Marketing channel behaviour and performance

Thesis submitted to University of Manchester for the degree of Doctor of Philosophy in the Faculty of Business Administration.

A major aim of this study is to offer a relatively comprehensive picture of marketing channel behaviour and performance. Given the statistical difficulties in testing a very large, comprehensive model to achieve this aim, two separate but overlapping models are proposed. One model specifically addresses behaviour in marketing channels, while the other integrates key behavioural outcomes with objective measures of performance. Within the second model the personalities of key boundary managers are introduced to help explain the linkages between the behavioural outcomes and economic outcomes. Empirical research to test the models was undertaken in one large marketing channel­ Post Office Counters Ltd. (POCL). This research focuses on privately owned agencies-the sub-post office--and only those with a viable second business. The marketing channel is otherwise similar in form to many franchises and agencies. The agents are owners-managers of small businesses in which they are the key decision makers. POCL provided a sample of 2,400 post offices with a second viable business. The research followed a dyadic design. A self-completion questionnaire, which had been previously validated, was sent to the sub-postmaster in each outlet. The area manager acting on behalf of the principal for each of the outlets for which a return was received was sent a mirror image questionnaire (the same questionnaire about the relationship but rephrased to be relevant to the principal). A total of 497 cases that contained complete data for the dyad represent the sample used in this study. The questionnaire was designed to measure a number of constructs. These were: the satisfaction of the sub-postmaster- the agent- with the overall channel relationship; the balance of power in their relationship with POCL-the principal; wh ther POCL were using power coercively or non-coercively; trust (using two measures, honesty and benevolence); conflict, and the personality of both the agent and the area manager. Conflict is treated as a dyadic measure. The personality instrument was the NEO Five­ Factor Inventory (NEO-FFI; Costa and McCrae 1992), a standard, well-validated personality test. This assesses the five broad dimensions of personality: Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness. Two measures of performance-business and operational-were computed . based on data supplied by Post Office Counters Limited. Structural Equation Modelling was used to test the two hypothesised models and to obtain robust information about how constructs interrelate. The behavioural model adds to the streams of research on power and on trust. Past research on trust has emphasised the investigation of antecedents other than that of power structure and power use. The present study has extended the knowledge of the impact of power (power asymmetry, coercive power, and non-coercive power) on critical behavioural outcomes-co-operation, conflict, and satisfaction-by showing that trust is an important mediating variable. The second model, the model of channel member behaviour and performance also makes several contributions to the channels literature. One is to play a part in the integration of the economic and the behavioural perspectives on the channels literature. The work is unique in that it combines the channels literature with the personality and industrial/organisational psychology literatures to improve the understanding of channel phenomena. The main contribution of this study is the confirmation that the personality traits (Openness to Experience, Extraversion, and Agreeableness) of key interface managers helps to explain channel member satisfaction, conflict, and performance. Concretely, the personality dimension Agreeableness proved useful to help explain the controversial link between conflict and performance. The findings reveal that personality is relevant to help predict not only the attitudes and behaviour of channel members, but also objective, non-perceptual and non-self reported dimensions of performance (e.g., sales growth).