The Dark Side of Team Social Cohesion in NPD Team Boundary Spanning

1 The dark side of team social cohesion in NPD team boundary spanning Pilar Carbonell School of Administrative Studies, Atkinson Building, Room 282, 4700 Keele Street, Toronto M3J 1P3, Canada. E-mail: [email protected], Tel: 1-416-736-2100 ext. 66303. Ana I. Rodríguez Escudero Facultad de CC.EE. y Empresariales, Avda. Valle Esgueva, nº 6, 47011 Valladolid. E-mail: [email protected], teléfono 983184394 This is the accepted version of the manuscript: Carbonell, P. y Rodríguez-Escudero, A.I.. (2019): The dark side of team social cohesion in NPD team boundary spanning, Journal of Product Innovation Management, 36 (2), 141-171. https://doi.org/10.1080/19368623.2019.1564106 Funding The authors are grateful for the financial support of the Junta de Castilla y León (Spain), project reference VA112P17, and the Ministerio de Economía, Industria y Competitividad, Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016, project reference ECO201786628-P. 2 Introduction In today’s hypercompetitive and fast-paced environment, new product development (NPD) performance depends, more than ever, on a team’s ability to span boundaries and forge relationships with parties external to the team, a phenomenon known as team boundary spanning (Ancona et al., 2009). Team boundary spanning represents a team’s actions to establish links and manage interactions with individuals and groups inside and outside the organization with the purpose of coordinating NPD activities, obtaining information and resources and building support for the team and the NPD project (Ancona et al., 2009; Joshi et al., 2009). Research in NPD has shown that team boundary spanning activity positively influences a number of new product outcomes such as adherence to budgets and schedules and project technical quality (Ancona and Caldwell, 1992; Howell and Shea, 2006). This study offers new insights on the influence of team boundary spanning on the success of new products by examining its impact on new product competitive advantage. Product advantage has been shown be the strongest predictor of new product success in Henard and Szymanski (2001) and Evanschitzky et al. (2012) meta-analyses of success factors in product innovation. Team boundary spanning connects a NPD team to highly valued resources from external parties such as information, feedback on progress and support which, according to research in NPD, can increase a team’s ability to develop new products with a competitive advantage (Garcia et al., 2008; Kim and Atuahene-Gima, 2010, Potter and Lawson, 2013; Veldhuizen et al., 2006). However, for boundary-spanning teams, having access to external information and resources may not be enough to deliver product advantage, unless team members integrate the information and resources gained through their boundary spanning work into the NPD project (Maurer et al., 2011). On this point, recent work on boundary spanning suggests that boundary 3 spanning teams can be differentially receptive to information and resources coming from outside depending on their psychological characteristics (Dokko et al., 2014; Oh et al., 2004; 2006). As it is argued in detail in this article, one psychological characteristic likely to negatively influence boundary spanning teams’ motivation to use external resources is team social cohesion. Team social cohesion refers to the degree to which team members “are attracted to one another and are motivated to stay in the group” (Bettenhausen, 1991, p. 361). Within the literature in social identity, it has been argued that team social cohesion induces a strong team identity among group members and in doing so it elicits intergroup bias (Petersen et al., 2004; Rosenberg and Treviño, 2003), a systematic tendency to evaluate one’s group (the in-group) and its members more favorably than other groups (the out-groups) (Hewstone et al., 2002). Drawing on this research, the current study contends that when social cohesion is high, the intergroup biases inherent in highly cohesive groups, would prejudice boundary spanning teams against absorbing and using the information and resources generated through their boundary spanning work, ultimately limiting the potential of these activities to facilitate new product competitive advantage. Social identity research has argued that socially cohesive groups will exhibit higher intergroup bias in situations in which external actors present a perceived threat to the group’s collective identity (Turner and Pratkanis, 1998). Because cohesive groups present a strong group identity along with a strong desire to protect it, in instances where the group identity is threatened by external groups, cohesive groups would tend to maintain their positive image by displaying stronger negative biases toward outgroup members (Turner et al., 1992; Turner and Pratkanis, 1998). For a socially cohesive NPD team, external task interdependence and project newness are two situational project characteristics susceptible of creating situations such the one described above (Glynn et al., 2010; Janssen, 2004). External task interdependence refers to a 4 situation in which a NPD team has to rely on external parties to successfully complete their project. A NPD team working under conditions of high external interdependence depends on the cooperation and information of individuals and groups external to the team to effectively do its job (Drach-Zahavy and Somech, 2010). For highly cohesive teams, the work constraints and loss of control that arise from having to consult and collaborate with individuals and groups outside the team can be perceived as a threat to the group’s sense of identity (Glynn et al., 2010; Langfred, 2000). Regarding project newness, literature on resistance to change has contended that teams in charge of highly innovative projects do often fall into conflict with and experience resistance from other individuals inside and outside the company due to the wide range of changes brought about by this type of projects (Janssen et al., 2004). For highly cohesive groups, such resistance from outgroup members is likely to be viewed as an attack on the group’s objectives and therefore on their collective identity (Janssen, 2004). Drawing on the view that external task interdependence and project newness can provide a context in which external actors are perceived as threats to the identity of highly cohesive teams, the current study expects team social cohesion to have a stronger negative moderating effect on the relationship between team boundary spanning and new product competitive advantage when external task interdependence and project newness are high than when they are low. The current