and 2) the ability of the new ventures to attract external financing at
founding. We use the career histories of the founders, including the identities of past employers, to
examine how differences in the prominence of the established firms affects the strategy and financing of
new ventures.
The remainder of this paper is organized as follows. In the next section, we outline a series of arguments
linking the entrepreneurial prominence of established firms to the characteristics of new ventures. We
then discuss the characteristics of our data and the methods used to test our hypotheses. The presentation
of results is followed by a discussion section.
Entrepreneurial Prominence and Firm Advantage
Students of the link between career histories and entrepreneurship have examined how the accumulation
of human capital through career histories influences the formation of new ventures. In particular, scholars
have emphasized how job experiences shape the technical and managerial skills of (potential)
entrepreneurs. For example, Jones-Evans (1996) examines how the occupational backgrounds of
technological entrepreneurs affect the skills they bring to their ventures. He finds, for example, that
entrepreneurs from academic research settings have strong technical skills but low levels of managerial
competence. In a similar vein, Chandler (1996) examines how the past experiences of founders affect the
success of new ventures. He finds that new venture performance improves to the extent that there is
Coming from Good Stock: Career Histories and New Venture Formation
6
similarity between the task environment of the new venture and the task environment faced by the
entrepreneur in his or her previous job. Similarly, performance tends to improve to the extent that the
skills required in the new venture are similar to those previously developed (Chandler 1996; see also
Chandler and Jansen 1992; Cooper, Gimeno-Gascon and Woo 1994). Finally, Shane (1999) demonstrates
how differences in the past experiences of a set of entrepreneurs shapes their conceptions of the
opportunities associated with the same technological innovation.
As this brief review suggests, the main focus in past research on careers and entrepreneurship has been on
how careers shape the human capital available to entrepreneurs. However, the effects of careers on social
capital are neglected in existing research (Aldrich and Zimmer 1986). Entrepreneurial activity depends
on access to ideas and resources, and such access is differentially available to individuals occupying
different positions in the social structure (Burt 1992). One of the key determinants of an individual’s
position in social structure is her career history, in particular her affiliation with different employers.
Employers shape the personal networks of their employees, expose them to new ideas, endow them with
valuable resources and confer implicit credentials upon them. At the same time, established firms are
differentiated from each other both horizontally – by virtue of being engaged in different activities – and
vertically – by virtue of being more or less visible in different arenas. The nature of the resources
available to employees therefore typically will differ according to the structural position of the employer.
Therefore, to understand how the social structure of established firms affects entrepreneurial behavior, we
must consider the consequences of such differentiation among established firms.
We conceive of the social structure of existing firms as a set of positions hierarchically ordered according
to the prominence of their occupants. Network theorists suggest that an actor’s prominence in a social
network is a function of centrality – the extent to which the actor is extensively involved in relations with
other actors (Knoke and Burt 1983). Prominence garners both informational and reputational benefits for
the actor. In recent years, organizational sociologists have applied this notion of prominence to
understanding organizational behavior and industry dynamics (Podolny 1993; Podolny, Stuart and
Coming from Good Stock: Career Histories and New Venture Formation
7
Hannan 1996; Stuart, Hoang and Hybels 1999). In this paper we are particularly interested in the
landscape of existing firms as it relates to the generation of new firms; thus, we focus on the
entrepreneurial prominence of firms in the existing social structure. Established firms acquire
entrepreneurial prominence by virtue of their being tied to a relatively large number of new ventures.
Unlike strategic alliances (Stuart, Hoang and Hybels 1999) and many other types of interorganizational
ties, these ties generally are not created intentionally by established firms; rather they are formed by
virtue of employees leaving to found new ventures.
Entrepreneurs must be adept at executing two roles: 1) scanning the environment for opportunities and
devising strategies to take advantage of them; and 2) ensuring and managing the flow of resources —
such as capital, supplier relationships and customers — to the venture such that it may pursue its business
strategies successfully. Prior employment experience shapes the capabilities of entrepreneurs with respect
to these two roles. Our argument is that entrepreneurs benefit if they launch a venture from a prominent
position in this social structure. Employment by a prominent firm benefits the entrepreneur in two ways:
1) centrality in entrepreneurial networks makes it easier to identify entrepreneurial opportunities and to
act to exploit the opportunities; and 2) the prominence of prior employers helps to reduce the perceived
uncertainty of a new venture for external constituents.
Entrepreneurial opportunities arise when a prospective entrepreneur receives new information that, when
combined with knowledge already possessed, can be translated into something of value (Shane 1999). As
such, the potential that an opportunity will be discovered is related to both the stock of knowledge an
actor possesses and the flow of new information. This implies that human capital differences can only
partially explain the entrepreneurial process. Structural differences in access to new information must
also be considered.
Network theorists have demonstrated that the quantity and quality of the information an actor receives is a
direct function of the actors social network (e.g., Granovetter 1973; Burt 1992). Entrepreneurially
Coming from Good Stock: Career Histories and New Venture Formation
8
prominent firms, by virtue of their network centrality, will be exposed to stronger flows of new
information about technologies, emerging markets, and unmet customer needs. High quality information
will pass through prominent firms in high volume and at a fast rate; thus, employees have a higher
propensity to make the necessary information combinations and recognize opportunities. But recognizing
the opportunity is only the first step in creating a new venture. Prospective entrepreneurs must take action
to transform an opportunity into a venture. Here, we believe that nascent entrepreneurs in
entrepreneurially prominent firms vicariously benefit from the experiences of those entrepreneurs who
preceded them. As employees exit to launch new ventures, they likely deposit knowledge about the
appropriate steps and methods for building an enterprise with former colleagues and coworkers. First,
coworkers rarely immediately sever ties when colleagues change employment. Second, organizations
have memories residing in long-tenured employees about the actions and activities of former employees.
Thus, tactical knowledge of entrepreneurship – which law firms to call, which financiers to meet, where
to locate offices – becomes part of the stock and flow of information available to employees of
entrepreneurially prominent firms.
Although the information benefits of working for a prominent employer can help people to identify
entrepreneurial opportunities and allow them to take appropriate steps towards becoming an entrepreneur,
prospective founders still face substantial obstacles to launching the new venture. Entrepreneurs must
successfully mobilize the resources of wealth, power and legitimacy necessary to realize their vision
(Stinchcombe 1965; Aldrich and Zimmer 1986). Doing so requires overcoming information asymmetries
that make it difficult for external resource providers to assess the quality of a new venture or its founders
ex ante. These problems are exacerbated to the extent that a venture wishes to pursue a new, unproven
strategy. Under these conditions, external actors are likely to arrive at an estimate of the quality of the
venture by considering more easily observable attributes that are thought to be associated with the quality
of the venture (Stuart, Hoang and Hybels 1999; Podolny 1993; Spence 1974).
Coming from Good Stock: Career Histories and New Venture Formation
9
One source of information on the quality of the venture lies in the prior accomplishments of the founding
team members. Here, the career histories of entrepreneurs enter through a consideration of the
experiences and skills that have accumulated through the career. Studies suggest, for example, that
venture capitalists are particularly interested in the background experiences and managerial capabilities of
entrepreneurs (MacMillan, Siegel, and SubbaNarishma 1985; Goslin and Barge 1986; but see Hall and
Hofer 1993). Indicators of technological competence might include educational credentials and patents
held.
A second class of information on the quality of new ventures is reputational, and focuses on the identity
of the entrepreneurs themselves. Sociologists have long maintained that individual reputations are in part
constructed from the identities of the parties with whom a person associates (Blau 1964). In particular,
individual reputations benefit from association with prominent actors (Goode 1978). These reputational
advantages in turn facilitate the mobilization of resources and social action. Sociologists of science, for
example, argue that scientific careers are enhanced to the extent that young scholars are affiliated with
prominent individuals in the field (Merton 1968). Latour (1987) suggests that the reception accorded to
new ideas depends on the prominence of the scientist’s associates. Podolny and Stuart (1995) show that
other actors are drawn to innovations that are advanced by actors whose prior technological contributions
are perceived as important.
In a study of entrepreneurial ventures in biotechnology, Stuart, Hoang and Hybels (1999) demonstrate
that the prominence of a venture’s alliance partners is positively associated with its performance. In their
model, new ventures with prominent affiliates benefit from an implicit transfer of status from the
affiliates. In the eyes of third parties, association with high-status partners functions as a guarantee as to
the quality of the venture. Affiliation with prominent partners therefore gives firms an advantage in the
competition for customers, suppliers and employees.
