Productivity and convergence in Norwegian container seaports: An SFA-based Malmquist productivity index approach

Highlights

• We explore productivity convergence of container seaports using SFA-based Malmquist productivity index.

• Norwegian and comparable seaports in the Nordic countries and the UK are examined.

• Annual average total productivity is 9.7%, but technical efficiency regressed by −3.2%.

• Norwegian seaports are best performers with regard to scale efficiency change.

• Convergences are found; seaports with initially lower productivities progressed faster than those with initially higher.

• Convergences exist; seaports with initially lower productivities progressed faster than those with initially higher.

Abstract

This paper examines the productivity of Norwegian container seaports and the extent to which the initially less productive seaports are converging/diverging in productivity in relation to the initially more productive seaports. The rationale of the paper is that productivity assessment of Norwegian seaports is scarce in the literature, and studies of convergence/divergence in container seaports are missing in the literature on transportation. The methodology used is the stochastic frontier analysis (SFA)-based Malmquist productivity index ($\mathit{MPI}$) to assess productivity. To infer productivity convergence/divergence over the years, Beta (Β) and Sigma (Γ) types of convergence/divergence are used. The data used are for the 2002–2014 period and include comparable container seaports from the Nordic countries and the UK to increase the discriminatory power in the analyses and to compare the Norwegian container seaports with their peers. The results attest that (i) there has been annual average total productivity improvement across seaports of approximately 9.7%, (ii) the productivity improvement observed has been due to improvement in pure efficiency change of approximately 11% and scale efficiency improvement of approximately 3.2% and a decrease in technical change of approximately 3.2% annually, (iii) Norwegian seaports are also experiencing productivity improvement and are best performers with regard to scale efficiency change, and (iv) there is an indication of convergence among seaports with respect to all productivity indices, suggesting that seaports with initially lower productivity indices progress faster than those with initially higher indices. A major conclusion is that the overall productivity of seaports has improved, and the productivity growth rates have converged.

Introduction

In the past quarter century, there have been significant changes in the seaport industry worldwide, which have challenged seaport management both in terms of objectives that are set and tools that can be utilized to realize those objectives (Suykens and Van de Voorde (1998); Cullinane, 2002, Cullinane and Wang, 2010a, Cullinane, 2019). For instance, The Norwegian National Transport Plan (NTP) for 2018–2029 has proposed incentives for transferring the transport of goods from roads to sea in the Norwegian territory; such a shift involves shortsea container shipping as a means of reducing overall transportation costs as well as reducing emissions. By that time, the volume of goods transported by sea as measured in billion-ton kilometers will be a formidable 58% of all goods transported in the territory. It is clear, then, that if the objectives of the NTP are to be realized effectively and if the growth in container flows continues, seaports must play their role, which is to operate as efficiently and as productively as possible. These observations are not limited to the case of Norway but rather reflect worldwide trends where the international literature on the performance of seaports has recognized that the efficiency of seaports is a prerequisite for a well-functioning supply chain and transportation systems (Bichou and Gray (2005); UNCTAD (2016) and; Cullinane (2019)).

In addition to the observations above, it has been observed in the literature that the contribution of seaports to international competitiveness has increased tremendously over the last decade, irrespective of country (Hung et al. (2010); UNCTAD (2016). These trends have led transportation and logistics scholars to question whether seaports are adapting to the new conditions, that is, whether they are striving to be as efficient and as productive they can be and what potential there is for efficiency and productivity improvements. Scholars have attempted to address this question by using frontier approaches to technical efficiency (TE) measurements, where seaports are compared against each other as measured by a distance to a given frontier. The underlying rationale for this type of comparison is that poorly performing seaports can learn from their peers or the best performers to increase efficiency and/or productivity and thereby to improve overall supply chain performance (Lai et al. 2002).

Despite the existence of TE studies of seaports in the literature (Panayides et al. (2009); Odeck and Bråthen (2012); and Woo et al. (2012)), there are still some shortcomings in the literature, and some of them are specific to Norway. These shortcomings are as follows: (1) Measurement of the efficiency of Norwegian seaports is very scarce and limited to only two studies: Schøyen and Odeck, 2013, Schøyen and Odeck, 2017. They used a data envelopment analysis (DEA) approach to measure TE, and their data encompassed the short periods of 2002–2008 and 2009–2014, respectively; thus, they do not illuminate how performance has further evolved in the most recent years or with the use of an alternative approach to DEA. Furthermore, the elasticity of seaport outputs with respect to the various inputs has not been addressed in the case of Norway; elasticity is readily addressed using the Stochastic Frontier analysis (SFA) approach. (2) Productivity assessments of seaports using the Malmquist productivity index ($\mathit{MPI}$) are virtually missing in the literature apart from the study by Schøyen and Odeck (2017), which used a DEA-based $\mathit{MPI}$ to measure the productivity of Norwegian and some Nordic and UK seaports. Here, too, it should be added that an alternative method such as SFA is interesting to consider because the results are not necessarily independent of the method used. Moreover, studying productivity using $\mathit{MPI}$ offers insight into three mutually exclusive components of productivity that may help clarify the reasons for observed productivity improvement/regression and enable a study of convergences/divergences in productivities, as explained below. (3) No current study in the literature of seaport efficiency has yet explicitly examined the notion of convergence or divergence as defined in the literature of economics with respect to the productivity of seaports, which would reveal how seaports are converging or diverging in relation to each other in terms of productivity. Although efficiency and productivity studies over time may implicitly address convergence/divergence, it has not been done according to the common definition found in the literature of economics; see, for instance, Barro and Sala-i-Martin, 1995, Sala-i-Martin, 1996. (4) Finally, with respect to the three points above, it is necessary to compare Norwegian seaports with their peers in the Nordic countries and the UK. This latter issue is interesting to investigate in the wake of the changing circumstances within which seaports operate, which may be different across countries, although the physical operational conditions for seaports are comparable.

The aim of this paper is to explore the four shortcomings addressed above with seaports in Norway as the focus. However, because the Norwegian seaports considered are few in number, comparable seaports from other Nordic countries and the UK are included in the analysis as a means of increasing the discriminatory power of the method used. Furthermore, the inclusion of comparable Nordic and UK seaports makes it possible to compare the performance of Norwegian seaports versus international peers that operate under more or less the same physical conditions. The data used are for the seaport production period of 2002–2014. The approach used to assess efficiency in the analysis is SFA and its extension to the $\mathit{MPI}$ to analyze productivity improvement or regression. To analyze convergences in productivity, the well-known β- and Γ -types of convergence (Sala-i-Martin (1996)) are used.

The rest of the paper is organized as follows: Section 2 provides a brief literature review, Section 3 describes the data, and Section 4 describes the methodology used. The empirical results are presented in Section 5, and Section 6 provides concluding remarks.