study makes several contributions to the literatures on team boundary spanning, NPD and social identity. First, findings from this study shed new insights on the significance of team boundary spanning to the success of new products by revealing that boundary spanning activities is beneficial to achieving new product competitive advantage. Secondly, by investigating how boundary spanning’s effectiveness is contingent on the level of team social cohesion, this study not only addresses recent calls to examine how team internal characteristics 5 interact with team boundary spanning to affect team performance (Marrone, 2010) but also provides a new perspective into the impact of team social cohesion on NPD teams. Thus, within the field of NPD, social cohesion has been primarily viewed as a desirable criterion for team formation and a predictor of new product performance (Im et al., 2013; Nakata and Im, 2010; Shaner et al., 2016). The current study departs from existing research in that it exposes a dark side of team social cohesion for NPD teams engaged in boundary spanning activities. Drawing on social identity theory (Hogg and Abrams, 1988; Petersen et al., 2004; Tajfel and Turner, 1986), the current study argues that team social cohesion can undermine the positive effect of team boundary spanning on new product competitive advantage by decreasing the propensity of team members to consider and integrate the information and resources gathered through their boundary spanning work. To our knowledge, this is the first study to propose a negative synergy between team boundary spanning and team social cohesion. Lastly, the current study expands research on social identity and intergroup bias by proposing and demonstrating that project newness and external task interdependence bring about situations in which external groups present a threat to the collective identity of socially cohesive groups. Thus, while research in this field has argued that, in instances where the collective identity of socially cohesive groups is threatened, these groups are expected to show greater intergroup bias (Rosenberg and Tevino, 2003; Rempel and Fisher, 1997), empirical evidence on this subject has been inconsistent (Turner and Horvitz, 2001). According to Turner and Pratkanis (1998), one explanation for these contradictory findings lies in the nature of the threats used in the literature. In particular, it has been noted that the threats used in a number of studies, while alluding to some form of crisis, presented very little consequences for the effective functioning of the group and thus lacked realism (Turner and Pratkanis, 1998). The current study maximizes the possibility of observing 6 threat effects by examining two situational contexts with genuine potential to pose serious threats to the identity of socially cohesive teams, mainly external task interdependence and project newness. Building on the view that external task interdependence (Glynn et al., 2010) and project newness (Janssen, 2003 and 2004) give rise to situations in which socially cohesive teams may perceived that their sense of identity is under threat, it is argued that social cohesion’s detrimental effect on the relationship between team boundary spanning and new product competitive advantage would become more pronounced when external task interdependence and project newness are high than when they are low. Theoretical framework The theoretical model depicted in Figure 1 builds upon research on team boundary spanning, NPD and social identify theory. Marrone (2010) defined team boundary spanning as a team’s effort to establish links and manage interactions with individuals and groups inside and outside the organization in an effort to facilitate team goals and objectives. Team boundary spanning encompasses an array of different activities, mainly ambassador, scouting and task coordination (Ancona and Caldwell, 1992; Marrone, 2010). Ambassador activities pertain to building support for the project and the team by talking the NPD project up and forming relationships with external parties. Task coordination activities involve negotiating ad-hoc resources, co-solving problems, and coordinating NPD activities with groups inside and outside the organization. Scouting activities are directed at gaining project-related information such as stakeholder’s expectations and market and technological opportunities and trends (Ancona et al., 2009; Marrone, 2010). In this study, team boundary spanning is expected to have a positive effect on new product competitive advantage (see Figure 1). The proposed positive effect is based on NPD research suggesting that 7 access to external knowledge and resources can increase a NPD team’s ability to develop new products with competitive advantage (Garcia et al., 2008; Kim and Atuahene-Gima, 2010, Potter and Lawson, 2013; Veldhuizen et al., 2006). Social identity theory asserts that individuals derive satisfaction and value from their group memberships and hence groups and their members are motivated to adopt and maintain positive group identities (Tajfel, 1982; Tajfel and Turner, 1986). Such preference to uphold a favorable group identity is believed to create and sustain intergroup bias (Aberson et al., 2000; Hewstone et al., 2002; Leonardelli and Brewer, 2001, Rosenberg and Treviño, 2003). The more salient a group membership is to its members, the more motivated group members will be to protect or even enhance that identity by showing attitudes and behaviors that favor one’s group and derogate other groups (Hewstone et al., 2002; Tajfel and Turner, 1979). According to social identity theory, highly cohesive groups, where individuals have high team identification (Brockman et al., 2010), are too expected to show intergroup bias (Hogg and Abrams, 1988; Petersen et al., 2004; Tajfel and Turner, 1986). High cohesiveness induces a strong sense of unity among members of a group; this perception of unity can be accompanied by a strong sense of differentiation with respect to an outgroup (Dion, 1973). “Once generated, this differentiation of ingroup from outgroup may strongly dominate individuals’ perceptions and behavior in intergroup situations evoking contrasting modes of behavior and negative attitudes toward outgroups” (Dion, 1973, p. 169). The current study draws on this research to propose a negative moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage. In particular, we argue that when team social cohesion is high, the intergroup biases present in highly cohesive groups would prejudice boundary spanning teams against absorbing and using information and resources stemming from those 8 activities, ultimately limiting the potential of team boundary spanning