Since economic capital is less accessible to novice entrepreneurs, these entrepreneurs are especially dependent on non-economic—social, human and cultural—capital. Established firms may provide these forms of capital, especially when firms cluster spatially. In a small Swedish community, all 29 established firms, as well as seven novice entrepreneurs were investigated with respect to their local networking. The findings are that tightly knit local networks provide strategic advantages for established firms on the one hand, while on the other hand offering novice entrepreneurs rich sources of non-economic capital. Embedding business networks in economic and social institutions enforces the potential of the local community as an incubator for entrepreneurship.

Entrepreneurship is about emergence, the continuous organizing of resources according to opportunity. In this endeavor, entrepreneurs use several sources of capital—potential input in the venturing process. In the economic literature, primarily financial and physical capital are addressed. Economic capital is tangible, controlled by ownership and appears on every balance sheet. It enforces the ideal economist image of the firm as an independent actor on the market. Contrasting these ‘autonomous’ forms of capital, ‘non-economic capital’ provides resources/services that are initiated, maintained and enforced as outcomes of exchanges between firms and other socio-economic agents. Thus, non-economic capital is not controlled by ownership but instead, jointly by dyadic and more complex social relations. Since the business world in general and enterprising activity in particular is socially enacted, we argue that non-economic capital is of special interest in business venturing.

Non-economic capital is genuinely relational. It mobilizes human faculties such as social responsiveness, ingenuity, trust and commitment (Borch, l992). Thus, not only a calculative interaction rationale but ideational and/or affective ones (Sjöstrand, l992) also have to be applied in order to make the venturing process intelligible. Firms and their representatives build informal and personal networks bridging business and private life spheres (Halinen & Törnroos, 1998). The latter networks appear as an important source of non-economic capital (Weijland, 1994).

The most recognized non-economic resource is social capital (e.g., Putnam, 1993; Burt 1997; Cohen & Field 1999; Nahapiet & Ghoshal, 1998; Portes, 1998). It is usually associated with mutual trust meaning that two (or more) agents have faith in each others’ ability and good will in exchange operations. The trust an entrepreneur brings into a relationship may be associated with her-/himself as a person, her/his emerging organization or an institutional setting, for example, the professional or trade association to which the entrepreneur/the firm belongs. Here we are especially interested in the role of spatial context as a provider of trust and trust-supporting norms (e.g., Lyon, 2000). Such trust does not only encourage local exchange but also adds to the general trustworthiness in the external environment of individual local (new) entrepreneurs and the local business community to which s/he belongs at large.

Human capital is the second form of non-economic capital we address. It includes codified knowledge as well as skills and, especially when associated with its practice, also attitudes and values (e.g., Light & Gold, 2000; Moog, 2000). Human capital as formal education is often generated out of social capital (Coleman, l988). Building human capital conducive to entrepreneurship, i.e., the practical skills and ambition needed, is indeed a social process (e.g., Neance, 1998). Preferably, this process starts as primary socialization at the dinner table during childhood, followed-up by the lifelong experiential learning associated with the practice of entrepreneurship. Then information sharing is natural; it does not make the informant less knowledgeable. There is generally an increasing awareness of the social dimension of learning, i.e., the accumulation of human capital. Local clusters of small firms, thus, represent arenas for individual and collective learning and entrepreneurship (Becattini, 1990). In regional studies, the concern for social learning is expanded far beyond the industrial district. The unique posture of human capital is assumed to offer global competitive advantages to the location and its firms (e.g., Markusen, l997; Flora, 1997; Maskell et al., 1998).

Informal human capital conducive for entrepreneurship is closely associated with cultural capital, our third form of non-economic capital (Berger, l991; Davidsson, 1995; Light & Gold, 2000). An entrepreneurial occupational culture includes, for example, nearby access to role models in the family and a business community dominated by independent-business values. Some preconditions for desirability and feasibility of entrepreneurship, such as adequate skills and values, are partly ‘enculturated,’ i.e., ascribed to the subject when being born into a community where entrepreneurship is a way of life. Cultural capital also may be intentionally acquired by making oneself eligible as a member of a (local) business community and its practices for example.

