Structural antecedents of institutional entrepreneurship in industrial networks: A critical realist explanation

Industrial Marketing Management 42 (2013) 405–420

Contents lists available at SciVerse ScienceDirect

Industrial Marketing Management

Structural antecedents of institutional entrepreneurship in industrial networks: A critical realist explanation Paul Matthyssens a, b, 1, Koen Vandenbempt a, b, 2, Wouter Van Bockhaven a,⁎ a b

University of Antwerp, Prinsstraat 13, B-2000 Antwerp, Belgium Antwerp Management School, Sint Jacobsmarkt 9-13, B-2000 Antwerp, Belgium

a r t i c l e

i n f o

Article history: Received 24 January 2012 Received in revised form 6 November 2012 Accepted 6 November 2012 Available online 1 March 2013 Keywords: Critical realism Institutional entrepreneurship IMP Change in business networks Case-study research

a b s t r a c t In response to the ongoing debate in industrial network (IMP) literature regarding the manageability of networks, the present paper explores how a critical realist epistemology can facilitate a more multilayered explanation of collective change. In line with recent literature, we combine IMP with neo-institutional theory and identify the cognitive and normative boundary conditions for unmanaged change in networks. As such, we make a mid-range contribution to IMP by delineating structural antecedents of its interface with institutional entrepreneurship in explaining collective change in networks. The mechanisms underlying the transition from emergent to intentional change are illustrated with a case study describing difficulties experienced by Dutch steel wholesalers in realigning their business strategy with market conditions even when it is necessitated by upstream and downstream pressures. This paper is one of few examples of critical realist case-study research to guide scholars on how to translate this epistemological orientation into methodological choices. © 2013 Elsevier Inc. All rights reserved.

1. Introduction Over the years marketing scholars have searched for a unifying theoretical basis for their discipline. The 1983 Journal of Marketing special issue on “Marketing theory”, for instance, explicitly emphasized alternative perspectives on marketing. Since Bartels (1968) seminal call for more consideration of behavioral and sociological elements in a ‘general theory of marketing’, other marketing theorists as well have criticized marketing's “over-reliance on micro-economics” (Arndt, 1985; Cunningham & Sheth, 1983). From the ensuing anti-positivist disputes, realism arguably arose triumphant, especially in industrial marketing (Easton, 2002). There, the industrial networks' tradition (IMP3) explicitly contradicts the ontological assumptions of mainstream traditions rooted in micro-economics (Snehota, 2003; Welch, 2001). Although proficient in articulating mid-range explanations, the field has lacked an integrative philosophy. This study takes a critical realist perspective to business networks (Morais, 2003; Ryan, Tahtinen, Vanharanta, & Mainela, 2009) to uncover the contingent nature of their evolution. It illustrates how this ‘epistemological orientation’ (Easton, 2002) supports a generalizable theoretical mid-range contribution from case-studies. Critical realism ⁎ Corresponding author at: University of Antwerp, Department of Management, Belgium. Tel.: +32 32 75 50 17; fax: +32 32 75 50 79. E-mail address: [email protected] (W. Van Bockhaven). 1 [email protected] 2 [email protected] 3 Acronym for the Industrial Marketing and Purchasing group, which is interchangeable with the markets-as-networks tradition or the industrial networks' perspective (McLoughlin & Horan, 2002). 0019-8501/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.indmarman.2013.02.008

(CR) assumes a reality beyond and independent of our knowledge and that observable events are manifestations of invisible generative mechanisms (Bhaskar, 1978). It upholds a complex world view, wherein countless ‘real’ objects act interdependently with each other through heterogeneous causal and contingent relationships, under various boundary conditions (Sayer, 1992). The complexity of these interdependencies in social reality's open system makes outcomes highly variable, context-specific and never entirely predictable given that many of them remain unknown (Archer, 1995). IMP thinking, seeing industrial markets as complex networks (McLoughlin & Horan, 2002; Ritter, Wilkinson, & Johnston, 2004), fits with a CR orientation. Emphasizing interaction, social exchange and the indeterminacy of networks, IMP considers marketing interactions as emergent, unpredictable and less controllable (Ford, Gadde, Håkansson, & Snehota, 2011). Recently, however, the assumption of emergence has become the subject of a lively debate in IMP (Golfetto, Salle, Borghini, & Rinallo, 2007; Rampersad, Quester, & Troshani, 2010), with new studies arguing for the possibility of intentional networks (Möller & Rajala, 2007). IMP researchers are increasingly ‘infusing’ institutional entrepreneurship in studies describing how networks go from the traditional ‘emergent’ state (Håkansson, 1992) into an intentional ‘mode’ (Möller et al., 2005). Ritvala and Salmi (2008), for instance, describe how campaigns for institutional change result in the emergence of new interaction rules and collaborative platforms. These subsequently drive changes to the institutional structure (Ritvala & Salmi, 2010). An institutional lens might thus further our understanding of change dynamics in networks by illuminating the ‘structural’ mechanisms (Brito, 2001). However, these studies describe institutional entrepreneurship for

406

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

non-profit issues. As most industrial networks get by just fine without institutional entrepreneurship, its relevance for industrial marketing is not indubitably established. For a critical realist, separately acknowledging agency and structure (Archer, 1995), it would be interesting to examine these structural powers and conditions causing a network to adapt by intentional rather than distributed action. Little research has investigated the contingent boundary conditions of the traditional IMP explanations (Ford et al., 2011; Lundgren, 1992) of collective change in networks. Hence, we aim to address the following research question: under which boundary conditions does collective change4in networks require an intentional ‘mode’ and thus institutional entrepreneurship? By exploring the boundary conditions shaping the overly embedded network context where adaptation is congested and institutional entrepreneurship (Dimaggio, 1988) becomes warranted, we delineate and extend IMP theory. We contribute to collective change literature by investigating antecedents to institutional entrepreneurship in the complex context of business networks. We thereby focus on the cognitive and normative rather than the well-established regulative institutional contingencies (Vermeulen, Büch, & Greenwood, 2007; Welch, 2001) and consequently indicate which forces of persistence within networks should be counteracted by institutional entrepreneurship. Such a mid-range contribution would also have substantial practical relevance, as e.g. globalization and sustainability motivate firms to respond with innovation collectively (Greer & Lei, 2012). When competitive rivalry in business markets becomes a war between networks (Möller & Rajala, 2007), it is useful to know whether one's network will respond organically or needs institutional entrepreneurship. The paper continues by first overviewing the hierarchic relationships between critical realism, industrial network theory and neoinstitutionalism and evaluating if they should be combined. It then reports the methodology and explicitly relates it to suggestions from CR epistemology, as opposed to alternatives. In line with this specific CR-inspired research structure, the case is then presented by describing the problem motivating the transition to institutional entrepreneurship. In this extreme case, Dutch steel wholesale companies (SWC) feel an obvious need to change collectively, but are not able to do so without institutional entrepreneurship. A subsequent analysis section probes for different sets of causal mechanisms which lead to the failure of unmanaged collective change. The explanation section then abstracts the generalizable causal configuration at work in this case into theoretical propositions of boundary conditions.

2. Theoretical foundations 2.1. Critical realism as a metatheory CR as a philosophy of science stratifies reality into three overlapping domains (Bhaskar, 1978). The real domain consists of intransitive objects or ideas (Ryan, Tähtinen, Vanharanta, & Mainela, 2012), which have the generative power to cause certain effects. The actual domain consists of the objective events that manifest due to these different causal powers, their contingencies and their interactions (Sayer, 1992). So the presence of a real object might manifest in different ways or not at all, depending on the complex configuration of causal powers that act in open systems like business networks. Third, the

4 Construed here as strategic change, a difference over time in the alignment of a system with its external environment (Rajagopalan & Spreitzer, 1997). Such change would encompass reactive adaptations and more proactive initiatives to shape or anticipate the market or institutional environment and can be either continuous or discontinuous, depending on the extent of misfit with the environment (Meyer et al., 1990).

empirical domain consists of the subjective experiences of these events and ideas. CR is not a general theory explaining all problems in industrial marketing. Rather, it is a metatheory, an ‘under-labourer’ (Bhaskar, 1978) that offers indications of how the world is to be seen (ontology). It is an ‘epistemological orientation’ (Easton, 2000) and therefore has implications as to how theory about reality can be developed (epistemology). By itself, however, CR does not explain any of the problems that marketers are actually confronted with (Easton, 2000; Morais, 2003). Much like a computer language is a necessary precondition that determines the possibilities and limits of software applications to actually solve a problem, so is a meta-theory a necessary but insufficient condition for the scientific causal explanation of events. Just like specific software programs are required to provide solutions for practical problems, a metatheory requires specific theories as ‘flashlights’ (Ferraro, Pfeffer, & Sutton, 2009) to explain actual events. CR puts ontology first (Lawson, 2009), seeking accurate causal explanation that illuminates structure and agency elements separately (Archer, 1995). Therefore, CR would promote the ‘metatriangulation’ (Lewis & Grimes, 1999) of different yet complementary ‘regular’ theories which amply expose one dimension while acknowledging the other to avoid atomism. To add new insights into the manageability dilemma, this paper combines the industrial networks and institutional entrepreneurship perspectives following recent papers on collective change in business networks (Ritvala & Salmi, 2010, 2011). Although the main selection criterion for these specific theories was their fit with the research problem, they are also remarkably suitable for use in a CR study. Arisen from extensive empirical insights refuting mainstream assumptions (Håkansson, 1982; Johanson & Mattson, 1987), IMP's depiction of business relationships is realistic and open-ended (McLoughlin & Horan, 2002). Institutionalism likewise adheres to CR's preference for ontological fit by explaining strategy in terms of e.g. power, symbols and self-interest (Scott, 2008), thus accounting for the rationally suboptimal organizational arrangements found outside classrooms (Ferraro et al., 2009; Hodgkinson & Rousseau, 2009). Consistent with CR's epistemological preference for stratified causal explanation rather than prediction, IMP (Akgun, Lynn, & Yilmaz, 2006) and institutional (Iederan, Curseu, Vermeulen, & Geurts, 2011) researchers are interested in the cognitive and social mediators of business processes. From a CR perspective, combining two lenses to increase explanatory power makes sense, as long as their incommensurability is refuted (Miller & Tsang, 2010). In the following section, they are hence analyzed for their complementarity, implying (1) compatibility of assumptions (Geels, 2010) and (2) synergy in their combination (Arora & Gambardella, 1990). 2.2. The compatibility of institutionalism and the industrial networks' perspective The industrial network approach and institutional theory implicitly share some fundamental assumptions such as the salience of interdependence (Welch, 2001) and the need for mutual benefit in inherently open-ended business relationships (Möller & Halinen, 1999; Powell, 1990). Snehota (2003, p.2) argues that: “the market-asnetwork perspective is clearly siding with the perspective on markets as an institution”. However, to combine them in a CR analysis, in-depth assessment of their complementarity is recommended (cf. supra). This section analyzes the compatibility of their assumptions (Table 1). Table 1 also compares the explanation of change in both theories, which enables evaluating their synergy. This is the second condition for complementarity and is further discussed in Section 2.3.3. To compare both theories' assumptions, one can generically divide them along a voluntarism–determinism and a micro–macro dimension (Astley & Van de Ven, 1983). Neo-institutionalism and IMP could be considered voluntaristic on the first and relativist towards

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

407

Table 1 Compatibility of IMP and institutional theory. IMP Epistemological level of explanation Agency (voluntarism)

Rationality

Atomism Opportunism Actor homogeneity

Role of change

Change process

Neo-institutionalism

Assumptions Relative individualism: viewed from focal organizations & relationships. Delimited by outcomes of distributed agency and contextual influences. Motivated by expected rewards & interactions. Bounded by relational norms & incomplete network pictures; includes psychosocial & evaluative considerations. Actors, Resources and Activities (ARA) are interdependent in networks. Self-interest & calculative behavior, but trust leads to persistent relationships. Functionally specialized heterogeneous network actors.

