Development of interfirm network management activities: The impact of industry, firm age and size

Abstract This article investigates the structural characteristics of firms that promote activities involving partners who coordinate with each other to achieve common or individual goals. The article also aims to verify empirically whether these activities generate advantages for companies embedded in relationships by examining the effects of industry, age and size on interfirm network management activities in a sample of Spanish companies operating in several industries and belonging to networks. The results show differences according to the life cycle stage: growth or maturity. Only the relation between interfirm network management activity and performance has been confirmed in both samples. The findings point to the need to consider the industrial environment when analysing firms’ networking decisions because the situations they face differ in mature or growing industries.


INTRODUCTION
R elationships are often regarded as the foundations for success in a more global, uncertain and competitive environment (Morgan & Hunt, 1994), and networks constitute the frameworks for all activities that take place in business relationships (Mattsson, 1997). This paper specifically focusses on the firm's network, which directly influences the flow of resources across the firm's boundaries. The firm's network consists of its set of direct, dyadic, informal ties and the relationships between these ties, with the firm at the centre of the network as the focal actor (Hite & Hesterly, 2001). Informal ties comprise relationships defined as implicit, personal, generic and not fixed by any legal arrangement (Rank, 2008). Formal ties that prevail in strategic networks (Jarillo, 1988) are defined as being explicit, impersonal and functionally specific relationships among firms (Rank, 2008).
Researchers have considered two main perspectives in order to study firms' networks and their effects: the structural and the managerial perspectives. The structural perspective examined how the structure of networks and quality of ties affected resource flow and influenced business behaviour (Hoang & Antoncic, 2003;Batjargal, 2006). Traditionally, this perspective used three dimensions: the first dimension focussed on the structure of the network and the properties of the position occupied by the agent in the network (structural dimension), the second dimension summarized the characteristics of the agent's relations, such as confidence and longevity of the link, (relational dimension) and the network management tasks identified by Ritter, Wilkinson, and Johnston are not entirely appropriate (Ritter, Wilkinson, & Johnston, 2002, 2004Ritter & Gemünden, 2003). In these situations, all firms in a network will be simultaneously involved in its ongoing management, and the resulting strategy is coproduced by their actions. In this work, we focus specifically on this situation that has not been addressed in the literature; precisely because in business networks where informal ties prevail, it is more usual for firms to face situations where it is difficult a priori to identify their network strategy. Consequently, in this paper we propose slightly different network management tasks to the ones proposed by Johnston (2002, 2004). We examine the following tasks below: interfirm knowledge sharing, resource sharing, coordination, conflict resolution and adaptation between network members. We refer to these network management tasks as interfirm network management activities (INMAs). Thus, our first supposition in this paper is that INMAs help create an effective coworking environment that enables firms to use the potential shared benefits of networking to enhance their own performance by facilitating their adaptation to customer needs. We adopt a marketing focus (Helfert, Ritter, & Walter, 2002) placing network firms' customers satisfaction as an important element in determining common benefits. We are aware that other strategic elements may influence common benefits and justify the development of INMAs like, for example, technological learning, but they lie outside the scope of this work. Given the lack of research in this area, our first research question is:

RQ1 Do INMAs influence firm performance?
Inspired by the contingency perspective (Chandler, 1962;Miles & Snow, 1978;Miller & Friesen, 1982;Covin & Slevin, 1989), in this research, we argue that a company's ability to engage in INMA will depend, in part, on its organizational resources. INMA tend to be resource-consuming activities, therefore the development of INMA will be, to some extent, limited by its resource base. Firms with abundant resources may have a greater capacity than those with sparse resources to engage in INMA. Although different variables have been defined as proxies of firm's resources (e.g., Covin & Slevin, 1989), as others before in the network context we use size and age (Håkansson, 1982). However, a negative effect may be also identified if we use these variables. Bigger and older firms usually develop routines that diminish their flexibility to respond to changes required by INMA to adapt to customer needs (Autio, Sapienza, & Almeida, 2000). Thus, our second research question is: RQ2 Do firm age and size contribute to the development of INMAs?
The contingency perspective in management also argues for the need to consider environmental characteristics as important determinants of management activities (Chandler, 1962) and has received substantial empirical support (Walter, Kellermanns, Floyd, Veiga, & Matherne, 2013). Past research makes it clear that the nature of industries evolves over time through their life cycle (e.g., Levitt, 1965;Grant, 2010). Contingency theory suggests that the management elements that determine firm adaptation to customer needs will be reconfigured as the life cycle shifts from one stage to another. Although the literature contains a significant body of research supporting this influence (Karniouchina, Carson, Short, & Ketchen, 2013), none of these studies have accounted for the potential effects of changes in life cycle stages on comanagement activities such as INMAs. Consequently, our third research question is: RQ3 Do industry life cycle stages influence the development of INMAs?
By highlighting the importance of INMAs, this study extends previous network management research mainly focussed on relationship-specific tasks and cross-relational tasks (Ritter, Wilkinson, & Johnston, 2004) to include insights into how firms in a network develop conjoint bottom-up management activities. Moreover, gaining additional insights into how firms contingency variables Maria Ripollés and Andreu Blesa (size and industry life cycle stages) can contribute to the development of INMAs will enable us to better understand firms' networking activities from a managerial perspective.
Furthermore, this study also provides suggestions for researchers when considering variables like industry life cycle, company age and size as control variables. In short, we propose a conceptual model to explain firm performance that relates age, size and industry life cycle with INMA and INMA with firm performance. The following section presents the theoretical background and the relationships between the structural factors studied and INMA. Then, the method for analysing our hypotheses is explained, followed by a discussion of the results. Finally, the conclusions, the implications, the limitations and proposed future research developments are presented.