In Table 1, we outline some contrasting characteristics of economic and non-economic capital. Non-economic capital is often brought up in the context of small firms in general and underprivileged groups in particular, including nascent/novice and ethnic entrepreneurship. Different modes of financial bootstrapping, i.e., mobilizing the owner-manager’s trust-building capabilities, is crucial for most small, young firms (Winborg, 2000). Mentors bring refined knowledge. Ethnic entrepreneurs benefit from all three modes of non-economic capitalizing when compensating for their marginal position in the economy (see Light & Gold, 2000).

Non-economic capital is, however, generally important in market exchange. Since Granovetter’s seminal article (l985), there has been an increasing awareness of the social embeddedness of economic activity (see for example Uzzi, 1997). Networking, and the trust associated with it is generally recognized as an alternative regime to the market and the hierarchy. Increasingly, corporations build strategic alliances and other partnerships in order to reduce uncertainty and create unique resource profiles. In the service economy, human capital has been upgraded and the emerging experience economy seems to promise a paradise for knowledge workers and network entrepreneurs. The increased relevance of cultural capital is obvious in the present economy where global competition calls for local clustering (e.g., Porter, l998). In spite of the indisputable relevance of non-economic capital, there is, though, a need for further inquiry into how such capital is accumulated and structured.

Concerning form, we are especially concerned with how the (social) relations carry non-economic capital pattern into different networked structures. Social capital is not just embedded in individual dyadic relations, but within broader collectives as well. Regarding social/trust capital, this has been a special concern for research into regional-development, in particular industrial districts (see Putnam, l993). As indicated, human capital is associated both with dyadic exchanges of experiences and with networked ‘communities of practice’ (Wenger, l997). Cultural capital is genuinely collective, appearing as the sediments of human interaction.

We propose that the shared non-economic capital controlled by clusters of existing firms that are socially embedded will be accessed by those novice entrepreneurs who are accepted as members of the cluster. Similar to how the settlement of imbalances in a dyadic trust relationship may be delayed, the repayment of a debt a novice entrepreneur owes to a business community may be postponed. Thus, in favorable cases, a locality may amplify access to the non-economic capital that is so important for nascent and novice entrepreneurs.

In order to investigate how (collective) non-economic capital is created, defused and used, we need to model the way this capital is formed. In the literature, the difference between form and structure is not systematically dealt with; concepts are used metaphorically. The general notion of embeddedness itself on the one hand refers to the fact that individuals/organizations are nested in webs of (social) relations, and on the other, underlines that the exchange in these relations is ideational/genuine (personal) as well as calculative (impersonal).

The metaphorization of the phenomenon is especially elaborate in the literature on spatial clusters of firms. To some, the different forms of collective non-economic capital appear ‘in the air’ (Marshall, l920), to others, it is just generally ‘sticky’ (Markusen, l997), while still others point out that it is ‘untraded’ or ‘localized’ and therefore difficult to imitate (Maskell et al., l998). What seems to be lacking in the literature, though, is a way to represent the relational feature of all non-economic capital. Such insight is needed, whether we want to help individual firms building capitalizing strategies, or increase available non-economic capital in society. In this study, we therefore outline an operational model for mapping different kinds of local non-economic capital.

In the next section, we develop a network model of the making and diffusion of non-economic capital in a small-business community setting. The model and associated methodology maps exchange contents indicating different forms of capital, as well as the way firms and further local agents interrelate in network patterns. This modeling makes it possible to understand how novice entrepreneurs may be fed by, as well as feed, an entrepreneurial community. In the subsequent section, we present findings concerning the embeddedness of established and novice firms in the business cluster as well as in the institutional setting. The paper concludes with a discussion concerning ways to improve methodology and bridge to policy-making and practice.

The way relationships are structured in networks is as essential as their content (Lyon, 2000). Embeddedness and the associated creation and use of non-economic capital appear in different layers. Here we adopt the concepts of first- and second-order embeddedness (see Johannisson et al., 2000). First-order embeddedness concerns the integration of individual firms in business-to-business networks. Second-order embeddedness refers to the relations firms have with economic and social institutions (see Table 2).