to deliver competitive advantage. Finally, social identity theory suggests that socially cohesive groups will exhibit stronger biases toward outgroup members in situations where group identity is threatened by out-groups (Turner et al., 1992; Turner and Pratkanis, 1998). Because cohesive groups exhibit a strong group identity along with a strong desire to protect their identity, in instances where the collective identity is threatened by out-groups, cohesive groups would tend to maintain their positive image by displaying stronger negative bias toward out-groups (Turner et al., 1992; Turner and Pratkanis, 1998). Extending this line of research, the current study proposes that external task interdependence and project newness create such instances in which external groups can pose a threat to the identity of socially cohesive teams (Glynn et al., 2010; Janssen, 2004), and thus socially cohesive teams are expected to exhibit higher in-group favoritism and out-group derogation when external task interdependence and project newness are high rather than when they are low. Approaching external task interdependence and project newness as situations in which out-group members can pose a significant threat to the identity of highly cohesive teams is something relatively new in the innovation literature. One exception is Glynn et al. (2010), which contended that high identification within the team may lead individuals to view external task interdependence as a threat to their team’s identity. Regarding project newness, Janssen and colleagues (Janssen 2003; Janssen 2004; Janssen et al., 2004) have asserted that while it is expected that new ideas introduced by innovative employees generate resistance from other employees, to highly job-involved individuals, resistance from co-workers is likely to be interpreted as an identity-relevant value clash. Building on the view that external task interdependence (Glynn et al., 2010) and project newness (Janssen, 2003 and 2004) give rise 9 to situations in which other teams can pose a challenge to the identity of one individual or team, the current study argues that social cohesion would exert a stronger negative moderating effect of team cohesion on the relationship between team boundary spanning and new product competitive advantage when external task interdependence and project newness are high than when they are low. Hypotheses Team boundary spanning and new product competitive advantage New product competitive advantage is defined as “a product’s perceived superiority relative to competitive products” (Song and Montoya-Weiss, 2001; p. 65). Team boundary spanning activity is expected to positively impact new product competitive advantage based on the following reasons. First, building relationships with existing and potential customers can give NPD teams an opportunity to better understand customers’ needs and wants and to clarify expectations (Ancona and Caldwell, 1988), improving their ability to deliver superior products (Li and Calantone, 1998). Also, establishing relationships with suppliers, distributors, and research centers can improve a NPD team’s access to up-to-date market and technological information and trends, which is deemed critical to create new products with innovative and superior features (Li and Calantone, 1998). Furthermore, boundary spanning activities can provide NPD teams with timely and critical feedback from groups inside and outside the company on the NPD project, which has been shown to heighten a NPD team’s ability to produce superior new products (Menon et al., 1997). Therefore, we propose that: H1: Team boundary spanning will have a positive effect on new product competitive advantage Moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage 16 Measurement scales Research in team boundary spanning has generally viewed team boundary spanning as consisting of three different activities, i.e., ambassador, scouting and task coordination activities, each of them taking place inside the organization (i.e., intrafirm) as well as outside the organization (i.e., extrafirm) (Ancona et al., 2009; Marrone, 2010; Sleep et al., 2015). In keeping with this view, we operationalized team boundary spanning as a formative second-order construct composed of six reflective first-order dimensions, mainly intrafirm ambassador, extrafirm ambassador, intrafirm scouting, extrafirm scouting, intrafirm task coordination and extrafirm task coordination. Measurement items of the six dimensions of team boundary spanning were adapted from Ancona and Caldwell (1988 and 1992) (see Table 2). The appropriateness of the initial pool of 30 items for capturing the multidimensionality of the team boundary spanning construct was tested by performing a series of exploratory factor analyses (principal component with promax rotation). An iterative process was used to eliminate singleitem factors, items with communalities estimates of 0.6 or lower, items with factorial loadings of 0.4 or lower and items with high cross-loadings (Nunnally and Bernstein, 1994). The final factor analysis showed a 27-item, seven-factor solution with eigenvalues greater than 1, which together explained 72.76% of the variance. The exploratory factor solution included two factors related to ambassador activities, three factors about scouting activities and two factors pertaining to task coordination activities. Observe that intrafirm scouting activities were split into two dimensions; one related to scouting general information about the project and another pertaining to scouting specific marketing and technical information for the project. To further asses the scales’ factor structure, we performed a confirmatory factor analysis on the seven-factor solution (SmartPLS 3.1.5). All 27 items had item loadings greater than 0.50 on 17 their respective factors thus, all items were retained (see Table 2). As shown in Table 2, Cronbach alphas and composite reliability values of all the team boundary spanning dimensions exceeded the standards of 0.70. Also, for all the dimensions, the values of the average variance extracted were above 0.50, hence indicating convergent validity (Hair et al., 2013). (Insert Table 2 here) To statistically validate the formative character of the seven dimensions of team boundary spanning scale, we checked for multicollinearity among its dimensions using the varianceinflation factor (VIF). The VIF values of the dimensions were below the cut-off value of 10 (max. VIF: 2.198). Thus, there are no concerns about collinearity issues. Next, we examined the significance of the contribution of each dimension to the main construct. Fit of the formative measurement model was good as evidenced by the fact that the outer weights