Current Issues in Shipping, Ports and Logistics [1 ed.] 9789054878582

Citation preview

CURRENT ISSUES IN SHIPPING, PORTS AND LOGISTICS  Theo NOTTEBOOM (ed.)

Cover design: Stipontwerpt, Antwerpen Print: Silhouet, Maldegem © 2011 Uitgeverij UPA University Press Antwerp UPA is an imprint of ASP nv (Academic and Scientific Publishers nv) Ravensteingalerij 28 B-1000 Brussels Tel. +32 (0)2 289 26 50 Fax +32 (0)2 289 26 59 E-mail: [email protected] www.upa-editions.be ISBN 978 90 5487 858 2 NUR 804 / 780 Legal deposit D/2011/11.161/021 All rights reserved. No parts of this book may be reproduced or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the author.

TABLE OF CONTENTS

 Introduction

1

PART 1

CURRENT ISSUES IN SHIPPING

9

Chapter 1

The Role of the Shipper in Decarbonising Maritime Supply Chains Robert Woolford and Alan McKinnon

11

Chapter 2

Shipping Companies’ Awareness and Preparedness for Greenhouse Gas Regulations: a Korean Case Sang-Yoon Lee and Young-Tae Chang

25

Chapter 3

The Determinants of Tanker Price in the Chinese Shipbuilding Industry Liping Jiang

51

Chapter 4

Bunker Costs in Container Liner Shipping: are Slow Steaming Practices reflected in Maritime Fuel Surcharges? Pierre Cariou and Theo Notteboom

69

Chapter 5

Integrating Intangible Resources in Strategic Co-Operations of Container Lines: Ships Agents’ Perspective Indika Sigera, Stephen Cahoon and Jiangang Fei

83

Chapter 6

Piracy off the Horn Of Africa: Impact on Regional Container Services on the Middle East-Africa Trade Paul De Coster and Theo Notteboom

103

Chapter 7

World LNG Shipping: Dynamics in Markets, Ships and Terminal Projects Siyuan Wang and Theo Notteboom

129

PART 2

CURRENT ISSUES IN THE ANALYSIS OF FLOWS AND NETWORKS

155

Chapter 8

The EU-Hanseatic Trading Rim in the 2009 World Crisis Anneleen Claessens, Evrard Claessens, Vesna Stavrevska and Franky To

157

Chapter 9

Estimating Global Container Flows Using a Strategic Network Choice Model Lóránt Tavasszy, Michiel Minderhoud, Jean-François Perrin, T. Notteboom

167

Chapter 10

Network Position and Throughput Performance of Seaports César Ducruet, Sung-Woo Lee and Ju-Mi Song

189

i

Current Issues in Shipping, Ports and Logistics

Chapter 11

North Africa and International Trade Networks: Port Investments and Market Opportunities Claudio Ferrari, Francesco Parola and Alessio Tei

203

Chapter 12

Applying Port Development Models Upstream: the Case of the Yangtze River Port System Albert Veenstra and Theo Notteboom

221

Chapter 13

Emerging Global Networks in the Container Terminal Operating Industry Theo Notteboom and Jean-Paul Rodrigue

243

Chapter 14

In Search of Routing Flexibility in Container Shipping: the Cape Route as an Alternative to the Suez Canal Theo Notteboom

271

PART 3

CURRENT ISSUES IN TERMINAL OPERATIONS AND PERFORMANCE

297

Chapter 15

Quay Efficiency of Container Terminals: Comparison between Ports in China and its Neighboring Ports Tao Chen

299

Chapter 16

Measuring and Analysing Terminal Capacity in East Africa: The Case of the Seaport of Dar Es Salaam John Layaa and Wout Dullaert

315

Chapter 17

Developing a New Technique for Evaluating Service Quality of Container Ports Kai-Chieh Hu and Paul T-W Lee

337

Chapter 18

The Size Distribution of Container Terminals: An Inter-Quartile Range Analysis Vicky Kaselimi and Theo Notteboom

357

Chapter 19

Exogenous Determinants of Land Productivity of Container Terminals Dries Verbraeken and Theo Notteboom

379

PART 4

CURRENT ISSUES IN LOGISTICS

389

Chapter 20

Shipment Consolidation and Container Booking Planning for Sea Freight Forwarders Manoj Lohatepanont, Pakorn Rattanatsuwan, Supaporn Siripraprut and Nantavat Kamolthiptrakul

391

Chapter 21

Managing Speed and Reliability in Freight Transport Wout Dullaert and Luca Zamparini

405

Chapter 22

Taking Replenishment Capacity into Account in a Basestock Inventory Policy Thomas Dubois, Birger Raa and Wout Dullaert

423

ii

Table of Contents

Chapter 23

The Development of Time Driven Activity-Based Costing Models: A Case Study in a Road Transport and Logistics Company Sirirat Somapa, Martine Cools and Wout Dullaert

431

Chapter 24

China’s Logistics Service Enterprise Structure: Institutional Impediment to Multimodalism Booi H. Kam and Peter J. Rimmer

447

PART 5

CURRENT ISSUES IN PORT DEVELOPMENT AND GOVERNANCE

465

Chapter 25

Transport Node Governance in a Changing World: The Institutional Reform of Tianjin Port Eva Chen Shou, Adolf K.Y. Ng and Athanasios A. Pallis

467

Chapter 26

The Impact of Institutional Structure on the Development of Inland Waterway Transport in the Pearl River Basin Jinyu Li, Theo Notteboom and James Wang

483

Chapter 27

The Next Step on the Port Generations Ladder: Customer-Centric and Community Ports Matthew Flynn, Paul T-W Lee and Theo Notteboom

497

Chapter 28

The Development of Bangkok Port Kamonchanok Suthiwartnarueput

511

Chapter 29

The Spatial Distribution of Port Activity: Theoretical and Methodological Notes for Modelling and Research Erick Leal, Theo Notteboom and Ricardo Sánchez

521

Chapter 30

Small and Medium-Sized Ports (SMPs) in Multi-Port Gateway Regions: The Role of Yingkou in the Logistics System of the Bohai Sea Lin Feng and Theo Notteboom

543

Chapter 31

An Institutional and Regulatory Framework for Dry Port Development Linda Spies and Theo Notteboom

563

About the Authors

587

iii

Current Issues in Shipping, Ports and Logistics

iv

INTRODUCTION

 Theo NOTTEBOOM (book editor)

1 | Motivation for this book Shipping, port and logistics activities are indispensable in today’s global economy. They are a major source of added value creation and employment and have an important role to play in facilitating global trade. At the same time these activities need to be embedded in their local multi-stakeholder environment. Rising globalisation with powerful and relatively footloose players, extensive business networks and complex production and consumption patterns have a dramatic impact on freight transport and related logistics. The new environment creates a high degree of uncertainty and leaves managers active in shipping, ports or broader supply chain management with the question on how to respond effectively to market dynamics and regulatory issues. The shipping, port and logistics industry is challenged to be efficient and to expand its services in line with increased customer requirements while at the same time lowering the environmental footprint and guaranteeing high security and safety standards. This book bears the title ‘Current Issues in Shipping, Ports and Logistics’. With this book the contributors unravel some of the most pressing challenges to shipping, ports and logistics, so as to advance and update our thinking on the developments in this field. This book brings together 31 contributions from a team of more than 50 authors. The scholars involved are coming from 14 nations and six continents giving the book a truly international view on the challenges to shipping, ports and logistics. The book covers five parts: current issues in shipping, current issues in the analysis of flows and networks, current issues in terminal operations and performance, current issues in logistics and current issues in port development and governance. Most of the contributions in this book were presented during the Asian Logistics Round Table conference held in Antwerp on 2 and 3 December 2010 (www.ua.ac.be/alrt2010). The conference was organized by the Institute of Transport and Maritime Management Antwerp (ITMMA) of the University of Antwerp (www.itmma.ua.ac.be) with the support of the Antwerp Port Authority (www.portofantwerp.com). Marc Van Peel, President of the Antwerp Port Authority, and Eddy Bruyninckx, CEO of the Antwerp Port Authority, personally saw to it that the conference could be held at the ‘Harbour House’, the headquarters of the port of Antwerp. PortEconomics.eu acted as a supporting organization. 1

Current Issues in Shipping, Ports and Logistics

The ALRT 2010 conference brought academics together to exchange ideas on the conference theme 'Analyzing logistics and transport networks in a changing world economic geography'. The conference served as an excellent forum to advance this book project. The event also incorporated the results of past ALRT conferences held at Kainan University in Taoyuan (Taiwan) in 2007 and 2008 and the ALRT 2009 conference held at Inha University in Incheon (South Korea). The next ALRT conference will take place in 2012 at the University of British Columbia (UBC) in Vancouver. Founded in 2007, ALRT is an international cooperation program between academic research centers to address research questions linked to transport and (maritime) logistics, mainly in Asia. Member universities include:  Department of Maritime and Logistics Management - Australian Maritime College AMC (Australia);  RMIT University (Australia);  Department of Logistics and Shipping Management and International Shipping Management and Logistics Research Center - Kainan University (Taiwan);  Logistics Research Centre - The Hong Kong Polytechnic University (Hong Kong, China);  Jungseok Research Institute of International Logistics and Trade and Asia Pacific School of Logistics - Inha University (South Korea);  International Logistics Management Division - Chulalongkorn University (Thailand);  Center for Transportation Studies - University of British Columbia (Canada);  Transport Research Institute - Edinburgh Napier University (UK);  Logistics Research Centre - Heriot-Watt University (UK);  ITMMA - University of Antwerp. The Greater China Supply Chain and Logistics (GCSCL) network acts as observer. The papers in this multidisciplinary book add value to existing literature on shipping, ports and logistics and will help to initiate further systematic thinking on these and related research themes.

2 | An overview of the contributions The first part of the book focuses on current issues in shipping with contributions on environmental aspects in shipping, piracy, shipbuilding, pricing strategies of shipping lines, strategic co-operation in the shipping industry and the LNG shipping market. The first chapter, authored by Robert Woolford and Alan McKinnon, considers maritime supply chain carbon emissions from the perspective of the shipper. The shipper, as the purchaser and user of transport services can influence the intensity of carbon emissions emitted through supply chain decisions. The paper proposes a framework based upon eight key parameters that link the economic output of an economy or industry to the carbon emissions produced through the activity of moving the goods it produces.

2

Introduction

Chapter 2 by Sang-Yoon Lee and Young-Tae Chang looks at the level of awareness and preparedness of Korean shipping companies for the currently discussed GHG reduction issues in the international shipping sector. The paper identifies differences between large and small/medium size shipping firms in terms of recognition and ability for green shipping regulation. In addition, the paper attempts to find linkages between the level of awareness and preparedness of Korean shipping companies and their preferring market based measures. The determinants of tanker price in the Chinese shipbuilding industry are addressed by Liping Jiang in Chapter 3. The analysis, employing Principal Component Regression (PCR) approach, indicates that the shipbuilding cost has the most significant impact. While increases in other five factors, namely crude oil price, time charter rate, shipbuilding capacity utilization, price cost margin, and ship export credit rate, have descending order of influences. The paper includes simulations to investigate what would happen to the Chinese tanker price when major changes would occur at the level of the factors identified. The contribution by Pierre Cariou and Theo Notteboom deals with the linkages between bunker costs in container liner shipping, slow steaming practices and fuel surcharge policies of shipping companies. Slow steaming has been implemented by the main liner shipping companies since 2008. Through an extensive analysis of liner service characteristics, fuel costs and fuel surcharges this paper provides an answer to three research questions (a) How significant are slow steaming practices in container liner shipping?; (b) What is the impact of slow steaming on fuel consumption and liner service characteristics?; and (c) To what extent has slow steaming changed the relation between fuel costs and fuel surcharges imposed on shippers by shipping lines? Indika Sigera, Stephen Cahoon and Jiangang Fei give an overview of the results of an empirical study conducted in mid 2010 that investigated how intangible resources are integrated during strategic co-operation among container lines. The activities of several forms of strategic co-operation among container lines are reviewed. The empirical study found that the integration of intangible resources is varied among these strategic co-operation; in particular, they differ in M&As. The study also found that the main factors determining the integration of intangible resources were antitrust and other regulations, structure of strategic co-operation and motives for forming them. Piracy off the Horn of Africa is the central theme in Chapter 6 by Paul De Coster and Theo Notteboom. The authors provide an in-depth analysis of the impact of piracy on the cost base and operations of container shipping between the Middle East and East/South Africa. The paper describes the piracy problem off the Somali coast, the scope of the danger zone, the trend over 2009 and the consequences of making a detour in an attempt to avoid the piracy area. The real cost implications of detours are compared for different levels of charter hire and fuel cost. Incremental costs for containers and other indirect consequences are also being highlighted.

3

Current Issues in Shipping, Ports and Logistics

The last chapter in Part 1 deals with world LNG shipping. Siyuan Wang and Theo Notteboom explore current dynamics in the LNG shipping market focused on three aspects: the evolution of LNG short-term shipping and market structure, the growth of the LNG fleet and the development of terminal projects. It is found that with the increase of LNG trade, short-term shipping is getting more common, and that the size of LNG tankers and the scale of terminals are also growing accordingly in order to benefit from economies of scale. The second part of this book deals with current issues in the analysis of flows and networks. Seven chapters discuss various aspects related to trade flows and the configuration of global transport networks. Anneleen Claessens, Evrard Claessens, Vesna Stavrevska and Franky To analyse the EU-Hanseatic trading rim in the 2009 world crisis. This chapter addresses the ancient Hanseatic region from the North Sea to the Baltic as an informative yardstick for external trade and internal traffic analysis. The authors demonstrate that for some EU Member States a heady boom was followed by a crushing reversal of external trade in the 2009 crisis. Other member states escaped from the dramatic downturn, and managed to grow even faster than they did before, though only for specific products. All this may be explained by net scale factors, effective marketing efforts and due strategies in hinterland coverage and transport infrastructure. Finally, a post-crisis update is bound to model the structural recovery across trading trails and industries. Chapter 9 by Lóránt Tavasszy, Michiel Minderhoud, Jean-François Perrin and Theo Notteboom presents a specification, estimation and application of a strategic network choice model for global container flows. The paper presents a strategic model for the movement of containers on a global scale in order to analyse possible shifts in future container transport demand and the impacts of transport policies thereon. Chapter 9 also introduces a scenario analysis to understand the impact of slow steaming, an increase in land based shipping costs and an increased use of large scale infrastructures on port throughput. César Ducruet, Sung-Woo Lee and Ju-Miang Song examine the network position and throughput performance of seaports. They propose a set of novel indicators describing the relative position of seaports in the worldwide maritime network of container shipping, which are distinguished among five categories: circulation (calls, vessels, and operators), foreland (distance to other ports and distribution of connections), connectivity itself (number of connections to other ports), centrality (betweenness and eccentricity), and neighbourhood (strength and clustering indices). Chapter 11 adds an African dimension to the analysis of flows and networks. Claudio Ferrari, Francesco Parola and Alessio Tei analyse how the infrastructural development of North-African ports will affect the international trade patterns on Europe-Far East routes. The main socio-economic trends of the region are discussed as well as the possible impact the emerging hub ports could have on the carriers’ deep-sea services. The future development of this economic region could generate additional cargo traffic volumes and the progressive growth of the North-African port range, enhancing the Mediterranean competitiveness as a whole. 4

Introduction

The next paper brings the reader to the Yangtze river basin in China. Albert Veenstra and Theo Notteboom apply port development models to the Yangtze river port system. It is argued that the Yangtze River system is going through a regionalization phase, mainly but not exclusively in relation to the port of Shanghai. The paper addresses the dynamics in the Yangtze River ports system by analyzing the level of cargo concentration and the degree of inequality in operations of the container ports. Chapter 12 also assesses observed differences in development of ports in different areas along the river (upstream/downstream) and reflects on the role of ownership structures in shaping regional load centre networks. Theo Notteboom and Jean-Paul Rodrigue discuss the emerging global networks in the container terminal operating industry. They analyse the similarities or differences among terminal locations, the processes leading to the expansion of terminal operating groups and the interactions they maintain as nodes within the global freight distribution system. It will be demonstrated that terminal operators show varying degrees of involvement in the main cargo handling markets around the world. The paper also unravels the corporate geography of the investment strategies of global terminal operators. Chapter 14 is focused on the competition between the Cape route and the Suez Canal. The Suez Canal plays a pivotal role in today’s global container shipping network. Theo Notteboom examines to what extent and for which trade lanes the Cape route could develop into a competitive alternative to the Suez route. The market potential of the Cape route is analyzed using a distance analysis, a transit time analysis and a generalized cost analysis on a set of trade lanes. The results show that the Cape route has the potential to serve as an alternative to the Suez route on eleven trade lanes. The third part of this book is dedicated to current Issues in terminal operations and performance with a particular focus on container terminals. Chapter 15 by Tao Chen compares the differences between the factors affecting the quay efficiency of terminal operators in China and its neighboring ports. The paper can help terminal operators, especially the ports neighboring China, in reviewing their operations planning and management procedures to improve quay efficiency. John Layaa and Wout Dullaert measure and analyse terminal capacity in East Africa in Chapter 16. The port of Dar es Salaam in Tanzania serves as a case study. The authors use standard queuing models to measure capacity utilisation. The results show that terminal capacity for the terminals considered is being underutilised and that as a result ships are being subjected to unnecessarily long waiting times. Chapter 17 introduces a new technique for evaluating service quality of container ports. Kai-Chieh Hu and Paul T-W Lee propose a so- called improvement effort index to evaluate service quality factors in the container port sector from the perspective of container shipping lines by integrating Kano’s model and importance-performance analysis. The new index is applied to evaluate service quality of major Asian container ports. The results confirm that the top five quality attributes that container liner shipping lines are not satisfied with are: port congestion, inappropriate settlement of 5

Current Issues in Shipping, Ports and Logistics

accident claims, lower transparency in pricing negotiation and administrative process, lack of monitoring system of port services for a port user’s satisfaction, and low level service of cargo claims and a port user’s needs. Vicky Kaselimi and Theo Notteboom discuss the size distribution of container terminals. They argue that the optimum scale of a terminal is guided by the “preferred” scale. Conventional calculation methods are not relevant for the study of the “preferred” scale of a container terminal. In the absence of more appropriate estimation approaches, the authors in this paper explore the possibilities of using a proxy method based on the size distribution of container terminals for calculating the “preferred” scale. Chapter 19 is centered on the exogenous and endogenous determinants of land productivity of container terminals. The paper examines the land productivity related to port terminals and develops a broad fitting framework to categorize all external influences and to benchmark terminals. Using a survey, Dries Verbraeken and Theo Notteboom categorize external factors based on three distinct characteristics: the amount of influence a terminal operator can exert on a given factor, the projected time of effective implementation of changes and the relative expected effect of the factor on overall productivity. Current issues in logistics are centre stage in part 4. The first chapter of this book section looks at shipment consolidation and container booking planning for sea freight forwarders. Manoj Lohatepanont, Pakorn Rattanatsuwan, Supaporn Siripraprut and Nantavat Kamolthiptrakul argue that shipment consolidation and container booking are two interrelated decisions, which can best be made simultaneously to obtain optimality. The paper proposes an integer programming formulation for the Integrated Shipment Consolidation and Container Booking Model. The model is tested using real data from a major forwarder in Thailand with operation out of Laem Chabang port in Thailand. Wout Dullaert and Luca Zamparini examine the impact of transport speed and reliability, measured by the average and variance of the lead time, on inventory costs. It is shown how reducing variability does not necessarily reduce costs and might in fact increase the costs of safety stock, depending on the shape of the demand during lead time distribution and targeted service level. The impact of transport reliability on safety stock costs can therefore differ significantly and offers a novel explanation for the wide variety of value of reliability figures obtained in empirical transport research. The ideas are illustrated by means of a flexible simulation framework, capable of estimating the value of time and the value of reliability for a real-life case. Chapter 22 studies periodic review inventory systems in which replenishments are capacitated. This capacity restriction implies that the order-up-to level may not always be reached at each replenishment, such that additional safety stock is needed to achieve the same service level as in the uncapacitated case considered in the literature so far. To determine the required level of safety stock, and hence the order-up-to level, Thomas Dubois, Birger Raa and Wout Dullaert propose an iterative procedure. A 6

Introduction

computational experiment is reported that illustrates both the impact of a restricted replenishment capacity on the required safety stock level, and the effectiveness of the proposed iterative method at determining this. Chapter 23 looks at time driven activity-based costing models (TDABC) with application to road transport and logistics. TDABC is aimed to overcome the disadvantages of the traditional activity-based costing and seems particularly useful for road transport and the logistics sector. Sirirat Somapa, Martine Cools and Wout Dullaert find that small firms can benefit from TDABC due to the use of simplified parameters to capture the disparity in operations. They recommend that transaction data recording should be designed thoroughly before the actual implementation in order to enhance the accountability and efficiency of TDABC models. In the last chapter of part 4 Booi Kam and Peter Rimmer argue that China’s logistics service enterprise structure is an institutional impediment to multimodalism. Drawing on the findings of three years of investigation on the logistics industry in China, this paper shows that the institutional set-up of China’s logistics service industry, which continues to nurture the growth of the single truck-owner operators, poses a significant impediment to multimodal transport development in China. The last part in this book deals with current issues in port development and governance. Chapter 25 brings the reader to China with a case study on the institutional reform of Tianjin port. Eva Chen Shou, Adolf Ng and Athanasios Pallis explain that port devolution schemes are varied across countries, which led to various approaches applied in shaping devolved institutional structures for port governance during reforms. Using path dependence theory and a case study on Tianjin port, empirical evidence was found that the following reform was considerably constrained by the previous reform, which shaped the exclusive path of the individual port in institutional reforms. The next chapter also focuses on institutional issues in China with an analysis of the impact of the institutional structure on the development on inland waterway transport on the Pearl River. Jinyu Li, Theo Notteboom and James Wang aim at closing a gap in existing literature by examining the institutional structures in the Pearl River Delta and their contribution to the existing problems in making inland waterway transport a more competitive transport mode in the region. Matthew Flynn, Paul Lee and Theo Notteboom propose an extension of the port generations typology first developed by UNCTAD in 1994. This paper suggests that a “four generations” framework is not enough to reflect the port functions required by community and needs of port users in the rapidly evolving globalised economic system. This paper aims to develop a conceptual port ladder framework on “Fifth Generation Ports” characterized by the emergence of world-class customer-centric and community ports. Chapter 28 by Kamonchanok Suthiwartnarueput shows the specific challenges for the development of Bangkok port in Thailand. The paper presents the appropriate future 7

Current Issues in Shipping, Ports and Logistics

role and direction of the port and analyses related land use issues in terms of cargo operations and other related business like logistics activities.

both

Erick Leal, Theo Notteboom and Ricardo Sanchez propose theoretical and methodological notes in view of modeling the spatial distribution of port activity in a scenario where concepts such as knowledge and capabilities jointly with containerisation, networks and logistics integration are omnipresent. Evolutionary economic geography is presented as a useful theoretical framework to deal with the limitations of traditional models in port system development. In Chapter 30, Lin Feng and Theo Notteboom focus on the role of small and medium sized ports (SMPs) in multi-port gateway regions with a case study on the role of Yingkou Port in the Bohai Sea region. The authors argue SMPs are challenged to take up a more prominent and indispensable role in transport and logistics systems. The functional and strategic position of Yingkou is analyzed using multivariate analysis of the characteristics of the direct and extensive hinterland and the port’s connectivity to the hinterland. Fuzzy Comprehensive Appraisement (FCA) is used to make concrete proposals on how the logistics system in and around Yingkou can be improved. Finally, Linda Spies and Theo Notteboom present a discussion on the concept ‘institutional and regulatory framework’ in relation to dry port development. They explore the hierarchical structure and characteristics of an institutional framework and investigate which policy tools should be included in a regulatory framework. The theoretical considerations are supported by two case studies: the Venlo Trimodal Container Terminal in the Netherlands (Europe) and the Inland Container Depot of Lat Krabang, Thailand (Southeast Asia).

8

PART 1

 Current Issues in Shipping

9

Current Issues in Shipping, Ports and Logistics

10

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

CHAPTER 1 The Role of the Shipper in Decarbonising Maritime Supply Chains

 Robert WOOLFORD and Alan McKINNON

Abstract This paper introduces a new piece of research that considers maritime supply chain carbon emissions from the perspective of the shipper. The emissions from international shipping have been the subject of considerable research in recent years. Much of this research has concentrated on the design and operation of ocean going vessels. This research proposal takes a wider perspective and considers the whole of the maritime supply chain including the land-based logistics that form a part of the end-to-end supply chain. The shipper, as the purchaser and user of transport services can influence the intensity of carbon emissions emitted through the supply chain decisions he makes. This paper proposes a framework based upon 8 key parameters that link the economic output of an economy or industry to the carbon emissions produced through the activity of moving the goods it produces. The framework provides a means to assess the potential for CO2 reduction arising from different supply chain configurations. The paper discusses the eight parameters along with some constraints and supply chain trade-offs. This research furthers knowledge on the subject of maritime logistics by considering emissions from a whole supply chain perspective and explores how the shipper may influence them.