The model also proposes that some relations and associated analyses mainly concern organizational exchange, while others originate in personal exchange. In new and small firms, though the novice entrepreneur and small-business owner-manager epitomize their firms as much as business and social concerns intertwine.

Operationalizing the localized business/institution networks, we follow a methodology that was originally developed by Johannisson et al. (1994), and Johannisson et al. (2002). Each respondent (business leader) was presented a list of all local firms and asked simply to mark those with whom they had relationships. This produced a number of adjacency matrices that provided information about which relationships were realized, out of the 406 (29(29-1)/2) potential (symmetric, reciprocated) relations between the firms in each strand. The strands building dyadic relations between firms/business leaders (first-order embeddedness) are identified in Table 3a.

In order to identify and measure second-order embeddedness, i.e., the relations between firms and economic and social institutions, we adopted a two-mode network analysis. That is, we created adjacency matrices with 29 rows (the responding firms) and 20 columns for the addressed economic institutions and 47 for the identified social institutions. These relations are unidirectional as only the business leaders’ responses and not those of the representatives for institutions are included. For technical reasons however, transposing the matrices symmetrizes all single-stranded ties between firms and associations.

The strands concerning the economic institutions paralleled those in the business-to-business network (with the exception that ‘acquaintance’ may concern anybody in the management of the institution). Table 3b presents the strands of the network constituted by firms/entrepreneurs and social institutions.

Three kinds of network variables were identified in order to map first-order embeddedness: a) frequencies of local network ties of firms; b) density measures of the local business networks; and c) firm centrality in the business-to-business (commercial) network. We used three measures of centrality: degree (number of dyadic relations to others), closeness (direct and indirect relations to others), and betweeness (gate keeping role) (see Appendix). Regarding second-order embeddedness, we only report frequencies of ties between firms/businesspersons and economic/social institutions (see also Johannisson et al., 2000). Thanks to the support of a regional small-business development organization, it was possible to cover the whole population of local firms with regards to network data. Background information concerning firms included, for example, size, market location and performance. Twenty-six firms provided such background information. Quantitative network approaches have often been criticized for limiting themselves to well defined variables, at best mirroring structure but not process (see for example Mönsted, 1995). However, the nearly complete network data available in this case provide unique opportunities to uncover structures that would have remained hidden in a necessarily focused qualitative approach. The technical network analysis was carried out by the UCINET V computer package (Borgatti et al., 1999) and the bivariate analyses used Minitab (State College, 1999).

The Empirical Base: The Lammhult Industrial Community

Lammhult is a small community with about 2,000 inhabitants located half an hour’s drive north of the regional center of Växjö in Southern Sweden. In Lammhult, there are about one hundred businesses, whereof the 29 largest ones are organized in a local business association. Together, these 29 firms employ about 600 persons, whereof the largest two about one hundred each. Additional local firms, most of them in trade and private services, only provide another 100 jobs (besides the 250 jobs in the public sector). The local trade association has conducted a number of initiatives that have considerably vitalized the community over the last five years and promoted it nationally. Lammhult was chosen as the study site because of our privileged access to the businesses and local organizations/institutions. The firm study concerns all the 29 members of the local trade association and 7 new firms (some data are only available for four firms).

In order to position the Lammhult cluster, we compare it with the Anderstorp cluster, part of the most advanced industrial district in Sweden (The Gnosjö region) (see Johannisson et al., l994 and Table 4). Surprisingly, we found that the two clusters seem quite similar. The firms are of the same average size and in both locations managed by founders in the same proportions. Lammhult and Anderstorp are also parts of the same wider regional setting dominated by small (family) businesses. It has to be kept in mind, however, that Anderstorp is considerably larger—twice as large with respect to population and, regarding number of firms (in each case the members of the local trade association), five times as large. The local business network studied in Anderstorp has 67 members, in Lammhult 29 members. In addition, a much larger proportion of firms are owned by their managers in Anderstorp. For example, the two largest firms in Lammhult are externally owned.