of all dimensions were significant at p<0.01 (see Table 3). (Insert Table 3 here) Established multi-item scales were used to measure the rest of the study’s constructs. Team social cohesion was operationalized with a 4-item scale taken from (Sethi et al., 2001). External task interdependence was measured with three items adapted from Sethi (2000) and project newness and product competitive advantage were each measured with 3-item scales adapted from McNally et al. (2010). The study includes two control variables for new product competitive advantage: firm size and product innovation. Firm size can be considered as a proxy variable of market power and financial resources; larger organizations typically have enough R&D, marketing and financial resources to successfully develop and commercialize new products and services (Ali et al., 1995). Firm size was measured with the number of employees. Product innovation refers to the weight that a company assigns to the value of new products for 18 creating and retaining a competitive position (Zahra, 1993). Firms with high product innovation activity are more likely to collect and disseminate customer and competitor intelligence so that product innovations provide greater value added for customers (Siguaw et al., 2006). Product innovation was measured with three items adapted from Zahra (1993). Measurement items and statistics that confirm the reliability (Cronbach alpha and composite reliability) and validity (average variance extracted) of the scales are shown in Table 4. (Insert Table 4 here) Discriminant validity was assessed by examining the square root of the average variance extracted for each construct. Discriminant validity is evidenced when the square root of the AVE for each construct exceeds the corresponding correlations between that construct and any other constructs (Fornell and Larcker, 1981). All possible pairs of constructs passed this test. Henseler et al.’s (2015) heterotrait-monotrait (HTMT) ratio of correlations technique also indicated discriminant validity, as all HTMT ratios were well below the conservative threshold of 0.85. Table 5 shows the square roots of AVE on the diagonal, the correlation values below the diagonal and the HTMT ratios above the diagonal. (Insert Table 5 here) Data analysis and results The proposed model was tested using Partial Least Squares Structural Equation Modeling (PLSSEM) algorithm (Smart PLS 3.1.5). PLS-SEM is preferable to covariance-based SEM (CBSEM) because it better handles small sample sizes and does not make any distributional assumptions to estimate the model (Hair et al., 2013; Reinartz et al., 2009). A further strength of PLS relevant for this study is the ability to handle reflective and formative measures in a single model (Hair et al., 2013). A bootstrap test (1,000 sub-samples) was used to generate the standard 19 error and t-values of the parameters. Option “no sign change” was employed for model estimation because it results in the most conservative outcome (Hair et al., 2013). Variables were introduced into the model hierarchically as blocks. First, we included the direct effects of team boundary spanning, control variables and moderators on new product competitive advantage (Model 1). Then, we added the two-way interaction effect of team boundary spanning and team social cohesion (Model 2). Next, the three-way interaction effect of external task interdependence was included (Model 3) followed by the three-way interaction effect of project newness (Model 4), along with all two-term interaction terms. We chose to incorporate external task interdependence and project newness into separated models in order to control for the increase of Type I error that happens when multiple moderator effects of correlated variables are investigated (Cohen et al., 2013; Frazier et al., 2004). Multicollinearity was examined using VIF statistics and found to be acceptable in all cases. The maximum VIF values for Models 1, 2, 3 and 4 were respectively, 1.196, 1.210, 1.769 and 1.341, which are well below the cut-off point of 10 (Hair et al., 2009). Table 6 contains the results for Models 1-4. Results for Model 1 in Table 6 show a positive relationship between team boundary spanning and new product competitive advantage (β=0.14, p<.05), which provides support for H1. Also, in support of H2, we found a significant and negative two-way interaction effect of team boundary spanning and team social cohesion (β=-0.24, p<.01, Model 2) i . Following Aiken and West (1991)’s procedure, we calculated the effect of team boundary spanning on new product competitive advantage at one standard deviation below and above the mean of team social cohesion. Results showed a positive and significant effect of team boundary spanning on new product competitive advantage at low level of team social cohesion (β =0.44, p<.01). The effect of team boundary spanning on new product competitive advantage became insignificant at 20 high levels of team social cohesion. Results in Models 3 and 4, respectively show significant and negative three-way interaction effects of team boundary spanning, team social cohesion and external task interdependence (β=-0.28, p<.01), and team boundary, team social cohesion and project newness (β=-0.19, p<.01), thus H3 and H4 are supported. (Insert Table 6) Next, we employed the partial derivative’s approach proposed by Schoonhoven’s (1981) (Figures 2 and 3) and the Aiken and West’s (1991) procedure to determine whether the two-way interaction effect of team boundary spanning, and team social cohesion changed over the range of the external task interdependence and project newness variables. Figure 2 graphically depicts the two-way interaction effect of team boundary spanning and team social cohesion over the range of values of external task interdependence [Eq.(1): (d new product competitive advantage /d team boundary spanning)x(1/d team social cohesion)= -0.18 -0.28 x (external task interdependence)]. The results in Figure 2 shows that the negative moderating effect of team social cohesion on the team boundary spanning-new product competitive advantage relationship becomes more pronounced as external task interdependence increases. Following Aiken and West (1991)’s procedure, we calculated the moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage at one standard deviation below and above the mean of external task interdependence. Results showed a non-significant interaction effect of team social cohesion and team boundary spanning at low level of external task interdependence (-1SD). The interaction effect however became significant and negative at high levels (+1SD) of external task interdependence (β =-0.52, p<.01). Figure 3 depicts the two-way interaction effect of team boundary spanning and team social cohesion over the range of values of project newness [Eq.