Knowledge and Ignorance of Connections between Relationships

In this paper we review extant studies and report new empirical insights regarding connected relationships within a focal firm’s network horizon. Furthermore, we discuss possible reasons for how and why knowledge and ignorance of different types of connections are dispersed in industrial structures. The Industrial Network Approach and the concept of connected relationships “One critical specification in all approaches developed to analyze managerial problems involves the interface between the firm and its environment” (Anderson, Håkansson and Johanson 1994, p. 2). The Industrial Network Approach as represented by e.g. Axelsson and Easton (1992), Håkansson and Snehota (1995), Ford (1997), and Dubois (1998) focuses explicitly on the interface between the firms and its environment, and it does so by having business relationships as the focal unit of analysis. However, within the Industrial Network Approach, as the name suggests, relationships are not the only unit of analysis paid attention to. The network structures of which relationships form part are an important unit of analysis. It may be argued that the most basic assumption within the Industrial Network Approach is that of business relationships being connected thereby giving rise to the type of industrial structure conceptualised as industrial networks. The issue of connected relationships within the Industrial Network Approach seems to have been inspired by research within the field of sociology. Within Social Exchange Theory, Cook and Emerson (1978) discuss exchange networks as a set of two or more connected exchange relations. They define the concept of connection in the following way: “Two exchange relations are connected to the degree that exchange in one relation is contingent upon exchange (or nonexchange) in the other relation. The connection is positive if exchange in one is contingent upon exchange in the other. The connection is negative if exchange in one is contingent upon nonexchange in the other” (Cook and Emerson 1978, p. 725). Based on laboratory experiments regarding power, equity and commitment in exchange networks, Cook and Emerson (1978, p. 737) conclude “Needed in place of a theory tied to perfectly competitive (i.e. uncommitted) markets is a theory of network structures tied together by repetitive exchange with specified partners.” Within Economic Sociology, Granovetter (1992, p. 33) discusses the same issue in terms of embeddedness, stating that “‘embeddedness’ refers to the fact that economic action and outcomes, like all social action and outcomes, are affected by actors’ dyadic (pairwise) relations and by the structure of the overall network of relations. ... The structural aspect is especially crucial to keep in mind because it is easy to slip into ‘dyadic atomization’, a type of reductionism.” Similarly, Grabher (1993) have seconded this view pointing out the problem of ‘dyadic reductionism’, i.e. treating a dyad as if it existed separate from its context. Hence, we may say that within the Industrial Network Approach, ‘dyadic reductionism’ is avoided by conceptualising relationships as being connected. However, even though the concept of connection is important within the Industrial Network Approach, only a few studies have examined ‘dyadic’ business relationship in the context of other relationships. These studies have shown that firms (and relationships) do not exist in isolation, but rather that they are connected to a network context through direct and indirect relationships (Laage-Hellman 1989; Axelsson and Easton 1992; Blankenburg 1992; Anderson, Håkansson and Johanson 1994; Havila 1996; Pedersen 1996; Ritter 1999 and Holmen and Pedersen 2000). This implies that in addition to characterising individual relationships, the studies have shown that single relationships are also affected by and affect developments in other relationships. Thus, a single business relationship exists both in itself and, at the same time, is embedded in a context, through its connections to other relationships. The purpose of this paper is to (a) review extant studies regarding connections between relationships and (b) report new insights into a focal firm’s network horizon. Furthermore, we (c) discuss possible reasons for how and why knowledge and ignorance of different types of connections are dispersed in industrial structures, and (d) suggest how further research can be carried out into this phenomenon What do we know about connections within the Industrial Network Approach? Most empirical studies of connections within the Industrial Network Approach have started out by exploring which types of actors represent potential third parties to a relationship, i.e. affect and/or are affected by a focal business relationship. In her excellent licentiate thesis, Blankenburg (1992) studied 85 supplier-customer relationships in the IMP2 project. She uses the following classification of potential third parties: (1) supplier’s suppliers, (2) competitive suppliers, (3) supplementary suppliers, (4) other customers, (5) customer’s customers, (6) other units of the customer’s firm, (7) other units of the supplier’s firm, (8) bank or financial organisations, (9) law firms or legal organisations, (10) consultants or research institutes, (11) trade unions or other social bodies, (12) governmental agencies, (13) international organisations, and (14) other relevant organisations connected to a focal business relationship. Anderson, Håkansson and Johansson (1994) also focused on a focal dyadic relationship and potential connected third parties. They use the following categories: (1) supplier’s supplier, (2) other supplier unit, (3) competing supplier, (4) supplementary supplier, (5) other customers, (6) customer’s customer, (7) other units in focal customer firm, (8) other units in the focal supplier firm, (9) other ancillary firms and (10) third parties in common. Hence, Anderson, Håkansson and Johanson (1994) use fewer third-party categories than Blankenburg (1992) but, to a large extent, focus on the same types of third parties. We may conclude that some insight has been gained into connected relationships and third parties. Firstly, we now know that more or less all types of organisations can be a third party to a connected relationship! Hence, relationships to all types of third parties may be connected to a focal relationship. Secondly, we know a little about how many connected relationships firms are aware of on average. The cross-sectional study by Blankenburg (1992) is the only one which has focused on this issue. In relation to 85 supplier-customer relationships, she identified 312 connected relationships in total 101 connected via the supplier and 211 connected via the customer. This implies that, on average, firms are aware of 3.7 connected relationships/third parties, 1.2 connected relationships via the supplier and 2.5 connected relationships via the customer. It is, however, important to be aware that no specific counterparts were identified in the study by Blankenburg (1992). Instead, the respondents were