Some of the network data concern the global personal network of the owner manager/CEO. In both locations the business leaders have discussed their business with about the same number of alters over the last six months. Anderstorp’s portion of contacts in the local context, however, is higher than that of the Lammhult network, probably due to the much greater number of potential talk mates. In Lammhult, firms have a larger portion of their important contacts in other parts of Sweden and abroad.

Identifying the Lammhult Business Community as a Networked Setting

Networks between firms in Lammhult are very dense with respect to the portions of potential dyadic relations that are realized (see Table 5). Regarding acquaintance ties, 46.8% means that out of 406 potential dyadic relations, 190 are enacted. Compared with the mature industrial district Anderstorp, the Lammhult business community is much denser networked. One obvious explanation is that in Lammhult, the firm population (and the community itself) is considerably smaller. In a smaller socio-economic setting, the overview of potential network mates (‘alters’) is of course easier to achieve and the awareness of mutual dependencies is higher as well.

In the same way as in industrial districts in general, the business structure in Anderstorp is quite focused (mainly plastics and light engineering industries). This presumably facilitates business exchanges and the creation and diffusion of human capital. Nevertheless, the Anderstorp firms thus are considerably less densely interrelated than the firms in Lammhult are, in spite of the fact that there only nine firms belong to the same industry (furniture).

The other firms in Lammhult represent a wide variety of industries. While networking in Anderstorp is mainly spontaneous, different measures have been taken in Lammhult, both by private and public agents, in order to activate exchange between the firms themselves and between the firms and other social and economic institutions.

When uncovering local networking behavior it is important to find out the character of realized relationships. Are the relations single-stranded or do they combine different strands? If so, do business and social concerns intertwine? Table 6 shows that the Lammhult cluster stands out with respect to the social embeddedness of economic exchange. Close to half of all existing ties between firms include strong friendship relations (friendship here means combining acquaintance and talk strands). In addition, two out of three commercial ties are intertwined with such friendship ties. These relations represent genuinely personal-business relationships. Although the industry in Lammhult is far more heterogeneous than in Anderstorp, the portion of relations including taking advice from others (professional ties) is as high as in the industrial-district community.

A complex tie combines all strands and the data suggests that a large portion of the Lammhult relations make such strong ties. Nevertheless, when asking for new business opportunities mediated by business partners and vice versa, i.e., opportunities shared with local colleagues, only a few cases were reported. This suggests that trust is personal indeed and that direct relationships are important when it comes to enacting new business opportunities (see Uzzi 1997).

Testing the Homogeneity of Networking Activity

In the l994 study, the variation in networking activity was analyzed according to various contingencies such as founder/non-founder management, venture size, market scope and venture performance (Johannisson et al., 1994, Table 5). The general lesson was that there is little variation in networking due to these background variables. This suggests an egalitarian and munificent local business community, which invites every member to network and where everybody appreciates networking. The situation is close to what Borch (l992) addresses as communicative networks. In Lammhult, we analyzed the sensibility of the networking findings as well, also including ownership and habitual entrepreneurship as contingencies.

In addition, the performance variable was split into two parts—perceived growth and perceived profitability (if compared with other businesses in the same sector). Differences according to these seven (dichotomized) contingency variables were tested on all four single-stranded networks according to Table 3a. Out of the overall 28 t-test analyses thus made, only three turned out to be significant (p = 0.05). This suggests that Lammhult, as much as Anderstorp, is a very homogenous business community with respect to local organizing. The three significant differences suggest that founder-managers, when compared with non-founders, are less interested in doing business locally, for example outsourcing or having professional exchanges with local colleagues. The reason is probably a strong need for independence. Also, not surprisingly, business leaders in firms that are internationally oriented are personally acquainted with fewer local colleagues than others; their available time is obviously restricted (Burt, 1997).

Strategic Local Networking and Firm Performance

When testing for differences in local networking (number of single-stranded ties) according to profitability and growth indicators, the analysis weakly indicates (p = 0.07 - 0.08) that business leaders in successful firms are more involved than others in general exchanges of experiences locally (talk network). We have expanded on this indicative finding by introducing a more complex network variable, centrality. This is a structural measure. Such a variable identifies how networks pattern, for example, by making the individual firm more or less nested and influential in the local network. A central position provides an opportunity to accumulate social capital, although this calls for considerate networking (Lyon, 2000). Thus, three different indicators of centrality in local business networks were introduced. See Appendix for details.