Coming from Good Stock: Career Histories and New Venture Formation
10
We argue that the prominence of prior employers plays a similar role in reducing the perceived
uncertainty of a new venture. External actors use information on previous employers to make inferences
about the likelihood that the founders will build a successful venture. Third parties suffer from an
information asymmetry that makes it difficult to assess the true abilities of potential entrepreneurs ex
ante; this asymmetry is analogous to the one faced by employers when making employment decisions. In
this setting, founders who come from employers that are established incubators for entrepreneurial talent
benefit from this association. In other words, the prominence of previous employers may function as an
indicator of the quality of the prospective founder (Spence 1974). This supposition -- that employees of
prominent firms are, on average, of higher quality -- may indeed be correct. If more prominent firms are
also more successful, for example, they can devote greater resources to attracting and retaining skilled
personnel, and they may invest more in training. Furthermore, highly skilled individuals can be expected
to prefer employment with prominent employers, improving the pool of candidates for employment at
such firms. (Note that we are careful here not to argue that employers explicitly seek to certify their
employees; nor, that they have a reputational incentive to ensure that they hire only highly qualified
employees. Unlike alliance partners, who may put their reputations at risk when associating with a new
venture (Stuart, Hoang and Hybels 1999), we do not see the reputations of prior employers as being at
risk in the entrepreneurial ventures of their employees.)
Finally, employees of entrepreneurially prominent firms are advantaged because more is likely to be
known about them in the entrepreneurial community. Just as centrality in information networks may help
founders gain access to information and resources, it also helps diffuse information about founders and
their new ventures. The experiences and accomplishments of prospective founders will be more widely
recognized if they come from prominent employers. In this respect, affiliation with prominent employers
may reduce the information asymmetries faced by a new venture directly.
For each of these reasons, then, third parties may infer that founders from more prominent employers
possess, on average, greater skills and have a higher probability of success in their new ventures. Because
Coming from Good Stock: Career Histories and New Venture Formation
11
of this, we suspect that startup proposals from employees of prominent firms will a priori seem more
promising and hence receive more attention from external actors. Thus, external actors will have a
greater level of confidence in the ability of such founders to have success in the new venture.
Employment by prominent firms, in short, should reduce the perceived uncertainty of a new venture.
This idea is consistent with early statements by Stinchcombe (1965:146-7) regarding entrepreneurship,
where “…the patterns of trust and of mobility of resources which determines whether resources can be
moved to innovators are socially patterned.” In this paper we propose a specific source of theses social
patterns – past employers.
Hypotheses
Our arguments suggest that potential entrepreneurs secure informational and reputational benefits by
virtue of having once been employed by a prominent firm. Furthermore, these benefits will be amplified
in the face of uncertainty. In order to test these assertions, we first examine the level of uncertainty
involved in the kinds of ventures entrepreneurs start. We distinguish between ventures that pursue
innovation strategies and those that pursue other strategies. Most of the organizational strategy typologies
employed by empirical scholars allow for this distinction (e.g. Miles and Snow 1978; Porter 1980; Miller
1986). One theme across all of the typologies is the importance of differentiating firms that are exploiting
an existing market from those that are creating a new market. Following Maidique and Patch (1982), we
believe that this is an especially salient distinction for technology-based firms. In our definition, firms
pursuing an innovation strategy are seeking to win a technology race in a new niche. These firms are
attempting to gain competitive advantages by being the first to develop and exploit new, hitherto
unproven technologies. By contrast, incrementalist startups build upon existing products and
technologies, and seek to gain competitive advantage through technical enhancements, superior marketing
and customer service, and/or cost advantages. A critical difference between innovative and incrementalist
startups lies in the degree of uncertainty associated with a new venture of each type. High levels of
Coming from Good Stock: Career Histories and New Venture Formation
12
uncertainty characterize innovative startups, as the core products and technologies around which they are
built are of unknown value. The level of uncertainty for incremental ventures is correspondingly lower,
since external actors can more readily judge a venture’s promise by reference to existing firms. The
information asymmetries involved in assessing a venture’s quality are exacerbated for innovative
ventures, and founders of innovative ventures must therefore overcome a greater degree of perceived
uncertainty regarding their firm’s prospects.
Identifying opportunities for innovative ventures requires a good understanding of the future path of
technological progress and a wealth of information about technological alternatives. Entrepreneurs from
prominent firms should be at an advantage in both respects. The reputational benefits of being affiliated
with a prominent employer should make entrepreneurs more successful at reducing the perceived
uncertainty of their ventures. As a result, the informational and reputational benefits of working for a
prominent firm should translate into a superior ability to identify opportunities for innovation. Thus we
hypothesize that
Hypothesis 1: The prominence of prior employers will be positively related to whether a firm
pursues an innovative strategy.
Our second hypothesis concerns the ability of entrepreneurs to secure external financing at the time of
firm founding. In particular, we are interested in whether entrepreneurs from prominent firms are better
able to acquire resources from third parties. We suspect that the willingness of third parties to invest in a
highly uncertain venture during its infancy, before it has any track record by which to be assessed,
depends on the perceived quality of the new venture. There are three ways in which a brand new venture
can have higher perceived quality: 1) its founders have high levels of human capital; 2) it has a product
which can be independently evaluated; and 3) it has ties to prominent firms that serve as endorsements.
Coming from Good Stock: Career Histories and New Venture Formation
13
While in our analyses, we will attend to all three mechanisms, we are primarily interested in the third,
thus we hypothesize that
Hypothesis 2: The prominence of founding team’s prior employers will have a positive effect on
the probability of a new venture obtaining external financing at the time of founding.
Data and Methods
Data on new ventures
The data for this study are from the Stanford Project on Emerging Companies (SPEC). SPEC is a
stratified random sample of 173 young high-technology firms in Silicon Valley.
2
The sample is drawn
from the population of firms listed in Rich’s Everyday Sales Prospecting Guide (1994) and The
Technology Resource Guide to Greater Silicon Valley (1993/1994) and supplemented with firms from the
Silicon Valley business press that were too young to appear in published directories. SPEC is a
longitudinal study of organizational evolution with emphasis on formal systems and practices. In order to
minimize recall bias and to guarantee that the entities under consideration could potentially have the need
for formalized structures and systems, age and size criteria were used to define the population. Firms
included in the study were no older than 10 years and had at least 10 employees at the time of sampling.
At the time of sampling, the average firm was 7.3 years old and had 89 employees. The sample included
2
For details on the data collection and coding methods, see Burton 1995; Baron, Burton and Hannan 1996; Hannan,
Burton and Baron 1996. These publications describe the original sample of 100 firms for which data was gathered
in the summer of 1994. The sampling and data collection strategies were replicated in the summer of 1995 to
supplement the sample with an additional 72 firms (See Baron, Hannan and Burton 1999, 2000 for more
information).
Coming from Good Stock: Career Histories and New Venture Formation
14
firms that ranged in age from 2 to 12 years and in size from 9 to 2042 employees. The SPEC research
team conducted interviews with founders, CEOs, and senior managers responsible for human resources,
gathered survey and archival data, and compiled detailed organizational histories for each of the firms in
the study.
Bhide (1999) reports that 60% of start-ups fail within the first six years and that even those which survive
remain small. Based on these statistics, the SPEC sample is likely biased towards successful start-ups.
The nature of the sampling frame means that the firms under investigation have achieved some minimum
scale and longevity. However, it is important to note that attempts were made to include younger firms,
precisely to minimize survivor bias. Furthermore, despite this data limitation – which plagues virtually
all survey or interview-based organizational research – the SPEC sample has some noteworthy
advantages for our purposes. First and foremost, it includes firms pursuing different strategies, with
different sources of capital. Second, it is geographically constrained, which increases the probability that
multiple founders will have held positions in a given employer and thus generate variation in our measure
of prominence. For these reasons, we believe the sample is appropriate for testing our ideas.
The first dependent variable of interest in this paper is whether or not the firm was founded to pursue a
technological innovation strategy. Trained MBA and doctoral students conducted semi-structured
interviews with a founder of each of the firms asking him or her to describe the core competence of the
firm at founding. The open ended response (supplemented in some cases by early press reports, product
announcements, business plans and prospectuses) comprised the raw data that was used to categorize each
of the firms into one of four strategic archetypes: Innovators, Enhancers, Marketers and Low-Cost
Producers (see Hannan, Burton and Baron 1996). Innovators are firms that seek to gain first-mover
advantages by winning a technology race. Enhancer firms seek to produce a product similar to other
companies, but employ a general modification or enhancement to gain competitive advantage. Marketers
seek competitive advantage through superior sales, marketing or customer service. Finally, Low Cost
Producers are firms that seek cost advantages through cost efficient production techniques, relationships
Coming from Good Stock: Career Histories and New Venture Formation
15
with low cost suppliers, or economies of scale. The three latter strategies all revolve around extending
existing products or services. For the analyses presented here, we collapse the latter three categories into
one category, thereby focusing on the distinction between innovators and incrementalists. In light of
suggestions that entrepreneurs may selectively recall their company’s history (Bhide 1999), the use of a
retrospective measure of strategy may seem problematic. However, we feel confident that our measure
captures the difference between innovators and incrementalists with a high degree of accuracy. In
particular, respondents were not asked to classify their strategies themselves; rather researchers coded
strategies based on business plans, prospectuses, and articles from the business press describing the
industry. Furthermore, Hellman and Puri (1999) perform a number of post-hoc analyses of the same data,
including linking patenting activity to strategy and finding that innovators accumulate larger patent
portfolios, which increase our confidence in the measure.