Codea

Relative holism: viewed from organizational field.

R

Delimited by collective norms and habits. Motivated by identities, shared meanings & values.

R

Bounded by cognitive & normative meaning systems; includes political & social values.

C

Organizational fields conform to shared norms & are mimetically isomorphic. Actors can also influence their field. Self-interest, but symbols, myths & taken-for-granted norms guide behavior. Functional roles, identities & positions determined by legitimacy & embeddedness.

R

Explanation of change Regaining fit with actor interests or demands on a higher Organic adaptation that keeps the ‘living’ contextual level to maintain systemic health in evolutionary network system healthy in an evolutionary cycles. process. Evolutionary, emergent, mutual adaptation, Revolutionary; infrequent phases of replacement of the continuous. institutional template when institutional misfit reaches a critical level, intermitting with phases of isomorphism.

R R

C

I

Based on Baraldi et al. (2007), Brennan (2006), Dimaggio (1988), DiMaggio & Powell (1983), Ford et al. (2011), Lawson (2009), Mattson (1987), Oliver (1992), Scott (2008) and Welch (2001). a C = convergent (shared explanatory dimensions); R = reconcilable (shared outcome, yet different explanatory factors); I = incommensurable (mutually exclusive).

the other. 5 They are non-deterministic by accepting the possibility of adaptation, but contextualize it in network stabilizers (Sutton-Brady, 2008) or field processes. From the micro–macro level of explanation, neo-institutionalism and IMP are respectively holist and individualist. IMP centers on interfirm exchange relationships to explain the resulting network dynamics (Ford et al., 2011). Institutionalism on the other hand, describes the collective dynamics of change in organizational fields – the entire set of horizontal nets operating towards a market – and their context (Oliver, 1992; Scott, 2008). Nonetheless, they explicitly acknowledge the other level as well. In IMP, the distributed actions of individual actors and dyads will be distorted by chained reactions and eventually holistic network level processes, (Azimont & Araujo, 2007). Likewise, individual agency is embedded in neo-institutional concepts of power, self-interest and institutional entrepreneurship (Scott, 2008). Moreover, institutionalists focusing on processes of embeddedness (Baum & Oliver, 1992; DiMaggio & Powell, 1983), view the lines between context and content as explicitly permeable and in IMP they are intertwined at the network level (Baraldi, Brennan, Harrison, Tunisini, & Zolkiewski, 2007). Both theories hence acknowledge the interplay between different levels of analysis. As neo-institutionalism and IMP are reactions against mainstream economics, one might further compare their compatibility by considering their positions on the latter's assumptions. As can be seen in Table 1, both theories are convergent in largely rejecting rationality. They extend it with risk avoidance to cope with actors' rationality bounds (Loasby, 2000) and uncertainty. They do so by either creating institutions (rules, conventions) and relying on conventional wisdom (Kondra & Hinings, 1998) or embedding one's ‘incomplete’ organization in networks (Ford et al., 2011). Their explanations overlap at least on the cognitive and social dimensions.

For most assumptions, however, they are reconcilable. The initial IMP studies' findings emphasize the pervasiveness of trust, continuity and interdependence between market actors (Baraldi et al., 2007; Snehota, 2003). Institutional theory builds on the social structure of organizations and markets, such as the normative codes and takenfor-granted beliefs that lead to conformism in organizational fields (Scott, 2008; Welch, 2001). As such, both perspectives reject classic and institutional economics' assumptions of pure opportunism and atomistic maximization behavior (DiMaggio & Powell, 1983; Gadde, Huemer, & Håkansson, 2003; Uzzi, 1997). These rejections are, however, explained through different factors. The concepts of organizational fields (DiMaggio & Powell, 1983) and business networks (Halinen, Salmi, & Havila, 1999) are furthermore explicitly inclusive in considering stakeholders beyond the traditional factor or competition markets. As such, both theories differentiate between types of actors and by adding these additional types of interdependencies, they further stress the inappropriateness of atomistic assumptions. In sum, when comparing the assumptions of IMP and institutionalism, our analysis did not yield incommensurable assumptions. In fact, some similarities are found due to their shared rejection of the unrealistic assumptions of mainstream economics. Nonetheless, there are substantial differences in the explanations of both theories. Their respective individualist and holist lenses and the different mechanisms they focus on will, for instance, lead them to explain network position (Brito, 2001; Welch, 2001) or the role of meaning structures (Greenwood & Suddaby, 2006; Mouzas, Henneberg, & Naudé, 2008) differently. In what follows, we discuss how their explanations of change, despite having a compatible evolutionary underpinning, describe a rather mutually exclusive change process. It is argued that this allows their combination to explain more actual situations and thus raises their explanatory value. 2.3. The explanation of change in IMP and institutional theory

5

Whereas old institutionalism is macro-deterministic, neo-institutional theory pertains to the macro-voluntary quadrant, while acknowledging the potential for individual agency (institutional entrepreneurship).

To assess whether there is a synergy between IMP and institutionalism, one needs to establish whether combining one with the other raises its marginal explanatory value (Arora & Gambardella, 1990). To this end, the basis for each theory's explanation of change is briefly

408

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

presented. Then, it is argued here that these illuminate nonoverlapping individual and holistic facets of change processes in networks, thereby compensating each other's blind spots. It is furthermore asserted that (only) their combination helps overcome major fallacies in competing explanations, thus improving explanatory value. Both perspectives share a similar evolutionary outlook on the role of change (Table 1; Brennan, 2006). Neo-institutionalism and IMP consider it to be involved with maintaining systemic health by decreasing institutional misfit (Kondra & Hinings, 1998; Söllner, 2002) or keeping the ‘living’ network intact (Brito, 2001). Change in institutionalism originates from actors' dissatisfaction or contextual inducements (Greenwood & Hinings, 1996), much like it originates either from struggles inside the network or from evolutionary forces outside in IMP (Håkansson, 1992). In both, the change will have effects on individual behavior as well as on the overarching structures. The following explanations of change according to each theory, however, will illustrate that the processes described are rather mutually exclusive. 2.3.1. Industrial networks' explanation of change In contrast to mainstream management literature, where change is commonly treated either as individual agency or as exogenous occurrences (Meyer, Brooks, & Goes, 1990), IMP stands out by illuminating the microfoundations of collective change. IMP provides an explanation of how adaptive change is distributed in markets (Snehota, 2003), a black box in classic economics. It illuminates how actor bonds, resource ties and activity links in networks are conducive of mutual adaptation to changes (Håkansson & Ford, 2002; Macbeth, 2002; Mattson, 1987). The working of these dynamics has been referred to as that of a ‘living’ structure (Brito, 2001; Ritter et al., 2004), in which actions initiated by one actor via interaction ripple through the network and cause suited reactions by connected actors. The exact pattern of this ‘rippling’ effect is unpredictable and the consequence of the distributed calculative actions of individual yet interdependent firms (Azimont & Araujo, 2007). Hence the strategy process is rather emergent (Baraldi et al., 2007). However, in response to a common discontinuous threat, network actors might also coordinate more intensely (Lundgren, 1992). A more recent stream of literature contests the traditional view of change as emergent adaptation and argues that networks can be governed intentionally (Heikkinen, Mainela, Still, & Tahtinen, 2007; Möller et al., 2005; Rampersad et al., 2010). However, as emergence, indeterminacy and distributed agency are deeply engrained in the IMP tradition (Ford et al., 2011), the latter seems an unlikely candidate to explain ‘network management’ by itself. The recently espoused combination with institutional entrepreneurship (Ritvala & Salmi, 2010, 2011) might offer an alternative and it might hence be useful to identify the boundary conditions in which it applies. 2.3.2. Institutional theory on change in networks Institutionalism traditionally explains organizations as striving for legitimacy among their peers (Zimmerman & Zeitz, 2002). With neo-institutionalism's focus on cognitive and social processes (Welch, 2001), firms hence strive for cultural and structural isomorphism and adherence to shared norms and reciprocal expectations (DiMaggio & Powell, 1983; Meyer & Rowan, 1977). Cognitive and normative institutionalization mechanisms maintain the status quo in networks and can hence be a force of inertia (Seo & Creed, 2002; Uzzi, 1997). Change is consequently less frequent, yet more revolutionary in pace and scope (Greenwood & Hinings, 1996). However, embeddedness through a central network position can be a facilitator of change (Greenwood & Suddaby, 2006; Polidoro, Ahuja, & Mitchell, 2011). Lack thereof might constrain non-central actors' attempts to press their interests into a new institutional arrangement. When evolving to ‘neo’-institutionalism (Scott, 2008), the institutional explanation of change became more open to proactive agency,

rather than contextual inducements (Greenwood & Hinings, 1996). Change in neo-institutionalism is driven by institutional misfit (Kondra & Hinings, 1998), the misalignment of an institutional arrangement with its context or the interests of actors in it (Greenwood & Hinings, 1996). As such, concepts like power, values and expectations are also recognized as drivers in the evolution of organizational fields. This recently led to the development of the institutional entrepreneurship concept. Institutional entrepreneurship presents change as a consequence of the interest in a new institutional template (Maguire, Hardy, & Lawrence, 2004) by (collective) “organized actors with sufficient resources” (Dimaggio, 1988, p.14). It thus emphasizes a proactive agency element which is rooted in the empirical domain instead of the induced, ‘actual’ aspects of change from traditional institutional analysis. Yet, it can still be shaped by the extant institutional template (Leca & Naccache, 2006) or induced by outside pressures (Greenwood & Suddaby, 2006). In fact, it explicitly deals with the challenge for actors to fundamentally reshape the context in which they are embedded (Seo & Creed, 2002). As such, it makes a contribution towards an analysis of change that is mindful of the distinct influences of agency and structure as forces of change and stability.