1 | Introduction In recent years, there has been a steep increase in the amount of research undertaken on CO2 emissions from shipping. The vast majority of the new studies have focused on shipping operations and examined what ship designers, ship builders and the operators of the vessels can do to reduce these emissions. The sea leg, however, is only part of a maritime supply chain. Shippers (or cargo owners) sending their goods by sea are more interested in the overall carbon footprint of the door-to-door movement of their products than in emissions from the marine link in the chain. Confining the analysis of vessel-related emissions also ignores the fact that there is an interaction between the shipping operation and land-based logistics. In measuring and

11

Current Issues in Shipping, Ports and Logistics

managing CO2 emissions associated with international trade, it is, therefore, advantageous to adopt a wider supply chain perspective. Shippers assuming responsibility for Scope 3 emissions, under the Greenhouse Gas Protocol, must report emissions made by carriers working on their behalf (WBCSD/WRI, 2004). Some companies are now committing themselves to carbon reduction targets at a supply chain, rather than internal company, level. As a result of globalization, more and more of their supply chains cross international frontiers and involve a shipping link. The CO2 emissions from this link therefore become a key element in their carbon management strategies. How much influence, however, can they exert on the carbon intensity of this link? Shippers’ demands, after all, exert a strong influence on the level of service provision and utilization of vessel capacity. Most of the previous research on carbon mitigation strategies for shipping, however, has adopted a supply-side perspective and under-estimated the contribution that users of maritime services can make. We have recently embarked on a new study focusing on the shippers’ role in reducing carbon emissions from supply chains containing a deep-sea container movement. It is seeking the views of other stakeholders in the maritime supply chain on the role that shippers can play. To what extent are their efforts to improve efficiency and cut carbon emissions currently constrained by the logistical requirements of shippers? If these requirements were relaxed, how large would be the resulting carbon savings?

2 | Scope of the research The research is confined to UK-based shippers who use deep-sea container services for exporting and importing consignments. Their degree of influence over the shipping operation largely depends on the terms of the trade (i.e. Incoterms). A large proportion of UK containerized exports are now sold on a delivered price or CIF basis giving the British exporter control over the transport operation as far as the foreign market. This situation is reversed for imported goods, with much of the responsibility for the deep-sea movement retained by the foreign shipper. Many large UK retailers, however, have now extended control of their import supply chains back to the source country and can therefore influence the nature of the global shipping operation. Some large shippers deal directly with shipping lines, though many others employ freight forwarders to purchase shipping services on their behalf. As intermediaries between shipper and shipping line they influence mode and carrier choice, consignment routing and the degree of traffic consolidation, all of which impact on CO2 emissions. Many port operators have also been assuming wider logistical responsibilities in an effort to transform their ports from terminals into ‘supply chain hubs’. At the same time, they are trying to reduce their carbon footprints. These two initiatives are closely inter-related and require the co-operation of shippers. In the course of the research, several deep-sea shipping lines and providers of port feeder services will also be consulted to explore, from their perspective, what shippers can do to help them decarbonise their operations. 12

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

3 | Conceptual framework In research undertaken for the Commission for Integrated Transport, the International Transport Forum1 and a previous project on Green Logistics, we have developed and refined a framework within which the potential for reducing CO2 emissions from freight transport can be assessed. The framework has been adapted to the proposed research on the decarbonisation of the maritime supply chain (Figure 1). It essentially maps the relationship between the output of an economy or industry sector and the amount of CO2 emitted in the course of its distribution, by sea and feeder modes, to customers. It is constructed around a series of eight key parameters discussed in the next section. Figure 1. Conceptual Framework

1

Commission for Integrated Transport and International Transport Forum, op.cit. 13

Current Issues in Shipping, Ports and Logistics

4 | Key factors that shippers can influence 4.1. Choice of transport mode This choice is made at two levels. At a global level, some shippers have a choice of shipping consignments solely by sea or on a combined sea-air service. Altering this balance in favour of the pure shipping option can substantially cut CO 2 emissions on the global transit. At the regional level, feeder movements to the deep-sea ports by sea, rail or road can generate, respectively, twice, 2-3 times and 6-8 times as much CO2 per tonne-km as deep-sea shipping2. These figures are indicative rather than absolute, particularly so in the cases of road and rail transport. In the case of road transport, congestion can have a significant impact on the scale of emissions. It has been suggested that a 40-tonne articulated truck stopping twice per kilometre raises fuel consumption by a factor of 3 (IRU, 1997). In the case of rail freight, there are two commonly used modes of traction, namely diesel and electric. In terms of emissions per tonne-km moved these emissions are typically to the order of 20 to 40g CO2 in the case of diesel traction and 15 to 20g CO2 for electric. The matter is complicated further, as the mode of electricity generation will have a direct effect on transport emissions where electric rail traction is used. In countries where nuclear power and / or renewable energy account for a large share of electricity generation, such as France or Switzerland, the emissions associated with rail transport will be extremely small, in fact arguably less than using ocean transport. The French environmental agency, ADEME claim an emissions figure of just 1.8 g CO2 per tonne-km for electrically hauled rail freight using the French rail network. Notwithstanding the outlying ADEME example of French electric rail traction, a recent study (McKinnon and Piecyk, 2010) on behalf of CEFIC, the European Chemical Industry Council determined emissions factors for different transport modes as detailed below in Table 1 as current best estimates for the European transport sector. Table 1. Carbon intensity factors for different transport modes Transport Mode Road Rail (average figure) Short sea Deep-sea container Deep-sea tanker Air

Emission Factor / g CO2 per tonne-km 62 22 16 8 5 602

Short-sea or coastal shipping has the potential to lower emissions from feeder movements quite considerably. There are however issues with the use of short sea shipping, certainly within the context of the UK that need to be considered. First, the UK Government’s road building programme over the past 50 years has encouraged the development of plentiful low cost road transport. Second, as many industries have 2

Commission for Integrated Transport op. cit.

14

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

sought to reduce inventory investment through just-in-time practises the need for more frequent and smaller deliveries has become the norm; a requirement that is more easily satisfied by road. Third, much of the UK’s port infrastructure has evolved to support the transhipment of goods from ocean going vessels to road or rail. Transhipment facilities for transferring containers to a coaster or barge are not as readily available in the UK. Finally, it has been reported (Saldanha and Gray, 2002) that the short sea shipping industry in the UK is a low-profile industry that does not actively market itself as part of a multi-modal transport service and as a result it remains largely invisible to much of the shipper community. While the use of shortsea shipping makes sense in terms of emissions, its use will usually add an extra link to the transport network making it only suitable for use where port facilities are close by. Although much shorter than the deep-sea voyage, feeder-related emissions can still represent a significant proportion of the door-to-door total. The choice of feeder mode often rests with the shipping line, though can be influenced by the shipper. Table 2 presents the emissions resulting from a hypothetical movement of a 40 foot container carrying a 22 tonne3 payload between Glasgow and Shanghai through the port of Felixstowe using road and rail feeder services and based on the emissions factors shown in Table 1. This type of journey is not unusual given that both of the UK’s major container ports (Felixstowe and Southampton) are located in the South of England. Table 2. Relative feeder movement emissions for a 22 tonne consignment moving between Glasgow and Shanghai. Movement Feeder (Road) Feeder (Rail) Deep-sea

Distance / km 670 670 19080

Emission Factor / g CO2 per tonne-km 62 22 8.4

Total CO2 / kg 914 324 3526

Fraction of total / % 21 8.4 -

It can be seen that switching from road to rail feeder services could in such a case reduce carbon emissions from transport by 13%. If low carbon electric rail traction was used, the carbon reductions would be higher still. In the UK, where road transport is dominant, the exploitation of lower carbon intensity feeder service movements could yield a significant reduction in transport emissions. It is clear that energy consumption comes at a cost. Better utilisation of energy can lead to a reduction in emissions per unit load moved. This is frequently accompanied by a reduction in cost. UK retailer Boots in conjunction with Maersk reduced the amount of air transport used for their inbound supply chain from the far East. The results (Barnes, 2008) are presented in Figure 2. The figure shows improvements made between 2004 and 2007 in terms of CO2 reduction per cubic metre of goods moved. The emissions were reduced by 29%. In addition the associated costs were reduced by 21%. If and when energy prices rise, the economic benefits of using lower carbon intensity transport will increase further.

3

Typical value, from McKinnon and Piecyk op cit. 15

Current Issues in Shipping, Ports and Logistics

Figure 2. Summary of carbon reduction for Boots Far East inbound supply chain

4.2. Choice of carrier According to the Clean Cargo benchmarking exercise (Clean Cargo Working Group, 2008), CO2 emissions per TEU-km vary across a sample of ten major lines between 76g and 118g. Within each of the main feeder modes, there can also be substantial differences in the carbon intensity of individual carriers. Carrier selection can therefore have a significant impact on total carbon emissions from companies global supply chain. A significant factor that determines the length of haul is that of the choice of port, both for loading and unloading. The shippers’ main priority is moving goods from factory to consumer at an acceptable cost. In practice, the shipper has a relationship with a carrier or freight forwarder, but ultimately it is the carrier that chooses the port and hence determines the specific route used. The carrier will often make a considerable strategic investment in port infrastructure, perhaps building a dedicated terminal to service its ships. The high capital expenditure required will mean that the carrier is usually tied to a particular port for a length of time; i.e. the choice of port is a decision that is infrequently made, and once made is unlikely to be altered in the short to medium term. Although the choice of port is one area that is outside the shippers’ direct sphere of influence, the choice of carrier clearly can be controlled and so the shipper can have some indirect influence over the choice of port, although doing so may involve a fundamental change in the carrier selection process. 16

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

4.3. Average handling factor In their passage through the supply chain, consignments are loaded on and off vehicles several times. As a consequence the ratio of the tonnes-lifted statistic to the weight of goods produced (also known as the ‘handling factor’) can serve as a crude measure of the number of links in the maritime supply chain. Shipping goods directly from a UK deep-sea port, rather than trans-shipping them through a European mainland port, effectively removes a link from this chain. Port-centric logistics and ‘distribution centre (DC) bypass’ strategies can also streamline the landward supply chain, reducing CO2 emissions. There is relatively little data in the public domain on carbon emissions from port activity, making a comparison difficult to perform. However, research conducted by the EPA in the United States of America (EPA, 2004) suggests that in-port emissions could be significant. They estimate that emissions from on-dock equipment (trucks, locomotives and handling equipment) may be as high as 33 – 50% of emissions associated with visiting ocean going vessels. The nature of international shipping, particularly with regard to operating under a flag of convenience makes emissions from ocean going vessels difficult to control. However, emissions arising from shoreside equipment fall into the host nation’s national inventory and therefore present an opportunity to reduce emissions. If these emissions are as high as the EPA’s research suggests then the carbon savings could be significant and certainly warrant investigation, although it is worth remembering that data collection is unlikely to be straightforward since much of the equipment operated within a port setting will be owned and accounted for by the terminal operators and not the port owners. In recent years, container ships have become ever larger as carriers have sought economies of scale. At the same time ports have expanded their businesses to cope with the demand of global trade. The result is often crowded terminals and congested road networks surrounding port facilities. The location of a port is usually fixed as a function of physical geography; there are only a limited number of places suitable for deep water berthing. One potential solution to this problem is the use of a ‘dry port’. This scores well on two counts. First, the demand for road transport in the vicinity of the port will be greatly reduced. Second, the use of an inter-modal link between the sea port and the dry port provides a much less carbon intensive transport solution. The decision to develop a dry port is clearly outside the scope of the shipper community, although in a world where the implications of carbon are becoming an important issue, the users of shipping services could favour a lower carbon option. There is also a strong case to make for port-centric logistics. Ports are increasingly becoming more than the traditional trans-shipment hubs that they once were. In recent years, many of the world’s larger port operators have developed their port centric logistics activities to include co-located distribution centre facilities. Given the trade imbalance between Europe and the Far East, locating distribution centres within port infrastructure can be beneficial. By emptying containers at the port, the number of empty container miles can be reduced. Sainsbury PLC is reported (Mangan and Lalwani, 2008) to have saved 140 road miles per TEU entering the UK via its distribution centre at Felixstowe. 17

Current Issues in Shipping, Ports and Logistics

4.4. Average length of haul This is the mean length of each link in the supply chain and essentially converts the tonnes-lifted statistic into tonne-kms. In a maritime context, the two key links are the feeder movement to the deep-sea port and deep-sea voyage. CO2 emissions from the former are a function of the shipper’s port, mode and carrier choices. Shippers can also influence the deep-sea routing through their overseas distribution or import strategies and choice of shipping line. A service from the far-East to the UK via Suez is approximately 6,000km shorter than via Panama or the Cape of Good Hope. Routing through Suez in this case has the potential to reduce transport emissions by approximately 30%. In recent years, concern has been expressed about the effects of globalization on the average distance that products are transported by sea and other modes. The so-called ‘food miles’ debate has highlighted the environmental problems created by the wider source of food products. One obvious way in which shippers could decarbonise the shipping sector would be to confine their sourcing and distributing to smaller areas (i.e. ‘relocalising’). The research is not examining this option as it raises wider issues about economic development, the liberalization of international trade and the total life-cycle emissions of products sourced from different locations. The study is taking the pattern of international trade flows as given and exploring ways of reducing carbon emissions per tonne-km rather than in absolute terms.

4.5. Average loading of laden container This is usually expressed as the average payload weight, though as lower density consignments ‘cube’ out before they ‘weigh’ out, it is desirable also to take account of freight volumes. The loading of container ships and their carbon intensity is often measured in terms of container (TEU) numbers, ignoring the extent to which containers are actually loaded. This loading is something that shippers can strongly influence through sales, pricing, ordering and packaging practices, their choice of container sizes and willingness to group consignments. Major retailers can have control over the loading of imported goods into containers in foreign markets, in some cases picking and packaging them in store-ready formats. The maximum load weight of containers is also partly constrained by the choice of feeder mode and is higher on rail and ships than on the road network. The use of a port-centric logistics strategy could effectively enhance container loading, by allowing containers to arrive at the port of destination fully laden and not constrained by the maximum gross road weight limit. Such an approach by Asda-Wal Mart resulted in filling an additional 70,000 m3 of container space resulting in 1,200 fewer TEU movements per annum.

18

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

4.6. Repositioning of empty containers This can be expressed as the distance that containers travel empty on deep-sea and feeder services. Empty movements represent a substantial under-utilisation of container carrying capacity on ships. UNESCAP estimates that roughly a fifth of the containers passing through international container ports are empty. This proportion can be much higher on corridors with a pronounced directional imbalance in traffic flow. It is particularly important issue for the UK, given its large deep-sea trade imbalance. The large trade imbalance between the Far East and the West has been a factor in world shipping for several decades. In the late 1960s, shipping giant Sealand responded with the introduction of a round the world liner service in an attempt to minimise empty container movements. While partially successful, the scale of the trade imbalance will inevitably mean that some empty container movements are necessary. More recently, the use of collapsible containers has been proposed (Shintani et al., 2010) as a solution to the trade imbalance problem, although at present the collapsible container is not in commercial use, and furthermore, their introduction is likely to require specific handling equipment. The UK Department for Transport has published statistics (DfT, 2008) that report that 52% of all containers leave the UK empty. Despite this large number of empty outbound container movements anecdotal evidence obtained from the Scotch Whisky industry (which relies heavily on its export markets) suggests that it can be difficult to obtain sufficient empty containers to meet the demand for outbound services. If this evidence proves typical then it would suggest that there is a lack of communication between the carriers and the UK shippers. While finding backloads is generally to be considered the responsibility of the shipping line, shippers can help to create backhauling opportunities by reconfiguring their supply chains. Providers of feeder services and forwarders can also use triangulation more effectively to exploit backhaul capacity. Where there is a pronounced traffic imbalance between regions A and B, a carrier can increase the probability of obtaining a backload by routing the returning vehicle via region C.

4.7. Energy efficiency Within the maritime supply chain most of the energy is consumed in the transport operation (kWh per tonne- or TEU-km), though allowance must also be made for energy used in container handling and ancillary activities at ports (kWh per tonne- or TEU). Shippers have little direct influence over the energy use by ships and ports, though can effect it indirectly through the logistical demands that they impose on maritime transport, in particular the required speed of delivery. The IMO has estimated that cutting average ship speed by 10% would cut CO2 emissions from shipping worldwide by 23%. The wider logistical effects of such deceleration need to be investigated, however. It is possible, for example, that shippers might switch to faster and more carbon-intensive feeder modes to compensate for the increase in deep-sea transit time. Alternatively, rather than using faster feeder services, the use of more deep-sea vessels can compensate for slow steaming. In the current economic downturn, this has been the approach adopted by the world’s major shipping lines. It 19

Current Issues in Shipping, Ports and Logistics

has been shown (Notteboom and Vernimmen, 2009) that a China – Northern Europe liner service typically requires 8 vessels in the loop. Reducing the vessel’s cruise speed by 10% raises this requirement to 9 vessels. Even deploying the additional vessel still results in a significant absolute CO2 reduction of 16%. In the current economic climate this increase in vessel numbers can be achieved from the existing fleet, but an economic recovery to pre-recession trade levels will require a significant ship building programme if the practice of slow-steaming is to be maintained without other changes to the supply chain configuration. However, there are signs emerging that slowsteaming is here to stay (Sustainable Shipping, 2010a). Shipping giant Maersk has indicated that it plans to continue slow-steaming even when the global economy recovers, since the current high fuel costs are not likely to come down. By reducing the responsiveness of replenishment systems it might also increase the demand for warehouse space, although scale of the issue is not clear. There is evidence to suggest that the slow-steaming results in improved service reliability which reduces the need for buffer stock in the supply chain which may offset any increase resulting from an increased lead-time. Total CO2 emissions from refrigerated containers also rise on slower voyages. The wider supply chain effects of what the World Economic Forum (2009) calls ‘despeeding’ need to be fully analysed.

4.8. Carbon intensity of the energy used This varies with the type of fuel used by the feeder modes and the nature of the electricity powering rail freight operations and port handling; these points have been discussed earlier in the paper. Consideration must also be given to opportunities for ‘cold ironing’ i.e. use of shore-side electricity by vessels while in port. It could be argued that the overall scale of maritime emissions is small, but it is clear that in the absence of effective control, they are set to grow in both absolute and relative terms. It is estimated (ICCT, 2007) that by 2020 in the EU, greenhouse gas emissions from shipping will exceed all land based emissions. Ships in port traditionally rely on their auxiliary engines to provide domestic power whilst alongside. The practice of ‘cold ironing’ enables a vessel to use a shore-side supply of electricity in place of this auxiliary power. If this power is drawn from a grid fed with low carbon electricity from nuclear or renewable sources, it can significantly reduce CO2 emissions. It should be emphasised that cold-ironing will not be effective in every case. It has been suggested (Sustainable Shipping, 2010b) that if China, which is heavily reliant on coal generated electrical power, was to switch to shore-side power, hotelling related emissions could increase by as much as 38%. The same study suggests hotelling emissions reductions to the order of 10-15% are typically achievable through the adoption of cold-ironing. There are a number of issues to overcome regarding the mode of power generation and the global standardisation of shore-side power. Shippers cannot influence this carbon intensity parameter directly though they can favour operators using lower carbon energy in their selection of carriers and ports. By helping to alter these eight critical parameters, directly or indirectly, shippers can decouple the quantity of goods they distribute from the related amount of CO 2 emitted by the maritime supply chain. 20

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

5 | Constraints on shippers’ effort to decarbonise maritime supply chains A lack of information about carbon emissions for various nodes and links in the chain make a comparison of the options difficult for the shipper to undertake. For each of the possible initiatives either led or approved by shippers we could consider the possible constraints: 5.1 Switch to lower carbon transport modes for feeder service Switching container movements from road to rail or water is clearly desirable from a carbon perspective and evidence suggests that the movements need not be long in order to be cost-effective from a financial perspective (DfT, 2009) although the relatively high fixed costs usually require large volumes to make the service economically viable. However, road transport has a number of advantages from the shippers’ perspective. First, it is convenient. Every business will be connected to a road network, but most will not have direct rail or water access, meaning that a road journey is required to move goods to the railhead or port. Second, road transport is very price competitive, especially in the current market, and particularly where volumes are small. Road transport also offers greater flexibility than rail or water. The running of additional trucks to meet a short-term need is relatively simple and absorbed into the road traffic network. Rail requires dedicated and specific network time and has to operate within the existing rail timetable. Water transport is reliant on berthing infrastructure to load and unload. In both cases, the network restrictions mean that rail and water transport is far less flexible than road. 5.2 Switch to carriers with lower carbon-intensity values on feeder and deep-sea services While it is clear that rail or short sea feeder movements will offer a carbon emission reduction compared to road, the data required to make a modal comparison is fragmented and lies in the domain of various different enterprises in the supply chain. Shippers lack access to benchmark data on shipping lines CO2 emissions. Further, infrastructure constraints and network capacity would make moving to low-carbon feeder services impractical in many cases. 5.3 Improving container loading both on export and import consignments The use of a port centric logistics approach, or rail / short sea feeder services are options for improving container loading although additional investment or a fundamental change in supply chain configuration may be required. 5.4 Rerouting of containers to minimise CO2 emissions from feeder and deep-sea services Shippers usually leave routing decisions to the carrier or freight forwarder and do not generally have an active role in the choice of route.

21

Current Issues in Shipping, Ports and Logistics

5.5 Adjusting logistical systems to accommodate longer maritime transit times At face value, extended maritime transit times require more inventory in the supply chain. However, the pipeline inventory is not the whole story. There is also the issue of buffer stock. On the one hand, extending the delivery lead-time will increase the required buffer stocks. However, it has been reported that slow steaming has had a positive impact on liner service reliability, i.e. less variation in delivery time. This reduced variability has the opposite effect and will reduce buffer stock requirements. Research by Saldanha et al. (2009) make the important observation that the value density of the goods in transit has a significant impact on logistics costs. For a low value density product (they use a figure of $6,750 per TEU), pipeline stock typically represents 3.5% of logistics costs. If the value density is high ($220,000 per TEU) the portion of the logistics costs rises to around 35%. So, while slowing the deep-sea movement will reduce carbon emissions, it can add significantly to inventory and total logistics costs. These cost trade-offs are important considerations when assessing the wider supply chain implications of low steaming. The new study will model trade-offs that shippers have to make in implementing these measures. Where possible, these trade-offs would be monetised and estimates made of the implicit prices of a tonne of CO2 that would be required to incentivise adoption of the measure.

6 | Conclusions The shipper, as the purchaser and user of transport services can have influence on the scale of emissions arising from the physical distribution network. This research proposes a conceptual framework (defined in Figure 1) to measure the use of carbon in maritime supply chains and will go on to measure the effect of the different supply chain decisions on carbon emissions. The research focused mainly on the movement of containerised goods. The framework considers eight key parameters that affect level of carbon emissions across the maritime supply chain. The shipper is not in a position to exert a direct influence on all of the parameters; for example, the shipper may choose the carrier, but the carrier will determine the specific details of route. There are areas however over which the shipper may have a direct influence. In this category, the nature of the port feeder movements can be critical, particularly in the UK where road haulage, a relatively carbon-intensive mode predominates. Given infrastructural constraints a wholesale modal shift away from road is neither feasible nor desirable, but where practical the reduction of CO2 emissions could be significant. There is lack of easily accessible benchmarking data. This not only applies to the shipper but virtually every other player in the maritime supply chain. Emissions data, where available, usually resides within the enterprise responsible for those emissions. This research will attempt to provide a measure of emissions corresponding to the 8 key parameters. The research must also consider the cost-benefit trade-offs that will 22

Woolford and McKinnon – Role of Shipper in Decarbonising Maritime Supply Chains

result from making changes to the maritime supply chain; perhaps additional inventory to offset less frequent delivery schedules or infrastructure investments to make use of alternative feeder movements or implement a port-centric logistics strategy. Management of the maritime supply chain is essentially cost-driven. Although the triple bottom line is a consideration for many organisations, if the financial bottom line is not healthy then the business suffers. Therefore, if carbon is to become a serious consideration for the business community, two things must happen. First, carbon must be given an economic value. Second, the supply chain must be considered as a whole and not a series of independent processes. If these two conditions are fulfilled, an optimum supply chain configuration could be found. This research will examine the extent to which shippers can optimise, in carbon terms, their use of the maritime supply chain.

