Strategic Organization http://soq.sagepub.com/

The more the merrier? The effect of group size on effectiveness in SME funding campaigns Joakim Wincent, Daniel Örtqvist, Jessica Eriksson and Erkko Autio Strategic Organization 2010 8: 43 DOI: 10.1177/1476127009360402 The online version of this article can be found at: http://soq.sagepub.com/content/8/1/43

Published by: http://www.sagepublications.com

Additional services and information for Strategic Organization can be found at: Email Alerts: http://soq.sagepub.com/cgi/alerts Subscriptions: http://soq.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://soq.sagepub.com/content/8/1/43.refs.html

>> Version of Record - Mar 5, 2010 What is This?

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

Article

The more the merrier? The effect of group size on effectiveness in SME funding campaigns

Strategic Organization 8(1) 43–68 © The Author(s), 2010 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1476127009360402 http://soq.sagepub.com

Joakim Wincent

Luleå University of Technology and Umeå School of Business, Sweden

Daniel Örtqvist

Luleå University of Technology, Sweden

Jessica Eriksson

Umeå School of Business, Sweden

Erkko Autio

Imperial College Business School, UK

Abstract To overcome resource shortages, companies can subscribe to groups that, in competition with other groups, seek to obtain external funding for the joint development of innovations. In this context, the authors argue smaller groups are better equipped to be successful in external funding campaigns. Based on five-year panel data from a sample of 53 Swedish groups of small and medium-sized companies, the authors find support for a claim suggesting that due to the adverse effect of group size on governance and internal cohesion, the costs associated with group size will outweigh benefits, which reduce the ability to compete for external funding. Consistent with their expectations, the authors find that the adverse effect of group size on fundraising effectiveness is mitigated by internal and external governance devices including the presence of external directors on the group board and interlocking board memberships, and by a bottom-up group formation process.

Keywords funding campaigns, fundraising, group size, innovation, resource mobilization, size paradox

Introduction A significant body of research has demonstrated the value of joining forces to develop innovations (Bougrain and Haudeville, 2002; Monge et al., 1998; Oxley and Sampson, 2004; Powell et al., 1996; Stuart, 2000). Inspired by the benefits and motivated by the need to circumvent resource constraints, intentionally formed groups of geographically proximate small and medium-sized

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

44

Strategic Organization 8(1)

companies (SMEs) operating in the same industry sector appear particularly inclined to organize themselves in groups to pursue joint innovation projects (Gomes-Casseres, 2003; Grindley et al., 1994; Hoang and Antonic, 2003; Human and Provan, 1997; Jarillo, 1988; Mauer and Ebers, 2006). In order to finance collective innovation, SME groups frequently seek external funding from public sector agencies (Chaston, 1995; Hanna and Walsh, 2002; Shin, 2004; Thorgren et al., 2009). For example, over a three-year period in Denmark, Chaston (1995) observed more than 2000 firms, organized in groups, competing with other groups for external funding. However, such groups face considerable internal and external challenges when they compete for limited resources such as funding. There are reasons to expect these challenges to grow as a function of the size of the group, because under certain circumstances larger groups are more vulnerable to collective action problems and dispersion of effort (Monge et al., 1998; Olson, 1965), which deduct from SME groups’ ability to compete effectively for external funding. This size issue is important in practice, given that the size of SME groups may vary considerably, from three participants to more than 50 or even 100 participating SMEs. Our objective in this study is to explore how group size matters for SME group funding effectiveness, and which factors mitigate the effect of SME group size on its external fundraising performance. We address this by drawing on group and collective action literatures to develop and test a contingency model. Our argument rests upon the paradoxical effect of group size on the effectiveness of collective action. We argue that although a larger size of the SME group brings many benefits, greater size may also burden the group if not appropriately designed. We argue that the complicating effect of group size is particularly important for groups that compete for limited external resources, where the fruits of collective action may not be shared evenly among group constituents, and where group constituents are also potential competitors. In such situations, which typically characterize SME groups, it is particularly important to invest in enhancing group cohesion in order to prevent group size from becoming a burden that deducts from the group’s ability to compete effectively. To examine the paradox of SME group size, we use a unique dataset of five-year panel data on 53 Swedish groups of SMEs that compete on the same market for external funding. Our dataset covers virtually the entire population of such SME groups in Sweden, and we were able to collect complete data from this population over the entire period of study. This dataset enabled several contributions to past research. Inspired by prior research on groups and the collective action literature (Astley and Fombrun, 1983; Oliver, 1993), we consider a theoretical model that builds on the realization that different group sizes usually result in different organizational formations and preconditions for action. In extending these insights, we suggest two distinct solutions to overcome the burden of group size. These operate through their effect on group cohesion and on the ability of the group to access and act upon external resources. In our empirical study, we predict and test two- and three-way interaction effects which suggest that the influence of group size on success in the amount of external funds retrieved is moderated by the process by which the group is initially formed, as well as the characteristics of its administrative function. By so doing, our study seeks several contributions to prior research. We demonstrate the critical effect of group size on the success of collective efforts in general and on SME groups’ obtaining of external funding in particular. Previous studies have largely focused on various direct influences on collective action platforms’ ability to mobilize resources and on success (Rowley and Moldoveanu, 2003). In this study we consider group characteristics that mitigate the influence of group size on its performance, thereby extending previous literature by providing a contingency perspective to SME groups’ effectiveness for obtaining external funding. Specifically, we show that the effect of group size on performance is mitigated by the number of external members in the

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

45

Wincent et al.

board of the group; the presence of board interlocks; as well as the process with which the group is initially formed. In the following, we develop specific hypotheses on the effect of group size, as well as on the moderating contingencies.

Theory and hypotheses The main influence of group size on external funding The literature on groups and collective action suggests a number of parameters that influence the ability of groups to achieve desired objectives. One central parameter is group size (GomesCasseres, 2003; Oliver, 1993; Oliver and Myers, 2003; Zald and Ash, 1966). Recognizing the complexity in the group size effectiveness relationship, we notice there is a paradox of group size when considering how groups can stand to gain or lose from an increasing number of members (Monge et al., 1998; Oliver, 1993). Several studies have shown that strong, manageable and dense groups are able to obtain greater funding (see e.g. Andrews, 2001). This implies that in large groups, (too) many ties between individuals in social networks can hamper collective action (Siegel, 2009) and thus funding effectiveness. Larger groups of individuals as well as organizations tend to be more complex and require more internal role specialization (McCarthy and Zald, 1977) or support to function effectively in terms of, for example, decision-making and equal participation (Benbasat and Lim, 1993), which ultimately limits their ability to achieve shared goals (see DeNardo, 1985). Because smaller groups are generally more cohesive than larger groups, it may be easier for smaller groups to agree on shared goals and foster a shared collective identity that harmonizes with the identities of each member in the group (Olson, 1965). These aspects facilitate coordination and collective action (see e.g. Gamson, 1995; Klandermans and deWeerd, 2000; Monge et al., 1998; Polletta and Jasper, 2001). A smaller group size may also make it easier for the group to ‘speak with one voice’, and thus influence the target with a coherent message (Dominelli, 1996). This capability may be gradually lost with increasing group size, as the likelihood of inconsistent messages issued by group members tends to increase with group size. Received research has also highlighted benefits associated with larger group size. For example, a larger number of group members may confer greater political power, thus helping convince targeted external stakeholders (McAdam, 1982). Larger groups may speak with a louder voice and demonstrate potentially more impressive support for the shared goals pursued. Larger groups are also able to harness more resources for the pursuit of shared goals, thereby increasing the potential external effectiveness of the group (Oliver, 1993; Zald and Ash, 1966). Ultimately, the size of the group should thus provide better chances of achieving the collective goals, provided that the strength in numbers is successfully converted into effective action (DeNardo, 1985). Supporting these notions, the study by McCarthy and Wolfson (1996) showed that the size of task committees contributed positively to the amount of funding obtained in campaigns for local government. In the preceding paragraphs, we have considered the benefits and costs of group size at a general level and noticed the complex and even paradoxical state of group size. However, the types of mechanisms evoked (e.g. coordination, cohesion and decision-making) imply that both the benefits as well as costs may accrue differently in different types of groups, possibly altering the relationship between group size and achievable benefits depending on how the group is composed (GomesCasseres, 2003). We notice that groups of SMEs created with the objective of raising funds for shared innovation activities differ in important ways that are likely to alter the relationship between group size and

