When container specialisation makes a difference: an efficiency analysis of the Spanish port authorities

ABSTRACT

There is evidence of specialisation around containerised cargo but also a lack of literature regarding its impact on ports’ performance. This paper studies the relationship between the level of containerisation and port efficiency and investigates whether the containerisation–efficiency relation varies according to the degree of containerisation. To this end, a two-stage procedure is proposed. In the first stage, the efficiency of 26 Spanish port authorities is estimated by meta-frontier DEA techniques. In the second stage, the effect of containerisation on port efficiency is analysed. The results show evidence of threshold effect of container specialisation on port efficiency. Specialisation in container traffic is beneficial for medium-high-complexity ports when they reach a level of specialisation of 40% in container traffic.

KEYWORDS:

• Efficiency
• container ports
• specialisation
• competitiveness

Note

This article is a revised and expanded version of a paper entitled Wy container specialisation make a difference? An efficiency analysis of the Spanish Port Authorities presented at 15th World Conference on Transport Research, Mumbai, 27 May 2019.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The concept of decision-making units (DMUs) is explained later in section 3.2

2. The choice the model orientation should be done based on the level of control Port Authorities have on inputs and outputs variables. As argued by Cullinane et al. (2006) we consider that Spanish Port Authorities operate in a regulated sector characterized by endogenous inputs and exogenous outputs.

3. Statistical information about input quantities are not imputed to the activities for which they are employed, so not including passengers can lead to biased results.

4. Variables’ definition comes directly from the sources of data (annual reports of Puertos del Estado and port authorities)

5. $\bar{x}=\frac{{\sum }^{x_{it}}}{N\ast T}$ where $x_{it}$ represents the value of the variable $x$ for the $i_{th}$ port authority in the $t_{th}$ period, $N$ is the number of port authorities and $T$ the number of periods in the panel of data.

6. $\sigma =\sqrt{\frac{{\sum }^{{(x_{it}-\bar{x})}^2}}{N\ast T}}$ where $x_{it}$ represents the value of the variable $x$ for the $i_{th}$ port authority in the $t_{th}$ period, $N$ is the number of port authorities and $T$ the number of periods in the panel of data.

7. The Min. and Max represents the minimum and maximum value of the variable in the panel of data, respectively.

8. The concept of decision-making units (DMUs) is explained later in section 3.2

9. The choice the model orientation should be done based on the level of control Port Authorities have on inputs and outputs variables. As argued by Cullinane et al. (2006) we consider that Spanish Port Authorities operate in a regulated sector characterized by endogenous inputs and exogenous outputs.

10. Statistical information about input quantities are not imputed to the activities for which they are employed, so not including passengers can lead to biased results.

11. Variables’ definition comes directly from the sources of data (annual reports of Puertos del Estado and port authorities).

12. $\bar{x}=\frac{{\sum }^{x_{it}}}{N\ast T}$ where $x_{it}$ represents the value of the variable $x$ for the $i_{th}$ port authority in the $t_{th}$ period, $N$ is the number of port authorities and $T$ the number of periods in the panel of data.

13. $\sigma =\sqrt{\frac{{\sum }^{{(x_{it}-\bar{x})}^2}}{N\ast T}}$ where $x_{it}$ represents the value of the variable $x$ for the $i_{th}$ port authority in the $t_{th}$ period, $N$ is the number of port authorities and $T$ the number of periods in the panel of data.