اثر مکانیسم های معتبر  هماهنگی بر بازارگرایی در مشارکت کانال های نامتقارن

An important outcome of a firm's interorganisational relationships is the attainment of resources and capabilities required to attain a competitive advantage. Although market orientation has been considered one of those critical resources, there is little understanding regarding how it is created and nurtured in business relationships. In the context of asymmetrical channel partnerships dominated by a downstream leader, in this paper we present a model where the upstream follower's market orientation is promoted as a consequence of the leader's use of authoritative mechanisms of coordination. Some hypotheses are presented and empirically tested in a sample of agricultural cooperatives that maintain channel partnerships with second-order marketing cooperatives. The results indicate that formalization, participation, input control, and behaviour control are coordination mechanisms that lead to improvements in market orientation. Implications for interorganisational theory and managerial activities are discussed at the end of the paper.

The end of the 20th century has witnessed a growing competitive pressure in most industries, forcing companies to seek more creative and flexible means for meeting competition. Many firms have faced these challenges participating in interorganisational alliances to achieve the critical resources they need to obtain competitive advantage (Dyer & Singh, 1998). One of those critical resources is market orientation, which has been considered as a causal antecedent of a firm's overall perceived performance (Jaworski & Kohli, 1993), profitability Hooley et al., 2000, Narver & Slater, 1990 and Pelham, 1999, market performance (i.e., sales growth, new product success, market share, and marketing and sales effectiveness; Hooley et al., 2000, Matsuno & Mentzer, 2000, Pelham, 1999 and Slater & Narver, 1994), superior customer value (Slater & Narver, 2000), and sustainable competitive advantage (Hunt & Morgan, 1995). This justifies the relevance of studying how companies can attain higher levels of market orientation. Some internal factors have been postulated as antecedents of a firm market orientation (see for example Kohli & Jaworski, 1990). However, to the best of our knowledge, only Siguaw, Simpson, and Baker (1998) and Baker, Simpson, and Siguaw (1999) empirically analyse market orientation in business relationships. Using reference group theory, Siguaw et al. found that the distributor's market orientation is positively influenced by its supplier's market orientation—i.e., the distributor would be more market oriented “…to comply with its reference group's norms or in meeting some perceived standard” (p. 101). Baker et al. unveiled market orientation as an antecedent of the relational dimensions of trust, cooperative norms, and commitment. The main purpose of this research is to examine how a firm's market orientation is improved as a result of its channel relationships. The revision of marketing literature does not show any study that empirically analyses the effects of the interorganisational processes and structure on the partners' market orientation. This lack of empirical research is surprising given the theoretical consensus on the potential utility of interorganisational relationships to obtain market intelligence (Kohli & Jaworski, 1990), to offer superior customer value (Slater & Narver, 2000), and to generate relational quasi-rents (Dyer & Singh, 1998). Moreover, researchers in the area of relational marketing have stressed the importance of relationships to obtain and develop the critical resources and capabilities of the company Anderson et al., 1994 and Hunt, 1997. Therefore, the model presented and tested in this article will contribute towards filling this gap in the literature of interorganisational relationships. This is addressed within the particular setting of channel partnerships in the commercialisation of fresh fruit and vegetables. During the last decades, channels of distribution for these products in the European Union have witnessed a growing concentration in the retailing sector, accompanied by centralized buying decision-making (Brookes, 1995). In this context, the distance between farmers and consumers hindered farmers from accessing accurate and immediate market information, delaying their adaptation to the fast pace of changes in retailing and consumers demands (Dijk & Mackel, 1994), and causing asymmetry of information that could be opportunistically used by retailers (Wathne & Heide, 2000). The creation of cooperatives that integrate farmers from the same geographical area, hereafter named first-order marketing cooperatives (1OMC) was, in most of the European countries, one of the initial reactions in order to reduce both production costs (Sargent, 1991) and transaction costs (Ollila, 1994). 1OMCs are responsible for commercialising their farmers' production, although other services can also be offered (i.e., supplies, technical support, etc.). However, even though they represent a first movement toward higher levels of integration in the agricultural sector, their local orientation limits the volume and number of products they can offer to their clients. Compared to their larger competitors, 1OMCs' competitiveness is hindered by their inability to provide the service required by the distribution chains at a reasonable cost (e.g., assortment, larger temporal coverage of the campaign, strict reliability and stability of the deliveries). As a result, the cooperative sector responded with higher levels of integration (Sargent, 1991). Whereas in some countries (e.g., Germany) merging has been the route chosen by 1OMCs to become larger, in others (e.g., Spain) second-order marketing cooperatives (2OMC) were constituted to commercialise all, or a portion, of the production of various first-order cooperatives. Considered as cooperatives of cooperatives, the 2OMCs are marketing specialists and, by means of their selling and distribution network, they assume the function of serving the market demands, with the retailing chains as their main clients. Consequently, they are dedicated to commercialising their first-order partners' production by establishing, developing, and maintaining relationships with the distribution chains in the national and international markets. The law establishes some conditions that