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

46

Strategic Organization 8(1)

realistically achievable benefits. The type of SME groups that we focus upon in our study are typically (although not exclusively) formed to pursue economic benefits by generating funding for joint innovation. Therefore, the prospect that some group participants may benefit significantly more than others creates potential for tension (Monge et al., 1998) that is likely to accrue with group size. Further, SME groups are typically formed among potential competitors (Grindley et al., 1994). Although the groups pursue shared goals, they may have different ideas about the scope of cooperation, and there is always a possibility that the group action may end up benefiting some more than others (Khanna, 1998; Wincent, 2008). To gain their benefits, SMEs are likely to have to invest their resources ex ante in order to achieve shared goals, creating opportunities and temptation for free-riding (see e.g. Grindley et al., 1994; Monge et al., 1998; Olson, 1965; Powell et al., 1996). These aspects mean that, in order to function, SME groups have to invest significant attention and resources to build and nurture trust, establish and reinforce shared norms of reciprocity and create norms that punish opportunistic behaviour. As the scope and opportunities for freeriding tend to increase rapidly as a function of group size, it is likely that complexities accruing with SME group size rapidly outweigh potential benefits (see Garciá-Canal et al., 2003). A practical illustration of such diseconomies of scale is the fact that most collective action platforms created by only two firm constituents, i.e. joint ventures, tend to fail (Das and Teng, 1998) and that the complexity in a collaboration increases in alliances with several partners (Garciá-Canal et al., 2003; Robson et al., 2008; Zeng and Chen, 2003). This argumentation implies that, ceteris paribus, smaller SME groups should have advantages over larger SME groups in generating external funding. Because innovation is characterized by high uncertainty, and because of additional complexity, it should be as difficult, if not more so, for a larger SME group to organize successful fundraising campaigns. We therefore hypothesize: 1: Group size influences the amount of external funding obtained by SME groups: smaller groups are generally better able to obtain funding than are larger groups.

hypothesis

The moderating effect of external members and interlocks in the group’s administrative function In the preceding, we have argued that in collective action platforms characterized by physically limited potential benefits (i.e. limited funding to be shared by participants), such as SME groups, complex interactions and decision-making among constituents and potential conflicting interests among constituents, the potential benefits of larger group size tend to be outweighed by potential dysfunctionality. The performance of larger SME groups thus hinges on their ability to mitigate potential cooperative problems and conflicts of interest among its constituents, as well as the ability to reinforce cohesion and coordinate collective action. We consider two such coordination-enhancing and conflict-mitigating mechanisms. First, the number of external directors appointed to the board of the SME group should enhance the group’s internal cohesion as well as its ability to prioritize external relationships. In a related vein, interlocking board relationships with other network organizations should enhance the ability of the SME group to compete against other groups for limited external resources. Second, we examine the possibility that the initial organization of the SME group, specifically, its composition through a bottom-up process, will be more conducive to coherence and cohesion than will an initial organization through a top-down process. The group and collective action literature has recognized the importance of the ‘involvement on the part of individuals and organizations from outside the collectivity’ for the group’s external

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

47

Wincent et al.

effectiveness (McCarthy and Zald, 1977: 1216). The administrative function of an SME group can include external representatives, i.e. persons not affiliated with constituent SMEs. The presence of externals in the administrative function of an SME group provides a governance-enhancing device that can enhance both internal cohesion and external effectiveness. In terms of internal cohesion, external members are less likely to carry vested interests, which would enable them to spot and suggest optimal internal collaborations within the group without being constrained by past collaborations and relational impediments (Burt, 2005). In terms of external effectiveness, external members can enhance the legitimacy of the group in the eyes of key external stakeholders. For instance, in the case of non-profits, Galaskiewicz and Wasserman (1989) found that elite attitudes towards a non-profit increased the likelihood of it receiving donations. It can hence be assumed that campaign success is more likely when external board members provide a possibility to form alliances with ‘elites’ (McAdam, 1996). In SME groups, external members are normally invited to the administrative function because of their social network, reputation, or competence. External members of the administrative function can therefore provide a greater number of links to diverse other units (Burt, 1980), thereby improving the potential to access resources (Bougrain and Haudeville, 2002; Kapferer, 1969) and transmit information from a variety of sources (Lynall et al., 2003). We posit that the effectiveness of external members in the group administration depends on the size of the SME group, with the effect that they help mitigate size-related disadvantages in larger SME groups. In small SME groups, internal cohesion is rarely a major issue. As the group size grows, however, internal coalitions are more likely to form around leading companies, and the constituent SMEs may get locked into such coalitions. Hence, in larger groups, structural holes would be more likely to emerge, deducting from internal cohesion. Because external members are free from pre-existing relational impediments, they would be able to more objectively generate links within the group between ‘people with different skills, information, and experience’ and thus bridge structural holes (Reagans and Zuckerman, 2001: 504; see also Burt, 2005). External members within the group administration may also be able to more effectively mitigate problems brought about by group size in external relations. Larger SME groups typically face more complex webs of external stakeholders than do smaller SME groups, which typically tend to address and deal with only one funding agency. As the number of potentially addressable external opportunities increases with the size and related heterogeneity of the SME group, external members may more effectively help the group to make sense of and use related information as well as opportunities recognized (Oliver and Marwell, 1988). In a related vein, because larger SME groups are able to address a potentially larger number of external opportunities, the diverse information and expertise carried by external members of group administration may be utilized more effectively than in smaller groups which may find it bothersome to make use of all opportunities and relationships (see Ahuja, 2000). As the web of potential external stakeholders (i.e. potentially addressable funding opportunities) grows larger and more complex with group size, the legitimacy-enhancing effect of external members of group administration is likely to grow to be more relevant than what would be the case in smaller SME groups, where the funding landscape is less complex. The aforementioned considerations suggest that external members of group administration may more effectively enhance internal cohesion in larger SME groups. In addition, in larger SME groups, external members of group administration may more effectively contribute to external legitimacy, and their skills and strategic insight may be used more effectively than in small SME groups. Because of these effects, we predict that the number of externals appointed to the board moderates the relationship between group size and the amount of external funding obtained:

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

48

Strategic Organization 8(1)

2: The number of externals on the board of the group moderates the negative influence of group size on the amount of external funding obtained: the negative effect of group size on its ability to raise funding decreases as the number of external board members increases.

hypothesis

We also expect that, ceteris paribus, group boards whose members are also on the boards of other groups (i.e. interlocking directorates) will be more effective in obtaining external funding. A number of studies have found that interlocks are important (e.g. Carpenter and Westphal, 2001; Schoorman et al., 1981). An SME group with a large number of interlocks would be more centrally positioned among groups, thereby signalling that the group is part of an elite and enhancing its prestige and reputation (Davis et al., 2003; Diani, 2003). Further, network centrality is likely to be an indication of prior experience of networking and therefore an improved ability to collaborate and manage collaborations (Powell et al., 1996). Board interlocks would also signal higher competence and strategic insight of the board members (Rowley and Moldoveanu, 2003), enhance the exchange of resources and information (Pfeffer and Salancik, 1978), as well as improve vertical and horizontal coordination (Schoorman et al., 1981) and social cohesion (Mizruchi, 1996). Evidence supporting such notions was provided by Gould’s (1991) study of the 1871 Paris commune, which showed that linkages among neighbourhoods enhanced access to resources and information, and that the efficacy of these linkages was dependent on cohesion among leading participants, as expressed in dense, informal social ties within neighbourhoods. Compared with external board members, interlocks provide additional advantages, as they provide insider information about relevant competing campaigns pursued by other groups. Such information from the support base enables groups to differentiate themselves from other groups (Zald and McCarthy, 1980) when targeting a particular funding call. Such differentiation is a result of enhanced interorganizational coordination and specialization, through which the interlocks facilitate the distribution of work among the groups involved, thereby reducing competition between the SME groups concerned (Palmer, 1983; Schoorman et al., 1981). In the preceding, we have considered the benefits of board interlock on a ceteris paribus basis. There is reason to expect, however, that board interlocks will not be equally effective in SME groups of different sizes. Specifically, we suggest that interlocks should have a significant role in mitigating the negative costs that are associated with an increasing group size in strategic SME networks. As in the case of external members of the group administration, their greater insight and knowledge regarding alternative funding opportunities should be particularly effective in helping large groups to navigate complex funding landscapes. Board members with many interlocking ties are likely to have experience with more funding agencies, making them better able to select among and pursue the multiple funding opportunities available for larger SME groups. For larger SME groups, the strategic positioning within the funding landscape is more important than for smaller SME groups, because they will be looking for a greater share of the limited resources made available for SME groups in general. The strategic insight of members with interlocking ties should come in handy when positioning large SME groups correctly so as to maximize success in fundraising (Davis et al., 2003). Smaller SME groups may not be able to fully utilize the broader skills and information provided by board interlocks, because their limited resources restrict the number of addressable funding opportunities (Oliver and Marwell, 1988). Limited resources in combination with many interlocks may lead to a situation in which the overall opportunity for strategic action is curtailed, and the existence of many interlocks may therefore have a negative effect on the group’s ability to obtain funding. We therefore anticipate that having many interlocks is particularly important for larger SME groups:

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

49

Wincent et al. 3: The number of interlocks in the group board moderates the negative influence of group size on its ability to raise external funding: the negative effect of group size on the amount of funding obtained decreases as the number of interlocks increases.

hypothesis

The role of group formation in moderating the size-effects on external funding In the preceding paragraphs, we have considered the moderating effect of governance devices on the fundraising success of SME groups. These operate mainly (although not exclusively) through their effect on the group’s external efficacy. Here we posit that the formation process of the group will also moderate the effect of group size on its fundraising efficacy. The internal organization of this group will carry long-lasting traces of its formation process, that is bottom-up or top-down processes. Although both approaches can be effective in forming the SME group, studies show that the forces and processes towards formation elicit different end results (Tell, 2000). Groups created through a top-down formation process are typically initiated by a limited number of group participants to implement a fixed agenda that the group members need to subscribe to when joining. In such groups, it tends to be the agenda that drives group composition and not vice versa (Chaston, 1995; Huggins, 2001; Varamäki and Vesalainen, 2003). Groups developed by a bottom-up formation process grow organically, with additional actors joining the group over time on a voluntary and rather ad hoc basis. The group’s agenda is likely to evolve over time, in parallel with the group structure (Ahlström-Söderling, 2002). It has been argued that bottom-up processes strengthen the dialogue among group constituents and thus foster a stronger group identity (Fukugawa, 2006; Scully and Creed, 2005), internal cohesion and increase the time and resources devoted by constituents to activities within the group (Halme and Fadeeva, 2001; Huggins, 2001; Weber et al., 2008). As the group’s agenda evolves organically, it is more likely to balance the varying needs of the different constituents, thereby reducing goal dissonance. Overall, therefore, bottom-up processes can reduce problems of commitment and motivation among group members (Fukugawa, 2006; Weber et al., 2008). One can expect that, ceteris paribus, a bottom-up process should be more helpful than a top-down process for an SME group’s ability to attract external funding. Although potentially important as such, we argue that a bottom-up process is even more important for large SME groups than for small SME groups. Previously, we have observed that group size is likely to be negatively associated with internal cohesion. The larger the group size, the more difficult it is to agree on a shared agenda and goals among different members (for a practical example, see Grindley et al., 1994, on the development of the SEMATECH consortium). In small SME groups, agendas are easier to set, and cohesion-enhancing devices are less likely to be needed (Rowley and Moldoveanu, 2003). Larger SME groups can gain more from enhanced cohesion because of the scope of potential synergies that can be exploited among group constituents (Monge et al., 1998), in the form of enhanced commitment, increased willingness to invest time and resources and stronger identification with the funding campaign. These benefits will mitigate the costs associated with larger group size more effectively than in smaller groups simply because the scope of potential synergies is greater. We argue that the situation may be the inverse in small size groups, where a top-down approach may actually, and counterintuitively, be preferable to a bottom-up approach. Whereas it is difficult to build a strongly cohesive large group, a group with fewer members may be able to reap the benefits of a top-down approach because there are fewer interests to combine, and negotiations to reconcile differences are therefore less complex (AhlströmSöderling, 2002). A bottom-up approach may even be detrimental for smaller groups in pursuit of external funding, as the cohesiveness in such groups is likely to be very strong to the point of

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

50

Strategic Organization 8(1)

inbreeding (Oliver and Myers, 2003), which prevents smaller groups from recognizing and addressing funding opportunities and funding calls. Thus, we hypothesize: 4: The formation process of the group moderates the negative influence of group size on the amount of external funding obtained: the negative effect of group size on its ability to raise funding will be smaller with bottom-up formation

hypothesis

Three-way interactions: group size, board composition and formation process In addition to two-way interactions between governance devices, internal organization and group size, there is also reason to expect that the internal organization (i.e. bottom-up formation process) of the group is particularly important for larger groups that fail to implement governance devices (i.e. fail to assign external members to the group board or foster board interlocks). The crux of this argument rests on the importance of group architecture for successful campaigns (McCarthy and Wolfson, 1996). As we have suggested, the architecture of groups hinges on several aspects, each of which contributes in isolation and in interaction with the size of the group. As such, we argue that both aspects of the internal organization (i.e. bottom-up process) and governance devices (i.e. number of externals appointed to the board and board interlocks) jointly influence each other and contribute to the potential success of the group in obtaining external funding. The number of externals appointed to the board and board interlocks contribute to group architecture by enhancing larger groups’ potential to coordinate resources effectively (Kapferer, 1969). They also facilitate identity creation and differentiation relative to other groups, aspects that are necessary to achieve success in funding campaigns (see Rowley and Moldoveanu, 2003). The internal organization – bottom-up formation – also influences identity formation in the group, facilitates the development of, and commitment to, shared goals and strengthens the members’ willingness to contribute their time and resources to funding campaigns (Halme and Fadeeva, 2001; Huggins, 2001). Although both enhanced access to external resources through governance devices and the internal organization of the group are important, success in funding campaigns more likely depends on the existence of one characteristic of the architecture if the other is lacking. For instance, in a top-down process, the contribution to joint identification and commitment by enhanced access to external resources can replace the lack of commitment and group identification that results from a discordant internal organization. In line with our previous argumentation that smaller groups have more natural advantages, and that larger groups often need to overcome issues related to control, management and organization, we expect the potential gains from relational and formational characteristics to be especially prominent in larger groups. Therefore, we propose the following hypotheses: 5: The number of external members on the board of the group and the formation process of the group jointly moderate the influence of group size on the amount of external funding obtained: when larger groups do not have board externals, a bottom-up formation is more important for improving the amount of funding obtained.

hypothesis

6: The number of interlocks in the group’s board and the formation process of the group jointly moderate the influence of group size on the amount of external funding obtained: when larger groups do not have board interlocks, a bottom-up formation is more important for improving the amount of funding obtained.

hypothesis

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

51

Wincent et al.

Research methods Research setting and data collection procedure The sample of this study consists of five-year panel data from 53 groups of Swedish SMEs created to seek funding for shared innovation activities. Our single-country dataset has important merits. All the SME groups in our sample operated in the same market for public funding and dealt with the same set of funding agencies. This design eliminates exogenous variance in our data. Because SME groups mostly deal with public sector funding agencies, the use of single-country data was important in order to eliminate variance in institutional conditions, such as differences in regulatory regimes. Furthermore, all 53 SME groups in our dataset were intentionally formed groups of geographically proximate SMEs operating in the same industry. The limited variance in firm size reduces the risk of variance in groups’ internal processes due to the type of constituents. The use of groups made up of geographically proximate SMEs that operated in the same sector minimized heterogeneity in resource mobilization processes. It is also worth noting that our dataset is longitudinal panel data, which enabled us to control for group-specific effects, thereby reducing potential problems with endogeneity. No complete register of SME groups existed at the time when we started this research. Fortunately, as Sweden is a small country, we were able to contact all relevant funding agencies and develop a full register by ourselves. The population used for our study was identified by contacting municipal authorities in every Swedish county, as well as Sweden’s industrial development centre (IUC) and the Swedish state-owned development company ALMI Företagspartner. We also contacted the Swedish technology development agency NUTEK. Through these overlapping queries, we identified 53 groups that fit our definition of SME groups (see also Human and Provan, 1997) and had shown some interest in obtaining funding. Because we searched all relevant Swedish authorities and organizations, some regional and others national, and because the mandates of various authorities overlapped with one another to a significant degree, we are confident that our initial list covered virtually the entire Swedish population of SME groups that matched our selection criteria. Once we had identified the SME groups, we contacted a representative of each group by telephone and by mail. These representatives often worked full-time for the groups and maintained relevant registers and secondary data necessary for our research. All groups agreed to participate. Thus, our dataset carries complete data from the entire population of Swedish SME groups that met our initial selection criteria. The data used for this study covers a five-year period, with observations from each group for each year. The result was 265 observations based on the five-year annual data on the 53 groups. In order to validate and cross-check the data obtained from the SME groups, we also accessed reports and data directly from the principal funding agencies. These efforts ensured that we have complete, cross-checked panel data over the five-year period of study. The average group age was nine years, and 20 percent had received more than 1,000,000 SEK (US$143 000; US$1 ≈ 7 SEK) in external funding. In terms of geographical distribution, the groups were rather equally spread across Sweden, which indicates that there should be at least some competition among groups in the study to obtain external funding. On average, the groups had 24 active members. One of the typical groups in the sample – YWOOD – consisted of some 50 SMEs with a common focus on developing new products for the wood industry and experimenting with complementary product concepts. The core idea of the group was to combine inputs from individual SMEs to provide a complete range of products so that potential customers could use the network to satisfy all their needs of wood-based products. The group also aimed to have a competitive and innovative product base to offer to global retailers. To meet this shared goal, the participating firms had applied for external funding to develop both manufacturing equipment for processing wood in

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

52

Strategic Organization 8(1)

new ways, as well as a range of new products that the member firms have subsequently exported with good success.