2.3.3. Synergies between IMP and institutional theory IMP and institutionalism recognize influences from the other's level of analysis (cf. supra), yet those are somewhat black-boxed. By these traditions' scholars' own admission, institutionalism is illequipped to explain individual agency (Greenwood & Hinings, 1996; Oliver, 1992) and IMP labels the holistic processes in networks ‘indeterminate’ distributed agency (Azimont & Araujo, 2007; Ford et al., 2011; Macbeth, 2002). By focusing on actor ties, resource bonds and activity links as the microfoundations of change diffusion in networks (Håkansson, 1992), IMP compensates institutionalism's agency blind spot. In neoinstitutionalism, individual agency phases intermit rather sequentially with structural processes of institutionalization and deinstitutionalization (Oliver, 1992). By detailing the cognitive, normative and regulative mechanisms (Scott, 2008) driving organizational fields as such, this tradition might help IMP researchers make sense of collective processes in networks. By employing a holist and an individualist lens, these theories combined enable analytical duality, and thus avoid the constructivist fallacy of conflating agency and structure (Archer, 1995; Ryan et al., 2012). According to CR, agency and structure influence each other but do so sequentially and on different levels (Archer, 1995). Theoretical explanations should thus be able to separate them analytically (Leca & Naccache, 2006; Mutch, 2007). Agency is separately recognized in IMP and neo-institutionalism, yet bounded by social constructions (Welch, 2001) and by evolutionary cycles (Brennan, 2006). There are hence clear synergies in combining the concept of institutional entrepreneurship with the industrial networks perspective to explain collective change in networks. Delineating its boundary conditions, we argue, will furthermore also advance the manageability debate in IMP.

3. Methods This paper's methodology is reported following the four main interdependent and overlapping tasks of CR case research: design, investigation, analysis and explanation (Ryan et al., 2012). Throughout these 4 tasks, the methodological choices made are elaborately justified from a CR perspective and special attention is paid as to how they are specific for CR designs as opposed to alternatives. These epistemological differences are summarized in Appendix A.

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

3.1. Design As CR emphasizes causal explanation, the epistemological and methodological strategies are dependent on the ontological issue addressed. Hence, CR researchers are free to select the design and methods that best fit their particular research question and the reality they try to explain. To answer our research question, we draw on the findings of an industry case-study on value creation and differentiation efforts in the Dutch steel chain. Observing that firms did not arrive at direly needed change autonomously, the industry association became interested to learn why these firms jointly and individually failed at reorienting their business model. A single industry case study fits our intention of illustrating a theoretically peculiar extreme situation (Boisot & McKelvey, 2011; Siggelkow, 2007) and making an analytical generalization based on it (Yin, 1984). Tightly scoped (Eisenhardt & Graebner, 2007), abductive (Dubois & Gadde, 2002) case research fits our aim to extend theory and explore yet unclear underlying causal explanations (Shadish, Cook, & Campbell, 2002). CR case research is to some extent congruous with the moderate constructivist method (Järvensivu & Törnroos, 2010). Hence, the methods we used are somewhat similar in the investigation phase, yet the finality of CR research is of a more analytical nature, which is apparent in the following phases. CR studies differ from constructivist ones in that they are transcendental (Bhaskar, 1978), aiming for analytical generalization (Easton, 2000). CR design hence builds on triangulation and abstraction, which are further explained.

3.2. Investigation Concerning the investigation task, we present this study's sought after data types and approach to data collection, framed in a CR perspective on case-study research. CR research shares the concern for contextualized social and cognitive processes of interpretive designs (Matthyssens & Vandenbempt, 2003) as these processes construct the empirical reality experienced by respondents and researchers (Bhaskar, 1978). CR research is equally inclusive of the contextual detail, complexity and richness that characterize interpretive designs (Langley, 1999), so as to portray a comprehensive and accurate picture of the event and its context. As such, we also provide a thick description and employ an abductive design systematically iterating between data, context and theoretical frameworks to capture as many of the different layers of data as possible. As such, interview protocols were based on an analysis of secondary reports and were enriched and benchmarked for face validity in focus groups with expert managers. Still consistent with the aim of descriptive accuracy, CR research allows a combination of such eclectic data (Langley, 1999), combining variables, events, and interpretations, to increase explanatory power. However, CR researchers intentionally combine these different types of data for triangulation purposes. Triangulation implies the interrelating of different respondents, data sources and theories to evolve towards a better fitting explanation of the phenomena under study. During the investigation, this means that empirical insights are continuously contrasted with each other, as all observations are by necessity theory-laden (Sayer, 1992). ‘Horizontal’ triangulation, contrasting rather than describing different subjective experiences, is not systematically performed by constructivist case researchers (Easton, 2000). We attempted to achieve horizontal triangulation by adopting an embedded design (Yin, 1984), allowing triangulation through multiple perspectives and by systematically combining (Dubois & Gadde, 2002) desk research, focus groups and interviews with theory in three iterative cycles (Fig. 1). Moreover, CR qualifies that these different types of eclectic data respectively reflect the real, the actual and the empirical level and distinguishes between objective and subjective data. This refers

409

to the further analytical challenge of ‘vertical’ triangulation, which is discussed in the next section. The sequential, phased approach (Fig. 1) enabled gradually deepening out previous findings. Data in the first step generally consisted of industry-specific (academic, consulting, financial and industry association) reports. These were complemented with a quantitative model to identify the main drivers of steel demand in stage 2 and with an analysis of wholesalers' websites in stage 3. As CR researchers should analyze the commonalities and differences in respondents' interpretations, focus groups were employed to distinguish what Calder (1977) labels intersubjective (shared) from the intrasubjective (individual, diverging) interpretations. Focus groups always started exploratory (intrasubjective) to extend desk research, then progressed to an experiential phase (intersubjective) to benchmark how these elements should be interpreted from a steel wholesale company (SWC) perspective. They ended clinically (intrasubjective) to elicit unconscious assumptions (Fern, 2001). In the first two rounds, mini-focus groups (Fern, 2001) were employed with 5 to 6 participants, allowing more in-depth discussion (Fern, 1982). Participants were top management from the main SWC segments who served in one of the industry association's boards or committees, as well as the chairman of a large downstream industry association. They were recognized as industry captains well beyond their industry by supply chain and other business partners (complementary material suppliers, knowledge institutions, consultants, …). A larger design, with 12 SWC managers, was adopted in round 3 to provide a rich benchmark on SWCs' strategic inferences upon confrontation with results from the market analysis and interviews from the previous phases. Focus groups were always moderated by 2 of the authors, so that one could facilitate smooth progression and the other could monitor coverage of all relevant issues (Morgan & Krueger, 1998). Issues of observer dependency were minimized because most participants had prior contacts with at least one of the moderators in previous studies and consulting work. Although most information coming from focus groups is at best pre-scientific (Calder, 1977), they were instrumental in immersing the researchers into the context, acquiring cooperation from industry experts and contrasting our own interpretations with theirs. They also served to integrate relevant trends and benchmark assumptions into the interview guide and the volume forecast model. This cooperation from industry experts provided access to managers throughout the supply chain for 36 rich and detailed in-depth interviews (respondents in Appendix B, interview guides in Appendix C), lasting between 1 and 2.5 h, conducted to a point of data saturation (Eisenhardt, 1989). The challenge for CR case research in the investigation phase lies in fully exploring these rich data while maintaining an analytical focus on the objects studied. To support such balance between open-endedness and scope, interviews were semi-structured. Moreover, summaries were made of each interview on the same or the following day based on interview recordings and field notes, immediately extracting themes and contrasting with previous findings, so as to feed following interviews. This first-order interpretation even started during the interviews, when one of both interviewers made elaborate field notes of 3 to 7 pages, schematically representing responses as well as indicating related themes and concepts in the margin. Interviews focused on identifying the need for strategic change and the causes of inaction in stages 1 and 2 and on drivers of scarce successful collective change in stage 3. 3.3. Analysis CR research aspires to theorize from field data to the identification of causal mechanisms through analytical generalization (Easton, 2000; Ryan et al., 2012). It acknowledges that said mechanisms are out of phase with actual events and that the latter are likewise out

410

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

Fig. 1. Research stages and steps.

of phase with ‘empirical’ knowledge thereof. Besides ‘horizontal’ triangulation of subjective experiences, this additionally requires what Harrisson and Easton (2004) refer to as ‘peeling the onion’. In line with the layers needing to be removed, this results in two tasks: (1) ‘vertical triangulation’ of events and interpretations to distinguish subjective from objective and (2) abstraction (Lawson, 2009) to deconstruct holistic ‘event’ data into their constitutional ‘real’ mechanisms, liabilities and conditions (Sayer, 1992). 1 Peeling off the empirical: Vertical triangulation Like constructivism, CR acknowledges the subjectivity of empirical data, yet distinguishes it as an additional layer obscuring the underlying events and mechanisms (Easton, 2000). Hence, vertical triangulation of event data and interpretations is first required so as to peel empirical experiences from actual events. We dealt with this by systematically analyzing commonalities and differences in secondary and perceptual data and by then identifying the bases of such differences. The interview rounds were useful to broaden the range of interpretations and deepen out their objective and intersubjective cores through confrontation and specific case examples. Vertical triangulation between subjective experiences and actual events was also supported by combining sources informing SWC management with independent data sources (e.g. OECD, Eurostat, national statistics) which SWCs typically did not access. Peeling off the empirical also meant exposing the researchers' subjective assumptions through dialogue (Bhaskar, 1978). To this end, we confronted our a priori interpretations of secondary data with a panel of experts' opinions in focus groups. In the interviews, we also confronted managers with our interpretations of the findings to that point by presenting them at the end of each interview, so as to elicit qualifications and competing explanations. Furthermore, interpretations were consistently discussed between the two interviewers right after the interview and subsequently reviewed and discussed with the third author as well, based on the interview summary. These steps proved instrumental in exposing our assumptions.