INMAs AND FIRM PERFORMANCE
In business networks, where informal ties prevail, the managerial challenge is that the firm mainly has to cope with managing interactions taking place in multiple relationships, which may be with partners not entirely of the firm's choosing and have been in operation for some time. Therefore, each partner has a history that exerts an influence on how things are done (Ritter, Wilkinson, & Johnston, 2004). In these situations, firms need to develop different cross-relational tasks to the ones proposed by Ritter, Wilkinson, and Johnston. We have identified five INMAs firms in a network need to perform to successfully meet customer's needs: interfirm knowledge sharing, resource sharing, coordination, adaptation and conflict resolution (Helfert, Ritter, & Walter, 2002).
Interfirm knowledge sharing is defined as the set of activities performed jointly by firms in the network enabling them to obtain valuable information from their customers and conjointly develop solutions for improving their offerings. These activities enable network partners to streamline the flow of customer information across organizational boundaries (Shih, Hsu, Zhu, & Balasubramanian, 2012), in turn improving firm's agility and adaptability to new customer needs (Robson, Skarmeas, & Spyropoulou, 2006). Knowledge sharing within a network allows a firm to acquire information about its relationship partners, including their resources, needs, capabilities, strategies and other relationships (Johanson & Vahlne, 2009). Such information-sharing activities allow organizations to expand their customer knowledge pool, deliver value-added products or services, detect emerging opportunities and capture business benefits in a hypercompetitive business environment (Shih et al., 2012). The process of creating knowledge is not separate from the other activities in business relationships; rather it is embedded in them. Knowledge accrues not only from the firm's own activities, but also from the activities of its partners, and as those partners also have other relationship partners with whom their activities are coordinated, the firm is indirectly engaged in a knowledge creation process that extends far beyond its own horizon. Thus, a network of business relationships provides a firm with an extended knowledge base (Kogut, 2000). Effective knowledge-sharing activities enable network partners to streamline the flow of customer and market information, money and products across organizational boundaries, in turn improving the agility, adaptability and predictability of the network. These activities are a critical factor for collaborative resource coordination, allocation and integration across different members of the network (Kim, Umanath, Kim, Ahrens, & Kim, 2012).
In addition to these practices, business networks offer their members a portfolio of services designed to overcome the competitive weaknesses of individual firms. Services shared among members could range from negotiating and purchasing from suppliers, marketing, personnel development, to financial services, quality management, inventory optimization and market research. Each network is able to define the services most relevant to its members (Wegner & Padula, 2010).
Interorganizational coordination refers to synchronization of partners' actions (Mohr & Nevin, 1990). Network coordination can be seen as routines for integrating network activities (Löfgren, Tolstoy, Sharma, & Johanson, 2008). Industrial Marketing and Purchasing project studies show that Development of interfirm network management activities relationships usually involve a number of managers who work together to coordinate their firms' activities and create interrelated routines (Cunningham & Homse, 1986). This coordination comprises the establishment, use and control of formal rules and procedures and the exertion of informal influence (Helfert, Ritter, & Walter, 2002). Grandori and Soda (1995) cite a set of practices that involve the planning, communication and evaluation of strategies. These functions must be modified to suit the dynamics of networks, which are kept in operation by constant negotiation processes. Moreover, evaluation of the results provides information that feeds back to the management of the network and should result in changes (Wegner & Padula, 2010). Awareness that the network partner may face disadvantages in return for defective behaviour motivates the actor to fulfil the implicit and explicit rules of networking (Fink & Kessler, 2010).
Adaptation refers to the activities firms must adopt to meet partners' special needs or the ability to adapt to new circumstances (Helfert, Ritter, & Walter, 2002). Adaptation processes include relationship-specific investments in areas such as technology, products/services, manufacturing processes, logistics, administration, employee qualification or financing (Hallén, Johanson, & Seyed-Mohgamed, 1991;Claycomb & Frankwick, 2010). Harrigan (1988) showed that partnerships are more likely to succeed when partners possess complimentary missions and resource capabilities. Compatibility in terms of resources is the key issue for performance outcomes. Therefore, coordinating and adapting the activities of a network will help to make resource compatibility a source of superior performance.
The use of constructive conflict resolution mechanisms extends the notion of coordination because these mechanisms address extraordinary, nonstandard situations, which are bound to occur in every long-term relationship (Ruekert & Walker, 1987). Interaction/network theory declares that organizations linked by cooperative interaction processes employ other noncontractual processes associated with conflict, coexistence, collusion and competition (McLoughlin & Horan, 2000). In relationships characterized by a desire to establish and maintain long-term, collaborative efforts, managers favour productive conflict resolution mechanisms because they are less volatile. Constructive conflict resolution requires a timely reaction to conflict, a readiness to compromise and a sense of justice. Constructive mechanisms contribute to a relationship, strengthen each firm's identification with the other, and increase cooperation. Firms developing long-term, collaborative relationships engage in joint problem solving because integration satisfies more fully the needs and concerns of both parties (Claycomb & Frankwick, 2010). Joint problem solving to resolve conflict leads to mutually satisfactory solutions, thereby enhancing relationship success (Mohr & Spekman, 1994).
Successful relationships tend to exhibit processes characterized by high levels of joint participation, cooperation, effective communication and productive conflict resolution. Consequently, in this paper we propose that network-driven performance is associated to the development of INMAs.