Dependent variable Our dependent variable was the natural logarithm of the yearly amount of external funding across years 2000–4. We found that the correlation between the years was, on average, .66, which indicates that the groups were not always successful in their funding campaigns and not necessarily able to attract funding each year. No clear patterns or trends existed in the data. Overall, the groups were most successful in obtaining funding in 2003 and less successful in 2000, but firm-level and group-level inspections revealed no linear trend in the dataset in this regard.

Predictor variables To test hypothesis 1, the main predictor used was group size, measured as the number of member firms in the group, using the amount of funding as the dependent variable. For hypothesis 2, we calculated the mean-centred interaction between group size and the number of external members in the group board, measured as the number of people on the group’s board who were not from a member firm of the group. To test hypothesis 3, we interacted group size (mean-centred) with interlocking directorates (mean-centred), measured as the total number of positions that board members held on other group boards. These measures are consistent with the traditional measures in the board literature (Boyd, 1990). Overall, the range of the variables was wide. Some 45 percent of the groups had external board members, and 17 percent had at least one board member with a position on another group board. As expected from prior research (Ahlström-Söderling, 2002), there was some variance in the number of active member firms across years. The correlation across years was, on average, .55. Moreover, 11 percent of the SME groups had more than 50 members, and 50 percent had fewer than five members, which indicated significant variance in group size. To test hypothesis 4, we used an approach similar to that of Doz et al. (2000) and calculated bottom-up formation as a dummy variable, where 1 corresponds to groups in which firms initiate group formation and 0 corresponds to a top-down initiative, such as institutions or governments. We then interacted this variable with group size. With respect to formation processes, 38 percent of the SME groups originated with a bottom-up approach. We calculated three-way interactions by interacting mean-centred values of group size, formation process and the number of external members in the group board to test hypothesis 5, and by interacting group size, formation process and interlocking directorates to test hypothesis 6.

Control variables From the primary data collection, we adopted several control variables. To address the issue of unobserved heterogeneity and to control for the fact that prior success may enhance the network’s reputation with potential sources of funds and increase its ability to obtain more funds in the future, we included the lagged effect of external funding, measured as the natural logarithm of the amount of external funding at a one-year lag of the dependent variable. To account for the potential difference in self-reliance and resource accumulation, we also included a control for expenditures on R&D activities in the form of the amount of group

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

53

Wincent et al.

internal spending on innovation activities. We also included size of the group board, measured as the total number of board members (Goodstein et al., 1994), and age of group, measured as the number of years since the group had been initiated. The latter two can be considered proxies for legitimacy, and they capture possible disturbing variance that we might have obtained otherwise by using the previously described independent variables. We acknowledge that there may be political factors involved when external funding decisions are made. Government and public agencies may be more inclined to sponsor certain groups on the basis of premises that go beyond our hypotheses and what the groups themselves cannot control, which ultimately could make a given group more or less attractive for obtaining funding. To control for such variance when testing our hypotheses, we were able to access annual public reports from the potential fundraisers and other public registers that contain information that could have constituted a basis for such a decision. This enabled us to control for political bias with the following variables. First, it was observed that the geographical position of a group potentially influenced the degree of funding obtained. In certain peripheral regions, Swedish government policy allowed higher rates of funding. Therefore, we controlled for the geographical location of the SME group. With data from Statistics Sweden, we included measurements for each group on the regional gross domestic product (RGDP), calculated as current prices in millions of Swedish kronor (US$100 ≈ 700 SEK), and the sum of regional compensation, a measurement of political efforts to equalize differences in tax potential (income equalizing) and structural conditions (cost equalizing) across different regions. Although different, both variables reveal the availability of funding in the regions, whether from own earnings (e.g. RGDP) or policy efforts (e.g. regional compensation). We also controlled for whether the groups were located in prioritized regions by a dummy variable (1 = groups located in prioritized regions). This measure was developed from official political reports published by NUTEK (www.nutek.se). Finally, we controlled for regional competition by estimating the number of groups located in each region. We did so because competition may influence the group’s potential to attract external funding. We calculated how many groups were located in each county, and then used this value as a measure of competition for the respective group. We calculated all regional measures at the county level.

Analytic approach Because our observations were not independent, we consolidated our data over the five-year window. To analyse the data we therefore used a random-effects generalized least squares model, which corrected for first-order autoregressive (AR) disturbance terms. This method is effective for correcting endogeneity problems because the model enabled us to account for biased estimators due to possible heteroscedasticity and serial correlation, as well as nonrandom distribution of residuals (Bascle, 2008; Hamilton and Nickerson, 2003). We did not use a fixed-effects specification because the variance of the independent variables between the years was limited and the formation variable (bottom-up or top-down) was stable across years. Moreover, the Hausman test failed, indicating that random effects were more appropriate. Our main tests of the hypotheses are based on annual panel observations without any time lag between the independent variables and the dependent variable, the amount of external funding obtained each year. Because the output or success of a funding campaign for the groups studied is likely to occur within a one-year time frame, we considered this approach to provide the best reflection of the phenomenon studied.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

54

Strategic Organization 8(1)

Results Table 1 reports the correlations, means and standard deviations for all variables. The average group obtained some 900,000 SEK (or US$128,571; US$1 ≈ 7 SEK) in external funding each year, on average. External funding varied between 0 percent and 60 percent with the highest amount obtained in a single year of 15 million SEK (equivalent to US$2.1 million). Yet, only about 20 percent of annual funding receipts exceeded 1 million SEK (equivalent to US$140,000) in yearly funding. The average group consisted of 24 firms and spends about 700,000 SEK of internally raised funds on joint innovation activities (equivalent to US$100,000). The distribution shows that 50 percent of the groups consisted of some five members, and that the third quartile starts at some 25 members. The board size averages 5.5 members with some 40 percent having a single representative to some 25 percent having more than eight members. On average, the boards comprised 1.5 external members. However, this varied between no external members (60 percent of total) to between two and 13 external members (25 percent of the total). The average number of board members having interlocking directorates was approximately 0.45. Some 12 percent of the boards had between one and 12 interlocking directorates. Finally, 38 percent of the groups were initiated bottom-up. Overall, the results show significant correlations between the independent and dependent variables. Foremost, we found a significant and negative relationship between group size and external funding in accordance with our hypothesis (r = −.46; p < .001). Further, our evidence also shows that the direct effects of expected moderating variables were significantly related to external funding. Accordingly, the number of external members of the board was negatively related to external funding (r = −.21; p < .01). The threat of multicollinearity is low: the highest correlation coefficient is about .65 and is given for the relationship between the control variable of in-group R&D spending with group size. Table 2 presents the results of the hypothesis tests. The first model includes the control variables, where previous external funding and R&D spending shows up as significant effects on external funding. The results indicate that previous external funding increases the odds of receiving large external funding, and also that groups which raise more money for innovation activities internally in general receive smaller amounts of external funding. The second model adds direct effects of group size, board externals, board interlocking directorates and bottom-up formation. Results indicate that the significant control variables exhibit the same strength and that, in addition, group size is significantly and negatively related to external funding (β = −.13; p < .001), thereby supporting hypothesis 1. We also performed a set of tests to explore the validity of our conceptual arguments about size. Specifically, when omitting the inclusion of board externals, board interlocking directorates and bottom-up formation to the second model we noticed a difference in the magnitude of the influence of group size on external funding (β = −.21; p < .001). In support for the outlined contrasting effects of size, not keeping the direct effects of board externals, board interlocking directorates and bottom-up formation constant strengthened the negative relationship between group size and external funding. The third model adds a number of two-way interactions to test hypotheses 2 and 3. As evident, the two-way interaction model also reports two of the hypothesized interactions to be significantly related to external funding. In further evaluation of the hypothesized relationship, we calculated and plotted two-way interactions. In plotting such relationships we computed the three levels of high, medium and low (for both the continuous main effect as well as the continuous moderating variable) using the mean as the medium value, one standard deviation above the mean as the high mean and one standard deviation below the mean as the low mean (following Aiken and West,

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

.09

2096.95 -.56***

917.05

N = 265. *p < .05; **p < .01; ***p < .001. Two-tailed significance.