(2): (d new product competitive 21 advantage /d team boundary spanning)x(1/d team social cohesion)= -0.13 -0.19 x (project newness)]. In keeping with H4 prediction, the results in Figure 3 shows a more negative interaction effect between team social cohesion and team boundary spanning as project newness increases. Relatedly, Aiken and West (1991)’s procedure revealed negative and significant moderating effects of team social cohesion on the relationship between team boundary spanning and new product competitive advantage at high (+1SD) levels of project newness (β=-0.51, p<.01) and a non-significant moderating effect (-1SD) at low levels of project newness. (Figure 2 and 3 here) Additional analysis In this work, team boundary spanning was modelled as a second-order formative construct consisting of three different activities, i.e., ambassador, scouting and task coordination, each of them taking place inside as well as outside the organization (Ancona et al., 2009; Marrone, 2010; Sleep et al., 2015). Because of the formative nature of team boundary spanning, ambassador, scouting and task coordination activities are not expected to be strongly correlated. In this context, it was deemed appropriate to conduct additional analyses to determine whether the model proposed in Figure 1 would hold for the three activities of team boundary spanning. Accordingly, we re-run the proposed model for each activity of team boundary spanning separately. The results from these analyses, which are shown in the Appendix, are highly consistent with the results associated with the model that combines the three boundary spanning activities into one single construct. All two-way interactions (i.e., ambassador x social cohesion, scouting x social cohesion and task coordination x social cohesion) and five of the six three-way interactions (exception made for scouting x social cohesion x project newness) were significant at p<0.05. In sum, the model proposed in Figure 1 is valid for all components of the team 22 boundary spanning scale. Discussion Prior research has suggested that team boundary spanning can have a positive effect on the performance of NPD teams (Ancona and Caldwell, 1992; Howell and Shea, 2006). By establishing linkages and managing interactions with parties inside and outside the organization, boundary spanning allows NPD teams to access highly valuable external resources needed for the successful completion of the NPD projects (Ancona and Caldwell, 1992; Marrone 2010). Findings from this study extends research on this subject by revealing a positive effect of team boundary spanning on new product competitive advantage. Our results are consistent with research on NPD suggesting that product advantage depends, in part, on a NPD team’s access to external knowledge and resources (Li and Calantone, 1998; Kim and Atuahene-Gima, 2010). Although team boundary spanning is conducive to achieving new product competitive advantage, these actions may not always deliver the expected performance (Marrone, 2010). The current study makes an initial attempt to examine factors that can reduce team boundary spanning positive effect on new product competitive advantage by proposing and testing a negative moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage. Team social cohesion has generally been considered a desirable criterion for team formation and a predictor of new product performance (Im et al., 2013; Nakata and Im, 2010; Shaner et al., 2016). Results from this study confirm the above findings by showing that team social cohesion has a positive effect on new product competitive advantage. However, unique to this study is the finding that team social cohesion can reduce the positive effect of team boundary spanning on new product competitive advantage. Thus, the study’s results show that whereas team boundary spanning has a positive and 23 significant effect on new product competitive advantage when team social cohesion is low; the effect becomes insignificant when social cohesion is high. As argued earlier, teams with high levels of social cohesion are prone to show positive negative outgroup biases (Petersen et al, 2004; Rosenberg and Treviño, 2003). Such biases are likely to prejudice group members against absorbing and using information and resources stemming from boundary spanning activities, ultimately limiting the potential of these activities to deliver competitive advantage. Moreover, the study’s results reveal that project newness and external task interdependence accentuates the negative moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage. Both project newness and external task interdependence are argued to create a context in which socially cohesive groups would perceive external groups as threats to their collective identity. Regarding project newness, NPD teams developing highly innovative NPD projects are more likely to experience conflict with and resistance from groups and individuals inside and outside the organization because of the wide-ranging changes brought about by these types of products (Janssen, 2004). Accordingly, for highly cohesive groups developing high-innovative products, resistance from outgroup members may be conceived as an attack to the team’s collective identity in the sense that resisting groups may obstruct or impede a team’s efforts to successfully advance the highlyinnovative project. External task interdependence can also accentuate socially cohesive groups’ negative bias against external groups. External task interdependence creates work constraints for the NPD team as team members have to coordinate with others to be able to complete their job. Also, external task interdependence may cause teams to lose their distinctiveness and question their status within the organization, threating the team’s collective identity (Glynn et al., 2010; Langfred, 2000). Accordingly, under high project newness and external task interdependence, 24 highly cohesive teams will tend to disregard externally derived information and resources to the detriment of the new product performance. Interestingly, our results suggest that team social cohesion does not have a significant moderating effect on team boundary spanning when project newness and external task interdependence are low, which suggests that it is the perception of a threat what activates the negative moderating effect of team social cohesion. Implications Theoretical contributions Our findings extend previous research in several ways. First, the study contributes to research in team boundary spanning. In her review article of past research on