And yet, non-equity cooperative entries do improve international performance: uncovering the role of networks’ social capital

Drawing on social capital theory, this study analyses the moderating role of social networks in the relation between international new ventures’ choice of non-equity cooperative entries and international performance. The research adopts an alternative point of view that considers that the development of social capital is dependent upon what actions an individual or group of people carry out to build and maintain social capital. Three relational norms have been associated with effective interaction among network partners, namely, informational exchange, organizational coordination and social conflict resolution. Through these relational norms, firms can co-create the structure of the social network and define what the network benefits and social capital are. Data gathered from a sample of international ventures operating in several industries support the idea that networks’ social capital endows international new ventures with informational advantages and experiential knowledge, which are important to reduce the problems associated with the non-equity entry mode choice when partners do not come from their networks. The results point to the need for INVs’ entrepreneurs to engage in establishing routines that enable them to develop management activities in coordination with their network members. The findings provide entrepreneurs of INVs with contextual evidence for making successful foreign market entry decisions.


Introduction
While progress has been made in exploring the role of networks' social capital (NSC) in international new ventures' (INVs) organizational processes (Anderson et al. 2010;Casson and Giusta 2007), questions still arise (Jantunen et al. 2008;Knight and Liesch 2016;Schwens et al. 2017). One of these questions is how NSC influences the relationship between non-equity cooperative entry modes (NCEMs) and INVs' international performance (Laufs and Schwens 2014;Knight and Liesch 2016;Schwens et al. 2017;Servantie et al. 2016). In fact, international entrepreneurship literature supports the thesis that NCEMs are the most convenient entry modes for early internationalization firms (Berg et al. 2008;Gabrielsson et al. 2008). However, normative literature on cooperation (Williamson 1985) advises that international new ventures might face an important challenge when using NCEM, namely, property right protection (Paul et al. 2017), suggesting the need to consider network social capital as an important safeguard mechanism (Coviello and Munro 1997). In this regard, the topic of social capital and the part it plays in the development of strategic alliances is a relevant line that warrants further research (Berg et al. 2008). This paper seeks, at least partially, to move this research line ahead.
Network social capital is supposed to influence INVs' international performance by means of trust and relational norms (Afandi et al. 2017;Liu et al. 2009), but in previous research network social capital has been mainly associated with the "goodwill/trust" that may exist in cohesive networks in which common values, language and goals are shared (Anderson et al. 2010;Casson and Giusta 2007). Nevertheless, network social capital results from firms' civic collaboration (Hessels and Parker 2013) and comes into being through a unique path-dependent process in which not only the structure of the network, the relational tie-factors and the shared values are critical (Anderson et al. 2010;Casson and Giusta 2007), but also the pattern of relations in the network (Afandi et al. 2017). Moreover, the characteristics of the network structure, the position an actor occupies in the network and the shared language or goals are, in some way, a function of the actors' network relational norms (Gulati et al. 2000), that is, the network pattern of relations that emerges from firms' informal commitment with their networking activity (Fink et al. 2008). The network pattern of relations is studied in this paper through the firms' informal engagement in information exchange, inter-organizational coordination and social conflict resolution mechanisms with one another (Dyer and Singh 1998;Gulati 1995;Liu et al. 2009;Ripollés and Blesa 2017).
Relational norms provide entrepreneurs with precise guidance on systematizing relationships inside social networks and creating a feasible environment in which NSC comes about (Adler and Kwon 2002;Portes 1998). As these effects have been largely neglected in the international entrepreneurship literature, a need exists to explore the role of network social capital from a perspective focused on networks' relational norms. Consequently, this paper proposes further theorybuilding research that delineates and investigates the role of network social capital through relational norms in the NCEM-INVs' international performance relationship. In this way, it extends the literature on NCEM and INVs' international performance (which has traditionally centred its attention on analysing firm size and age as the principal variables) by considering network social capital as a significant mediator variable . Additionally, this work advances previous international entrepreneurship research, which has mostly paid attention to the strategic consequences of NSC on INVs' knowledge base development (De Clercq et al. 2012;Lindstrand and Hånell 2017), but less on the entry mode strategy (Berg et al. 2008). Furthermore, unlike previous research lines that have analysed NSC from a structural perspective (Anderson et al. 2010;Casson and Giusta 2007), our study takes an instrumental approach in order to offer a more exact interpretation of how NSC is generated.
