Report No. 10543-EAP Pacific Islands Transport Sector Study (In Seven Volumes) Volume l: Transport Issues -A Regional Perspective March 1993 Infrastructure Operations Division Country Department III East Asia and Pacific Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. EXCHANGE RATES (June 28, 1991) Fiji F$1.00 = US$0.667 Kiribati A$1.00 - US$0.765 Solomon Islands SISl.00 = US$0.368 Tonga T$1.00 = US$0.760 Vanuatu VtlOO.00 = US$0.884 Western Samoa WS$1.00 = US$0.425 Note: Unless otherwise noted, all prices in this report are in mid-1991 prices. FOR OMCIL US ONLY ACRONYMS AND ABBREVIATIONS ADB - Asian Development Bank AIDAB - Australian International Development Assistance Bureau AMI - Airline of the Marshall Islands ASAS - Australian Staffing Assistance Scheme ASPA - Association of South Pacific Airlines CAAF - Civil Aviation Authority of Fiji CAD - Civil Aviation Department CAF - Currency Adjustment Factor CCNI - Compania Cbilena de Navegacion Interoceania CCS - Chief Container Services CEMA - Commodities Export and Marketing Authority of the Solomon Islands CGM - Compagnie Generale Maritime et Financiere CNCO - China Navigation Co CPD - Central Planning D)epartment CPO - Central Planning Office CSAV - Compania Sud Americana de Vapores DCC - Development Coordinating Committee DPM - Department Ports and Marine DRT - deadweight registered tonnes E.A. - equivalent annual EC - European Community ESCAP - Economic and Social Commission for Asia and the Pacific FAL - Freight All Kind FCL - full container load FIA - Friendly Island Airways GNP - gross national product IBRD - International Bank for Reconstruction and Development ICAO - International Civil Aviation Organization IDA - International Development Association IFAD - International Fund for Agricultural Development IMF - International Monetary Fund JICA - Japanese International Cooperation Agency LCL - less than container load LTB - Land Transport Board MCTW - Ministry of Communications, Transport and Works MOF - Ministry of Finance MOL - Mitsui-OSK Lines MOT - Ministry of Transport MOTC - Ministry of Tourism and Communications MOW - Ministry of Works MPM - maximum permitted mileage NCHP - Navale & Commerciale Havraise Peninsulaire NDB - non-directional beacons This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. NYK ..,1pon Yusen Kaisha OAA - Orient Airlines Association ODA - Official Development Assistance OECD - Organization of Economic Cooperation and Development PAF - Ports Authority of Fiji PAFCO - Government owned fish canner PFL - Pacific Forum Line POL - Polish Ocean Lines PMC - Pacific Member Country PWD - Public Works Department PVU - Plant and Vehicle Unit RERF - Rev_o Equalizion ReOve Fund RTD - Road Transport Department SAS - Scandinavian Airlines System SBN - Societe Bourbonnaise do Navigation SCI - Shipping Corporation of India SCC - Shipping Corporation of Kiribati SCP Shipping Corporation of Polynesia SEAL - Scandinavian East Africa Line SIPA - Solomon Islands Port Authority SMEC - Snowy Mountains Engineering Corporation SMN - Societe Mauricenne de Navigation SMTM - Societe Nationale Malgache de Transports Maritimes SOM - shortest operated mileage SNC - Societe de Navigation Cannaise SPDC - Special Projects Development Corporation STABEX - Export Earnings Stabilization System TCB - Transport Control Board TEU - twenty-foot equivalent unit UNDP - United Nations Development Programme UTA - Union de Transports Aeriens VFR - visiting friends and relatives WPAS - Western Pacific Air Services WSSC - Western Samoa Shipping Corporation WSAA - Western Samoa Airport Authority PREFACE This Pacific Islands Transport Sector provides an assessment of transport Study (PITSS) has its origin in a desk review of infrastructure and its maintenance, documents transport in the South Pacific region that was serious deficiencies, conflrms that looking after undertaken in 1990. That review revealed that existing assdts should take priority over new seor-wide assessment of transport was investment, and outlines actions to establish extremely limited and confirmed the desirability sound irwo cturo manaement. (Because of of preparing a formal transport sector report for its size and relative development, Fiji is not the region-the first such report by the World included in Part II). Part 111 compares air and Bank. At the time (1991) the study was sea transport costs to the South Pacific region undertaken there were six Pacific Island with other Island regions, including the Indian countries that were members of the Bank, Ocean. The hypothesis that transport costs for namely, Fiji, Kiribati, Solomon Islands, Tonga, the South Pacific are excessively high, relative Vanuatu and Western Samoa (although to other regions, is not supported by the membership was extended to the Marshall evidence. Part IV explores cooperation in Islands in 1992 and the Federated States of regional transport, specifically in aviation. Micronesia is in the process of joining the Numerous cooperative arrangements among the Bank). The report examines the transport airlines of the region have evolved to cope with sectors in the six present Pacific Island member the small thin market conditions. Areas where countries as of 1991. Governments can facilitate further gains in efficiency of air services are highlighted. Efficient transport services are vital to the support of economic growth and social well Volume Two of this report contains a being in these Pacific Island member countries series of transport sector surveys for the sbL (PMCs). Accordingly, sectcr performance is Pacific island member countries. These surveys examined and areas of major concern are provide background on the institutional identified. The main objective is to provide an arrangements, market conditions, operations, assessment of the transport sectors in the PMCs, regulatory regimes, finance, cost recovery, and to indicate areas which warrant priority investment and maintenance, planning processes, attention by PMC Governments. external assistance, and major problems in each country's transport sector. Each country survey The study is reported in two volumes: (except Fiji) is supplemented by a Maintenance Volume One - A Regional Perspective on Annex which sets out specific details of the Transport Issues, and Volume Two - Country transport infrastructure maintenance situation in Surveys. Volume One presents an analysis of the country. transport issues which prevail across the region. Part I provides a rtgional overview of the status, The report is based on an initial mission problems, and priorities concerning the transport visit to the region in early 1991, which allowed sectors of the six countries. On the basis of this around 4 days in each country, and a follow-up overview, three major transport issues were mission to five countries, of similar duration, in selected for in-depth examination. Part II September 1991. A separate brief field visit to survey selected regional airlines was undertaken in October 1991. A draft of this report (Volume One and each individual country survey) was discussed with the Government of each PMC during the period June-July 1992. A special thanks is extended to the government officials and industry representatives in the PMCs for their cooperation, kind assistance, and valuable comments during the course of this study. Ih tdy wa strucwred and managed by Colin Gannon (Senior Economist, EA3IN) with substantial assistance from consultants, who were financed under the Australian International Development Assistance Bureau (AIDAB) South Pacific Facility. Ihe principal authors are Colin Gannon; David Bray and Ian Gordon (Parts I, II and the Country Surveys); Paul Wait and Richard Bullock (Part III); Peter Forsyth and John King (Part IV). Helpful comments on an earlier draft were received from several Bank colleagues, as well as staff members of a number of international assistance agencies, including ADB, AIDAB and ESCAP and are gratefully acknowledged. Secretarial support in putting the report together was provided by Daphne Glass. PACIMIC ISLANDS TRANSPORT SECTOR STUDY VOLUME ONE: TRANSPORT ISSUES- A REGIONAL PERSPEC'IVE CONTENTS EXECUTIVE SUMMARY ..................................... . viii Pan I. REGIONAL OVRVIEW OF TE TRANSPORT SKrR 1. INTRODUCTION ....................................... 3 A. OverviewContext ..................................... 3 B. Background to Economic Development ........................ 3 C. Performance and Prospects .......S......................... S 2. ECONOMIC CONTEXT .................... 6 A. Demand for Transport Services ............................. 6 3. INSTIrTIONAL STRUCTURE ................... 9 A. Government Structure ................................... 9 B. Agency Responsibilities and Capacities ........................ 9 C. Private Sector Participation ................................ 12 D. Public Finance ....................................... 13 E. Aid Environment ...................................... 13 4. TRANSPORT SECTOR DEVELOPMENT NEEDS ....... ........... 16 A. Information and Planning ................................. 16 B. Investment and Maintenance ............................... 17 C. Institutional Issues ..................................... 18 D. Management of Environmental Impact ......................... 20 Part II. MAMNENANCE MANAGEMENT OF TRANSPORT INFRASTRuCTURE 1. INTRODUCnON ....................................... 25 A. Context ............................................ 25 B. The Maintenance Issue .................................. 25 C. Maitenance Management Framework ......................... 26 - ii - 2. NATURE OF THE MAINTENANCE ISSUE ......... .. ........... 30 A. Transport Infrastructure Inventory ........................... 30 B. Maintenance Expenditure ................................. 39 C. Implications of Inadequate Maintenance ........................ 45 D. Maintenance Funding and Cost Recovery ....................... 54 E. Conclusions ......................................... 58 3. MAINTENANCE DEVELOPMENT AND IMPLEMENTATION ........ .. 59 A. Components of a Maintenan.'e System ......................... 59 B. Institutional Development ................................. 59 C. Implementation Requirements .............................. 63 D. Action Plan ......................................... 65 Part III. COMPAISON OF REGIONAL TRANSPORT COSTS 1. INTRODUCTION ....................................... 75 A. Background and Objectives ................................ 75 B. Approach ........................................... 76 2. ASSESSMENT OF SHIPPING SERVICES ........... ..0........... s A. Shipping Services to the South Pacific ......................... 80 B. Structure of Pacific Islands Market Competition ................... 82 C. Shipping Services to the Indian Ocean Islands .................... 85 D. Assessment Methodology ..................... ......... 85 E. Assessment of Shipping Rates .............................. 90 F. Conclusions ......................................... 96 3. ASSESSMENT OF AIR SERVICES .......................... .. 98 A. Air Services in the Pacific Region ........................... 98 B. Assessment Methodology ................................. 99 C. Assessment of Air Fares ................................. 102 D. Conclusions ......................................... 105 Part IV. REGIONAL AIRLINE COOPERATION 1. INTRODUCTION AND BACKGROUND ........................ 113 A. Problems of Smallness and Remoteness ........................ 113 B. The Cooperative Response: Proposals and Actual Arrangements ....................................... 114 C. The Structure of the Analysis .............................. 115 2. ISSUES IN AIRLINE COOPERATION ........... .. ............. 116 A. Cooperation as a Management Decision ........................ 116 B. Tbe Costs of Cooperation ................................ 116 C. Efficient Cooperation ................................... 116 * lii - 3. CONSTRAINTS TO AIRLINE COOPERATION ,............. 119 A. Reasons for Non-cooperation .119 13. Market Failures at the Airline Level .......................... 119 C. Government Objectives: The Interaction of Tou-ism and Aviation .120 D. Institutions and Incentives .122 E. Negotiation Costs ...... ............................... 124 4. EFFICIENCY, COOPERATION AND COMPETITION .125 A. Giving Meaning to Efficiency .125 B. Cooperation and Efficiency .126 C. The Decision to Cooperate ............................... 127 5. COOPERATION BETWEEN AIRLINES IN THE SOUTH PACIFIC .129 A. Cooperation between Airlires .129 B. Cooperation in the South Pacific ............................ 129 C. Patterns of Cooperation .130 D. Problems with Cooperation .................... . 134 E. General Assessment . .......................... 136 6. ASSESSING THE GAINS FROM IMPROVED COOPERATION . .137 A. Indicators of Inadequate Cooperation .137 B. Airfares and General Cost Levels .137 C. Aircraft Utilization .139 D. Load Factors .140 E. Unremunerative Sectors .... .............................. 140 F. Schedule Inconvenience .142 G. Other Gains from Cooperation .142 H. Conclusion .143 7. AN ENVIRONMENT AND INCENTIVES FOR EFFICIENT COOPERATION . 145 A. The Constraints on Cooperation .145 B. Supporting an Environment for Cooperation .147 C. Issues to be Resolved .150 8. CONCLUS'ONS .152 BIBLIOGRAPHY ............................................ 157 - iv - LIST OF TABLES PART I Table 1.1 Paciflc Islands-Population and Demographic Trends .... .......... 4 2.1 Pacific Islands-Visitor Arrivals, 1988 ....................... 7 3.1 Pacific Islands-General Institutional Arrangements .... .......... 10 3.2 Pacific Islands-Fiscal Indicators, 1985-89 ..................... 14 4.1 Pacific slands-Areas Readily Amenable to Cost Recovery for Government Infrastrcture and Service Provision .... ......... 20 PART JR Table 1.1 Pacific Island Countries-Summary Management Informatior. Inventory ....................................... 29 Table 2.1 Pacific Islands-Transport Infrastructure Inventory, 1991 ............ 31 Table 2.2 Pacific Islands-International Comparison of Road Subsector Indicators .32 Table 2.3 Pacific Islands-Infrastructure Replacement Value, i991 .34 Table 2.4 Pacific Islands-Marine and Aviation Indicators, 1988/1991. 35 Table 2.5 Pacific Islands-Infrastructure Maintenance Overhang, 1991 .38 Table 2.6 Pacific Islands-Transport Infrastructure Maintenance Costs, 1991 ...................... .............. 42 Table 2.7 Pacific Islands and Comparator Countries-Assessed Unit Unit Annual Road Maintenanse Costs, 1991 .43 Table 2.8 Pacific Islands-Capital, Maintenance and User Costs of Infrastructure .47 Table 2.9 Pacific Islands-Effect of Maintenance on Infrastructure Capital Cost and Economic Life .48 Table 2.10 Pacific Islands-Costs and Benefits of Improved Road Maintenance, 1991 .49 Table 2.11 Pacific Islands-Transport Infrastructure and Development Expenditure, 1985-1988, 1991 .52 Table 2.12 Infrastructure and ODA-Australia and New Zealand, 1988 .53 Table 2.13 Pacific Islands-Transport Sector Cost Recovery, 1901 .56 .V. PART m Table 2.1 South Pacific Iands "Pausing Servien, 1991. .... .. SO Table 2.2 Pacific Islands-Dedicaed Extra-Regional Servica, 1991 ....... 82 Table 2.3 Pacific Islands-Regional Shipping Opeators, 1991 ...... 83 Table 2.4 Idian Ocean Islands-PPwing Sevicas Calling, 1991 ... . . 86 Table 2.5 Indian Ocean Ishlns-Dedicated Extra-Regional Servica, 1991 ........................................... 86 Table 2.6 Pacific and Comparator Ilands Average Frequency of Major Contaor Servica, 1991...... .. 89 Table 3.1 Paciflc blands from the Pacific Rim-Interational Airline Servico, 1991 . 101 Table 3.2 Aircraft Typ nd Typial Sat Capcity 103 Table 3.3 Paciflc and Comparator Regions-Air Fares, Sector Distances ad Capacity, 1991 ...... ..... 104 ivA_~r w Table S.1 Cooperation Between South Pacific Airlines, 1991 . . 130 Table 5.2 Cooperation Betwee South Paciflc and Pacific Rim Airlines, 1991 ..131 Table 6.1 Cost Recovery in South Pacific Airlines, 1990 .138 Table 6.2 South Pac-ific Airlines-Aircraft Utilization Rates, 1990 .139 Table 6.3 South Pacific Airlines-Load Facors, 1989 or 1990 .140 Table 6.4 Sharing Arrangements on Sectors Without Traffic Rights .141 LIST OF BOXES PART II Box 2.1 Maintenance, Infrastructure, and User Costs .40 Box 3.1 Continuity of Organizational Arrangements ..61 PART IV Box 1.1 Aircraft Sharing Arrangements in the South Pacific . .115 Box 2.1 The "Freedoms" of the Air ..117 Box 3.1 Some Mfyths About Tourism ..121 Box 4.1 Competition and Cooperation: Some Traffic Trade- Offs. 127 _ vi - Box S.1 Choice of Partner: South Pacific Airline or Pacific Rim Airline? ........... ........................... 132 Box 5.2 A History of Cc .erative Involvements: The Csse of Ansett ............ ................................ 134 Box 5.3 Airline Cooperation in the South Pacific: Some Ventures That No Longer Exist ................................. 135 Box 6.1 Traffic Rights and Seat Purchases .......................... 141 Box 6.2 Parallel Schedules .................................... 143 Box 7.1 A Regional Airline: Advantages and Disadvantages .... ........... 148 Box 7.2 A Pacific Aircraft Leasing Company: Advantages and Disadvantages ..................................... 149 LST OF FRGURES PART IX Figure 1.1 Infrastructure Maintenance Management .27 Figure 2.1 Pacific Islands-Infrastructure Value vs GNP, 1991 .37 Figure 2.2 Pacific Islands-Actual vs Assessed Maintenance .44 Figure 3.1 Management, Finance and Delivery Roles for Maintenance .60 PART II Figure 1.1 Pacific Ocean Islands-PMCs and Comparators .77 Figure 1.2 Indian Ocean Comparator Islands .79 Figure 2.1 Actual Distance an I Freight Rates-PMCs and Comparators .88 Figure 2.2 Pacific Islands-Estimated Shipping Costs and Actual Freight Rates .91 Figure 2.3 Pacific Islands Region-Actual Freight Rates and Distance .94 Figure 2.4 Pacific Islands Region-Actual Freight Rates and Traffic Density .95 Figure 2.5 Pacific Islands and Comparators-Shipping Rata Analysis Results .97 Figure 3.1 Pacific and Comparator Regions-Average Economy Fares .106 Figure 3.2 Pacific and Comparator Regions-Average Excursion Fares .106 Figure 3.3 Pacific and Comparator Regions-Estimated and Actual Economy Fares .107 Figure 3.4 Pacific and Comparator Regions-Estimated and Actual Excusion Air Fares .107 PART IV Figure 6.1 Traffic Rights and Network Efficiency ....................... 142 - vii - LIST OF ANNEXES PART II Annex 11. 1 Management Information for Infrastructure .................. 165 Annex 11.2 Data and Methodological Issues in the Analysis of Infrastructure Maintenance .167 PART III Annex M.1 Statistical Data and Sources .169 Annex m.2 Maps of Shipping Services for Pacific Islands, Comparator Countries and Pacific Rim .175 Annex HI.3 Schedule of Shipping Route Tariffs and Volumes .187 Annex I1.4 Shipping Cost Model .191 Annex m.5 Statistical Analysis of Shipping Rates - Detailed Results .196 Annex 111.6 Air Fares, Distances and Traffic Volumes for Pacific Island and Other Routes, 1991 .198 Annex 11.7 Air Service Routes for Pacific Islands, Comparator Countries and Pacific Rim .202 Annex 1.8 Maps of Air Services for Pacific Islands, Comparator Countries and Pacific Rim .215 Annex 111.9 Airline Codes .223 Annex HI.10 Results of Statistical Analysis of Air Fares for PMCs and Comparators ......... 