Analysis of Operating Efficiency of the Iberian Peninsula Container Ports. A Multiple Regression Analysis and DEA in a Second Stage

Abstract

ABSTRACT

The aim of this paper is to obtain the most consistent values of efficiency of Iberian Peninsula container ports from 2008 to 2014. This study applies DEA methodology to a sample of the 16 Iberian Peninsula container ports they account for more than 85 per cent of the total container traffic, during the five-year period of 2008-2014.

Firstly, multiple linear regression models have been applied for all possible combinations of explanatory variables and using the number of manipulated TEUs as an independent variable which has been considered as the most relevant of the outputs chosen in the efficiency analysis. The aim has been to represent a variable response (TEUs) according to different possible explanatory variables. The explanatory variables used are: Number of Cranes (NG), Area of the Terminal (ST) and Docks with draft of more than 14 meters (M14). The indices that allow us to measure the quality of the results obtained are the corrected multiple determination coefficients (R²) and the Fisher-Snedecor (F) statistic. It can be observed that the best results for both coefficients are obtained in those hypotheses where the explanatory variable corresponding to the Number of Cranes (NG) is included.

In a second Stage Data Envelopment Analysis (DEA) with variable return to scale (BCC) has been used to obtain the standard efficiency, using output-oriented option in order to determine efficiency rankings of Iberian Peninsula container ports. The results show that the ports with significant import/export traffic seem to exhibit lower levels of technical efficiency than those ports specialized in transhipment operations. On the other hand, in those cases in which the transhipment traffic has increased, the technical efficiency has diminished.