The paper starts with a review of the existing literature and the role played by NCEM in INVs. We then explain the rationale underlying our hypotheses and describe the method used to analyse them. Next, the results of the empirical analysis are introduced. These findings support the idea that networks' social capital endows INVs with informational advantages and experiential knowledge that is important to reduce the problems related to the choice of non-equity entry modes, even in cases in which partners do not come from their networks. The paper continues with the discussion and implications for scholarship and for firms. Finally, the limitations and proposed future research developments close the paper.

Concepts development
International performance One generally accepted definition of INVs is that proposed by Oviatt and McDougall (1994, p. 49). According to these authors, INVs are "business organizations that, from inception, seek to derive significant competitive advantages from the use of resources and the sale of outputs in multiple countries". Their early international activity is the main defining characteristic of these firms (Coviello and Jones 2004;Oviatt and McDougall 2005), and therefore, international new ventures' performance is highly related to their international activity (Hollender et al. 2017). Accordingly, this paper focuses on international performance, which refers to the firm's performance associated with INVs' international activity in the last three years.
Non-equity collaborative entry modes An entry mode can be defined as "a structural agreement that allows a firm to implement its product/market strategy in a host country" (Sharma and Erramilli 2004, p. 2). Pan and Tse (2000) developed a conceptual model which distinguishes between two kinds of entry modes: equity and non-equity. The first involves the investment of resources to set up foreign operations and to enable companies to get closer to international markets (Pan and Tse 2000). On the other hand, non-equity modes (e.g. export or collaborative modes) are much less investment-intensive entry modes but fall short in providing foreign market closeness (Pan and Tse 2000). Accordingly, in this paper, non-equity collaborative entry modes are those nonequity entry modes involving cooperation between international new ventures and foreign ventures.
Network social capital and relational norms Hite and Hesterly (2001, p. 277) defined networks as the "firm's collection of direct, dyadic, informal ties and the relationships between these ties, with the firm at the middle of the network as the focal actor". Informal ties consist of relationships that are tacit, particular and, in general, are not set up in accordance with any kind of legal agreements (Rank 2008). Consequently, relationships among these firms are informal and embedded in civic engagement (Afandi et al. 2017), maintained with a view to attaining common benefits (Kingsley and Malecki 2004) and are "self-organizing" because no one appears to be in charge (Casson and Guista 2007). Scholars agree that these networks are important for international new ventures' early international entry and growth (Oviatt and McDougall 2005;Jones et al. 2011).
Social capital has been the main theory used to explain the benefits that international new ventures can extract from their networking activity. Network social capital can assist these firms in realizing opportunities, acquiring resources, gaining legitimacy and achieving other desirable outcomes and strategies (Jones et al. 2011;Light and Dana 2013), such as organizational flexibility (Yousaf and Majid 2018). It has been viewed as a way to unlock the resources inside the network, but also as a resource in its own right (Afandi et al. 2017;Gedajlovic et al. 2013;Hohenthal et al. 2014). The core intuition guiding network social capital research is that social participation in networks can contribute to performance mainly by means of trust and relational norms (Afandi et al. 2017;Liu et al. 2009). Social participation contributes to generate trust, forgiveness and reciprocal relationships between network members (Adler and Kwon 2002). The "goodwill" that others show towards us through trust can be used for other purposes, such as information gathering or advice (Coleman 1988;Nahapiet and Ghoshal 1998), which in turn contributes to increase the performance of the networked firms.
Social participation also contributes to generate relational norms among the networked firms. Relational norms are understood here as the emergent pattern of relations derived from firms' engagement with their networking activity (Fink et al. 2008). This commitment does not derive from a formal agreement among firms within the network, but instead from the investments that the firms voluntarily decide to make in their networking activity in order to improve their results (Galkina and Chetty 2015).