224 PART IV Annex IV. I Air Service Route Networks-Pacific Islands and Pacific Rim .230 Annex IV.2 Broader Objectives and Aviation Cooperation .231 _ _~~~~~~~~~~~~~~~~~~~~~~~~~ - viii - EXECUTIVE SUMMARY TE CONWTET FOR TRANSPORT IN significantly lower for Solomon Islands. Tm}S SOUTH PACIFIC REGION Remittances from abroad are also very substantial for Tonga and Western Samoa. The rich cultural traditions exert major influences, 1. Spedal Charateristics of the Pacific especially on the use of land, and, in keeping Islands. The Pacific Island World Bank with the extended family structure, the attitude member countries (PMCs) have a number of to provision of social services (for example, to geographic, demographic, cultural, and isolated communities), endowment characteristics whicli condition their prospects for economic development. The 3. Transport Circumstances. The special countries are situated in the center of the South geographic, economic, and social characteristics Pacific region yet, as small groups of islands, of the PMCs give rise to circumstances which they are geographically remote from their major shape and constrain the nature and cost of markets around the Pacific Rim and in Europe. transport services. The remote and dispersed Their land areas are small and Solomon Islands locations of the islands, combined with the small and Fiji account for three-quarters of the total of size of their populations and small scale of their 62,000 square kilometers. Several PMCs, and economic activities, implies that transport especially Kiribati, involve many tiny islands services involve low volumes of traffic over spread across vast areas of ocean. While relatively long distances, i.e., the transport climate is conducive to a wide range of markets for passenger and freight movements are activities, the PMCs are very vulnerable to "thin". These "thin" markets prevail for both tropical cyclones. Western Samoa has international and domestic transport. Such experienced two major cyclones in the last two transport markets are more costly to serve, years. principally because there are economies of scale-especially in transport infrastructure but 2. Although natural resource endowments also in vessels and aircraft. Moreover, in thin vary considerably among the PMCs, in many markets, frequency of service (ship calls, respects their resource bases are better than most scheduled flights) will be lower; this lower small island economies. Overall, agricultural quality of service can reinforce the low levels of commodities, fish, raw materials, and tourism transport demand for both passengers and cargo. dominate exports; outside Fiji manufacturing is In some cases, such as domestic transport very small. Service sectors are basic although service to isolated communities, the level of quite well developed in Fiji and Vanuatu. demand may be so low that provision of the Living standards are fairly sound; official GNP service is not commercially viable. per capita across the PMCs is in the low- to middle-income range. With the exception of 4. The cost of transport services for the Fiji, a large share of national income comes PMCs, and in particular, international shipping from foreign aid, although the share is freight rates, is also increased by the difference -ix - in size and composition of import and export many cases, determination of the appropriate cargoes. Imports are mainly containerized design standard for airports, ports and roads is manufactured goods while export volumes are likely to dominate adoption of the design smaller, and with the exception of Fiji, almost capacity. exclusively bulk agricultural commcdities/raw materials. As a consequence, there is an Transport Development in the 1980s imbalance in inward/outward vessel capacity, which tends to increase transport costs. 7. The Existing Stock of Infrastructure. However, in some cases, rates for bulk export The main development in the transport sectors of capacity may be set at relatively low levels, the PMCs over the last decade has been depending on voyage patterns and alternative investment in infrastructure. Overall, transport cargo availability. facilities and networks in the PMCs are now well established. The overwhelming majority of 5. Transport services involve both transport infrastructure-roads, wharfs and "vehicles" (vessels, aircraft, cars and trucks, jetties, airports and airfields, and associated etc.) and fixed infrastructure (ports, airports, facilities (such as terminals and storage roads, etc.). Fixed infrastructure is more sheds)-is provided by government. Relative to "'lumpy' than vehicles since the indivisibilities their size and foreseeable needs, a substantial are more signiflcant. The size, number, and stock of transport infrastructure now exists in all frequency of use of vehicles can be adjusted to PMCs. The extent of this stock does vary; it is traffic volume much more easily than relatively low in Solomon Islands. The bulk of infrastructure-especially where traffic volumes the transport infrastructure has been put in place are quite low. Moreover, minimum with finance from bilateral donor grants. levels/standards of some infrastructure (for example, runways and berths) are necessary on 8. Capacity, Standards and Demand. technical grounds. In addition, given the Demand for transport, including supporting geographic circumstances of most PMCs, infrastructure, has in general, increased including the long access distances over water, moderately. However, with the possible the choice of aircraft and vessel types is exception of Fiji, there are no significant constrained. However, the PMCs are located in capacity constraints, delays or congestion. In proximity to some major trans-Pacific and South general, major new investments are not required, Pacific regioiial shipping and air routes which although replacement of some facilities may be can be drawn upon to a limited degree, needed. (In Western Samoa, the two major especially in shipping. Indeed, while recent cyclones have caused substantial damage technological changes in transport have allowed to infrastructure.) In addition, infrastructure transport costs to be most substantially reduced design standards are typically adequate to high. where traffic volumes are high, at the same time In aggregate, the stock of existing infrastructure these changes have also widened transport has become quite high relative to foreseeable options, and lowered transport costs, for the needs, levels of national income, and PMC PMCs, although to a much lesser extent. capabilities to sustain the entire accumulated stock. 6. For the PMCs, the combination of low traffic volumes and lumpiness of some 9. Influence of Technology. Advances in infrastructure (principally in marine and transport technology, especially in container aviation) implies that utilization of infrastructure vessels and wide-body aircraft, have influenced will be low and unit fixed costs will be high. It the design of infrastructure, especially for follows that very close attention needs to be international ports and airports; in some PMCs given to infrastructure investment decisions. In facilities have been introduced to accommodate x - these technologies. However, with the exception of fish and timber resources. (For Fiji, the of Fiji, international traffic volumes are broader economic base requires more relatively low and actual use of large aircraft, in comprehensive planning of transport particular, is very limited. Consequently, the development.) Therefore, a challenge for the utilization and productivity of most major future is to address these fundamental infrastructure facilities in the PMCs is low. In intersectoral issues. Specifically, attention needs domestic transport, traffic flows are also modest, to be given to sector-wide and intersectoral but in most cases the choice of transport planning and to reflect their implications in technology is appropriate to the small volumes. project appraisal. Local airfields are unsealed coral/gravel or grass surfaced and small turbo-prop aircraft are used. 11. Management of the Existing Interisland/coastal shipping is very basic and Infrastruct6re. The major challenge now vessel age is high but services are generally confronting PMC Governments in their transport tailored to demand. Road vehicle ownership is sectors is looking after the valuable stock of high and social mobility important in the PMCs transport infrastructure that has been established. and investment in roads has been substantial. Accordingly, any new infrastructure investments However, in some PMCs road standards should be subject to stringent economic (especially sealed roads) are somewhat high, appraisal; they should be evaluated against a given the low levels of traffic. strategic view of the sector, and against the priority needs which prevail for maintaining Challenges for Transport Sector Management valuable existing facilities. Primary attention in the 1990s needs to be focused on the efficient operation, maintenance and, where appropriate, 10. Intersectoral Balance. Development of r%placement of existing assets. the transport sector in the PMCs has taken place in a somewhat ad hoc manner. Moreover, the 12. A key management issue which all PMC level of infrastructure investment has been high. Governments face is how to obtain the best These factors suggest that attention should be value from their existing infrastructure. This given to assessing intersectoral balance. In involves developing capability to take decisions particular, it is important to now gauge the on the most effective allocation of scarce marginal productivity of additional expenditure maintenance resources. In order to guide these in transport infrastructure relative to other decisions, priorities will need to be established sectors. Excessive allocation of resources to the for the maintenance of existing facilities-based transport sector also distorts intersectoral on cost-effective delivery of maintenance and linkages. The relatively high level of transport realistic assessment of the benefits expected. investments in the PMCs is likely to have shifted The PMCs would benefit by sharing investment responses of agents in other sectors. maintenance experiences and by utilizing a In general, the high level of investment in similar approach to maintenance management. transport infrastructure has influenced the spatial distribution of population and economic activity; 13. Reconstruction or upgrading of existing concentration may have been less than optimal to assets should be evaluated thoroughly, in the support economic growth. The spatial outcome, same way as new ("green field") investments. in turn, has shaped the demand patterns for For small projects, guidelines (traffic volumes, transport. Transport development should serve unit costs, standards) should be established to (on the basis of economic/social merit) the facilitate cost-effective appraisal. Selected post- development role of key growth sectors in the project evaluation should be undertaken to test PMCs, for example, tourism, smallholder appraisal and to identify major lessons (for agricultural enterprises, and prudent harvesting example, the pervasive need for institutional -xi - strengthening and inadequacies in project should form integral parts of the management appraisal, design and selection.) component. 14. The main issues that need to be addressed 18. In the short-term, maintenance priorities are weak maintenance management, excessive should be based on meeting appropriate deterioration of assets due to past inadequate operational standards. However, as maintenance, lack of information to articulate rehabilitation and reconstruction decisions arise, priority maintenance needs, inadequate budget a full reappraisal of each facility should be support for economically justified maintenance, undertaken. In some cases it may be appropriate limited delivery effectiveness, and substitution of to modify existing infrastructure, in particular to donor-financed rehabilitation for continuous increase the standard of some facilities, and maintenance. allow the standard of others to be reduced. Such adjustments in the existing infrastructure 15. Maintenanee of Infrastructure. enable maintenance efforts to be directed to Increased awareness is needed to overcome the priority uses. Without such adjustments, approach to maintenane often adopted at mantenance expenditure may be attracted to less present. Given the long life of most transport valuable facilities and entrench distortions, not infrastructure, there is a tendency to defer or only in the use of scarce maintenance resources, minimize expenditure on its maintenance. This but also in the demand and modal choice by Is understandable but it causes infrastructure to users. deteriorate more rapidly, and to a greater extent, than it should. This imposes increased current 19. Cost Recovery for Infrastructure. User operating costs on users and leads to higher life charges for infrastructure are limited, utilization cycle costs of the infrastructure. As a result, the of assets is low, and total cost recovery in most total costs of providing and using the transport cases is poor. Indeed, existing levels of system are excessive. The higher transport transport costs, including international rates, operating costs increase domestic prices and need to be considered against the prevailing low impair much needed export competitiveness. level of cost recovery for PMC infrastructure. The share of the full cost of priority 16. In all PMCs, funding and execution of infrastructure in total transport costs is relatively maintenance of infrastructure falls well short of low for international transport but can be high what would be required to keep the existing for many domestic services. Current levels of valuable assets in good condition. The efficient cost recovery and user charges are not only low, level of maintenance should be established based but vary across modes. Over time this distorts upon prioritization of assets. the efficient level and choice of mode. 17. Maintenance obligations, and their 20. The main issues are achieving balanced implications for recurrent budget support, should and increased cost recovery, and efficient be highlighted in all project proposals, and infrastructure use across all modes. Addressing evaluated against realistic prevailing delivery these issues poses special difficulty where the effectiveness, and financial feasibility. An level of use (for example, road traffic, air action plan for improving maintenance of passenger movements and cargo volumes) is transport infrastructure should be developed for low, and an important part of the infrastructure each PMC. These plans should be designed (roads, airfields, jetties) involves a high around three key components: management, proportion of fixed costs. This is relatively finance, and delivery of the maintenance task. more important in the domestic area, particularly Monitoring, as well as programming (based for the social purpose of providing reliable upon an appropriately scaled information base) accessibility. - xii - 21. User Charges. The structure and level church organizations), exert some discipline on of indirect and direct user charges should be air fares. In general, the capacity and coverage reviewed for all infrastructure and better of domestic transport services are adequate and targeted for both increased cost recovery and responsive to market demands. As a result, consistency across modes. On efficiency and transport costs, by and large, reflect efficient capacity grounds, a target should be set based on operations conditions. However, excessive user the efficient level of maintenance costs for each costs prevail where infrastructure condition is mode. Special attention should be given to the poor due to inadequate maintenance. evaluation and cost-recovery of roads, wharfs and airfields that provide important basic access 24. The main issues involved in improving but at low utilization and low feasible cost performance of domestic transport service recovery. Su:h evaluation should not only operations, are regulatory reform (especially in consider cost-effective provision of accessibility, domestic/interisland shipping), efficient but also incentives for relocation or expanding provision of non-commercial transport services local services. to remote communities, increasing commercial autonomy of government-owned operators (for hmproving Effidency of Domestic Transport example, airlines, ports, shipping) and Services expanding opportunities for private sector involvement (for example, in domestic 22. Enabling Entry and Exit to Markets. shipping). Primary attention should be given to Although there are some areas where the removing legislated barriers to entry. Existing efficiency of domestic transport services can be entry controls by licensing of domestic/ improved, performance in domestic transport interisland shipping operators serve no efficient markets generally appears adequate. Surface economic purpose; rather they assist collusion, modes (road and water) play the central role in weaken price competition and blunt service the PMCs. In the land transport subsector, delivery. Such controls should be discontinued. Governments have established entry conditions Restrictive licensing of operators by that are typically liberal. Freight markets government, as a means of securing the supply involve low barriers to entry, typically lenient of non-commercial services to remote operator licensing, and small sunk costs (i.e., communities by cross-subsidization, is inefficient the markets are contestable). Licensing of and distortive. Such services should be treated public transport operators appears, in general, as a part of an explicit package of policy benign. As a result, market structure conditions measures designed to provide acceptable levels are conducive to efficiency in operations. of social services for small/remote communities. Domestic interisland shipping and aviation are typically subject to entry control, although for 25. Government Involvement in the shipping, where contestability is strong and Operation of Transport Services. In all collusion weak, it is often loosely enforced in PMCs, in addition to their primary role as practice. In some PMCs there may be owners and operators of transport infrastructure inefficiencies and foregone government revenues (roads, airports and ports), public agencies from controlled entry and flawed competitive remain involved to some degree in the operation bidding for franchise arrangements (for example, of transport services-especially as ship in stevedoring). operators and air carriers. PMC Governments should shift such service operations outside 23. Domestic air services are subject to tight direct government economic control through safety and economic regulation. However, corporatization (subject to commercial directives internal political pressures, and the presence of set by Government) or transfer to private limited services operated by NGOs (for example, ownership. - xiii - 26. Although the markets for transport