According to all indicators of centrality used, the data suggest that firms managed by non-founders occupy a central position. Several of these firms are externally controlled. Firms that are particularly central in the Lammhult small-firm cluster have their most important contacts and dominant markets in the small-business region surrounding Lammhult and their business leaders have a national personal network that is larger than average. While the literature generally proposes that growth and profitability are associated with pro-action on external markets, little is said about the importance of interaction locally in order to increase performance. Our analysis unambiguously reports a strong relationship between having a central position in the local network on one hand, and firm growth and profitability on the other. (This conclusion remains when the two dominant factors are excluded from the analysis.) Thus, global competitiveness seems to be dependent on intense local collaboration, envisioning a ‘glocal’ strategy.

New Firms and Their Access to Non-Economic Local Capital

The analysis so far has demonstrated that established firms in Lammhult are generously provided with non-economic capital. This brings us to another major question. To what extent is non-economic capital made available to nascent and novice entrepreneurs? In order to address this issue, we extended our network analysis to include new firms in the Lammhult community. Seven new small firms were identified. The oldest firms started their operations about five years ago, while the youngest one at the time of the inquiry (February 2001) had only been in operation for five months. Their production is as varied as that of the existing small-business community. None of the new firms has more than six full-time employees. Most of the novice entrepreneurs are locals with family ties to the established firms. This suggests that they are integrated in the established local business network. In Table 7, we compare the average number of relations between the established firms themselves and the relations between the new firms and the established firms. The general finding is that the new firms are as tied into the established business network as its mature members are. Obviously, in order to overcome liabilities of newness, the novice entrepreneur takes advantage of the fact that social capital is locally accumulated. On one hand, it is not surprising that the new firms are not yet fully integrated in the business network. On the other hand, regarding the professional network, new firms have 7.8 local relations, 50% more than the established firms (5.2). This indicates that new firms use the human capital embedded in the local business network to enforce their entrepreneurial competencies.

Using the vocabulary introduced in Table 2, novice entrepreneurs in Lammhult are well integrated in the business community (first-order embeddedness). Regarding second-order embeddedness, we have shown that institutions may even act as bridges between firms not yet connected in the business network in Lammhult (Johannisson et al., 2000). If new firms can benefit from this second-order embeddedness as well, the value of the local context as an incubator will increase.

Next we therefore report how established as well as new firms and their business leaders businesspersons relate to (20) economic and (49) social institutions in the Lammhult context. (The findings concerning the new firms must be treated with care since they are based on responses from only four firms). Our findings shows that established firms do not often use the potential contributions of non-economic capital that economic institutions may provide. Business’ exchange with institutions is even more modest. In fact, new firms are in most aspects more active than established firms in networking with economic institutions.

As regards the established firms, they are again outperformed by the novice entrepreneurs who in all investigated respects are more active building and using non-economic capital. Evidently, social institutions encourage active participation, and it seems that new firms perceive this as an advantage in their venturing career. At social gathering attended by peer entrepreneurs, they have the opportunity to expand their local business network.

Our inquiry into the Lammhult small-firm cluster has demonstrated that the complementary and interrelated networks contain localized non-economic capital for business venturing. The findings indicate a need for additional theories other than those brought by concepts such as ‘strength of weak ties’ (Granovetter, l973) and ‘bridging structural holes’ (Burt, 1997). Firms gain strength individually as well as collectively by combining multiple ties in different networks. Thereby social capital as well as human capital are generated and made available for realizing new ventures. At the same time, a business culture favorable to entrepreneurship is enacted. Strong local connectivity between firms and second order embeddedness in the institutional setting, however, is only one side of the coin. Since many business leaders in Lammhult do not live in the community and since several of the most influential businesses are not locally owned, global frames of references are also present in Lammhult. These may provide a joint bridging over structural holes, combining the benefits of strong local ties and weak global ties. This statement is supported by the indisputable finding that the most successful firms in Lammhult are the ones that rank highly with respect to centrality, i.e., the most socially embedded ones.