the relationship between first- and second-order marketing cooperatives must fulfil. For example: (1) A 2OMC must be founded for at least two 1OMCs, which are its owner-user; (2) one partner cannot own more than 30% of the 2OMC's capital; (3) at least two-thirds of the 2OMC's Board of Directors must be elected by the General Assembly among the 1OMCs' Board of Directors; (4) all the 1OMCs have a vote in the 2OMC's General Assembly, although the statutes can determine a weighted right of participation according to the relative volume of business of each partner. Moreover, these statutes determine the minimum percentage that the 1OMC must commercialise through the 2OMC and, when this is lower than 100%, the 1OMC is allowed to use alternative channels. These legal characteristics and the fact that each company is specialized in vertically differentiated functions, shares compatible goals, works for mutual benefit, and has a high level of interdependence, allow us to qualify this type of relationship as a channel partnership (Mohr & Spekman, 1994). In channel relationships, authoritative mechanisms of control involve “one party in the relationship using its position or power to control the activities of the other party (Weitz & Jap, 1995; p. 306)”. In our interorganisational setting, the 2OMCs have power as a result of their larger dimension (reward and coercive power), their proximity to the market that provides them with more and better information (information power), their marketing expertise and responsibility to the clients (expertise power), and the contractual conditions to which the 1OMCs must agree (legitimate power). This makes the relationship asymmetrical, with the 2OMC as the leader (source) that controls those 1OMC's (target) activities that are important to be able to offer value to the partnership's clients (i.e., the retailing chains). Research into all the possible interorganisational antecedents of market orientation is beyond the scope of this study. Moreover, this paper does not attempt to develop a complete theory of market orientation in interorganisational relationships. We contribute to the marketing literature by (1) providing new insights on how authoritative mechanisms of coordination benefit the target's market orientation in asymmetrical interorganisational relationships; (2) considering simultaneously the effects of multiple types of authoritative coordination mechanisms; and (3) analysing those aspects in the relational setting of channel partnerships. To address these objectives, we begin by providing the conceptual framework in which a set of research hypotheses is formulated. We then discuss the research design, including survey context, sampling procedure and data collection, and measurement properties. The empirical results of the hypotheses testing are then reported. Finally, we discuss results, its theoretical and managerial implications, and the limitations.

The proposed structural model is specified from the hypothesized relationships in Fig. 1, discussed in the text as H1, H2, H3 and H4. The path linking output control and market orientation is estimated in order to obtain a complete picture of the effects of all the authoritative mechanisms of coordination. Conventional maximum likelihood estimation techniques were used to test the model (Jöreskog & Sörbom, 1996). The fit of the model is satisfactory (χ2=375.48, df=233, p<0.001; GFI=0.84; CFI=0.92; RMSEA=0.068; TLI=0.91; IFI=0.93), thereby suggesting that the nomological network of relationships fits our data—another indicator of support for the validity of these scales To provide greater confidence in our model specification with market orientation treated as a second-order factor, we test our theoretical model (MT) against alternative model specifications (MA). Anderson and Gerbing (1988) recommend this procedure and suggest the use of a chi-square difference test (CDT) to test the null hypothesis: MT−MA=0. Compared with a less parsimonious MA, a nonsignificant CDT would lead to the acceptance of the more parsimonious MT. Table 4 reports a nonsignificant change in chi-square between our model and the CFA, leading us to consider MT as a better specification. We also contrast MT with an alternative model where market orientation is not specified as a second-order construct. Again, the better fit shown by the alternative model is not significant, which leads us to stay with the more parsimonious MT. Finally, in our theoretical model we propose a differentiated effect of input, behaviour and output control. However, from a theoretical point of view, a second-order construct of unilateral control could be conceivable (Heide, 1994). Therefore, we compare MT with an alternative model where both market orientation and unilateral control are considered second-order constructs (two second-order factors model). In this case we are comparing MT with a more parsimonious model. The CDT presents a p<0.001 which allows us to consider the alternative model's fit as significantly worse. As a result of the different model comparisons, we can be confident about our theoretical model specification.In terms of our hypotheses (Table 5), the finding for H1 (formalization→market orientation; γ11=0.19, p<0.10) suggests that formal rules are interpreted by the 1OMCs in an enabling way—i.e., as a valuable resource that improves their market orientation. In our sample, there are reasons to believe that formalization is actually being used in an enabling way rather than exclusively as a tool to reduce the target's possibilities of shirking. This is due to the fact that the three forces cited by Adler and Borys (1996) as promoters of the enabling logic are present in our empirical setting: 1. “The enabling logic has considerable and growing legitimacy in the broader culture” (p. 82). In the cooperative sector, where the joint cooperation is a normative rule, this is even more obvious. 2. “The advantages of the enabling logic can be bolstered by the demands of the task environment and notably by the intensification of competitive pressure” (p. 83). The market description presented in the introduction of this article proves the applicability of this principle to our empirical context. 