small, thin, and, for freight movements, service operations in the PMCs are small, sunk imbalanced by type, direction, and time. costs and entry/exit risks are limited; vessels and Against these adverse cost circumstances, aircraft can be shifted readily among routes (or international shipping services are subject to sold). Thus, while the withdrawal of competition (particularly by non-conference government operators may result in few private "tramp" operators), and the national airlines operators in markets at any one time, threat of have developed creative use of cooperative easy entry can be expected to exert discipline on arrangements to counter their lack of individual the prices charged, and service quality offered, scale economies. by incumbents. This reinforces the need to remove subsidies and legislated (economic) 29. Government Involvement in Transport barriers to entry or exit. Nevertheless, given the Enterprises. Some government-owned small size of the private sector in most PMCs, it enterprises, principally national airlines and would be desirable for Governments to monitor shipping companies (such as the Pacific Forum service conditions and provide a mechanism for Line), can face precarious financial positions, the registration of user concerns. partly due to government pressure to serve objectives such as tourism (and other exports), 27. Environmental Issues. Environmental at the expense of commercial performance. negative impact of transport facilities and Also, although limited, some government-owned Services in the PMCs is, to date, limited. operators receive subsidies either directly or However, the local natural environments are through the provision of equipment (such as fragile in some locations (for example, coral vessels and aircraft) to governments as part of atolls), and fixed infrastructure can disturb aid programs. This distorts transport markets natural processes (for example, water flows in and weakens the ability of private operators to lagoons and in logging areas). Import of provide competitive services. hazardous materials (for example, petroleum) places a premium on maritime safety. The 30. National airlines, which serve almost impact of transport activity on the environment exclusively international routes, should be given should be addressed in each PMC's national financial autonomy and directed to operate as environment action plan. efficient commercial enterprises. It is not efficient for governments to use airlines directly Improving Efriciency of Regional Transpor. to pursue broad objectives such as tourism Services development, employment creation and social services; it would be more efficient if PMC 28. International Transport Costs. By and Governments addressed these objectives directly. large, regional and international transport tariffs Airlines, together with other enterprises, can be (ocean freight rates and air fares) are consistent induced, if judged appropriate, to serve specific with tariffs for other similar countries and tourism or social aims through specific payments regions. International transport costs for the or direct inducements from governments. PMCs are indeed higher than for some other regions, but not excessive. The higher level of 31. On balance prices and quality of service international transport costs experienced by the for international transport are in line with PMCs stems primarily from exogenous efficient levels. However, given the location of circumstances that effect transport cost the PMCs, their range of export and import structures, including market size, distance, source markets, and the low volumes of traffic traffic density, transhipment requirements, and that are likely to prevail, international transport variations in demand by direction and over time. costs can be expected to remain relatively high. International transport markets for the PMCs are There is some prospect that air fares on some - xiv - routes may have potential for reduction, if Many of the National Planning Offices are aware substantial expansion in tourism can be of these problems but have been unable to exert achieved. However, in general, the delivered much influence. At the siTr 3 time, Government cost of imports, and the f.o.b. price received for responsibilities aretypica!y fragmented by mode exports, will reflect transport costs that are high (aviation, roads, marine) and assigned to relative to other countries, including those separate departments or agencies. A sector-wide countries with which the PMCs compete for perspective does not emerge and modal exports on international markets. The higher departments do not have adequate capability or international transport costs increase production staff to develop or screen well designed costs, reduce receipts and have squeeze margins. proposals. Such capability would carry weight It follows that production costs in the PMCs in establishing priority programs and financing must be sufficiently lower (for essentially arrangements for the sector. homogenous commodities) than competitors with better market access, if they are to be export 34. Sector Management and Institutional competitive. Therefore, emphasis should be Support. To address effectively the key issues given to fostering low cost local production and in their transport sectors, PMC Governments efficient internal transport. need to take action on two broad fronts. First, each PMC Government should concentrate its CHiANGING RoLE OF GovERNMENr scarce managerial and professional resources on overall management of the sector. Second, each 32. Government Planning and Project PMC Government should establish a program of Evaluation. PMC Governments have not given institutional strengthening and staff training sufficient attention to strategic planning of their which is systematically designed for, and transport sectors. Long-term needs, integrated with, the necessary managerial alid commensurate with projected derived demands professional skills. Each Government's for transport from key sectors for growth, such management role should focus on the following as tourism and agriculture, are not soundly tasks: examined. Assessment of modal alternatives (for example, port expansion vs. road * establishment of a sectoral framework and upgrading) are often overlooked. Within modes, intersectoral coordination, based on the alternative standards for infrastructure are not role of transport in supporting the adequately assessed to identify minimum overall country's strategy for economic growth; costs. Even at the project level, economic appraisal is inconsistently applied. Evaluations * policy formulation on a sector-wide basis are often carried out (primarily for multi-lateral with c_nsistency across modes; and major bilateral projects), but not for the many projects which are small or directed to * adoption of explicit guidelines for the rehabilitation. preparation, evaluation, implementation, and future maintenance of projects, 33. Project proposals and development together with indicators of their programs are often shaped more by recipient consistency with the sectoral framework desires, and external financing possibilities, than and, in particular, their contribution to priority needs. PMC Governments have tended the country's growth strategy and/or their to be reactive to proposals, rather than cost-effectiveness in serving social proactive. PMC Governments should set objectives; such guidelines should also priorities and a strategic agenda for sustainable require explicit assessment of the development of the transport sector as well as operating, mainteiuance and replacement assess reallocation of aid to productive sectors. - xv - cost obligations of accepted projects and Implications for Donor Support their financing; 37. Generous assistance from donors has * identification of priorities for maintaining enabled most PMCs, and especially Tonga, and rehabilitating existing facilities Vanuatu and Western Samoa, to build up a together with a program of financing that substantial stock of valuable transport includes both cost recovery and external infrastructure over the last decade or so. Road assistance; investment has received the major share (around one-half), although airport development has been * safety regulation and enforcement; very substantial in Tonga, Vanuatu and Western Samoa. Transport infrastructure development * formulation of guidelines for protection of has been relatively modest in Solomon Islands. environmental quality, and the By and large, the capacity of existing transport development and exercise of local infrastructure is now more than adequate to capability to assess and address the serve foreseeable development needs. environmental impact of projects. 38. Overseas Aid Within a Sectoral * resource planning, for the sector, Framework. Donors should work with each including manpower needs and training PMC Government to establish a sectoral for all modes; and framework for strategic development, policy formulation, and overall management of their D monitoring sector performance. transport sector. To utilize this framework, donors should assist Governments strengthen 35. For transport service operations, institutional capacities, assemble key information Governments should concentrate on fostering an to enable monitoring of sector performance, 'enabling environment", with incentives and assess policy options which can improve market conditions which promote efficiency efficiency in the provision of transport services, through entry, exit, and pricinig and service and ensure that project proposals are evaluated freedoms for private operators 'r autonomous in a sector-wide context, are sustainable, and are government-owned enterprises. For transport consistent with economic and social objectives. infrastructure, PMC Governments should move to corporatize revenue-earning agencies and 39. Reorientation of External Assistance. introduce clear commercial directives coupled There is a pressing need for a major shift in the with explicit accountability (including a schedule allocation of external assistance. In general, of target levels of cost recovery). overseas aid should be redirected away from new investments to looking after the existing 36. The small domestic markets in the PMCs valuable stock of trarport infrastructure assets. typically will support only relatively few This represents a substantial reorientation of aid operators. Moreover, "natural barriers" to entry programs, and a major challenge for donors and exist to some degree, for example, island the PMCs. If future aid is to be used most remoteness. Therefore, it would be prudent for effectively, the focus of donors needs to shift to Governments to establish a means for the providing assistance that will enable registration of user concerns (for example, a Governments to manage better their transport "Transport Ombudsman Office"). Such an sectors, especially existing public infrastructure. office should develop independent capability and Within the transport sectors of the PMCs, authority to monitor and assess charges and assistance for establishing effective management service standards and report directly to the of all transport infrastructure should be assigned Government at Ministerial level. highest priority. Public sector managerial skills - xvi - are scarce and support is needed to attract, government to the private sector (including select, train, and retain managerial staff. devolution of force account works to private Arrangements should allow consistent and contractors and, for routine tasks, to local continuous development of maintenance communities). Aid to government agencies for management procedures in each PMC. maintenance equipment should be reviewed, and Contracting-in of professional staff (expatriate or ways of assisting development of the private local), and use of 'twinning' arrangements with sector established. overseas experienced agencies, are likely to be the best approaches in the short to medium term. 42. Cohrence of Donor Support. Since the Effective management of infrastructure will PMCs share closely the technical, economic and require, for each PMC, the establishment of key social aspects of these efforts, there is information covering inventories of assets, considerable advantage to regional cooperation maintenance costs, technology options, an combined approaches. Donors should give standards, and user costs and benefits. This is more serious attention to ways to coordinate an essential early task for which donor assistance their assistance. It may be desirable for donors would be very valuable. Its scope should be to concentrate their individual efforts on specific strictly limited to assembling the basic subsectors (and possibly geographic areas) to information needed to guide core management facilitate procurement, establish identification, decisions-especially identification of the best and promote stable commitment to an agreed use and maintenance of existing assets, program of development and assistance. expenditure priorities, budget requirements, standards, and work specifications. 40. External Assistance for Maintenance. Primary financial and managerial responsibility for the maintenance of infrastructure must be increasingly assumed by each PMC. At the same time, donors should focus their future assistance primarily on how best they can contribute to the care of the most valuable existing facilities. In this connection, new ways need to be established which enable external assistance to provide broad support for the various types of maintenance projects. Governments and donors will need o assess priorities and funding requirements for existing infrastructure and negotiate new financing arrangements, including the balance between donor funds and recurrent budget commitment by each PMC. 41. Effective Maintenance. Efficient use of maintenance resources requires effective delivery of maintenance tasks, preceded by sound quality control at the time of initial construction. Donors have been assisting in these areas. However, future assistance should be directed primarily to the transmer of responsibility from PART I Regional Overview of the Transport Sector - 3 - CHAPTER 1 INTRODUCTION A. OvVMwM CoMm are d A4 O(ow volumes and rdeadvy long distances) their composition of trade Is 1.1 Background. Tbe discussion in Part I of bmbolanced (exports are largdy limied to bulk this report provides a regional overview of the primary products and tourism srvices, whoie transport sector In the six Pacific Island Member imports are principally manufactured goods); Countries (PMCs) of the World Bank, namdy, they are vulnerable to natUwal disastes, Fiji, Kiribati, Western Samoa, Tonga, Vanuatu especially cyclones; andforeign asistCaCjl and the Solomon slands, based upon individual arc a wry large par of nationaOl income. Whilo sector surveys for each country.' there are differences across the six island nations, these characteristics shape the structure 1.2 The study as a whole focuses on three and performance of each country's trasport major transport issues in the region: sector, and are at the root of many of the maintenance of transport infrastructure (which is priority transport problems which the countries examined in Part 11), international transport costs share. Accordingly, various influences of these (Part IHo) and regional airline cooperation characteristics are examined in detail in this (Part IV). The overview in this first partof the study. report provides a regional perspective of the status of the transport sector in the six 1.4 Geography. The six countries occupy a countries,and the overall context for examining central position in the Pacific, stretching from these three issues. the Solomon Islands and Vanuatu (north-east of Australia) to the Line Islands of Kiribati in the longitude of Hawaii. The total land area of all B. BACKGROUND TO ECONOMIC countries is only 62,000 km2, of which the DEVELOPMENr Solomon Islands and Fiji account for 45 percent and 30 percent respectively. By comparison, the 1.3 A combination of several characteristics PMCs Exclusive Economic Zones comprise 7.6 distinguish the Pacific Island countries from million km2, of which Kiribati controls nearly other developing countries. These one half. (A map showing the location of these characteristics need to be recognized in countries is provided at the end of this volume.) considering the prospects and constraints for their economic development. The countries are 1.5 The countries range in character from very small; they are remote from their major Kiribati, which consists of 33 small, generally export and import markets; their populations are low-lying coral atolls spread over a vast area of dispersed, and often spread over many islands. the Pacific to Western Samoa with two large The international and domestic transport markets inhabited volcanic islands separated by a narrow -4- strait. Sea transport Is vital for economic and 1970, and Solomon Islands, Kiribati and social integration in Kiribati, while land Vanuatu .n the period 1978-1980. All countries, transrport plays a greater role in Western Samoa. with the exception of Vanuatu which was governed but a condominium administration (UK 1.6 Population. The population of all and France), have inherited an English-based countries is small, ranging from 67,000 in civil service. Kiribati to 730,000 in Fiji (see Table 1.1). Western Samoa, Tonga and, to a lesser extent, 1.8 External Assistance. With the exception Kiribati, have large proportions of their of Fiji, all countries are heavily dependent on populations which have emigrated abroad. This external financial assistance for their emigration has resulted in a significant loss of development programs as current income of the young people with skills, but provides financial national governments is barely sufficient to meet remittances of considerable magnitude relative to current expenditure. Annual official dometic inm. Both the Solomon Islands and development assistance is high, ranging (in ss Kiribati face rural to urban (or remote terms) from US$53 per capita in Fiji, to US$249 community to central community) population per capita for Kiribati, during the period 1980- drift which will influence the structure of 1987. In 1988, for example, the six countrles transport services in the future. There is also with a combined population of 1.5 million potential for this situation to develop in Vanuatu. received US$220 million in official development assistance. 