Past research seems to suggest that network social capital should emerge in networks in which social participation is organized through three mainly relational norms: in networks in which firms engage in information exchange, inter-organizational coordination and social conflict resolution mechanisms with one another (Dyer and Singh 1998;Gulati 1995;Liu et al. 2009;Ripollés and Blesa, 2017). Through these relational norms, the networked firms provide them not only with informational advantages (Dyer and Singh 1998) but also with relational experience (Lindstrand et al. 2011) that can be used to increase performance in other organizational collaborative environments (Hohenthal et al. 2014), such as nonequity cooperative entry modes.
Interferences may also exist between trust and relational norms. Trust among network partners helps to diminish the potential dissemination risks that may exist among them, thereby enforcing their engagement with network exchange information, inter-organizational coordination and social conflict resolution mechanisms (Liu et al. 2009). On the other hand, commitment with these activities can contribute to generate trust among the networked firms as a result of mutual awareness and reputational concerns (Gulati et al. 2000). The choice of entry mode influences international new ventures' international performance because it entails a certain amount of resource investment in diverse markets, each of them with different degrees of risk, control and returns (Canabal and White, 2008;Morschett et al. 2010). Research has stated that international new ventures are found to use a broad variety of modes of entry, concerning both equity and non-equity (Aspelund et al. 2007;Gabrielsson et al. 2008;Ripollés et al. 2012), and that both have associated potential benefits and risks (Paul et al. 2017). However, non-equity formal relations with foreign partners seem to be international new ventures' preferred entry modes because they are less resource intense, provide greater flexibility and can help these firms to complement their capabilities (Jones et al. 2011;Knight and Liesch 2016). In fact, research suggests that resource scarcity, in addition to lack of foreign knowledge and limited market legitimacy, are the main arguments supporting the relationship between nonequity cooperative entry modes and INVs' international performance (Fink et al. 2008;Gabrielsson et al. 2008). Therefore, we expect to confirm this relationship in our study: Hypothesis 1: Non-equity cooperative entry modes contribute to INVs' international performance.

The moderating role of relational norms
A number of advantages can be gained from using nonequity cooperative entry modes in terms of both resource investment and the development of a competitive advantage in international markets. Non-equity entry modes, however, may still spread the risk of losses arising from opportunistic behaviour carried out by partners (Berg et al. 2008;Burgel and Murray 2000). Even though international new ventures could anticipate future partners' opportunistic behaviour, for these firms, it is not easy to protect themselves by means of patents or other legal ways (Williamson 1985), as their weak capacity to negotiate would limit the effectiveness of those instruments (Berg et al. 2008).
A non-equity collaborative relationship must be cultivated and upheld while trade occurs if efficiency is to be achieved, and INVs might encounter important challenges when it comes to doing so. For example, in nonequity cooperative entry modes, increased access to markets is obtained by INVs in exchange for taking the decision to surrender some control (Coviello and Munro 1997), which forces them to trust their partners' goodwill (Berg et al. 2008). International new ventures have no direct contact with their foreign markets. Information is provided by middlemen, who may filter it to suit their own interests. As a result, the knowledge base of INVs could be compromised, and with it the sustainability of their competitive advantage (Aspelund et al. 2007).
Consequently, for these constrained firms, it is relevant not only what resources a cooperative relationship generates but also how costly that relationship is (Hessels and Parker 2013), which brings to the fore the important role that INVs' NSC may play (Berg et al. 2008).
The moderating role of network information exchange activity International new ventures need information about potential non-equity cooperative foreign partners and about their suitability. Additionally, potential partners must be sure that cooperation will produce reciprocal gain, even with the notable shortcoming of international new ventures' lack of legitimacy (Berg et al. 2008). The commitment of the networked firms to exchange information with one another can be a reliable source. Dyer and Singh (1998) suggest that information exchange among members in a network can help to create a trustful atmosphere that facilitates inferences about the trustworthiness of the information transferred. It can lessen opportunism by making it more probable that such behaviour will be discovered and that the information will expand quickly throughout the network (Gulati et al. 2000). Because reputation takes time to build but can easily be destroyed, the exchange of information among networked partners can create strong disincentives for opportunistic behaviour (Gulati 1998).