-.12

.22***

0.49

0.38

-.09 -.11 -.04 -.08 .01

-.14*

.02

.05

.14*

-.13* -.14* .18** .19** .07 .03 .19** .17** .01 -.14*

.41*** -.11 -.19**

.28***

.14* -.26*** -.46*** -.21** -.02

.34*** .16** .17**

-.05 1.00 .23*** .31*** 1.00 .12 .18** .16** 1.00 .19** .16** .52*** .17** 1.00 .01 .09 .04 -.01 .04 1.00

1.00

.50***

-.14* .52***

.47*** 1.00

-.30**

6.37 .44*** 7.44 .33*** 73.85 .65*** 2.57 .19** 1.65 -.01

.08

.02

3.81 5.51 23.81 1.50 0.45

10

5.86

9

7.53

8

0.45 -.10

7

0.72

6

-.31*** 1.00

5

2908.12

4

10949.84

3

1960.54 1. 00 59343.04 -.04 1.00

2

724.68 68101.26

1

  1. R&D spending   2. Regional gross domestic product   3. Sum of regional compensation   4. Location in prioritized region   5. Regional competition   6. Group age   7. Board size   8. Group size   9. Board externals 10. Board interlocking directorates 11. Bottom-up   formation 12. External funding

SD

Mean

Variable

Table 1.  Descriptive statistics and correlation matrix

-.10

1. 00

11

Wincent et al. 55

56

Strategic Organization 8(1)

Table 2.  Influences on external funding

Model 1



Coef.

SE

Model 2 Coef.

Model 3 SE Coef.

Model 4 SE Coef.

SE

Controls and independent variables Previous external .55*** (.05) funding R&D spending -.29*** (.05) Regional gross .02 (.05) domestic product Sum of regional .01 (.06) compensation Location in prioritized .03 (.06) region (.06) Regional competition .01 Group age (.05) .06 Board size -.08 (.05) Group size Board externals Board interlocking directorates Bottom-up formation

.53***

(.05)

.23***

(.04)

.21***

(.04)

-.30*** .03

(.06) (.05)

-.43*** -.03

(.05) (.06)

-.58*** -.05

(.05) (.05)

.02

(.06)

-.03

(.06)

-.04

(.05)

.03

(.06)

.04

(.07)

.05

(.06)

-.01 .05 -.06 -.13** -.06 -.02

(.06) (.06) (.06) (.04) (.06) (.05)

-.11 .03 .06 -3.64*** .17 .23

(.07) (.07) (.06) (.28) (.15) (.13)

-.03 .16 -.01 -3.38*** .21 .28*

(.06) (.05) (.05) (.24) (.13) (.11)

.05

(.06)

.12

(.07)

.21***

(.05)

.65***

(.14)

3.67***

(.36)

Two-way interactions Group size * Board externals Group size * Board interlocking directorates Group size * Bottom-up formation Board externals * Bottom-up formation Board interlocking directorates * Bottom-up formation

-.23

(.14)

1.79***

(.15)

1.72***

(.14)

-.16*

(.08)

-.20**

(.07)

-.11

(.10)

-.07

(.08)

.50

(.35)

Three-way interactions Group size * Board externals * Bottom-up formation Group size * Board interlocking directorates * Bottom-up formation Constant R2 χ2; d.f.; sig.



.03 (.05) .70 298.20;9;***

.03 (.05) .70 277.48;13;***

-.22*** (.06) .68 463.82;18;***

N = 265. Standard errors are in parentheses. *p < .05; **p < .01; ***p < .001. Two-tailed tests.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

-2.01***

(.23)

-.60*

(.24)

-.25*** (.05) .77 829.29;20;***

57

Wincent et al.

1991). In support of hypothesis 2, the number of board externals moderates the influence of group size on the amount of external funding obtained. The positive coefficient for the interaction term of board externals and group size (β = .65; p < .001) implies, and simple slope tests affirm, that additional externals on the board positively influence the funding outcomes of large groups (slope = 3.88; T-value = 9.94; p < .001) but negatively influence the funding outcomes of small groups (slope = −3,46; T-value = −9.30; p < .001). As revealed, and in line with hypothesis 2, larger groups gain more from appointing externals in terms of achieving external funding compared to smaller groups who in fact gain more from not appointing externals. As such, smaller groups might be better off without a high number of externals. This is also depicted in Figure 1. In contrast, hypothesis 3, which posited that the number of board interlocks moderates the influence of group size on the amount of external funding obtained, was not supported (β = −.21; p > .05). However, we found significant support for hypothesis 4 and the interactive influence between group size and formation process on the amount of external funding in both models (β = 1.79; p < .001). Based on the coefficient and simple slope tests, top-down initiatives compared to bottom-up initiatives positively influence the funding outcomes of large groups (slope = 1.93; T-value = 12.73; p < .001) but negatively influence the funding outcomes of small groups (slope = −1.51; T-value = −10.79; p < .001). As Figure 2 also illustrates, the form of the interaction was consistent with our hypothesis. In larger groups, a bottom-up formation increases the amount of funding obtained; in smaller groups, the slope is slightly negative. The fourth model adds the two three-way interactions. Overall, there is convincing support in the results for the two three-way interactions as posited in hypotheses 5 and 6. Both threeway interaction terms are negative, as we had expected, and the inclusion of the interaction terms contributes to an increase in explained variance (R2 = .77). The model as presented in Table 2 supports hypothesis 5, which posited that the number of board externals and the group

Figure 1.  Group size and board externals interaction

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

58

Strategic Organization 8(1)

Figure 2. Group size and formation process interaction

formation process jointly moderates the influence of group size on the amount of external funding obtained. We found evidence that when larger groups do not have board externals, a bottom-up formation is critical to improving the amount of funding obtained. Figure 3 also displays graphical support for this interaction. Table 3 reveals slope analysis according to Aiken and West (1991) for the four slopes included in Figure 3. All four slopes are significant and negative, thus supporting that increasing group size is negative for external funding regardless of the moderating influences focused upon in this research. However, the table also shows that given the moderations the magnitude of the negative effects differs between the slopes. In accordance with hypothesis 5, the table shows that the worst strategy with increasing group size for achieving external funding is to have few board interlocking directorates and to have a top-down formation process. The most rewarding of the four strategies seems to be to have a bottom-up formation process (i.e. a high bottom-up formation), yet the additional difference by number of board interlocking directorates seems to be marginal at best. Table 4 reports the simple slope difference test following procedures by Dawson and Richter (2006). This test shows that there is no significant difference in the two strategies building on having a bottom-up formation process (i.e. high bottom-up formation). However, in support of hypothesis 5 we find that strategies to reduce negative effects on external funding for groups with increasing size which builds on increasing levels of bottom-up formation and board interlocking directorates generally outperform strategies that build on top-down formation processes and a low number of board interlocking directorates. We found convincing support for hypothesis 6 in the results, which indicated that the number of board interlocks and the group formation process jointly moderate the influence of group size on the amount of external funding. We found support for our predictions, which suggest that when larger groups do not have board interlocks, a bottom-up formation is critical to increase the amount of funding obtained. Figure 4 also illustrates support for this

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

59

Wincent et al.

Figure 3.  Group size, board externals and formation process interaction Note: Dependent variable: external funding.

Table 3.  Simple slope analysis – group size, board interlocks and formation process interaction (1) at high (2) at high (3) at low (4) at low

Slope Board interlocking directorates Board interlocking directorates Board interlocking directorates Board interlocking directorates

and high and low and high and low

T-value Bottom-up formation = –1.76 Bottom-up formation = –4.00 Bottom-up formation = –1.56 Bottom-up formation = –6.20

–3.46*** –3.06** –7.85*** –12.17***

Note: Simple slope analysis according to Aiken and West (1991).

Table 4.  Simple slope difference test – group size, board interlocks and formation process interaction Differences between the slopes (1) and (2) (1) and (3) (2) and (4) (3) and (4) (1) and (4) (2) and (3)

ptwo-tailed = .000*** ptwo-tailed = .816 ptwo-tailed = .011* ptwo-tailed = .000*** ptwo-tailed = .000*** ptwo-seitig = .001**

t = 4.026 t = –.233 t = 2.565 t = 8.339 t = 5.866 t = –3.224

Note: Simple slope difference tests according to Dawson and Richter (2006).

interaction. Table 5 reveals the slope analysis for the four slopes included in Figure 4. The table shows that both strategies involving a bottom-up formation process cannot be significantly established to have an influence on external funding. However, both strategies involving a top-down formation process are negatively related to external funding. In further support of hypothesis 6, we find the worst strategy for larger groups is to have a top-down formation process and a minimum of board externals. Table 6 reports the Dawson and Richter (2006)

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

60

Strategic Organization 8(1)

Figure 4.  Group size, board interlocks and formation process interaction Note: Dependent variable: external funding.