1.7 InstitutIons. The PMCs have had different periods since independence in which to 1.9 Living standards in the PMCs are develop their institutional frameworks and public moderate despite many constraints. GNP per and private sector capabil;ties. Western Samoa capita is at the upper end of the low-income to gained independence in 1962, Fiji and Tonga in middle-income range. Social indicators compare Table 1.1: PACIFIC ISLANDS-POPULATION AND DEMOGRAPHIC TRENDS Total Natural Net Net Fertility Population Migration Population Population Rate growth rate rate growth rate 1988 1980 1988 1980-88 1980-88 1980-88 ('000) (percent) (percent) (percent) Fiji 732 3.5 3.2 2.2 -5.2 1.8 Kiribati 67 4.6 4.3 2.3 -4.3 1.9 Solomon Islands 304 7.2 6.6 3.5 0.0 3.5 Tonga 101 4.8 4.2 2.2 -18.8 0.4 Vanuatu 151 6.1 5.8La 3.1 -2.5 2.9 Western Samoa 168 5.6 4.7 2.9 -25.7 0.3 La 1987. Source: World Bank (1991a). - - favorably with developing countries of the same positions and to provide professional and or higher levels of inceme. The social system specialist skills. The problem is compounded in of all countries, which is based on extended countries, such as Western Samoa and Tonga, family or kinship ties, and the strong subsistence which have a high rate of emigration, sector of the economy, result in a low incidence particularly of highly educated and experienced of poverty. professionals for whom ease of entry into developed countries of the Pacific Rim and their higher wage rates are attractive. C. PERFORMANCE AND PROSPECTS2 1.13 The public sector occupies a dominant 1.10 Growth performance has been modest position in the economy of all PMCs. Aid flows during the 1980s, despite some of the highest are channeled through the public sector and are inflows of per capita development assistance. used primarily for public sector projets. The As a group, the six PMCs recorded an average public sector, therefore, has taken the lead in growth rate in real GNP of only 0.6 percent per performing economic activities whereas, for annum during 1980-1988, while population grew historic and cultural reasons, private initiative at 2 percent per annum. Natural disasters have has been lacking. As a result, the role of public had an impact (though not a dominant influence) agencies aud enterprises in the PMCs is large; on growth performance; all of the island this places a considerable burden on the fiscal economies except Kiribati suffered major budget. To address this burden, public devastations during the 1980s. Political events enterprise tariffs/revenues need to be reviewed also affected the economic performance of Fiji to cover operating and replacement costs. and Vanuatu during this period. Improvements in operating efficiency are also needed. In this connection, moves to 1.11 All six countries have potential for corporatization and privatization of specific sustainable higher growth based upon stimulation functions to increase management effectiveness of private investment in productive sectors and require careful assessment of such aspects as a supportive government policy environment. market structure and potential monopoly power, The limitations, however, are substantial. availa. 'e skills, cultural capability, monitoring Prospects in traditional primary .ommodity capability and accountability, and the functioning export areas appear weak. All countries face of the capitai market. limited natural resources, in most cases dispersed land areas, small, fragmented and 1.14 Except in Fiji, most employment in the remote markets, and, as a consequence, high PMCs stems from traditional primary internal and external transport cost structures.? sources-agriculture, forestry and fisheries. In Vanuatu, Solomon Islands, Western Samoa, 1.12 Human resources present a fundamental Tonga, and Kiribati a large share of agricultural constraint to development in both the public and production is in the form of smallholder- private sectors. All countries, including Fiji, operated, partially-subsistence holdings. rely substantially on expatriates to fill key -6 - CHAPTER 2 ECONOMIC CONTEXT A. DEMAN FOR TRANSPORT SERvicEs 90 percent of exports are to the EEC. Copra and fish account for over 95 percent of all Internatlonal Trade and Transport exports by volume and value. Copra shows significant variation on a year by year basis and 2.1 In all PMCs the principal direct linkages the total export volume has mirrored this of the transport sector are with three key variation. activities: trade, tourism and agriculture. Trade flows for freight are characterized by the 2.4 The principal exports from the Solomon substantial imbalance between imports and Islands are copra, palm oil, fish and timber, the exports (up to 6 times), the imbalance in last two categories accounting for 60 percent of containerization of import and export cargoes, the export trade (by value) in 1988. Japan and and the imbalances between sources of supply of Asian countries are the principal importers of import commodities and destination of export commodities from the Solomon Islands, and commodities. Australia is the principal source of imports. The trade deficit is substantial and in 1988 was 2.2 In contrast to most Pacific Island equivalent to about 70 percent of import countries, imports and exports (by value) to Efjj earnings. are approximately in balance. Exports are dominated by sugar, although other categories of 2.5 Exports from Tonga have grown rapidly exports (for extmple, fish, forestry products) in the second half of the 1980s, due largely to have grown more rapidly. The UK is the single increased trade in vanilla beans, fish and squash. largest export destination for Fiji, accounting for Imports have increased steadily, to the extent about a third of total exports. Other trade is that the value of imports exceeded exports by six focused on the Pacific rim, in particular, times in 1988. This imbalance is also reflected Australia, New Zealand and increasingly, Asia. in volume terms. The Pacific rim countries are More than 46 percent of imports come from the main source of imports and the major Australia and New Zealand. However, from the destination of exports from Tonga. standpoint of transport services, Fiji (and the other PMCs) experience imbalances particularly 2.6 The principal exports from Vanuatu are as bulk movements dominate exports and copra, cocoa, beef and timber. There has been container movements dominate imports. no sustained growth in any major export commodity during the 1980s, whereas imports 2.3 For Kiribai the trade deficit in 1988 was have increased over the period. For 1989, the equivalent to three times export earnings. value of imports exceeded exports by almost five Imbaiance of flow by origin and destination is times. This imbalance in shipping requirements far greater. The Asia-Pacific region is the major is further reflected in trading relationships. supplier of Kiribati's requirements, whereas over Countries in the Asia/Pacific region (New -7 - Zealand and Australia, in particular) are the However, there is scope for developing some main sources of imports and European countries higher value specialized crops in several PMCs the main receivers of exports. (for example, vanilla in Tonga). 2.7 Exports from Western Samoa have shown Tourism little growth over the latter half of the 1980s due to the volatility of copra prices, declining 2.9 Tourism is underdeveloped, with the demand for coconut oil, and quality control exception of Fiji, but suffers from remoteness problems with bananas. Cyclone Ofa is from major source countries, high transport and expected to have cut the volume of export crops on-ground costs, limited resort facilities, and by 40-50 percent in 1990 and Cyclone Val strong competition from well-developed (December 1991) has further affected export alternative destinations. As shown in Table 2.1, potential. The volume of imports has tended to Fiji is the dominant tourist destination; the increase during the 1980s, so that in 1989 the number of visitor arrivals in 1988 being more volume of imports exceeded exports by nearly than double the combined total of the other six times. Asia and Pacific Rim countries are countries. Australia is the principal source the main sources of imports and the South market, followed by New Zealand and the USA. Pacific Islands region, with Europe, the main The figures for Kiribati, Tonga and Western receiver of exports. Samoa are affected by visiting friends and relatives (VFR) traffic from New Zealand, other 2.8 In all PMCs, development of traditional Pacific countries and Australia. Tourism to the crops in the agricultural sector has been beset by region increased during the 1980s, albeit from a low export demand and international prices and very low base. Tourist arrivals in Fiji and the higher share of export (f.o.b.) values Vanuatu declined towards the end of the decade associated with domestic and international as a result of civil and political difficulties, and transport costs.' As a result, residual returns to the effects of several severe cyclones growers have become marginal or even zero. (Vanuatu).5 Table 2.1: PACIFIC ISLANDS-VISITOR ARRiVALs, 1988 No. of Visitors by Nationality (%) Visitors /a USA/ Europe Other Country ('OOOs) Australia NZ Canada & UK Japan Pacific Other Fiji 208.2 36.1 10.3 28.4 13.9 1.6 6.8 2.9 Kiribati 3.4 13.0 8.0 13.0 7.0 6.0 42.0 11.0 Solomon Islands 10.7 36.2 12.1 8.5 7.2 5.5 22.2 8.3 Tonga 19.5 15.0 24.0 22.0 12.0 2.0 13.0 12.0 Vanuatu 17.5 54.8 8.3 3.6 4.0 2.9 19.9 6.5 Western Samoa 49.1 9.5 27.6 7.1 6.1 n.a. 47.2/b 2.5 La Exclude cnise ship passongers. /b Includes visitors from American Samoa. Source: Country reports. -8 - Domestic Transport Requirements 2.13 Land transport demand in Fiji is heavily influenced by the dominant size of the island of 2.10 Internal transport requirements are Viti Levu, and the quite widespread distribution influenced strongly by the geography, terrain of its economic activity and population. The and spatial distribution of population of each land area of the island is small and there is the country. Inter-island shipping is an essential potential for considerable intra-island road service and performs the major role in the travel. Traffic volumes and trip patterns of road movement of freight and passengers in Kiribati, transport can be expected to expand as roads are the Solomon Islands and Vanuatu and, to a upgraded. Even with existing road conditions, lesser but still essential degree, in Fiji and there is an increasing number of full containers Tonga. Sea transport across the strait between reported to be carried by road from Suva to the two main islands of Western Samoa (Upolu Lautoka to use the latter as an export port. and Savaii) is of importance but overall plays a lesser role than the land transport subsector. 2.14 In Tonga, land transport demand is relatively high on Tongatapu, where two-thirds 2.11 Inter-island air services provide a crucial of the population live. Agricultural products are means of communication for business, tourist carried by land transport to Nuku'alofa (the and urgent personal travel, and for the capital) for consumption by the urban population movement of high value, fragile or urgently and for export through the port. Demand for required commodities and equipment. Air and land transport is lower on Vava'u (the other maritime services each provide valuable major island) and minimal elsewhere. functions within all PMCs. 2.15 In Western Samoa, some 70 percent of 2.12 Land transport serves a complementary the population lie in Upolu (where the capital role to sea and air transport outside of the urban Apia is located) and the fertile coastal plains centers in Kiribati, the Solomon Islands and have been developed extensively for agricultural Vanuatu. Its main functions are to provide uses. Savaii, which is separated by a sea strait connection between village communities or from of 21 km from Upolu, is less developed. Both communities to wharves, beach landing areas, islands have effective road systems for personal airstrips and regional facilities. Vehicle travel and freight movement. numbers are low and average traffic volumes on many roads are commonly less than 10 vehicles per day. - 9 - CHAPTER 3 INSTITUTIONAL STRUCTURE A. GOVERNMwN STRucTUE (Kiribati, Tonga, Vanuatu, Western Samoa) they provide construction and maintenance services to 3.1 All countries have a national government other government organizations responsible for administration, although a process of marine and aviation functions. Land transport decentralization of government administration services are provided by the private sector in all and provision of public services by provincial the PMCs. governments has been in progress for some time in the Solomon Islands and Fiji. Vanmatu and 3.3 In the marine subsector statutory Kiribati have a system of local government authorities have been created in Fiji (Port administration which is involved in local Authority of Fiji) and the Solomon Islands development issues, the maintenance of some (Solomon Islands Port Authority) which are local public assets and the administration of local responsible for the operation of the major services. Western Samoa and Tonga have international ports. Elsewhere public service centralized government administrations. organizations are responsible for the administration of the main ports. Although not yet implemented, Kiribati has passed legislation, B. AGENCY REsPONSI1ELrTiEs and Western Samoa has a draft bill, to establish AND CAPAcMEs a port authority. Stevedoring services may be provided by the port organization, by contract Major Transport Functions with private enterprise for the provision of this service within the port, or through stevedoring 3.2 Responsibility for functions in the companies competing for the service directly transport sectors of the PMCs is reasonably well with ship's agents. The responsibility for the defined.6 The general institutional arrangements operation and maintenance of outer island in each country are illustrated in Table 3.1. In wharves and jetties is less clear.' general, the PMCs have developed similar institutional arrangements for the planning, 3.4 Internal marine transport services construction and maintenance of transport (excluding the informal sector) are provided by infrastructure. Without exception, central each Government, the private sector, or both. government responsibilities for land transport Government-owned or supported domestic in.frastructure are administered through shipping services are provided in all PMCs, with ministries or departments of Government. These the exception of Vanuatu where inter-island organizations cover a range of public works services are provided solely by private functions (for example, water supply, public enterprise. Private enterprise provides the buildings) and their responsibilities are not majority of services in the Solomon Islands; confined to road matters. In many cases government services are directed to commercially unattractive routes. A similar - 10- Table 3.1: PAcIFIC ISLANDs-GENERAL INSriTUmIONAL ARRANGEMENTS Mode Land Marine Aviation Infra- Infra- Infra- Country structure Services structure Services structure Services Fiji PSO P SA P + PSO SA GC + P Kiribati PSO p PSO GC + P PSO GC Solomon Islands PSO p SA P + PSO PSO GC + P Tonga PSO p PSO GC + P PSO GC Vanuatu PSO p PSO p PSO GC Western Samoa PSO P PSO GC SA GC PSO Public Service Organization (Ministry, Department) P Private Enterprise SA Statutory Authority or Corporation GC Government Company (wholly or substantially owned by government), Government Corporation Source: Country reports. situation exists in Fiji. In Western Samoa the Transport Regulatory Functions Government is the sole operator, while in Kiribati and Tonga there is limited private 3.6 Regulation of transport services is a enterprise competition. Adequate knowledge of function of government ministries or the size, role and performance of the informal departments in all of the PMCs. The degree of provision of maritime services is extremely regulation and the level of enforcement are limited for all PMCs. (Recent studies of variable. iiiterisland shipping, primarily financed by ADB, are revealing the importance of informal 3.7 In the land transport subsector all shipping.) countries with the exception of Kiribati, do exercise some control over the operation of 3.5 Operation and maintenance of aviation public transport services (taxis and buses) infrastructure is a responsibility of statutory through entry restriction, route service licensing authorities in Fiji and Western Samoa and or both. Fare levels for the provision of undertaken by government departments (or services are typically set by government. By ministries) elsewhere. Internal aviation services and large, entry control is liberal, there is little are provided by government owned (or evidence of high rents or excess demand. Entry controlled) airlines in all six countries. Private control and tariff regulation is not applied to the airlines operate only in Fiji and Solomon road freight industry. Vehicle weight and Islands. dimension controls exist in all countries but few (if any) resources appear to be directed to the enforcement of regulatory limits. While - I1 - localized, the suspected levels of road damage recommended by a study of rationalization of and potential safety problems arising from inter-island services (ADB, 1986). Inter-island overloading of freight vehicles are matters for freight rates, however, are set by the concern in most countries. (Kiribati has only a Government's Prices Control Board. small freight vehicle fleet and limited road system-yet a road safety problem, especially 3.9 The administration and enforcement of for pedestrians in some locations). All PMCs air safety regulations and the provision of air have vehicle registration and driver licensing traffic services are the responsibility of systems. Some countries (for example, Western go"rrnment departments (or ministries) in all Samoa, Solomon Islands, Vanuatu) operate PMts. With the exception of Fiji and the vehicle inspection systems for both private and Solomon Islands, the provision of public aviation public service vehicles as a component of the services is confined to organizations owned or vehicle registration system. Vehicle registration controlled by government. The level of charges are uniformnly low and unrelated to the domestic airfares is a sensitive issue in all PMCs costs of provision and maintenance of road and increases in fares require govermnent systems; overall cost recovery is low. Overall approval. the performance of the land transport subsector is weakened principally by inadequate cost Corporatization and Privatization of recovery and enforcement of vehicle loading; Transport Functions market structure conditions are reasonably conducive to workable competition and 3.10 There is a growing attraction to the satisfactory performance. corporatization or privatization of transport functions in the public sector. Fiji, Kiribati and 3.8 The administration of marine legislation the Solomon Islands Governments have instituted relating to the operation and safety of vessels on specific commercialization or privatization national registers is carried out by government policies. Other countries are considering the departments. Some form of entry control formation of statutory authorities, principally in (licensing) is applied to the formal inter-island the aviation and marine subsectors, to operate shipping system in all PMCs. Route licensing is facilities on a commercial basis. applied in the Solomon Islands and Vanuatu, where shipping services are provided mainly or 3.11 In Fiji proposals have been made for wholly by the private sector, but there appears privatization of the Government shipping fleet to be little surveillance or enforcement of route and shipyard (geared to regional vessels), both compliance. However, the