International new ventures recognize two possible scenarios: (1) that NCEMs' partners are part of the firm's network and (2) that NCEMs' partners are not among their contacts (Ripollés and Blesa 2017). In both cases, the commitment of the networked members in exchanging information with each other can offer access, timing and referrals, which lowers the likelihood of suffering from partners' opportunism (Burt 1992). In the first scenario, information exchange can provide international new ventures with information on the capability and appropriateness of the members of their networks to establish non-equity cooperative contracts. Through this information exchange, international new ventures have an opportunity to evaluate their partners' intentions and motives (Gulati 1998;Coviello and Munro 1997) in a timely way. Focusing on the second of these scenarios, it seems reasonable to think that international new ventures will turn to their network partners in search of advice and help when thinking of choosing new nonequity cooperative partners to enter foreign markets. Through their network partners, INVs can obtain finegrained information about the capabilities and likely behaviour of a potential partner, who is not a member of the INV's network of contacts. Moreover, international new ventures can communicate to common partners any bad behaviour by either NCEM's partner through the information exchange activity, which serves as an effective constraint for both and decreases the risk of dissemination and opportunism (Gulati 1998). Through the information exchange activity among the network members, a reputational lock-in could be created whereby good behaviour of non-equity cooperative partners is guaranteed through an interest in local reputation.
By exchanging information with their network members, INVs have an opportunity to evaluate possible partners' intentions and motives. As Al-Laham et al. (2008, p. 349) suggested, "transacting with firms with whom information is available is less risky than transacting with firms whose collaborative behaviour is unknown". Therefore: Hypothesis 2: Information exchange among the members in a network positively influences the relationship between non-equity cooperative entry modes and INVs' international performance.
The moderating role of inter-organizational coordination Inter-organizational coordination refers to the extent to which network partners commit themselves to the informal establishment and use of common procedures (Helfert et al. 2002) for adaptation between network partners' activities (Löfgren et al. 2008). Being aware that network members may face disadvantages for defective behaviour makes them abide by the implicit rules of social cooperation (Fink and Kessler 2010). Therefore, inter-organizational coordination helps to increase solidarity among the networked partners. Solidarity refers to the expectation that network members will generally act to increase mutual benefit and engage in coordinated actions (Lumineau and Henderson 2012). Through the network members' commitment to inter-organizational coordination, firms have an opportunity to assess their network members' intentions, motives and cooperative capabilities, as well as their own capability to cooperate, and to learn what behaviour is satisfactory when adopting cooperative strategies (Ripollés and Blesa 2017). This relational experience creates a basis for foresights about future partners' behaviour and makes it easier to draw inferences about their conformity with the rules of nonequity collaborative agreements (Dyer and Singh 1998).
In addition, this experience in relationships can also be valuable when NCEM partners have not been members of previous networks. In this situation, experience in conducting relationships in a network setting can be used in new cooperative agreements (Hohenthal et al. 2014). That is, when a partner does not come from INVs' networks, firms do not have any prior background to enable them to outline the phases with which the objectives of non-equity cooperative entry modes are fulfilled. With no previous experience, international new ventures see their commitment in inter-organizational coordination among their network members as a very useful experience that can help them to specify allowable functional borders in new collaborative agreements (Dyer and Singh 1998). Thus, inter-organization coordination among network members can help INVs to manage MCEMs, even though partners do not come from their networks. Therefore: Hypothesis 3: Inter-organization coordination among the network members positively influences the relationship between non-equity cooperative entry modes and INVs' international performance.