SIZE, AGE AND INMA
The development of INMA requires companies to have sufficient human and organizational resources and these resources are usually associated to firm size and age (e.g., Greiner, 1972).
Large firms are more resource rich than small and medium enterprises. Large firms may also have a longer term view towards investments, allowing them to keep operating to assess their viability, even if they are incurring losses. Institutional theory emphasizes institutional environments, which include cognitive and sociological elements, such as shared norms, standards and expectations (DiMaggio & Powell, 1991;Scott, 1995). This institutional environment is an underlying driving force behind organizational activities because of an organization's desire for legitimacy (Martinez & Dacin, 1999). Large size tends to legitimate organizations, to the extent that large size is interpreted by external Maria Ripollés and Andreu Blesa stakeholders as an outcome of an organization's prior success (Baum & Oliver, 1991). From an institutional perspective, large firms tend to attract disproportionate attention from the public. Large firms are arguably more concerned than small and medium enterprises about the downside effect on their reputation associated with the dissolution of their alliances. To maintain a favourable public image, large firms may hesitate to terminate unprofitable relationships. The dependence of small and medium enterprises' on large partners for resources and legitimacy gives the large partners bargaining power over the small and medium enterprises partners and places them in a position to influence network management. From an institutional perspective, profitability is less visible than survival because it is difficult for the public to obtain financial information. So, in terms of their public image, large firms are more concerned about network survival (Lu & Beamish, 2006). Therefore, factors from either economic or social perspectives point to increased efforts from large companies to contribute positively to network management (Lu & Beamish, 2006).
Although the literature review reiterates that networks and relationships are important for firms of all sizes because they enable firms to link activities and tie resources together (Coviello & Munro, 1995;Chetty, 2003), they seem especially important for small firms, who face many more challenging obstacles to survival and growth than larger firms, primarily owing to the constraints on their organizational resources and capacity (Luo, Zhou, & Liu, 2005). Largeness promotes insularity (March, 1981), complacency and inertia (Hannan & Freeman, 1984), and resistance to adaption (Aldrich & Auster, 1986). Small firms' greater flexibility, response speed (Katz, 1970), and tendency to constantly monitor the environment for threats and opportunities (Aldrich & Auster, 1986) usually enhances swiftness of strategy implementation and customer understanding. Small firms have also been found to make active use of interorganizational relationships to facilitate growth (Coviello & Munro, 1995) by, for example, outsourcing key marketing activities traditionally held within the organization. Coviello, Brodie, and Munro (2000) demonstrate that smaller firms are more relational than larger firms in their approach to marketing communication and primary customer contact, investment in marketing resources, and the level at which marketing activities are conducted in the firm. Therefore, smaller firms appear to place more emphasis on direct relationships with other players in a network. This behaviour added to the constraints of small companies will foster the development of coordination, adaptation and knowledge-sharing routines in their interfirm networks, whereas the independency and resource availability of large firms will discourage sharing activities that are perceived to be developed more efficiently in an independent way. Small firms' lack of power in interfirm networks will encourage them to promote conflict resolution mechanisms that improve the network atmosphere, whereas large firms will be more tempted to use the power of their size inside the network. Finally, small firms will take more advantage of network resource availability than large firms who usually have less need for those resources. Consequently, Hypothesis 2 Company size has a negative influence on INMAs.
INMA development requires professionals with experience, and also internal organizational processes to provide support. For example, a firm can only become involved in the joint development of activities to exchange information on customers if it has previously developed internal customer information management processes to facilitate the exchange of that information with other network members.
Nevertheless, another effect is also possible. Time, as signified by the age of firm, impacts on its strategy and its ability to change. Time is history and represents the specific, dated context of a firm. Boeker (1989) demonstrates that both the age of the firm and its history limit the available strategic spectrum. He also shows that firms with one specific dominant strategy are unlikely to change it, even if performance is poor. This type of analysis matches the notion of organizational inertia as identified by Hannan and Freeman (1984). Companies' reluctance to change in adulthood is likely to be a barrier to network adaptation activities. Conflict will probably arise in the relationships, making coordination among partners more difficult and, consequently, reducing knowledge-sharing routines. Inertia also Development of interfirm network management activities makes it difficult to find satisfactory ways of solving inherent conflict in networking. Young companies usually need resource availability which encourages them to find partners to cover that need. Thus, young firms will be more willing to maintain knowledge-sharing routines particularly focussed on market demands, coordinate them, adapt to their partners and establish conflict resolution mechanisms. Therefore, young firms will have a higher propensity to contribute to INMAs than mature firms.
Hypothesis 3 Company age has a negative influence on INMAs.