The case study of the Lammhult community shows the importance of embedded and networked localities for entrepreneurship. The mixture of different networking settings adds a higher value to business operations. Through this continuous interaction, economic actors gain insight regarding the capabilities of the firms, their reputations and strategic intents, and thus build rich non-economic capital. Firms/entrepreneurs that are less likely to behave opportunistically and innovations are quickly diffused. Obviously, not only business-to-business networks make the local web that operates into an amplifier of local entrepreneurial initiative. Institutional networks contribute to this enforcement. This broadens the scope for adopting a networking strategy when vitalizing communities that have not yet achieved the critical mass needed for takeoff.

To the same extent that ventures need entrepreneurs to initiate and organize renewal and entrepreneurs need triggering events to get started, communities may need community or social entrepreneurs to mobilize collective entrepreneurial processes (Neance, 1998). The new economy seems to enforce these tendencies. Business and local community development, both mobilizing non-economic capital, have to be made to integrate, whether spontaneously as in industrial districts or deliberately with the help of ‘civic entrepreneurs’ (Henton et al., 1997). In fact, even the networked Lammhult cluster has its process facilitator making strategically important contributions to a largely self-organized business community. This opens up for genuinely ‘glocal’ as well as virtual strategies, presumably combining private and public initiatives, where local networking provides a launching pad for firms operating on global markets.

CONTACT: Bengt Johannisson, Växjö University, SE-351 95 Växjö, Sweden; (T) +46-470-70 85 13; (F) +46-470-830 92; bengt.johannisson@ehv.vxu.se

Becattini, G. (1990) “The Marshallian Industrial District as a Socio-economic Notion.” In Industrial Districts and Inter-firm Cooperation in Italy, eds. F. Pyke, G. Becattini & . Sengenberger. Geneva: ILU.

Berger, B., ed. (l991) “The Culture of Modern Entrepreneurship.” In The Culture of Entreprenurship ed. B. Berger, Chapter 1. San Francisco, CA: ICS Press.

Borch, O. J. (l992) “Small Firms and the Governance of Interorganizational Exchange.” Scandinavian Journal of Management 8(4): 321–334.

Borgatti, S., M. Everett & E. Freeman (1999) Ucinet 5.0, Version 1.00. Natick: Analytic Technologies.

Burt, R. (1997) “The Contingent Value of Social Capital.” Administrative Science Quarterly 42(2): 339–365.

Coleman, J. S. (1988) “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94(2):95–120.

Cohen, S. S. & S. Field (1999) “Social Capital and Capital Gains in Silicon Valley.” California Management Review 41(2): 108–130.

Davidsson, P. (l995) “Culture, Structure and Regional Levels of Entrepreneurship.” Entrepreneurship and Regional Development 7(1): 41–62.

Flora, J. L. & J. Sharp  (1997) “Entrepreneurial Social Infrastructure and Locally Initiated Economic Development in the Nonmetropolitan United States.”  38(4): 23–44.

Granovetter, M. (1973) “The Strength of Weak Ties.” American Journal of Sociology. 78(6): 1360–1380.

Granovetter, M. (1985) “Economic Action and Social Structure: A Theory of Embeddedness.” American Journal of Sociology 91(3):481–510). Boston, MA: Harvard Business School Press.

Halinen , A. & J. Törnroos (1998) “The Role of the Embeddedness in the Evolution of Business Networks.” Scandinavian Journal of Management 14(3): 187–205.

Henton, D. M. J. & Walesh, K. (l997) Grassroot Leaders for a New Economy. How Civic Entrepreneurs are Building Prosperous Communities. San Francisco, CA.: Jossey ass.

Johannisson, B., O. Alexanderson, K. Nowicki & K. Senneseth (1994) “Beyond Anarchy and Organization: Entrepreneurs in Contextual Networks.” Entrepreneurship and Regional Development 6(2): 329–356.