2 Abstraction: Deconstructing events & identifying causal dynamics As CR, unlike positivism, distinguishes between the actual and the real, there is also a need to deconstruct events and interpretations into the causal mechanisms and contingencies that activate and actualize them (Ryan et al., 2012; Sayer, 1992). It does not equate the constant conjunctions measured in event regularities with causal laws, as the presence of a real object should not always coincide with the manifestation in the ‘actual’ domain of the effect it has the generative power to produce. Abstraction differs from the positivist notion of theoretical isolation (Lawson, 2009) in that objects and mechanisms are studied within their context rather than isolating them from the contingencies and conditions that might affect their operation. Abstraction iteratively develops and tests causal logics that best fit the triangulated observations in a process of retroduction, a logical reflection “on not only what happened, but what could happen or what hasn't happened” (Ryan et al., 2012, p.302). The iterative aspect in this study meant that coding occurred in three phases, resulting in the data hierarchy presented in Fig. 2, inspired by Corley and Gioia's (2004) approach. 1st-order coding was inspired by original assumptions about the problem and followed directly from the data. 2nd-Order coding came from confronting the data with theories that typically offer insight into supply chain issues, e.g. dominant logic, industrial networks, organizational behavior, resource dependence or industrial organization. The aggregate constructs materialized once it became clear that the interface of IMP and institutional theory offered the most fitting theoretical frame for this case. 3.4. Explanation The task of explanation involves searching for and selecting the best fitting causal explanation for the events observed (or not). CR differentiates between the explanatory value of theories given a certain research problem, unlike constructivism's judgmental relativism (Miller & Tsang, 2010; Ryan et al., 2012), or the positivist belief that one theory is universally superior to another (Popper, 1965). Aggregate

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

411

Fig. 2. Coding structure.

constructs (Fig. 2) hence followed from our final selection of the combined IMP–institutional entrepreneurship framework. Rather than sequentially following the prior phases, this task runs throughout the entire project and involves systematically iterating between data and theories to capture the best fitting explanations. So, even though this activity is most visible in the retroductive final coding of aggregate constructs (Fig. 2), it is already present in the first-order coding of the empirical, theory-laden, observations. To sufficiently both show and interpret data (Pratt, 2009), we insert separate analysis and explanation sections. These respectively illustrate how the analytical inference of 2nd order and aggregate constructs is grounded in the case story, and derive generalized propositions by relating these constructs to extant theories. 4. Investigation: Problem description 4.1. The industry context The case focuses on the Dutch steel wholesale industry, consisting of steel distributors that keep a physical stock to meet contracted or ad hoc steel demand. Power in the Dutch steel chain is high upstream and downstream as well in markets with big OEMs, with SWC residing in the squeezed center of this hourglass model (Table 2). Depending on the end-market, the supply chain is either structured in elaborate tiering systems or coordinated by the OEM. Typically the midstream from the steel mill towards the OEM would

consist of respectively service centers, wholesalers, jobbers, product manufacturers and systems suppliers. The main downstream applications for steel in the Dutch economy are steel constructions (42%), steel processing (13%), mechatronics (11%) and shipyards & offshore (7.6%). Except for construction, these industries consist mainly of innovative SMEs that compete on technological capabilities in highvalue segments. As to our focal type of companies, SWCs, following three types of Dutch SWC exist in increasing degree of volumes per contract and delivery flexibility: regional wholesalers, large independent wholesalers and branded factory ‘outlets’. In reality there is also a fourth type, steel traders. As they are essentially spot market steel brokers, they are not embedded in the local networks and hence not central to this study, although we did have interviews with three such traders. SWCs are generally SMEs, with the exception of the few factory outlets, which are also the only ones regularly serving international customers. Regional wholesalers supply small volumes to local customers, like contractors or installers and are highly flexible to special requirements regarding form, size, finishing or delivery. Independent SWCs distribute industrial volumes to diverse sections of the market, with often a small number of specializations and significant serviceaddition. Serving the more mobile and heterogeneous middle field in the industry, these SWCs are most prone to the change dynamics described in this case and thus the main focus. Branded factory outlets, operated by the steel mills, supply large industrial buyers with their own primary processing capacity (e.g. automotive).

412

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

Own calculations, based on a report from Beddows (2010). Eurometal, 2010. 25% more of them reporting substantial YTD revenue losses than in other segments in the first semester of 2009 (Ruis, van den Berg, & Meijaard, 2009).

‘big three’ miners' 40-year old benchmark pricing mechanism. European players became pure price-takers in a market governed by iron ore oligopolists and Chinese producers. The recession and globalization put pressure to evolve on the entire steel chain. A senior manager of Europe's largest producer, for instance, reported being “in the old world of internal competition between production and factory outlet with one leg and in the consolidated new one with another” (see respondent P1 in Appendix B). In the SWC segment this was especially felt by the larger independent SWCs. 6 These pressures manifested in search behavior among SWCs even more. Factory outlets, for instance, became more interested in relatively smaller volumes than before, “trying to become more flexible in their offerings” (J2). Among independent SWCs, this search behavior lead to innovation attempts as well. Yet, they were generally unsuccessful, even those initiated collectively or carried by heavyweights in the chain. This was for instance the case when a large producer introduced a new steel sheet product optimized for high velocity laser cutting. The superior finishing and material of this product allowed service centers and jobbers lower turnaround times. It was very promising in terms of added value, was technologically sound and there was a clear demand downstream. Nonetheless, the project ran aground in distribution, where SWCs' deviation from prescriptions considerably destroyed demand and profitability. Open innovation attempts oriented at the commercial challenges of SWCs neither yielded much success. For instance, a forum to improve the distribution of steel sheet material to the Dutch industry was initiated collectively by SWCs. It gathered support from external partners, such as consultants, technical universities and OEMs to provide SWCs with knowledge and other resources to become better at their logistical function. Neither the industry association, nor partaking managers, report any noticeable success in the projects undertaken. In retrospect, the common denominator distinguishing between successful and unsuccessful projects seems to be whether they are oriented towards sustaining the traditional SWC function or at working towards an entirely different organizing template. One individual SWC was able to increase its position and its value captured in the market by acquiring its own production capability. As such, it gained new knowledge and foremost a new image. It was now perceived as a producer, which did provide a legitimate position to initiate innovations in the chain. This was also true for the scarce successful attempts at collective change. A project centered on steel construction applications and another on steel conservation, turned out to be examples of embryonic institutional entrepreneurship. They involved collective change with diverse actors from within and beyond the traditional supply chain to address a challenge for the entire supply chain and beyond. Rather than seeking to acquire knowledge or resources to better conduct their business, SWCs strikingly took another role and specifically aimed at developing a new value-added business model integrating the efforts of the entire industry towards different end customers. Although these projects would certainly be of interest for future studies on successful institutional entrepreneurship, these cases are not further developed here. Rather, the case study focuses on the antecedent conditions necessitating such institutional entrepreneurship in the first place.

4.2. Failed innovation attempts

5. Analysis

The Dutch steel chain faced structurally decreasing volumes due to substitution. Capital investments decreased due to past delocalizations and downstream consolidation. Meanwhile, the Chinese economy soared and its share of global crude steel demand rose from 15.7% of global demand in 2003 to 46.4% in 2009, while the EU27 market share dropped from 23.1% to 11.3% (World Steel Association, 2010). Chinese producers consequently overturned the

As it turns out, the innovation attempts in the SWC industry segment failed because of being out of touch and lacking legitimacy. The analysis section, guided by Fig. 2, examines the nature of this

Table 2 Characteristics and trends in the Dutch steel chain. IRON ORE MINERS

Structure • Global oligopoly (C4-index global trade: 65%a). ‘Big 3’ (Vale, RioTinto & BHP): 61.2%. Impact on the chain • Determine industry's pace and price (capture 41% of upstream value globallyb), but lost some power to Chinese buyers: now flexible pricing & trend from bilateral to online trading. Trends • Persistent price rises due to increasing demand; fear of depleting reserves.

Steel producers

Structure • From “loosely coupled” towards integrated producers. • From fragmented national companies towards global intra- & inter-organizational consolidation. Impact on the chain • Price takers globally, but still capture 28% of global upstream valuec; heavyweights in Europe. Trends • Flexible pricing forced on them: from yearly to monthly prices. • Some backward integration to mitigate dependence. • Chinese producers are becoming bigger and upgrading to higher qualities. • From production to market focus; increasing importance of own distribution unit for market access & towards combining scale with flexibility and added value in services.

Midstream

Structure • Fragmented with many specialized SMEs; the role of SWC depends on end-market. • SWCs control substantial share of steel towards industry (45–60%). Alternatives: steel mills (direct sourcing) or traders. Impact on the chain • SWCs: Mainly customer-driven & reactive distribution role (determined by upstream and downstream parties): neither innovative nor responsive to new competence demands. • System suppliers have most power, this decreases fast upward in the chain. Trends • Funneled relay of flexible pricing and lack of price transparency. High price volatility requires increased attention to organization of logistical processes.

OEMs

• Typical markets: steel construction, automotive, machinery & equipment, shipyards, agro-technical, high-tech, mechanical engineering. • Mostly high specialization (low volume, high mix), technological innovation and customer-specific solutions to counteract customer switching in global competition. Seeking differentiation. • Internationally oriented and market-driven. Often nationally anchored but at least a pan-European customer base and further internationalizing. • Orchestrate demand-driven value chain because of end-market access.

a

b c

6 25% more of them reporting substantial YTD revenue losses than in other segments in the first semester of 2009 (Ruis, van den Berg, & Meijaard, 2009).

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

413

Table 3 The added value gap: increasingly different logics and templates.

Key competences Key activities Partners Revenue model Customer interaction Value proposition

SWC business model

Downstream business model

Delivery speed & flexibility, market knowledge, material specialization Storage, primary cutting & processing, bundling, bulk-breaking, delivery Supply chain Volume-push Defensive & reactive Product+: basic services; cutting into small or special sizes; special qualities; speedy & flexible delivery

Product & technological innovation, build-to-concept/specifications Codevelopment, engineering, production, design Value networks Value-pull; working the market Cooperative; coordination Custom solutions; integral business model; ‘unburdening’

misfit. It also analyzes how the mechanisms of cognitive and normative institutional inertia appear to have caused a vicious cycle that leads to this misfit. We will describe how the observed items (Fig. 2) were experienced in the supply chain.