licensing is of which are operated by the Marine Department vulnerable to effective entry control and should in the Ministry of Infrastructure Public Works be removed. Aside from some incumbent-vested and Maritime. The Government of Kiribati is interests, the main resistance to relaxation stems committed to a policy of privatization of major from withdrawal of operators obliged at present public enterprises as a means of reducing costs to serve remote non-commercial islands by cross and encouraging greater private sector subsidization. Explicit contracting by participation. An Act has been passed to create Government for such services would be a Kiribati Ports Authority. In addition, the preferable. Freight and passenger rates are decision to separate the Shipping and the Port essentially determined by individual operators Divisions of the Shipping Corporation of although the Governments do exert influence on Kiribati will require a more commercial tariffs through freight rates set by the approach to the provision of shipping services commodity marketing authorities in each which to date have been cross-subsidized from country. Route licensing of inter-island services port operations. The Solomon Islands is not practiced in Fiji, although it has been Government has initiated a process to privatize - 12 - some public sector functions and to corporatize time. Some devolution of powers, including others. The way in which privatization of public transport functions, has occurred but has not sector activities might occur is still to be proved satisfactory because of resource developed. Indications are that a statutory limitations and regional institutional capabilities. authority will be proposed for the operation of Decentralization of transport functions is being the major airports (Henderson and Munda) and reassessed, particularly for road and airfield for general aviation policy, air navigation and maintenance, and the operation of some air safety. A Bill to establish a Western Samoa government inter-island vessels. Fiji has a local Ports Authority is in final draft form. When government system based on fourteen provinces established the Port Authority would take but responsibility is directed to village responsibility for all port and ferry facilities and development and rural health. A policy of for navigation aids in Western Samoa. The devolution of nominated functions to Local Authority will be subject to the general policy of Government exists in Vanuatu (as set out in the Cabinet in the exercise of its function, but would Decentralization Act, 1990). Transfer of retain revenue raised to meet operating costs, functions has been confined to village matters interest and loan repayments and provide for although the transfer of ownership and depreciation and general reserves. maintenance responsibility for local transport infrastructure (roads, wharves, airfields) is under 3.12 Capacity and Constraints. A program consideration. A system of local government for improving the performance of statutory administration, consisting of 17 Island Councils organizations including corporatization and and two Town Councils, has developed in privatization of selected functions should be Kiribati. Responsibility is confined to local developed. However, its feasibility and administration and participation in local effectiveness will depend crucially on existing development issues. capabilities and market conditions. There is little tradition of private initiative in the PMCs, 3.14 In general, considerable fragmentation of the capacity for entrepreneurial risk taking is the planning, construction and maintenance, limited, the effective commercial management of operations and regulatory functions remains in major enterprises is not well developed (with the the transport sector. Kiribati and Western exception of Fiji), and local finance and capital Samoa have tended to concentrate transport markets are shallow. In addition, the smallness sector responsibilities within one or two of markets increases the potential of (private) ministries, but other countries have a wider monopoly power and hence the need for dispersal of functions. Given the scale of the facilitating entry, effective monitoring (for functions, the ministerial concentration of example, using international norms), and, if responsibilities is feasible and would allow more necessary, real price reduction directives. effective use of the scarce executive, Donors should develop means by which professional, and administrative resources within assistance can be given to the development of the transport sector. Furthermore, private sector capabilities in the PMCs. amalgamation of policy functions would promote better intra- and inter-sectoral planning and Devolution of Central Government Functions coordination. In some PMCs (for example, Fiji and Vanuatu) a transport capability or unit 3.13 As noted above, in the Solomon Islands within the central planning agency is utilized. a process of decentralization of government administration and the provision of government C. PRIVATE SECTOR PARTICIPATION services through a provincial government system (seven provincial Governments and the Honiara 3.15 The private sector provides land transport town Council) has been in progress for some services in all PMCs and is the sole, or - 13 - principal, provider of inter-island shipping D. PUDLC FINANCE' services in Vanuatu, the Solomon Islands and Fiji. It has a minor role in the provision of 3.19 Tho general fiscal indicators of inter-island shipping services in Kiribati and Government revenue and current expenditure are Tonga. Private sector participation in the set out in Table 3.2. Current governmental provision of air transport services is confined to expenditures are high in relation to GDP but Fiji and the Solomon Islands (see Table 2.1). revenue has been generally sufficient to cover The private sector has an additional role the expenditures, with the exception of the (although mainly subsidiary) in the provision of Solonon Islands and Vanuatu. services to public service organizations and government statutory authorities or corporations 3.20 The ratios of tax revenue to GPD range responsible for the transport sector. from 18 percent for Kiribati to 31 percent for Wetern Samoa, with the other countries 3.16 Extesiv use is made of privato grouped in th 19-21 percent range. Aside from contractors for road construction and Kiribati, non-tax revenue to GDP ratios are low, rehabilitation works in Fiji, the Solomon Islands ranging from 2 percent to 10 percent. With the and Western Samoa. Private sector contractors exception of Fiji, tax revenue is drawn from a are also used for some road maintenance workcs narrow tax base. Tax on international trade and in Fiji and Vanuatu and to a minor extent in transactions constituted the largest source of Western Samoa. In Kiribati and Tongsa all road revenue in all countries except Fiji, where taxes construction and maintenance is carried out by on income and profits dominate. As is common departmental force account operations. Some in developing countries, the transport sector is a limited use is made of community-based labor- major source of tax revenues-from taxes on intensive means for construction or maintence fuels and import duties on vehicles and parts. of minor roads in the Solomon Islands and Vanuatu. 3.21 The dominance of the public sector in all countries (with the exception of Fiji) is indicated 3.17 Major construction works for ports and by the ratio of government expenditures to GDP, airports are generally undertaken by private which ranges from 39 percent to 85 percent. sector contractors in all countries. Where public sector organizations (rather than statutory authorities) are responsible for marine and E. AID ENYRoONMT aviation infrastructure, the organization and supply of construction and major infrastructure 3.22 Official development assistance (ODA) to maintenance services are normally undertaken by the South Pacific region has been substantial and the government public works agency. levels of aid on a gross flow per capita basis are among the world's highest. For the smaller 3.18 International contractors are usually countries official development assistance has employed for major works, with domestic accounted for 25 to 50 percent of GDP; for Fiji private sector construction organizations the figure Is only 3 to 5 percent. employed as subcontractors. With the exception of Fiji, domestic private sector contractor, 3.23 Grants from bilateral donors and the EEC professional, and technical resource capabilities form the major part of the aid to the PMCs, for are limited. both capital development and technical assistance. Traditional donors have been the United Kingdom, New Zealand and Australia, with Japan and the EEC contributing significantly since the mid 1980s. At 1991, - 14 - Table 3.2: PACIFIC ISLANDS-FRSCAL INDICATORS, 1985-89 Solomon Western Fiji Kiribati Islands /a Tonga /b Vanuatu Samoa ------------------------------- (as percentages of GDP) ----------------- Revenue 23.8 46.4/e 23.4k 29.9 25.8 40.2 Tax revenue 19.4 17.5 21.1 20.3 20.5 30.6 Nontax revenue 4.4 28.9 2.2 9.5 5.3 9.6 Grants 1.0 36.7 7.6 19.3 25.8 15.6 Expendituro and net lending 28.0 84.6 38.7 49.2 S4.0 52.4 Current expenditure 23.0 46.8/d 26.1 27.2 36. ILd 21.3 Development expenditure and net lending 5.0 37.8 12.6 22.0 17.8 31.2 Current balance, before grants 0.8 -0.4 -2.8 2.7 -10.4 18.9 Oveall balance, before grants -4.2 -38.2 -15.3 -19.3 -28.2 -12.3 E:ncins Extemal grants 1.0 36.7 7.6 19.3 25.8 15.6 External borrowing -0.5 .. 6.7 2.6 1.5 2.1 Domestic nonbank 2.9 .. 0.3 -0.5 1.1 1.4 Domestic bank 0.8 .. 0.7 -2.1 -0.2 -6.8 Memorandum Item Overall balance, after grants -3.3 -1.Se -7.8 -0.1 -2.4 3.3 /a 1985-88. /b 1985-86 - 1988/89. /c Includes a very small amount of capital revenue. /d Includes technical assistance. k In Kiribati, if reinvested RERF income were included in nontax revenue, the figures for total revenue, nontax revenue, current balance before grants, overall balance before grants, and overall balance after grants would be 58.0, 40.5, 11.2, -26.6 and 10.1 respectively, as percentages of GDP. Source: World Bank (1991a), Table 2.4. about three-quarters of all aid to the region came 3.24 Estimates of the sectoral allocation of aid from these five donors; the World Bank and indicate that some 40 percent is directed to the Asian Development Bank provided 6 percent of primary sectors and to industry; the remainder to the net ODA. The United Nations, through its sectors including infrastructure and human agencies, also provides considerable technical resource development. In the transport sector, assistance. aid requests and finance have been directed, generally, to meeting perceived deficiencies in infrastructure and to training. - 15 - 3.25 Aid requests for transport have reflected the almost exclusive concentration of planning in most PMCs on the public sector. In many cases, little account has been taken of the ability of government to sustain the infrastructure development, either on the basis of cost recovery or general budget support. Nor has much attention been given to private sector development, including the informal subsector. - 16 - CHAPTER 4 TRANSPORT SECTOR DEVELOPMENT NEEDS A. INFORMAnON AND PLAN G is more difficult to draw together a cross- sectoral approach to project identification or a 4.1 Data Base. Adequate information on the comprehensive strategy to guide development. existing stock of transport assets in the PMCs The role of donors in undertaking project and their status is difficult to establish. assessments and developing strategic plans Inventories of transport assets are either through sectoral studies appears to have inhibited inadequate, out-of-date or non-existent. the need for development of a planning Inventory information which does exist is not competence in sectoral agencies. Within the based on consistent reporting standards and transport sector, with the exception of contains little or no information on the quality or international aviation, there is limited condition of assets. Data on system capacity, discernable planning and programming capacity existing and projected traffic demand, and costs in the PMCs, although capacity does vary across and reviews by function, are not generally countries and subsectors. available, or are not in a readily useable form. Typically the order of accuracy of statistics is 4.3 In many PMCs project preparation not indicated. provides neither a clear indication of priority nor an adequate economic appraisal of projects. A 4.2 Planning. All of the PMCs, with the national transport development plan has been exception of the Solomon Islands, use a multi- completed recently in Vanuatu (Wilbur Smith year plan as the central planning instrument. 1990). The plan proposes a series of priority Usually it is the responsibility of the ministries projects over a 15-year period but does not to undertake planning activities and to submit enunciate a strategic approach to transport project proposals to the central planning development. An inter-island transport plan is organization for review and integration into being finalized for Kiribati (Danport, 1991) and national development plans. In the main, central similar studies are proposed for Vanuatu and planning tends to be reactive, rather than guiding Tonga. Western Samoa, unlike other PMCs, sectoral agencies in their planning. A primary has given considerable attention to transport objective appears to be the identification of planning for both individual modes and multi- projects for consideration by aid donors with the modal transport development. Transport result that development plans are concentrated planning in Fiji has been undertaken mostly on on the public sector. The planning organizations a modal basis, but a national transport sector provide a central point of contact for donors, "Master Plan" is currently proposed. Neither although donors may approach Governments Tonga nor the Solomon Islands have undertaken through different ministries. Where national transport plans. The need for a national responsibilities for multi-lateral and bi-lateral aid t&ansport planning framework is most important projects are assigned to different agencies (which for the Solomon Islands. An initiative to has recently occurred in the Solomon Islands), it undertake a national transport plan which - 17 - resulted from an Integrated Transport Survey in rehabilitate infrastructure when it is let decay to 1987, did not proceed, although work has the point of failure. proceeded (under EEC assistance) in identifying and undertaking road rehabilitation priority 4.7 In all countries it is evident rehabilitation projects for the Solomon Islands since then. For and reconstruction of existing infrastructure is all PMCs there is a need for a coherent and taking place well before the normal replacement realistic transport policy and development period. A significant backlog of warranted strategy to be systematically established for their maintenance and rehabilitation has resulted from transport sectors. past underfunding of maintenance- notwithstanding the additional reconstruction needs as a result of the substantial damage to B. INVETMENT AND MAINENANCE facilities from natural disasters, such as Cyclones Ofa and Val in Western Samoa." 4.4 Capaclty. There is no evidence to suggest that the present coverage or designed 4.8 Further investment in new transport capacity of the major elements of transport infrastructure, in general, is not a priority for infrastructure inhibit economic development in supporting economic growth in any PMC. A any of the countries. In most cases (excluding priority requirement is the effective management for example, some roads in urban areas), the of existing transport systems, based upon a coverage and capacity of each nation's transport careful review of existing infrastructure and the infrastructure appears sufficient to meet transport identification of those assets that should be demands well through the 1990s, provided that maintained (and at what standard), together with the assets are adequately maintained. Existing priorities for the allocation of scarce demands on transport infrastructure are generally maintenance resources. well below design capacity levels.9 4.9 Maintenance. Road maintenance 4.5 Investment. Large levels of investment continues to be a key transport issue in all have been made in transport infrastructure in the PMCs, with the exception of Kiribati, where PMCs during the 1980s." The main objectives land transport is important but small in scale. have been to develop areas of export potential Warranted road maintenance funding for primary products and tourism, to provide requirements are generally in excess of linkages to markets and trade centers and to prevailing recurrent budget support.'2 Yet enhance social mobility. The results, in general, existing institutional limitations do not allow have been mixed; substantial infrastructure has present levels of maintenance funding to be been put in place, in many cases, but demands deployed in a fully effective manner. Moreover, for its use has been modest. warranted maintenance funding requires confirmation of the economic justification of 4.6 The attraction of aid funding for transport maintenance specific items of infrastructure. infrastructure development has not been supported by provision of recurrent funding for 4.10 Land access constraints (which are maintenance. The apparent willingness of shaped by the social and cultural values placed donors to fund rehabilitation projects has created on land ownership) are often a contributing a 'moral hazard' problem: external aid, when factor to the problems of road maintenance. perceived as infrastructure "insurance", weakens Discharge of water from the road reserve is attention to domestic financial and institutional often inhibited, resulting in ineffective drainage capacity to meet maintenance needs; the PMCs and premature pavement failure. The problem have found that they can rely on donors to is magnified by the use of roadside verges for planting by local communities, and by the failure - 18 - to control natural growth along the roadside. * associated management systems which With the exception of Kiribati, there appears to enable scheduling and programming of be appropriate legislation for the reservation of maintenance effort; land for road purposes. * equipment resources to perform the tasks; 4.11 Problems also exist with the maintenance of landings and wharves on outer islands and of * organizational resources and management some major ports. The maintenance of landings competency to deliver the services and wharves requires specific attention in the efficiently and effectively; Solomon Islands and Vanuatu and adequate maintenance funding needs to be provided for * the financial resources to meet the port infrastructure in all countries; in the program costs. Solomon Islands and Fiji port authorities are able to allocate funds for maintenance needs. None of these elements has been addrossed satisfactorily, to date, with the possible 4.12 There is a lack of maintenance of exception of