INDUSTRY LIFE CYCLE STAGE AND INMAs
The structure of an industry evolves continually, driven by technological, economic and competitive changes. Industry life cycle is commonly used to study industries (Levitt, 1965;Miles, Snow, & Sharfman, 1993;Grant, 2010), because it provides a criterion for classifying industries according to their stage of development. The process of choosing a classification scheme and putting industries into different categories leads to consideration of what is important in an industry and the aspects in which industries are similar and where they differ. In fact, life cycle stage may negatively affect the amount of strategic variety found in an industry (Miles, Snow, & Sharfman, 1993). Following similar research (Andersson, 2004), our study focusses on the growth and maturity stages of an industry's life cycle. When considering the effect of industry life cycle on enhancing INMAs, a central issue is that the strategic objective underlying firms' network activity is to improve their adaptation to their customers' needs. Building on past theory and research, we expect that INMAs focussed on customers' satisfaction will be important in both stages, but that their relative importance will vary according to the industry life cycle stage. In growth stages firms will motivate their INMAs in order to reduce technological uncertainty; but in mature stages businesses will focus their networking efforts on how to improve business offerings to meet new customer demands.
In growth stage periods by definition almost no dominant competitive strategy or product standards exist (Miles, Snow, & Sharfman, 1993). This period is characterized by high technological uncertainty; consequently, until a dominant technological design emerges, there are advantageous conditions for establishing informal technological networks (Pyka, 2000). In this context, the firm's networking activities do not focus mainly on customers and how to develop new offers to satisfy their needs, but on technological factors to reduce technological uncertainty.
The growth stage is characterized by accelerating market penetration as technical improvements and increased efficiency open up the mass market (Levitt, 1965). Increasing market saturation causes the onset of the maturity stage. Once saturation is reached, demand is wholly for replacement (Grant, 2010). In the later stages, market knowledge becomes critical for avoiding company decline. In this situation, interfirm knowledge sharing, resource sharing, coordination, adaptation and conflict resolution activities concentrated on consumers' needs merit special effort. In order not to fall behind one's competitors, it is important to obtain the latest market information. It is also important to gain access to sophisticated and demanding buyers (Porter, 1980). Thus, the progression of this stage will foster cooperation inside the network focussed on discovering new customer demands. As the industry advances towards its end customer focussed INMAs gain value. Decreasing sales will give rise to the need to discover and adapt to new customer demands. Therefore, Hypothesis 4 Industry effects on INMAs will be stronger in the maturity stage than in the growth stage.

METHODOLOGY
The purpose of this study is to analyse the role of INMAs in firm performance. In addition, we study the influence of firm size, age and industry life cycle stage on the development of INMAs. As such, the Maria Ripollés and Andreu Blesa current study involves a multiindustry empirical examination of firms. Data were gathered from a sample of Spanish companies operating in several industries and belonging to an interfirm network. According to Grant (2010), it is likely that an industry will be at different stages of its life cycle in different countries. Therefore, it is advisable to restrict the analysis to only one country, in order to allow comparisons between industries.
Firms were selected from 2010 Dun and Bradstreet Database. Companies had to belong to a network; understanding network as informal relationships among at least three independent companies, in such a way that all the companies have focal relationships with and know the other companies and their activities inside the network (Schoonjans, Van, Cauwenberge, & Bauwhede, 2013). In addition, firms could not be subsidiary or affiliated companies. Only independently owned and operated firms were included in our sample. This process gave a total population of 9,439 companies. The field research was carried out during the second quarter of 2010 and the final sample consisted of the 400 companies that responded to the questionnaire.
For the field research, interviewee collaboration was requested, together with confirmation of the e-mail address. After the questionnaire had been sent out, follow-up contact was made by telephone to increase the response rate. The questionnaire was posted on the internet and an e-mail with a link to it was sent to each manager. Table 1 summarizes the main characteristics of the sample.
To test for nonresponse bias, the responses of early and late respondents were compared. Analysis of the t-test showed no significant differences (p = .05 level), indicating an absence of nonresponse bias (Armstrong & Overton, 1977).

Measuring instruments
The current study relies on previous research for items to measure key constructs. Items were adapted from previous studies by changing words and sentences to enhance understanding in the Spanish context. Table 2 displays specific items used to measure the constructs and their respective factor loadings and t-values.
Industry's life cycle stage Beal and Lockamy (1999) used the following measures to identify industry life cycle stage: (1) growth in the industry's sales during the past 5 years; (2) level of demand for the industry's products; (3) stage of development of the industry's products; (4) level of diffusion of information about the industry's products; (5) plant capacity of the industry's firms over the past 5 years; (6) current price levels of the industry's products; (7) growth in the different types of distribution channels for the industry's products over the past 3 years; and (8) level of the industry's advertising expenditures over the past 3 years. Following their procedures, each author, based on individual analyses of the respondents, assigned an industry life cycle stage to each of the firms: growth or maturity. Then a value from 1 to 5

Loadings t-value
Interfirm knowledge sharing (Helfert, Ritter, & Walter, 2002 Maria Ripollés and Andreu Blesa that assessed which phase of the stage the industry was in (1 being the earliest and 5 the latest) was assigned to each company. In total, 119 firms were assigned to the growth stage and 279 firms to the maturity stage. Two firms could not be assigned owing to missing data.