The moderation role of social conflict resolution mechanisms Social conflict refers to the degree to which network partners have competing interests, preferences and practices that cannot be easily reconciled. The use of social conflict resolution procedures addresses exceptional non-regular situations, which are bound to happen in lengthy relationships (Ruekert and Walker 1987). Organizations related by social interaction employ non-contractual methods linked with conflict, concurrence, cooperation and rivalry (McLoughlin and Horan, 2000). When partners engage in social conflict resolution, it increases their solidarity with the relationship (Helfert et al. 2002) and reinforces trust among network collaborators (Lumineau and Henderson 2012). Therefore, social conflict resolution mechanisms better fulfil the needs and interests of networked firms (Claycomb and Frankwick 2010) and lead to common satisfying solutions, which in turn makes the relationship more likely to be successful (Mohr and Spekman 1994). Social conflict resolution calls for an appropriate response to conflict, a willingness to negotiate and an awareness of the need for fairness (Helfert et al. 2002). The development of social problem-solving provides international new ventures with important experiential knowledge that can be used when conflict with their non-equity entry mode partners arises. Consequently, it seems logical to expect that international new ventures accustomed to carrying out social conflict resolution mechanisms among their network members will try to replicate them when establishing formal agreements, which applies to different non-equity cooperative entry modes. This is especially so because INVs are aware of the significant potential risks that are posed by conflicts with their partners that they are unable to resolve satisfactorily. Indeed, at stake would be not only the loss of a market but also a possible deterioration in their competitive advantages (Burgel and Murray 2000;Melén and Rovira 2009). Reducing conflicts also helps to meet the requirements of formal collaboration and leads to highlevel performance (Li et al. 2013). Therefore, the international new ventures' participation in networks in which partners engage themselves in social conflict resolution mechanisms gives international new ventures a valuable relational experience that can be used when conflicts with their non-equity cooperative partners appear, as well as when they differ from network partners. As posited earlier, the relational experience gains in one context can be easily transferred to other cooperative settings. Hence: Hypothesis 4: Social conflict resolution mechanisms among the members in a network positively influence the relationship between non-equity cooperative entry modes and INVs' international performance.

Study context
Two hundred Spanish INVs operating in several sectors and members of a network were interviewed to gather the data. As in Wu et al. (2018), in this paper for a firm to be considered as participating in a network, it had to state that it had social relationships with at least two other independent domestic or foreign companies, which also have an informal relationship between them (Ripollés and Blesa 2017).
The Dun and Bradstreet Database from the year 2010 was used to select firms for the survey according to three criteria. First, a firm was considered a new venture if it was four years old or less . Second, more than 25% of firms' annual sales are required to come from abroad (Harveston et al. 2000). Lastly, the firms should be not part of a business group. This resulted in 2019 Spanish international new ventures that could be considered candidates for inclusion in the sample.

Data collection
Managers were contacted by phone and asked to participate in the survey and to provide their email addresses. The scales were translated from English into Spanish, and a pre-test was conducted with 10 managers. Following suggestions put forward by the managers, some of the items were slightly reworded. Each of the managers was then sent an email with the link to the questionnaire posted on the Internet. The responses were collected during the fourth quarter of 2010 and the first quarter of 2011. After three reminders, the final sample consisted of 200 INVs. A summary of the main characteristics of the sample can be seen in Table 1.

Measuring instruments
Non-equity cooperative entry modes were conceptualized as exporting through an independent agent and other contractual agreements. Brouthers and Nakos (2004) suggest that SMEs entering markets where the country risk level is perceived to be high tend to prefer non-equity cooperative modes of entry. Consequently, country risk was considered when measuring nonequity cooperative entry modes by three items that gather the number of times they had entered a new high-, medium-or low-risk country, according to data from the "International Country Risk Guide", as has been performed in previous research (Rasheed 2005).
Authors in favour of objective performance measures state that they suffer less from common method bias (Hollender et al. 2017). However, it has been found that, for new firms, it is not always possible to supply objective measures of performance (Woodcock et al. 1994).
In addition, researchers (Geringer and Hebert 1991;Hollender et al. 2017) have agreed that objective performance measures correlate well with subjective ones, suggesting the reliability of subjective measures in the context of new ventures and small firms. Consequently, subjective performance measures were used. Managers have been previously employed to measure the overall performance of entrepreneurial firms and have proven to be a reliable source of information (Brush and Vanderwerf 1992). Managers were therefore asked about the extent to which they were satisfied with the following items concerning their international activity during the previous three years: market share, market access, sales volume, know-how development, profitability, image development and general performance.
The degrees of information exchange, interorganizational coordination and social conflict resolution among the networked members were measured using an adaptation of Helfert et al.'s (2002) scale, proposed within the context of relational marketing. Participants assessed their perceptions on the degree to which the members of their network are committed to information exchange, inter-organizational coordination and social conflict resolution ( Table 2).