The second dependent variable is whether the firm received external financing at the time of founding.
New ventures have a variety of alternative sources of capital. Some entrepreneurs self-finance the early
start-up phase using their own personal assets. Other entrepreneurs have a source of revenue or cash
flow, such as a licensing agreement or a consulting contract, that finances the venture. Still others are
able to mobilize the resources of friends and family to support the early stages of a new enterprise. Many
must seek capital from external third parties such as venture capitalists, private investors (so-called
“angels”), corporate investors, commercial or investment banks, other financial institutions such as
pension funds or insurance companies, or the government. These alternative funding sources vary in the
extent to which they are willing to associate with risky ventures and in the price at which they provide
capital. Venture capitalists and private investors anchor the end of the continuum that finances the most
uncertain enterprises at the highest price (see Roberts and Stevenson (1991) for an overview of start-up
financing). Not surprisingly, most of the start-up firms that received external funds at inception obtained
the funds from these high price sources. Thus, the task of attracting external financing was one of
persuading investors who were in the business of evaluating risky ventures.
Coming from Good Stock: Career Histories and New Venture Formation
16
Information on the financing history of each of the SPEC firms was collected via a combination of public
and proprietary databases, SEC filings and annual reports, internal company documents and a survey
instrument that was sent to the most senior finance executive at each of the firms.
3
In this paper we focus
on whether or not an entrepreneurial venture received funds from any outside investor at inception. We
do not distinguish types of investors, nor do we differentiate so-called “seed money” from first or second
round financing, nor do we account for differences in the amount of financing. Instead our variable is
simply a measure of whether the founding team had any amount of money from any third party at the
earliest moments of the firm’s existence. Of course it is interesting to note that the vast majority of
external investors in the SPEC sample were venture capitalists (71%) with some angels (13%) and
corporate investors (13%). Only one of the SPEC firms borrowed start-up funds from a commercial bank.
The invested amount for all stages of investment (for the subset of firms for which we have the data)
ranged from $10,000 to $30,000,000 with one firm receiving $100,000,000 in cash and stock as part of a
merger agreement. Excluding this outlier firm, the average investment size is just under $2.5 million, and
the amounts of initial investments are one the small end of the distribution.
We are interested in whether a founding team can persuade third parties that their venture is promising
before they have begun significant operations and have a tangible organizational track record that can be
3
The financing history data collection effort was led by Professors Thomas Hellmann and Manju Puri of the
Stanford Graduate School of Business (Hellmann and Puri 1999). Sixty-six firms (38%) responded to a finance
history survey that was addressed to the senior executive responsible for finance. Data for a large number of the
sample firms was available from commercially available databases that track the venture capital industry. 107
(62%) of the SPEC sample firms had records in the Venture One database (see Gompers and Lerner 1997 for a
discussion of this database); 95 (55%) had records in the Venture Economics database (see Lerner 1995 for a
discussion of this database). Additional information was gleaned from the founder interview transcripts as well as
archival research in the business press.
Coming from Good Stock: Career Histories and New Venture Formation
17
evaluated. Having an external third party provide financial capital is evidence that the team was
successful. Ideally, we want to capture those firms that have external funds at inception. In practice, this
is difficult to operationalize. There are ambiguities and inaccuracies inherent in both the founding dates
and the financing dates. Most scholars define the birth of a firm as the date that it was legally
incorporated. However, in constructing organizational life histories for very young firms we discovered
that many had substantial lives – ones that involved full-time employees and/or revenues from sales –
well before the founders ever approached a lawyer to incorporate them. For this reason we define the
founding date as the earliest possible date that there is any indication that a new organization exists,
including legal incorporation, having a full-time employee, or selling a product. This sets the start-up
clock to begin at the earliest possible moment. Similarly, financing dates recorded in commercial
databases and on our surveys were often recorded in terms of annual “quarters.” Even in cases where a
precise date was given, we understood this to be the date a financing deal closed, rather than the moment
when negotiations commenced or even when a “handshake” or verbal agreement was reached. Thus, in
order to accommodate these recording inaccuracies, we coded a firm as having external financing at
founding if it was received within three months of the founding date.
4
Data on Founders’ Careers
We have further augmented the SPEC data with information on the career histories of each of the
founding team members of the SPEC companies. As part of the data collection process, the SPEC
research team interviewed and surveyed a founder of each of the firms. This informant was asked to
4
We tested alternative intervals. The results when we more strictly define the date of financing are weaker, since
there are fewer positive outcomes, but in the same direction. We obtain statistically significant findings that are
substantially equivalent to those reported when we expand the financing interval to be within the first six months of
founding. We chose to report the analyses from the slightly more conservative three-month interval.
Coming from Good Stock: Career Histories and New Venture Formation
18
identify, by name, the other members of the founding team. This list of founding team members was then
verified through archival research of public documents as well as internal company records available to
the research team. Among the 173 SPEC firms, founding team size ranged from 1 to 12 (with an average
team size of 3). For each founding team member, SPEC research assistants searched a number of archival
sources, including SEC filings, company documents, newspaper articles and profiles, electronic databases
such as Lexis/Nexis, and internet archives in order to reconstruct each founder’s job history prior to
launching the new venture. For each of the 527 founders, we attempted to collect information on all jobs
held prior to the start of the new venture including the position held and the name of the employer. We
contacted the human resources department for 20% of the firms and confirmed the founders prior place of
employment. The career history data collection process generated a list of 1252 positions in 438 distinct
prior employers. In our data, the number of prior jobs held by a given founder ranges from 0 to 9. Of the
527 founders we identified at least one prior employer for 420.
5
Our key independent variable, employer
prominence, requires that we have data on at least one prior employer for at least one member of the
founding team. We were unable to collect any career history or educational background information for
any of the founders at 9 of the 173 SPEC companies; thus our sample is reduced to 164 firms.
5
We confirmed that there were at least 38 additional founders who began working at the SPEC firm directly from
school and thus their number of prior employment ties was truly 0. For the remaining 69 it is difficult to ascertain
whether missing data arises because the founder had no prior jobs, or whether the experience was simply not
reported in our sources. We suspect that there is some bias toward large, established firms being mentioned in press
accounts about the individuals in our sample; employers that are less important in the eyes of the media may not be
mentioned in newspaper stories and press releases. We attempt to account for this problem in our analyses by
replicating the models using different numbers of prior jobs. At a minimum, it is important to note that since we
were unable to administer job history interviews to the founders, these data are imperfect records of the career
histories of the SPEC entrepreneurs.
Coming from Good Stock: Career Histories and New Venture Formation
19
Measuring the Prominence of Past Employers
The central predictor of interest is the prominence of each founder’s past employers. Our measure of
prominence should capture the extent to which an existing firm is visible to those engaged in
entrepreneurial activity (cf. Knoke and Burt 1983). Given the diversity of industries represented by our
firms, it is difficult to think of a single dimension along which all of the firms can be unambiguously
ranked. Asset- or revenue-based size measures may have some applicability to for-profit organizations,
but are difficult to apply to universities, for example. Measures of technological prominence based on
patents (Stuart, Hoang and Hybels 1999) may also be a plausible basis for ranking firms. Again,
however, difficulties arise with respect to cross-industry comparisons and with respect to the best way to
characterize the prominence of firms operating in multiple industries. More importantly, it is unclear
what criteria third parties in an entrepreneurial context use to assign prominence to existing firms, and
whether these criteria are consistent across industries. Ideally, we would want an independent reputation
survey of all existing firms completed for each of the years from 1982 to 1992 when sample ventures
were founded. Unfortunately, we know of no such survey.