References Barnes, I. (2008) Carbon Auditing Supply Chains – Making the Journey. Presentation to the MultiModal 2008 conference, Birmingham Clean Cargo Working Group / BSR (2008) 2008 EPS Scorecard (2007 Results) Department for Transport (2008) Transport Statistics Report: Maritime Statistics 2008. London Department for Transport (2009) Choosing and Developing a Multi-modal Transport Solution, Freight Best Practice. London Environment Protection Agency (2004) Port Emission Inventories and Modeling of Port Emissions for Use in State Implementation Plans (SIPs), White Paper #3, EPA Contract No. 68-W-03-028 ICCT (2007) Air pollution and greenhouse gas emissions from ocean going ships. Washington, DC International Road Union (1997) Driving towards sustainable development, Geneva Mangan, J., Lalwani, C. (2008) Port-centric logistics. International Journal of Logistics Management, 19(1), 29-41 McKinnon, A., Piecyk, M. (2010) Measuring and Managing CO2 emissions. CEFIC Notteboom, T.E., Vernimmen, B. (2009) The effect of high fuel costs on liner service configuration in container shipping. Journal of Transport Geography, 17, 325-337 Saldanha, J., Gray, R. (2002) The potential for British coastal shipping in a multimodal chain. Maritime Policy and Management, 29(1), 77-92 23

Current Issues in Shipping, Ports and Logistics

Saldanha, J.P., Tyworth, J.E., Swan, P.F., Russell, D.M. (2009) Cutting Logistics Costs With Ocean Carrier Selection. Journal of Business Logistics, 30(2), 175-195 Shintani, K., Konings, R., Imai, A. (2010) The impact of foldable containers on container fleet management costs in hinterland transport. Transport Research Part E, 46, 750763 Sustainable Shipping (2010a) at http://www.sustainableshipping.com/news/i97270/High_speed_era_a_thing_of_the_ past, accessed, 21 Oct 2010 Sustainable Shipping (2010b) at http://www.sustainableshipping.com/news/i93159/Cold_ironing_can_increase_CO2_e missions, accessed 20-Oct-2010 WBCSD / WRI (2004) Greenhouse Gas Protocol: Corporate Accounting and Reporting Standards World Economic Forum / Accenture (2009) Supply Chain Decarbonisation. Geneva

24

CHAPTER 2 Shipping Companies’ Awareness and Preparedness for Greenhouse Gas Regulations: A Korean Case

 Sang-Yoon LEE and Young-Tae CHANG

Abstract The main objectives of this study are to investigate the level of awareness and preparedness of Korean shipping companies for the currently discussed GHG reduction issues in international shipping sector and to identify any significant difference between large and small/medium size shipping firms in terms of recognition and ability for green shipping regulation. In addition, it has been attempted to find any significant linkage between the level of awareness and preparedness of Korean shipping companies and their preferring market based measures. The empirical analysis employing a questionnaire survey methodology reveals that the degree of awareness of Korean shipping firms on overall green shipping regulation is low and the level of recognition and understanding on the major agenda and debating issues discussed at the IMO is also quite low. In particular, the degree of awareness of small and mediumsized shipping companies is significantly lower compared to the large firms. Concerning the level of preparedness or ability corresponding to green shipping shift, the Korean shipping firms seem to be in a very poor condition and the majority of small/mediumsized companies even have not been suited at the starting line yet while, on the contrary, the large firms have cumulated necessary abilities even if not sufficient. In addition, it has been found that the majority of Korean shipping firms support carbon taxation rather than ETS as market based measure regardless of their size and the level of awareness and preparedness.

25

Current Issues in Shipping, Ports and Logistics

1 | Introduction Global warming is one of the most challenging issues facing human race because it requires long-term, broad and complicated processes and involves many different viewpoints and political positions. After the announcement of the United Nations Framework Convention on Climate Change (UNFCCC) in 1992, many efforts have been made to reduce greenhouse gas (GHG) emissions across all the industries over the world. Although the Kyoto Protocol (1997) required legally binding GHG emission targets to the industrialized countries for their domestic emission sources, there has not been explicit progress on the GHG emissions from international shipping, which is estimated to have emitted 870 million tons, about 2.7% of total global emissions of CO2 in 2007 (IMO, 2009). The Kyoto Protocol had delegated GHG mitigation framework for international shipping to the International Maritime Organization (IMO) and its sub-organization, the Marine Environment Protection Committee (MEPC) has designed various measures and mechanisms to be adopted. It is anticipated that the technical, operational and market based measures proposed by the IMO are likely to be adopted in the foreseeable future with the method of adding to MARPOL (International Convention for the Prevention of Pollution from Ships) Annex VI or making new conventions. Korea has a 2.62% share of world trade in terms of value and possesses 3.63% ownership of fleet in terms of dead weight tonnage (dwt) in 2008, ranked 5th position in the world (UNCTAD, 2008 cited in IMO, 2009); but Korean shipping companies just started to observe this serious movement and are not deemed to have responded properly. The main objectives of present study are to investigate the level of awareness and preparedness of Korean shipping companies for the currently discussed GHG reduction issues in international shipping sector and to examine if there is any significant difference between large and small/medium size shipping firms in terms of recognition and ability for green shipping regulation. Moreover, it aims to find out if there is any significant linkage between the level of awareness and preparedness of Korean shipping companies and also if there is any preferred market based measure by the companies. Once the new measures to regulate GHG emissions from international shipping are adopted, these are likely to bring about enormous impacts on the technical structure of ships and operating practices as well as the long-term business strategy setting of shipping companies. Therefore, this can be a new paradigm shift in the international shipping industry. Awareness is the initial condition and stage that organizations select

26

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

a proper direction and formulate an effective strategy. Consequently, it should be a very critical issue to examine how well and how much shipping companies recognize and understand the upcoming paradigm shift pertinent to the GHG regulation. Nevertheless, there has been no empirical research on shipping firms’ awareness and preparedness for green shipping regulation thus far - only one interim paper by Giziakis and Christodoulou (2010) was presented in a recent International Association of Maritime Economists (IAME) conference targeting Greek shipping companies. Therefore, the current study is the first empirical research to examine the degrees of awareness and readiness of shipping companies for green shipping regulation in the Korean context and, in consequence, to identify the gaps between policy makers and industry players in terms of viewpoint, recognition and implementation.

2 | Literature review: main trend and critical points The United Nations Framework Convention on Climate Change (UNFCCC) was signed in 1992 and entered into force in 1994. Under the convention, parties collect and share data, launch national strategy to mitigate emissions and cooperate for the adaptation to climate change (IMO, 2009). The Kyoto protocol (1997) contains legally binding greenhouse gas emission targets and mandatory limits on greenhouse gas emissions for 37 industrialized countries and the European community. These so called Annex I parties should reduce their overall GHG emission level by at least 5% below 1990 levels between 2008 and 2012. The Kyoto protocol proposes three mechanisms to achieve cost effective emission reduction, which are international emission trading (IET), the joint implementation (JI) program, and the clean development mechanism (CDM). In an IET system, a developed country which will emit more than its assigned amount is allowed to buy GHG emission reductions from another developed nation which emits less than its assigned amount (Article 17); while in a JI program, an Annex I party may purchase emission reduction units (ERUs) by carrying out a JI project within the Annex I region (Article 6). In addition, a developed nation may acquire the certified emission reductions (CERs) from nonAnnex I countries by means of a CDM project (Article 12) (Woerdman, 1999). It should be noticed that CDM has different aspect from IET and JI since it was designed to reduce the GHG abatement cost by allowing developed nations to reduce emissions in developing countries, where GHG mitigation could be achieved with relatively cheaper means. In the real world, it is reported that by the end of 2004, 120 CDM transactions had been recorded and more players moved into the CDM market especially after the Marrakesh Accord in 2001 (Wang and Firestone, 2010). Researchers have presented

27

Current Issues in Shipping, Ports and Logistics

different perspectives about the effectiveness of those mechanisms. Woerdman (1999) has argued that JI and CDM are more effective, efficient and politically acceptable than IET, while some authors have asserted that the IET system is more environmentally sound and cheaper than JI or CDM (Tietenberg, 1992; Grubb et al., 1998; Zhang and Nentjes, 1999 cited in Woerdnam, 1999). Although the Kyoto Protocol requires nations to establish CO2 mitigation policy proposals for domestic land-based emissions sources, there has been little progress in the international shipping and aviation sectors (Corbett et al., 2009). International shipping and aviation are not covered by the Kyoto Protocol or the Copenhagen Accord, due to ‘lack of reliable emission data and lack of an agreed approach for defining responsibility by country’ (SBSTA/INF.2 2005 cited in Giziakis and Christodoulou, 2010). The United Nations has delegated the reduction framework of greenhouse gas emissions from international shipping to the IMO. According to the article 2.2 of the Kyoto Protocol, “the parties included in Annex I shall pursue limitation or reduction of emissions of greenhouse gases not controlled by the Montreal Protocol from aviation and marine bunker fuels, working through the International Civil Aviation Organization and the International Maritime Organization, respectively.” The IMO has 166 member states and its governing body is the Assembly which meets every two years. In between Assembly sessions a Council, consisting of 32 member states elected by the Assembly, acts as the governing body. The technical and legal work is carried out by five committees which are the Maritime Safety Committee, the Marine Environment Protection Committee (MEPC), the Technical Co-operation Committee, the Legal Committee, and the Facilitation Committee. Among them, MEPC is responsible for drafting relevant regulations to prevent ships from polluting the ocean and the atmosphere (Lin and Lin, 2006; Stopford, 2009). MEPC has been developing technical and operational measures to mitigate GHG emissions from international shipping and also proposed market‐based instruments to provide incentives for the shipping industry to comply with these measures (IMO A.963(23) 2003; MEPC 59/4 2009 cited in Giziakis and Christodoulou, 2010). The technology based approaches, so called technical measures, include propeller redesign, anti-fouling measurers for hulls, low-carbon fuels, and improved engine operations. However, limitations of these measures have required operational changes and demand management to achieve mitigation targets in a more cost-effective manner. Slow steaming and voyage optimization are typical examples of the operational approaches or operational measures and some simulation models present

28

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

significant reductions in CO2 emissions (e.g. Corbett et al., 2009) and fuel costs (e.g. Notteboom and Vernimmen, 2009). According to Corbett et al. (2009) emissions reduction across a range of containership routes can be reached up to 70% when the speed is halved. As for the market-based measures, two main instruments, namely international fund (carbon tax) and emission trading scheme (ETS) have been proposed, but there are very controversial view points about which measure is more effective for the international shipping sector. In addition, there has been sharp conflict in discussion on the decision whether the proposed measures can be embraced in MARPOL Annex VI or not. MARPOL is the main international convention in IMO covering the prevention and minimization of pollution by ships from operational or accidental causes. It is a combination of two treaties adopted in 1973 and 1978 and updated by amendments. It currently has six technical annexes, among which Annex VI deals with ‘prevention of air pollution from ships’ (Stopford, 2009). The conflict in IMO stems from fundamentally different positions between developed nations asserting ‘no more favorable treatment (NMFT)’ and developing countries advocating ‘common but differentiated responsibility (CBDR)’. The principle of CBDR under the UNFCCC recognizes the differences of countries in the contributions of GHG reduction according to their responsibility and capability. However, it is asserted by the developed nation group that the IMO regulatory framework on GHG emissions should be applicable to all ships, irrespective of the flags they fly since about three quarters of the world tonnage of all merchant vessels engaged in international trade is registered in non Annex I countries of the Kyoto Protocol, therefore, it has limited effect for any regulatory regime to act only on the remaining portion (IMO, 2009). The principle of NMFT is a fundamental condition when an IMO instrument has entered into force, and ‘port state control (PSC)’ is often used to enhance more global compliance efforts regardless of the flags of ships. In other words, the PSC system can control and investigate any suspicious vessels entering the territory of the party states to IMO convention irrespective of whether the nation of the flag vessel is a party to the convention or not. Therefore, one possible case could be considered that the proposed measures will be included in MARPOL Annex VI by developed nations and implemented with PSC in their territorial seas. The United Nations Convention on the Law of the Sea (UNCLOS 1982) allows coastal states to legislate for the ‘good conduct’ of ships in their territorial seas. The Convention lists specific cases in which legislation is permitted, which are related to ‘safety of navigation’, ‘protection of navigational aids’, ‘preservation of the environment and prevention’, ‘reduction and control of pollution’, and ‘the prevention of

29

Current Issues in Shipping, Ports and Logistics

infringement of customs and sanitary laws’, etc. The port state control movement was caused by the increasing number of ships registered under the flags of convenience and the recognition that some of these flags were not strictly enforcing international maritime regulations. In the real world, regarding current MARPOL conventions, some maritime countries adopt port state control. An ocean-going ship may be inspected, and even punished, if the convention has been violated during its stay in the harbor of a port state country (Stopford, 2009). As a matter of fact, this kind of action was taken by the European Union for their international aviation sector. In 2008, the EU decided to include international aviation in the already existing EU-ETS market. From 2012, allowances will be required for all international flights landing at, and departing from, any airport in the EU. Some studies have been conducted to investigate different effects of this action on the European aviation industry, which have presented some negative aspects. For instance, Scheelhaase et al. (2010), through their empirical estimation, have asserted that the European network carriers are likely to encounter competitive disadvantage after the introduction of EU-ETS, and Mayor and Tol (2010) have examined the changes in international flows of tourists and showed that the tourist industry shrinks in the EU affected by the price increase with very small amount of emission reduction. From the literature, it is clear that no studies have been conducted yet to investigate the awareness and preparedness of shipping companies on upcoming regulatory measures on the international shipping. Accordingly, this study intends to fill the gap in the literature.

3 | Research methodology 3.1 Questionnaire survey process 3.1.1 Questionnaire design The questionnaire for the survey was developed in the following phases. First, after considering critical issues, measures and international regulations about greenhouse gas, which have been discussed in the United Nations and the International Maritime Organization, one survey form was designed consisting of three parts: general information about respondents and their companies; questions asking Korean shipping companies’ awareness of green shipping policy; and items measuring Korean shipping firms’ current ability for dealing with green shipping issues. Second, the questionnaire was pre-tested by visiting five shipping companies in Korea to check whether the expressions and terms on the form were easy for the respondents to understand and

30

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

answer, and also if any important questions were missing or not. Third, the survey form was revised to incorporate the findings from the pre-test. From the pre-test, it was found that some of the questions were not easy to answer because those items asked very detailed information requiring considerable amount of time to collect data, and also that other important items were missing in the form. Therefore, the questionnaire was modified to be answered with respondents’ general knowledge and perspectives, and to add the missing factors. Finally, the revised questionnaire was tested again (post-test) with the same number of companies as in the pre-test. This post-test ensured that all the expressions were clear and easy to understand and that all of the important items were included. Accordingly, the questionnaire was finalized containing 31 items for the awareness of green shipping policy, which are divided into conceptual issues and debating issues, and the preparedness issues for green shipping regulations, as shown in Table 1, Table 2 and Table 3, respectively. The questions were asked in ‘yes’ or ‘no’ or five point Likert scales, where point 1 is ‘absolutely not aware’ or ‘very bad’ and point 5 is ‘fully aware’ or ‘very good’. Table 1. Awareness for conceptual issues Conceptual issues

Basic knowledge

Principle and practical method under the U.N.

International shipping specific conventions and measures

AC1. Greenhouse gas AC2. United Nations Framework Convention on Climate Change AC3. Kyoto Protocol AC4. Copenhagen Accord AC5. Framework Act on Low Carbon, Green Growth (Korea) AC6. Common But Differentiated Responsibilities AC7. Carbon taxation AC8. Emission Trading Scheme AC9. Clean Development Mechanism AC10. Joint Implementation program AC11. The United Nations Convention on the Law of the Sea AC12. International Maritime Organization AC13. No More Favorable Treatment AC14. International Convention for the Prevention of Pollution from Ships AC15. MARPOL Annex VI AC16. Marine Environment Protection Committee AC17. Energy Efficiency Design Index AC18. Energy Efficiency Operational Indicator AC19. Market Based Measurement AC20. Port State Control

31

Current Issues in Shipping, Ports and Logistics

The conceptual issues were asked with twenty items. First, five questions are about basic knowledge on important international agreements for GHG emissions reduction and also corresponding domestic act in Korea, which was adopted recently in domestic transport with the title of ‘Framework Act on Low Carbon, Green Growth’. The following five items include the principle and practical methods of GHG emissions reduction proposed through international conventions. The next ten indices are related to the international shipping specific issues and measures for CO2 emission reduction. Meanwhile, Korean shipping firms’ awareness for currently debating issues were identified using eleven questions, which are mainly related with authority issues for CO2 reduction frameworks in the international shipping and aviation sectors, the conflict of two principles of the UN and the IMO, and the participation and implementation of MARPOL Annex VI. Questions on the debating issues were developed based on the circumstances of the IMO at the time of writing this paper. Table 2. Awareness for debating issues Debated issues AD1.The UN has delegated the reduction framework of GHG emissions from international shipping to the IMO AD2.The IMO should submit the reduction framework of GHG emissions from international shipping to UN until 2012 AD3.The authority of reduction framework of GHG emissions from international shipping can be transferred to UNFCCC in the future if IMO does not provide for it AD4.Joining MARPOL Annex VI is voluntary AD5.Inclusion of technical measure for new vessel design in MARPOL Annex VI has been debated AD6.Inclusion of operational measure about present vessel operation in MARPOL Annex VI has been debated AD7.Inclusion of market based measure in MARPOL Annex VI or new convention for market based measure has been debated AD8.Conflict of two principles of UN and IMO, i.e. CBDR and NMFT AD9.EU adopts ETS for GHG emissions reduction in international air transport AD10.Framework Act on Low Carbon, Green Growth adopts ETS for domestic transport in Korea AD11.Participants of MARPOL Annex VI can adopt PSC to non signing nations in the case of revision of MARPOL Annex VI

Table 3 presents twenty indices to measure the status of Korean shipping companies in terms of preparedness and ability for upcoming green shipping regulations, which are 32

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

benchmarked from various green shipping reports published by global container shipping groups such as Maersk, CMA-CGM, Evergreen, Hapag-Lloyd, Hamburg Sud, OOCL, NYK, MOL, K-line, Yang Ming, COSCO and Hanjin. Table 3. Preparedness and capability items Preparedness/Ability Contextual Aspect

P1.CEO's recognition of the importance of green shipping P2.Monitoring international trends in green shipping P3.Benchmarking advanced and/or competing shipping firms P4.Investment ability and/or financial source for green shipping P5.Pre-investment and experiences for green shipping P6.Corporate level strategic approach to CO2 emissions reduction P7.Establishing long-term target for CO2 emissions reduction P8.Publishing regular report on CO2 emissions reduction performance

Operational Aspect

P9.Measurement of emission amount from possessing vessels P10.Management of operations information for possessing vessels P11.Establishing specialized division for green shipping P12.Collaboration between divisions for green shipping issues P13.Qualified human resource for green shipping P14.Outsourcing of green shipping issues to specialized organizations

Strategic Aspect

P15.Establishing corresponding framework to technical measurement P16.Implementing corresponding framework to technical measurement P17.Establishing corresponding framework to operational measurement P18.Implementing corresponding framework to operational measurement P19.Establishing corresponding framework to market based measurement P20.Implementing corresponding framework to market based measurement

3.1.2 Data collection A survey targeting Korean shipping companies was conducted in order to empirically analyze the degree of Korean domestic ocean carriers’ awareness and readiness to green shipping regulations adopted under the initiative of the UN and the IMO. The questionnaire instrument previously explained was distributed through internet mail to 181 shipping firms registered in the Korean Ship Owner Association in mid May, 2010. After the first round of mailing the survey forms, a reminder was sent twice by fax and email and in some cases the research team called the company offices. In spite of all these efforts, only twenty-one valid responses could be collected after excluding duplicated responses from the same shipping companies; therefore, the response rate was merely 11.6%. However, this response rate could provide a basis for drawing out 33

Current Issues in Shipping, Ports and Logistics

generalizations as the respondents are the major players in the Korean shipping industry accounting for the bulk of the shipping market. In addition, it should be noticed that this low response rate may reflect small and medium-sized Korean carriers’ low interest and readiness for green shipping movements.

3.2 Research framework After the survey data were collected, a three-step research framework was designed to analyze the data, as shown in Figure 1. First, from the aggregated survey results, a descriptive analysis was implemented to identify the degree of awareness of Korean shipping firms regarding green shipping related issues categorized into conceptual issues and debating issues. In addition, it is attempted to measure how effectively Korean shipping companies are preparing for the upcoming green shipping paradigm using 20 items selected from world leading ocean carriers activities. The second step is to compare the level of awareness and readiness between large and small/mediumsized shipping firms. It is generally recognized that the large shipping carriers have much more advanced information and knowledge in contemporary issues and possess superior ability to prepare for and respond to them. At the second stage, it was tried to examine if there are significant gaps between the two groups in terms of green shipping awareness and preparedness. These comparative analyses may provide meaningful implications for the establishment and implementation of green shipping policy from the national perspective as well as from the international viewpoint. At the final stage, the relationships between Korean shipping firms’ preference on marketbased measures and the degrees of their awareness and preparedness are investigated. Economic theories advocate ETS is a more effective system compared to a carbon taxation system in terms of certainty and administrative costs, and the Kyoto Protocol initially selected ETS as an emission reduction mechanism. However, the shipping industry has its own specific characteristics and market structure and individual firms or nations may have different preferring positions. In addition, it has been argued that relatively advanced and large companies may prefer ETS to carbon taxation because ETS could provide new opportunities and competitive edge for them. The authors have tried to investigate whether the differences in the awareness and preparedness could exert any significant influence on shipping firms’ preferences on MBM or not using binomial logit model.

34

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

Figure 1. Research framework

Step 1

Descriptive Analysis on Awareness and Preparedness for Green Shipping Issues: Conceptual/Debating Issues and Capability

Step 2

Comparison of Green Shipping Awareness and Preparedness between Large and Small/Medium Size Firms: T-Test

Step 3

Relationships between Preferring MBM and Degrees of Awareness/Preparedness: Binary Logit Regression Model

4 | Survey results 4.1 Demographic profile of sampled companies This section mainly presents the demographic statistics resulting from the questionnaire survey in order to provide a general picture of survey participants and their responses to the questions. Chatfield (1985) has asserted that the initial data analysis is critical for most statistical investigations, not only for exploring and summarizing data, but also for model formulation employing more advanced statistical techniques at the later stage of the analysis process. Table 4 shows the characteristics of the respondents and the companies they represent. The characteristics of respondents were analyzed by identifying their work experience in the shipping industry. More than half of the respondents have work experience in the shipping industry for more than 10 years. Related to the respondents’ job grade, a total of 76.2% marked their positions in the managerial and executive categories. Considering the statistical characteristics of the respondents’ position and work experience, it was assumed that the respondents had sufficient knowledge about their firm’s activities to provide accurate and reliable information. Regarding company

35

Current Issues in Shipping, Ports and Logistics

information, the respondents were commonly asked to indicate their total sales value, the number of full time employees, main service type and area. Firstly, 80% of companies had total sales value of more than 10 million US dollars and 25% of them had total sales value of over one billion US dollars in 2009. Secondly, 38% of companies had less than 100 full time employees and 14% of firms had over 2000 full time workers in 2009. Thirdly, of the twenty-one companies, container shipping lines and dry bulk carriers were equally eight companies, respectively (38%*2 = 76%), and the remaining five companies (24%) were oil and chemical carriers. Regarding main service region, the share of intra Asia service was 67% while the long haul service shared 33%. This indicates that the majority of responding companies were small and medium-sized firms mainly providing short sea shipping service. Table 4. Descriptive statistics of sampled companies Industry Experience

Full time employees

Total sales value

Main service

Years

Share

USD Mil.