Table 5.  Simple slope analysis – group size, board externals and formation process interaction (1) at high (2) at high (3) at low (4) at low

Slope Board externals Board externals Board externals Board externals

and high and low and high and low

T-value Bottom-up formation = .00 Bottom-up formation = –3.32 Bottom-up formation = .58 Bottom-up formation =–10.78

.001 –7.62*** 1.33 –24.73***

Note: Simple slope analysis according to Aiken and West (1991).

Table 6.  Simple slope difference test – group size, board externals and formation process interaction Differences between the slopes (1) and (2) (1) and (3) (2) and (4) (3) and (4) (1) and (4) (2) and (3)

ptwo-tailed = .000*** ptwo-tailed = .000*** ptwo-tailed = .000*** ptwo-tailed = .000*** ptwo-tailed = .000*** ptwo-seitig  = .000***

t = –4.964 t = 4.910 t = 16.795 t = 63.975 t = 14.028 t = 5.075

Note: Simple slope difference tests according to Dawson and Richter (2006).

simple slope difference test. The best of the four alternative strategies is to have a bottom-up formation process and a low number of externals on the board. The next best strategy for a larger group seeking external funding is to have a bottom-up formation process and a high number of externals on the board.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

61

Wincent et al.

Yip and Tsang (2007) have outlined the risks of misinterpreting results when dummy variables are involved in interactions. Therefore, in addition to the preceding analysis presenting the base approach simultaneously including main effects (intercept term) and single interaction effects of the dummy variable, we also performed an analysis of the partition approach of the slope terms. Following the logic of the partition approach, we entered interaction terms with reverse coding of the dummy variables. The results supported the previous findings. In accordance, we find one negative significant interaction effect between group size and top-down formation processes (β = −1.82; p < .001) compared to a positive interaction term between group size and bottom-up formation (β =1.72; p < .001). In an attempt to validate the three-way interactions, we performed post-hoc analyses in two subsamples where the original sample was divided with respect to formation processes, which is the dummy variable that could potentially cause these problems. Multi-group analyses revealed that the path coefficients of the included direct and interaction effects were in expected directions and significantly different between the two formation processes (χ2 = 91.76 with 5 d.f., p < .001), further strengthening a difference in interactions depending on formation logic.

Discussion and conclusion This research addresses the lack of attention paid to how SMEs organize themselves in groups to obtain funding for joint development of innovations that they would otherwise have difficulty developing on their own. In acknowledging that obtaining funding is an important issue for many groups of SMEs working on joint innovation, and in an attempt to contribute to the thus far limited understanding of this topic, we suggested that the size of such groups is an important determinant of their success. Using the literature on groups and collective action processes, we elaborated on the paradox of size: while large groups have more resources, they are at the same time more likely to run into collective action problems that impair their ability to obtain funding. We shed light on these issues by developing a set of hypotheses to suggest that, in general, smaller groups are likely to be more successful than larger ones in obtaining external funding. We elaborated on two mechanisms for how larger groups can overcome the burden of size: either by enhancing the internal organization through formation processes or by enhancing the access to external resources through governance devices. Analysing a full population sample of groups of Swedish SMEs that competed for external funding by using annual observations of the key variables of interest over a five-year period, we found strong support for our main arguments. First, overall, smaller groups were more successful in obtaining external funding. Second, although larger groups are less successful, they can improve their chances of obtaining funding if they align with external contingencies and configure themselves appropriately. These findings have several important implications. With several key variables controlled, the findings consistently show that smaller groups are in better shape to obtain external funding. However, we also found that larger groups can improve their odds substantially by taking actions to mitigate the shortcomings of their size. Larger groups that rely more heavily on incorporating a greater number of board externals outperformed those that did not in terms of the amount of external funding obtained. We also found that larger groups did better when a bottom-up approach characterized the formation process. Confirming our hypothesized relationships and pointing out the urgency of a good configuration for larger groups, our results clearly suggest that when larger groups do not have board externals, a bottom-up formation is critical for increasing the amount of external funding they obtain. We also found such evidence, though not as strong, for larger groups and group board interlocks. When larger groups did not report board interlocks, a bottom-up formation was critical to increasing the amount of external funding obtained.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

62

Strategic Organization 8(1)

Overall, this implies we also found support for our underlying assumption suggesting that smaller groups do not profit from the same contingencies or configurations as do their larger counterparts. The influences of a small group size on the amount of external funding obtained were reduced by board externals or reliance on a bottom-up process. This means that small size benefits such as ‘speaking with one voice’ and internal cohesion can be threatened by a more diverse number of links to other units and an increased access to outside resources from a variety of sources due to using persons not affiliated with the constituent SMEs. However, this also suggests that a bottom-up approach may facilitate cohesiveness to the point of inbreeding, which restricts them from effectively recognizing and addressing funding calls, thus ultimately restricting their odds of obtaining funding. In our research on why groups of various sizes may be differently equipped to obtain external funding and success in funding campaigns, we relied on group and collective action literature. In our attempt, we outlined a set of mechanisms to help explain how larger groups could improve their chances of obtaining external funding. Our results suggest that, in relation to funding campaigns, several contingencies mentioned in the group and collective action literature facilitate larger groups, whereas smaller groups, by virtue of their smallness, have natural advantages of swiftness and cohesiveness and thus do not require the same tools for infrastructure or political support. Consequently, some fundamentals mentioned in the group and collective action literature (e.g. creation of a joint identity, clear target goals for cohesiveness in the group) suffer from strengths and weaknesses with respect to size. We found rather strong support for our contingency and configurational tests of ideas, which suggests that future management research could benefit from paying more attention to it. Our study has also highlighted important ways in which SME groups differ from the kinds of collective action that have been studied previously. Specifically, our study has highlighted the observation that for groups that are formed to influence external stakeholders to obtain physically limited resources, problems associated with group cohesion may rapidly outweigh the political benefits of group size. When the outcomes of political action are physically limited, this rapidly gives rise to both free-riding problems and potential conflicts of interest. Thus, although SME groups can be studied through the lens of group and collective action literature, important differences also exist. This aspect may provide a conduit for SME group studies to contribute back to theory itself by addressing the effect of underlying objectives and specific governance challenges on the effectiveness of groups and collective action. Our findings have strong practical implications for managers in SME groups when they make decisions on how to organize the group to obtain external funding. For example, managers involved in firms that are part of larger groups should be aware of their need to improve collective action and implement appropriate governance devices to further the effectiveness of their group size. Our research suggests that larger SME groups are often disadvantaged as compared to smaller groups, for whom group cohesion is seldom a major problem. Not surprisingly, larger groups received much bigger payoffs for efforts to design structures and positions, a suggestion that is in line with prior research on collective action and group size (Oliver and Marwell, 1988). Still, smaller groups obtained relatively greater external funding no matter which action they took. Although more research is needed before any solid conclusions can be drawn, larger groups may be disadvantaged and may need to always consider how to work to mitigate the problems following an increasing group size. The management in smaller groups may have enough to overcome the natural advantages for effective composition of group members and the limited size of numbers mitigating the exploitation of large-scale innovation or attracting large market players for their joint products.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

63

Wincent et al.

A common trend in innovation research has been to discuss the failure rates of funding for innovative projects, especially for innovative projects among SMEs (Chesbrough and Teece, 1996; Freel, 2000). However, recent research (this study included) has found that, for SMEs, organization in groups may provide a better solution for funding larger programmes of innovative work, because such constellations reduce monitoring costs for funding agencies while providing a mechanism to share risks among firms by sharing expenses, gaining additional knowledge and developing joint standards. As such, this study also has implications for funding agencies that implement SME support programmes at regional and national levels. Specifically, the study provides advice on how to best configure groups to achieve success in obtaining external funding. Funding agencies can use the results of this study to understand the complexities of size, configuration and formation processes, and then use this knowledge in selection processes and to advise firms on how to better receive external funding. Importantly, although we tracked a sample of groups and tested the hypothesized relationships using adequate statistical procedures, the data covered a rather long time span (which reduces the possibility of not detecting other influences), and several limitations warrant further investigations. First, we acknowledge that context and time-dependent characteristics may influence the effectiveness of collective action. For instance, the political context of the study can influence the results of the study. The political and economic system of Sweden and the European Union builds on policy interventions. Typically, policy interventions in the areas of business focus on fostering entrepreneurship and innovation through a diverse set of funding agencies. Although this context is useful for a study such as the present one, the results may not be as obvious in political economies that rely less on policy interventions and have fewer policy programmes for external funding. We suggest that future studies should seek to replicate this study in heterogeneous samples of SME groups from different countries and in different time frames to examine the role of cultural and political context and of time-frame dependence. Moreover, we have established our results in a specific sample of SME groups (following the definition of Human and Provan, 1997). Although such groups are recurrently used in studies and have been deemed appropriate, there are also other network or group forms (see e.g. Jones et al., 1997, for a list and review of different types) and organizational forms (e.g. different forms of associations) in which the results of this study are potentially valid. We leave the examination of the contextual boundaries for the validity of our results to future research. Some other potential limitations of the study pertain to the measurement and selection of key variables. Although we carried out a meticulous literature review to trace and select a few variables that were likely to influence how group size influenced success in funding campaigns, the relationship between group size and funding success may still be influenced by additional contingencies. Therefore we recommend that future studies expand on the variables we have tested here to map those that are most prominent in influencing the success of SME funding campaigns. In a related vein, we also suggest that future studies examine the success of funding campaigns using multiple dependent variables, where success in funding is one of several indicators of the success of the campaign. We suggest a closer examination of both tangible and intangible assets (Freeman, 1979) as dependent variables, and the examination of potential interrelationships between these different categories of assets. We also recommend that future studies include innovative output as a final consequence in modelling how configuration of groups influences success in terms of funding, which in turn likely influences innovative output. These limitations aside, given the large number of SME groups that seek external funding for innovation programmes, it is surprising that prior research has focused only marginally on what leads to success in such campaigns. In this article, we draw on insights from group and collective