For these reasons, we choose to arrive at the prominence of past employers inductively, based on the
observed pattern of entrepreneurial activity in our sample. We measure a firm’s prominence by the extent
to which it has been a source of entrepreneurial ventures. Firms that generate a lot of new ventures
should be more visible to other actors in the entrepreneurial community. In order to construct this
measure, we start with a binary matrix of ties between the 164 new ventures in the SPEC sample that have
some prior career data for the founders and the 438 past employers of all of the founders. A “tie” is
formed if any member of the founding team had worked for the past employer. Thus entries in the cells
i,j of this matrix are 0 if there is no founding member at start-up i who worked for the past employer j,
and 1 if there is at least one founder who worked for the past employer j. Summing across the rows
generates a count, for each start-up, of the number of prior employers represented on the founding team.
Coming from Good Stock: Career Histories and New Venture Formation
20
a x
i ij
j
?
?
Summing down the columns of this matrix generates a count, for each past employer, of the
number of startups in the SPEC sample that have emanated from that employer.
b x
j ij
i
? ?
?
1
We subtract 1 from b
j
to exclude the focal start-up; if a past employer has only generated one start-up
(each has generated at least one), it will have a prominence score of zero. For each venture in the SPEC
sample, we then generate a measure of the prominence of all of the past employers by summing the b
j
across each of the past employers represented on the founding team.
6
Using the observed entrepreneurial activity in our sample to measure the prominence of past employers
may strike some as tautological, given that we are seeking to explain entrepreneurial activity. There are
two reasons why we believe this is not so. First, we do not use this measure to predict the rate of
entrepreneurial activity, but rather as a predictorof the characteristics of the entrepreneurial activity. We
do not see a necessary connection between our measure of prominence and whether entrepreneurs pursue
innovation strategies, much less whether they are able to secure external financing at start -up. Second,
we conceive of the SPEC sample of start-ups as generating a sample of past employers of start-up
founders, where the past employers are represented proportional to the entrepreneurial activity that they
generate. Our strategy thus parallels the National Organizations Survey, which generated a sample of
organizations by asking randomly selected individuals to name their employers (Kalleberg, Knoke,
6
This measure of prominence will increase on average with the number of prioremployers recorded for a founding
team. In order to account for this, we control for the number of past employers in the models.
Coming from Good Stock: Career Histories and New Venture Formation
21
Marsden and Spaeth 1996; see also McPherson 1982). This sampling procedure – termed probability
proportional to size samples – are statistically optimal for populations where the elements vary widely in
size (Sudman 1976). We believe that replications of this procedure for new high -technology ventures in
Silicon Valley would generate similar lists of past employers.
The career history data collection process generated a list of 1252 positions in 438 distinct prior
employers. If people were described as being in self-employment (such as doctors or independent
consultants), or for some other reason the firm was not identified, the “prior firm” was coded as missing.
Missing prior firms account for 87 of the 1252 positions (6.9%). The remaining positions are infirms
that range from familiar high-technology employers in Silicon Valley – such as Hewlett-Packard, Intel
and Apple – through academic institutions – such as Stanford and Harvard – to the military and less well-
known firms. Despite the diversity of firms, there is a surprising degree of concentration in
entrepreneurial activity (see Table 1). For example, 6 prior employers dominate the list (IBM, Hewlett -
Packard, Stanford University, Apple Computer, Intel, and National Semi conductor) and 69 SPEC firms
(46% of the sample) have at least one founder who worked at one of these six firms. It is also worth
noting that, while this list captures many large Silicon Valley employers, it is not collinear with size. One
of the largest employers, Lockheed, with over 21,000 employees in 1990, does not appear on the list of
prominent firms. Furthermore, for firms such as Apple Computer (5,700 employees) and Sun
Microsystems (7,700 employees) the prominence measure appears unrelated to s ize. Apple’s prominence
score is double that of Sun’s (12 compared to 6).
7
7
Numbers of employees by firm is based on data from September 1990 and was reported in a San Jose Mercury
News article, “Largest Employers” printed Monday, January 14, 1991 on page 2C.
Coming from Good Stock: Career Histories and New Venture Formation
22
Control Variables
In addition to the prominence of past employers, we control for a number of other characteristics of the
SPEC companies and their founders (since we are studying the firms at their inception, there are few
organizational characteristics to measure). In the external financing models, we control for how far along
each company is in the entrepreneurial process by including an indicator as to whether or not they had a
completed product ready for shipment within six months of founding. Having a working product, or even
a product prototype, is one way that a firm can reducing the perceived uncertainty for external
stakeholders. We believe that this approach to reducinguncertainty will be particularly effective for firms
pursuing an innovation strategy; thus, we include an interaction term. We also control for industry for
several reasons. We want to capture differences in the need for capital (medical devices compani es on
average should require more initial capital than software companies) and the attractiveness of an industry
in the capital market. We also need to account for different baseline levels of innovativeness across
industries. Finally, industries are notequally represented in the sample. In the analyses presented in this
paper we include dummy variables for three broad industries: medical -related (including medical devices
and biotechnology), networking and telecommunications, and semiconductors. The omitted category
consists primarily of computer hardware and software companies, electronic component manufacturing
companies, and contract research and development firms.
Past research suggests that career experiences shape the propensity and ability ofindividuals to launch
entrepreneurial ventures. We control for a number of such experience-related characteristics in our
models. First, we control for the number of founders with prior entrepreneurial experiences. Second, we
include measures of the number of founders with prior senior management experience. Some evidence
indicates that venture capitalists take into account the management experience of entrepreneurs
(MacMillan, Siegel, and SubbaNarishma 1985). We also control for the number of founder s with
experience in sales/marketing or finance in order to control for the possibility that the perceived quality of
a team may be related to the presence of functional diversity.
Coming from Good Stock: Career Histories and New Venture Formation
23
Finally, we control for the general human capital of the founders by including education level in our
analyses.
8
Specifically, we measure the proportion of the founding team that has advanced degrees (i.e.,
more than a B.A.) We also consider the possibility that third parties might look to tangible measures of
accomplishment asan indicator of the quality of the founding team; thus, we collected information on all
of the patents granted to each of the founders in our sample prior to the launch of the new venture.
9
We
interpret this as a measure of the technological or innovative competence of the founding team
members.
10
Table 2 presents descriptive statistics for all variables in the models. Bivariate correlations are presented
in Appendix A. As is apparent from Table 2, almost half of the ventures in our sample pursue an
innovation strategy. Slightly over a third of the SPEC companies have external financing at the time of
founding, while approximately one-sixth have a product within the first six months. On average, 3.4
different employers are represented on each founding team, which is slightly more than the mean
8
The models that we report in this paper include only the education control variables. The findings are equivalent
when we include age as a proxy for experience; however our sample size is dramatically reduced due to the
difficulty in locating reliable birthdates for the founders.
9
Patent data for each individual was collected through the U.S. Patent Office’s web sitehttp://www.uspto.gov
10
We also collected information on the number of citations to each founder’s portfolio of patents; however, this had
no effect in our models.
Coming from Good Stock: Career Histories and New Venture Formation
24
founding team size (2.95).
11
The vast majority of prior employers (338 of 438) have a prominence score
of 0. The maximum prominence score for a prior employer is 21. Aggregating across all prior employers
for a team yields a prominence score range from 0 to 52 with the average SPEC firm earning 9.6 points.
Results
We argue that firms pursuing innovative strategies face higher levels of uncertainty. Evidence of the
higher level of uncertainty surrounding innovation strategists can be found in Table 3, which cross-
classifies the initial strategy of the ventures in our sample by whether or not the firm had external
financing at founding. While almost half of the incrementalist firms had external funding at the outset,
only 28% of those firms pursuing an innovation obtained such funds. This difference is statistically
significant. By construction, incrementalist firms are operating in known market niches where there are
already established entities. There is both an identifiable market opportunity and a means to assess – and
benchmark – the quality of the product or service being offered by the new venture. Neither is possible
for innovative firms. Thus, we interpret this table to support our claims that third parties, such as venture
capitalists, are less willing to provide initial funding to new ventures that pursue high -risk and uncertain
innovation strategies.
11
It is possible that our data collection strategymisses firms that are prominent in an entrepreneurial context but that
do not garner media attention. For these reasons, we conducted the analyses using only the immediately prior job
for each founder, using three prior jobs per founder, and using all available data. The results conform to our
hypothesized expectations; however, the prominence distribution is greatly constrained in the first case and
dramatically skewed in the latter. We report the intermediate choice, allowing up to three prior jobs for each
founder, in this paper.
Coming from Good Stock: Career Histories and New Venture Formation
25
In Table 4, we present logistic regression estimates of the determinants of a ne w venture’s strategy at the
time of founding. We focus our discussion on the fourth column of results. The parameter estimates
suggest that the past career experiences of founders have an impact on their choice of strategy. Teams
with a lot of experience in sales or finance are, as might be expected, less likely to pursue innovation
strategies. Graduate education also has a positive impact on the decision to pursue an innovationstrategy.