Johannisson, B., M. Ramírez-Pasillas and G. Karlsson (2002) “Institutional Embeddedness: The Case of the Lammhult Industry.” Forthcoming in Entrepreneurship & Regional Development.

Ligth I. & S. Gold (2000) Ethic Economics. San Diego, CA: Academic Press.

Lyon, F. (2000) Trust, Networks and Norms: The Creation of Social Capital in Agricultural Economies in Ghana. World Development 28(4):663–681.

Markusen, A. (1997): “Sticky Places in Slippery Space: A Typology of Industrial Districts.” Economic Geography 73(2):293–313.

Marshall, A (l920/l979) Principles of Economics. Eighth edition. London: MacMillan.

Maskell, P., H. Eskelinen, I. Hannibalsson, A. Malmberg & E. Vatne (l998) Competitiveness, Localised Learning and Regional Development. London: Routledge.

Monsted, M. (1995) “Process & Structures of Networks: Reflection on Methodology.” Entrepreneurship & Regional Development, 7(.3): 193–213.

Moog, P. (2000) “Human Capital and its Influence on Entrepreneurial Success.” Paper presented at the Rent XIV Conference on Entrepreneurship and Small Business, Prague, Czech Republic, November 23–24.

Nahapiet, J. & S. Ghoshal (1998) “Social Capital, Intellectual Capital, and the Organizational Advantage.” Academy of Management Review 23(2): 242–266.

Neance, M. B. (1999) “Entrepreneurs in Emerging Economies: Creating Trust, Social Capital, and Civil Society.” Annals of the American Academy of Political & Social Science 565: 148–152.

Porter, M. (1998) “Clusters and The New Economics of Competition.” Harvard Business Review 76(6): 77–90.

Portes, A. (1998) “Social Capital: Its Origins and Applications in Modern Sociology.” Annual Review of Sociology 24 (1): 1–24.

Putnam, R. D. (1993) Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press.

Sjöstrand, S. (1992) “On the Rationale of Irrational Institutions.” Journal of Economics Issues 26(4): 1007–1040.

Starr, J. C. & I. C. McMillan (1990) “Resource Cooptation Via Social Contracting: Resource Acquisition Strategies for New Ventures.” Strategy Management Journal 11: 79–92.

State College (1999) Minitab Reference Manual: Release 13 for Windows. State College, PA: Minitab, Inc.

Uzzi, B. (1997) “Social Structure and Competition in Inter-firm Networks: The Paradox of Embeddedness.” Administrative Science Quarterly 42(1): 35–67.

Wasserman, S. & K. Faust (1994) Social Network Analysis: Methods and Applications. Cambridge, MA: Cambridge University Press.

Weijland, H. (1999) “Microenterprise Cluster in Rural Indonesia: Industrial Seedbed and Policy Target.” World Development 27(9): 1515–1530.

Wenger, E. (l998) Communities of Practice. Learning Meaning, and Identity. Cambridge, MA: Cambridge University Press.

Winborg, J. (2000) Financing Small Businesses—Developing our Understanding of Financial Bootstrapping Behavior. Lund/Halmstad: School of Management and Economics/SIRE.

Centrality Measures

In order to locate the most prominent actors in a network, a large number of definitions of centrality have been proposed. We use the centrality measures: degree, closeness and betweenness (See for example Wasserman & Faust, 1994, p. 46). These centrality measures are also provided as output from network computer packages such as UCINET V. The following are the definitions considered in our estimations.

Degree: The number of vertices adjacent to a given ‘u’ vertex in a symmetric graph is the degree of that vertex. For non-symmetric data, the in-degree of a vertex u is the number of ties received by u and the out-degree is the number of ties initiated by u.

Closeness: The farness of a vertex is the sum of the lengths of the geodesics (shortest paths) to every other vertex. The reciprocal of farness is closeness centrality.

Betweenness: Let bjk be the proportion of all geodesics linking vertex j and vertex k which pass through vertex i. The betweenness of vertex i is the sum of all bjk where i, j and k are distinct and j < k. Betweenness is therefore a measure of the number of times a vertex occurs on a geodesic.