Protecting relational assets: a pre and post field study of a horizontal business combination

As evidence of this point, the retailer members of the cooperative under study are free to purchase from any other sources of supply that they wish. The dividends that they receive from the cooperative, however, are directly tied to their purchases from the cooperative.

The contract breaches that we study here are the breaches of contract between the retailer-owner-principals of the cooperative and the management-agents who actually operate the cooperative. The agents’ failure to generate the benefits promised pre-business combination is a breach of the normative principal-agent contract with respect to the combination.

Critics argue contract breach cannot be distinguished from the related constructs of: (1) unmet expectations, (2) inequity, and (3) dissatisfaction (Guest 1998). Expectations are generalized beliefs about what parties can expect from their relationships with each other, while normative contracts pertain to specific beliefs formed by the promises that the parties make to one another (cf. Robinson 1996; Robinson and Rousseau 1994). Turnley and Feldman (2000) empirically demonstrated the discriminant validity of the contract breach and unmet expectations constructs (n.b., Turnley and Feldman (2000) purportedly investigate contract violation. However, their operationalization of that construct is more consistent with our conceptual definition of contract breach). Contract breach and inequity are also closely intertwined (Pate et al. 2003). However, “[w]hereas equity is evaluated by considering ratios of inputs and outcomes, breach of contract is determined by comparing inputs and outcomes relative to what was promised” (Morrison and Robinson 1997, p. 242). Thus, unkept promises distinguish between perceptions of inequity and perceptions of contract breach. Finally, (dis)satisfaction, as we view it here, is a channel member’s affective response to an overall assessment of its relationship with an exchange partner (Geyskens et al. 1999). Turnley and Feldman (2000) have empirically shown contract breach and dissatisfaction to be distinct constructs.

In-role performance is defined as how well an employee performs those activities described in its job description. Extra-role performance (often termed citizenship behaviors) is comprised of those activities an employee performs beyond his/her job description.

Our focus here is upon channel member dependence rather than channel relationship interdependence. We do this because Kim and Hsieh (2003, p. 109) argue that “[d]istributor and supplier dependence must be decoupled and managed separately.”

This second cooperative resulted from a merger of two retail cooperative chains prior to this investigation.

Peter et al. (1993) argue that difference scores should generally be avoided, largely because they “are typically less reliable than other measures” (p. 661). However, Willett (1988, p. 368) counters that, because reliability coefficients of difference measures are often misinterpreted, such measures are not necessarily unreliable. Following Willett’s (1988) recommendation, we create our measure of contract breach from differences between individual items rather than from differences between scales. Because we construe our measure of contract breach as a formative scale, reliability coefficients used to assess reflective measures are not appropriate (Diamantopoulos and Winklhofer 2001).

A good example of this is our propensity to exit (PTE) measure. We began with three items for this measure; however, the purification process indicated that two items should be dropped. Thus, the two items reflecting propensity to exit during the next six months or during the next year were eliminated. Their poor statistical performance can be traced to their conceptual redundancy with the remaining item, propensity to exit during the next two years.