As predicted by Hypothesis 1, the prominence of past employers has a positive impact on the propensity
to pursue risky strategies. A one-standard deviation increase from average prominence of the founding
team’s past employers increase the odds of pursuing an innovation strategy by a factor of 1.65. This
result is consistent with our claim that entrepreneurs benefit from being associated with prominent
employers. Our measure of the prominence of past employers captures the extent to which firms are at
the center of entrepreneurial activity. Thiscentrality in entrepreneurial networks can have both
informational and reputational benefits which make it more likely that employees of prominent firms will
pursue innovation strategies.
It is difficult to differentiate the information and reputation acc ounts as explanations of the prominence
effect. The information story suggests that employees of prominent firms take advantage of ideas and
innovations that they are exposed to in the course of their work. Their employer may not be aware of
these ideas, or may not be interested in pursuing them. In order to explore the role that such exposure
may play, we turned to the explanations given by founders in response to the question, “What was the
catalyst or impetus for founding the company?” Out of these open-ended responses, we coded whether
the founder indicated that the idea that formed the basis of the new venture had come from work being
done at a prior employer. Of the firms for which we have suchinterviews (N=131), 23% mentioned that
projects they had undertaken in the context of a prior employment setting as the catalyst for starting the
venture. In separate models, we included a dummy variable indicating whether such a project with a prior
employer was the impetus for the new venture. This variable had no effect on the probability of pursuing
Coming from Good Stock: Career Histories and New Venture Formation
26
an innovative strategy, and had a negligible effect on the relationship between the prominence of past
employers and venture strategy.
We also experimented with a different measure of employer prominence, in part because of our lingering
concern over whether our effects are driven by differences in the size of prior employers. Size may be
relevant since it has been shown to affect organizational innovation processes (Cohen and Levin 1989).
We do not have direct measures of employer size. Instead, we created a dummy variable indicating
whether the prior employer was listed in the Silicon Valley 100, an a nnual listing of the largest firms in
Silicon Valley produced by the San Jose Mercury News. In separate analyses (available from the
authors), we experimented with various ways of including information on the number of prior employers
listed in the Silicon Valley 100. None of these affected the propensity of firms to pursue an innovation
strategy, and the effects of entrepreneurial prominence were robust throughout the different
specifications.
In Table 5 we turn our attention to the determinants of external financing at founding. These estimates
are from logistic regression models of whether or not a venture had external financing within three
months of founding. As the cross-classification in Table 3 suggested, firms pursuing an innovation
strategy are less likely to secure external financing at start-up. In the second model, we introduce an
interaction effect between the firm’s strategy and whether or not they had a product at founding. We see
that these two variables have a complex effect on the li kelihood of external financing at founding. The
main effect of the product variable indicates that firms pursuing an incrementalist strategy are less likely
to secure external financing at founding. This may seem counterintuitive. However, it is importa nt to
note that our dependent variable primarily captures infusions of venture capital, which comes at a higher
cost than traditional sources of capital (such as bank loans). Since incrementalists are operating in
established markets, those with a productin hand have the least need for this more expensive type of
financing. In fact, they may be able to generate sufficient revenue from sales to mitigate the need for any
external financing. Turning to innovation strategists, the interaction effects sugges ts that these firms,
Coming from Good Stock: Career Histories and New Venture Formation
27
having a product makes it more likely that the firm will receive external financing at founding. Unlike
incrementalist firms, however, innovative startups are more in need of venture capital due to the
uncertainty surrounding the market for their products.
In the next two models, we include measures of the experiences and achievements of the founding team.
First, we see that prior founding experience has no effect on the odds that a new venture will receive
external financing at founding. This may be due to the fact that our measures captures only whether or
not a founder had been involved with a prior start -up, but nothing about the outcome. If the prior
founding experiences have had negative outcomes, third parties may be hesitant to invest in another
venture. Alternatively, if the prior founding experience had been successful, and the entrepreneur has
“cashed out,” his or her own personal wealth may obviate the need for external financing in the early
stages of the firm. It is also worth noting that prior founding experience is significantly correlated with
senior management experience – which has a positive effect on the odds of attracting external
stakeholders. Founding teams whose members include at least one with prior senior management
experience are more likely to secure external financing at founding. This is consistent with studies
showing that venture capitalists value the management experiences of entrepreneurs when evaluating
proposals (MacMillan, Siegel, and SubbaNarishma 1985). Neither the innovative ability of the founders,
as measured by the number of patents held, nor graduate credentials have a significant effect.
12
We see in the fifth model in Table 5 that the prominence of past employers initially has no signifi cant
effect on the odds of securing external financing at startup, suggesting no support for Hyothesis 2.
However, this model does not take into account the different levels of uncertainty associated with
innovation strategies and incrementalists. We expect entrepreneurial prominence to be especially
beneficial when the perceived uncertainty of the venture is high, such as when a firm pursues an
12
We tested for an interaction effect with the strategy of the firm; it was not significant.
Coming from Good Stock: Career Histories and New Venture Formation
28
innovative strategy. In the final model (model 6) in Table 5, the effect of employer prominence differs
for the two types of firms. Among firms pursuing an innovation strategy, employer prominence has the
expected positive and statistically significant effect on the odds of securing external financing at startup.
For innovative ventures, where the quality of theventure team is arguably of greatest importance,
employer prominence has a significant effect on the ability of the founders to secure resources from
external providers.
13
(Separate analyses (not shown) using the Silicon Valley 100 measures discussed
above had no influence on the pattern of results.) This supports our claimthat the reputational benefits of
employer prominence reduces the perceived uncertainty of new ventures and facilitates entrepreneurial
activity.
Discussion and Conclusion
The analyses presented in this paper provide evidence supporting the claim that career histories shape the
entrepreneurial process. First, functional and educational backgrounds influence initial strategic choices,
and management experience is important to external stakeholders. Entrepreneurs with advanced degrees
establish firms with innovation strategies, but entrepreneurs with sales or finance experience are less
likely to pursue an innovation strategy. Entrepreneurs with senior management experience havemore
legitimacy with external constituents and are more likely to obtain external financing. These findings are
consistent with work on human capital and the importance of career histories on the formation of new
ventures. Our work moves beyond these findings, however, to address the importance of social capital
13
Arguably, the firms with the greatest uncertainty surrounding their quality are innovation strategists without a
product at the time of founding. This suggests a three-way interaction between strategy, product at founding and
employer prominence. We tested for this interaction in a separate model, not shown here. Employer prominence
has no significant effect for innovators with a product, but does have a significant effect for innovators withou t a
product. This is consistent with our argument.
Coming from Good Stock: Career Histories and New Venture Formation
29
for entrepreneurs. Entrepreneurs setting out from prominent employers have both information and
reputation advantages over those who emanate from less prominent firms. It is important to note n ot only
what experiences and background entrepreneurs have but also where these experiences come from. The
information and reputation advantages that accrue from social capital allow entrepreneurs from prominent
firms to pursue more risky ventures, such as founding a firm dedicated to establishing a new product or
market. The reputational capital derived from being affiliated with a prominent employer also allows
entrepreneurs to reduce the perceived uncertainty of their venture, thereby facilitating the acquisition of
resources from third parties. Risky ventures (those pursuing an innovation strategy) that emerge from
prominent employers are more likely to obtain external financing.
While we believe our analyses are persuasive, they are limited in certain respects. First, our data do not
allow us to distinguish between a desire to launch a new venture pursuing an innovation strategy, and the
ability to do so. This makes it difficult to specify clearly the mechanism by which employer prominence
influences the choice of initial strategy. Specifically, we cannot confidently determine whether
individuals from prominent employers are more likely to launch innovative ventures because they are
privy to superior information, or because they benefit from t he prominence of their employers in
convincing third parties to support the venture. Distinguishing between these accounts would require a
more detailed study of proposed entrepreneurial ventures and the process by which they move from initial
concepts to nascent firms. Despite this limitation, what we do know is important: entrepreneurs from
prominent employers launch more innovative ventures, and those ventures are more likely to obtain
external financing.
