Too much power or no power: when does intermediary’s power result into better wine and happier farmers?

Abstract

The study analyzes the trading relationship performance between farmers and intermediaries and the factors shaping it, with a focus on intermediary’s power, based on a structured survey of vineyard farmers in Kosovo. Confirmatory factor analysis is employed to develop measures for the study latent variables, and ordinary least squares regression is used to test the hypothesis. To further validate the results, machine learning (i.e. random forest) is used to model the factors affecting the relationship performance between farmer and intermediary. The results show that when the intermediary has considerable (excessive) power, it leads to low trading relationship performance with farmers. Also, when the intermediary has little power, the relationship performance with farmers behaves in a similar way. The main contribution of this paper is to further illuminate the debate on the role of power in business-to-business relationships, in that it points out an alternative explanation; stating that there is an optimal level zone that power needs to exist, in order to achieve above average trading relationship performance. Outside this zone, either low or excessive/high intermediary’s power results in poor relationship performance.