equipment resources for which regional airstrips and of air navigation aids, adequate utilization and maintenance is a particularly in those countries which require problem. extensive domestic aviation networks to serve remote communities. Tonga, Western Samoa and Vanuatu have attracted substantial aid C. INsTrITmoNAL MISUES investment in international airport facilities and the Solomon Islands and Kiribati are seeking 4.14 Allocation of Functional donor support (concessional loans) for major Responsibilities. A number of countries have upgrading of their facilities. These investments devolved some functional responsibilities to are perceived by the countries as important provincial or local government. Given the small catalysts to economic development, in particular size and limited capacities of both central and for their tourist industry. Development of provincial governments, devolution is likely to airport infrastructure, with the exception of Fiji, be highly ineffective. On top of this, financial has been in advance of complementary support is typically lacking. For example, in investment in tourist facilities. Only in Fiji is Vanuatu and the Solomon Islands maintenance of tourism a major economic activity. In the other local rcads and outer island wharves and jetties PMCs growth in tourism has been at best is neglected. These may link vital services to modest. The capacity to fund the operations and local communities yet these communities may maintenance of airport facilities and support not be equipped to maintain them. systems, even with appropriate cost recovery policies in place, is a matter of concern and 4.15 Human resources remain a fundamental requires examination. constraint in both the nrivate and public sectors. Historically, aid donors have provided 4.13 Improved maintenance of the valuable considerable technical assistance to the public stock of transport infrastructure in the PMCs sector, by direct provision of professional and calls for the adoption by each PMC of an asset technical staff and through training and management strategy. The key elements of such development programs. the results have not met a strategy are: the objective of developing organizations capable of efficiently managing their assigned functions * an inventery data system from which it is using indigenous staff. The development and possible to forecast future cyclical retention of executive, professional, technical, maintenance needs and costs; and senior administrative staff remains a key - 19 - issue in all countries. Results from the requirement to meet certain objectives set by provision of expatriate staff in professional and legislation or policy, state authorities and technical positions, through technical assistance enterprises should be required to provide programs, have been variable, and at the best appropriate annual reports. temporary. Lack of trained counterpart staff, continuous turnover of personnel in key 4.18 Cost Recovery and User Charges for positions, and variability in the number and the use of transport infrastructure or government quality of the personnel supplied jeopardize skill transport services has been given little attention transfer and the development of cohesive and in the PMCs. Table 4.1 sets out areas in which motivated organizations. Consideration needs to cost recovery policies appear appropriate. There be given to examining past experience, is no evident cost recovery policy for road identifying lessons, and formulating better infrastructure or for services provided to the methods of providing institutional strengthening land transport industry in any PMC, though the and development of human resources. level of cost-recovery has been examined in Fiji. Land-based public transport services are 4.16 The matching of training with the provided by private enterprise with varying existing and projected skill requirements remains governmental regulatory control. largely unresolved. Kiribati has initiated a national manpower planning and training 4.19 Both the Solomon Islands Port Authority program; other countries do not appear to have (SIPA) and the Port Authority of Fiji (PAF) done so. Substantial opportunities are available operate their facilities commercially, although for training overseas and considerable numbers return on capital investment is low. A of PMCs nationals travel overseas for tertiary commercial, or cost recovery policy in setting training on specialist courses. The port charges is not evident for major ports appropriateness of the training, the extent to elsewhere. There appears to be no user charge which the skills are retained within the country systems in place for outlying jetties and or the public service, and the opportunity cost of wharves, with the exception of Western Samoa the present arrangements warrants a full review. which applies charges for the use of the inter- These matters are under review. (See World island ferry terminals. It is not evident that fees Bank, 1992.) Consideration also needs to be or charges for marine surveys or for the given to the manner in which assistance can be provision of navigation and safety aids have provided to the private sector to promote the been designed to meet specified cost recovery development of necessary managerial, objectives. Where government shipping services commercial, and technical skills. are provided the level of cost recovery ranges from below direct operating costs (Solomon 4.17 Government Accountability. Only in Islands) to meeting apparent full commercial Fiji, the Solomon Islands and Tonga, do obligations (Western Samoa). In other cases departments of government provide annual there is underprovision for replacement of reports setting out the performance of their capital items. functions and responsibilities. Thus, there is little formal accountability to, and of, 4.20 Landing fees and other charges are government. There is little public exposure of applied for major airports in all cn-%, and in policies and assessment of the effectiveness of some instances for use of domestic airstrips. efficiency of operations. The requirement for With the exception of Western Samoa and Fiji, statutory authorities to provide annual reports is where airport authorities operate all or some less clear, but in general this does not appear airports, organizations with commercial mandatory in practice. In view of the greater objectives have not been established. It is not autonomy of action, and in most cases, the evident that charges for air navigation and air - 20 - Table 4.1: PACIFIC ISLANDS-AREAS READILY AMENABLE TO COST RECOVERY FOR GOVERNMIENT INF1ASTRUCrURE AND SERVICE PROVISION Land Marine Aviation Infrastructure Roads Ports, Wharves Airports, Airstrips La Industry services Vehicle Inspection Vessel Surveys, Flight Services, Regulatory Port and Marine Air Navigation Systems /a Navigation Aids, Airport Aids la Services Government Public Transport Inter-lsland International Transport Systems Services Shipping Services Services, Domestic Services La FUlI cost recovery may not be feasible and/or efficient. safety services have been designed specifically to construction of transport projects and the recover the cost of service provision. Available operation of transport activity. Although evidence suggests that national airlines owned or environmental issues and their relationship to supported by governments are commercially transport are outside the scope of this report, oriented on international services, but that fare there are several areas for which emerging constraints applied to government operated concerns with this relationship should be noted. domestic air services result in operating losses. For example, causeway construction and lagoon water quality in Kiribati; logging road 4.21 Community Service Obligations. A construction, run-off, and the ecology of virgin fundamental problem in all PMCs is the areas in Solomon Islands; and shipping provision of transport services (normally movements, maritime safety, and coral reef shipping and air) to outer islands where markets protection in all PMCs. are thin and services are not commercially viable. A minimum standard of service needs to 4.23 Management of the natural environment be assessed and cost-effective arrangements is a general issue for the PMC Governments. established. In general, it will be desirable for The quality of their environments is important to the government to remove itself from direct the culture of these nations, as well as a major provision of such services and to facilitate factor influencing the demand for tourism. private sector provision on a contractural basis. Accordingly, it is important and timely for PMC Goverunents themselves to have available the basic capacity to appreciate, scrutinize and D. ENVRoNMENTAL IMPaCrs manage environmental impacts that might be associated with all projects, including those 4.22 The natural environments of the PMCs within the transport sector. Procedures need to are quite fragile and therefore vulnerable to be in place to ensure that new transport projects various forms of intervention, including the (including reconstruction) are designed taking -21 - into account potential environmental impacts; * coastal zone management and protection such impacts should be subject to local issues, especially in relation to assessment and reflect public participation. This conservation of mangroves, coral is an important area and warrants appropriate communities, and beach erosion. Government attention. However, it does not require, nor should it be used as a basis for, 4.25 PMC Governments have prepared significant additions to administrative resources. national reports (with assistance from UNDP and ADB) for the United Nations Conference on the 4.24 Specific areas of environmental impact in Environment and Development (June 1992). the PMCs that may result from projects in the These reports contain only minor Jiscussions of transport sector, and warrant special treatment in the environmental impact of transport activity. impact assessment and monitoring include: In addition, under the South Pacific Regional Environmental Program (SPREP) with UNDP * unsustainable, induced development in and ADB funding, the PMCs are preparing sensitive environmental areas such as environment management plans. The Bank's forests and coastal zones; upcoming Regional Economic Report for the Pacific Islands will also cover environmental * water quality issues relating to soil issues, based upon a synthesis of existing erosion and surface and near-shore water analyses. These efforts should serve to satisfy quality, and tidal interchange of water in IDA requirements for each PMC to prepare an lagoons; Environmental Action Plan (EAP) by June 30, 1993. Specific areas of potential transport * water use conflicts, particularly conflicts impacts should be addressed in the final EAPs. that affect in-stream and subsistence uses of surface and coastal waters; - 22 - Endnotes 1. Preliminary information on development needs in the transport sector was developed in a desk study of the transport sector in each PMC (Wiorld Bank, 1989). A survey mission to each country was undertaken early in 1991. The mission team members were Colin Gannon (Senior Economist and mission leader), David Bray (consultant-Fiji and Tonga) and Ian Gordon (consultant-Kiribati, Vanuatu, Solomon Islands and Wcstern Samoa). Detailed surveys of the transport sector in each country are presented in Volume Two of this report. 2. The Bank Country Study, World Bank (199 la), provides a detailed review of the economy of each PMC. 3. The level of transport costs for the region, in comparison to other similar regions, is examined in Part HI of this study. 4. Regional international shipping costs are discussed in Part III of this report. 5. Aspects of interrelationships between transport and tourism are discussed further in Part IV of this study. 6. In some PMCs the allocation of responsibility between central and local government is contentious. This is particularly the case for the maintenance of local roads, wharves and airfields. (See further comment below.) 7. Maintenance problems and their effects are discussed in Part II of this report. 8. World Bank (1991) presents a more detailed review of public sector finances in the PMCs. 9. Designed standards, and the associated designed quality of service for users, do not prevail due to undermaintenance. 10. Individual country transport sector surveys are presented in Volume Two of this report. 11. An assessment of damage to economic infrastructure and rehabilitation needs in Western Samoa, following Cyclone Val in December 1991, is being undertaken. 12. See Part 11 of this report, below. - 23 - PART II Maintenance Management of Transport Infrastructure - 25 - CEAPTER 1 INTRODUCTION A. CONTEXT B. THE MAENANCE ISSUE 1.1 Part II of this study has as its point of Sustainable Infrastructure and Warranted departure the series of PMC country by country Maintenance transport sector surveys.' The accumulation of a large stock of transport infrastructure in the 1.3 Proper assessment of the adequacy of PMCs, relative to their economic and existing maintenance requires determination of institutional capacity, is identified as a common the economic merit of preserving, and at what and serious problem; its dimensions are not well standard, individual pieces of lnown and its consequences do not appear to be infrastructure-sections of road, particular appreciated.2 Accordingly, infrastructure--its wharves, and specific airfields. Maintenance is sustainability and its maintenance-is examined warranted only if each separate piece of below for five of the Pacific Island States infrastructure, to which it is applied, is a namely, Kiribati, Solomon Islands, Tonga, valuable asset, i.e., the item of infrastructure Vanuatu and Western Samoa.3 generates economic benefits which exceed the costs of its long-term preservation and use. 1.2 The objective of Part II is to outline a Such assessments require an extensive set of strategy for management of infrastrucure and its micro level "asset appraisals". maintenance, approprtiate to the region.4 The approach adopted to meet this objective involves 1.4 It is possible, however, to gauge the four steps: sustainability of the stock of transport infrastructure at a macro level. Estimates of the (a) estimation of the stock of existing physical scale and replacement plus maintenance transport infrastructure, costs, in aggregate, can be examined relative to a country's national income and geographical (b) assessment of the sustainability of the extent, and these relativities can be compared existing stock, across similar countries. For all PMCs the existing stock of transport infrastructure is (c) documentation of the current status of extensive and (as shown below) non-sustainable. maintenance and its consequences, and 1.5 This situation has arisen primarily from (d) provision of a frameworkfor institutional inadequate sector investment planning and development of infrastructure project appraisal, inadequate attention to future management. maintenance obligations, and large external grant assistance. As a consequence, current maintenance decisions require identification of infrastructure for which maintenance is - 26 - warranted. This is beyond the analysis of 1.9 Prior to project decisions recurent maintenance presented here which is directed at maintenance cost requirements are not given estimating the stock of existing infrastructure, sufficient attention in feasibility studies. Once a the maintenance expenditure that would be project is completed, the effects of under- required to keep it in good condition, and maintenance of transport assets are often not therefore, the sustainability of the existing stock. transparent in the short term and the On the basis of this analysis, a strategy to intermediate effects, which give rise to address the overall manage-nent of infrastructure increasing transport user costs as the system and its maintenance can be advanced. decays, are often ignored or underestimated in setting budget priorities. In the longer term full Inadequate Maintenance and Its consequences have sometimes been masked by Consequences external assistance for replacement. 1.6 Maintenanceexpenditureandmaintenance 1.10 In general, in the PMCs, maintenance of effectiveness are inadequate in the PMCs. There existing assets is the key issue, not new are several causes: the accumulated stock is investment. At present, policies of PMC very large relative to their size, recurrent budget Governments do not appear to reflect this resources are user charges are low, delivery circumstance nor a high priority for a systematic effectiveness is weak, and donors finance approach to assess maintenance needs to assist rehabilitation and reconstruction.5 recurrent budget planning. Nevertheless, the outcome is inadequate maintenance-even for warranted infrastructure. Influence of External Assistance 1.7 Adequacy of maintenance for transport 1.11 Since infrastructure priorities involve the infrastructure in developing (and indeed preservation of warranted existing infrastructure, developed) countries is not a new issue. donor assistance needs to be redirected to this Numerous studies by multi-lateral and bi-lateral task. At the same time, the primary assistance agencies have set out the global nature responsibility for looking after assets rests with of the maintenance problem, and provided each PMC. Therefore, donor assistance for specific analyses through sectoral or subsectoral maintenance needs to be combined (for example, transport studies in various countries (see, for on a matching basis) with increased mobilization example, World Bank, 1989 and 1990). of domestic resources for maintenance, for example, through increased user charges. 1.8 Deterioration of infrastructure can result in substantial adverse consequences to a nation's economic performance. Poor maintenance not C. INFRASTUCTuRE MANAGEMENT only increases infrastructure costs it also FRAmwoR imposes other costs and reduces economic efficiency. For example, inadequate 1.12 An overall strategy for infrastructure maintenance of infrastructure leads to higher maintenance, covering management, financing vehicle operating costs, reduced productivity of and delivery does not exist. 