3. “Increasing complexity often increases the relative advantage of the enabling form” (p. 83). The retailers' demands for better products, more controlled production and manipulation programs, and faultless service stimulate 2OMCs to implement formalization in an enabling rather than in a coercive logic.Continuing with the effects of the bureaucratic structure, participation is found to be positively associated with market orientation (γ12=0.39, p<0.01). This confirms that participation creates an adequate environment that enhances those behaviours that lead to a better orientation to the market. H3 is also supported, with input control (training) significantly associated with an increase in the target's market orientation (γ13=0.33, p<0.05). This suggests that the postulated learning and motivational effects of training are present in our relational setting. Also, behaviour control exerts a positive and significant impact on target's market orientation (γ14=0.27, p<0.10), confirming H4. We did not present any hypothesis about the effects of output control on market orientation. The results obtained for the parameter γ15 (−0.04, p>0.10) suggest that the exercise of this type of control does not influence the target's market orientation, in accordance with the discussion presented in the theoretical section. Our sample was composed of 161 1OMCs that were associated to 2OMCs as described in Table 6. This may cause concerns about the intra-2OMC variation in the variables. The authors' knowledge of the sector led us to be confident about the fact that the differences among the constructs were due to real-world differences, rather than to just perceptual noise. However, to be more secure about this, we phoned three senior executives of 2OMC who confirmed our first impression. Although general guidelines exist, each relationship in the same 2OMC is governed in a customized fashion. Differences are based on the type of product (fruits, citrus, and vegetables are characterised by different temporal and commercial specifications), the dimension and capabilities of the 1OMC, its commitment to the partnership, etc. For example: 1.Those 1OMCs that participate in programs of “premium quality” have much higher levels of authoritative governance than others that are also integrated in the same 2OMC, but just producing standard quality. 2.The 1OMCs that only produce fruits (e.g., apricot, grape, peach, etc.) have a relationship with the 2OMC with larger time lags than the cooperatives that produce vegetables. As a consequence the development of relational behaviours is more complicated with those 1OMCs that only produce fruits. 3.The 1OMCs that produce products very important for the 2OMC in terms of volume (e.g., orange) or value (e.g., lettuce, melon) are generally subjected to higher levels of authoritative governance than those that produce less important products for the 2OMC (e.g., cucumber, onion).Continuing with the effects of the bureaucratic structure, participation is found to be positively associated with market orientation (γ12=0.39, p<0.01). This confirms that participation creates an adequate environment that enhances those behaviours that lead to a better orientation to the market. H3 is also supported, with input control (training) significantly associated with an increase in the target's market orientation (γ13=0.33, p<0.05). This suggests that the postulated learning and motivational effects of training are present in our relational setting. Also, behaviour control exerts a positive and significant impact on target's market orientation (γ14=0.27, p<0.10), confirming H4. We did not present any hypothesis about the effects of output control on market orientation. The results obtained for the parameter γ15 (−0.04, p>0.10) suggest that the exercise of this type of control does not influence the target's market orientation, in accordance with the discussion presented in the theoretical section. Our sample was composed of 161 1OMCs that were associated to 2OMCs as described in Table 6. This may cause concerns about the intra-2OMC variation in the variables. The authors' knowledge of the sector led us to be confident about the fact that the differences among the constructs were due to real-world differences, rather than to just perceptual noise. However, to be more secure about this, we phoned three senior executives of 2OMC who confirmed our first impression. Although general guidelines exist, each relationship in the same 2OMC is governed in a customized fashion. Differences are based on the type of product (fruits, citrus, and vegetables are characterised by different temporal and commercial specifications), the dimension and capabilities of the 1OMC, its commitment to the partnership, etc. For example: 1. Those 1OMCs that participate in programs of “premium quality” have much higher levels of authoritative governance than others that are also integrated in the same 2OMC, but just producing standard quality. 2. The 1OMCs that only produce fruits (e.g., apricot, grape, peach, etc.) have a relationship with the 2OMC with larger time lags than the cooperatives that produce vegetables. As a consequence the development of relational behaviours is more complicated with those 1OMCs that only produce fruits. 3. The 1OMCs that produce products very important for the 2OMC in terms of volume (e.g., orange) or value (e.g., lettuce, melon) are generally subjected to higher levels of authoritative governance than those that produce less important products for the 2OMC (e.g., cucumber, onion). To test whether our theoretical constructs have the posited effects above and beyond that captured by the 2OMC's name or the type of product, we performed an OLS regression introducing nine dummy variables. Considering the list of 2OMCs reported in Table 6, six variables were used to introduce the 2OMC's name in the regression. In this case, the 12 cooperatives in the category “others” are the base case. To control the effect of the type of product, a value of 1 is given to Fruit when the 1OMC exclusively produces fruits (20.8%), Vegetables when the 1OMC exclusively produces vegetables (27.7%), and Citrus when the 1OMC exclusively produces these type of products (19.5%). Those cooperatives that commercialise a mix of those categories are considered the base case to control the effect of the type of product. The results presented in Table 7 show that none of the dummy variables have a significant effect on market orientation, whereas the positive and significant effects of formalization, participation, input and behaviour control are maintained.