Second, while our interpretation of these results emphasizes the benefits of prominent structural locations,
we are sensitive to alternative explanations that point to the possible effects of unobserved heterogeneity
among founders. It is possible that the observed effects of entrepreneurial prominence are due to
unobserved characteristics of established firms and the employees they attract. For example,
entrepreneurially prominent firms may attract employees whose personal characteristics make them
Coming from Good Stock: Career Histories and New Venture Formation
30
particularly likely both to pursue innovativeventures and to win the confidence of external investors. As
with any such claim, we cannot rule out with certainty that the findings can be attributed to unobserved
heterogeneity. However, we feel confident that we have measured and controlled for seve ral of the most
important individual-level characteristics that can most plausibly be thought to affect the outcomes we
examine. Our models include measures of the patenting activities of the founders, their educational
backgrounds, their prior work experiences and their past entrepreneurial activity. Moreover, we have no
a priori reasons to expect that the firms identified as entrepreneurially prominent in this sample should
differ systematically in their recruitment behavior. At the same time, we beli eve that an important and
promising line of future research would be to explain why firms differ in the rate at which they generate
new ventures through employee departures. The limited amount of work that has been done in this area
suggests that such variations can be traced to differences in internal promotion chances, reward levels,
technological emphases and managerial practices (Freeman 1986; Brittain and Freeman 1986). A full
understanding of how established firms shape entrepreneurial behavior must attend to both the cause and
the consequences of entrepreneurial prominence.
Finally, the diversity of firms in the SPEC data set, although useful for understanding a broad set of
organizations, has certain shortcomings. Ideally, in addition to our emplo yer prominence measure, we
would have an exogenous measure of the prominence of past employers. The broad set of industries
represented in the sample make such a measure difficult to generate. For studies of new ventures within a
single industry, measures of technological or innovative prominence may be appropriate. Stuart, Hoang
and Hybels (1999), for example, measure the prominence of alliance partners using counts of citations to
a firm’s patent portfolio. The development of exogenous measures of prominence requires confidence
about the criteria by which members of the entrepreneurial community rank existing firms. To our
knowledge, this topic is unexplored in the existing literature. Furthermore, an exogenous measurement of
prominence has its own problems. No clear dimension exists on which we could compare the prominence
Coming from Good Stock: Career Histories and New Venture Formation
31
of a biotechnology firm with the prominence of a hardware firm. As such, single industry studies may be
more appropriate places to develop exogenous measures of prominence.
We began this paper by arguing that the landscape of existing firms shapes the entrepreneurial process.
We believe our results demonstrate that patterns of entrepreneurial activity are shaped by the social
structure of existing organizations (Stinchcombe 1965; Aldrich and Zimmer 1986). Our work is,
therefore, an important complement to studies showing how the general scarcity of resources affects the
formation of new firms (Hannan and Freeman 1989). In particular, these findings stress the importance of
hierarchical differentiation in the social structure of organizational populations (Podolny 1993; Podolny,
Stuart and Hannan 1996; Stuart, Hoang and Hybels 1999). We know from this existing work that
prominence dictates future patterns of affiliation, firm sur vival, and performance. We find additional
benefits accruing from prominence: firms emanating from prominent firms are more innovative.
For network theorists, our work further confirms the importance of network position. Entrepreneurs with
prominent past employers occupy a privileged place in the social structure, and their position garners
important advantages with respect to access to resources and information. What we add to the network
literature is an examination of how the network of existing organizations impacts the new venture, and by
extension the new venture network. The new ventures that spawn from prominent employers may occupy
a more prominent position in their own network. The access to ex ternal funding immediately connects
these innovative new ventures into an exclusive network of organizations. The innovative strategies of
these firms may lead them to higher visibility in their own industries. And the fact that they emerge from
prominent others may imprint them with positional advantage from the very beginning (Stinchcombe
1965). Prominence may not only be fairly stable over time, it may transfer from one organization to the
other through entrepreneurs. Furthermore, the impact of the new venture’s lineage may have implications
far beyond founding. These possibilities offer intriguing directions for future research.
Coming from Good Stock: Career Histories and New Venture Formation
32
We noted earlier that organizational researchers have grown increasingly interested in the role managerial
careers play in shaping organizational behavior and industry dynamics. Most research in this tradition
focuses on how career histories shape individual experiences and abilities (Boeker 1997; Sørensen 199 9).
Our research emphasizes that careers have important reputational consequences as well. In this respect,
the identity of a person’s employers (and perhaps other institutional affiliations) assumes primary
significance. Organizational reputations transfer to individual reputations. Inferences about the talents
and abilities of individuals are constructed from their histories of affiliation with employers. This
parallels studies of scientific careers, which have documented that the prestige of the university a person
attended has a positive effect on the prestige of the first job (Hurlbert and Rosenfeld 1992). Our results
suggest, however, that the effects of institutional or organizational prestige extend beyond the signals
associated with educational credentials and encompass the firms and other organizations that people move
through in the course of their careers. Moreover, the effects of institutional prestige extend beyond their
impact on individual life courses. The role that hierarchical differentiation among organizations plays in
both individual career dynamics and organizational populations, and the interconnections between the two
levels of analysis, is an important arena for future research.
Future research should examine other benefits of entrepreneurial prominence. We find a link between
prominence and innovative strategies and external funding, but prominent past emp loyers may continue to
impact internal organizational decisions through means like the recruitment of personnel from prominent
firms. Ventures spawned from prominent employers may be more likely to go public successfully, or
they may be more likely to be acquired by a larger, more established firm attempting to increase their own
prominence. These various research possibilities point out how disentangling where imprinting ends and
path dependence begins offers a challenge to futureresearch.
Our research contributes to a greater understanding of what differentiates new ventures. In order to
understand the emergence of innovative new ventures, we need to know where they come from in the
network of existing organizations. Past employer prominence offers firms a significant advantage in the
Coming from Good Stock: Career Histories and New Venture Formation
33
struggle for survival and success. We tie new ventures into the existing social structure and point out that
a new venture is more than a compilation of skills and experiences, but it emerges from other
organizations with positions in the social structure. Without incorporating the existing social structure
into our understanding of new ventures, we cannot hope to understand why one venture survives and
another fails, much less why the occasional venture succeeds beyond all expectations. Despite the rapid
rate of new venture formation, the ever changing technology, and the considerable hurdles new ventures
face, the underlying stability of the social structure offers a means to understand and keep up with the
changing organizational landscape.
Coming from Good Stock: Career Histories and New Venture Formation
(34)
References
Aldrich, Howard and C. Marlene Fiol. 1994. "Fools Rush In? The Institutional Context of Industry
Creation." Academy of Management Review, 19, 4: 645-670
Aldrich, Howard, and Catherine Zimmer. 1986. “Entrepreneurship Through Social Networks.” Pp. 3-23
in Donald L. Sexton and Raymond W. Smilor (eds.), The Art and Science of Entrepreneurship,
Cambridge, MA: Ballinger.
Baron, James N., M. Diane Burton, and Michael T. Hannan. 1996. “The Road Taken: Origins and
Evolution of Employment Systems in Emerging Companies,” Industrial and Corporate Change
5(2):239-275
Baty, Gordon B., William M. Evan and Terry W. Rothermel. 1971. “Personnel Flows as
Interorganizational Relations.” Administrative Science Quarterly 16:430-443.
Benjamin, Beth A. and Joel M. Podolny. 1999. “Status, Quality, and Social Order in the California Wine
Industry.” Administrative Science Quarterly 44(3):563-589.
Bhide, Amar. 1999. The Origin and Evolution of New Businesses. New York: Oxford University Press.
Blau, Peter M. 1964. Exchange and Power in Social Life. New York: Wiley.
Boeker, Warren 1988. “Organizational Origins: Entrepreneurial and Environmental Imprinting at the
Time of Founding.” Pp. 33-51 in Glenn R. Carroll (ed.) Ecological Models of Organizations. Cambridge,
MA: Ballinger.
Boeker, Warren. 1997. “Executive Migration and Strategic Change: The Effect of Top Manager
Movement on Product Entry.” Administrative Science Quarterly 42: 213-236.
Brittain, Jack W. and John Freeman. 1986. “Entrepreneurship in the Semiconductor Industry.” Mimeo.
University of California, Berkeley.
Coming from Good Stock: Career Histories and New Venture Formation
(35)
Brockhaus, Robert H., Sr., and Pamela S. Horwitz. 1986. “The Psychology of the Entrepreneur.” Pp.
25-48 in Donald L. Sexton and Raymond W. Smilor (eds.), The Art and Science of Entrepreneurship,
Cambridge, MA: Ballinger.
Burt, Ronald S. 1992. Structural Holes. Cambridge, MA: Harvard University Press.
Burton, M. Diane. 1995. The Emergence and Evolution of Employment Systems in High Technology
Firms. Unpublished Ph.D. Dissertation, Department of Sociology, Stanford University.
Chandler, Gaylen N. 1996. “Business Similarity as a Moderator of the Relationship between Pre-
Ownership Experience and Venture Performance.” Entrepreneurship Theory and Practice 20: 51-65.
