Determinants of container port choice in Spain

For the economic and financial evaluation of port investment projects, it is important to know the demand function of a port's services. The objective of this study is to establish such a demand choice function for the Spanish container port services. The function is derived from the coefficients of a port choice model, for which a multinomial logit model is used and of which the coefficients are estimated with regression analysis. The variables tested concern inland transport cost, ocean transport costs and broad proxy variables for quality of service. Information on container import and export flows for 2007 is obtained from the Spanish Treasury Department. The linear regression analysis is based on differences of utilities of alternative routings of containerised cargoes compared to those routed via the port of Valencia. The obtained results are satisfactory in terms of model fit. The estimated coefficients can be used to assess the impact of changes in costs of container flows routed via a port on a port's market share. A demand choice function for the port can be derived by systematically doing so. An example is presented for the port of Valencia.     

Testing port choice models using physical and monetary data: a comparative case study for the Spanish container trades

ABSTRACT

Transport costs are useful explanatory variables in port choice research. Nevertheless, the availability of such information usually poses a problem. Thus, the formulation of an alternative approach, to be used as a proxy of these variables, would be desirable. The objective of this study is to improve the analysis of container port choice using logit models by adopting physical non-monetary indicators based on maritime distance and ship size. The statistical tests of logit models on port choice using these new variables are compared with the result of using cost variables for the same data set of choice positions. The statistical outcome is good and allows us to present this new approach as a contribution to the literature on port choice modeling.