Share

Employee

Share

Types

Share

2000

14%

Sum

100%

Sum

100%

Sum

100%

Figure 2. Response firms’ main service region 35%

30%

25%

20%

29%

10%

19%

19%

3%

3%

2%

2%

Russia

2%

Africa

2%

Mediterranean Sea

Western Europe

Central/South America

North America

South East Asia

Taiwan

Japan

China

36

2%

Eastern Europe

5% 2%

0%

Australia

10%

5%

Middle East

15%

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

Next, in order to identify some managerial characteristics of the response firms, several vessel operating items were asked: the numbers of present and newly ordered vessels, the average age of present vessels and the proportion of ballast voyage. First, 40% of sampled companies were small firms operating less than 10 vessels and 40% were medium-sized carriers operating 11 to 50 vessels, and the other 20% were large carriers operating more than 50 vessels. Second, 70% of sample firms reported the average age of their vessels was between 5 to 15 years and the other 30% ranged between 15 and 20 years. Third, 76% ordered 5 to 10 new vessels and 18% expected delivery of 10 to 20 newly built vessels. Concerning ballast voyage proportion, 43% of carriers (mainly container lines) answered around 10% while 44% (mainly dry and liquid bulk carriers) answered more than 30%. The high rate of ballast voyage implies inefficient fleet management and may give disadvantage to the carriers when they correspond to the operational measurement, i.e. EEOI. Table 5. Descriptive statistics of sampled companies Number of present vessels

Average age of present vessels

Number of new building vessels

Proportion of ballast voyage

Vessels

Share

Year

Share

Vessels

Share

Proportion

Share

200

10%

>25

0%

>30

0%

Sum

100%

Sum

100%

Sum

100%

4.2 Descriptive analysis on awareness and preparedness 4.2.1 Awareness of green shipping policy The average score of each awareness variable about conceptual issues was calculated and plotted on a graph shown in Figure 2, where, point 1 means ‘absolutely not aware’, point 2 ‘hardly aware’, point 3 ‘slightly aware’, point 4 ‘moderately aware’ and point 5 ‘fully aware’. The results reveal that Korean shipping firms’ awareness level for the basic knowledge about some critical international/domestic agreement/regulation is marked around ‘slightly aware’ while the degree of recognition for international shipping related organization and regulations such as IMO, MARPOL, PSC is positioned 37

Current Issues in Shipping, Ports and Logistics

around ‘moderately aware’. However, the awareness for more specific and technical regulations and measures such as JI, CDM, UNCLOS, MEPC, technical/operational/market based measurement is revealed very poor.

and

Figure 3. Average scores on conceptual awareness items 5.0

4.1

4.0

4.0 3.7

3.7 3.2

3.0

3.6 3.3

3.2 3.0 2.7

3.0 2.9 2.6

2.6 2.2

2.0

2.6

2.6

2.5

2.2 2.2

1.0 AC1

AC2

AC3

AC4

AC5

AC6

AC7

AC8

AC9 AC10 AC11 AC12 AC13 AC14 AC15 AC16 AC17 AC18 AC19 AC20

For currently debated issues under the MEPC, Korean carriers have moderately monitored main trends, but do not recognize well the possibility that “the authority of reduction framework of GHG emissions from international shipping can be transferred back to the UNFCCC if the IMO does not provide it”. In addition, it is also not recognized by the Korean carriers that that “EU adopted ETS for GHG emissions reduction in international air transport” and “Framework Act on Low Carbon, Green Growth in Korea adopted ETS as well for domestic transport”. This result reflects the lack of strategic awareness and preparedness for disputed market-based measures among the Korean shipping companies. Figure 4. Awareness for debating items AD1 AD2 AD3 AD4 AD5 AD6 AD7 AD8 AD9 AD10 AD11 0%

10%

20%

30%

40%

50%

Know

38

60%

Don't know

70%

80%

90%

100%

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

From the above, it can be interpreted that Korean shipping firms have not realized well the importance and seriousness of CO2 emission measures which will be internationally adopted in the foreseeable future. In particular, the degree of awareness of small and medium-sized carriers for green shipping issues is quite low. 4.2.2 Preparedness for green shipping policy Before answering the questions about the degree of readiness and ability for green shipping regulations, the carriers were asked to evaluate the anticipated impact of GHG emission measures on their competitiveness. The majority (63%) answered that these measures would impose a burden and weaken their competitiveness while only three carries answered they could utilize green shipping issues as a new source of competitiveness. The remaining four carriers evaluated that these measures could give a burden on operating costs but would not exert any negative influence on their competitiveness. Figure 5. Average scores on green shipping capability items 5.0

4.0

3.7 3.2 3.0

2.8

2.8

2.7

2.6

2.5

2.3 2.0

2.0

2.0

1.8

2.1

2.1

2.1

2.3

2.1

2.2

2.1

2.0 2.0

1.0 P1

P2

P3

P4

P5

P6

P7

P8

P9

P10

P11

P12

P13

P14

P15

P16

P17

P18

P19

P20

With respect to green shipping capability items, 20 items were marked on a five-point Likert scale from ‘very bad’ (1) to ‘very good’ (5). For those indices Korean shipping firms expose a very poor score - only two items ‘CEO's recognition of the importance of green shipping (3.2)’ and ‘management of operations information for possessing vessels (3.7)’ were above average while the other 18 items were between point 1.8 and point 2.8. Korean carriers showed an extremely poor position on the items ‘preinvestment and experiences (1.8)’ and ‘regular report (2.0)’. In addition, the scores of overall items standing for operational and strategic aspects were around point 2, which implies that Korean shipping companies could not be expected to correspond effectively to international green shipping measures and regulation unless the current position would be improved in the short term.

39

Current Issues in Shipping, Ports and Logistics

4.3 Comparison between large and small/medium size firms A total of 21 respondents were categorized into two groups, i.e. five large carriers with more than one billion USD in sales value in 2009 and sixteen small/medium-sized shipping companies with less than that amount. The categories were created in order to analyze any significant differences in the levels of awareness and preparedness for green shipping issues between the two groups. The comparison presents clear differences between large carriers and small/medium-sized companies. The degree of awareness of small/medium-sized shipping firms was significantly lower compared to large carriers. Different from previous analysis targeting all companies, the level of awareness of large shipping carriers was very high across all items. In contrast, small/medium-sized companies were marked below point 3 on 14 indices and, in particular, below point 2 on EEOI and MBM. For a more rigorous comparison between the two groups, a t-test method was used. The t-test results showed that there existed statistically significant gaps in the degree of awareness for green shipping issues between the two groups. Figure 6. Comparison of awareness for conceptual issues between shipping firms AC1

4.3

2.9

AC3

4.0

3.0

AC4

4.0

2.3

AC5 AC6

4.4

2.6

4.8

1.9

AC7

4.8

2.4

AC8

4.4

2.4

AC9

1.9

AC10

1.9

AC11

1.9

3.2 3.2 3.0

AC12 AC13

4.6

2.0

5.0

3.3

AC15

5.0

3.1

AC16

4.6

2.8

AC17

2.0

AC18

2.0

AC19

2.0

4.6 4.6 4.2

AC20

4.8

3.8 2.0

3.0 Large

40

5.0

3.8

AC14

1.0

4.6

3.4

AC2

Small/Medium

4.0

5.0

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

In addition, all the large carriers recognized the detailed agenda and disputed points in green shipping issues currently discussed in the IMO and the MEPC. However, the small/medium-sized shipping companies did not sufficiently follow up the proposed issues. As regards carriers’ evaluation for CO2 reduction regulation which will be adopted in international shipping, 75% of small/medium-sized carriers were worrying about some negative effects on their competitiveness. Large carriers mainly expected no negative influence or considered this development as a potential new source of competitiveness. Table 6. Independent samples t-test for green awareness between shipping companies Levene's Test for Equality of Variances

t-test for Equality of Means

F

Sig.

t

D.F.

Sig.

AC1. GHG

0.434

0.518

3.484

19

0.002

AC2. UNFCCC

0.686

0.418

3.220

18

0.005

AC3. Kyoto Protocol

1.357

0.258

2.456

19

0.024

AC4. Copenhagen Accord

1.314

0.266

4.245

19

0.000

AC5. Framework Act (Korea)

0.534

0.474

5.851

19

0.000

AC6. CBDR

1.452

0.243

6.139

19

0.000

AC7. Carbon taxation

4.245

0.053

5.409

19

0.000

AC8. ETS

1.900

0.184

4.291

19

0.000

AC9. CDM

1.292

0.270

2.487

19

0.022

AC10. JI

2.106

0.163

2.617

19

0.017

AC11. UNCLOS

1.445

0.245

2.718

18

0.014

AC12. IMO

5.209

0.034

6.333

15

0.000

AC13. NMFT

0.195

0.664

5.674

19

0.000

AC14. MARPOL

6.950

0.016

6.260

15

0.000

AC15. MARPOL Annex VI

8.134

0.010

6.536

15

0.000

AC16. MEPC

1.670

0.213

4.010

18

0.001

AC17. EEDI

0.007

0.933

7.293

19

0.000

AC18. EEOI

0.007

0.933

7.293

19

0.000

AC19. MBM

0.299

0.591

5.695

19

0.000

AC20. PSC

3.715

0.069

2.247

19

0.037

Note: Equality of variances are not assumed in IMO, MARPOL and MARPOL Annex VI

41

Current Issues in Shipping, Ports and Logistics

Figure 7. Comparison of position for green shipping regulation between shipping firms 80%

75%

70% 60% 50%

40%

40%

40%

30%

20%

20%

19%

10%

6%

0%

Additional costs and weaken competitivenss

Additional costs but not weaken competitivenss Large

New source of competitivness

Medium/Small

Figure 8. Comparison of green shipping capability between shipping firms 5.0

P1

2.6

P2

2.3

P3

2.3

P4 P5

4.2 4.2 4.0

2.3 2.8

1.5

P6

3.8

2.2

P7

3.4

2.0

P8

2.8

1.7

P9

3.8

2.1

P10 P11 P12 P13

3.6

1.6 3.4

1.7 3.2

1.8

P14

1.8

P15

1.8

P16

3.8 3.4 3.6

1.8

P18

1.8

P19

1.8

1.0

3.2

1.8

P17

P20

3.4 3.0 2.8

1.7 2.0

3.0

4.0 Large

42

4.6

3.4

Small/medium

5.0

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

Table 7. Independent samples t-test for green shipping capability between shipping companies Levene's Test for

t-test for Equality of

Equality of Variances

Means

F

Sig.

t

D.F.

Sig.

0.158

0.696

3.727

19

0.001

0.686

0.418

3.220

18

0.005

1.357

0.258

2.456

19

0.024

1.314

0.266

4.245

19

0.000

0.233

0.635

7.114

19

0.000

7.596

0.013

6.455

15

0.000

6.950

0.016

6.260

15

0.000

3.715

0.069

2.247

19

0.037

0.720

0.407

5.523

19

0.000

0.563

0.463

2.054

18

0.055

7.487

0.013

6.267

15

0.000

2.318

0.145

3.818

18

0.001

P13. Qualified human resource for green shipping

2.158

0.158

6.117

19

0.000

P14. Outsourcing of green shipping issues

0.509

0.484

5.639

19

0.000

0.007

0.933

7.293

19

0.000

0.299

0.591

5.695

19

0.000

1.900

0.184

4.291

19

0.000

2.680

0.118

2.550

19

0.020

1.292

0.270

2.487

19

0.022

4.960

0.038

7.701

16

0.000

P1. CEO's recognition of the importance of green shipping P2. Monitoring international trends in green shipping P3. Benchmarking advanced and/or competing shipping firms P4. Investment ability and/or financial source for green shipping P5. Pre-investment and experiences for green shipping P6. Corporate level strategic approach to CO2 emissions reduction P7. Establishing long-term target for CO2 emissions reduction P8. Publishing regular report on CO2 reduction performance P9. Measurement of emission amount from possessing vessels P10. Management of operations information for possessing vessels P11. Establishing specialized division for green shipping P12. Collaboration between divisions for green shipping issues

P15. Establishing framework to technical measurement P16. Implementing framework to technical measurement P17. Establishing framework to operational measurement P18. Implementing framework to operational measurement P19. Establishing framework to market based measurement P20. Implementing framework to market based measurement

Note: Equality of variances is not assumed in the items of P6, P7, P11, P20. 43

Current Issues in Shipping, Ports and Logistics

Regarding green shipping preparedness items, the two shipping groups presented significant gaps across all the indices. In the case of large carriers, though some scores were ranked below point 3 (i.e. pre-investment and experiences for green shipping; publishing regular report for CO2 emissions reduction performance; and implementing corresponding strategy to market based measurement), overall preparedness/ability level was revealed to be rather good. In addition, the scores of preparedness level items were generally lower than those of awareness level indices. Meanwhile, the preparedness scores of small/medium-sized carriers were definitely poor. Only one item ‘management of operations information for possessing vessels’ was above point 3, which implies that it is urgently needed for Korean small/medium-sized shipping firms to seek strategic approaches to reinforce their green shipping related ability. Meanwhile, in order to find statistically significant differences between the two groups, a t-test method was adopted. The t-test results revealed that there were significant gaps in 15 items for the green shipping capability between the two groups. 4.4 Relationships between preferring MBM and degrees of awareness/preparedness The final interest in the current study is to explore the reasons why a certain shipping firm preferred an emission trading scheme while another company supported a carbon taxation system. The survey instrument asked respondents to mark their preferred market-based measure among two options, i.e. ‘carbon taxation’ and ‘emission trading scheme’. The result shows that the majority of respondents supported carbon taxation (81%), while only four carriers preferred ETS (19%). This preference structure seems to be maintained even if the shipping market recovers in the future. About the reason why the majority of Korean shipping firms did not support ETS, it is reported that carriers found major difficulties in the following areas: ‘inferiority in technology and investment source’; ‘lack of ETS infrastructure and knowhow’; ‘difficulties in fair allocation of CO2 emission right to carriers’; ‘time and cost burden for calculating CO2 emission amount from present ships’; ‘complexity in defining responsible bodies about emission right and obligation’; ‘uncertainty in ETS market’, etc.. On the contrary, carriers felt that carbon tax would be relatively simple and easy to implement and that considerable amount of the tax burden could be transferred to shippers. Table 8. Preference on market-based measures Under Current Market

44

Under Market Boom

Carbon taxation

17 (81%)

Carbon taxation

17 (81%)

ETS

4 (19%)

ETS

4 (19%)

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

Figure 9. Difficulties in implementing an emission trading scheme Difficulties in fair allocation of CO2 emission right to carriers, 10%

Inferiority in technology and investment source, 18%

Time and cost burden for calculating CO2 emission amount from present ships, 13%

Difficulties in cost transfer to shippers, 6% Lack of ETS infrastructure and know how , 20%

Short preparation period, 3% Conflict betw een CBDR and NM FT, 4% Complexity in defining responsible bodies about emission right and obligation, 15%

Uncertainty in ETS market, 10%

In order to find out any significant influential relationship between the degrees of awareness/ preparedness for green shipping regulation and preferred market based measures, binomial logit model was employed. In the model, MBM is the dependent variable with a value of ‘zero’ when a company prefers the carbon taxation or ‘one’ if a firm supports ETS, while the awareness and preparedness levels are explanatory variables. First, we examined the causal relations between the preferring MBM system and the degree of awareness on conceptual issues which were classified into three categories, i.e. ‘basic knowledge (XA1)’, ‘principle and practical method under UN (XA2)’, and ‘international shipping specific conventions and measures (XA3)’. Next, the connection between the preference on the MBM system and preparedness level was investigated employing three independent variables, which are ‘contextual aspect (X P1)’, ‘operational aspect (XP2)’, and ‘strategic aspect (XP3)’. Before implementing empirical tests, it is necessary to validate the measurement model. The term of validation is used to mean demonstration of the measures’ validity and reliability (Olsen, 2002). Validity can be understood as “whether what we tried to measure was actually measured” (McDaniel and Gates, 1999). Validity is composed of four sub-dimensions, namely content validity, unidimensionality, convergent validity and discriminant validity (Mentzer and Flint, 1997; Steenkamp and van Trijp, 1991). Content validity is a precondition towards establishing the correspondence between theoretical constructs and measurement items (Mentzer and Flint, 1997). However, there is no rigorous way to test content validity (Dunn et al., 1994) because it mainly depends on a subjective judgment of the researcher (Churchill, 1992; Garver and 45

Current Issues in Shipping, Ports and Logistics

Mentzer, 1999). Nevertheless, if the constructs are created from a comprehensive analysis of the relevant literature, content validity can be certified (Churchill, 1979, Ahire et al., 1996). In this study, the items of awareness and preparedness were carefully selected from abundance of literature sources and screened by industry experts and, therefore, it can be asserted that the content validity is ensured. The remaining three validities can be best tested by confirmatory factor analysis (CFA); however the present study includes only 21 companies, a number that is not sufficient for adopting the CFA method. Meanwhile, reliability means “the internal consistency of the items that are used to measure a latent construct” (Dunn et al., 1994), which can be tested by Cronbach’s alpha, composite reliability of construct and average variance extracted. For the same reason, in the present study only Cronbach’s alpha was used for reliability testing. Although the CFA technique was not adopted, the proposed regression model may provide valid and reliable results because the selection and categorization of variables were made by a comprehensive literature review process and experts’ screen, and because the scores of Cronbach’s alpha for six variables were extremely high as shown in Table 9. Table 9. Reliability test results Awareness

Variables Cronbach’s alpha

Preparedness

XA1

XA2

XA3

XP1

XP2

XP3

0.947

0.923

0.939

0.966

0.947

0.985

Table 10. Binomial Logit regression results Regression 1

B

S.E.

Wald

D.F.

Sig.

XA1

4.567

2.251

4.115

1

0.043

XA2

-4.478

2.280

3.858

1

0.050

XA3

0.924

1.841

0.252

1

0.616

Constant

-8.130

4.342

3.505

1

0.061

Regression 2

B

S.E.

Wald

D.F.

Sig.

XP1

4.677

3.139

2.221

1

0.14

XP2

-25.356

16.254

2.434

1

0.12

XP3

18.032

12.821

1.978

1

0.16

Constant

-4.737

3.002

2.489

1

0.11

Table 10 presents the results of the regression analysis. Regarding the connection between the preferring market-based measures and green shipping regulation awareness, two awareness variables are shown to have significant influences on the preference system. The regression results imply that Korean shipping firms may 46

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

support ETS rather than carbon taxation if they only have some basic concepts or information; but they may select a carbon taxation system if they obtain more detailed and specific knowledge on the principles and mechanism for GHG mitigation. In addition, it could be an interesting finding that the awareness about international shipping related institutes and regulation had no significant influence on the preference system. Next, the regression result reveals that there was no significant relationship between the level of preparedness for green shipping regulation and preference for a specific market-based measure. As a matter of fact, Maersk, who has made a lot of efforts to prepare for green shipping regulation and presented the most advanced capability in technical and operational approach for GHG emission reduction, officially supports a carbon taxation system. This may imply that a high degree of awareness and/or ability for green shipping does not necessarily mean a preference for an emission trading scheme.

5 | Conclusion and policy implications The main findings and some policy implications can be summarized as follows. First, the empirical analysis reveals that the degree of awareness of Korean shipping firms on overall green shipping regulation is low and, in addition, the level of recognition and understanding on the major agenda and debated issues discussed at the IMO is also quite low. In particular, the degree of awareness of small and medium-sized shipping companies is significantly lower compared to the large firms. On the competitiveness issue after introducing green shipping regulation, small and medium-sized companies are worried that the new regulation will impose additional costs and weaken their price competitiveness. This recognition may be based on the competitive structure in Northeast Asia where the small and medium-sized Korean firms are marginally competing with low cost Chinese companies and they do not want any external shock influencing the current market structure. Third, concerning the level of preparedness or ability to shift to green shipping, Korean shipping firms seem to be in very poor condition. In particular, many small/mediumsized companies even have not started their preparation for green shipping yet, while on the contrary, the large firms have cumulated necessary capabilities even if not sufficient yet. In addition, when we consider that the responding small/medium-sized companies possess their own regional reputations and sizable sales values and fleets, the other small carriers who have not responded to our survey may have a much lower interest in and capability for green shipping regulation. This result implies that the international institutes and national governments should adopt a very careful

47

Current Issues in Shipping, Ports and Logistics

approach to small and medium-sized shipping companies when they establish and implement green shipping policies and regulations. Meanwhile, the majority of Korean shipping firms support carbon taxation rather than ETS as a market-based measure for GHG emissions reduction from international shipping regardless of their size and the level of awareness and preparedness. One of the interesting findings of the empirical study is that shipping firms may support ETS rather than carbon taxation when they only have initial information on green policy but when they get more advanced knowledge on the principles and mechanism for GHG mitigation, their preference changes to carbon taxation system. In addition, it is another interesting finding that the awareness for international shipping related regulation may have no influence on the selection between carbon taxation and ETS. The current research has some limitations. Most of all, the sample size is not large enough to conduct more rigorous validity tests. Therefore, it would be recommendable to collect more cases from other companies even if they are small and do not have a keen interest in green shipping policy. In addition, it would be an interesting research to compare the positions on green shipping regulations among shipping firms from different countries. For instance, Giziakis and Christodoulou (2010) shows that Greek shipping companies have a high level of awareness of the efforts made towards the adoption of a regulatory framework for GHG emissions from ships and asserted that carbon taxation is a more environmentally effective tool than an ETS. The present study focused on the Korean case and did not give much attention to the positions of foreign countries. Therefore, a more international study may provide a broader and deeper understanding about the shipping industry which faces a new paradigm shift. References Abrell, J. (2010) Regulating CO2 emissions of transportation in Europe: A CGE-analysis using market-based instruments, Transportation Research Part D,15, 235-239 Ahire, S.L., Golhar, D.Y. and Waller, M.A. (1996) Development and validation of TQM implementation constructs, Decision Sciences, 27(1), 23-56 Bakker, S., Coninck, H. and Groenenberg, H. (2010) Progress on including CCS projects in the CDM: Insights on increased awareness, market potential and baseline methodologies, International Journal of Greenhouse Gas Control, 4, 321-326 Chatfield, C. (1985) The initial examination of data, The Journal of Royal Statistic Society, 48(3), 214-253

48

Lee and Chang – Shipping Companies and Greenhouse Gas Regulations

Churchill, G.A. (1992) Better measurement practices are critical to better understanding of sales management issues, Journal of Personal Selling and Sales Management, 7(2), 73-80 Corbett, J.J., Wang, H. and Winebrake, J.J. (2009) The effectiveness and costs of speed reductions on emissions from international shipping, Transportation Research Part D, 14, 593-598 Dunn, S.C., Seaker R.F. and Waller M.A. (1994) Latent variables in business logistics research: scale development and validation, Journal of Business Logistics, 15(2), 145172 Garver, M.S. and Mentzer, J.T. (1999), Logistics research methods: Employing structural equation modelling to test for construct validity, Journal of Business Logistics, 20(1), 33-57 Giziakis, C. and Christodoulou, A. (2010) Environmental awareness and practice concerning maritime air emissions: the case of the Greek shipping industry, Proceeding of 2010 annual conference of the international association of maritime economists, Lisbon, July 2010 Grubb, M., Michaelowa, A., Swift, B., Tietenberg, T., and Zhang, Z.X. (1998) Greenhouse Gas Emissions Trading, Draft, UNCTAD, Geneva Haar, L.N. and Haar, L. (2006) Policy-making under uncertainty: Commentary upon the European Union Emissions Trading Scheme, Energy Policy, 34, 2615-2629 International Maritime Organization (2009) Second IMO Greenhouse Gas Study 2009 Hofer, C., Dresner, M.E. and Windle, R.J. (2010) The environmental effects of airline carbon emissions taxation in the US, Transportation Research Part D, 15, 37-45 Knapp, S. and Franses, P.H. (2009) Does ratification matter and do major conventions improve safety and decrease pollution in shipping?, Marine Policy, 33, 826-846 Lin, B. and Lin, C.-Y. (2006) Compliance with international emission regulations: Reducing the air pollution from merchant vessels, Marine Policy, 30, 220-225 Mayor, K. and Tol, R.S.J. (2010) The impact of European climate change regulations on international tourist markets, Transportation Research Part D, 15, 26-36 McDaniel, C. and Gates, R. (1999) Contemporary Marketing Research (4th Edition), South-Western College Publishing Mentzer, J.T. and Flint, D.J. (1997) Validity in logistics research, Journal of Business Logistics, 18(1), 1-25

49

Current Issues in Shipping, Ports and Logistics

Notteboom, T. and Vernimmen, B. (2009) The effect of high fuel costs on liner service configuration in container shipping, Journal of Transport Geography, 17, 325-337 Olsen S.O. (2002) Comparative evaluation and the relationship between quality, satisfaction and repurchase loyalty, Journal of the Academy of Marketing Science, 30(3), 240-249 Scheelhaase, J.D. and Grimme, W.G. (2007) Emission trading for international aviation – an estimation of the economic impact on selected European airlines, Journal of Air Transport Management, 13, 253-263 Scheelhaase, J., Grimme, W. and Schaefer, M. (2010) The inclusion of aviation into the EU emission trading scheme – Impacts on competition between European and nonEuropean network airlines, Transportation Research Part D,15, 14-25 Steenkamp, J.B.E.M. and van Trijp H.C.M. (1991) The use of LISREL in validating marketing constructs, International Journal of Research in Marketing, 8(4), 283-299 Stopford, M. (2009) Maritime Economics (3rd Edition), London: Routledge Tietenberg, T. (1992) Relevant experience with tradeable entitlements, in Combating Global Warming: Study on a Global System of Tradeable Carbon Emission Entitlements, New York, United Nations, pp. 37-54 Victor, D.G. and House, J.C. (2006) BP’s emissions trading system, Energy Policy, 34, 2100-2112 Wang, C. and Corbett, J.J. (2007) The costs and benefits of reducing SO2 emissions from ships in the US West Coastal waters, Transportation Research Part D, 12, 577-588 Wang, H. and Firestone, J. (2010) The analysis of country-to-country CDM permit trading using the gravity model in international trade, Energy for Sustainable Development, 14, 6-13 Webster, M., Paltsev, S. and Reilly, J. (2010) The hedge value of international emissions trading under uncertainty, Energy Policy, 38, 1787-1796 Woerdman, E. (2000) Implementing the Kyoto protocol: why JI and CDM show more promise than international emissions trading, Energy Policy, 28, 29-38 Zhang, Z.X. (2001) The liability rules under international GHG emissions trading, Energy Policy, 29, 501-508 Zhang, Z.X. and Nentjes, A. (1999) International tradeable carbon permits as a strong form of joint implementation. In Eds. Skea, J. and Sorrell, S., Pollution for Sale: Emissions Trading and Joint Implementation. Cheltenham: Edward Elgar, pp.322-342

50

CHAPTER 3 The Determinants of Tanker Price in the Chinese Shipbuilding Industry

 Liping JIANG

Abstract In this paper we present, for the first time, the analysis of tanker price determinants in China using vessel prices signed by major Chinese shipyards in actual shipbuilding contracts. This allows us to investigate the relationship between price and determinants in the Chinese shipbuilding industry by including generic market factors as well as Chinese elements. The analysis, which employs Principal Component Regression approach, indicates that the shipbuilding cost has the most significant positive impact on tanker price; increases in five other factors, namely the crude oil price, the time charter rate, the shipbuilding capacity utilisation, the price cost margin, and the ship export credit rate, have influences in descending order. We assert that the most important role that the shipbuilding cost plays is mainly attributed to the large proportion of cost in tanker price and the crude oil is of product price inelasticity of demand. In addition, simulations are performed to investigate what would happen to the Chinese tanker price under major changes in determinants. This paper has implications for Chinese shipyards, customers in shipbuilding industry and industry policy makers.