7he entire stock of port operations, and results in foregone net existing infrastructure is not sustainable without benefits from airfield closures. These additional rationalization; priorities need to be established costs are avoidable and reduce economic and some infrastructure downgraded or performance. Although quantification of the even.ually withdrawn. effects of under-maintenance of transport infrastructure is well developed for roads,6 it is less systematically advanced for other modes. -27 - Towards a Strate for Man _ u or Mangement Information for Effeeye Infrstructure and Its Mainteance Maintenance 1.13 A basic strategy for improved 1.16 Management of infrastructure, including management of existing transport assets in the setting priorities, and programming activities PMCs requires: requires basic information.' The level of information appropriate to the maintenance * asset inventory information and management task in the PMCs is rudimentary, monitoring capability; and its cost is low. The existing level of maintenance information may be depicted by * identification of the set of existing assets drawing upon guidelines developed recently by which warrant preservation, and at what the Bank (World Bank, 1990). The guidelines stadard; provide for a m geent informatdon system which might be progresivel imy pleme to * information procedures and capity to enable effectivo reporting on the status of the forocast warranted cyclical mantenance system and the basic planning, programming and and replacement costs, taking into account rowurcing of routine and periodic maintnance alternative sandards; nd of asset rehabilitation. Although the guidines are directed to the magement of a * scheduling and programming of road system, the principles are equally maintenance activities; applicable to the marine and aviation subsectors. 3 organizational resources, management 1.17 To manage infrastructure, the starting capacity and technical skills to ensure point is knowledge of the stock. Asset maintenance is delivered efficiendy and in inventories that do exist are largely the result of a cost-effective manner, including use of project studies; the inventories are based on a the private sector; and specific need and are neither comprehensive nor uniform and they have become increasingly out 3 financial resources, including cost- of date. Thus, information at any level typically recovery from users and donor assistance, requires considerable effort in putting together to meet program costs. fragments of data and/or estimation. 1.14 Management decisions on these areas 1.18 A consolidated assessment of the affect total transport costs and impact on the prevailing status of maintenance information for performance of a country's economy (see each country is summarized in Table 1.1. This Figure 1.1). To date, the basis for such table reveals the lack of data on transport decisions has been highly inadequate; often infrastructure. No Ministry or department holds decision emerge by default. an adequate inventory of assets at the most aggregated level (with the exception of the 1.15 In focusing on a strategy directed at Solomon Islands Port Authority) and there is no efficiency in the maintenance of infrastructure detailed information which would serva there are three main components: management, management needs at the network, project or financing and delivery. These components are operational levels. examined in Chapter 3, below.7 1.19 An outline of the management information, classification, and depth suitable to the PMCs is preented in Annex 1. - 28 - Eigure 1.1: INFRASTRUCTURE MAINTENANCE MANAGEMENT Economy Physical Environment ll I ~~ ~~1 V , ,=,, Transport Services Service Requiremeante lInventory Transport Status w l ~~~~~of Asset of Capital l Assets Register Assets Vehiclo Cost of - Operating Infrastructure ! ! ; Costs Maintenance w Needs | . >| Life Cycl Infrastructure Total t | | * |Llf Cyle nfrstrctue || Tft Cost to Users Maintennance Costs ! of Infrastructure Total Transport System Cost !77Hi1 Key Management Decisions - 29 - Table 1.1: Pacdfic bland Countries-Summary Management Information Inventory Roads Ports Aviation Network & Network & Network & Sectoral Operations Sectoral Operations Sectoral Operations Kiribati U U P U P U Solomon Islands U U ALa ALa P U Tonga P U P U P U Vanuatu P U U U P U Western Samoa P U P U A U La Excludes local wharves and jetties outside SIPA control for which no information is available. A - Acceptable basic information available. P - Partial basic information available. U - Information unavailable. Source: Mission estimates. *30 - CHAPTER 2 NATURE OF THE MAINTENANCE ISSUE 2.1 The presentation in this Chapter covers determination of core information needs for five principal tasks: maintenance management. (a) to establish an inventory of transport 2.3 Government-Owned Facilities. infrastructure in each of the five PMCs; Consideration in this report is limited to transport infrastructure in the public sector; this (o) to estimate a replacement cost valuation represents the overwhelming majority of for each inventory; facilities in the PMCs. Privately-owned fixed transport infrastructure is limited to a few (c) to gauge the sustainability of the existing private wharves and jetties in the Solomon stocks of infrastructure and to assess Islands and Vanuatu, and some private airfields regional patterns; in the Solomon Islands. Of course, from the standpoint of the provision of transport services, (d) to assess the maintenance levels that the private/public ownership distinction is an would be associated with keeping the artificial one. entire stock of existing infrastructure in good condition, (if setting this was The Existing Stock of Transport warranted); Infrastructure (e) to compare assessed maintenance and 2.4 A physical inventory of the existing stock actual maintenance expenditure; of transport infrastructure in the five PMCs is presented in Table 2.1 The most extensive data (f) to set out the implications of inadequate are for roads. Sealed roads are less than maintenance; and 3 percent of the total length of roads, except in Western Samoa where they account for (g) to identify revenue from user charges and 14 percent. Earth roads represent a majority of the existing levels of cost recovery by total road length in Western Samoa, Kiribati and mode. Tonga, reflecting the ability to construct reasonable graded roads in the primarily coral/sand soils in these countries. By contrast, A. TRANSPORT InRAsTRuCnlRE in the Solomon Islands and Vanuatu, where the INVENTORY more difficult terrain and loam/clay soils make base grading difficult, a much smaller share of 2.2 Compilation of a basic inventory of roads are earth formed. The major role of inter- transport infrastructure is the essential basis for island water and air transport in the Solomon addressing all of these tasks; it also allows Islands and Vanuatu is also shown by the large number of formal jetties and airfields in each of - 31 - Table 2.1: PAcIFc ISLANDs-TRNSPORT INFRASTRUCrURE INVENTORY, 1991 Solomon Wstern Total (or Kiribati Islands Tonga Vanuatu Samoa Average) Road length (km) Sealed 36 101 109 lS 300 661 Unsealed engineered 289 700 492 926 522 2,929 Earth 496 S00 1,273 719 1,250 4,238 1282 "/^ 1.301 18 1. 2Ji72 2t2A Shar of Road Leogth (9) Sealed 4 8 6 7 14 (8) Uneled engineered 35 S4 26 53 25 (37) Earth 60 38 68 41 60 (54) Did ~ ~ ~ ~ ~~0 iQ 100 l 100 100 (100 Madi Major ports 1 2 1 2 1 7 Major jetties 1 30 4 16 2 53 1211 2 2 _6 Aviation Airports with sealed nways 2 1 1 2 1 7 Other airfields 14 21 5 26 2 68 -Ta 1 22 6 28 a 75 L& Excludes road length on Kiritimati; the relatively high non-sealed road lengths as reported for Kiribati warrant further confirmation. Sour: Mission estimates. these countries. Among the other countries, it 2.5 The physical coverage of the existing is likely that there are more jetties of infrastructure in each PMC can be gauged significance (particularly in Tonga) however, broadly for the road subsector, since roads are there is no formal documentation. The length of described by type and length. For aviation and road in proportion to land area, population and marine facilities, aggregate physical capacity is GNP is greatest in Tonga, and least, generally less amenable to assessment; for these modes, by a considerable margir., in the Solomon total replacement cos: is a more suitable Islands.' The substantial stock of road aggregate measure-see below). In relation to infrastructuie in most PMCs stems in part from roads, most PMCs have reasonably substantial the provision of accessibility to a population physical networks, especially on their main which is widely distributed and low in density. islands. Table 2.2 presents figures on road Tablt 2.2: PACIFIC ISLANDS-INTERtNATIONAL COMPAJLSON4 OF RoAD SuBSHcToRi INDIcAToRs /a Pp-Land Population GNPI ODA/ Poad ReA~ g oad RoadRplc-R rd latnArea (density Capita caita CaLpitaiiA C GWF4 iieUzmwd Relcmn ntcVle (kin') (eP1eIki~ (USS) (US)M m) (m/0 r Qnl00(kmi vehicle VIeG-NP* Reg. Vehicle GNP4 (um') people) us$'0) (kavb (5) (US$iOOOiveh)(Vdh/USSnm) /b /b /C~~ Li Lf h/ South Pacific (thoumnds) Fji 32 11,270 40 1,540 74 4,994 273 11.5 4.4 0.13 Kirbt 67 710 100 650 243 721 1,156 12.3 17.9 0.79 ii 17:9 23 P44m ew Ginea3,800 463,000 1 700 17 23,14 40 5.5 7.7 0.55 SIoWun Islands 304 27,990 12 430 192 1.301 46 3.9 1.6 0.36 30 12.6 24 TOnge 101 720 142 g00 116 1,174 2,603 11.3 20.0 0.40 69 13.9 50 Vamjatu 151 12,190 13 120 260 1,760 144 10.7 12.3 0.31 56 17.3 32 Wester Samoa 168 2,530 60 600 112 2,072 732 12.2 17.1 0.33 91 16.9 54 Caribbean, Americas and Europe Barbados 256 430 691 4 340 .. 1,670 3,114 6.4 1.5 .- Belize 114 22,100 a 1,150 137 1,9810 17 12.5 10.9 .- Costa Rica 2,500 51,000 49 1,190 11 2,150 559 11.4 9.6. c ~~~~695 9,000 76 3,590 .. 6,130 759 10.2 2S . 2mitius ~5,400 21,000 193 320 35 4,000 143 0.7 2.3 .- Jamaica 2,200 11,000 200 1,150 70 17,700 1,609 1.0 7.0 . Afiica and Inian Ocean Comoros 451 2,3 200 340 116 950 425 2.2 6.6... . Co.oV 1,10 342,00 6 1,140 75 10,940 32 6.1 5.4 ..... Mamritus 1,100 1,850 500 1,490 57 2,59 1,391 2.5 2.3. .. . Svsazilnd 761 17,000 41 100 53 2,820 166 4.0 5.0 .. Regiom (millions) Eantem and Southem Africa 201 is 310 ..53 2.9 11.5 ..35 Wester Africa (excl. Nigeria) 90 10 270 . 41 3.6 9.4 . 33 Fast Asiasandthe Pacific 1,350 102 710 ..1,127 1.1 2.6 Is 1 South Asia 1,001 197 220 ..328 1.? 6.5 ..13 Enuroe, Middle Eas & W.Afuica 232 36 1,795 ..162 4.5 3.1 ..20 LAti Anmac and the Caribean 350 19 1,190 ..117 6.3 3.1 . 13 TOWa (mici. Nigeri) 3,327 52 I...11 2.3 3.9 ..16 Some data relate to differest years and hence nominal conyrisoas need to be iaiempreted with cmauion. 1911 population sad GNP for South Pacific countris except PNG. Population mad unadjustd GNP in 1914 fiw FNG and other couies, and median GNP/capita in 1984 for Li = 1911 as data available. Ad kod in 1991 for South Pacific countries except PNG; 1914 for PNG and countries in othe regions. rep~j~o in 1991 for South Pacific countres except PNG; 1914 for PNG and coujiries in other SGN inI nltdby 5 S p.a. to indicative 1991 prices for South Pac tic countries except FNG; 1914 da¶CFi4G and for couggies in other region.. g.Re *jfe vehicles for laefts year available. 1991 for the South Pacific countries excep PNG; 1914 for PNG and other countries and regioma. Replaceimen vabantious are eambinked in TAbl 2.3 below. Sousces: Word Dak (1991a), (1919, (191), andmission estimae. - 33 - densitv t;r the tive PMCs, and provides a costs to value of international trade flows (on coniparison otf these tigures with other countries, average 2.9 percent) indicates that full recovery in the region, as well as some other selected of port costs may be possible. The main port countries. In general, the density of roads by accounts for the lion's share of the total marine various measures (land area, population, subsector replacement cost in all PMCs, except vehicles, and GNP) is high in the PMCs, Vanuatu. Port performance measures need to be relative to the other countries. This suggests established to examine cost recovery for these that the road infrastructure in the PMCs is main facilities and to confirm their sustainability. excessive relative to aggregate demand for road use, and the capacity of the economy to sustain 2.9 Aviation activity also varies considerably the road stock. This matter is examined in among the PMCs." The replacement cost of greater depth below for all transport modes, aviation infrastructure at the main international based on estimates of the replacement cost value airport of each PMC is generally high in of the infrastructure. proportion to international passenger movements, though expected growth in Valuation of Existing Infrastructure international passenger movements (particularly in Vanuatu), and the high proportion of travel 2.6 Infrastructure replacement cost valuations between Western and American Samoa will are summarized in Table 2.3. The capital stock affect the overall comparative costs. values reveal that transport infrastructure in the Sustainability of all existing aviation PMCs is not only high in broad physical terms, infrastructure is problematic; it requires detailed it is also inordinately high relative to the assessment of the prospects for cost recovery capacity of each economy to sustain it. On a and the economic warrant of major airport uniform annual basis, the capital replacement facilities. cost value of transport infrastructure is estimated to represent 10.0 percent of GNP, on average, 2.10 The inordinately high levels of transport across all five PMCs. For individual PMCs, infrastructure in the PMCs strongly indicate that this share of GNP varies from a high of the replacement of many existing infrastructure 13.1 percent in Western Samoa, to a low of 5.3 facilities cannot be warranted-and, in addition, percent in Solomon Islands. that the maintenance of such facilities may not be justified (at least at existing standards). 2.7 The stock value of transport Large flows of external assistance to the PMCs, infrastructure is also very high relative to other relative to their domestic economies (as indicated countries. For example, Table 2.3 shows that in by the levels of ODA per capita shown in ,he road subsector, where some comparable data Table 2.2) have resulted in an accumulated stock are available, road density, in terms of its of transport infrastructure which is non- replacement cost value to GNP, is of the order sustainable. As this entire stock matures (and of two to three times higher in the PMCs perhaps ever increases), full replacement relative to the other countries. through overseas aid can also be expected to become non-viable. Unwarranted existing 2.8 For the marine and aviation subsectors, facilities can cause further inefficiencies: Table 2.4 provides indicators of infrastructure replacement values relative to use. International (a) in attracting unjustified scarce trade is a high proportion of GNP in the PMCs maintenance resources, and most of this trade is by sea. The replacement cost of marine assets, in proportion (b) in influencing demand, notably affecting to total international trade, varies across the the best choice of infrastructure by users, PMCs.'° The ratio of marine annual capital and - 34 - Table 2.3: PACIFIC ISLANDS-INDAS NRUCTURE REPLACEMEa VALUE, 1991 La Solomon Western Total (or Kiribati Islands Taup Vanuetu Samoa Averge) Replacement Stock Value (USS million) Roads 18.7 45.7 64.9 79.8 106.7 315.9 Marine 9.3 23.8 27.6 38.0 27.2 125.9 Aviation 3.0 14.0 30.8 55.7 32.2 136.7 Share of Stock Value (%) Roads 60 55 53 46 64 55 Marine 30 28 22 22 16 22 Aviation 10 17 25 33 19 24 lau 1QQ00 100 1. 1.00 Equivalent Annual Replacement Value (USS million) lb Roads 1.5 4.6 6.4 8.0 10.1 30.7 Marine 0.7 2.0 2.4 2.9 2.1 10.0 Aviation 0.5 1.4 3.0 S.9 3.0 13.8 IsQ 2U u ILl Ii! IS.2 S4. Annual Replacement Value (as % of GNP)k Roads 3.2 3.1 6.9 S.2 8.7 5.7 Marine 1.5 1.3 2.6 2.1 1.8 1.8 Aviation 1.1 0.9 3.2 4.2 2.6 2.5 IPAI s. La 12: .1= lid 1 Stock Value per Capita (USS) 27963 Roads 150IS 643 528 635 399 Marine 139 78 273 252 162 1S9 Aviation 45 46 305 375 192 173 _W4 27J L I IJ 2 2u Unit Replacement Cost for Roads (USS'000/km) 22.8 35.1 34.6 45.3 51.5 40.4 Principal Facility as % of Total Replacement Value /o Marne 92 71 78 45 81 69 Aviation 49 69 87 42 9s 67 Population ('000) /t 67 304 101 151 168 791 GNPUSS million) 46.3 150.5 92.6 138.9 115.7 544.0 GNPpor capita (US) 690 49S 917 920 689 687 /a 1991 prices unless induated. Equivalent annual replacement cost values computed at 7 pent discount rate over economic lives a lic ble to each type of infrastructure. (See Country Maintenanc Annoxeo in Volume Two of this report). -mes values apply for maintaining the assets in good condition (i.e., with _ssed levels of maintence, see Table 2.7 bedoW%8 lGNP o, inflated by 5% per annum to indicative 1991 pric. nin 1988 increased to 1991 at the ave g t rte for the peiod 1980-88. placement value of single largest facility (the intternatona airport ad port) as a share of the total red acement value of infastructure in the mod. If Allowce s been made for approximate value of unsad rod/tacks ao Kiritimti. Sources: World Bank (1991a) and mission edsima. - 35 - Table 2.4: PACIFIC ISLANDS-MARINE AND AVIATION INDICATORS, 1988/1991 Solomon Western Total or Kiribati Islands Tonga Vanuatu Samoa Average International trade (US$ millicn) L 27.5 186.2 56.5 66.2 88.4 424.8 Share of GNP (%) 59 137 68 59 89 66 Value/Capita (USS) 351 590 548 485 534 533 Marine: Annual Capital Cost of Marine Assets (USS million) 0.74 1.93 2.24 3.08 2.20 12.39 % of Le value of international trade /b 2.70 1.00 4.00 4.70 2.50 2.90 Aviation: International Air Passenger Movements ('000) Li 27 30 67 40 186 66 Movements per USS million of GNP La 147 233 721 290 1,597 613 Annual Replacement Value of Principal Airport (USS per international passenger movement) L. 35 27 39 57 16 35 Li Estimates for 1988/59; merchandize imports and exports only. Lb Replacement value of marine assets, on an equivalent annual capital cost basis (7 percent discount rate over 30 years; capital recovery factor 0.081) divided by total merchandize imports and exports in 1988 in USS inflated by 5 percent per annum to indicative 1991 prices. /c 35 percent of visitor arrivals in Western Samoa are from American Samoa. Visitor arrivals account for half of international passenger raovements through Western Samoa. Ld For 1988. /e Replacement value in 1991 prices. International passenger movements (visitors and nationals) estimated for 1988 (see tables in Country Surveys). Sources: World Bank (1991a) and mission estimnates. resources can only be made after the existing (c) in distorting investment responses of stock of infrastructure is rationalized, i.e., agents in other sectors which use facilities which are warranted and justify transport services. maintenance are clearly identified. Therefore, high priority needs to be given to the Characteristics of the Existing Stock of reassessment of all existing substantial Infrastructure infrastructure and to the identification of those facilities that are no longer warranted (including 2.11 Since substantial transport infrastructure those for which modification, such as reduction may not be warranted, interpretation of the in standard, is also not warranted). Proper patterns of investment across the PMCs needs to allocation of, and priorities for, maintenance be made with caution. Notwithstanding this, - 36 - estimated patterns are largely as expected. 