Direct and indirect effects of coercive power in the commitment -trust theory of relationship marketing

This study examines the effects of coercive power in relational exchange. Specifically, the direct and indirect effects of coercive power are examined with respect to the key variables in the Commitment-Trust Theory of Relationship Marketing (Morgan and Hunt, 1994). These key variables include trust, relationship commitment and cooperation. A significant negative effect is found between coercive power and trust. Significant indirect effects of coercive power are also found for cooperation through trust and for relationship commitment through trust. A method is also introduced for examining extreme cases of the independent variable using structural equation modeling. The sample is split into cases based on high and low coercive power. Then separate structural models are examined for each case. This study shows that in the high coercive power case there is no significant association between trust and cooperation nor between relationship commitment and cooperation. This suggests that there are other variables driving relationships besides the trust-based relationships suggested by Morgan and Hunt (1994). There are both trust-based and power-based relationships. However, coercive power alone is not found to be a driver of relationship outcomes. In the case of high coercive power, the Commitment-Trust Theory fails to generalize. The concept of examining extreme values for split case sub-samples introduced in this paper may help to improve the generalizability of structural equation models in future studies.