Company age
Company age was measured by subtracting the year of the field work (2010) from the year of incorporation.

Company size
Company size was measured through number of employees.

INMAs
An adaptation of the scale proposed by Helfert, Ritter, and Walter (2002) was used.

Company performance
In situations where firms are hesitant to provide objective performance data, collecting subjective data provides researchers with a better ability to understand the values that a manager may place on performance (Hult et al., 2008). There is evidence to suggest that subjective and objective measures are positively associated (Shoham, 1998) and that subjective measures of performance can accurately reflect objective measures (Lumpkin & Dess, 2001). Furthermore, management assessments of a firm's performance appear to be guided more by their subjective perceptions than by objective measures (Madsen, 1989). These arguments would seem to support the adoption of subjective measures to assess international performance. Furthermore, Johnson and Kaplan (1987) outlined the limitations of economic measures and proposed that a selection of noneconomic indicators should be employed. These measures should be based on organizations' strategies, and include measures of manufacturing, marketing, research and development. Thus, to measure international performance, we adopted a subjective approach in order to improve the response rate. Globally, seven items were used to measure recent performance.

Validity and reliability of the scales
As the aim of our analysis is to describe the validity of indicators as measurement instruments of INMA and performance scales, the confirmatory initial model was adjusted following the indications of Jöreskog and Sörbom (1993). Items INMACon3 (Δ = −0.13, t = −3.96 p < .001) and INMAIks3 (Δ = 0.14, t = 2.68 p < .01) were eliminated from the scale because they did not reach a λ of 0.5. The validity analysis results show good fit indexes. Table 2 displays the list of items, their sources, their respective standardized factor loadings and t-values, and results of reliability and validity tests. The positive and significant loadings confirm convergent validity of our measures. Results also show α reliability, composite reliability and average variances extracted. In order to test the discriminant validity between the scales, the confidence interval test was used (Anderson & Gerbing, 1988). According to this test, the value '1' should not appear in the confidence interval of the correlations between the scales in the same level of analysis. Table 3 shows the results of this test, which were satisfactory in all cases.

RESULTS AND DISCUSSION
To test the conceptual model, we use a structural equation modelling approach. In order to test Hypothesis 4, the sample was divided in two parts according to whether the companies were in the growth or maturity stage of their industry life cycle. This procedure also enables Hypotheses 1-3 to be tested in two different  (Marks & Kamins, 1988), the INMA measurement scale was narrowed down averaging the items in the construct. Table 4 shows the descriptive statistics and correlations, and Figure 1 displays the results of the structural models analyses.
The results show differences according to the life cycle stage considered. Only the relation between INMA and performance has been confirmed in both samples. This finding points to the need to consider the industrial environment when analysing firms' networking decisions because the situations they face differ in mature or growing industries. As expected, the results show a positive relation between INMA and performance in both cycle stages (Δ = 0.38, t = 3.31 p < .001 for growth stage and Δ = 0.43, t = 5.98 p < .001 for maturity stage). This result supports Hypotheses 1 and underlines the importance of firms getting involved in the development of INMAs to manage their networks. This finding contributes to the literature on firm's network management (Möller & Halinen, 1999;Ritter, Wilkinson, & Johnston, 2002, 2004Ritter & Gemünden, 2003) by showing the importance for firms of developing cross-relational management tasks, which do not necessarily respond to a strategy planned by top management. Ritter et al. (Ritter, Wilkinson, & Johnston, 2002, 2004Ritter & Gemünden, 2003) recognize that in some situations it may not be possible a priori to determine the firm's network strategy, because it will emerge out of the interactions between firms in the network. This situation, however, has not been specifically contemplated by these authors. We have also confirmed the influence of developing INMAs on firm performance regardless of the life cycle in the industry in which the firm is operating. Thus, the results encourage us to propose that INMAs could be included in the firm's network management capability construct developed by Ritter et al. (Ritter, Wilkinson, & Johnston, 2002, 2004Ritter & Gemünden, 2003). The network management capability is referred to 'as the firm's capability to mobilize and coordinate the resources and activities of other actors in the network' (Möller & Halinen, 1999, p. 417). Johnston (2002, 2004) analyse the degree of network management capability through the development of relationship-specific and cross-relational management tasks. Our findings also encourage us to consider INMAs when analysing a firm's network management capability because they focus on nondeliberate