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

64

Strategic Organization 8(1)

action literatures, and thus arrive at theories that acknowledge the important role of group size for funding levels, as well as the joint influence of configuration of internal organization and external access to resources on this relationship. In doing so, we acknowledge two- and three-way interactions, something that prior group and collective action literature has rarely, if ever, considered. References Ahlström-Söderling, R. (2002) ‘Some Thoughts Based on a Longitudinal Analysis of Business Networks in the Swedish Wood Manufacturing Industry’, paper presented at the International Council for Small Business World Conference 2002, San José, Puerto Rico, September. Ahuja, G. (2000) ‘Collaboration Networks, Structural Holes, and Innovation: A Longitudinal Study’, Administrative Science Quarterly 45(3): 425–55. Aiken, L. S. and West, S. G. (1991) Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage. Andrews, K. T. (2001) ‘Social Movements and Policy Implementation: The Mississippi Civil Rights Movement and the War on Poverty, 1965 to 1971’, American Sociological Review 66: 71–95. Astley, W. G. and Fombrun, C. J. (1983) ‘Collective Strategy: Social Ecology of Organizational Environments’, Academy of Management Review 8(4): 576–87. Bascle, G. (2008) ‘Controlling for Endogeneity with Instrumental Variables in Strategic Management Research’, Strategic Organization 6(3): 285–327. Benbasat, I. and Lim, L.-H. (1993) ‘The Effects of Group, Task, Context, and Technology Variables on the Usefulness of Group Support Systems: A Meta-Analysis of Experimental Studies’, Small Group Research 24: 430–62. Bougrain, F. and Haudeville, B. (2002) ‘Innovation, Collaboration and SMEs Internal Research Capacities’, Research Policy 31: 735–47. Boyd, B. (1990) ‘Corporate Linkages and Organizational Environment: A Test of the Resource Dependence Model’, Strategic Management Journal 11(6): 419–30. Burt, R. S. (1980) ‘Cooptive Corporate Actor Networks: A Reconsideration of Interlocking Directorates Involving American Manufacturing’, Administrative Science Quarterly 25: 557–82. Burt, R. S. (2005) Brokerage and Closure: An Introduction to Social Capital. Oxford: Oxford University Press. Carpenter, M. and Westphal, J. (2001) ‘The Strategic Context of External Network Ties: Examining the Impact of Board Appointments on Board Involvement in Strategic Decision Making’, Academy of Management Journal 44: 639–60. Chaston, I. (1995) ‘Danish Technological Institute SME Sector Networking Model: Implementing Broker Competencies’, Journal of European Industrial Training 19(1): 10–17. Chesbrough, H. W. and Teece, D. J. (1996) ‘When is Virtual Virtuous? Organizing for Innovation’, Harvard Business Review 74(1): 65–73. Das, T. K. and Teng, B.-S. (1998) ‘Between Trust and Control: Developing Confidence in Partner Cooperation in Alliances’, Academy of Management Review 23(3): 491–512. Davis, G. F., Yoo, M. and Baker, W. E. (2003) ‘The Small World of the American Corporate Elite, 1982– 2001’, Strategic Organization 1(3): 301–26. Dawson, J. F. and Richter, A. W. (2006) ‘Probing Three-Way Interactions: The Development and Application of a Slope Difference Test’, Journal of Applied Psychology 91: 917–26. DeNardo, J. (1985) Power in Numbers: The Political Strategy of Protest and Rebellion. Princeton, NJ: Princeton University Press. Diani, M. (2003) ‘Leaders or Brokers? Positions and Influence in Social Movement Networks’, in M. Diani and D. McAdam (eds) Social Movements and Networks, pp. 105–22. Oxford: Oxford University Press.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

65

Wincent et al.

Dominelli, L. (1996) ‘Deprofessionalizing Social Work: Anti-Oppressive Practice, Competencies and Postmodernism’, British Journal of Social Work 26: 153–75. Doz, Y. L., Olk, P. M. and Ring, P. S. (2000) ‘Formation Processes of R&D Consortia: Which Path to Take? Where Does it Lead?’, Strategic Management Journal 21(3): 239–66. Freel, M. S. (2000) ‘Barriers to Product Innovation in Small Manufacturing Firms’, International Small Business Journal 18: 60–80. Freeman, L. C. (1979) ‘Centrality in Social Networks: Conceptual Clarification’, Social Networks 1: 215–39. Fukugawa, N. (2006) ‘Determining Factors in Innovation of Small Firm Networks: A Case of Cross Industry Groups in Japan’, Small Business Economics 27(2): 181–93. Galaskiewicz, J. and Wasserman, S. (1989) ‘Mimetic Processes within an Interorganizational Field: An Empirical Test’, Administrative Science Quarterly 34(3): 454–79. Gamson, W. A. (1995) ‘Hiroshima, the Holocaust, and the Politics of Exclusion: 1994 Presidential Address to the American Sociological Association’, American Sociological Review 60: 1–20. García-Canal, E., Valdés, A. and Ariño, A. (2003) ‘Effectiveness of Dyadic and Multi-Party Joint Ventures’, Organization Studies 24(5): 743–70. Gomes-Casseres, B. (2003) ‘Competitive Advantage in Alliance Constellations’, Strategic Organization 1: 327–35. Goodstein, J., Gautam, K. and Boeker, W. (1994) ‘The Effects of Board Size and Diversity on Strategic Change’, Strategic Management Journal 15(3): 241–50. Gould, R. V. (1991) ‘Multiple Networks and Mobilization in the Paris Commune, 1871’, American Sociology Review 56: 716–29. Grindley, P., Mowery, D. and Silverman, B. (1994) ‘Sematech and Collaborative Research: Lessons in the Design of High-Technology Consortia’, Journal of Policy Analysis and Management 13(4): 723–58. Halme, M. and Fadeeva, Z. (2001) ‘Small and Medium-Sized Tourism Enterprises in Sustainable Development Networks’, Greener Management International 30: 97–113. Hamilton, B. and Nickerson, J. (2003) ‘Correcting for Endogeneity in Strategic Management Research’, Strategic Organization 1(1): 53–80. Hanna, V. and Walsh, K. (2002) ‘Small Firm Networks: A Successful Approach to Innovation’, R&D Management 32(3): 201–7. Hoang, H. and Antonic, B. (2003) ‘Network-Based Research in Entrepreneurship: A Critical Review’, Journal of Business Venturing 18: 165–87. Huggins, R. (2001) ‘Inter-Firm Network Policies and Firm Performance: Evaluating the Impact of Initiatives in the United Kingdom’, Research Policy 30(3): 443–58. Human, S. E. and Provan, K. G. (1997) ‘An Emerging Theory of Structure and Outcomes in Small-Firm Strategic Manufacturing Networks’, Academy of Management Journal 40: 368–403. Jarillo, C. (1988) ‘On Strategic Networks’, Strategic Management Journal 9: 31–41. Jones, C., Hesterly, W. S. and Borgatti, S. P. (1997) ‘A General Theory of Network Governance: Exchange Conditions and Social Mechanisms’, Academy of Management Review 22(4): 911–45. Kapferer, B. C. (1969) ‘Norms and the Manipulation of Relationships in a Work Context’, in J. C. Mitchell (ed.) Social Networks in Urban Situations: Analyses of Personal Relationships in Central African Towns, pp. 181–244. Manchester: Manchester University Press. Khanna, T. (1998) ‘The Scope of Alliances’, Organization Science 9(3): 340–55. Klandermans, B. and deWeerd, M, (2000) ‘Group Identification and Political Protest’, in S. Stryker, T. J. Owens and R. W. White (eds) Self, Identity and Social Movements, pp. 68–92. Minneapolis: University of Minnesota Press. Lynall, M. D., Golden, B. R. and Hillman, A. J. (2003) ‘Board Composition from Adolescence to Maturity: A Multitheoretic View’, Academy of Management Review 28: 416–31.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