5.1. Institutional misfit More dynamic downstream markets, such as mechatronics and offshore had since long been competitively challenged to look for new ways of creating value. These firms thrived on their innovative capacity, leading to a perceived gap in capabilities and focus with SWCs'. Although buyers were looking for knowledge and development partners, SWCs were not considered “knowledge beacons” (SI3). Commenting on the growing distance with their suppliers, SI2 asserted: “the bulk of the added value we produce comes from our engineering, not the quality of our materials”. As buyers' business models and technologies became more technologically advanced, their demands shifted accordingly (Table 3) and SWCs' mundane logistical and primary processing functions diminished in the total value of end products. While purchasing criteria increasingly became legitimacy measures like speed, flexibility, and reliability in traditional downstream markets, a large middle class of underserved buyers arose in the medium volumes, high value-added buyer segment. They demanded application specialization, intelligent delivery solutions and advice on markets and materials in return for the distributor's premium. Also, supply chain management, market forecasting and financial hedging became increasingly valued. Yet, SWCs hardly responded with appropriate offering upgrades, leading buyers to seek alternatives like acquiring cutting installations to allow purchasing directly from steel mills, collective buying or procuring outside Europe via traders. This, for instance, manifested in SWCs' dropping share of Dutch steel distribution from 50 to 40% in 2009 (Industry association SFN). Growing more and more detached from the inert steel chain, OEM purchasers had “little to no affinity with steel anymore” (O9). Operating in basically a completely different type of environment, they focused on the demands posed by it, which had little relevance with the value propositions and the issues put forward by SWCs. This lack of attention for SWCs' value proposition was furthermore institutionalized through the development of tiering structures, in which “supply chain management… [was] outsourced to contract manufacturing partners” (O3), hence to the next tier. Finally, the inertia and disconnectedness of the steel chain left them unprepared for the demands posed by the rapidly changing global steel market. With regard to globalization the historic steel chain had become a laggard: “steel is one of the biggest and oldest industries in the world, but only in its infancy concerning globalization because of the inherited national structures” (P1). The increasing capacity in the East was considered a threat, especially as “Chinese producers will not focus on lower qualities for long anymore; their production facilities are at least as modern and being built with European knowhow”

(T2). This was also visible in the higher-end stainless market, where respectively 12 and 13 of the top 20 global long and flat steel producers were Asian (Moll, 2011). Despite the flexibility demanded by ever shorter price cycles where “risk is pushed ever further downstream” (O1), SWCs failed to develop the ability to cope, not really knowing which way to turn. Managers' sensemaking about these demands remained mostly vague (e.g. we have to “move from thinking in terms of volumes towards focusing on ‘specials’ more and more” (S7)) or just confused (e.g. “pressured into becoming a ‘candy store’ with a broad assortment, while at the same time becoming a product specialist” (T3)). All these factors together created a gap that blocked successful organic responses from the SWCs, which were hopelessly behind their supply chain counterparts and customers. The next section will go deeper into the causal mechanisms that allowed things to get so far.

5.2. Normative institutionalization Steel is a commoditized product, but tight control over the chain and high entry barriers allowed producers to make a profit by protecting distribution channels .To maintain control over the supply chain, producers conservatively restrict access only to reputed distributors and keep supply scarce. As SWCs sell more than half of producers' capacity, they depend on each other. 7 Due to long periods of structural stability actor bonds “were cemented historically” (P1), implying that relationships within the steel chain were fixed and entry by outsiders was considered unfeasible. One inbound trader (T1), with a + 50 years track record stated that even for him “it is impossible to penetrate the canalized European steel chain, despite our long history, network and accomplishments”. He only knew of one firm (T2) that had been able to conquer a spot among the happy few and even then because they were actually a sort of management buy-out from existing SWCs. For SWCs, this meant that rather than the value one added, one's “main competency is its access to producers and its reputation” (T3), the latter mainly referring to consistency and reliability. This conservative, rigidly canalized nature of the Dutch steel chain turned it into a protective cocoon where business development virtually stood still. Making matters worse, downstream buyers were equally conservative due to issues with reliability of quality and delivery among others. Reliability was the most mentioned reason for not wanting to further cooperate with SWCs. Buyers often complained about SWCs tendency of “buying-in wildly” (O9) and especially in high-grade applications, the “difficulty in finding properly certified steel” (SI2). Besides quality issues, non-transparent pricing gave SWCs an opportunistic and untrustworthy image.

7 Crowding in the middle field creates tension between SWCs and steel producers, who are pushed to look for new business in the middle field, but at the same time have to rely on the SWCs to distribute most of their production volumes. As the latter closely monitor their markets and have the possibility to switch suppliers, producers face a realistic threat of retaliation as market access is becoming increasingly important.

414

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

Their suppliers and buyers also have a significant advantage in bargaining power over SWCs due to their size, contribution to the value of the end product and, in general, their position in the strategic supply network. Due to their low impact and unreliability, SWCs were most often considered bottleneck suppliers, ideally replaced by a better alternative when possible. SWCs' perceived network position was hence restricted by their suppliers and buyers to connecting supply and demand in a responsive manner. Even independent SWC, although rich in supply chain connections and the most entrepreneurial SWCs, were not perceived as potential co-creators by their customers, or as one buyer put it: “To be honest, we have never even considered wholesalers for innovation” (O5). In sum, the gap between buyers' normative expectations and SWCs' performance left them no room to really change their position in the supply chain. Hence, even the proactive initiatives of the most entrepreneurial SWC aimed to “help our customers win contracts” (S2), only led to “customer bonding, but no willingness to pay for the extra value delivered” (S2).

involved in the study reported being prepared to go into such requests, yet clearly did not want to share more about the arrangements he had already worked out. SWCs considered entering or developing new markets nearly unfeasible, time-consuming, they did not know the market and had little confidence in it. Upstream parties repeatedly complained about SWCs' external locus regarding innovation: “All the innovation in steel has to come from us. Wholesalers are just not interested in that” (P2). As SWCs' passiveness regarding change and the negative normative reactions from the supply chain reinforce each other, they are captured into a position of learned helplessness, a persistence of passive maladaptive behavior (Martinko & Gardner, 1982). So, even when buyers did convey an interest for additional services, SWCs were “not interested in providing price forecasting, hedging or consulting about material characteristics”. They had closed their minds to the possibility of innovation or more intense collaboration. In contrast, we have talked to traders who were newcomers to the market and developed a supply line with Asian producers and an innovative logistical model for import in Europe in less than 5 years.

5.3. Cognitive institutionalism The majority of SWCs themselves were a conservative factor in the industry. They were often characterized as somewhat arrogant in that they did not add much value in the chain, but believed to be protected from the increasing pressures on the industry by their historically fixed relationships with producers. However, as the latter are forced to reorient, these lucrative relationships might be in contradiction with the actual proportions in the industry. Nonetheless, the prevailing logic was not challenged and the protection of the market had always provided a positive performance feedback. The high velocity laser cutting project, for instance, considerably increased jobbers' capacity and allowed a significant mark-up for the producers and the SWCs. The latter, however, persisted in their old “high-volume, low value-added” (S9) logic and cut corners on material storage and treatment, despite having received elaborate prescriptions from the steel mill. This caused corrosion and other problems in the steel processing, destroying the product's added value. The problem turned out to be not a matter of financial incentives or lacking competences, but merely a persistent logic, “wholesalers are so occupied with getting the best deal that they miss the big picture” (P1). So the producer was obligated to include penalty clauses and perform regular inspections to adjust the SWCs' behavior. All parties found SWCs responsive in their logistical role, but otherwise rigid. Following comment captures the dominant perception regarding SWCs' myopic mindset: “the problem with wholesalers is that they are so exclusively wholesale-minded and pay little attention to other elements like total-cost-of-ownership or financial hedging” (O4). Even now some still refuse to acknowledge a need to change. Decreasing volumes (demand in 2015 was projected in the best case at only 88% of the 2008 level, up from 84% in 2009) induce entry on SWCs' core business. Crowding is observed on lucrative niches in the middle field by SWCs, traders and steel mills' factory outlets. Traders were building their own service centers, SWCs building or buying their own production capacity and branded factory outlets becoming ever more flexible in volumes and service. Often, several roles are combined within one company and supply chain links are skipped. This results in a complex web of hard to distinguish actors and relations in which many are simultaneously suppliers, customers, competitors or project partners. SWCs invariably interpreted this evolution as a threat, which lead to defensive reactions. As suppliers, lateral partners or customers become new entrants, SWCs are reluctant to openly communicate or cooperate “We have asked them to be more transparent and to figure out innovative price control systems together, but they refuse to be open about it” (O7). SWCs instead resorted to protectionism and intransparency, responding to such requests with “What's in it for us?” (O9). In fact, only one of the SWCs

6. Explanation: Boundary conditions for institutional entrepreneurship in networks Our findings indicate how the Dutch steel industry has been marked by decades of relative stability which rendered it excessively conservative. Then, as the global socio-economic context required it to change radically, the midstream of the steel chain was slow to react and when it did, these attempts were largely unsuccessful, except for some small successes for those companies who had already been operating partly outside the SWC field. Emergent adaptation, as suggested by traditional IMP, did not occur. The additional explanations we abductively tried, building on dominant logic and competence-based barriers did not suffice either, as we observed able, open-minded and innovation-oriented SWCs try and fail as well. Further explanations based on resource dependence and industrial organization to account for SWCs' wedged position compensated these shortfalls somewhat. Ultimately, however, we were guided towards institutional entrepreneurship by the scarce successful collective change attempts and the realization that the threats faced by SWCs needed to be addressed on a meso-level. This accounts for the collaborative element and the structural constraints in its explanation of agency in. It urged us to explain why SWCs failed to acquire the necessary cooperation from their suppliers or customers, which in effect constitutes their network partners. We found that a number of normative and cognitive barriers blocked attempts to develop new business models in collaboration with network partners. We explain the boundary conditions of this extreme situation in terms of cognitive and normative institutionalization and the institutional misfit they result in. This section summarizes the generalized explanation derived from the industry case-study into testable propositions as to when the network structure becomes a boundary condition for institutional entrepreneurship, thereby relating the explanation to the theories it builds on. 6.1. The ‘real’ mechanisms: Cognitive and normative network institutionalism As expected, cognitive barriers were present in the Dutch steel industry, increasing its evolutionary rigidity. Innovation in the European steel market is historically sparse and within an old institutional logic of upstream technological innovation. Cognitive inertia (Hodgkinson, 1997), characterizes the historically conservative, stable Dutch steel chain. In our case, we find such cognitive persistence as a consequence of institutional logic (Scott, 1987), defensive routines (Macbeth, 2002) and learned helplessness (Martinko & Gardner, 1982). These factors of

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

cognitive inertia by themselves were insufficient to explain the persistent, increasing misfit in the steel chain. They were to an important extent caused by a lack of challenge in the protective cocoon that had been institutionalized through normative expectations for legitimacy by the conservatively selective producers. Firms conform to these normative prescriptions because legitimacy improves survival and growth probability (Baum & Oliver, 1992; Zimmerman & Zeitz, 2002). In networks, legitimacy is pursued by becoming embedded, which adds a relational component (Moran, 2005) and is hence conducive of competitive differentiation (Tsai & Echols, 2005), learning and innovation capacity (Powell, Koput, & Smith-Doerr, 1996). However, embeddedness of an organizational field also leads to taken-for-granted templates (Greenwood & Hinings, 1996) and dissuades intensive environmental scanning (Daft & Weick, 1984), thereby externalizing risk. Being overly embedded leads to decreasing marginal returns and thus becomes paradoxical (Uzzi, 1997), hindering value creation and adaptation logics. In the Dutch steel chain, with its ‘cemented’ ties, normative institutionalism left SWCs with limited degrees of freedom notwithstanding an immanent need for change. Moreover, the cognitive deficiencies further reinforced the normative walls surrounding the SWC field through negative feedback regarding their added value, their technological distance towards downstream parties and their troubles in attaining minimum quality standards. So, even if SWCs were open to communicate or cooperate, a lack of legitimacy in their potential partners' eyes prevented intensive information sharing or co-creative relationships to grow organically. This created a vicious cycle. This self-reinforcing cycle blocked proper working of the normal mechanisms for mutual adjustment in networks (see the IMP model of change; Section 2.3.1). Proposition 1. ‘Emergent’ cooperative adaptation is hindered when the network is simultaneously (a) normatively and (b) cognitively institutionalized. 6.2. The ‘actual’ state: Institutional misfit in the face of turbulence Cognitive and normative inertia caused a buildup of institutional misfit; a deviance from the value expectations, organizational template or institutional logic (Kondra & Hinings, 1998; Söllner, 2002) demanded by the network's context or stakeholders (Greenwood &