asked to what extent any of the different third party categories presented above affected the two parties in the focal relationship, as seen from the supplier; if for example four customers of the customer affected the focal relationship, they would only count as one. In addition to these two studies, research into connected relationships has been done within the fields of technical development, distribution and purchasing, respectively. Contrary to the above-mentioned studies, these studies employ a case methodology focusing on a few settings where specific connected relationships and third parties were studied and analysed. Laage-Hellman (1989) studied technological exchange between different firms and research units in industrial networks, and introduced the concept of triad as a first step towards a network analysis. With this as the point of departure, he analysed six empirical triad situations. In each of these he identified three specific counterparts in order to discuss how different counterparts were involved in technological development together with the focal firm, and how the focal firm’s relationships to these counterparts were connected. On the basis of this analysis, he introduced two types of ‘connection’ concepts: activity-based connections, i.e. structural dependencies which exists among different activities (and the resources used in these activities), and actor-based connections, i.e., dependencies caused by actors’ subjective acting within the given structure. Havila (1996) studied the changing role of intermediating actors in eleven non-overlapping international business-relationship triads, where each of the actors involved in the triads were identified. The main purpose of the study was to conceptualise different business-relationship triads, and how these change over time. Havila (1996) operated with two units of analysis: (1) the business-relationships triad, and (2) the role of the intermediating actor, and she related the changing nature of the second unit of analysis to changes in the first unit of analysis. On the basis of this, she distinguished between two types of triads: serial triads, i.e. where the supplier and the customer are connected through the intermediating actor and unitary triads, i.e. where all the three actors have relationships with each other. Havila (1996) did not explicitly discuss the concept of ‘connection’, but indirectly her discussion of triads contributed to the conceptualisation of connection within the Industrial Network Approach. Pedersen (1996) studied the development of supplier relationships and how these relationships were affected by connected relationships. In her study, she analysed three focal relationships each involving a focal firm and one of its most important suppliers. Furthermore, she studied how the development of these relationships was affected by different connected relationships to the focal firm or to the three suppliers (but always studied from the focal firm). On the basis of this Pedersen (1996) distinguished between effects of connected relationships in actor bonds, activity links and resource ties, as these are defined by Håkansson and Snehota (1995). In all the three case studies above, the authors identify specific third parties as opposed to categories of third parties as distinguished in the cross-sectional study by Blankenburg (1992). Furthermore, both in Laage-Hellman (1989) and Pedersen (1996) the content of what is affected in the focal relationships is identified and analysed. However, we still know little about, (1) the types of third parties and connected relationships firms are aware of as well as (2) the types of third parties and connected relationships firms are not aware of. Furthermore, we know little about (3) the type and depth of knowledge firms have regarding third parties and connected relationships, since the studies presented above either focus on the quantitative aspects (how many connected relationships), or only draw attention to some of the involved firms direct and connected relationships. In other words, we still have very limited insight into the network horizons of firms as well as the network contexts of relationships, and how relationships connect to form network structures. The network horizon of a firm is defined by Anderson, Håkansson and Johanson (1994, p. 4) as “how extended an actor’s view of the network is”. As they argue (p. 4), “actors have bounded knowledge about the networks in which they are engaged [..] . This is due to not only the network extending farther away from the actor but also the basic invisibility of network relationships and connections. The network setting extends without limits though connected relationships, making any business network boundary arbitrary.” In addition to the concept of network horizon, Anderson, Håkansson and Johanson (1994, p. 4) discuss the concept of network context, which they define as: “The part of the network within the horizon that the actor considers relevant is the actor’s network context.” This view of network context is based on Håkansson and Snehota (1989, p. 192) claiming that “it therefore seems useful to adopt the concepts of ‘context’ of an organization rather than its environment, when we want to refer to the entities that are related to the organization”. Furthermore, is seems that the concept of network context of a focal relationship in which two firms are involved comprises all those other organisations which one or both firms are directly related to. This is depicted in figure 1. Figure 1: Connected relations for firms in a dyadic relationship (Anderson, Håkansson and Johanson 1994, p. 3) We may note that as opposed to Håkansson and Snehota (1989) who include only direct relationships in a firm’s network context, Anderson, Håkansson and Johanson (1994) to some extent include all actors which the firm ‘considers relevant’. As opposed to the concept of network context, the concept of network horizon seems to refer to all those relationships in which a single firm is directly involved as well as all those other firms and relationships which the single firm is aware of. This implies that also direct counterparts of the firm’s direct counterparts may form part of a single firm’s network horizon. Due to the lack of empirical evidence of connections between relationships, it is not surprising that little theorising has yet been carried out in relation to connections among relationships, network contexts as well as network horizons. However, we contend that researchers within the Industrial Network Approach should be interested in exploring and explaining such issues, as they concern the sine qua non of industrial networks. Hence, we may conclude that in spite of the importance of connections between relationships within the Industrial Network Approach, little attention has been paid to investigating and explaining connections, network contexts and network horizons. Hence, we suggest that further research is needed into these matters. Focal Relationship Other Supplier Unit Third Parties in Common Other Units in Focal Customer Firm Other Ancillary Firms