team boundary spanning Marrone (2010) noted: “significant gaps still exist in our understanding of the team boundary spanning phenomenon” (p. 912). In particular, “surprisingly little empirical attention has been paid thus far to possible moderating conditions of the team boundary spanning-team performance relationship” (p. 927). Faraj and Yan (2009), and Gibson and Dibble (2013) are among the very few studies to empirically test and find moderators of this relationship. By investigating the moderating effect of team social cohesion in the relationship between team boundary spanning and new product competitive advantage, this study provides new empirical evidence about when team boundary spanning influences new product performance. Furthermore, by incorporating team social cohesion in our model, the current study supports the emerging recognition that team psychological factors are important to the effectiveness of a team’s boundary spanning efforts (Dokko et al., 2014). In addition, the current study brings a new perspective to the discussion of the effect of team social cohesion on NPD performance by examining team social cohesion in the context of intergroup relationships (i.e., team boundary spanning). Within the field of NPD, team social 25 cohesion has been studied primarily by researchers not so much interested in intergroup relationships but in intragroup processes (Brockman et al., 2010; Im et al., 2013). Within this perspective, it is noted that team social cohesion facilitates integration, communication and information sharing within the NPD team which, in turn, leads to higher new product performance (Im et al., 2013; Nakata and Im, 2010; Shaner et al., 2016). Moreover, wherever a negative effect of team social cohesion on new product performance has been identified, researchers have used groupthink theory to explain their findings (e.g., Brockman et al., 2010; Sethi et al., 2001). Thus, Brockman et al. (2010) noted that high levels of interpersonal cohesiveness can negatively impact new product performance due to team members’ unwillingness to express different opinions. Notwithstanding the previous studies, the current study departs from existing research in that it examines the effects of team social cohesion in the context of team boundary spanning (a.k.a. intergroup relationships). In doing so, the current study draws on social identity theory, which predicts that groups with high levels of social cohesion are likely to show high levels of intergroup bias (Hogg and Abrams, 1988; Petersen et al., 2004; Tajfel and Turner, 1986), to suggest that team social cohesion can undermine the positive effect of team boundary spanning on new product competitive advantage. Finally, the study’s results advance the emerging research on group identity and intergroup bias in team boundary spanning (e.g., Dokko et al., 2014; Richter et al., 2006) by showing that the moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage is accentuated by project newness and external task interdependence. Drawing on social identity research suggesting that socially cohesive groups exhibit a strong motivation to protect the group identity in situations where such identity is threatened by outgroup members (Turner et al., 1992; Turner and Pratkanis, 1998), we propose 32 Organization Studies 35(5): 703-726. Drach-Zahavy, A. and A. Somech. 2010. From an intrateam to an interteam perspective of effectiveness: The role of interdependence and boundary activities. Small Group Research 41(2): 143-174. Evanschitzky, H., M., Eisend, R.J. Calantone, and Y. Jiang. 2012. Success factors of product innovation: An updated meta-analysis. Journal of Product Innovation Management 29(S1): 21-37. Faraj, S. and A. Yan. 2009. Boundary work in knowledge teams. Journal of Applied Psychology 94(3): 604-617. Fornell, C. and D. Larcker. 1981. Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 18: 39-50. Frazier, P. A., A.P. Tix and K.E. Barron. 2004. Testing moderator and mediator effects in counseling psychology research. Journal of Counseling Psychology 51(1): 115-134 Gaertner, S.L., J.F. Dovidio, M.C. Rust, J.A. Nier, B.S. Banker, Ch.M. Ward, G.R. Mottola and M. Houlette. 1999. Reducing intergroup bias: elements of intergroup cooperation. Journal of Personality and Social Psychology 76(3): 388-402. García, N., M.J. Sanzo and J.A. Trespalacios. 2008. New product internal performance and market performance: Evidence from Spanish firms regarding the role of trust, interfunctional integration, and innovation type. Technovation 28(11): 713-725. Gibson, C.B. and R. Dibble. 2013. Excess may do harm: Investigating the effect of team external environment on external activities in teams. Organization Science 24(3): 697-715. Glynn, M.A., R. Kazanjian and R. Drazin. 2010. Fostering innovation in complex product development settings: The role of team member identity and interteam interdependence. 33 Journal of Product Innovation Management 27(7): 1082-1095. Goodman, P.S., E. Ravlin and M. Schminke. 1987. Understanding groups in organizations. In: Cummings, L.L., Staw, B.M. (Eds.). Research in Organizational Behavior 9: 121-173. Hair, J.F., W.C. Black, B.J. Babin, and R.E. Anderson 2009. Multivariate Data Analysis, Pearson, 7th edition. Hair, J.F., G.T. Hult, C.M. Ringle and M. Sarstedt. 2013. A Primer on Partial Least Square Structural Equation Modeling (PLS SEM). Sage publications, United Kingdom. Henard, D. H., and D.M. Szymanski. 2001. Why some new products are more successful than others. Journal of Marketing Research 38(3): 362-375. Henseler, J., C.M. Ringle and M. Sarstedt. 2015. A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science 43 (1): 115-135. Hewstone, M., M. Rubin and H. Willis. 2002. Intergroup bias. Annual Review of Psychology 53(1): 575-604. Hogg, M.A. and D. Abrams. 1988. Social identifications: A social psychology of intergroup relations and group processes. London: Routledge. Hogg, M.A. and D.I. Terry. 2000. Social identity and self-categorization processes in organizational contexts. Academy of Management Review 25(1): 121-140. Hogg, M.A. and S.C. Hains. 1996. Intergroup relations and group solidarity: Effects of group identification and social beliefs on depersonalized attraction. Journal of Personality and Social Psychology 70(2): 295-309. Howell, J.M. and C.M. Shea. 2006. Effects of champion behavior, team potency, and external communication activities on predicting team performance. Group and Organization 34 Management 31(2): 180-211. Im, S., M.M. Montoya and J.P. Workman. 