Estimating allocative efficiency in port authorities with demand uncertainty

This paper aims to analyse the impact of port demand variability on the allocative efficiency of Spanish port authorities during the period 1986-2007. From a distance function model we can obtain a measure of allocative efficiency using two different approaches: error components approach and parametric approach. We model the variability of port demand from the cyclical component of traffic series by applying the Hodrick-Prescott filter. The results show that the inclusion of variability does not affect the efficiency measures, except in the case of containerised general cargo. Moreover, we demonstrate that port authorities have excess capacity and their resources are misallocated. Finally we establish that the allocative inefficiency of Spanish port authorities is difficult to resolve given the limited substitution possibilities among the different pairs of inputs.