The firm-and industry-related factors that were controlled for include the size of the firm (number of employees), its experience with regard to exporting and the sector it belonged to. According to previous research, these factors can affect the internationalization processes and the international performance of INVs (Jantunen et al. 2008;Schwens et al. 2017). The analysis showed that the results obtained were not due to factors other than the independent variables included in the model (Table 3).

Control analyses
Several procedures were followed to control for common method bias . Respondents were assured complete anonymity, and they were asked to be as honest as possible in their answers. Additionally, this study used different response formats, including not only open questions but also Likert scales. A Harman's (1967) single factor test was conducted as a statistical measure. Unrotated factor analysis extracted seven significant factors with eigenvalues greater than one, which explained 67.11% of the variance (22.33%, 14.39%, 8.96%, 6.86%, 5.35%, 5.04% and 4.18%). Following Podsakoff et al. (2003), production in international markets was used as a marker variable that was theoretically uncorrelated to at least one variable in the conceptual model. The lowest correlation of the marker variable to the focal construct items was 0.232 (p > 0.95), which represents the upper bound for a potential common method variance. According to these results,   the influence of the common method variance can be ruled out.
To test for non-response bias, we compared the early and late responses with regard to all measures of the model constructs to be tested. Each data set was divided into thirds according to the time from initial emailing until reception of the completed questionnaire. The analysis of the t tests did not show any significant differences (p < 0.05 level) between the first and the last thirds, and accordingly, non-response bias was ruled out (Armstrong and Overton 1977). To ensure that certain features of the samples were not affecting the main results of the model variables, age, international experience, firm economic sector and size were used as independent variables to conduct MANOVA. No significant differences (p < 0.05 level) were found.

Scale validity and reliability
To measure the formative models of NCEMs and international performance, the source used for all the items was a review of related literature that was conducted in order to establish content and indicator specification (Jarvis et al. 2003). Furthermore, attempts were made to guarantee that the items fitted the conceptualization and covered all the relevant dimensions. The highest variance inflation in the indices of each scale (2.004 for NCEMs and 1.044 for international performance) was well below the acknowledged limit of 10 (Kleinbaum et al. 1988). Lastly, multi-indicator multi-cause models were estimated in order to test external validity (Jarvis et al. 2003). Market knowledge and clients' knowledge for NCEMs and design of unique products and development of high-quality products for international performance were used as reflective indicators respectively for each construct. Estimates gave good overall fits in both models.
Confirmatory analysis was used to evaluate the convergent validity as regards information exchange, interorganizational coordination and social conflict resolution. Item INEX3 was removed from the information exchange construct as it failed to reach a lambda of 0.5. The results of the validity analysis showed a high lambda in the rest of the parameters, a high significance at p < 0.001 in all t values and good fit indices. The scales also presented good indices of reliability with the exception of the variance extracted for the construct social conflict resolution. Anderson and Gerbing's (1988) confidence interval and the extracted variance (Fornell and Larker 1981) were used as the basis to test discriminant validity. Satisfactory results were obtained in all these tests (Table 4).