2013. Antecedents and consequences of creativity in product innovation teams. Journal of Product Innovation Management 30(1): 170-185. Janssen, O. 2003. Innovative behaviour and job involvement at the price of conflict and less satisfactory relations with co-workers. Journal of Occupational and Organizational Psychology 76(3): 347-364. Janssen, O. 2004. How fairness perceptions make innovative behavior more or less stressful. Journal of Organizational Behavior 25(2): 201-215. Janssen, O., E. Van de Vliert, and M. West. 2004. The bright and dark sides of individual and group innovation: A special issue introduction. Journal of Organizational Behavior 25(2): 129-145. Johnson, R.E., C.C. Rosen, C.C. and E. Djurdjevic. 2011. Assessing the impact of common method variance on higher order multidimensional constructs. Journal of Applied Psychology 96(4): 744-761. Joshi, A., N. Pandy and G.H. Han. 2009. Bracketing team boundary spanning: An examination of task‐based, team‐level, and contextual antecedents. Journal of Organizational Behavior 30(6): 731-759. Kim, N. and K. Atuahene‐Gima. 2010. Using exploratory and exploitative market learning for new product development. Journal of Product Innovation Management 27(4): 519-536. Langfred, C.W. 2000. The paradox of self-management: Individual and group autonomy in work groups. Journal of Organizational Behavior 21 (5): 563-585. Leonardelli, G.J. and M.B. Brewer. 2001. Minority and majority discrimination: When and why. Journal of Experimental Social Psychology 37(6): 468-485. 35 Li, T. and R.J. Calantone. 1998. The impact of market knowledge competence on new product advantage: conceptualization and empirical examination. Journal of Marketing 62(4): 13-29. Man, D.C. and S.S. Lam, 2003. The effects of job complexity and autonomy on cohesiveness in collectivistic and individualistic work groups: a cross‐cultural analysis. Journal of Organizational Behavior 24(8): 979-1001. Marrone, J.A. 2010. Team boundary spanning: A multilevel review of past research and proposals for the future. Journal of Management 36(4): 911-940. Maurer, I., V. Bartsch and M. Ebers. 2011. The value of intra-organizational social capital: How it fosters knowledge transfer, innovation performance, and growth. Organization Studies 32(2): 157-185. McNally, R.C., E. Cavusgil and R. Calantone. 2010. Product innovativeness dimensions and their relationships with product advantage, product financial performance and project protocol. Journal of Product Innovation Management 27: 991–1006. Menon, A., B.J. Jaworski and A.K. Kohli. 1997. Product quality: Impact of interdepartmental interactions. Journal of the Academy of Marketing Science 25(3): 187-200. Nakata, C. and S. Im. 2010. Spurring cross‐functional integration for higher new product performance: A group effectiveness perspective. Journal of Product Innovation Management, 27(4): 554-571. Nunnally, J.C. and I.H. Bernstein. 1994. Psychometric Theory, 3rd ed. New York: McGraw-Hill. Oh, H., G. Labianca and M.H. Chung. 2006. A multilevel model of group social capital. Academy of Management Review 31(3): 569-582. Oh, H., M.H. Chung, M. H. and G. Labianca. 2004. Group social capital and group effectiveness: The role of informal socializing ties. Academy of Management Journal 47(6): 860-875. 36 Petersen, L.E., J. Dietz and D. Frey. 2004. The effects of intragroup interaction and cohesion on intergroup bias. Group Processes and Intergroup Relations 7(2): 107-118. Podsakoff, P.M., S.B. MacKenzie, J.Y. Lee and N.P Podsakoff. 2003. Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879-903. Potter, A. and B. Lawson. 2013. Help or hindrance? Causal ambiguity and supplier involvement in new product development teams. Journal of Product Innovation Management 30(4): 794808. Reinartz, W., M. Haenlein and J. Henseler. 2009. An empirical comparison of the efficacy of covariance-based and variance-based SEM. International Journal of Research in Marketing 26(4): 332-344. Rempel, M.W. and R.J. Fisher. 1997. Perceived threat, cohesion, and group problem solving in intergroup conflict. International Journal of Conflict Management 8(3): 216-234. Riccobono, F., M. Bruccoleri and A. Größler. 2016. Groupthink and Project Performance: The Influence of Personal Traits and Interpersonal Ties. Production and Operations Management 25(4): 609-629. Richter, A.W., J. Scully and M.A. West. 2005. Intergroup conflict and intergroup effectiveness in organizations: Theory and scale development. European Journal of Work and Organizational Psychology 14: 177-203. Richter, A.W., M.A. West, R. Van Dick and J.F. Dawson. 2006. Boundary spanners' identification, intergroup contact, and effective intergroup relations. Academy of Management Journal 49(6): 1252-1269. Rindfleisch, A., A.J. Malter, S. Ganesan, and C. Moorman. 2008. Cross-sectional versus 37 longitudinal survey research: Concepts, findings, and guidelines. Journal of Marketing Research 45(3): 261-279. Rosenberg, A.S. and L.K. Treviño. 2003. A proposed model of between-group helping: An identity-based approach. Journal of Managerial Issues 15(2) (Summer): 154-174. Schoonhoven, C.B. 1981. Problems with contingency theory: Testing assumptions hidden within the language of contingency theory. Administrative Science Quarterly 26(3): 349-377. Sethi, R. 2000. Superordinate identity in cross-functional product development teams: Its antecedents and effect on new product performance. Journal of the Academy of Marketing Science 28(3): 330-344. Sethi, R., D.C. Smith and C.W. Park. 2001. Cross-functional product development teams, creativity, and the innovativeness of new consumer products. Journal of Marketing Research 38(1): 73-85. Shaner, M.B., L. Beeler and C.H. Noble. 2016. Do we have to get along to innovate? The influence of multilevel social cohesion on new product and new service development. Journal of Product Innovation Management 33(S1): 148-165. Siguaw, J.A., P.M. Simpson and C.A. Enz. 2006. Conceptualizing innovation orientation: A framework for study and integration of innovation research. Journal of Product Innovation Management 23(6): 556-574. Sleep, S., S. Bharadwaj and S.K. Lam. 2015. Walking a tightrope: the joint impact of customer and within-firm boundary spanning activities on perceived customer satisfaction and team performance. Journal of the Academy of Marketing Science 43(4): 472-489. Song, X.M, and M.M. Montoya-Weiss. 2001. The effect of perceived technological uncertainty on Japanese new product development. Journal of Academy Management 44: 61-80. 38 Tajfel, H. 1982. Social psychology of intergroup relations. Annual Review of Psychology 33(1): 1-39. Tajfel, H., and J.C. Turner 1979. An integrative theory of intergroup conflict. In W. G. Austin, & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-37). Monterey, CA: Brooks/Cole. Tajfel, H. and J.C. Turner. 