1|

Introduction

Shipbuilding is an important part of the maritime sector and provides the supply in the marine transportation system. It is a market where major yards compete internationally for orders at their quoted prices, which are the most important competition parameter. The price of shipbuilding plays an important role for shipyards in winning orders and for ship owners in making investment decisions. The price of shipbuilding is very much related to the state of shipping market. When there is a high demand for world seaborne trade, freight rates will be driven up by the limited transport capacity. This stimulates ship owners’ desire to expand their current fleets and thus order new ships at shipyards. It usually takes two to three years for a 51

Current Issues in Shipping, Ports and Logistics

ship to be delivered after placing an order (Stopford, 2009). For this reason, whether ship owners are prepared to take shipbuilding contracts and the price they are willing to pay, depend on their expectations of the future shipping market. If a long period of a thriving market is expected, investors are more likely to place an order. This is because the availability of shipyard capacity may be scarce and shipbuilding prices may be even higher in the future if the market continues to do well. With market confidence growing and orders rising, shipbuilding prices will be pushed up, and shipyards would expand their capacity to meet the shortage. However, when the market goes down, ship owners will face the market with a more reluctance and conservative mood. As a result, shipyards lie in a weak position. On the one hand, it is usually quite slow and difficult for shipyards to reduce the capacity which was expanded in times of prosperity or was supported the government subsidies. On the other hand, the overcapacity will lead to lower shipbuilding prices and acute competition. In addition to above demand-side factors that drive shipbuilding prices up and down, there are also supply-side factors at country, yard and ship levels that will influence the price of a new ship. These factors mainly include the national industrial policy, the production capacity, the shipyard’s cost of production, the currency fluctuation, the ship design, the vessel quality and the payment term. In essence, there is no single driving force behind the determination of shipbuilding price and the above factors have a combined effect. As the world’s shipbuilding center shifts from a high-cost capacity in Western Europe to low-cost shipbuilders in East Asia, China has taken a giant leap in the shipbuilding industry and developed into a strong shipbuilding nation in recent years. This paper particularly focuses on tanker in the Chinese shipbuilding, which is the second core product of Chinese shipyards and holds the second largest market share after South Korea. There are several reasons for the great achievement in the tanker sector. First, the industry experiences a considerable expansion in parallel with China’s accelerated economic growth and a rising demand for natural resources, especially crude oil. China has become a net crude oil importer since 1993 and the country’s demand for energy has continued to grow. In order to keep the competitiveness, China adopted a ‘National oil country carriers’ policy to build up the domestic tanker fleet and to meet the country’s growing demand for crude oil. In accordance with the policy’s objective, China's large-scale oil tanker fleet aims to carry more than 50% of imported crude oil for domestic use by 2010. In the long term, this policy not only ensures the oil transport ability and safeguards energy security, but also guarantees the expansion of tanker shipbuilding in China (OECD 2008). Second, Chinese shipyards gradually move into the production of tankers based on the experience in building less-difficult ships, like dry bulk carriers. Third, the low labour cost in China provides domestic shipyards with the advantages of competitive tanker prices, which is the major concern of tanker buyers. Understanding tanker prices, therefore, is important to Chinese shipyards and industry policy makers, and also has implications for Chinese shipyards moving into more advanced shipbuilding segments. Especially in recent years, the industry is confronted 52

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

with new situations that may challenge China’s competitiveness. We have seen a continuous increase of raw material costs, labour costs, and steel prices. A high level of government and private investment has unfortunately led to excess shipbuilding capacity. The RMB appreciation against the American dollar has also caught great attention within the Chinese shipbuilding community. Some rising questions include the following: Which factors actually determine Chinese tanker prices? How would tanker prices in China react to the rising costs and fluctuating freight rate? How can Beijing promote the tanker sector by introducing a new shipbuilding policy? Academic investigation is required to build a more comprehensive view of the price of Chinese tankers. Extensive research has investigated the dynamic behavior of world shipbuilding prices. Traditional approaches for price modeling are mainly based on the supply and demand equilibrium models (see Koopmans, 1939; Hawdon, 1978; Jin, 1993; Haralambides, 2005, among others). More recent studies apply portfolio theory first proposed by Beenstock (1998a) and he develops this idea further in following papers (Beenstock, 1998b; 1992; 1993). In contrast, limited scholarly attention has been paid to the Chinese shipbuilding price. This is due partly to the short development period of the shipbuilding industry in China compared to traditional shipbuilding powers. Furthermore, data on Chinese shipbuilding prices is not as detailed as world prices, since most of the shipping consultants only elaborate on the world shipbuilding prices. There is no systematic research on Chinese shipbuilding prices across different vessel types. Additionally, the existing evidence does not appear to take into account the impact of low costs and government support on Chinese shipbuilding prices. To fill the gap in the current literature, we collect shipbuilding contracts from the top 200 Chinese shipyards which in all account for 98.63% of national capacity. Historical contracts with prices are extracted and used to make a first economic analysis of Chinese shipbuilding prices for different vessel types. The paper contributes to the literature in a number of ways. First, this article enriches the growing but still limited amount of research on the Chinese shipbuilding industry, especially on the Chinese shipbuilding prices. Instead of using the time series of world shipbuilding prices, this paper, to the best of our knowledge, is the first academic work based on vessel prices signed by major Chinese shipyards in actual shipbuilding contracts. Second, the price model in this paper can account for generic market factors as well as the Chinese shipbuilding characteristics, for instance, Chinese shipbuilding cost and governmental financial support. The overall effect of cost is established through the construction of a Chinese shipbuilding cost index, which provides a more comprehensive description of the shipbuilding cost compared to one factor proxy used in existing literatures. A competition indicator is also constructed to measure the extent of competition that Chinese shipyards face in the world tanker shipbuilding market. In doing so, this model enables us not only to reveal the most important determinants of the Chinese tanker price, but also to draw a dynamic picture of the price movement under global and country-level factor changes. In addition, we provide a Principal Component Regression (PCR) approach which is new to the maritime economics field, and has been shown to be an effective method solving the problem of multicollinearity. Finally, the findings of this paper have implications for Chinese 53

Current Issues in Shipping, Ports and Logistics

shipyards, customers in shipbuilding industry and Chinese policy makers, and may also shed some light on the emerging shipbuilding nations who enter into the tanker shipbuilding sector. The remaining sections are organized as follows: section 2 reviews the extant literature on shipbuilding prices. Section 3 explains the econometric model and price determinants, section 4 describes the data. The Principal Component Regression model is introduced in section 5. Section 6 analyzes the result and discusses scenarios based on the price model. Finally, section 7 provides the conclusion.

2 | Literature review The topic of the shipbuilding price has received considerable attention in the maritime field. Earlier studies can be classified into two groups. One influential group is centered on supply and demand theory and Koopmans (1939) is among the earliest researchers who employ the theory to model the shipbuilding market. Koopmans argues that the time lag between the demand for shipping capacity and the actual availability of this capacity triggers expectations of the future market. Hawdon (1978) uses the current and lagged freight rates to represent different market conditions during the time lag. He also takes labour cost, cost of steel, and ship size into consideration, finding that that the current freight rate has a significant coefficient, whereas the lagged one is insignificant. Jin (1993) enriches the literature by not only utilising previous theory but also integrating it with a cost-based approach. More specifically, he models the relationship of tanker market factors regarding endogenous and exogenous variables, such as the shipbuilding cost, the shipyard capacity, and technological changes. Major factors affecting the tanker new building market are identified, but his approach of using the average number of employees in the Japanese shipbuilding industry as a proxy for the shipbuilding capacity is controversial, not to mention the small number of observations. One of most recent studies on shipbuilding prices is from Haralambides (2005) which utilises the econometric method of Structural Equation Model (SEM ), including short- and long-term impacts of variables. The second group of studies presents a very different point of view. Beenstock (1985) argues that the supply and demand theory is not appropriate for analysing ship price, because a ship has considerable longevity. He adopts the asset pricing approach and Rational Expectation Hypothesis (REH) into the maritime field as the starting point of his research. In the following cooperation with Vergottis (1989a; 1989b; 1992; 1993), Beenstock contributes several papers focusing on modeling the shipping market by using the asset pricing method. He argues that new building and secondhand ships are perfect substitutes only differing in age and thus that prices of new ships adjust to the expected prices of secondhand ships overtime.It should be noted that Beenstock’s argument has been debated by the following study. Haralambides (2005) explains that secondhand prices are volatile whereas new prices are relatively sticky and that the two prices are not perfectly correlated. In Adland (2007) the newbuilding order is a forward contract for the delivery of an age zero vessel in the future and not the value

54

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

of a vessel per se. For this reason, we adopt the classical supply and demand theory for analysing the price determinants of Chinese tankers. There are several recent studies on the Chinese shipbuilding industry due to its dramatic growth. Most of the literature on Chinese shipbuilding has mainly focused on product technology (Bai et al., 2007), shipbuilding management (Lu et al., 2000), labour costs (Chou, 2001), industrial policy (Song, 1990) and restructuring (Smyth, 2004). Using the actual prices from Chinese shipyards, this paper makes the first attempt to analyze the price determinants of tankers in the Chinese shipbuilding industry based on a traditional supply and demand view.

3 | Econometric model and price determinants The newbuilding prices are determined by the interaction of supply and demand in the market. In this section, we assume that tanker prices and observed quantity represent an equilibrium by the interaction of supply and demand in the tanker market. 3.1 The econometric model The supply of yard capacity for tanker Qts is a function of price Pt , an exogenous supply-side shifter Wt and an error term  t :

Qts  g ( Pt ,Wt ,  t ) Wt includes three variables indicating changes in the supply curve, which are shipbuilding capacity utilisation ( CAPA ), the shipbuilding cost ( C ) and the ship export credit rate ( CRATE ). The demand for the yard capacity for tanker Qtd is a function of price Pt , an exogenous demand side shifters Z t and an error term vt :

Qtd  h( Pt , Zt , vt )

Z t includes three variables indicating movements in the demand curve, which are time the charter rate (TRATE), crude oil price (OIL) and price-cost margin (PCM). It is assumed that the supply and demand of yard capacity for new tankers operate simultaneously to determine ship prices. A reduced form of price in equilibrium Qts  Qtd will relate price to determinants that influence both supply and demand, and is an error term.

Pt  f (CAPAt , Ct , CRATEt , TRATEt , PCM t , OILt , et )

3.2 The price determinants and hypothesis Six tanker price determinants are analysed below and they are assumed to be positively related with the tanker price.

55

Current Issues in Shipping, Ports and Logistics

3.2.1 Shipbuilding capacity utilisation (CAPA) Shipbuilding capacity is largely constrained by physical facilities. However, under the same condition, the output may also be different when yards produce at different productivity levels and product mixes (Strandenes, 1990). The reason is that the shipbuilding capacity utilisation rather than that the shipbuilding capacity varies. A low level of utilisation means that yards run at less than full capacity, leading to weak pricing power among shipbuilders. Conversely a situation of shipyards operating at almost full capacity implies that shipowners may have to pay higher prices for the limited berths. The utilisation of the shipbuilding capacity is measured by deliveries relative to the total capacity (Strandenes 1990). In this paper, capacity utilisation of Chinese shipbuilding is calculated by the Chinese annual delivery relative to China’s total shipbuilding capacity. Hypothesis 1 Shipbuilding capacity utilization of Chinese shipyards is positively related to the Chinese tanker price.

3.2.2 Shipbuilding cost (C) In terms of the traditional production theory, the producer will produce a positive amount as long as the price is at least equal to the average variable cost. If the prices are lower than that, then the producer will not produce at all (Wijnolst et al., 1997). The main variable costs for building ships contain the labour cost, the steel cost, and cost of equipment (including main engine). These three components account for 90% of the total variable costs and it is within these components that the largest difference will be found (Wijnolst et al., 1997). Stopford (2009) shows a rough cost structure of a 30,000 dwt bulk carrier which by modern standards is a small and simple vessel. 73% of the costs of a ship is variable cost which can be broken down by four main items. Steel represents about 18% of the variable cost, the direct labour 23%, the major equipment and engine 49%, and others 10%. The importance of each cost component varies according to different ship types and shipbuilders. It is obvious that the material cost is higher for high-outfit ships than simple cargo ones, and the steel cost is larger for bigger vessels compared to smaller ones. Emerging shipbuilding countries having labour cost advantages apparently enjoy a smaller share of labour cost. The price of each cost component also changes over time. Take steel for instance, there was a slow price growth trend in the world market since 2006 followed by a sharp surge in 2008 due to increasing iron ore prices and tightening steel supply. Labour wages are rising in China and South Korea, but slightly declining in Japan (ECORYS, 2009). To have a comprehensive understanding, a simple form of the cost index for Chinese shipbuilding can be expressed as follows:

C  ERATE * (W1 *ULC  W2 * S )  W3 * E where ULC is the unit labour cost index, S is the steel cost index, E is the cost index of equipment, ERATE is the RMB exchange rate against US dollar, and Wi (Wi  1, i  1,2,3) is the weight of the respective costs. Most of labour and steel costs occur in the Chinese currency, whereas the shipbuilding equipment has a high import content counted in US dollar. Therefore, we use ERATE 56

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

to convert the domestic labour and steel prices into US dollars instead. Data availability prevented us from constructing costs for individual shipyards, and instead an integrated national level of the Chinese shipbuilding cost is used. Among the cost components, variations in the labour cost between shipbuilders will bring on a great difference in the shipbuilding costs. It is mainly due to the fact that other components are available as products in the world market and technology in the tanker market is fairly equal (Wijnolst et al., 1997). The main strength of the Chinese shipyards is the significantly low wage, which provides a price advantage compared to other shipbuilders. However, the wage advantage enjoyed by China is partly offset by the low productivity compared to industry leaders such as Japan and South Korea (Lu, 2005). For this reason, labour costs need to be adjusted for productivity and the unit labour cost index ( ULC ), which measures the labour cost per unit of output is as follows: Wage ULC  V / No. where Wage is the average annual industrial wage, V is the annual industry’s added value and No. is the annual number of industrial employees. Unlike the domestic supplied labour and steel, the majority of Chinese shipbuilding equipment is imported. A high proportion of imported equipment will diminish the advantage of other low costs. The cost index of equipment can be written as I E r*D where I is the annual amount of China’s imported shipbuilding equipment, r the loading rate of imported equipment, and D the annual delivery of the Chinese shipbuilding industry. Hypothesis 2 Shipbuilding cost of Chinese shipyards is positively related to the Chinese tanker price.

3.2.3 Ship export credit rate (CRATE) Export credit for ships is a major government support and provides Chinese shipyards with financing channels for newbuilding. It can take the form of interest rate subsidy where the government provides a preferential rate for credit. Clearly the availability of credit rate support will lower the Chinese price. Hypothesis 3 Ship export credit rate provided to the Chinese shipyards is positively related to the Chinese tanker price.

3.2.4 Time charter rate (TRATE) The way in which time charter rate impacts the tanker price is quite straight forward. Shipbuilding orders are determined by the shipowner’s expectation of future earnings from new vessels. A higher freight rate renders it more profitable for ship owners to operate the vessels, leading ship owners to be more willing to place orders. It is

57

Current Issues in Shipping, Ports and Logistics

customary to assume that the time charter rates contain more information on the future markets compared to the spot freight rate. Hypothesis 4 The time charter rate of oil tankers is positively related to the Chinese tanker price.

3.2.5 Price-cost margin (PCM) The price-cost margin refers to the shipyard’s ability to set its price above its marginal cost. It is an indicator of the degree of competition in the market. The measure of price-cost margin (PCM) is defined as (p c ) PCM  i i * si pi where pi and ci represent the average price index and the marginal cost index of the Chinese tanker respectively. si denotes the Chinese market share of tankers in terms of new orders. In practice, the marginal cost is represented by the average variable cost; therefore, the cost index C will be adopted. The fiercer competition is reflected by the lower price-cost margin due to lower prices or a smaller market share (Creusen, 2006). If there is a low level of demand in the tanker market, then strong competition with major shipbuilding nations may force Chinese yards to reduce prices to the marginal cost level. Hypothesis 5 The price-cost margin of Chinese shipyards is positively related to the Chinese tanker price.

3.2.6 Crude oil price (OIL) Since the late 1960s, crude oil has replaced coal to become the main source of energy. Western countries’ burgeoning demand for crude oil significantly boosted the growth of oil seaborne transportation and supply of and demand for oil tankers. The price fluctuation of crude oil has great impact on the tanker prices. When there is a weak growth in the global economy, for example due to a financial crisis, the crude oil price will decline due to reduced demand for energy. Facing economic recession and low consumption of oil, the demand for tankers will become less and inevitably followed by a drop in tanker prices. On the contrary, the crude oil price will jump when there is a world economic recovery or a long winter. More oil tankers are needed to fulfill the demand for oil transport which will make the tanker price soar. As the oil trade boomed, shipowners and large shippers in oil exporting countries have accumulated sufficient capital and would have more enthusiasm in tanker investments, and it further increases tanker prices. Hypothesis 6

58

The crude oil price is positively related to the Chinese tanker price.

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

4 | Data collection In this study, the Chinese tanker prices are analysed by referring to five tanker types: VLCC (over 200,000 dwt), Suezmax (120,000-199,000 dwt), Aframax (80,000-119,000 dwt), Panamax (60,000-79,000 dwt), and Handysize (30,000-59,000 dwt) types. Both crude oil and product tankers are included in our study. VLCC and Suezmax are used for crude oil transportation. Handysize is used for product oil only and Aframax and Panamax are used for both types of oil. The sample period is from January 1995 to December 2009. Tanker shipbuilding orders are collected from the Clarkson Shipping Intelligence Network (Clarkson SIN) for the top 200 Chinese shipyards and tanker prices at contract time are reported in million American dollar. Independent variables use the data in corresponding contract months. For Chinese shipyards, there are only few tanker contracts in the late 1990s, several between 2000 and 2005, and a large amount between 2006 and 2009. Additionally, large tankers such as VLCC type only started to be built from 2001 due to gradual technological development. For the sake of an effective analysis, we pool the data of all five types and distinguish the size effect by dummy variables. Other shipping and shipbuilding data are also collected from Clarkson SIN. The time charter rate is the monthly average rate (thousand dollar /day) for the corresponding vessel type. The shipbuilding capacity utilisation rate is measured by the annual delivery of China ( CGT) relative to yearly Chinese shipbuilding capacity. The shipbuilding capacity is calculated as the maximum annual output (CGT) in China since 1991 (according to the definition used by Clarkson). The Brent Crude Oil Price (dollar/bbl) is also collected from Clarkson SIN. The data for average annual industrial wages (yuan / man year), the annual industry added value (yuan / year) and the number of employees are especially for the Chinese manufacturing of transport equipment industry. These data are collected from yearbooks of the Chinese National Bureau of Statistics and Chinese Ministry of Labour and Social Security. The monthly price (yuan / ton) of medium and heavy steel plate in China is collected from the MYSTEEL database. The annual import amount of shipbuilding equipment (American dollar) is collected from China Customs. The monthly exchange rate of yuan against the American dollar and the ship export credit rate are collected from the People’s Bank of China. According to previous studies (Wijnolst et al., 1997), we assume that labour, steel and equipment take various weights of the average variable cost for different tanker types. In general, a smaller vessel has a larger weight on equipment and a smaller weight on steel. In this paper, the weights of labour, steel and equipment account for 5%, 53% and 41% for a VLCC, 5%, 50% and 45% for a Suezmax, 5%, 46% and 49% for an Aframax ship, 6%, 43% and 51% for a Panamax ship, and 7%, 39% and 54% for a Handysize vessel. These are rough data of the cost structure; as they are depend heavily on ship type, ship size, as well asmore detailed design particulars, related to on-board transshipment facilities, cargo facilities and ship speed (Hopman and Nienhuis, 2009). In all, the shares of labour, steel and equipment for the Chinese shipbuilding cost in a specific year are calculated by weighted average of all types built in that year. Four 59

Current Issues in Shipping, Ports and Logistics

dummy variables are used to control for five vessel types. D1 equals 1 if VLCC, D2 equals 1 if Suezmax, D3 equals 1 if Aframax, D4 equals 1 if Panamax, and others is Handysize. Prices and costs in RMB are deflated by the Chinese CPI (1995=100) and those in US dollar are deflated by the US CPI (1995=100). There are 204 observations for each variable, but there is a limitation of our sample. A total of 50% of the data concentrate on the years 2006-2008 during which the market is clearly in a boom phase. In contrast, 40% of the data are from the start of the decade and only 10% of the sample stands for the post-financial crisis period. This unbalanced data structure will limit the explanatory power of our model, which will be improved in a future study. The bivariate correlation analysis indicates there is some correlation among price factors. The test results of VIF (larger than 10) and tolerance values (close to 0) in the multiple linear regression of price factors also suggest that there is a multicollinearity problem. The traditional multiple linear regressions cannot be used directly and therefore the Principal Component Regression approach will be used as an alternative way to model the tanker price.

5 | Methodology 5.1. Principal Component Analysis When highly correlated predictors are used in a multiple linear regression, the model can face a serious multicollinearity problem. This high correlation or multicollinearity can be even serious when the goal is to understand how the various independent variables impact the dependent variable. When high multicollinearity is present, confidence intervals for coefficients tend to be very wide and t-statistics tend to be very small. This can become the cause of incorrect rejection of variables, and numerical inaccuracies in computing the estimates of model coefficients (Jennrich, 1995). One method of removing multicollinearity in regression is to use the Principal Component Regression (PCR) approach (Jolliffe, 2002). PCR actually is the combination of Principal Component Analysis (PCA) and multiple linear regression techniques. The central idea of PCA is to transform a number of possibly correlated variables into a smaller number of uncorrelated principal components. A small number of components can be extracted and contains most of the information of the original data (Jolliffe, 2002). The component Ci is given by Equation (1)

Ci  i1 * Z ( X1 )  i 2 * Z ( X 2 )    ik * Z ( X k )(i  1,k )

(1)

th

where Ci stands for the i principal component, Z ( X k ) stands for the k th standardised independent variable,  ik stands for the component score coefficient of the k th standardised independent variable on the i th principal component.

60

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

5.2. Principal Component Regression By using PCA, the extracted components become ideal to use as predictors in a regression equation, as they have optimised spatial patterns and removed the possible multicollinearity (Al-Alawi, 2007). Therefore, Principal Component Regression can be expressed as Equation (2) Z (Y )  1C1   2C2    i Ci (i  1, p  k ) (2) where Z (Y ) is the standardised dependent variable, Ci stands for the i th principal component, and  i is the i th standardised coefficient of the standardized principal component regression equation. The component equation (1) will be applied to equation (2), and then the standardised linear regression equation will be yielded after sorting out all of the X i variables as Equation (3) Z (Y )  b1 * Z ( X 1 )  b2 * Z ( X 2 )    bk * Z ( X k ) (3) where, is the partial regression coefficient of the principal component regression equation, and then has to be changed to an unstandardised coefficient (Liu et al., 2003).

6 | Results 6.1. Principal Component Analysis SPSS 16.0 is used to run the Principal Component Analysis first and then the Principal Component Regression. In our case, the KMO test result is 0.388, which normally should be larger than 0.6. It indicates that there are some limits by using PCA to fully represent our data, which may due to the unbalanced structure of our dataset. The PCA involves several steps as follows. Table 1. Summary of the Principal Component Analysis Total Variance Explained Rotated Component % of % of ComponentTotal Variance Cumulative Variables C 1 C2 1 2.738 45.636 45.636 CAPA -.518 -.400 2

1.151

19.186

3

0.865

4

.735

5 6

64.822

C

.955

14.416

TRATE

.540

12.254

CRATE

.093

.469

7.815

PCM

.319

.042

.693

OIL

.875

.021

Component Score Coefficient Variables C1 C2 CAPA -.166 -.186 C

.462

-.191

.447

TRATE

.167

.216

-.799

CRATE

.219

-.616

.733

PCM

-.003

.477

.063

OIL

.414

-.143

61

Current Issues in Shipping, Ports and Logistics

First, six components are extracted from independent variables. Then, the varimax rotation technique is used to maximise the loading of a variable on one component and to produce a ranked series of factors (Al-Alawi et al., 2005). Based on the criteria that initial eigenvalues of principal components should be greater or equal to 1, we select the first two components as principal components, which together account for 64.8% of the total variance. Table 1 reports the result of the component extraction, the varimax rotation and the component score coefficient. Next, the component score coefficients in Table 1 are used to obtain the expressions of two principal components in Equation (4) and (5) C1  0.166 * Z (CAPA)  0.462 * Z (C )  0.167 * Z (TRATE )  0.219 * Z (CRATE)  0.003 * Z ( PCM )  0.414 * Z (OIL )

C2  0.186 * Z (CAPA)  0.191 * Z (C )  0.216 * Z (TRATE )  0.616 * Z (CRATE)

(4)

(5)

 0.477 * Z ( PCM )  0.143 * Z (OIL ) where C1 and C 2 stand for the two principal components, Z ( X k ) stands for the standardised independent variables.