415

Hinings, 1996). The increasing gap between the steel chain and the worlds of its technologically and internationally far more advanced customers had severe implications for the position of SWCs. As their behavior is directed by the issues in their environment (Ocasio, 1997), customers' attention increasingly averted from SWCs' value creation logics. The distance between these worlds was also visible in their technological expertise and eventually becomes institutionalized in tiering systems and more advanced business models. As the cognitive and normative institutionalization persists, the misfit continuously increases. This gradual buildup of misfit increases the discontinuity of the change momentum (Meyer et al., 1990). Exacerbated by the sudden strong exogenous change dynamic, this historically grown misfit overloads the ‘normal’ change diffusion structures in networks and contributes to their shutdown. Without the burning platform of this radical change demand, the steel chain can remain just doing what it had been for decades. Without the high degree of cognitive and normative institutionalization, the supply chain might just have needed to coordinate more intensely in its existing configuration (Lundgren, 1992). Proposition 2. Self-reinforcing cognitive and normative institutionalization in a network gradually increases institutional misfit, causing too high a change discontinuity for the institutionalized network to handle. 6.3. Towards an extended model Fig. 3 illustrates how our institutional change model in networks relates to the emergent, distributed model. Specifically, it extends Lundgren's (1992) description of how the actors–resources–activities (ARA) model reinforced with increased coordination translates discontinuous change into network adaptation. In our model, this mediation mechanism shuts down as cognitive and normative institutionalizations reinforce each other to destroy the elasticity of network ties. The change momentum then becomes too much for this ‘sick’ network to transmit. In this extreme situation, institutional entrepreneurship addressing the structural templates for organizing is required for the network to adapt. So when discontinuous change is necessitated by severe institutional misfit and the network is normatively and cognitively institutionalized, emergent adaptation is blocked and needs to be substituted by

Fig. 3. Change through institutional entrepreneurship in industrial networks.

416

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

institutional entrepreneurship. These two boundary conditions hence delineate the traditional industrial network explanation of change (Ford et al., 2011) and serve as an antecedent to institutional entrepreneurship in networks. This type of change requires devising entirely new platforms or structures to facilitate collective change and does not only occur following an immediate critical event, but can also be the consequence of a gradual buildup of institutional misfit. Our industry case showed a combination of both, as a buildup of decades resulted in radical actions triggered by the recent economic crisis. Proposition 3. In institutionalized networks, discontinuous strategic change caused by a high institutional misfit, needs to be unblocked through institutional entrepreneurship.

7. Conclusion Industrial network theorists appear to be divided on the question of network manageability (Golfetto et al., 2007; Rampersad et al., 2010). Some view networks as truly emergent, resulting from the open-ended interactions between actors who benefit from these relationships (Ford et al., 2011; Ritter et al., 2004). Others reject the assumption of ‘unmanageability’ (Heikkinen et al., 2007; Möller et al., 2002) and claim that “the potentiality for intentional ‘management of networks’ varies between different types of networks” (Möller et al., 2005, p.1275). In traditional IMP thinking, change can be categorized as evolutionary following van de Ven and Poole's (1995) typology. It is the consequence of interdependent distributed agency (Azimont & Araujo, 2007) and its outcome remains unpredictable. The change described here, introducing institutional entrepreneurship, is instead governed intentionally (Möller et al., 2005) and becomes ‘teleological’ (van de Ven & Poole, 1995), implying it is purposeful towards a known goal, despite holistic network or institutional constraints. The critical realist orientation underlying this study acknowledges both the agency and the structure stance and would consider denial of either as a form of conflation (Archer, 1995). This duality supports the possibility for networks that due to their simplicity or high centralization of power are manageable as well as for networks where firms can only ‘manage in’ them (Golfetto et al., 2007) because of their opaque, complex or vast nature. As such, the differences between these two versions of network theory can be considered as different ‘modes’ from a critical realist stance (Möller & Rajala, 2007). Both are plausible network forms, they just result from different configurations of generative mechanisms and contingencies (Bhaskar, 1978). Rather than a fundamental divide, the disagreement thus pertains to the degree of aggregation (Lundgren, 1992), determination (Möller et al., 2005) or complexity (Möller & Halinen, 1999) in the network. The present article contributes to this ongoing discussion by reconciling the potentiality of both types from a CR underpinning and by delineating the boundary conditions that warrant a transition to an intentional network ‘mode’. We find that when networks are in a self-reinforcing spiral of cognitive and normative institutionalization, they need to be unblocked through institutional entrepreneurship, as regular agency fails. This suggests criteria for managers to differentiate between competitive and institutional strategies and know which is appropriate when. Further research could hence make an important contribution to practice by further specifying signals of cognitive and normative institutionalization and by investigating which institutional entrepreneurship approaches then are effective, given the specific causes of institutionalization. Although we witnessed examples of institutional entrepreneurship, they were embryonic. A more longitudinal study could serve to produce more robust insights on best practices.

An intentional mode of ‘managing networks’ might require different competences, structures and roadmaps. Proposing institutional entrepreneurship for firms to achieve this transition and maintain initiative, rather than having change being imposed upon them, points for instance to the role of political and social capital, cross-sector platforms and stakeholder roadmaps. Further research combining these concepts with industrial network realities could inform strategies to address meso-challenges such as industry globalization, commoditization and decreasing industry margins. Introducing institutional entrepreneurship as another form of network strategy also opens an avenue for managerially relevant research into determinants and forms of successful institutional entrepreneurship in networks. These would be inspired by theories related to the concept (e.g. institutional theory, social movement theory, social capital theory). One could expect each institutionalized context to be a unique combination of several dimensions of persistence, thus requiring different responses. Hence, comparative research could focus on the interactions between specific mechanisms of network institutionalization and the specific dynamics that counteract them in multiple contexts. Specifically, researchers could investigate the role of resources (e.g. social capital or bridging ties) and enabling factors (e.g. industry associations, interest groups) for institutional entrepreneurship strategies. Seeing governance structures as a continuum of networks (Möller et al., 2005), sets aside the classical sharp divide between hierarchies and markets (Williamson, 1991). It points at network structures as encompassing the powers of both agency and structure, thus representing the more relevant categorizing frame for competition policy makers. The current study indicates that it would be beneficial to adapt competition and antitrust policies to the degree of institutionalization in an industry. In institutionalized industries, policies that foster collaboration and open interaction between institutions and commercial actors are likely valuecreating for society. As these commercial actors are often the initiators of institutional change, incentives for entrepreneurship also serve a social purpose. Finally, the current paper is also an illustration of how CR methodological principles (Easton, 2000; Ryan et al., 2012) can be translated into case-study research that makes a mid-range contribution. It demonstrates how a CR underpinning, by favoring selection of the ontologically most fitting theoretical lens, potentially increases some of the much needed relevance (Hodgkinson & Rousseau, 2009; Visconti, 2010) in empirical management studies. In this respect, further examples of CR-inspired studies and a debate on what should and should not be considered appropriate approaches to CR method are much welcomed to offer researchers more guidance in undertaking CR research. As the article is based on a single extreme case which serves as inspiration (Siggelkow, 2007) for the propositions made, it has some obvious limitations in terms of external validity. The proposed delineating dimensions of the interface between institutional entrepreneurship and the industrial networks perspective are only suggestions for further deductive testing. Moreover, comparative analyses in multiple contexts might lead to the identification of additional or completely different contingencies prompting institutional entrepreneurship.

Acknowledgments The authors thank the three anonymous reviewers and the special issue editors for their valuable suggestions, which substantially improved the paper. The authors also thank StaalFederatie Nederland for their financial support and their active cooperation in the study.

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

417

Appendix A. Characteristics of positivist, constructivist and critical realist research

Logical empiricism

Radical constructivism

Objective Finality of science

Prediction (what) Identifying universal laws in observations (of actual events)

Epistemological position Epistemological emphasis

Incommensurability

Understanding (how) Causal explanation (why) Expanding knowledge of reality; Meaning-making Bringing the empirical in phase with Accurate causal explanation, bringing the empirical ‘in phase’ with the real the actual Eclecticism Inter-ontology crossovers

Critical realism

Mainly deductive, preferentially quantitative; Falsification Statistical sophistication • Sampling logic for generalizing to populations/universes

Broadly-scoped inductive Grounded in data; Local knowledge creation (limited generalization)

Role of theory

Direct representations of reality that are either true or not

Simplicity Generality Accuracy Chief criterion of legitimacy

High High Low Internal validity

Narrative to depict social processes in terms of: • Sensemaking • Enactment Low Low High descriptive accuracy Fit with actual interpretations

Tightly-scoped/abductive Transcendence • Continuously and continually asking why • Distilling mechanisms and contingencies from events Triangulation • Data & theoretical grounding • Objective & subjective data • Dialectic/systematic combining • Replication logic & analytical generalization • Theoretical sampling for requisite variety Enlightenment: explaining the deep structures, drivers and contingencies underlying events. Theory is necessary to develop and test Moderate–low High High analytical/ontological accuracy Ontological fit

Based on Bhaskar (1978), Easton (2002, 2000), Geels (2010), Kwan & Tsang (2001), Langley (1999), Lawson (2009), Miller & Tsang (2010), and Ryan et al. (2012).