1986. The social identity theory of inter group behavior in S Worchel and WG Austin (Eds) Psychology of intergroup relations. Chicago: Nelson. Tellis, G.J., J.C. Prabhu and R.K. Chandy. 2009. Radical innovation across nations: The preeminence of corporate culture. Journal of Marketing 73(1): 3-23. Turner, M.E. and A.R. Pratkanis. 1998. A social identity maintenance model of groupthink. Organizational Behavior and Human Decision Processes 73(2): 210-235. Turner, M.E., A.R. Pratkanis, P. Probasco and C. Leve. 1992. Threat, cohesion, and group effectiveness: Testing a social identity maintenance perspective on groupthink. Journal of Personality and Social Psychology 63(5): 781-796. Turner, M.E., and T. Horvitz, T. 2001. The dilemma of threat: Group effectiveness and ineffectiveness under adversity. In M. Turner (Ed.) Groups at Work: Theory and Research: 445-470. Mahwah, N.J.: Erlbaum. Veldhuizen, E., E.J. Hultink and A. Griffin. 2006. Modeling market information processing in new product development: An empirical analysis. Journal of Engineering and Technology Management 23(4): 353-373. Williams, M. 2001. In whom we trust: Group membership as an affective context for trust development. Academy of Management Review 26(3): 377-396. Zahra, S.A. 1993. Environment, corporate entrepreneurship, and financial performance: A 39 taxonomic approach. Journal of Business Venturing 8(4): 319-340. 40 Figure 1 Conceptual Framework Team boundary spanning activities New product competitive advantage Team social cohesion External task interdependence Project newness 41 Table 1 Population and sample distribution by industry: Proportion test NAICS codes Industrial sector Amadeus directory Population Sample N % of total N % of total 311, 312 Food and beverages manufacturing 813 203 21.5% 23 16.4% 325, 326 Chemical and plastics product manufacturing 851 213 22.5% 36 25.7% 333 Machinery manufacturing 490 122 12.9% 15 10.7% 327, 331, 332 Non-metallic mineral product, primary metal and fabricated metal product manufacturing 781 195 20.6% 21 15.0% 334, 335 Computer, electronic, and electrical manufacturing 434 109 11.5%* 28 20.0%* 336 Transportation equipment manufacturing 417 104 11.0% 17 12.1% TOTAL 3786 946 100% 140 100% * Significant differences: p <0.05 48 Figure 2 The impact of external task interdependence on the effect that team social cohesion has on the relationships between boundary spanning (BS) activities and new product competitive advantage Eq.(1): (dProduct.adv/dBS) /dCohesion= -0.18 -0.28 x Task interdependence Figure 3 The impact of project newness on the effect that team social cohesion has on the relationships between boundary spanning activities and new product competitive advantage Eq.(2): (dProduct.adv/dBS) /dCohesion= -0.13 -0.19 x Project newness -0,7 -0,6 -0,5 -0,4 -0,3 -0,2 -0,1 0 0,1 0,2 0,3 0,4 -2,0 -1,0 0,0 1,0 2,0 -0,7 -0,6 -0,5 -0,4 -0,3 -0,2 -0,1 0 0,1 0,2 0,3 0,4 -2,0 -1,0 0,0 1,0 2,0 Project newness (dProduct.adv/dBS) /dCohesion (dProduct.adv/dBS) /dCohesion External task interdependence 49 Appendix Standardized Parameter Estimates for Ambassador Activities Model 1 Model 2 Model 3 Model 4 Hypothesized relationships Ambassador → NP comp. adv. Ambassador * Social cohesion → NP comp. adv. Ambassador * Social cohesion * External task interdependence → NP comp. adv. Ambassador * Social cohesion * Project newness → NP comp. adv. 0.13* 0.17* -0.18* 0.15* -0.21* -0.14* 0.16* -0.26** -0.22** Control relationships Firm size → NP comp. adv. Product innovation → NP comp. adv. Team social cohesion → NP comp. adv. External team interdependence→ NP comp. adv. Project newness → NP comp. adv. Social cohesion * External task interdependence → NP comp. adv. Ambassador * External task interdependence → NP comp. adv. Social cohesion * Project newness → NP comp. adv. Ambassador * Project newness → NP comp. adv. -0.12 0.28** 0.23** -0.09 0.17* -0.11 0.28** 0.23** -0.07 0.17* -0.09 0.28** 0.28** -0.05 0.15* -0.04 -0.04 -0.08 0.30** 0.20* -0.09 0.20* -0.09 -0.06 R2 NP competitive advantage 0.18 0.20 0.24 0.25 Significance levels: ** p<.01 (one-tailed test), * p<.05 (one-tailed test), + p<.10 (one-tailed test) Standardized Parameter Estimates for Scouting Activities Scouting Model 1 Model 2 Model 3 Model 4 Hypothesized relationships Scouting → NP comp. adv. Scouting * Social cohesion → NP comp. adv. Scouting * Social cohesion * External task interdependence → NP comp. adv. Scouting * Social cohesion * Project newness → NP comp. adv. 0.16* 0.22** -0.22** 0.20** -0.24* -0.13* 0.21** -0.26** -0.05 Control relationships Firm size → NP comp. adv. Product innovation → NP comp. adv. Team social cohesion → NP comp. adv. External team interdependence→ NP comp. adv. Project newness → NP comp. adv. Social cohesion * External task interdependence → NP comp. adv. Scouting * External task interdependence → NP comp. adv. Social cohesion * Project newness → NP comp. adv. Scouting * Project newness → NP comp. adv. -0.10 0.25** 0.24** -0.07 0.16* -0.11 0.28** 0.25** -0.04 0.17* -0.07 0.26** 0.30* -0.02 0.15* -0.04 -0.11 -0.07 0.30** 0.27** -0.05 0.20* -0.09 -0.05 R2 NP competitive advantage 0.19 0.23 0.27 0.28 Significance levels: ** p<.01 (one-tailed test), * p<.05 (one-tailed test), + p<.10 (one-tailed test) 50 Standardized Parameter Estimates for Task Coordination Activities Task coordination Model 1 Model 2 Model 3 Model 4 Hypothesized relationships Task coordination → NP comp. adv. Task coordination * Social cohesion → NP comp. adv. Task coordination * Social cohesion * External task interdependence → NP comp. adv. Task coordination * Social cohesion * Project newness → NP comp. adv. 0.03 0.04 -0.22** 0.07 -0.25* * -0.33** 0.10+ -0.25** -0.29** Control relationships Firm size → NP comp. adv. Product innovation → NP comp. adv. Team social cohesion → NP comp. adv. External team interdependence→ NP comp. adv. Project newness → NP comp. adv. Social cohesion * External task interdependence → NP comp. adv. Task coordination * External task interdependence → NP comp. adv. Social cohesion * Project newness → NP comp. adv. Task coordination * Project newness → NP comp. adv. -0.11 0.27** 0.23** -0.06 0.17* -0.10 0.28** 0.21** -0.04 0.18* -0.07 0.26** 0.31** -0.09 0.17* -0.04 0.06 -0.07 0.24** 0.24** -0.08 0.19* -0.09 -0.05 R2 NP competitive advantage 0.16 0.21 0.30 0.28 Significance levels: ** p<.01 (one-tailed test), * p<.05 (one-tailed test), + p<.10 (one-tailed test) 51 i At the suggestion of one reviewer, we also examined the possibility of a curvilinear moderating effect of team social cohesion on the relationship between team boundary spanning and new product competitive advantage. However, we did not find support for such effect. The interaction effect of boundary spanning and the squared term of team social cohesion was not significant.