6.2. Principal Component Regression The two principal components are subsequently used as independent variables in a multiple regression model. Furthermore, four dummies are included to examine the impact of vessel type. Because all dummies have to be jointly included or excluded in the model, the ‘enter’ regression method is chosen for the dummy group and component group respectively. The PCR results are summarised in Table 2. Table 2. Result of Principal Component Regression Coefficients

Std. Error

t

Sig.

(Constant)

-0.820

.041

-19.871

.000

C1 C2 D1 D2 D3 D4 R Square Adjusted R Square

.306 .098 1.835 0.792 0.579 0.226 0.944 0.942

.020 .020 .061 .082 .053 .056

15.177 4.779 30.008 9.712 10.964 4.028

.000 .006 .000 .000 .000 .000

Tolerance .702 .683 .344 .771 .582 .619

VIF 1.425 1.464 2.910 1.298 1.718 1.616

F (sig.) 552.350(.000)

a: Dependent Variable : Z score(p)

The F statistics shows that variables have significant impacts on the tanker price. The model can explain 94.2% of total variance and it indicates that the regression line fits the observation well. The coefficients of variables are all statistically significant. We 62

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

apply the principal component equations (4) and (5) into the PCR model (2). The final model with six original independent variables can be present as Equation (6): Z ( Pi )  0.82  0.069 * Z (CAPAi )  0.123 * Z (Ci )  0.007 * Z (CRATEi )  0.072 * Z (TRATE i )  0.046 * Z ( PCM i )  0.113 * Z (OILi )  1.835 Di1  0.792 Di 2  0.579 Di 3  0.226 Di 4

(6) And the unstandardised coefficients model can be specified as Equation (7): Pi  18.162  17.177 * (CAPAi )  15.956 * (Ci )  1.318 * (CRATEi )  0.179 * (TRATEi )  23.772 * ( PCM i )  0.161 * (OILi )  53.342 Di1  23.012 Di 2  16.827 Di 3  6.573Di 4

(7)

The hypotheses are confirmed by the significant positive coefficients except that of CAPA. According to the standardised coefficient model, shipbuilding cost plays the most important role in determining the tanker prices. For example, a 10% increase in the Chinese shipbuilding cost will make the tanker price rise by 1.23%. It further proofs that shipbuilding cost is the decisive factor in the shipbuilding price and therefore world shipbuilding centers have always been relocated to lower cost countries. Other price determinants，namely the crude oil price, the time charter rate, the shipbuilding capacity utilisation, the price-cost margin, and credit rate are in decreasing importance. The crude oil price has the second most important positive effect on the tanker price. It indicates that higher crude oil price will raise the tanker price further, which proves that demand for crude oil is inelastic. In addition, the tanker price heavily depends on shipowners’ expectations on the state of freight market. A higher time charter rate for tankers, leads to a higher the return on ship investment. As a result, ship owners will be more willing to invest in tanker with higher prices. The shipbuilding capacity utilisation is found to have a moderately negative effect on the tanker price. It is worth noticing that the capacity utilisation of Chinese shipyards has remained at a high level since the beginning of the modern shipbuilding industry and the tanker price also showed an upward trend. From the global financial crisis at the end of 2008, capacity utilisation started to drop and fell to 54% in 2009. However, the VLCC sector particularly faced a tide in 2008 and earlier 2009. Thus there is no remarkable decrease in the average tanker price but an increase instead. A large share of data concentrates on the years 2008-2009 which obviously capture the negative correlation between price and capacity utilisation. The recent prosperity in the VLCC sector is the result of following aspects. First, South Korea, which is the largest tanker builder, has decided to replace all single-hull tankers before 2011 since South Korea’s worst oil spill occurred in December 2007. Second, given the requirement for single hull tankers and high charter rates achieved for bulk carriers, many ship owners also converted their VLCC into Very Large Ore Carriers (VLOC) which have a shorter delivery time and are cheaper to build. Third, millions of barrels of crude oil were stored on VLCCs at sea at that time due to record high prices for crude oil. Tanker owners

63

Current Issues in Shipping, Ports and Logistics

predicted that there may be a shortage of transportation capacity for crude oil due to the above reasons and therefore invested in VLCC building. The price cost margin has a significant positive but a smaller impact on tanker price. At present, South Korean shipyards hold an absolutely dominant position in tanker shipbuilding, especially for the ultra large and large tanker types. This can be reflected by the low price-cost margin in the Chinese tanker shipbuilding market. The lower the price-cost margin, the fiercer the competition Chinese shipyards face from South Korean builders, and the further the tanker price is suppressed. However, competition is not likely to count for as much as shipbuilding cost and market demand in determining the tanker price. The fluctuation of the ship export credit rate has a significant but the least pronounced effect. It confirms that the Chinese shipbuilding industry has relied less on government subsidies after transforming from a central planning system to a free market launched by the State Commission of Science and Technology for the National Defence Industry (COSTIND ) in 1999. Major shipyards are prompted to be more market-oriented and to take part in world competition by improving technology and quality. It is necessary to state that the preferential credit rate offered to foreign shipowners after the recent financial crisis makes a significant contribution to the Chinese shipbuilding industry. But the aim mainly focuses on attracting foreign orders and stabilizing shipbuilding demand instead of adjusting prices. At the same time, all coefficients of dummy variables are significant and prove the systematically different prices of five types of tanker. The bigger ship has a higher price level and the intercept for VLCC, Suezmax, Aframax, Panamax and Handysize in the unstandardised coefficient models are 71.504, 41.174, 34.989, 24.735 and 18.162 respectively. In general, the shipbuilding cost and the crude oil price have the greatest effect of all variables on the tanker price. The shipbuilding cost exceeds oil price as the most important factor due to two reasons. First, shipbuilding involves a large amount of investment and cost is the major component of the ship price and has a decisive impact on the price level. Another reason is that crude oil, as the engine of society and economic development, has a relative steady and inelastic demand. Therefore, the change of oil price has a relative smaller impact on the tanker price compared to the shipbuilding cost. What else can be implied from the model is that the degree of international competition in the tanker shipbuilding market has less influence on tanker prices than capacity utilisation, however, it affects more than national industrial policy. 6.3 Simulations and discussions Previously we tested the tanker price model. In this section, the model is used to carry out a few simulations to understand the likely price behaviour under various exogenous variable changes. Here, we choose the top three determinants as the impact variables and fix others in order to capture the most significant changes. The 64

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

reference year chosen is 2009 and eight scenarios are tested against the baseline scenario (see Table 3). The Chinese shipbuilding industry is facing heavy pressure from rising costs. For this reason the cost index will change from 50% to 100% levels compared to 2009. Because of American financial crisis, the time charter rate in the 2009 already touched a historically low level. We now assume that it will change at three different levels, namely 0%, 50% and 100%, compared to 2009. The crude oil price is also assumed to have an upward trend at 30% and 60%. The numbers given below are percentage changes over 2009 for the shipbuilding cost, the crude oil price, the time charter rate and tanker price. Table 3. Simulation results of price changes (%) Variables

Baseline

S1

S2

S3

S4

C

0

50%

50%

50%

50%

OIL

0

0%

30%

30%

TRATE

0

0%

0%

50%

VLCC

0

SUEZMAX

S5

S6

S7

S8

100% 100%

100%

100%

30%

30%

60%

60%

60%

100%

0%

0%

50%

100%

0.117 0.139 0.142 0.145 0.256 0.278

0.281

0.284

0

0.169 0.201 0.204 0.207 0.370 0.401

0.404

0.407

AFRAMAX

0

0.186 0.221 0.223 0.226 0.407 0.442

0.444

0.447

PANAMAX

0

0.223 0.264 0.267 0.270 0.487 0.529

0.532

0.534

HANDYSIZE

0

0.256 0.303 0.305 0.308 0.559 0.606

0.608

0.611

There clearly is a large increase in prices for all types under S1 to S8 compared to the baseline scenario. In S1 to S4, shipbuilding cost levels are the same but with a higher crude oil price and time charter rate, leading to the higher the tanker prices. S5 to S8 share the same characteristics. But the increase of the time charter rate does not push tanker prices up at the same extent compared to crude oil. We also notice that there is a big price gap between S2 and S5, S3 and S7, and S4 and S8 for all types, under which higher costs can lead to much higher prices. In all scenarios, smaller ships will have larger price changes under the same conditions.

7 | Conclusions In this study, we propose a price model for tankers in the Chinese shipbuilding industry, thus filling a gap in the literature on global shipbuilding prices. We utilise the actual vessel prices signed by major Chinese shipyards and propose Principal Component Regression approach for price modelling. In particular, the theoretical relationship between price and determinants is discussed by including generic market factors as well as Chinese elements. Several conclusions can be drawn. First, the shipbuilding cost is the most important determinant. Meanwhile, increases in the crude oil price, the time charter rate, the shipbuilding capacity utilisation, price-cost 65

Current Issues in Shipping, Ports and Logistics

margins, and the credit rate have effects decreasing in order of importance. The dynamic price simulations indicate that there is a likely big increase in the prices as a result of the growth in the shipbuilding cost, crude oil price, and time charter rate. Particularly, smaller vessels tend to have larger price increases under the same conditions. Second, the PCR approach is shown to be a good way of overcoming the problem of multicollinearity in the field of maritime economics. Dummies in the model successfully served to distinguish the systematic differences in vessel types. Third, investors and policy makers in the shipbuilding industry can benefit from applying the model when making decisions. It also has implications for emerging shipbuilding nations who are entering the tanker sector and join the arena of major shipbuilding nations. This paper mainly focused on the determinants of the Chinese tanker price without referring to how other shipbuilding competitors jointly set the world price and thus influence the Chinese price. We leave for future research the question of how competitors influence Chinese shipbuilding prices and competitiveness.

Acknowledgments This research is partly funded by the Chinese Scholarship Council and TORM Foundation.

References Al-Alawi, S.M., Abdul-Wahab, S.A. and Bakheit, C.S. (2008) Combining principal component regression and artificial neural networks for more accurate predictions of ground-level ozone, Enviromental Modelling & Software, 23, 396-403 Aland, R. and Koekebakker, S. (2007) Ship valuation using cross-sectional sales data: a multivariate non-parametric approach, Maritime Economics and Logistics, 9(2), 105-118 Bai, X.P., Niu, W. and Liu, C.M. (2007) A comparison of Chinese, Japanese and Korean shipyard production technology, Journal of Marine Science and Application, 6(2), 25-29 Beenstock, M. (1985) A theory of ship prices, Maritime Policy & Management, 12(3), 215–225 Beenstock, M. and Vergottis, A. (1989a) An econometric model of the world shipping market for dry cargo, freight and shipping, Applied Economics, 21, 339–356 Beenstock, M. and Vergottis, A. (1989b) An econometric model of the world tanker market, Journal of Transport Economics and Policy, 23, 263–280 Beenstock, M. and Vergottis, A. (1992) The interdependence between the dry cargo and tanker markets, Logistics and Transportation Review, 29(1), 3–38 Beenstock, M. and Vergottis, A. (eds) (1993) Econometric modeling of world shipping. London, Chapman & Hall.Brooks

66

Jiang - Determinants of Tanker Price in the Chinese Shipbuilding Industry

Chou, C.C. and Chang, P.L. (2001) Modeling and Analysis of Labour Cost Estimation for new building: The case of China New building Corporation, Journal of Ship Production, 17(2), 92-96 Clarkson SIN. www. clarksons.net Creusen, H. (2006) Measuring and Analyzing the competition in Netherlands, De Economist, 154(3), 429-441 ECORYS (2009), Study on Competitiveness of the European Shipbuilding Industry, http://ec.europa.eu/enterprise/sectors/maritime/files/fn97616_ecorys_final_report_o n_shipbuilding_competitiveness_en.pdf, accessed October 8th, 2009 Haralambides, H.E., Tsolakis, S.D. and Cridland, C. (2005) Econometric Modeling of Newbuilding and Secondhand Ship Prices, Research in Transportation Economics, 12, 65-105 Hawdon, D. (1978) Tanker freight rates in the short and long run, Applied Economics 10, 203–217 Hopman, J.J. and Nienhuis, U. (2009) The future of ships and shipbuilding - a look into the crystal ball. In Meersman, H., Van De Voorde, E., Vanelslander, T. (eds), Future challenges for the port and shipping sector (1), pp. 27-52 Meersman, H., Van De Voorde, E., Vanelslander, T. (eds) (2009) Future Challenges for the Port and Shipping Sector, The Grammenos Library, Informa, London Jennrich, R.I. (eds) (1995) An introduction to computational statistics-Regression Analysis, Prentice-Hall, Englewood Cliffs Jin, D. (1993) Supply and demand of new oil tankers, Maritime Policy and Management, 20, 215-227 Jolliffe, I.T. (eds) (2002) Principal component analysis, Springer NewYork Inc Koopmans, T.C. (eds) (1939) Tanker freight rates and tankship building, an analysis of cyclical fluctuations, Netherlands Economic Institute, Report No. 27, Haarlem, Holland Liu, R.X., Kuang, J., Gong, Q. and Hou, X.L. (2003) Principal component regression analysis with SPSS, Computer Methods and Programs in Biomedicine, 71, 141-147 Lu, B.Z. and Tang, A.S.T. (2000) China shipbuilding management challenges in the 1980s, Maritime Policy and Management, 27(1), 71-78 Lu, Z.D. (2005) Can China become No. 1 shipbuilding nation in 2015, Master Thesis, Erasmus University Rotterdam Organisation for Economic Co-operation and Development (OECD, 2008) The shipbuilding industry in China, Working Party on Shipbuilding 67

Current Issues in Shipping, Ports and Logistics

Smyth, R., Deng, X. and Wang, J.L. (2004) Restructuring state-owned big business in former planned economies: The case of China’s shipbuilding industry, New Zealand Journal of Asian Studies, 6(1), 100-129 Song, Y.B. (1990) Shipping and shipbuilding policies in PR China, Marine Policy, 14(1), 53-70 Stopford, M. (2009) Maritime Economics, 3rd Edition, Taylor & Francis Group Strandenes, S.P. (1989) Capacity utilization in shipbuilding and shipping, International Shipping Seminar, Bergen, August 1989 Wijnolst, N. and Wergeland, T. (eds) (1997) Shipping, Delft University Press, Delft

68

CHAPTER 4 Bunker Costs in Container Liner Shipping: Are Slow Steaming Practices Reflected in Maritime Fuel Surcharges?

 Pierre CARIOU and Theo NOTTEBOOM Abstract Slow steaming has been implemented by the main liner shipping companies since 2008. The reduction in vessel speed affects fuel consumption and should be reflected within the fuel surcharges paid by shippers. This article assesses if this was the case for the main outbound European container trades from the port of Antwerp. Through an extensive analysis of liner service characteristics, fuel costs and fuel surcharges this paper provides an answer to three research questions (a) How significant are slow steaming practices in container liner shipping?; (b) What is the impact of slow steaming on fuel consumption and liner service characteristics?; and (c) To what extent has slow steaming changed the relation between fuel costs and fuel surcharges imposed on shippers by shipping lines?

1 | Introduction Slow steaming, or the reduction in the sailing speed of maritime vessels, has become an increasingly common practice in container liner shipping as the amount and unit size of available vessel capacity rises and the price of fuel increases (Alphaliner, 2010a). Slow steaming can help to absorb vessel overcapacity as a slower commercial speed will require more vessels to maintain the same service frequency per liner service. Slow steaming has also proven to be an effective way to save on fuel costs and to restore liner shipping company profitability. Slow steaming is also claimed to reduce environmental emissions by ships at sea (Kollamthodi et al., 2008; Buhaug et al., 2009; Corbett et al., 2009; Cariou, 2011; Faber et al., 2010). However, slow steaming practices added a new source of contention between shippers and ship-owners regarding fuel surcharges, known as Bunker Adjustment Factor or BAF implemented by shipping lines since 1974 (Menachof and Dicer, 2001:143). Shippers’ organizations such as the European Shippers’ Council have objected for years that the way BAFs are determined is opaque, without uniformity, and involves a significant element of revenue-making (ESC, 2003: 20, ESC, 2006:20). The anticompetitive effect of BAF was already subject to studies shedding light on a tendency of BAF of amplifying bunker prices rises (Cariou and Wolff, 2006; Meyrick and Associates, 2008) impacting negatively consumers prices (Karamychev and van Reeven, 2009), and on the fact that 69

Current Issues in Shipping, Ports and Logistics

a combination of decreasing freight rates and fuel costs provide incentive to shipping lines to stall the downward correction of the BAFs (Cariou and Notteboom, 2011). Slow steaming added an additional dimension to the question whether fuel surcharges are a revenue-making instrument to shipping lines or only about cost recovery of incurred fuel costs. This article adds to former studies in incorporating the impact from slow steaming. It investigates if slow steaming practices on major trade lanes are reflected within the BAFs charged to shippers by shipping lines. The paper addresses the follow research questions:  How significant are slow steaming practices in container liner shipping?  What is the impact of slow steaming on fuel consumption and liner service characteristics?  To what extent has slow steaming changed the relation between fuel costs and fuel surcharges imposed on shippers by shipping lines? To answer these research questions, this paper presents first how fuel surcharges are set up by shipping lines. Section 3 presents a methodology for estimating the impact of slow steaming on the average fuel consumption of containerships, and consequently, on BAF. Section 4 applies the methodology to 618 vessels deployed in 104 services sailing from/to Europe in January 2010, and provides a comparison with 2008, the pre slow steaming era. Section 5 presents the results of a BAF vs. fuel costs analysis for 90 O/D relations using Antwerp as port of departure. Section 6 provides the conclusions and explores avenues for further research.

2 | Fuel surcharge practices since 2008 The application of fuel surcharges in liner shipping dates back to the liner conference era (Notteboom and Cariou, 2011). In principle, carriers cover basic bunker costs, while fuel surcharges only apply to changes above a certain level. Fuel surcharge practices have considerably evolved since the withdrawal in October 2008 of the European liner conferences block exemption (Regulation 4056/86). Their dismantling meant that container shipping lines calling at European ports were banned from collectively setting freight rates and other additional surcharges such as bunker and currency surcharges, and from publishing common tariffs. In doing so, this reduced the commonality amongst pricing structures and surcharges that existed before, with freight rates and surcharges being negotiated directly between shippers and shipowners and with container lines using sometimes diverging calculation methods for determining fuel surcharges. Despite these changes, guidelines still exist and are geared mostly for small shippers. For instance, Maersk Line published in early 2008 a formula for determining its BAFs, with the aim of creating more transparency (Maersk Line BAF calculator, 2010). The formula known as ‘Maersk Line BAF Calculator’ builds on two components: Bunker price changes in t x Trade specific constant so that:

70

Cariou and Notteboom – Bunker Costs, Slow Steaming and Fuel Surcharges

BAFt=(Bunker Pricet-Base)x(ConsumptionTEU/day)x(Transit Timeday)x(Imbalance Factort) Bunker price change is extracted from the difference in t between a representative basket of prevailing bunker prices in a specific trade (BPt) and a predefined Bunker base element for a trade (Base) or “normal” bunker cost already included in the freight rate. The trade specific component is function of the consumption in metric tons/TEU/day of a representative vessel, a transit time in days and an imbalance factor. To provide an example from Maersk Line BAF calculator, for a 20 foot dry container exported from Belgium to China (outbound) in November 2010, the reported BPt was 435 USD/ton, the Bunker base element equaled 65 USD/ton, the vessel consumption is 0.0256 mt/TEU/day, the transit time 35.6 days and the trade imbalance equal to 0.5. It led to a BAF of 2x[435-65]x[0.0256x36.5x0.5]= 345 USD to be paid for each FEU, a value close to the one retrieved from CMA-CGM on-line BAF calculator (370 USD/FEU). Delmas/OTAL (part of CMA-CGM group) indeed also developed since September 2008 its own BAF formula, following the dismantling of the Europe West Africa Trade Agreement (EWATA). Similarly, an average reference fuel oil price, fuel oil consumption per full TEU carried and an average fuel oil price in t-1 are used for calculation of the BAF applicable in month t+1. Another example relates to OOCL. Its fuel surcharge policy is based on specifics on trade lane, service loop, vessel size and round voyage capacity on a monthly basis. OOCL uses a neutral third party provider of bunker price information (Platts) for the major locations around the world and selected a number of representative vessels for calculating fuel consumption, a more manageable way than taking into account the actual consumption of all their operating vessels. In general terms, the formula is similar to Maersk Line or Delmas/OTAL. As for many other shipping lines, OOCL made a policy decision not to disclose the values for each component in the formula. If the bunker price in a month moves beyond the agreed band of USD 25 (either up or down), then it will trigger a recalculation of the total BAF payable in the following month (see Notteboom and Cariou, 2011). The new calculation method led to a BAF that is lower compared to the previous liner shipping conference environment. The new fuel surcharge calculators have not wiped out potential sources of contention between shippers and ship-owners. Shippers express concerns about the confidentiality of some inputs used in calculating the BAF. Examples include the projected cargo load for OOCL or imbalance factor for Maersk Line. The representative fuel consumption of vessels deployed on a specific trade is another major source of contention in the fuel surcharge calculations. Shippers face difficulties in verifying vessel consumption figures, which leads to some doubts in shippers’ circles about whether the fuel savings caused by slow steaming practices are fully reflected in fuel surcharges. Using the former example of a container shipped from Belgium to China, if the decision to slow steam a service reduces by 10% the vessel fuel consumption and is not factored in, this generates ceteris paribus around 34.5 additional USD per FEU (10% of $345) which for a typical service with 10 x 4,000 TEU vessels sums up USD 71

Current Issues in Shipping, Ports and Logistics

690,000 additional revenues per trip. However, these revenues are not without a cost (Kollamthodi et al., 2008; Corbett et al., 2009; Faber et al., 2010) as: (a) vessels are spending more time at sea reducing the annual payload; (b) in case of significant speed reduction, additional vessels are required to keep a weekly frequency in the ports of call (Notteboom et Vernimmen, 2008; Psaraftis et al., 2010) and (c) for shippers, intransit inventory costs increase with transit time (Efsen et al., 2010; Bergh, 2010; Cariou, 2011). Next section presents a methodology to assess the first two effects.

3 | The overall impact of slow steaming Using an extended version of Maersk Line BAF calculator, the BAF to be charged per FEU for a service s with n vessels can be estimated as follows: n ( FC  (1   s ) FCk , port ) BAFFEU  2.( BPt  Base) s k ,sea .Transit times .IFs (1) TEU s  k 1 With FC k ,sea  SFC k ELk kWhk (2)



And: FCk,sea FCk,port Rots IFs TEUs SFCk ELk kWhk



the fuel consumption at sea per day for a vessel k the fuel consumption in port per day the transit time in days with αs.Rots days at sea and (1-αs).Rots in ports the imbalance factor for service s the total capacity in teu deployed in a service s the Specific Fuel oil Consumption in g/kWh the Engine Load in % the engine power in kWh

Slow steaming impacts both on the fuel consumption of each individual vessel k (FCk,sea) and on the characteristics of a service s. Focusing on the first component, for containerships carrying more than 1,000 TEU which are using two stroke marine diesel engines, slow steaming reduces the main engine fuel consumption at sea (FCk,sea), with a limited effect for the auxiliary engine and consumption in port. Under normal condition, vessels were built for sailing at a speed close to design speed or an Engine Load between 70-90% of maximum continuous rate (MCR), a level at which the SFC is optimal - around 170-195 g/kW (MAN B&W Diesel A/S, 2008; Buhaug et al., 2009; Psaraftis et al., 2010; Faber et al., 2010). This value varies with the engine type and with weather conditions on route. The impact of slow steaming on fuel consumption depends on the magnitude of the speed reduction (MAN B&W Diesel A/S, 2008; Buhaug et al., 2009; Psaraftis et al., 2010; Faber et al., 2010). As long as the speed is reduced in small amounts up to a 10-15% reduction, the SFC remains fairly constant. As a rule of thumb, engine power is related to ship speed by a third power. When speed is reduced by more than 10% the SFC increases by up to 10%. This latter figure varies on the basis of engine characteristics, vessel type and engine age as engine retrofitting can limit the increase in SFC.1 1

According to one-year data gathered from a private operator for a 4,300 TEU containership with a modern engine, the SFC would only increase from 195 to 198 g/kWh and the fuel consumption at sea would fall by around 60%. 72

Cariou and Notteboom – Bunker Costs, Slow Steaming and Fuel Surcharges

The second impact from slow steaming is on the transit time and on the number of vessels to be deployed within a service (Notteboom and Vernimmen, 2008; Psaraftis et al., 2010; Cariou, 2011). The number n of vessels to add remains difficult to estimate as this primarily depends on what the shippers can bear in terms of increase in inventory costs (Bergh, 2010), and on the initial service characteristics in terms of the round voyage distance, the number and order of port calls, the frequency, the time buffers in the liner service, the fleet mix and the imbalance factor. As an alternative, some ports of call can also be dropped. Hence, a decision to opt for slow steaming requires a careful analysis of the trade-off between a positive impact resulting from the reduction in fuel consumption at sea and two negative effects: the need for additional vessels in case of significant reductions in speed; the increase in the time spent at sea, and therefore, in transit time. The final impact on BAF is then to be multiplied by differences in bunker prices, transit time and by the imbalance factor for a service or trade.