66

Strategic Organization 8(1)

McAdam, D. (1982) Political Process and the Development of Black Insurgency 1930–1970. Chicago, IL: University of Chicago Press. McAdam, D. (1996) ‘Conceptual Origins, Current Problems, Future Directions’, Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings, pp. 23–40. Cambridge: Cambridge University Press. McCarthy, J. and Wolfson, M. (1996) ‘Resource Mobilization by Local Social Movement Organizations: Agency, Strategy and Organization in the Movement against Drinking and Driving’, American Sociological Review 61: 1070–88. McCarthy, J. D. and Zald, M. N. (1977) ‘Resource Mobilization and Social Movements: A Partial Theory’, American Journal of Sociology 82(6): 1212–41. Mauer, I. and Ebers, M. (2006) ‘Dynamics of Social Capital and Their Performance Implications: Lessons from Biotechnology Start-ups’, Administrative Science Quarterly June: 262–93. Mizruchi, M. S. (1996) ‘What Do Interlocks Do? An Analysis, Critique, and Assessment of Research on Interlocking Directorates’, Annual Review of Sociology 22: 271–98. Monge, P. R., Fulk, J., Kalman, M. E., Flanagin, A. J., Parnassa, C. and Rumsey, S. (1998) ‘Production of Collective Action in Alliance-Based Interorganizational Communication and Information Systems’, Organization Science 9(3): 411–33. Oliver, P. E. (1993) ‘Formal Models of Collective Action’, Annual Review of Sociology 19: 271–300. Oliver, P. E. and Marwell, G. (1988) ‘The Paradox of Group Size in Collective Action: A Theory of the Critical Mass. II’, American Sociological Review 53(1): 1–8. Oliver, P. E. and Myers, D. J. (2003) ‘Networks, Diffusion and Cycles of Collective Action’, in M. Diani and D. McAdam (eds) Social Movements and Networks, pp. 173–203. Oxford: Oxford University Press. Olson, M. (1965) The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press. Oxley, J. E. and Sampson, R. C. (2004) ‘The Scope and Governance of International R&D Alliances’, Strategic Management Journal 25: 723–49. Palmer, D. (1983) ‘Broken Ties: Interlocking Directorates and Inter-Corporate Coordination’, Administrative Science Quarterly 28: 40–55. Pfeffer, J. and Salancik, G. R. (1978) The External Control of Organizations: A Resource Dependence Perspective. New York: Harper and Row. Polletta, F. and Jasper, J. M. (2001) ‘Collective Identify and Social Movements’, Annual Review of Sociology 27: 282–305. Powell, W. W., Koput, K. W. and Smith-Doerr, L. (1996) ‘Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology’, Administrative Science Quarterly 41: 116–45. Reagans, R. and Zuckerman, E. W. (2001) ‘Networks, Diversity, and Productivity: The Social Capital of Corporate R&D Teams’, Organization Science 12: 502–17. Robson, M. J., Katsikeas, C. S. and Bello, D. C. (2008) ‘Drivers and Performance Outcomes of Trust in International Strategic Alliances: The Role of Organizational Complexity’, Organization Science 19(4): 647–65. Rowley, T. J. and Moldoveanu, M. (2003) ‘When Will Stakeholder Groups Act? An Interest- and IdentityBased Model of Stakeholder Group Mobilization’, Academy of Management Review 28(2): 204–19. Schoorman, F. D., Bazerman, M. H. and Atkin, R. S. (1981) ‘Interlocking Directorates: A Strategy for Reducing Environmental Uncertainty’, Academy of Management Review 6(2): 243–51. Scully, M. A. and Creed, W. E. D. (2005) ‘Subverting our Stories of Subversion’, in G. F. Davis, D. McAdam, W. R. Scott and M. N. Zald (eds) Social Movements and Organization Theory, pp. 310–31. Cambridge: Cambridge University Press.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

67

Wincent et al.

Shin, D.-H. (2004) ‘Public Networks for SME Applications: Implications for Future SME Networks’, Human Systems Management 23: 151–9. Siegel, D. A. (2009) ‘Social Networks and Collective Action’, American Journal of Political Science 53(1): 122–38. Stuart, T. (2000) ‘Interorganizational Alliances and the Performance of Firms: A Study of Growth and Innovation Rates in a High Technology Industry’, Strategic Management Journal 21: 791–811. Tell, J. (2000) ‘Learning Networks: A Metaphor for Inter Organizational Development in SMEs’, Enterprise and Innovation Management Studies 1(3): 303–17. Thorgren, S., Wincent, J. and Örtqvist, D. (2009) ‘Designing Interorganizational Networks for Innovation: An Empirical Examination of Network Configuration, Formation and Governance’, Journal of Engineering and Technology Management 26(3): 148–66. Varamäki, E. and Vesalainen, J. (2003) ‘Modelling Different Types of Multilateral Co-operation between SMEs’, Entrepreneurship and Regional Development 15(1): 27–47. Weber, K., Heinze, K. L. and DeSoucey, M. (2008) ‘Forage for Thought: Mobilizing Codes in the Movement for Grass-Fed Meat and Dairy Products’, Administrative Science Quarterly 53: 529–67. Wincent, J. (2008) ‘An Exchange Approach on Firm Cooperative Orientation and Outcomes of Strategic Multilateral Network Participants’, Group and Organization Management 33(3): 303–29. Yip, P. S. and Tsang, E. W. K. (2007) ‘Interpreting Dummy Variables and Their Interaction Effects in Strategy Research’, Strategic Organization 5(1): 13–30. Zald, M. N. and Ash, R. (1966) ‘Social Movement Organizations: Growth, Decay and Change’, Social Forces 44: 327–41. Zald, M. N. and McCarthy, J. D. (1980) ‘Social Movement Industries: Competition and Cooperation among Social Movement Organizations’, Research in Social Movement, Conflict and Change 3: 1–20. Zeng, X.-P. and Chen, M. (2003) ‘Achieving Cooperation in Multiparty Alliances: A Social Dilemma Approach to Partnership Management’, Academy of Management Review 28(4): 587–605.

Author biographies Joakim Wincent is professor in business administration at Umeå School of Business and an entrepreneurship researcher at Luleå University of Technology. His research focuses largely on strategic interfirm relations, trust and reciprocal challenges, stress management of entrepreneurs and effectiveness implications of cognitive foundations and emotional experiences in new venturing. He has published in journals such as Academy of Management Review, Organizational Research Methods, Journal of Management Studies, Journal of Business Research, Journal of Business and Psychology and Journal of Applied Social Psychology. Address: IES, Luleå University of Technology, SE-971 87 Luleå, Sweden. [email: [email protected]] Daniel Örtqvist is assistant professor of entrepreneurship at Luleå University of Technology. His academic research mainly concerns the entrepreneurial identity, the establishment of new ventures and aspects of organized cooperation. He has published in a number of journals that include Journal of Management Studies, Scandinavian Journal of Management, Journal of Business Research, Journal of Business and Psychology and Journal of Applied Social Psychology. Address: IES, Luleå University of Technology, SE-971 87 Luleå, Sweden. [email: [email protected]] Jessica Eriksson is an assistant professor in entrepreneurship at Umeå School of Business, Umeå University. Her research interests are sustainable development and interorganizational dynamics, with a particular focus on how cooperative and competitive relationships influence innovation outcomes.

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012

68

Strategic Organization 8(1)

Most recent publications include articles in International Journal of Innovation and Sustainable Development and Competitiveness Review. Address: Umeå School of Business, SE-901 87 Umeå, Sweden. [email: [email protected]] Erkko Autio is professor and the QinetiQ – EPSRC Chair in Technology Transfer at Imperial College Business School. His research interests include high technology, international new ventures, learning and social capital. He has published in journals such as Academy of Management Journal, Academy of Management Review, Journal of Business Venturing, Research Policy, Strategic Management Journal and Journal of International Business Studies. Address: Imperial College Business School, Exhibition Road, London SW7 2AZ, UK. [email: [email protected]]

Downloaded from soq.sagepub.com by Monojit Banerjee on October 10, 2012