JOURNAL OF MANAGEMENT & ORGANIZATION
aspects of network management. The importance of our proposal is justified by the influence of INMAs on the performance of firms in the network. Furthermore, it could also be thought that in the cases of firms that can define a priori their network strategies, the development of INMAs would aid the introduction of these strategies when firms involved in a network have different goals. A firm's network management ability can only be understood in an ongoing, firm-wide process (Ritter, Wilkinson, & Johnston, 2004). Consequently, and based on seminal Mintzberg's studies, we argue that the firm's real network strategy will be the outcome of deliberate, intentional or rational cross-relational management tasks and of the result of developing INMAs to align its deliberate network strategy with the rest of network members (Mintzberg & Quinn, 1991). Furthermore, the development of INMAs can be viewed as a network-specific competence that varies among networks and can be an important source of competitive advantage for the network as a whole and for each firm in the network.
The findings in this work are also in line with those reported by Prashantham and Young (2011). These authors point out the importance of tie strength in the processes of assimilating and exploiting new knowledge. Our results, however, also show that stronger ties influence the processes of acquiring  and transforming new knowledge. In fact, as argued in this work, the development of INMAs helps to improve firms' information bases and facilitates their transformation. In contrast, Prashantham and Young (2011) indicate that in these stages of developing new knowledge weak ties would be more influential. In short, our findings appear to indicate that it is the development of INMAs that influences firms' absorption capacity, understood as a firm's capacity to uptake and integrate new external knowledge (Zahra & George, 2002), rather than tie strength. However, we want to emphasize that our results only point in this direction, because in this work we have not tested the relationship between the development of INMAs and a firm's absorption capacity. Our Hypothesis 2 proposed the existence of a negative effect of company size on INMAs. The results of the analyses show that the relationship has been confirmed in the firms classified as being in the growth stage (Δ = −0.21, t = −2.03 p < .05) but not in the case of the companies in the maturity stage (Δ = 0.14, t = 1.08). According to the results in growing industries, bigger companies discourage the development of INMAs. When the market is growing, big companies rely on the advantages of their size to make the most of good market conditions, promoting insularity and resisting interaction to other network members. This result is consistent with a large part of the literature on networks, which demonstrates their importance in bridging information gaps (Slotte-Kock & Coviello, 2009;Freeman, Hutchings, Lazaris, & Zyngier, 2010), in providing small-and medium-sized firms with market and technology knowledge (Slotte-Kock & Coviello, 2009;De Clercq, Sapienza, Yavuz, & Zhou, 2012), and in facilitating these firms' growth (Hite & Hesterly, 2001). Our results, however, do not allow us to confirm the same pattern of behaviour in the case of firms in mature industries.
Hypothesis 3 suggests the existence of a negative effect of company age on INMAs. Our results do not show that age can facilitate INMAs in any of the stages. Consequently, Hypothesis 3 cannot be confirmed. Contrary to what is commonly accepted in networking literature, the development of INMAs does not appear to need the support of internal organizational processes. Companies' reluctance to change in adulthood either hinders or fosters network adaptation activities. Young companies' need for resources does not appear to encourage them to participate in INMAs more than mature firms. This result, in line with the findings in Hite and Hesterly (2001), shows the importance of informal networks regardless of firm age. However, Hite and Hesterly (2001) point out that firm age does influence the structural characteristics of networks, whereas our study indicates that firm age cannot be considered an antecedent of the development of INMAs.
The results of our study confirm the influence of the development of INMAs on the performance of firms in a network regardless of the life cycle of the industry where they operate. Our results are not so conclusive, however, for the analysis of whether industry life cycle can be considered a contingent variable that influences which firms become involved in developing INMAs. Thus, in the growth stage, the relation between position in that stage and INMA shows a negative and significant effect, thereby indicating that this stage in the industry life cycle has a negative influence on INMAs. The results seem to suggest that when the market is growing, companies focus on obtaining the advantages of the stage. In the growing stages, individuals in firms are relevant resources and their interpretation of the environment is important (Maignan & Lukas, 1997). The search for efficiency in production and distribution is the determinant that guides networking in those industries. In the maturity stage, however, there is no significant relation with INMA. Consequently, Hypothesis 4 is only partially confirmed. This result could be owing to the fact that firms in a mature industry do not introduce new resources into the market.