4 | The impact of slow steaming on fuel consumption at sea Two sets of information are required to assess the impact of slow steaming on fuel consumption for a specific trade: (a) the number of services for which this strategy was implemented and how these services were affected, and (b) the vessel characteristics, in particular the reduction of the average fuel consumption as a consequence of slow steaming. To assess the extent of slow steaming per trade and the impact on fuel consumption, information was first gathered from three sources: from Alphaliner database (Alphaliner, 2010b) in January 2010 that was merged with data from the Lloyd’s Register Fairplay database (2009); and data on 90 outbound port-to-port relations with Antwerp as the port of loading in July 2008 and November 2010. The names of the shipping lines included in the dataset are not disclosed for confidentiality reasons. The initial data contains in Alphaliner database is for 174 liner shipping services and a total of 825 vessels deployed. The status of a service with respect to slow steaming was retrieved from comments in the database on liner service history. Services were then selected for 6 representative European container trades reducing the sample to 104 services with 618 vessels (table 1). For each trade, the mean age, design speed and engine power in kWh was then retrieved from LRF (2009). Europe/Far East is the first trade with 39 services - 37% of the 104 services - and with 273 vessels deployed - 44% of the 618 vessels. An interesting feature is disparities on the extent of slow steaming from one trade to another. For instance, 79.5% of Far Europe/Far East services are reported under slow steaming, contrary to services to Africa (6.3% of services). These results are roughly proportional, to the exception of services to Oceania, to vessel size and sailing distances. Regarding fleet structure, Far East is the trade on which the mean size of vessels is the largest, and North America, Oceania and Africa are trades for which vessels are the oldest. This later result is likely to influence the power needed, as age can be seen as a proxy of technology. Another important element to consider is differences in the structure of trade, and in 73

Current Issues in Shipping, Ports and Logistics

particular, the number of reefers. Information gathered from private sources stresses for instance that the consumption of the auxiliary engine for a typical 4000 TEU vessel increases from 4 to 20 tons due to the number of reefers carried. Table 1. Main characteristics of 174 European liner services in January 2010 Number Mean Age Design Engine Services % SS Vessels % SS TEU Speed kWh Africa 16 6.3 68 5.9 2662 9 21 23,570 Far East 39 79.5 273 79.5 7970 5 25 58,778 India/Pakistan 11 72.7 63 74.6 4509 7 23 39,202 Latin /South America 21 28.6 131 28.2 3251 7 22 27,639 North America 14 14.3 74 25.7 3983 11 23 32,971 Oceania 3 33.3 9 33.3 2940 10 22 24,427 SS = slow steaming Source: authors from Alphaliner database (January 2010) and LRF (2009) Table 2 provides additional information. It is based on a selection of 90 outbound services with Antwerp as a port of loading in July 2008 and October 2010. The port pairs considered are all connected via direct line-bundling services, meaning that no sea-sea transhipment takes place at intermediate hubs along the route. We distinguish two periods of analysis. The first period is June-July 2008, when the liner conference system still existed. As such, the case-study for the first period provides a snapshot of fuel surcharge practices in the liner conference era at a time when fuel costs reached unprecedented heights and when slow steaming was not yet implemented. The second period is October 2010 and is a time when slow steaming has been implemented. Indeed, if slow steaming practices already started in the summer of 2008, particularly on the Europe-Far East trade, to cope with the high bunker costs (as reported by Notteboom and Vernimmen, 2008), however, the full impact became visible in late 2009 and 2010. Indeed, more and more shipping lines decided to opt for slow steaming, not only to save on fuel costs but also to absorb the vessel surplus capacity created by the economic crisis. Information on the average one-way distance relates to the distance from Antwerp to the port of discharge, including the diversion distance to call at en-route ports of call is also estimated. The nautical distances were calculated using the Dataloy distance tables. In a few cases, up to seven ports of call are positioned between the loading port Antwerp and the port of discharge. At the other extreme, Antwerp sometimes acts as the last port of call in Europe while the port of discharge is positioned as the first port of call in the overseas service area. Table 2 also depicts the average transit times between Antwerp and the overseas destinations and the average vessel size per trade route. Both elements are key variables in determining the fuel consumption per container carried together with commercial speed of services. The commercial speed of the vessels was determined using shipping lines’ information on total transit times and port time. We decomposed the real transit time on a port-to-port basis into total sailing time, average port time 74

Cariou and Notteboom – Bunker Costs, Slow Steaming and Fuel Surcharges

per intermediate port of call and canal transit time. Differences in vessel size with values reported in table 1 are explained by differences in the characteristics of vessels deployed from Antwerp with those of services from Europe. Table 2. Main characteristics of services in July 2008 and October 2010 of the set of O/D relations considered with port of loading Antwerp Services Observations

Africa Far East India/Pakistan Latin/South America North America Oceania

Distance Size in TEU

15 24 9 23

In nm 4731 11183 7165 5765

2008 2525 7563 3963 3700

2010 3903 9308 4505 4180

12 7

5096 13136

4102 4283 2922 2653

Transit time Commercial speed in days in kt 2008 2010 2008 2010 17.5 17.8 20.1 19.6 25.6 29.1 22 18.4 20.9 24.8 21.3 19.1 17.3 18.1 20.7 19.8 16.6 42.9

17 40.4

20.3 20

19.5 20.1

Notes: (a) Including the diversion distance to call at en-route ports of call on liner service (b) Including total sailing time, total port time at intermediate ports of call on liner service and canal transits

Two markets experienced a decrease in commercial sailing speed. Europe/Far East with a significant reduction on average of 16% in speed and India/Pakistan with a mean decrease of 10%. Furthermore, a decrease in speed does not automatically increase proportionally the transit time as some ports are dropped for some services. Indeed, on most trade routes the average transit time, together with the average vessel size have increased between July 2008 and October 2010, indicating a trend towards the use of larger unit capacities sailing at slower speeds compared to their design speed. The high transit time is not only caused by slow steaming: the use of ever larger container vessels implies a longer total port time on the route since more and more containers need to be handled when the vessel calls at a port. The cargo volume increase is typically not offset by a higher terminal productivity, in net terms leading to more time spent in ports during a round voyage. Also a change in the order of port calls can have an impact on the total transit time between Antwerp and the overseas port of destination. Only Europe-Oceania has seen a decrease in transit time and vessel size. To estimate the ship’s average fuel consumption per trade in 2008 and in 2010, we retrieved information on the design speed and engine power of containerships from LRF database (2009). For the design speed, we considered the average value by vessel categories. For instance, containerships sailing from Antwerp to Africa in 2008 are on average of 2,525 TEU, and the 107 vessels with a carrying capacity between 2,000 and 3,000 TEU reported in LRF (2009) have an average design speed of 22.3 knots. To determine the engine kWh2, we approximated a log-linear relationship between

2

We also considered age but without significant results. A likely explanation is that vessel size already captures the influence of age. 75

Current Issues in Shipping, Ports and Logistics

engine kWh and TEU, with Engine kWh=exp2.97.TEU0.89 and R2=0.86. In our former case, it leads to an engine power of 21,444 kWh. We then estimated the fuel consumption per day using the design speed (22.3), the commercial speed (20.1 in 2008) and equation 2 for a SFC assumed to remain constant at 190 g/kWh. Fuel consumption is then due to engine power required and speed which is assumed to be related to ship speed by a third power. For our typical vessel sailing to Africa in 2008 at a commercial speed of 91% of design speed, (20.1/22.3), the mean fuel consumption per day at sea is 24x0.913x190x21,444/1000000= 74 tons of fuel burned by day at sea in 2008 (at design speed, the ratio is 1 instead of 0.91). Table 3 presents results on fuel consumption per day for all trades in 2008 and 2010. It also presents similar results using the fuel consumption per day/TEU reported in Maersk Line BAF calculator in November 2010. Table 3. Fuel consumption at sea of the main engine in July 2008 and October 2010 in tons/day 2008 2008 2010 Maersk at design at commercial at commercial Line* speed speed speed Africa 98 74 95 191 Far East 261 178 131 238 India/Pakistan 146 124 83 160 Latin/South America 138 107 86 218 North America 151 91 84 156 Oceania 111 83 77 106 * Reported value in November 2010 x by estimated size of containerships in Table 2 Differences between estimated values and reported value by Maersk Line can be explained by the characteristics of services for this company compared to services originated from Antwerp. However, several general conclusions can be drawn. Firstly, values reported by Maersk Line are closer to the fuel consumption at design speed, rather than on fuel consumption at commercial speed. Secondly, for some trades, namely Africa and Latin/South America a huge gap exists between estimated and reported values.

5 | Comparison between fuel costs, BAF and freight rates So far, the analysis focused only on one part of the equation: the impact of slow steaming on the average fuel consumption in metric tons per day. The analysis of the impact on BAF and its share within the total price to be paid by the line’s customers is more difficult to assess. For the former, to the base freight rate, a series of surcharges such as the BAF, the CAF (currency adjustment factor), the THC (Terminal Handling Charges), piracy surcharge (Gulf of Aden/Suez transit), port congestion surcharges (if any) and often also container-equipment related surcharges (e.g. demurrage charges, detention charges, equipment handover charges, equipment imbalance surcharge, 76

Cariou and Notteboom – Bunker Costs, Slow Steaming and Fuel Surcharges

special equipment additional for an open top container or heavy container, etc.) need to be considered. This section focuses first on the impact on BAF while the base freight rate emerges later in the analysis. Table 4. Estimated fuel costs and reported BAF in July 2008 and October 2010 July 2008 (IFO380 = US$ 585 per ton, MDO = US$ 1,125 per ton) Port of loading = Antw erp

Region of port of discharge

Africa Far East India / Pakistan Latin and South-America North America Oceania

Average fuel costs per FEU carried (a)

Average BAF per FEU 1 Oct 10

Difference BAF - fuel cost per FEU carried

Standard Deviation

US$ 1112 374 913 789 662 1691

US$ 1329 1003 847 1308 1195 1453

US$ 217 629 -66 519 533 -238

US$ 134 185 25 352 75 58

Minim um Maxim um Ratio BAF difference difference versus fuel BAF - fuel BAF - fuel cost per costs costs FEU carried

US$ -42 426 -83 11 296 -285

US$ 286 846 -33 1119 562 -176

Ratio 1.20 2.68 0.93 1.66 1.81 0.86

Base freight rate per FEU 1 Oct 10 US$ 1798 93 592 1628 371 1628

Ratio BAF versus base freight rate per FEU carried Ratio 0.74 10.82 1.43 0.80 3.22 0.89

October 2010 (IFO380 = US$ 435 per ton, MDO = US$ 680 per ton) Port of loading = Antw erp

Region of port of discharge

Africa Far East India / Pakistan Latin and South-America North America Oceania

Average fuel costs per FEU carried (a)

Average BAF per FEU 1 Oct 10

Difference BAF - fuel cost per FEU carried

Standard Deviation

US$ 684 184 458 464 431 1178

US$ 1077 238 738 1186 389 1407

US$ 393 54 280 722 -42 229

US$ 163 96 74 535 61 156

Minim um Maxim um Ratio BAF difference difference versus fuel BAF - fuel BAF - fuel cost per costs costs FEU carried

US$ 110 -84 192 -162 -129 104

US$ 531 116 362 1258 -1 396

Ratio 1.57 1.29 1.61 2.56 0.90 1.19

Base freight rate per FEU 1 Oct 10 US$ 1501 702 669 1828 1854 1841

Ratio BAF versus base freight rate per FEU carried Ratio 0.72 0.34 1.10 0.65 0.21 0.76

The comparison between our estimates on BAFs and those observed in 2008 uses a bunker price of US$ 585 per ton for the fuel grade IFO 380, to which a US$ 1,125 for marine diesel oil (MDO) was added. These figures relate to the average bunker price in Rotterdam in the month of June 2008. Average bunker prices in September 2010 reached US$ 435 per ton for IFO 380 and US$ 680 per ton for MDO. For each port-toport relation we included an imbalance factor retrieved as the mean value reported in Maersk Line BAF and similar values retrieved from the ratio between outbound-toinbound BAF charged by CMA-CGM in October 2010. The mean value is 1.56 for services from Europe to Africa, 0.44 to Far East and 0.98 to Latin/South America, 1.28 to North America the remaining two trades being with a factor of 1. We assumed that the same imbalance factors applied in July 2008. The fuel consumption by the auxiliary engine is assumed to be equal to 10% of the consumption of the main engine (EPA, 2000), to which 10 tons per day at sea were added in order to account for reefers for services to Latin/South America. Table 4 reports final estimates for the BAF values.

77

Current Issues in Shipping, Ports and Logistics

Figure 1. BAF, fuel costs and base freight rate per FEU – port-to-port relations with loading port Antwerp July 2008 (IFO380 = US$ 585 per ton, MDO = US$ 1,125 per ton) Average fuel costs (based on average bunker price June 08)

2000

Average BAF (1 July 08)

1800

Average base freight rate (July 08)

in US$ per FEU carried

1600 1400 1200 1000 800 600 400 200 0 Africa

Far East

India / Pakistan

Latin and North America South-America

Oceania

October 2010 (IFO380 = US$ 435 per ton, MDO = US$ 680 per ton) 2000 1800

Average fuel costs (based on average bunker price Sept 10) Average BAF (1 Oct 10) Average base freight rate (Oct 10)

in US$ per FEU carried

1600 1400 1200 1000 800 600 400 200 0 Africa

Far East

India / Pakistan

Latin and North America South-America

Oceania

Table 4 and Figure 1 bring together the main results of the analysis. Data relates to the transport of one FEU. The figures for BAF and the base freight rate were collected from freight forwarding companies and liner agencies in Antwerp. The following conclusions can be draw. First of all, the BAF per FEU carried is typically (much) higher than the average fuel costs per FEU that we estimated. These results confirm the earlier findings of Meyrick et al (2008) and Notteboom and Cariou (2011) who concluded that the BAF would involve an element of revenue-making for some trades. For June/July 78

Cariou and Notteboom – Bunker Costs, Slow Steaming and Fuel Surcharges

2008, the BAF turned out to be slightly lower than the fuel costs in only 19 of the 90 cases. In October 2010 this figure amounted to 14 cases, most of these on the EuropeNorth America trade. The results underline that the revenue-making character of BAF has not disappeared after the abolition of liner conferences and the wider adoption of slow steaming. On the contrary, four of the six trade routes considered see an even larger gap between BAF and actual fuel costs. The revenue-making characteristic of the BAF became more significant on the shipping routes from Antwerp to Africa (from a BAF/fuel costs ratio of 1.2 in July 2008 to a ratio of 1.57 in October 2010; mainly caused by high fuel surcharges to West African ports), Latin and South-America (from 1.66 to 2.56; mainly caused by BAF practices to destinations in Mexico and the Caribbean), India/Pakistan (from 0.93 to 1.61) and Oceania (from 0.86 to 1.19). Except for Indian/Pakistan, these trade lanes have not been subjected intensively to a shift towards slow steaming. The widening gap between the fuel surcharges and the actual fuel costs on the India/Pakistan route demonstrates shipping lines clearly have not passed on the fuel savings resulting from slow steaming practices on this trade to customers. Part of the explanation might relate to the increasing risks of delays in Indian ports as a result of increased concerns over port congestion. However, if such were the case then congestion surcharges should be used as a means to compensate for delays, not the fuel surcharges. As also a number of West-African container terminals are plagued by severe port congestion, a similar point can be made on the high BAF/fuel costs ratio on the Europe-Africa trade. The fuel savings resulting from significant scale increases in vessel size on the African route (see Table 2) have not resulted in a proportional decrease in fuel surcharges. The Europe-Far East and Europe-North America routes are the only trade routes that have seen a relative narrowing of the gap between BAF and actual fuel costs. Fuel surcharges on the Europe-North America trade are on average no longer sufficient to cover the fuel costs, meaning that part of the fuel costs must be absorbed in the base freight rate. The Europe-Far East route provides the most interesting results, particularly in light of evaluating the impact of slow steaming on fuel surcharge practices. In the summer of 2008 shipping lines were still strongly overcharging customers for the incurred fuel costs (ratio of 2.08). Bunker cost per ton peaked in the summer of 2008 and shipping lines seized this opportunity to charge disproportionately high fuel surcharges. The situation eased somewhat in 2010 with most shipping lines now overcharging customers for the incurred fuel costs with BAFs typically at 10% to 50% above fuel costs (average ratio of BAF/fuel costs of 1.29). The increased adoption of slow steaming on this trade combined with the deployment of larger vessels has reduced the fuel costs per unit carried. This development did not lead to a widening of the gap between BAF and these fuel costs. While fuel overcharging is still common practice, more of the fuel cost savings are passed on to customers than in July 2008. The broader adoption of all-in rates and the use of relatively moderate fuel surcharges suggest that the Europe-Far East trade is becoming a trade route where shipping lines seem to have tempered BAF revenue-making strategies. Shipping lines’ pricing practices on this trade route combined with a limited possibility for shippers to verify base data make it harder for shippers to prove that the savings generated by slow steaming are not passed on to them in an adequate way.

79

Current Issues in Shipping, Ports and Logistics

Variations exist in the difference between BAF and the estimated fuel costs per FEU (see minimum and maximum values in Table 4). The spread in observations is particularly high for Latin and South America. A further investigation of the data stresses that the observed spread is mainly the result of differences in shipping lines’ BAF policy for specific ports of discharge. The BAF strategy of shipping lines with respect to destinations in India/Pakistan, North-America and Oceania is more aligned.

6 | Conclusions This paper aimed at incorporating the impact of slow steaming in the ongoing discussion on fuel surcharge practices of shipping lines. We analyzed the relation between slow steaming practices and BAFs by focusing on three distinct research questions: (a) How significant are slow steaming practices in container liner shipping?, (b) What is the impact of slow steaming on fuel consumption and liner service characteristics?, (c) To what extent has slow steaming changed the relation between fuel costs and fuel surcharges imposed on shippers by shipping lines? Table 1 showed that slow steaming has become a common practice on the Europe-Far East trade while it also gained in importance on a number of other trade routes. Slow steaming practices were initiated in the summer of 2008, particularly on the Europe-Far East trade, as a response of shipping lines to fast rising bunker costs. However, the full impact became visible in late 2009 and 2010 as more and more shipping lines decided to opt for slow steaming, not only to save on fuel costs but also to absorb the vessel surplus capacity created by the economic crisis. This paper showed that slow steaming leads to longer transit times and more vessels per liner service, and significantly reduces fuel consumption of vessels deployed. A case-study including 90 port-to-port relations with the port of Antwerp as the base loading port demonstrated slow steaming has had some impact on the differential between fuel costs and the fuel surcharges imposed on shippers by shipping lines. The results underline that the revenue-making character of BAF has not disappeared after the wider adoption of slow steaming, but the results tend to differ according to trade route considered. The BAF revenue-making strategies of shipping lines have become weaker on the Europe-Far East trade, the main slow steaming trade, but stronger on the Europe-India/Pakistan trade, another major slow steaming liner route. On trade routes with a low slow steaming impact, the BAF typically outstrips the actual fuel costs by a factor of 0.5 to 1.5. The only noticeable exception is the Europe-North America trade with most shipping lines now no longer covering the fuel costs via BAF. One recurrent explanation for the fact that slow steaming has not lead to a closing of the gap between BAF and actual fuel costs is that slow steaming generated additional costs. Indeed, shipping lines had to incorporate more vessels within services in order to keep the weekly frequency. However, this seems not a valid reason to shippers as they are typically paying more BAF for a liner service that shows a poorer performance in terms of transit time.

80

Cariou and Notteboom – Bunker Costs, Slow Steaming and Fuel Surcharges

This paper does not pretend to provide a full answer to all pending issues in this area. While we could present a set of clear conclusions, there is room for further in-depth and comparative research on the relationship between BAF, slow steaming and the actual fuel costs. One obvious extension lies in broadening the scope of the case study to other regions, other shipping lines and other base ports. Such comparative research would reveal whether BAF policies are to some extent port-specific, carrier-specific or route-specific. Another field of further research lies in the analysis of the relationship between BAF, slow steaming and fuel costs on port pairs that are not linked to each other via direct services, but for which sea-sea transhipment in another port is needed before reaching the port of discharge (i.e. interlining, relay or hub-feeder systems). In our case-study, we have only considered direct liner services. References Alphaliner (2010a) Extra Slow Steaming to absorb over 2 percent of ship capacity, Alphaliner Weekly Newsletter, 2, 1-2 Alphaliner (2010b) Retrieved (January 2010) from http://www.alphaliner.com/ Bergh, I. (2010) Optimum speed – from a shipper’s perspective, Container ship up-date DNV, No 2 2010, 10-13 Buhaug, Ø., Corbett J., Endresen, O., Eyring, V., Faber J., Hanayama, S., Lee, D., Lindstad, H., Mjelde, A., Palsson, C., Wanquing, W., Winebrake, J., Yoshida, K., (2009), Second IMO greenhouse gas study, International Maritime Organization, London Cariou, P., Wolff, F-C. (2006) An analysis of Bunker Adjustment Factor and freight rates in the Europe/Far East market 2000-2004, Maritime Economics and Logistics, 8(2), 187201 Cariou, P. (2011) Is slow steaming a sustainable means of reducing CO2 emissions from container shipping? Transportation Research Part D, Forthcoming. CMA – CGM (2010), http://www.cma-cgm.com/eBusiness/BAFFinder/Default.aspx Corbett, J., Wang, H., Winebrake, J. (2009) The effectiveness and costs of speed reductions on emissions from international shipping, Transportation Research Part D, 14, 539-598 ESC (2003) The European Shippers’ Council’s submission to DG Competition of the European Commission on the review of Council Regulation 4056/86, European Shippers’ Council, Brussels ESC (2006) Response from the European Shippers’ Council to the information note on ‘Issues raised in discussions with the carrier industry in relation to the forthcoming Commission Guidelines on the application of competition rules to maritime transport services’ published by the Directorate General for Competition, European Shippers’ Council, Brussels, October Eefsen, T., Cerup-Simonsen, B. (2010), Speed, carbon emissions and supply chain in container shipping, Proceedings of the International Association of Maritime Economists Conference, Lisbon, 7-9 July

81

Current Issues in Shipping, Ports and Logistics

EPA (2000), Analysis of commercial marine vessels emissions and fuel consumption data, United States Environmental Protection Agency, February Faber, J., Freund, M., Köpke, M., Nelissen, D. (2010) Going slow to reduce emissions. Can the current surplus of maritime transport capacity be turned into an opportunity to reduce GHG emissions? Sea at Risk publishing, Retrieved (August 2010) from http://www.seas-at-risk.org/1mages/speed%20study_Final%20version_SS.pdf Karamychev, V., van Reeven, P. (2009) Why Fuel Surcharges may be Anticompetitive, Journal of Transport Economics and Policy (JTEP), 43(2), 141-155 Kollamthodi, S., Brannigan, C., Harfoot, M., Skinner, I., Whall, C., Lavric, L., Noden, R., Lee, D., Buhaug, Ø., Maritnussen, K., Skejic, R., Valberg, I., Brembo J., Eyring, V., Faber J. (2008) Greenhouse gas emissions from shipping: trends, projections and abatement potential, Final Report to the Shadow Committee on Climate Change, AEA Energy, September Lloyd’ Register Fairplay (2010) World shipping encyclopedia Maersk Line BAF calculator http://baf.maerskline.com/

(2010)

Retrieved

(June

2010)

from

MAN B&W Diesel A/S (2008) Propulsion Trends in Container Vessels, Copenhagen, Denmark Menachof, D.A., Dicer, G.N. (2001) Risk Management methods for the liner shipping industry: the case of the Bunker Adjustment Factor, Maritime Policy and Management, 28(2), 141-155 Meyrick and Associates (2008) Review of BAFs - Transatlantic and Europe/Far East trades, Report prepared for the European Shippers’ Council, Melbourne, May Notteboom, T., Cariou, P. (2011) Are Bunker Adjustment Factors aimed at revenuemaking or cost recovery? Empirical evidence on pricing strategies of shipping lines, in: Cullinane, K., International Handbook of Maritime Economics, Edward Elgar: Cheltenham, forthcoming (April 2011) Notteboom, T.E., Vernimmen, B. (2008) The effect of high fuel costs on liner service configuration in container shipping, Journal of Transport Geography, 17(5), 325-337 Psaraftis, H., Kontovas, C. (2010) Balancing the economic and environmental performance of maritime transportation, Transportation Research Part D, 15, 458-462

82

CHAPTER 5 Integrating Intangible Resources in Strategic Co-Operations of Container Lines: Ships Agents’ Perspective

 Indika SIGERA, Stephen CAHOON and Jiangang FEI

Abstract The discussion of this paper is based on an empirical study conducted in mid 2010 that investigated how intangible resources are integrated during strategic co-operations of container lines. Senior managers of agencies representing global container lines based in Sri Lanka are the respondents of this study. The paper utilises the resource based view (RBV) to discuss the resources in firms. The intangible resources are comprised of intangible assets (intellectual property assets, organisational assets and reputational assets) and capabilities (skills). The activities of several forms of strategic co-operations among container lines are reviewed, namely slot charters, liner conferences and shipping alliances through to mergers and acquisitions (M&As). The empirical study found that the integration of intangible resources is varied among these strategic cooperations; in particular, they differ in M&As. The study also found that the main factors determining the integration of intangible resources were antitrust and other regulations, structure of strategic co-operations and motives for forming them. Due to these factors, the close integration has been limited to few intangible resources. They are sailing schedules, business planning processes, operating and reporting processes. Further the study reveals that loose knitted strategic co-operations such as slot charters, pooling agreements, shipping alliances are more common than M&As among container lines.