Appendix B. Respondents

Producers

SWCs

Traders

Jobbers Systems integrators

OEM

End-users (B2B) Industry assoc. & knowledge centers

a

Code

Company description

Contact function

P1 P2 P3 P4 S1 S2 S3 S4 S5 S6 S7 S8 S9 T1 T2 T3 J1 J2 SI1 SI2 SI3 O1 O2 O3 O4 O5 O6 O7 O8 O9 O10 E1 E2 K1 K2 K3

Large global, EU rooted strip products Dutch rooted, small EU stainless German rooted, midsize EU long specials Large EU-rooted flat carbon & steel products Global producer factory distribution outlet Mid-large integrated independent distributora Large EU producer factory outlet Large independent multimetal distributor Midsized producer factory outlet Regional long & multiproduct distributor Midsized independent specialty distributor Large independent international tube distrib. Large indep. Benelux multimetal distrib. Small inbound trader Large outbound multimetal trader Medium inbound and regionalb trader Large steel sheet & structures processer Midsized steel cutting & finishing center Hightech applications & systems supplier Hightech precision components & systems Offshore component & systems supplier International steel office furniture OEM Large diversified steel construction group Hightech lithographic machinery producer EU-based truck & trailer OEM Agrarian equipment, machinery & systems Dutch ion & electron beam equipment OEM Potato processing machine builder Dutch-based international printer producer Bakery machinery & equipment builder Climate control systems OEM & services Food equipment reseller Maritime coating & finishing supplier Eurofer, European steel knowledge center NEVAT, component & systems association SFN, SWC trade association

Business development mgr Director Managing director Managing director NL Managing director, 2 business unit managers Director Managing director the Netherlands CEO Managing director CEO Marketing director Business unit manager CEO CEO Partner & CEO Account manager CEO Director Production mgr Managing director CEO & market intelligence manager Strategic purchaser CEO Procurement director Commodity manager; Project coordinator Strategic purchasing manager CEO Engineering mgr; logistics & production manager 2 Purchasing account managers Purchasing manager CEO Advisor Marketing manager Director market analysis & economic studies President Branch manager

Independent wholesaler that integrated backward by buying its own production site and forward by combining the knowledge of both activities to develop project capabilities. Inbound trader, who built its own primary processing installation, thus entering the SWC market.

b

418

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

Appendix C. Interview guides Appendix C. Interview guides

Birds-eye interview guide Company ID-kit: • # employees • International presence • Position in the chain • Revenues & evolution Metal sales • Types of steel/metal • Annual steel volumes • Share of SWCs in steel/metal sales Supply chain relationships & the role of SWCs • How are the relationships in the steel chain? What are they based on? • What are Key Success Factors in different segments of the steel chain? • How dynamic is each of these segments? What are the main drivers of this? • Describe the nature of your relationship with SWCs. • How could the metals/steel chain perform better? What does it require? • What is the current role of SWCs in the supply chain? How is it evolving and what about it should change? • What might be holding them back? Future (potential) of your industry and the industrial sector • Which are the major trends affecting the steel chain now and tomorrow? • How will they impact on the steel chain in the coming years? • How will Dutch/European companies be able to maintain their position? • Which activities will remain and which will delocalize? • What are the challenges for different segments of the steel chain? • Which options for improving does it have? Downstream interview guide Company ID-kit: • # employees • International presence • Position in the chain • Revenues & evolution Metal purchases • Impact of steel/metals (% of revenues) • Role of steel/metal (% in end product or in maintenance, repairs & operations) • Role of metal suppliers (on Kraljic matrix) • Steel volumes (total and per transaction) • Share of SWCs in steel/metal supply • Single or multiple sourcing? Main reasons/selection criteria? Added value potential of SWCs • In which areas do SWCs perform well/poorly? • Could SWCs take up additional value creation roles/services according to you? • What might be preventing them to do this? Desired role of SWCs: According to you, how should SWCs… (factors mentioned here differed according to preceding secondary analysis on the company's strategy etc.) • Increase their international services? • Cooperate differently? • Employ different business models? • Contribute to innovation? Supply chain relationships • Describe the nature of your relationship with your steel/metal suppliers. • Describe your transactions with SWCs and what Key Success Factors in them are. • How could the (metals) supply chain perform better? What does it require? • What are the functions for SWCs in that supply chain model? Future (potential) of your industry and the industrial sector • How will the industry evolve in the coming years? • How will your company and Dutch companies like it to be able to differentiate on the market? • How can continuity be maintained? • Which activities will remain and which will delocalize? Midstream interview guide Company ID-kit: • # employees • Market scale & international presence • Position in the SWC field (3 types) • Revenues & evolution Metal sales

Appendix (continued) C (continued) Birds-eye interview Midstream interviewguide guide • Types of products • How are you able to differentiate in your market? • How dynamic is your market? What are the main drivers of this? • Annual steel volumes Supply chain relationships & the role of SWCs • Describe the nature of your relationships with suppliers? What drives this? • Describe the nature of your relationships with customers? How would you like them to evolve? • What is your current role in the supply chain? How is it evolving and what about it should change? • What are the main barriers and drivers for such an evolution? • What are the main demands posed by your customers? How do you try to meet them? • How do you try to innovate? Describe some major examples. • Can you also give examples of failed projects? Specific trends in your industry • Which trends (are starting to) drive changes in your market most? • How do you interpret the following trends and prognoses (provided by us)? • How will this affect you? • What are the primary actions that should be taken to act on these trends? • To what extent should other parties be involved in these strategies?

References Akgun, A. E., Lynn, G. S., & Yilmaz, C. (2006). Learning process in new product development teams and effects on product success: A socio-cognitive perspective. Industrial Marketing Management, 35(2), 210–224. Archer, M. (1995). Realist social theory: The morphogenetic approach. Cambridge: Cambridge University Press. Arndt, J. (1985). On making marketing science more scientific: Role of orientations, paradigms, metaphors, and puzzle solving. Journal of Marketing, 49(3), 11–23. Arora, A., & Gambardella, A. (1990). Complementarity and external linkages: The strategies of the large firms in biotechnology. The Journal of Industrial Economics, 38(4), 361–379. Astley, W. G., & Van de Ven, A. H. (1983). Central perspectives and debates in organization theory. Administrative Science Quarterly, 28(2), 245–273. Azimont, F., & Araujo, L. (2007). Category reviews as market-shaping events. Industrial Marketing Management, 36, 849–860. Baraldi, E., Brennan, R., Harrison, D., Tunisini, A., & Zolkiewski, J. (2007). Strategic thinking and the IMP approach: A comparative analysis. Industrial Marketing Management, 36, 879–894. Bartels, R. (1968). The general theory of marketing. Journal of Marketing, 32(1), 29–33. Baum, J. A., & Oliver, C. (1992). Institutional embeddedness and the dynamics of organizational populations. American Sociological Review, 57, 540–559. Beddows, Hatch (2010). The traded market for ultrafine iron ore. 5th Annual EU Iron Ore Insights Conference. London. Bhaskar, R. (1978). A realist theory of science. Sussex: The Harvester Press. Boisot, M., & McKelvey, B. (2011). Connectivity, extremes, and adaptation: A power-law perspective of organizational effectiveness. Journal of Management Inquiry, 20, 119–133. Brennan, R. (2006). Evolutionary economics and the markets-as-networks approach. Industrial Marketing Management, 35, 829–838. Brito, C. M. (2001). Towards an institutional theory of the dynamics of industrial networks. The Journal of Business and Industrial Marketing, 16(3), 150–166. Calder, B. J. (1977). Focus groups and the nature of qualitative marketing research. Journal of Marketing Research, 14, 353–364. Corley, K. G., & Gioia, D. A. (2004). Identitiy ambiguity and change in the wake of a corporate spin-off. Administrative Science Quarterly, 49, 173–208. Cunningham, W., & Sheth, J. N. (1983). From the editors. Journal of Marketing, 46(Fall), 7–8. Daft, R. L., & Weick, K. E. (1984). Toward a model of organizations as interpretation systems. Academy of Management Review, 9(2), 284–295. Dimaggio, P. (1988). Interest and agency in institutional theory. In L. Zucker (Ed.), Institutional patterns and culture (pp. 3–22) Cambridge, MA: Ballinger Publishing Company. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147–160. Dubois, A., & Gadde, L. E. (2002). Systematic combining: An abductive approach to case research. Journal of Business Research, 55, 553–560. Easton, G. (2000). Case research as a methodology for industrial networks: A realist apologia. In S. Ackroyd, & S. Fleetwood (Eds.), Realist perspectives on management and organisations (pp. 205–519) London: Routledge. Easton, G. (2002). Marketing: A critical realist approach. Journal of Business Research, 55(2), 103–109. Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14(4), 532–550. Eisenhardt, K. M., & Graebner, M. E. (2007). Theory building from cases: Opportunities and challenges. Academy of Management Journal, 50(1), 25–32.

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420 Eurometal (2010, October). Quarterly report on steel trade nr 3. Fern, E. F. (1982). The use of focus groups for idea generation: The effects of group size, acquaintanceship, and moderator on response quantity and quality. Journal of Marketing Research, 19, 1–13. Fern, E. (2001). Advanced focus group research. Thousand Oaks, CA: Sage Publications Ltd. Ferraro, F., Pfeffer, J., & Sutton, R. I. (2009). How and why theories matter: A comment on Felin and Foss (2009). Organization Science, 20(3), 669–675. Ford, D., Gadde, L., Håkansson, H., & Snehota, I. (2011). Managing business relationships (3rd ed.). Chichester: John Wiley & Sons Ltd. Gadde, L. E., Huemer, L., & Håkansson, H. (2003). Strategizing in industrial networks. Industrial Marketing Management, 32, 357–364. Geels, F. W. (2010). Ontologies, socio-technical transitions (to sustainability), and the multi-level perspective. Research Policy, 39, 495–510. Golfetto, F., Salle, R., Borghini, S., & Rinallo, D. (2007). Opening the network: Bridging the IMP tradition and other research perspectives. Industrial Marketing Management, 36, 844–848. Greenwood, R., & Hinings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21(4), 1022–1054. Greenwood, R., & Suddaby, R. (2006). Institutional entrepreneurship in mature fields: The big five accounting firms. Academy of Management Journal, 49(1), 27–48. Greer, C. R., & Lei, D. (2012). Collaborative innovation with customers: A review of the literature and suggestions for future research. International Journal of Management Reviews, 14, 63–84. Håkansson, H. (Ed.). (1982). International marketing and purchasing of industrial goods. John Wiley & SonsChichester. Håkansson, H. (1992). Evolution processes in industrial networks. In B. Axelsson, & G. Easton (Eds.), Industrial networks: A new view of reality. London: Routledge. Håkansson, H., & Ford, D. (2002). How should companies interact in business networks? Journal of Business Research, 55(7), 133–139. Halinen, A., Salmi, A., & Havila, V. (1999). From dyadic change to changing business networks: An analytical framework. Journal of Management Studies, 36(6), 779–794. Harrisson, D., & Easton, G. (2004). Temporally embedded case comparison in industrial marketing research. In S. Fleetwood, & S. Ackroyd (Eds.), Critical realist applications in organisation and management studies (pp. 194–233) : Routledge. Heikkinen, M. T., Mainela, T., Still, J., & Tahtinen, J. (2007). Roles for managing in mobile service development nets. Industrial Marketing Management, 36(7), 909–925. Hodgkinson, G. P. (1997). Cognitive inertia in a turbulent market: the case of UK residential estate agents. Journal of Management Studies, 34, 921–945. Hodgkinson, G. P., & Rousseau, D. M. (2009). Bridging the rigour–relevance gap in management research: It's already happening! Journal of Management Studies, 46(3), 534–546. Iederan, O. C., Curseu, P. L., Vermeulen, P. A., & Geurts, J. L. (2011). Cognitive representations of institutional change: Similarities and dissimilarities in the cognitive schema of entrepreneurs. Journal of Organizational Change Management, 24(1), 9–28. Järvensivu, T., & Törnroos, J. A. (2010). Developing case study research with moderate constructionism and abduction: Conceptualization and a practical illustration. Industrial Marketing Management, 39(1), 100–108. Johanson, J., & Mattson, L. G. (1987). Interorganizational relations in industrial systems — A network approach compared with the transaction-cost approach. International Studies of Management and Organization, 17(1), 33–48. Kondra, A. Z., & Hinings, C. R. (1998). Organizational diversity and change in institutional theory. Organization Studies, 19, 743–767. Kwan, K., & Tsang, E. W. (2001). Realism and constructivism in strategy research: A critical realist response to Mir and Watson. Strategic Management Journal, 22(12), 1163–1168. Langley, A. (1999). Strategies for theorizing from process data. Academy of Management Review, 24(4), 691–710. Lawson, T. (2009). On the nature and roles of formalism in economics: Response to Hodgson. In E. Fullbrook (Ed.), Ontology and economics: Tony Lawson and his critics (pp. 189–231) : Taylor & Francis. Leca, B., & Naccache, P. (2006). A critical realist approach to institutional entrepreneurship. Organization, 13(5), 627–651. Lewis, M. W., & Grimes, A. J. (1999). Metatriangulation: Building theory from multiple paradigms. Academy of Management Review, 24(4), 672–690. Loasby, B. J. (2000). Market institutions and economic evolution. Journal of Evolutionary Economics, 10, 297–309. Lundgren, A. (1992). Coordination and mobilisation processes in industrial networks. In B. Axelsson, & G. Easton (Eds.), Industrial networks: A new view of reality. London: Routledge. Macbeth, D. K. (2002). Emergent strategy in managing cooperative supply chain change. International Journal of Operations and Product Management, 22(7), 728–740. Maguire, S., Hardy, C., & Lawrence, T. B. (2004). Institutional entrepreneurship in emerging fields: HIV/AIDS treatment advocacy in Canada. Academy of Management Journal, 47(5), 657–679. Martinko, M. J., & Gardner, W. L. (1982). Learned helplessness: An alternative explanation for performance deficits. Academy of Management Review, 7(2), 195–204. Matthyssens, P., & Vandenbempt, K. (2003). Cognition-in-context: Reorienting research in business market strategy. The Journal of Business and Industrial Marketing, 18, 595–606. Mattson, L. (1987). Management of strategic change in a ‘markets-as-networks’ perspective. In A. Pettigrew (Ed.), The management of strategic change (pp. 234–256) London: Basil Blackwell.