CONCLUSIONS
Evidence shows that business networks generate valuable benefits. INMA constitutes an additional objective for firms involved in business relationships, as a way of obtaining benefits in the shape of high levels of performance. In addition, INMA goes further than the leader company in the network and Development of interfirm network management activities involves the participation of all members in the activities needed to develop network management and obtain its benefits.
Most previous research has focussed on industry, age and size as control variables that should not be influencing the effect of the variables studied. Our research has considered these variables as factors that directly foster or inhibit the development of INMA and so, the study indicates the need for future networking studies to consider the influence of these variables in their hypotheses.
According to our results, added benefits of networking could be obtained by adopting comanagement activities. It is to be expected that developing INMA helps companies to extend their customer knowledge base. In a general sense, INMA development requires partners in a network to know each other's capabilities and share the same vision of the collaboration process as they work towards the common goal. Companies in networks not only share new market and customer information, but also share procedures that may help them to integrate the new knowledge in their knowledge base and exploit it. Therefore, a future line of research would be to explore the implications of INMA development for companies' absorption ability to further our understanding of the importance of networks in companies' success.
This study shows the importance of INMAs in a firm's network capability, because they are related to the problems firms have to face when rolling out their network strategy. However, the development of INMAs is also a network-specific capability, which varies among networks. From our research, we can conclude that the development of INMAs can be an important source of competitive advantage for the network as a whole. This capability has received relatively scant programmatic attention within network theory; therefore we suggest that future research in this area is needed to better understand the value of different types of networks. Different kinds of networks are typically assumed to function differently and have different capacities for extracting resources (Hite & Hesterly, 2001;Lechner, Dowling, & Welpe, 2006). Hite and Hesterly (2001) have distinguished between identity-based networks and calculative networks. Identity-based networks involve some type of personal identification with the other actor that motivates or influences economic actions. Calculative networks are primarily motivated by expected economic benefits. Different types of calculative networks can also be identified by the economic goals. Lechner, Dowling, and Welpe (2006) suggest that reputational networks, coopetition networks, marketing networks and technological networks are most important types of calculative networks.
From a managerial point of view, considering the structural factors analysed in our research could help firms to be aware of the forces that are driving their decisions and review the behaviours that, as in the case of being in the industry life cycle growth stage, are moving the company away from capabilities that could provide them with superior performance.
The results seem to suggest that research should consider not only the differences between particular industries, but also the differences between industry life cycle stages. This opens a new opportunity for generalizing results. In order to control for industry effects, most studies focus on a few industries, limiting the generality of their results. Our results show that differences between industries can also be observed at the life cycle stage, providing a higher degree of generalization because samples can be constituted by individuals from several industries. This approach also facilitates sampling. Although it might be difficult to obtain large enough samples from only one industry it seems easier to obtain answers from more respondents based on a life cycle.
These conclusions should be considered in the light of some limitations related to the method followed in our research. The sample for testing the hypotheses proceeded from a sample of Spanish companies thus, cultural and environmental factors affecting the activities inside the networks cannot be ruled out. Furthermore, although we received 400 responses to our questionnaire, the response rate was only 4.2%, and therefore insufficient to generalize the results to the population. Consequently, additional research in other countries and with representative samples would be helpful in order to generalize the results. We adopted a global perspective of looking at the network asking interviewees to refer their answers to the main network to which they belonged. Consulting only one member from a network for information on its activities could bias the data because they came from only one perspective of the situation. Nevertheless, as management activities are a shared behaviour for all members of the network, major differences in answers from respondents in the same network are not expected. A way of improving the collection of data from a network would be to identify all its members and interview all the agents involved in the relationships.
Although structural equation models allow testing of direct causal relations in a nonexperimental situation, there is still the problem of when an activity is implemented and when it is measured. Further research using longitudinal data is needed in order to test if the relationships established in this study have been affected by the cross-sectional design.