1 |Introduction Simatupang and Sridharan (2001) define strategic co-operations as being a combination of vertical and horizontal interactions between firms resulting in more flexibility by combining and sharing resources. These firms may be competing (active in the same supply chain) or be unrelated (Cruijssen et al., 2007). Successful firms view strategic co-operations as a window of opportunity to access their partner’s key capabilities and resources (Hamel et al., 1989). As a transfer of resources may occur 83

Current Issues in Shipping, Ports and Logistics

during an integration, the resource based view (RBV) is a useful means of identifying the resource interactions in strategic co-operations (Chatterjee & Wernerfelt, 1991; Das & Teng, 2000; Kogut, 1988). Of interest however, is that few research studies have focused on resource integration in strategic co-operations. The theory of RBV, developed by Wernerfelt (1984), explains how the resource heterogeneity of firms determines the more intangible-related service differentiation among them. As a result, intangible resources, rather than tangible resources, are recognised as contributing to resource heterogeneity due to them being valuable, rare, inimitable and non-substitutable (VRIN) (Amit & Schoemaker, 1993; Hall, 1992; Itami & Roel, 1987; Michalisin et al., 1997). Whereas the acquisition of tangible resources is clearly evident, intangible resources such as know-how, culture, or networks, which are people dependent, are not so easy to integrate (Dierickx & Cool, 1989).

2 | Purpose of the paper The purpose of this paper is to extend the discussion on intangible resources to the strategic co-operations of container lines by firstly reviewing the existing literature, and secondly by discussing the results of empirical research undertaken in mid 2010 on shipping agencies representing global container lines in Sri Lanka. This paper argues that in the present competitive global context, the effective management of intangible resources is an essential factor for container lines to successfully provide a reliable, punctual, safe, customised service with increasing frequency. Further, this paper asserts that the successful integration of intangible resources is an important consideration for the performance and growth of container lines involved in strategic co-operations.

3 | Strategic co-operations Lamber et al. (1999) and Cruijssen et al. (2007) identify four types of strategic cooperations among transport firms which differ depending on the level of integration, scope and intensity (see Figure 1). In an arm’s length strategic co-operation, communication is limited and although firms may co-operate over a long period of time, it usually involves a limited number of interactions because there is no strong sense of joint operations or commitment (Cruijssen, et al., 2007; Lambert, et al., 1999). Firms in a Type 1 co-operation differ in that there is a co-ordination of activities and planning although they tend to be limited in terms of time-span, extent, strength and closeness. The co-operation usually involves a single activity or division of each partner firm (Lambert, et al., 1999). The Type 2 strategic co-operation involves a longer term integration of each firms’ business planning involving multiple divisions or functions (Cruijssen, et al., 2007; Lambert, et al., 1999). In regard to the most closely integrated strategic co-operations, Type 3 firms have integrated their operations to a level that they regard each other as an extension of itself (Cruijssen, et al., 2007). This type of strategic co-operation is often recognised by researchers as an M&A. The varying level of integration and scope for transport firms as shown in Figure 1 can also be observed within the strategic co-operations of container lines. 84

Sigera, Cahoon and Fei – Intangible Resources & Strategic Co-Operation

Figure 1. Categorisation of strategic co-operations Loosely integrated

Arms length

Closely integrated

Type 1 co-operation

Type 2 co-operation

Type 3 co-operation

Source: adopted from Cruijssen et. al. (2007: 25)

3.1 Strategic co-operations among container lines The spectrum of strategic co-operations among container lines varies from looseknitted slot charters, liner conferences and shipping alliances through to M&As (see Figure 2). Figure 2. Strategic co-operations among container lines Loosely integrated

Slot charter

Liner conference

Closely integrated

Shipping alliance

Pool agreement (Joint service)

Consortia

M&As

According to Notteboom (2004) and Koay (1994) strategic co-operations among container lines can be categorised into three groups depending on their scope and level of integration - (i) trade agreements (liner conferences), (ii) operating agreements (for example slot chartering agreements, pooling agreements (joint service), shipping alliances and consortia) and, (iii) M&As. The level of integration of intangible resources and the scope of activities of container liner strategic co-operations vary depending on the motives for forming the strategic co-operations, the regulations, and the nature of the strategic co-operation. The main objectives for container lines forming strategic co-operations were identified by Song and Panayides (2002), Notteboom (2004) and Lu et al. (2006) as reduced costs, increased freight revenue, expanding service coverage, achieving economies of scale, gaining instant access to new markets, and increased freight revenue. To gain further understanding of the range and diversity of strategic co operations in container lines, each of the categories shown in Figure 2 are discussed in the following sections.

85

Current Issues in Shipping, Ports and Logistics

3.1.1 Slot charter A slot charter is a contractual agreement among partners exchanging space on board vessels under their control (Ryoo and Thanopoulou, 1999). This represents the simplest form of strategic co-operation, where the partners rely on minimal operational integration (Koay, 1994). Slot charter agreements are used by two groups of container lines (Koay,1994). The first group are the container lines not represented in a trade route but have the need to serve a liner route, without investing in any infrastructure. The second group are container lines which are already operating liner services in a particular trade route and have the need to increase their service frequency and space capacity (Koay, 1994). Such container lines will choose to enter into a slot charter agreement with another or a few other container lines. However, the container line, which sells space in the vessel, continues to retain full marketing and operational independence while the slot purchaser retains its marketing independence but forfeits any influence on the design and operations of the liner service (Koay, 1994). Since this form of partnership involves minimal integration between the partners, termination conditions for such agreements are usually more flexible compared to other forms of strategic co-operations (Ryoo & Thanopoulou, 1999). 3.1.2 Liner conference As defined by Tupper (2008: 5), a liner conference is ‘a group of two or more vesseloperating carriers, which provides international liner services for the carriage of cargo on a particular route or routes’, that ‘operate under uniform or common freight rates’. As per Brooks’ (1993) explanation, a conference structure is intended to limit ratebased, and in some cases capacity-based competition through common freight systems. Therefore, the main objective of setting up conferences was to minimise destructive and cut throat price undercutting among container lines. However, researchers such as Bennathan and Walters (1969), and Clyde and Reitzes (1998) argue that conferences could act as cartels because of the substantial scale of economies they possess. Although the concept of open conferences was introduced to the maritime world by the US Shipping Act 1916 (OECD, 2001), according to Sjostrom (2004), the first liner conference was the United Kingdom (UK) - Calcutta India conference, established in 1875. The Transatlantic agreement (TAA), which became active in 1993, is another example of a typical cartel agreement (Heaver et al. 2000). The rules about open conferences can be vague, for example, where some open conferences restrict entry by imposing an entry fee and by challenging the ability of the entrant to provide a common carrier service. A closed conference restricts membership based on the existing members’ requirements of pre-entrance qualifications, for example, the service frequency of the applicant’s service (Koay 1994). Due to the cartel concerns, which protected the ship owners, the European Commission banned liner conferences in 2008 completely by ending the existing block exemption on the general EU competition law (Tupper, 2008). 3.1.3 Shipping alliances According to Tupper (2008), a shipping alliance is a general term used to describe operational arrangements between two or more container lines in a global context. This may include, for example, vessel-sharing or slot charters. Consequently, this is an 86

Sigera, Cahoon and Fei – Intangible Resources & Strategic Co-Operation

arrangement whereby carriers agree to use each other’s vessels on certain routes for increased efficiency and cost savings. According to Ryoo and Thanopoulu (1999), operational synergies are the main attraction for container lines to join shipping alliances. Midoro and Pitto (2000) identify the operational synergies as being wider geographical coverage, possibility to perform vessel planning and co-ordination on a global scale, risk and investment sharing entry into new markets, and an increase in frequency of services. Accordingly, shipping alliances are groupings of operational integrations among container lines that can be further classified into pools and consortia, although, Thanopoulou and Ryoo (1999) identify alliances as being revamped consortia. The first global strategic alliance was set up in 1996 by the container lines of APL, OOCL, MOL and Nedlloyd. Their objective was to establish an integrated Europe-Far East service (Heaver et al., 2000). This was called the Global Alliance. In the same period, the Grand Alliance was formed by Hapag-Lloyd, P&O, NYK and NOL. However, during 1996-1998, two separate mergers occurred between APL and NOL, and P&O and Nedlloyd. After the merger, the new container line P&O Nedlloyd joined Grand Alliance, resulting in the disintegration of the Global Alliance due to a lack of partners. The remaining partners, Global Alliance, APL and MOL invited Hyundai to form a new global alliance, the New World Alliance. The Grand Alliance also became larger with the integration of MISC. With P&O Nedlloyd’s acquisition by Maersk in 2005, the market share of the Grand Alliance reduced. Meanwhile the three Asian based container lines Yanming, K-line and Cosco formed CKY Alliance, which in 2003 became the CKYH Alliance with the joining of Hanjin. As indicated in Table 1, CKYH Alliance currently has the largest market share. The market share variation indicates that while there has not been a large growth, alliances have helped container lines to maintain the market share. 3.1.4 Pool agreement (Joint service) As per Koay (1994), vessel pool agreements represent a more integrated form of operational alliance. Co-operation of this nature allows the partners to pool their vessels and operate a joint service. In a situation where the joint service consists of a few sub-services (multiple sub-service), vessels are utilised in the respective subservices based on the characteristics, rather than the ownership of the vessels (Koay, 1994). Therefore, vessel pooling allows ships of similar sizes and speed capabilities to be grouped together without considering the ownership of vessels. Vessel pools offer smaller ship owners the opportunity to pool resources to compete for large contracts of transport thus enabling ship owners to acquire the required critical mass to facilitate the bidding for such contracts without being exposed to unnecessary market risks (Tupper, 2008). The joint service operates using a vessel pool normally marketed as a separate service by each party. Accordingly, in each vessel all partners’ boxes are loaded according to previously decided slot allocation. The slot allocation is done on the basis of the number of vessels pooled by agreement of each party.

87

Current Issues in Shipping, Ports and Logistics

Table 1. The market share variation of the major shipping alliances Year 1997

Grand Alliance CKY Alliance New World Alliance

547,000 382,000 294,000

Market share% 12.45 11.50 9.50

2000

Grand Alliance CKY Alliance New World Alliance

692,551 649,709 446,381

13.45 12.62 8.67

2005

CKYH Alliance Grand Alliance New World Alliance

1,067,198 989,241 720,708

11.68 10.83 7.89

CKYH Alliance 1,264,640 Grand Alliance 1,251,016 New World Alliance 791,453 Source: adopted from Notteboom (2004)

11.93 11.80 7.46

2010

Alliance

TEU

3.1.5 Consortia Apart from mergers and acquisitions, consortia, are formed for specific route and trade requirements, as opposed to the global reach of alliances (Ryoo & Thanopoulou, 1999). A consortium is an agreement between two or more vessel operating carriers to rationalise their operations by means of technical, operational and/or commercial arrangements, with the exception of price fixing (Tupper, 2008). According to McLellan (2006), consortia in the early days of containerisation were tightly bonded by cost and revenue pooling and corporate union. However, many consortia became inefficient in the late 1980s because of the different objectives and commitments of participating container lines, delays in decision making processes, and the need for the provision of multimodal and logistics services (Ryoo & Thanopoulou, 1999). 3.1.6 Mergers and acquisitions M&As are the most integrated form of strategic co-operations where firms integrate their operations to a level where they consider the partner as an extension of itself; often the co-operation does not have a fixed date of completion (Cruijssen et al., 2007). M&As generate efficiency gains such as economies of scale, enhanced technical progress, or improved management efficiency (Tupper, 2008). During the last two decades, M&As among container lines have increased on a global scale (Brooks & Ritchie, 2006). Although, the majority of container lines acquired are regional operators by global lines, some significant carriers, including APL and DSR-Senator Line, were taken over by NOL and Hanjin respectively (Fossey, 2007; Tupper, 2008). P&O and Nedlloyd merged in 1997 to create P&O Nedlloyd, which later acquired Blue Star and Tasman Express Line (Fossey, 2007; Tupper, 2008). Evergreen became the second largest carrier in the world, in terms of TEU slots under its control, through the takeover of Lloyd Triestino in 1998. In 1999, Maersk acquired the international 88

Sigera, Cahoon and Fei – Intangible Resources & Strategic Co-Operation

shipping operations of SeaLand to form a company controlling 9.2 per cent of world container shipping fleet. After a decrease in M&As in the early 2000s, a renewed interest was led by a US$2.8 billion takeover of P&O Nedlloyd by Maersk to reach fleet capacity of approximately 1.8 million TEU (Fossey, 2007). The main motives for M&As between shipping lines are to access new markets, reduce competition, acquire tangible and intangible resources, and dispose of a loss making entity. However, most mergers and acquisitions have failed to create synergetic growth through the integration of resources (Birkinshaw et al., 2000) due to the differences in managerial or corporate culture (Lubartkin 1983; Scherer, 1980), and the resistance by members of both firms to change (Pitts, 1976).

4 | Strategic co-operations and resource based view The RBV develops the idea that ‘a firm’s competitive position is defined by a bundle of unique resources and relationships’ (Das & Teng, 2000: 1) and suggests that valuable firm resources are usually scarce, imperfectly imitable, and lacking in direct substitutes (Barney, 1991). Thus, trading and accumulation of resources becomes a strategic necessity and when efficient market exchanges of these resources are possible, ‘firms are more likely to continue alone, but market transactions of resources are imperfect or default mode’ (Das & Teng, 2000: 12). Further, efficient exchanges of resources are not possible on the spot market because some resources are not perfectly tradable, as they are either mingled with other resources or embedded in organisations (Dierickx & Cool, 1989). Hence, different kinds of strategic co-operations are adopted by firms to access these resources from partnering organisations.

4.1 Intangible resources As defined by Blair and Wallman (2001: 3), intangible resources are ‘non-physical factors that contribute to or are used in producing goods or providing services, or that are expected to generate future productive benefits for the individuals or firms that control the use of those factors’. Another significant feature of intangible resources is the difficulty in measuring them, an issue made more challenging to understand by some of their omissions from financial statements (Blair & Wallman, 2001). There are some features such as the difficulties to develop and duplicate that makes intangible resources more unique and heterogeneous when compared to tangible resources. However, firms will be in a better position to differentiate their services and be more competitive by better understanding and managing intangible resources rather than competing on similar factors such financial and physical resources which are commonly available for all firms in the industry. The range of intangible resources, as shown in Figure 4, can be categorised into intangible assets and capabilities (skills). Intangible assets are comprised of intellectual property assets, organisational assets, and reputational assets (Fahy, 2002; Galbreath, 2004; Hall, 1992).

89

Current Issues in Shipping, Ports and Logistics

Figure 4. An overview of intangible assets Intangible resources Intangible assets

Capabilities (skills)

Intellectual property assets  Copyright  Trade marks  Patents

Organisational assets  Contracts  Culture  HRM policies  Organisational structure

Reputational Assets  Brand reputation  Company reputation  Service reputation

Employee know how

Managerial know how

Routines

Relational abilities

Source: adopted from Hall (1992), Fahy (2002), Galbreath (2004)

4.2 Intangible resources of container lines Several researchers (as shown in Table 4) have studied service attributes (that is, intangible resources) of container lines. The main objective of these studies was to find the service attributes that determine the choice of container lines by shippers or consignees. Even though the focus of these studies was not from a resource point of view, the findings of such studies have been summarised from a resource perspective in Table 4. Apart from the study done by Lu (2007), the focus of other studies has been on identifying different service attributes, which determine the choice of container lines by shippers and consignees.

90

Sigera, Cahoon and Fei – Intangible Resources & Strategic Co-Operation

In February 2010, based on the above findings, Sigera et al. (2010) conducted the first stage of an empirical study to gain an understanding of how intangible resources are managed by agents representing major container lines in Sri Lanka. A mail survey was used, with a response rate of 51 per cent, equating to 46 respondents, all of which held senior management appointments and had significant experience in the shipping industry. The respondents attached high importance to intangible resources such as business planning processes, accurate documentation (capabilities), overall reputation of the company (reputational) and having a prompt response to shipper’s complaints (capabilities). The findings, as shown in Table 5, confirm the empirical research of Hall (1992), Galbreath (2004) and Fahy (2002) that found that the most valuable intangible resources are reputational resources, capabilities and organisational processes. Of interest is that intellectual property resources such as trademarks and licences were regarded as being relatively less important. This observation is also similar to the findings of Hall (1992), Galbreath (2007) and Fahy (2002). Furthermore, it was found in the Sri Lankan study that other intangible resources such as the ability to provide insurance services, having an employee retrenchment policy, and having a long term contractual relationships with inland transport firms were also ranked low. The Sri Lankan survey indicates that from the perspective of senior managers, not all intangible resources equally contribute to the competitive advantage of container lines. Table 5 shows the intangible resources that are most important for the success of container lines as identified by senior managers. As identified in the RBV, the competitive advantage of container lines depends on an understanding and utilisation of the VRIN resources in container lines. This confirms that some intangible resources are more valuable to container lines in comparison to other intangible resources. The findings of the current study are similar to other maritime-related research by Brooks (1980), Collision (1980), Thenagaku Jamaludin (1995), Chiu (1996), and Lu (2007). They found that accurate documentation, the company’s overall reputation, prompt response to shippers’ complaints, reputation of services offered, and a long term contractual relationship with shippers are all valuable capabilities (service attributes) that shippers value when selecting container lines. Further, respondents were asked to indicate the types of strategic co-operations that container lines have undergone in last ten years. As indicated in Table 6, the most common form of strategic co-operation was a slot charter with a frequency of thirty five and the least common were M&As with a frequency of nine. The findings further indicate that all container lines considered for the study have undergone at least one type of strategic co-operation and seventy eight per cent of them have under gone at least three types of strategic co-operations (Table 7). Therefore, it can be observed that strategic co-operations are common among container lines.

91

Current Issues in Shipping, Ports and Logistics

Collison (1984)

Suthiwartnarueput (1988)

Matear and Gray (1993)

Franckel (1993)

Gibson et al. (1993)

Tengku Jamaluddin (1995)

Chiu (1996)

Lu (2007)

Brooks (1983 )

Pearsons (1980)

McGinnis (1979)

Jerman (1978)

Service attributes (Grouped as resources)

Bardi (1973)

Table 4. Summary of service attributes by type of resources

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

Capabilities (skills) – intangible resources Frequency of service Transit time On time pickup and delivery Co-operation between carriers and users Speed and accuracy of documentation Competitive freight rates

*

*

*

*

*

*

*

* *

*

Quick cargo tacking Fast claims handling Directness of sailing Sales representative services Port coverage

*

*

*

*

*

*

* *

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

* * * *

* *

*

*

* *

*

*

Service reputation and intangible resources Reputation for quality service Loss and damage record Reliability of service/punctuality Willingness of long term contracts

*

*

* *

*

Source: adopted from Lu (2007)

92

*

*

*

*

*

*

*

*

*

*

*

*

* *

*

*

*

*

*

*

*

*

*

*

*

Sigera, Cahoon and Fei – Intangible Resources & Strategic Co-Operation

Table 5. Selected findings of Sri Lankan study from 2010 Intangible resources Mean

Standard Deviation 0.475 0.475 0.489

Business planning process 4.780 Accurate documentation 4.756 Company overall reputation 4.756 Prompt response to shippers 4.610 0.542 complaints Reputation of services offered 4.561 0.838 Long term contractual relationship 4.488 0.779 with shippers Courtesy of sales representatives 4.439 0.673 Licenses 4.415 0.741 Employee training policy 4.366 0.741 The operating and reporting structure 4.366 0.662 Ability of sales rep to handle problems 4.341 0.662 Ability to trace cargo 4.341 0.656 Reliability of advertised sailing 4.341 0.938 schedule Competitive pricing 4.317 0.825 On time pickup and delivery 4.244 0.722 Trade mark 4.200 0.831 Note: Mean scores are based on five-point Likert scale (1=Not important 5=Very important) Table 6. Frequency of strategic co-operations Strategic co-operation Slot charter Pooling agreement (Joint service) Shipping alliance Consortia Liner conference Merger and acquisitions

Rank 1 2 2 3 4 5 6 7 8 8 9 9 9 10 11 12

Frequency 35 34 25 24 19 9

Table 7. Number of strategic co-operations by container line Number of strategic co-operation by container line One strategic co-operation Two strategic co-operations Three strategic co-operations Four strategic co-operations Five strategic co-operations Six strategic co-operations

Percentage 100% 89% 78% 63% 11% 11%

93

Current Issues in Shipping, Ports and Logistics

5 | Integration of intangible resources Based on the results of the preliminary empirical study conducted by mail survey of container line agents, a second stage was developed to examine how the intangible resources identified by the Sri Lankan agents have been integrated in each type of strategic co-operation. To gain a greater depth of understanding, personal interviews were undertaken during mid 2010. Thirty six of the respondents from the preliminary study were selected to be interviewed based on a stratified sampling method. A 100 per cent response rate was achieved with an average of 45 minutes per interview. In order to equally proportionate the respondents, a limit of nine respondents was determined, as it was the least frequency registered in the first stage survey as indicated in Table 8. Table 8. Number of respondents for each type of strategic co-operation Strategic co-operation Shipping alliance Pooling agreement/Joint service Consortia M&A

Number of respondents 9 9 9 9

Based on the findings of preliminary study, four strategic co-operations were initially selected for the second stage of study. They are shipping alliance, pooling agreement/joint service, consortia and M&A. Slot charters were not considered for the second stage study due to their being minimal resource integration. Liner conferences were also not included because the preliminary study indicated that functioning liner conferences were currently non-existent due to anti-competitive laws in many regions of the world, which resulting in liner conferences being banned. Therefore, four types of strategic co-operations were considered for a second stage of study, as shown in Table 8 they are shipping alliances, pooling agreements/joint service, consortia and M&As. The personal interview questionnaire was categorised into seven sections. Section A focused on the role of the respondent in the formation of strategic co-operations, while Section B focused on the main features and motives that encouraged the formation of strategic co-operations. Section C examined the integration of 19 intangible resources in strategic co-operations. Section D questioned the processes adopted by container lines when integrating intangible resources during the forming of strategic co-operations whereas Section E asked which factors facilitated the integration of the intangible resources. Section F and G focused on post strategic cooperation economic performance and organisational success of the container lines. Table 9 shows respondents’ characteristics with respect to job title, years of working experience in the container liner shipping industry and size of the firm (number of employees). Results showed that all participants in the survey were senior managers including chairmen, managing directors, directors, and general managers thus ensuring 94

Sigera, Cahoon and Fei – Intangible Resources & Strategic Co-Operation

involvement in the decision making process of container lines. In order to further ascertain whether respondents understood the container liner industry, respondents were asked how long they had worked in the industry. Table 9 shows nearly 84 per cent of respondents had worked in the container liner industry for more than ten years, and some of them even had many years of working experience in regional offices of the container lines, suggesting that respondents had sufficient industry knowledge to answer questions and thus increase the reliability of the study and some of them had many years of working experience in the regional offices of the container lines. As 73 per cent of agencies had more than 20 staff, the senior managers were also aware of people-related issues (an important intangible resource) during a strategic co-operation. All these senior managers represent leading global container lines in the world and the cumulative market share of these container lines is more than 80 per cent. Table 9. Profile of respondents Job title Chairman Managing Director Director General Manager Senior Manager Working experience (years) 1-5 6-10 11-15 16-20