419

McLoughlin, D., & Horan, C. (2002). Markets-as-networks: Notes on a unique understanding. Journal of Business Research, 55, 535–543. Meyer, A. D., Brooks, G. R., & Goes, J. B. (1990). Environmental jolts and industry revolutions: Organizational responses to discontinuous change. Strategic Management Journal, 11, 93–110. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. The American Journal of Sociology, 83, 340–363. Miller, K. D., & Tsang, E. W. (2010). Testing management theories: Critical realist philosophy and research methods. Strategic Management Journal, 32, 139–158. Moll, M. A. (2011). Status quo in stainless steel 2012 and beyond. Maastricht: Steel & Metals Markets Research. Möller, K. K., & Halinen, A. (1999). Business relationships and networks: Managerial challenge of network era. Industrial Marketing Management, 28, 413–427. Möller, K., & Rajala, A. (2007). Rise of strategic nets — New modes of value creation. Industrial Marketing Management, 36, 895–908. Möller, K. K., Rajala, A., & Svahn, S. (2005). Strategic business nets—Their type and management. Journal of Business Research, 58, 1274–1284. Möller, K. K., Svahn, S., Rajala, A., & Tuominen, M. (2002). Network management as a set of dynamic capabilities. Dijon (France): 18th IMP Conference. Morais, R. (2003). A Critical realist interpretation of managerial contacts in industrial internationalisation. Lugano (CH): 19th IMP Conference. Moran, P. (2005). Structural vs. relational embeddedness: Social capital and managerial performance. Strategic Management Journal, 26, 1129–1151. Morgan, D. L., & Krueger, R. A. (Eds.). (1998). The Focus Group Kit (1st–6th ed.). SageThousand Oaks, CA. Mouzas, S., Henneberg, S., & Naudé, P. (2008). Developing network insight. Industrial Marketing Management, 37, 167–180. Mutch, A. (2007). Reflexivity and the institutional entrepreneur: A historical exploration. Organization Studies, 28(7), 1123–1140. Ocasio, W. (1997). Towards an attention-based view of the firm. Strategic Management Journal, 18(SI), 187–206. Oliver, C. (1992). The antecedents of deinstitutionalization. Organization Studies, 13(4), 563–588. Polidoro, F. J., Ahuja, G., & Mitchell, W. (2011). When the social structure overshadows competitive incentives: The effect of network embeddedness on joint venture dissolution. Academy of Management Journal, 54(1), 205–233. Popper, K. R. (1965). The logic of scientific discovery. New York: Harper Torchbook. Powell, W. W. (1990). Neither market nor hierarchy: Network forms of organization. Research in Organizational Behavior, 12, 295–336. Powell, W. W., Koput, K. W., & Smith-Doerr, L. (1996). Interorganizational collaboration and the locus of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41, 116–145. Pratt, M. G. (2009). From the editors — For the lack of a boilerplate: Tips on writing up and reviewing qualitative research. Academy of Management Journal, 52(5), 856–862. Rajagopalan, N., & Spreitzer, G. M. (1997). Toward a theory of strategic change: A multi-lens perspective and integrative framework. Academy of Management Review, 22(1), 48–79. Rampersad, G., Quester, P., & Troshani, I. (2010). Managing innovation networks: Exploratory evidence from ICT, biotechnology and nanotechnology networks. Industrial Marketing Management, 39, 793–805. Ritter, T., Wilkinson, I. F., & Johnston, W. J. (2004). Managing in complex business networks. Industrial Marketing Management, 33, 175–183. Ritvala, T., & Salmi, A. (2008). Actor mobilization and institutional change around a common issue: The case of fighting heart disease in Finland. Uppsala, Sweden: 24th IMP Conference. Ritvala, T., & Salmi, A. (2010). Value-based network mobilization: A case study of modern environmental networkers. Industrial Marketing Management, 39(6), 898–907. Ritvala, T., & Salmi, A. (2011). Network mobilizers and target firms: The case of saving the Baltic Sea. Industrial Marketing Management, 40(6), 887–898. Ruis, A., van den Berg, K., & Meijaard, J. (2009). Klant- en conjunctuurafhankelijkheid in het MKB metaal. Zoetermeer: EIM. Ryan, A., Tahtinen, J., Vanharanta, M., & Mainela, T. (2009). Putting critical realism to work in business relationship research. Marseilles: 25th Annual IMP Conference. Ryan, A., Tähtinen, J., Vanharanta, M., & Mainela, T. (2012). Putting critical realism to work in the study of business relationship processes. Industrial Marketing Management, 41, 300–311. Sayer, A. (1992). Method in social science: A realist approach. London: Routledge. Scott, W. R. (1987). The adolescence of institutional theory. Administrative Science Quarterly, 32(4), 493–511. Scott, W. R. (2008). Institutions and organizations (3rd ed.). Thousand Oaks, CA: Sage Publications, Inc. Seo, M. -G., & Creed, W. E. (2002). Institutional contradictions, praxis, and institutional change: A dialectical perspective. Academy of Management Review, 27(2), 222–247. Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Belmont, CA: Wadsworth. Siggelkow, N. (2007). Persuasion with case studies. Academy of Management Journal, 50(1), 20–24. Snehota, I. (2003). Market-as-network; So what? Paper presented at the 19th IMP Conference. Lugano, CH. Söllner, A. (2002). Individual behavior in cases of institutional misfit: The dirty hands of IMP-group members. Dijon, France: 18th IMP Conference. Sutton-Brady, C. (2008). As time goes by: Examining the paradox of stability and change in business networks. Journal of Business Research, 61, 968–973. Tsai, W., & Echols, A. (2005). Niche and performance: The moderating role of network embeddedness. Strategic Management Journal, 26, 219–238.

420

P. Matthyssens et al. / Industrial Marketing Management 42 (2013) 405–420

Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42(1), 35–67. van de Ven, A. H., & Poole, M. S. (1995). Explaining development and change in organizations. Academy of Management Review, 20(3), 510–540. Vermeulen, P., Büch, R., & Greenwood, R. (2007). The impact of governmental policies in institutional fields: The case of innovation in the Dutch concrete industry. Organization Studies, 28(4), 515–540. Visconti, L. M. (2010). Ethnographic Case Study (ECS): Abductive modeling of ethnography and improving the relevance in business marketing research. Industrial Marketing Management, 39(1), 25–39. Welch, C. (2001). IMP and the two traditions of institutionalism. Oslo: 17th Annual IMP Conference. Williamson, O. E. (1991). Comparative economic organization: The analysis of discrete structural alternatives. Administrative Science Quarterly, 36(2), 269–296. World Steel Association (2010). World steel in figures 2010. (Brussels). Yin, R. (1984). Case study research. Beverly Hills, CA: Sage Publications. Zimmerman, M. A., & Zeitz, G. J. (2002). Beyond survival: Achieving new venture growth by building legitimacy. Academy of Management Review, 27(3), 414–431. Wouter Van Bockhaven is a PhD candidate in the Department of Management (Faculty of Applied Economics, University of Antwerp, Belgium). His research focuses on the role of institutional entrepreneurship and network mobilization from a business actor perspective on innovation strategy. Preliminary results have been presented at Strategic Management Society, Industrial Marketing and Purchasing group, Society for Industrial and Organizational Psychology and European Institute for Advanced Studies in Management conferences.

Paul Matthyssens is Professor of Strategic Management at the Department of Management (Faculty of Applied Economics, University of Antwerp, Belgium) and at Antwerp Management School (Belgium). His research focuses on market strategy in international and business-to-business markets. His work has been published in journals such as Long Range Planning, Industrial Marketing Management, Psychology & Marketing, Journal of Business & Industrial Marketing, Journal of International Marketing, International Marketing Review, Journal of International Management, Strategic Organization, Technovation and Journal of Purchasing & Supply Management. Koen Vandenbempt is Professor of Strategic Management at the Department of Management (Faculty of Applied Economics, University of Antwerp, Belgium) and the Antwerp Management School (Belgium). His research focuses on the process of strategic reorientation efforts within industrial companies. His work has been published in journals such as Human Relations, Industrial Marketing Management, Journal of Business Research, Journal of Business & Industrial Marketing, Advances in Business Marketing & Purchasing and Advances in Applied Business Strategy.