Estimation of Dynamic Oligopolistic Interaction: The Case of the Banana Export Market

Abstract

In the past two decades, researchers in the field of industrial organization have focused on the use of static, structural econometric models to estimate the degree of market power in domestic industries. However, these studies fail to take into account the multi-period nature of markets. Therefore, it may be more appropriate to use a dynamic approach to estimating the degree of market power. This paper applies a linear-quadratic methodology, developed by Karp and Perloff (1989, 1993a), to the German market for banana imports. Assuming a quadratic objective function with linear demand and quadratic cost, and a Markov equation that captures the dynamic interaction among firms, a dynamic conjectural variations parameter is derived under open-loop and feedback strategies. Compared to the open-loop conjectural variations parameter, the feedback: conjectural variations parameter shows a greater degree of collusiveness. For the purpose of hypothesis testing, standard errors on the conjectural variations parameter are calculated using Taylor expansion and bootstrapping methods. The hypothesis that market structure is perfectly competitive/perfectly collusive is rejected, however, the hypothesis that firms behave in Cournot-Nash fashion could not be rejected.