Results
The effects of non-equity cooperative entry modes and international performance were examined by means of SEM (MLE procedure in LISREL). To test the moderator effects of information exchange, interorganizational coordination and social conflict resolution, the total sample was divided into two halves for each construct, according to their low (1-3.5) or high (4-5) position in an index calculated using the mean of the indicators. Hence, the effect of non-equity cooperative entry modes on international performance was tested in six samples: low information exchange sample (91), high information exchange sample (109), low inter-organizational coordination sample (95), high inter-organizational coordination sample (105), low social conflict resolution sample (96) and high social conflict resolution sample (104). This procedure allowed comparisons to be conducted because only two things differed between the low and high groups, Organizational coordination … discuss in collaboration who is doing what in an inter-firm context (Coordin1) … ensure that promises on all parts are fulfilled (Coordin2) … discuss the steps with which the aims of the relationships are fulfilled (Coordin3) Social conflict resolution … try hard to understand our firm's interests in the case of conflicts (Conflic1) … wait a considerable time in the case of conflicts to allow the situation to calm down (Conflic2) … try to establish a compromise which is acceptable for all sides when a conflict arises (Conflic3) Adapted from Helfert et al. (2002) Table 3 Means, standard deviations and partial correlation coefficients  namely, the data in the covariance matrices and the sample sizes (Jöreskog and Sörbom 1993). Table 5 offers a summary of the results obtained. The parameters show different effects of non-equity cooperative entry modes on INVs' international performance depending on the sample considered. Not all the proposed relations were confirmed in all the samples. The influence of non-equity cooperative entry modes on INVs' international performance, when significant, is quite low. Nevertheless, this is one of the few studies that have shown the existence of the aforementioned relation empirically. In fact, as posited in Hypothesis 1, NCEMs have a significant and positive effect on international performance for the total sample of INVs. However, although the analysis shows a significant effect of NCEMs on INVs' international performance when the level of information exchange is low, the fit of the model to the data presents serious problems with an RMSR far above the threshold of 0.08. On the other hand, the analysis shows a positive and significant effect in the case of high levels of information exchange (λ = 0.16, t = 2.89) with a good model fit (χ 2 /df = 1.35, RMSR = 0.05, GFI = 0.95, NFI = 0.98, CFI = 0.99, IFI = 0.99). Hypothesis 2 proposed a moderator effect of information exchange whereby a rise in the level of this variable would increase the positive effect of nonequity cooperative entry modes on INVs' international performance. The results of these analyses taken together allow the hypothesis to be accepted, as they show that, when high levels of information exchange are developed, the effects of non-equity cooperative entry modes on INVs' international performance are significantly positive, in contrast to the case of low levels of information exchange.
Hypothesis 3 suggested a higher influence of nonequity cooperative entry modes on INVs' international performance when the level of inter-organizational coordination was high. This influence is not significant in the context of low coordination but is found to be positive in the case of high coordination. These results show that inter-organizational coordination does not have any negative moderating effect on the relationship between non-equity cooperative entry modes and INVs' international performance, but there is a positive one when the level of coordination is high. Consequently, Hypothesis 3 has been confirmed.
Finally, both tests of Hypothesis 4 showed a poor fit of the data to the model, although the goodness-of-fit statistics were better in the high social conflict resolution  sample than in the low one. The results show a nonsignificant relation with NCEMs in the low conflict resolution sample, while that relation is significant and positive in the case of the sample of high social conflict resolution. Though these results should be considered with caution, they seem to point to a higher influence of high social conflict resolution activities on the effects of NCEMs on INVs' international performance.
6 Discussion and conclusions

Theoretical contributions
This study has demonstrated that international new ventures can take advantage of their networks when using non-equity cooperative entry modes, and that NSC plays an important moderating role in the relationship between non-equity cooperative entry modes and INVs' international performance. In general, we make several compelling contributions in this study. This study has empirically confirmed the positive relationship between non-equity cooperative entry modes and INVs' international performance, previously suggested in the literature (Aspelund et al. 2007;Burgel and Murray 2000;Coviello and Munro 1997;McDougall et al. 1994;Paul et al. 2017). However, our results add to this research by highlighting the need to consider moderating variables (Laufs and Schwens 2014); specifically, the importance of NSC as a moderating factor has been confirmed.
As acknowledged by previous research (e.g. Berg et al. 2008;Coviello and Munro 1997), we found NSC to be an important element contributing to the relationship between non-equity cooperative entry modes and INVs' international performance. In addition, we have  followed an instrumental perspective to study them. The focus has been on the informal relational norms that govern network relationships and not on the structural characteristics that networks with a high NSC potential are assumed to possess (Stuart and Sorenson 2007).
