This paper reports an investigation into the antecedents of commitment in non-Western industrial marketing relationships. The authors draw the antecedents from extant literature and posit that commitment is related to trust (integrity and reliability), communication quality, conflict, and similarity (social, ethnic, and economic). It is further argued that trust mediates the effects of communication, conflict, and similarity on commitment. As an extension, the authors examine the moderating effects of normative contracts (an implicit understanding of roles and responsibilities) on the construct interrelationships. The hypotheses are tested using data collected from approximately 150 industrial marketing relationships sampled from overseas Chinese firms. The results generally support the authors' framework; however, the mediating hypotheses are not supported. There is evidence of systematic differences in the effects of the studied antecedents on commitment and trust. Furthermore, a multigroup analysis provides evidence of significant moderating effects due to contracting mode. The study provides new insights into the theory and practice of industrial marketing.
Relationship commitment is a fundamental prerequisite for successful industrial marketing relationships Dwyer et al., 1987 and Gundlach & Murphy, 1993. Commitment enhances the effectiveness, productivity, and efficiency of relational exchanges (Morgan & Hunt, 1994). Commitment impacts on channel partner satisfaction and the financial performance of channel relationships (Mohr & Spekman, 1994). Committed parties focus on long-term outcomes and attempt to maximize profits over a series of transactions (Ganesan, 1994). A long-term outlook and focus on future goals is central to the concept of commitment Gundlach & Murphy, 1993 and Moorman et al., 1992. It is this focus on longer-term outcomes that makes investments in specialized assets possible Rindfleisch & Heide, 1997 and Williamson, 1985. It is clear that commitment is an important construct. Several recent industrial marketing studies have examined its antecedents in North American and European settings de Ruyter et al., 2001 and Goodman & Dion, 2001. However, few systematic attempts to model the antecedents of commitment in non-Western industrial marketing contexts are reported in the literature.
To address this important gap, this paper presents a model of the potential antecedents of commitment using a sample of buyer–supplier relationships from a non-Western setting. Trust (integrity and reliability), communication quality, conflict, and similarity (social, ethnic, and economic) are posited as predictors of commitment. It is further argued that trust mediates the effects of trust, communication, and similarity on commitment. As an extension of the basic model, the moderating affects of normative contracts on the construct interrelationships are examined. Normative contracts exist when the parties to a relationship have an implicit understanding of their roles and responsibilities (Lusch & Brown, 1996). The model is tested using data sourced from 152 buyer–supplier relationships sampled from overseas Chinese businesses. This is an appropriate context for study because a substantial proportion of economic activity occurs within the overseas Chinese business environment (Xin & Pearce, 1996). A rationale for the constructs and their hypothesized interrelationships are presented next (see Fig. 1). This is followed by a description of the research methodology and presentation of the results. The paper concludes with a discussion of the findings and their implications for the theory and practice of industrial marketing.
This study modeled some possible antecedents of commitment in a non-Western industrial marketing setting. Trust (integrity and reliability), communication quality, conflict, and similarity (social, ethnic, and economic) were modeled as predictors of relationship commitment. It was further argued that trust would mediate the effects of communication, conflict, and similarity on commitment. The empirical results are generally supportive of the conceptual framework; however, the mediating hypothesis is not supported. There is evidence of significant direct effects from communication and conflict to commitment. Furthermore, there is evidence of systematic differences in the effects of communication, conflict, and similarity on commitment and trust. Communication and conflict appear to be more important predictors of commitment than trust and similarity. Finally, the multigroup analysis indicates significant differences in the structural model estimates across high and low levels of normative contracting.
The estimates from the initial model supported the positive trust→commitment path. However, the relationship was attenuated to the point of nonsignificance in the full model. This result is inconsistent with a premise of relationship marketing theory and past research. It was argued that trust is a crucial prerequisite for commitment (Morgan & Hunt, 1994). Essentially, commitment entails vulnerability and exchange participants will seek only trustworthy partners. Past research supports the positive effect of trust on relationship commitment de Ruyter et al., 2001, Goodman & Dion, 2001 and Morgan & Hunt, 1994. Perhaps, the moderating effects of normative contracts provide a potential explanation for lack of consistent support for this relationship. The trust→commitment relationship tended towards significance in the low normative contracts subsample (P<.10), although the path was not significant in the high normative contracts group. It is possible to speculate that normative contracts act as a substitute for trust (cf. Lusch & Brown, 1996). That is, a mutual understanding of each other's roles can lessen the importance of making trust-based judgments about one's partner.
An unexpected result was the direct communication→commitment relationship. The effect of communication on commitment was strongly positive in the total sample and both subsamples. This finding indicates that communication is a direct corridor to commitment. There is some evidence of direct effects in past research. de Ruyter et al. (2001) and Morgan and Hunt (1994) posit communication→trust→commitment relationships. Morgan and Hunt's results are indicative of mediation. However, de Ruyter et al. find a direct effect from relationship characteristics (i.e., communication quality) to affective commitment. Next, there was consistent support for the communication→trust relationship. The effect of communication on trust was strongly positive and significant in the total sample and both subsamples. This finding reinforces the position that communication is key to building successful relational exchange (Mohr et al., 1996). It also extends past research by establishing the robustness of the communication→trust path in a non-Western setting.
Another unexpected result was the direct conflict→commitment path. It was argued that the negative effect of conflict on commitment would be mediated by trust. The direct effect of conflict was negative in the total sample and strongly negative in the low normative contracting subsample. However, the direct effect of conflict on commitment was nonsignificant in the high normative contracting group. This result is indicative of the moderating effect of normative contracting on the conflict→commitment relationship. The potential for conflict is present in many marketing relationships Dwyer et al., 1987 and Gaski, 1984. However, normative contracting appears to be an effective mechanism for lessening the destructive consequences of conflict. As predicted, the negative effect of conflict on trust was supported in the total sample and the low normative contacting subsample, although the negative conflict→trust path was attenuated to the point of nonsignificance in the high normative contracting group. This finding extends past research by isolating the conditions under which conflict erodes trust (cf. Anderson & Narus, 1990 and Morgan & Hunt, 1994). Conflict undermines trust in relationships where the parties lack a mutual understanding of their respective roles. Stated differently, the results indicate that normative contracts are a useful mechanism for lessening the destructive effects of conflict in exchange relationships.
As expected, the direct effect of similarity on commitment was nonsignificant in the total sample and the high normative contract group. However, the similarity→commitment relationship was negative and significant in the low normative contract subsample. The negative sign is unexpected and somewhat difficult to explain. It can be argued that dissimilarity would reduce an exchange partner's willingness to commitment to a relationship. The empirical results suggest that a different mechanism is at work. Normative contracts are a powerful tool for overcoming the apparent reluctance of parties with similar characteristics to form relationships. Finally, support for the similarity→trust relationship was mixed. The path was nonsignificant in the initial model but it was positive and significant in the full model. Normative contracting also moderated the effect of similarity on trust. The similarity→trust relationship was positive in the low normative contract group, but it was nonsignificant in the high normative contract subsample. Thus, it appears that normative contracting can act as substitutes for the positive effects of shared social, ethnic, and economic characteristics. Past research on similarity has produced mixed results Crosby et al., 1990 and Doney & Cannon, 1997. Again, it may be that the moderating influence of normative contracting provides a potential explanation for these inconsistencies.