Chandler, Gaylen N., and Steven N. Hanks. 1994. “Founder Competence, the Environmen t, and Venture
Performance.” Entrepreneurship Theory and Practice 18: 77-89.
Chandler, Gaylen N., and Erik J. Jansen. 1992. “Founders Self-Assessed Competence and Venture
Performance.” Journal of Business Venturing 7: 223-236.
Cohen, Wesley M. and Richard C. Levin. 1989. “Empirical Studies of Innovation and Market Structure.”
Pp. 1060-1107 in R. Schmalensee and R.D. Willig (eds.), Handbook of Industrial Organization, Volume
III. New York: Elsevier Science Publishers.
Cooper, Arnold C. 1985. “The Role of Incubator Organizations in the Founding of Growth-Oriented
Firms.” Journal of Business Venturing 1:75-86.
Cooper, Arnold C., F. Javier Gimeno Gascon, and Carolyn Y. Woo. 1994. “Initial Human and Financial
Capital as Predictors of New Venture Performance.” Journal of Business Venturing 9: 371-396.
Cooper, Arnold C., and W.C. Dunkelberg. 1986. “Entrepreneurship and Paths to Business Ownership.”
Strategic Management Journal 7: 53-68.
Coming from Good Stock: Career Histories and New Venture Formation
(36)
Freeman, John. 1986. “Entrepreneurs as Organizational Products: Semiconductor Firms and Venture
Capital Firms.” Advances in the Study of Entrepreneurship, Innovation, and Economic Growth 1: 33-52.
Gompers, Paul A., and Joshua Lerner. 1997. "Money Chasing Deals?: The Impact of Fund Inflows on
Private Equity Valuations" Harvard Business School Working Paper.
Goode, William J. 1978. The Celebration of Heroes: Prestige as a Social Control System. Berkeley, CA:
University of California Press.
Goslin, L., and B. Barge. 1986. “Entrepreneurial Qualities Considered i n Venture Capital Support.” In
R. Ronstadt, J.A. Hornaday, R. Petersen, and K.H. Vesper (eds.), Frontiers in Entrepreneurship
Research. Wellesley, MA: Babson College.
Granovetter, Mark S. 1973. "The Strength of Weak Ties." American Journal of Sociology 78:1360-1380.
Gunz, Hugh P. and R. Michael Jalland. 1996. Managerial careers and business strategies. Academy of
Management Review, 21(3):718-756.
Hall, John, and Charles W. Hofer. 1993. “Venture Capitalists Decision Criteria in New Venture
Evaluation.” Journal of Business Venturing 8: 25-42.
Hannan, Michael T., M. Diane Burton and James N. Baron. 1996 “Inertia and Change in the Early Years:
Employment Relations in Young, High-Technology Firms.” Industrial and Corporate Change 5(2):503-
536.
Hannan, Michael T., and John Freeman. 1977. “The Population Ecology of Organizations.” American
Journal of Sociology 82: 929-964.
———. 1989. Organizational Ecology. Cambridge, MA: Harvard University Press.
Hellmann, Thomas F. and Manju Puri. 1999. “The Interaction between Product Market and Financing
Strategy: The Role of Venture Capital.” Stanford Graduate School of Business Working Paper #1561.
Coming from Good Stock: Career Histories and New Venture Formation
(37)
Herron, Lanny, and Richard B. Robinson, Jr. 1993. “A Structural Model of the Effects of
Entrepreneurial Characteristics on Venture Performance.” Journal of Business Venturing 8: 281-294.
Higgins, Monica C. and Ranjay Gulati. 1999. “The Effects of IPO Team Ties on Investment Bank
Affiliation and IPO Success.” Harvard Business School Working Paper #00-025.
Hiltzik, Michael. 1999. Dealers of Lightning: Xerox PARC and the Dawn of the Computer Age. New
York, NY: Harper Business.
Hurlbert, Jeanne S., and Rachel A. Rosenfeld. 1992. “Getting a Good Job: Rank and Institutional
Prestige in Academic Psychologists'Careers.” Sociology of Education 65:188-208.
Jaffee, Adam B., Manuel Trajtenberg and Rebecca Henderson. 1993. “Geographic Localization of
Knowledge Spillovers as Evidenced by Patent Citations.” Quarterly Journal of Economics. 108:577-
598.
Jones-Evans, Dylan. 1996. “Experience and Entrepreneurship: Technology-Based Owner-Managers in
the UK.” New Technology, Work and Employment 11: 39-54.
Kalleberg, Arne L., David Knoke, Peter V. Marsden and Joe L. Spaeth. 1996. Organizations in America:
Analyzing their Structures and Human Resource Practices. Thousand Oaks, CA: Sage.
Kaplan, Jerry. 1995. Startup: A Silicon Valley Adventure Story. New York: Houghton-Mifflin.
Knoke, David and Ronald S. Burt. 1983 "Prominence." Pp. 195-222 in Ronald S. Burt and Michael J.
Minor (eds.), Applied Network Analysis. Beverly Hills: Sage Publications.
Latour, Bruno. 1987. Science in Action. Cambridge, MA: Harvard University Press.
Lerner, Joshua. 1995. “Venture Capitalists and the Oversight of Private Firms." Journal of Finance 50:
301-18.
Coming from Good Stock: Career Histories and New Venture Formation
(38)
Light, Ivan. 1972. Ethnic Enterprise in America: Business and Welfare among Chinese, Japanese and
Blacks. Berkeley: University of California Press.
Maidique, Modesto and Peter Patch. 1982. “Corporate Strategy and Technological Policy.” In M.
Tushman and W. Moore (Eds.), Readings in the Management of Innovation: 273-285. Marshfield, MA:
Pitman.
MacMillan, Ian C., Siegel, R., SubbaNarishma, P.N. 1985. “Criteria Used by Venture Capitalists to
Evaluate New Venture Proposals.” Journal of Business Venturing 1: 119-128.
McPherson, J. Miller. 1982. “Hypernetwork Sampling: Duality and Differentiation Among Voluntary
Associations.” Social Networks 3: 225-249.
Miles, R. E. and C.C. Snow. 1978. Organizational Strategy, Structure and Process. New York:
McGraw Hill Book Co.
Miller, Danny. 1986. Configurations of strategy and structure: Towards a synthesis. Strategic
Management Journal, 7(3):233-249.
Merton, Robert C. 1968 [1973] "The Matthew Effect in Science." Science 159: 56-63.
Pfeffer, Jeffrey, and Huseyin Leblebici. 1973. “Executive Recruitment and the Development of Interfirm
Organizations.” Administrative Science Quarterly 18:449-461.
Podolny, Joel M. 1993. "A status-based model of market competition." American Journal of Sociology,
98: 829-872.
Podolny, Joel M., and Toby E. Stuart. 1995 “A role-based ecology of technological change.” American
Journal of Sociology 100: 1224-1260.
Coming from Good Stock: Career Histories and New Venture Formation
(39)
Podolny, Joel M., Toby E. Stuart, and Michael T. Hannan. 1996. “Networks, Knowledge, and Niches:
Competition in the Worldwide Semiconductor Industry, 1984 -1991.” American Journal of Sociology
102: 659-689.
Porter, Michael E. 1980. Competitive Strategy. New York: Free Press.
Roberts, Michael J. and Howard H. Stevenson. 1991. “Alternative Sources of Financing.” Pp. 171-179
in William A. Sahlman and Howard H. Stevenson (Eds.) The Entrepreneurial Venture. Boston: Harvard
Business School Press.
Robinson, Peter B. and Edwin A. Sexton (1994) “The Effect of Education and Experience on Self-
Employment Success.” Journal of Business Venturing 9:141-156.
Rogers, Everett M. and Judith K. Larsen. 1984. Silicon Valley Fever: Growth of High Technology
Culture. New York: Basic Books.
Saxenian, Annalee. 1996. Regional Advantage : Culture and Competition in Silicon Valley and Route
128. Cambridge, MA: Harvard University Press.
Schumpeter, Joseph A. 1934. The Theory of Economic Development. Cambridge, MA: Harvard.
Shane, Scott. 1999. “Prior Knowledge and the Discovery of Entrepreneuri al Opportunities.”
Organization Science. Forthcoming.
Shane, Scott and Rakesh Khurana. 1999. “Career Experiences and Firm Foundings.” Working Paper.
Sloan School of Management, Massachusetts Institute of Technology.
Sørensen, Jesper B. 1999. “Executive Migration and Interorganizational Competition.” Social Science
Research 28: 289-315.
Sørensen, Jesper B. and Toby E. Stuart. 1999. “Aging, Obsolescence and Organizational Innovation.”
Administrative Science Quarterly, forthcoming.