Considering that NSC is also akin to relational norms is coherent with those authors who suggest the need to study the mechanisms underlying "trust" among the networked firms (Gedajlovic et al. 2013). Therefore, by moving the focus to the informal pattern of network relationships, this paper adds to previous research dealing with NSC and IE. In fact, previous research has associated NSC as a given in cohesive networks and has also suggested the existence of potential negative effects, such as the pressure of mental conformity, restricting escape from disappointing partners, lessening objectivity and protecting commonness (Adler and Kwon 2002;Li et al. 2013). Studying NSC through firms' involvement with network exchange information, inter-organizational coordination and social conflict resolution mechanisms might help us to understand how networks in which social capital exists overcome the abovementioned negative effects. Moreover, whereas past international entrepreneurship research has mainly associated NSC with informational advantages and studied its influence on INVs' knowledge development (Anderson et al. 2010;Casson and Giusta 2007), this paper has extended previous research by stating another important advantage associated with network social capital, called relational experience. By considering relational experienceexperience derived from being engaged with network information exchange, inter-organization coordination and social conflict resolution mechanisms-as an important advantage associated with NSC, we can potentially further advance our understanding with regard to networking and IE. This is in line with the Lindstrand and Hånell's (2017) research, in which they relate international social capital to the experience derived from establishing formal internationalization agreements with different types of firms.
Finally, the construct linking networks and NSC in non-equity cooperative entry modes in the INV context has never been operationalized or empirically tested using quantitative research (e.g. Berg et al. 2008;Coviello and Munro 1997). In addition, this paper has adopted a novel way to measure NSC because it has been operationalized through three relational norms: information exchange, coordination and social conflict resolution mechanisms among networked members. Although literature has identified both strategic and social networks, international entrepreneurship research has mainly focused on the role of strategic networks in foreign market entry strategies (Gabrielsson et al. 2008) or in international opportunity exploitation (Lindstrand and Hånell 2017). However, this research line is incapable of explaining the internationalization of firms without any network contacts to a particular market (Ojala 2009). In this situation, the networks studied here play an important role. The results obtained in this paper complement this research by showing their effect on the relationship between non-equity cooperative modes of entry and the international performance of international new ventures.

Practical implications for managers
From an applied perspective, this research points out the suitability of non-equity cooperative entry modes to manage international activities of international new ventures. Literature has highlighted the fact that non-equity cooperative entry modes could have serious negative consequences for international new ventures derived mainly from the lack of legitimacy in the market and their small size. The results obtained in this research show that these consequences can be substantially lessened or removed with a management of social relationships that contributes to raise the INVs' social capital. The relevant conclusion is that not only is it important to participate in cohesive or dispersed social networks, but that the member of that social network will commit to creating social capital. Additionally, the members of the social network have to share information, commit themselves to the informal establishment and use of common organizational procedures and adopt social mechanisms to solve conflicts. The development of these relational norms will help international new ventures to benefit from the intangible advantages related to their participation in social networks, such as informational advantages and relational experience. These advantages can be easily transferrable to formal cooperation, such as the development of non-equity cooperative entry modes.
Therefore, results point to the need for INVs' entrepreneurs to get involved in the complex task of setting up relational norms with the members of their network. Findings should provide entrepreneurs of international new ventures with contextual evidence for making successful foreign market entry decisions.

Limitations
These results should be taken with the reservations characteristic of this type of study. Due to the restricted availability of data that can be used to analyse the model proposed, this research focuses on a sample of only 200 Spanish firms. Although such an approach is not uncommon, the fact that the sample is made up of only Spanish INVs does limit the extent to which the findings can be generalized. Research in other countries may help to make the results more generalizable. Similarly, other studies might centre their attention on specific sectors.
The data were collected from individual subjective observations about networks, and thus, in some way, what is being captured is the entrepreneur's general perception of the network partners' investment in relational norms. One issue related to this procedure is whether only one person can provide adequate information about all the organization. Another possible problem is the fact that these managers gave answers about the activities carried out by a whole network of firms. Finally, as the fieldwork took place online, there is no way of telling who the actual respondent is.
Cross-sectional methods are a limitation for causal deductions. Testing the proposed model by means of longitudinal data and qualitative research would help to explain the relationships suggested. In this way, the points of view and judgements of several partners could be considered as a way to collect contrasted responses that are closer to the real network experience. Finally, cultural factors affecting relational norms cannot be ruled out and NSC has been proved to be affected by cultural norms (Li et al. 2013).