Coming from Good Stock: Career Histories and New Venture Formation
(40)
Spence, A. Michael. 1974. Market Signaling. Cambridge, MA: Harvard University Press.
Stinchcombe, Arthur. 1965. “Social Structure and Organizations.” Pp. 142-193 in James G. March
(ed.), Handbook of Organizations. Chicago: Rand McNally.
Stuart, Toby E., Ha Hoang and Ralph C. Hybels. 1999. “Interorganizational Endorsements and the
Performance of Entrepreneurial Ventures.” Administrative Science Quarterly 44:315-349.
Stuart, Toby E. and Olav Sorenson. 1999. “The Geography of Opportunity: Spatial Heterogeneity in
Founding Rates and the Performance of Biotechnology Firms.” University of Chicago Graduate School
of Business Working Paper.
Sudman, Seymour. 1976. Applied Sampling. New York: Academic Press.
Tushman, Michael L. and Philip Anderson. 1986. “Technological Discontinuities and Organizational
Environments.” Administrative Science Quarterly 31:439-65.
Coming from Good Stock: Career Histories and New Venture Formation
(41)
Table 1. Top 24 Entrepreneurially Prominent Firms in SPEC Sample
Firm # of SPEC Firms
IBM 21
Hewlett-Packard 16
Stanford Univ. 15
Apple 12
Intel 11
National Semiconductor 10
ROLM Corporation 9
AT&T 8
Sun Microsystems 6
UC Berkeley 5
Silicon Graphics 5
Ungermann Bass 5
AMD 4
Digital Equipment Corp 4
MIPS Computer Systems 4
NASA AMES 4
NIH 4
Xerox Corp. 4
Bridge Communications, Inc. 3
Chips and Technologies 3
Control Data Corporation 3
Texas Instruments Incorporated 3
U.S Navy and Naval Reserve 3
Harvard University 3
Note: SPEC firms can have multiple “parents.” For example, a SPEC firm may have one founder who
worked at IBM, HP, and GE and a second founder who worked at HP and Honeywell. The career
backgrounds are aggregated at the firm level; thus this SPEC firm has prior ties to four firms (IBM, HP,
GE, and Honeywell) and wouldbe listed in this table as coming from both IBM andHP.
Coming from Good Stock: Career Histories and New Venture Formation
(42)
Table 2. Descriptive Statistics
Variable Mean SD Min Max N
Innovation Strategist
0.491 0.501 0 1 159
External Financing within 3 mos.
0.370 0.484 0 1 138
Product within 6 mos.
0.185 0.389 0 1 168
Medical-related Industry
0.143 0.351 0 1 168
Networking & Telecom. Industry
0.202 0.403 0 1 168
Semiconductor Industry
0.107 0.310 0 1 168
Prior Founding Experience
0.560 0.860 0 5 168
Senior Management Experience
0.935 1.079 0 5 168
Finance or Sales Experience
1.185 1.434 0 9 168
Log Number of Patents
0.698 0.967 0 3.85 166
Advanced Degrees
0.784 0.355 0 1 151
Number of Past Employers
3.440 2.107 0 11 168
Ties to Prominent Firms
9.595 10.860 0 52 168
Coming from Good Stock: Career Histories and New Venture Formation
(43)
Table 3. Relationship between Strategy and External Financing
External Financing at Founding
Strategy: No Yes
Incremental 31 29 60
52% 48%
Innovation 51 20 71
72% 28%
Total 82 49 131
?
2
= 5.65 (1 d.f.), p< 0.02
Coming from Good Stock: Career Histories and New Venture Formation
(44)
Table 4. Logistic Regression Estimates of the Determinants of Founding Innovation Strategy (N=159)
Variable (1) (2) (3) (4)
Medical-related Industry 1.745** 1.255 1.132 1.399
(0.591) (0.641) (0.670) (0.709)
Networking and -0.420 -0.522 -0.621 -0.881
Telecom. Industry
(0.419) (0.448) (0.480) (0.523)
Semiconductor Industry 0.187 -0.215 -0.343 -0.328
(0.545) (0.632) (0.682) (0.746)
Log Number of Patents 0.418 0.453 0.372
(0.217) (0.234) (0.243)
Graduate Degrees 1.429* 1.525** 1.384*
(0.553) (0.569) (0.619)
Prior Founding Experience 0.185 0.183
(0.234) (0.248)
Senior Management 0.026 -0.135
Experience
(0.201) (0.226)
Sales or Finance -0.444** -0.556**
Experience
(0.165) (0.175)
Number of Past Employers 0.164
(0.136)
Employer Prominence 0.046*
(0.023)
Constant
N 159 144 144 144
Log-Likelihood -103.03 -86.586 -81.537 -75.453
Pseudo R-squared .065 .131 .182 .243
Coming from Good Stock: Career Histories and New Venture Formation
(45)
Table 5. Logistic Regression Estimates of the Determinants of External Financing at Start -up (N=128)
Variable (1) (2) (3) (4) (5) (6)
Medical-related -0.568 -0.572 -0.197 -0.309 -0.236 -0.066
Industry
(0.586) (0.591) (0.628) (0.654) (0.701) (0.720)
Networking and 0.581 0.404 0.525 0.373 0.587 0.300
Telecom. Industry
(0.461) (0.476) (0.498) (0.515) (0.538) (0.568)
Semiconductor -0.268 -0.235 0.289 0.125 -0.062 -0.209
Industry
(0.663) (0.674) (0.730) (0.755) (0.784) (0.811)
Innovation -0.469 -1.097 -1.488* -1.568* -1.697* -2.009*
Strategist
(0.527) (0.670) (0.746) (0.767) (0.789) (0.816)
Product -0.705 -0.993* -1.018* -1.079* -0.940 -2.286**
(0.399) (0.436) (0.479) (0.517) (0.533) (0.780)
Product * Innovation 1.891 2.252 2.695* 3.025* 3.201*
Strategy
(1.129) (1.182) (1.231) (1.284) (1.306)
Graduate Degrees 0.240 0.262 0.261 0.455
(0.629) (0.636) (0.655) (0.695)
Log Number of -0.403 -0.363 -0.338 -0.318
Patents
(0.242) (0.245) (0.249) (0.256)
Prior Founding -0.058 -0.023 -0.242
Experience
(0.285) (0.289) (0.314)
Senior Management 0.350 0.513* 0.545*
Experience
(0.214) (0.235) (0.247)
Sales or Finance -0.072 -0.009 0.025
Experience
(0.157) (0.161) (0.176)
Number of Past -0.235 -0.226
Employers
(0.144) (0.149)
Employer Prominence 0.006 -0.072
(0.022) (0.042)
Innovation Strategy 0.115*
* Prominence
(0.048)
Constant -0.145 0.068 -0.282 -0.116 -0.155 0.432
(0.357) (0.375) (0.454) (0.474) (0.521) (0.596)
N 129 129 121 121 121 121
Log-Likelihood -80.408 -79.022 -71.417 -69.964 -68.350 -64.817
Pseudo R-squared .056 .072 .1 .118 .138 .183
* p < 0.05 ** p < 0.01 (two-sided tests)
Coming from Good Stock: Career Histories and New Venture Formation
(46)
Appendix A: Bivariate Correlations
1 2 3 4 5 6 7 8 9 10 11 12 13
1 Innovation Strategist
1.00
2 External Financing
within 3 mos.
-0.19* 1.00
3 Product within 6 mos.
-0.25* -0.05 1.00
4 Medical-related
Industry
0.28* -0.11 -0.11 1.00
5 Networking & Telecom.
Industry
-0.14 0.18* -0.01 -0.21* 1.00
6 Semiconductor Industry
0.01 -0.03 -0.07 -0.14 -0.17* 1.00
7 Log Number of Patents
0.24* -0.14 -0.19* 0.32* -0.11 0.19* 1.00
8 Graduate Degrees
0.27* -0.04 -0.18* 0.09 -0.08 0.13 0.18* 1.00
9 Prior Founding
Experience
0.06 -0.01 -0.10 0.11 0.05 0.00 0.05 0.05 1.00
10 Senior Management
Experience
-0.03 0.15 -0.11 0.02 0.18* 0.00 -0.04 0.01 0.39* 1.00
11 Finance or Sales
Experience
-0.18* 0.03 -0.05 -0.05 0.04 -0.02 -0.01 0.00 0.19* 0.39* 1.00
12 Number of Employers
0.21* -0.07 -0.15 0.11 0.25* 0.01 0.14 0.08 0.31* 0.48* 0.36* 1.00
13 Employer Prominence
0.24* -0.05 -0.01 -0.14 0.11 0.10 0.07 0.15 0.04 0.14 0.17* 0.52* 1.00
doc_308279261.pdf