2.14 Some past studies have identified While aviation and marine transport are intrinsic rehabilitation needs for selected infrastructure in components of the transport task in the Pacific terms of so called "maintenance overhang", i.e., Island States, roads account for the majority of the rehabilitation requirements resulting from the value of transport assets in each PMC (from past inadequate maintenance (brsed on original almost two-thirds in Western Samoa to a little design standards) to restore warranted assets to under one-half in Vanuatu). The spatial "good" condition. "Working" estimates of distributionof population shapes modal shares of "overhang" has been assembled based on these infrastructure. For example, the concentration studies. " Estimates of the maintenance of activity on the two main islands in Western overhang are shown in Table 2.5.'5 The Samoa is reflected in the high share in roads, estimated overhang is greatest in the road sector and low share in marine assets.'2 The recent where it is equal to 40 percent of the completion of major airport terminal and runway replacement cost of road infrastructure in the facilities in Vanuatu and Tonga account for the Solomon Islands, and about a quarter of the very large proportion of the value of aviat.on replacement cost in Vanuatu and Western infrastructure in these countries. Samoa. The identified overhang is less in the marine and aviation sectors. This reflects, in 2.12 The large investment in ports, jetties and part, the more heavily engineered nature of airports in Vanuatu and Tonga in the 1980s is assets in these subsectors and the greater reflected in the very high value of infrastructure pressure, for economic and technical reasons, to per capita, and as a share of GNP, in these ensure major assets in these subsectors, meet countries. Corresponding to this is the low international operational standards (for example, imputed average economic productivity of the those set in aviation by ICAO). There is also infrastructure. On a per capita basis, the less information available in the marine and polynesian countries of Western Samoa and aviation subsectors to establish the extent of the Tonga, and to some extent Vanuatu, have maintenance overhang. As stressed, not all established a larger stock of transport existing infrastructure may warrant infrastructure, relative to GNP, than Solomon rehabilitation. Islands and Kiribati (see Figure 2. 1).'3 Comparable figures covering transport Overall Assessment of the Existing Stock of infrastructure valuations are not available for Transport Infrastructure other countries. 2.15 The stocks and replacement costs of 2.13 The replacement value of infrastructure transport infrastructure in the five PMCs gives no indication of the current condition examined are inordinately high. There is a status of existing facilities. Except for recently critical need to assess the extent to which the completed works, the current "value" of PMCs can sustain this infrastructure from facilities will be considerably lower than the domestically generated economic activity. There replacement cost due to deterioration with age is a need to determine a strategy to ensure and lack of maintenance. In Vanuatu, efficient maintenance for warranted 20 percent of engineered roads were rated as infrastructure. Elements of the strategy should being in good condition in 1989, with about include: equal proportions of the remainder rated as fair or poor. This is the only available systematic * Assessment of the total quantity of information on the physical condition of existing infrastructure which can be transport infrastructure in the five PMCs. sustained, and identification of those - 37 - Ilau...L.' VACIInC ZSLAND-ITVRASTRUCT3 VALYK TS Off, 1991 1200 - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ v ir tu 1000- Western 'a~~~~~~~~ao Soo 600 A 6 0 400 _ Z t~~~~~~~iribatli Solos on Islands 200 0 00400 Soo 600 700 Soo GNlP per capita, US$ - 38 - Table 2.5: PACIFIC ISLANDS-INFRASTRUCTURE MAINrENANCE OVERHANG, 1991 La Solomon Western Total or Kiribati Islands Tonga Vanuatu Samoa Average Total Value Roads (US$ million) 3.1 18.4 7.6 19.1 24.2 72.5 (US$ per km) 3,775 14,153 4,055 10,852 11,680 9,262 M ine (USS million) 1.8 n.i/b 0.2 3.5 n.i 5.6 Aviation (USS million) 1.2 1.3 n.i 0.1 n.i 2.6 Total 6 1197 7 8 24.2 Q, As % of Replacement Value Roads 16 40 12 24 23 23 Marine 20 n.i 1 9 n.i - Aviation 24 9 n.i 0 n.i Ttal 18 X _ 15 X La Approximate cost (in 1991 prices) of rehabilitating existing infrastructure to good condition, from which it can be treated with regular routine and periodic maintenance. Lb n.i indicates no maintenance overhang identified or indicated; for example in Western Samoa, Apia Port is being redeveloped, and cyclone damage eclipses overhang at other ports. Source: Mission estimates. infrastructure facilities which are With the exception of the identification of warranted, together with their appropriate warranted facilities, these elements are analyzed, standards. in turn, below. * Determination of the optimal level of The Meaning of Assessed Maintenance maintenance for warranted infrastructure which minimizes capital, maintenance, 2.16 Reassessment of existing infrastructure and user costs (on an equivalent annual and appropriate standards are outside the scope basis). of this study.. Therefore, the examination of maintenance presented below, is based * Targets for cost recovery and appropriate necessarily on the identified entire stock of user charges. transport infrastructure in each PMC. The maintenance that would be associated with * Institutional capacity to support effective keeping these existing stocks in good condition, maintenance. is estimated. This is termed 'assessed - 39 - maintenance". Obviously "assessed Assessed Maintenance maintenance" does not represent warranted or required maintenance. The purpose of 2.18 Optimal maintenance expenditure is that estimating assessed maintenance levels is level which, if effectively applied, would result in the lowest total transport system costs. Total (i) to probe further the non- system costs are the sum (on an equivalent sustainability of existing annual basis) of the life-cycle costs 16 for infrastructure, at its present size, infrastructure (sustained in relation to an in terms of its implied "assessed" established standard, over its potential economic maintenance obligations, and life) plus user operating resource costs. Total system costs are set out conceptually in Box 2.1. (ii) to use the existing stock as a These relationships are not well understood for working basis to examine actual conditions in the PMCs and therefore "optimal maintenance expenditures in the maintenance" (for warranted assets) cannot be PMCs, estimate the implications determined. The approach adopted herein is for total transport system costs of pragmatic; that level of maintenance which 'inadequate" maintenance, i.e., would keep infrastructure in good condition shortfalls between actual and (assessed maintenance) is estimated for the assessed maintenance, and assess existing stock. The reductions in total transport cost recovery for infrastructure in system costs that could be achieved by the sector. increasing maintenance from actual to assessed levels are estimated. B. MAINTENANcE EXPENDImuRE 2.19 The annual costs of maintenance are analyzed below for each mode. An annual Actual Maintenance and Accounting Systems expenditure of assessed maintenance for roads is estimated using an average annual cost per 2.17 Ministries and departments in the PMCs kilometer for each of the three categories of road follow usual government accounting practices covered in the road inventories. The assessed which identify expenditure by type (for example, annual costs of maintenance for marine and labor and materials) and not by function (for aviation facilities are estimated on a cruder example, construction and maintenance). basis. "Best practice" maintenance costs, in the Therefore, it is not possible to determine directly absence of estimates of the resources required to actual expenditure on maintenance (or whether undertake routine and periodic maintenance, are it has been provided at minimum cost for the estimated by applying a "rule of thumb" prescribed standard). Rather, estimates are percentage factor to the estimated replacement made by taking the proportions of inputs such as cost of the asset. This maintenance factor labor and materials assigned to maintenance, as includes periodic maintenance as an average against construction and rehabilitation. In no equivalent annual value; this spreads the case is it possible to establish a maintenance "lumpy" nature of maintenance over the history for specific facilities which, together economic life of the infrastructure.'7 with condition assessment, would assist estimates of future maintenance needs and evaluation of 2.20 Acceptance of Existing Assets. The additional standards. assessed maintenance expenditure estimates developed below are those associated with all existing infrastructure facilities and their associated design standards; it is not implied that assessed maintenance is economically warranted .au3;pu.dzz eamu"u m Tu uS 3800 1831d10f . . *aWU@w^ . I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ . .... '..:. %*0: * o$" w , .. '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. . . . . *tU>=l@ wFI~~~~~~~~~~.... ...° . .... .' ....... ... . .. ... ... . t~~~~ ~ ~~~~ ~ ~~~~~~~~~~~~~~~~~~~~ .S ............ 3800~~~~~~~~~~<' ''"' '\ ' '' ''''"W0 ler~~~~~~~~~~~. . w° .. . ..' . .. . .N .... . ' ...',,''.''.'.' , \ ,,,~~~~~~~~~~~~. ,' '''''...\ ,', .,''.i......'f ' ' ' ' ' ' ' ', , '" " '' ''' ":, "'': ......... ' . ' ' ' '' ' '.'' ,, ' ., ''''' '. ' ::::' '' .': . ': : :........... :: . ... ;tU^t: ... tY . ..Slt:::>::::*:t~~~~~~~~~~~~~~~~~~*::::2m:.:.m:::wA::::::: :: ::: :::::::: :: :::::::: :::::::::..... .... . ............ : . . : : w :. §aaet::7w:::*t"::^e:::*Ry,.. ....... ...... ... ............................ ......... : : ::. ... ..:::. ... .. ......... ::..:: ,,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. ., :'. ''.''. .... . '..,... .. ' '. .. . .. . .. . " ' ............ .........'' "'' '' '' 1Z~U2VP :: ~O : : :-: --- : : .:: . :. .: .:,.:.:: - - ::.: ::.:.._::.:. .. : :. :: :: ::: ::: :.:::::. :: :::~~~~~~~~.. .... ................ --. ... .. ,..:,z,n,x,,,..~ ~ ........ .l .. ... . ...... I... ......... . .. . ..... w n ':... ...... . .. ... .......... ''~~~~~~~~~~~... : . , ::., ,f' ' ... . .... ... .......... ......... ....v...... . .... :. w"yP''''' . .... . h .!I p . im-k.r: .. : :weu . ,, ,,,li t,ql ,l5~~~~~~~~~~~~~.. . . .. .. e::,, ...... . ..! .... ........... . .................. . ............. e,.|, p.a "4 ! t ~~~~~~~~~~~~~~~~~~~~~~~~~.X. M.3 .2", -ttm :::V"a IVMA. -~~~~~~~~~~~~10 .." ... ... .. .............. .... . . . ..."" .. ~~~~~~~~~~~~~~~... . . . . .. . . . . . ...... ...... ........ ............ .... ... .............. . ...... ........ . ...... ............. .......... . . . .. .. :: ::. : .. ::.: :.:: : .:::: ...... : : ::.: :.:: ::: :: ,: ,:: :::~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. .. . . ... . . .. ... .. .. .. .. '""S~~~ww3m, 0A#.$ $°w $fu4 #0 1 D $0$ n o.~~~~~~~~ q" Ela 830Gb%. amF& 4@@° t"$aiS3q : 2" v @ Z z zo: p: w: 9: :¢ ,'0~~~~:T.Wl' .AP..P @.:.... ..... . ....... . ... .. . .. . . .. .. .. .... ... ... ..... .. 9 .. . ,t .... .... . ... .. . .. . ... ... .......## U3,1t n2 3*' S @333Lt 3 ''''''' '''' m~T~1"V ~3 3 . .. . . . . . .. .. ........ .... .. .... ............. ...... .... . . .. . ...... :: . .................. ........... ............ .....~~~... ........ . ................... .................... ....................... ....... ..... ........... :: . q : ia .: u GbZ3G 1 ,, .. .................... ... . ....... '*ZJ4~~d2gOr " so; '00t0#**0.........U.... ..2... .. ........ v: :. . . . ........:.. .. . ...::: :: -o onpUrnasid up ei 1'3Z9 01 303 U0233 3331403d(, '*Z*h~. ~uu3u o ~ ............... O's .30 pd*3 ~l 3 u 3 t~ w*w~#04~9~WfW* .. .... ... . . .. 1 . L.. ......~2 a l .b. 5 I i ~ ~ ..... .....3 .. u. w .. ... zw.. p AT7~~~~~~40Z ..$i ~~~~~~~~~ m aanqu~~~~~~~ F .... . ...5 ..... .. .. .O. ..... 33SO~~~~ USA GUY 'IIAZOWASYILEUZ. .'..E... ... ... .........E... ............ -41 - or socially justified; rather, assessed Vanuatu Public Works Department indicated maintenance levels reveal the scale of the support for this approach.) maintenance issue. As stressed above, existing transport infrastructure in its entirety is non- 2.22 Assessed expenditure on routine and sustainable; demand for some facilities may have periodic maintenance of roads in the five PMCs shifted since its original construction, some is estimated to be on average 2.9 percent of facilities may have been over-or under-designed GNP (see Table 2.6). Actual road maintenance originally, and some facilities with relatively low across all PMCs is, on average, 31 percent of economic or social merit may have been assessed maintenance, and I percent of GNP.' introduced. Both assessed and actual expenditure on road maintenance are high relative to the size of the Road Maintenance Expenditure PMC economies. Yet estimated actual expenditure on maintenance falls well short of 2.21 Actual and assessed maintenance costs for assessed maintenance in all PMCs; acta all transport infrastructure, including roads, as expenditure ranges from only 17 percent of estimated for 1991, are presented in assessed maintenance in Vanuatu to 69 percent Table 2.6.1" The assessed annual maintenance in Kiribati. Current expenditure on road cost for road maintenance is, as a proportion of maintenance in Kiribati is, at 1.4 percent, a infrastructure replacement cost, high in the comparatively low proportion of the Solomon Islands and Vanuatu. This arises from Government's recurrent expenditure; expenditure the high proportion of engineered roads in the at the level of assessed maintenance is estimated two countries (see Table 2.1). Road to require 2.0 percent of government outlays; maintenance costs in these two countries are also this is a lower proportion than that estimated for relatively high because of the greater need to the other four PMCs.2' In these four PMCs, move maintenance equipment between a number assessed road maintenance expenditure is of major islands, and soil and geographic between two and six times the existing level. conditions which increase the cost of road The assessed levels of road maintenance range maintenance. Particularly high unit maintenance from 6 percent of government expenditure costs reported for Vanuatu further raise the cost (Tonga) to 27 percent (Vanuatu). Such levels of of maintenance as a proportion of asset road maintenance expenditure are non- replacement value. Unit road maintenance sustainable; even with increased cost recovery; costs, by road type, as assessed for the five maintenance obligations also indicate that PMC, and for two comparator countries (Fiji substantial rationalization of existing roads and and Thailand) are presented in Table 2.7. standards is required. Clearly unit maintenance costs for unsealed engineered roads and earth roads in Vanuatu are Marine Maintenance Expenditure inordinately high and they call for further review; they are 120 percent and 220 percent, 2.23 Assessei maintenance expenditure respectively, greater than for similar roads in the associated with marine infrastructure is relatively Solomon Islands. They lead to somewhat low, being, in aggregate for all five PMCs, only atypical conclusions for Vanuatu, therefore, the 13 percent of total assessed maintenance road maintenance situation in Vanuatu is placed expenditure for the road subsector (see in perspective by supplementing the reported Table 2.6).' Vanuatu figures with results obtained by appiying the road maintenance unit cost levels 2.24 The gap between actual and assessed from Solomon Islands."9 (Subsequent to this maintenance expenditure is least in the Solomon analysis, discussions with the Government of Islands, where the commercially oriented and largely autonomous Solomon Islands Port - 42 - Tabl 2.6s PACIFIC ISPORT IIIaT.ucTU3 x Uh C COSTS, 1991 solon Western Total or Kiribati Islands Tonga Vanutu tL l, is Averag_ La Assessed Maintenance Aseessed AnnAl Maintenance Cost (US$ mdllion) loads 0.3 2.5 1.5 5.6 (3.51 3.0 16.2 111.41 Marie 0.1 0.5 0.7 0.5 0.3 2.1 Aviation 0.1 0.2 0.4 0.5 0.4 1.9 DI"a li LI L LI ILLI fIIL AisI Assooeed Annual Matemace Cost (S of Irntr ctuRe placmat Vablu) loads 1. 6 .1 2.3 10.7 (4.53 25 5 .1 13.61 Marin 0.6 2.0 2.6 1.4 1.3 1.7 AviLation 2.1 1.5 1.3 1.3 1.4 1.4 biga Li id Lak L LLII LI L LLZI Aaeege Aal MaIntesaace Cost (S of SU) Roads 0.6 1.9 1.6 6.0 (2.41 2.6 2.9 12.13 lkrIm 0.2 0.3 0.5 0*4 0.3 0.4 Avlatioa 0.2 0.1 0.4 0.5 0.4 0.3 Insi Li La LI Ai ILAI Li UALLII Assessed Annual Mainteance Cost (S of Governamt vove) loads 1.59 6.00 5.09 21.17 (8.53 5.71 0." (0.31 Marln 0-29 0.94 2.14 1.15 0.56 1.43 Aviatimo 0.37 0.59 1.25 2.07 0.50 1.07 2W LII LB ILi ALM 15.41 7 i 1.33 14&.U Actual Maintenance IzpendLture (US$ millon) Lk Road 0.19 0.99 0.83 1.49 1.49 4.99 Marin 0.02 0.26 0.10 0.09 0.13 0.59 Aviation 0.03 0.07 0.10 0.22 0.11 0.54 Actual Maintenance (a S of Assessed Mantenance) Roads 69 35 55 17 (391 49 31 (441 Marine 34 53 14 17 36 25 Aviation 54 34 24 29 25 25 DS&L u 21~i in B 111 II Il Lin Actual SaintenaCD e (s S of Goverrmnt Owosno) a Roads 1.1 2.1 2.8 3.6 2.5 3.0 Marin 0.1 0.5 0.3 0.2 0.2 0.4 Aviation 0.2 0.2 0.3 0.6 0.2 0.3 Z2a Li LI Li ia Li Li L The road mateOnacsO estimatee for * auatu are Liordinatel 1 gh sa the source data requtrer review. The figures *ow I I theu, reprosent road maintenae coats for *amatuY , based upon the unit maIttenac c o st for Solomonalsands. See para. 2 22 In the text. Lk lasd on maitenase In 1991, or latest available year ia 1991 prles. Ic lasd on recurrent rev_ee in 1959 inflated by 5 perent per m to iledative 1991 prlce. Sorcesm Mislm astimtes sod World Bank (1I91a). - 43 - Table 2.7: PACIFIC ISLANDS AND COMPARATOR COUNTRIES-ASSESSED UNIT ANNUAL ROAD MAINTENANCE COSTS, 1991 (US$ per kilometer) Solomon Western Comparator Countries Road Type Kiribati Islands Tonga Vanuatu Samoa Fiji ;a Thailand /a Sealed 2,000 4,500 2,600 { 4,170 4,250 3,000 { 6,800/a Unsealed/Graveled { 6 m pavement 4,000 3,650 4,000 2,400 3 m pavement 200 2,100 1,100 2,000 2,580 Earth Formed 80 1,000 620 1,000 250 1,000 700 Average 370 2,150 800 4,880 1,450 - /a Estimated from total maintenance requirements for mix of road types, based on recent Bank projects. Source: Mission and Bank estimates. Authority is in a position to exercise cost increased maintenance expenditures should be recovery measures, budget for maintenance, and determined by reviewing the stock of marine control of its assets. However, there is very low infrastructure. expenditure on maintenance of other marine assets in the Solomon Islands and overall actual Aviation Maintenance Expenditures marine maintenance expenditure is one-half the assessed level. Actual maintenance expenditures 2.25 Assessed maintenance expenditure in the are well below assessed maintenance levels in dviation sector is, as with the marine sector, Vanuatu (with two major ports and sixteen small in absolute terms. Notwithstanding the jetties) and Tonga (where substantial investment need to meet international standards for airside has been made in port facilities at Nuku'alofa). facilities, actual expenditure is generally only Much of the investment in marine facilities in between a quarter and a third of that estimated these two countries has been made during the as associated with sustainability of all aviation 1980s and, so far, the effects of deferred assets. Low maintenance in the past has been maintenance are small. Assessed maintenance is feasible, in part, because Vanuatu, Western expressed here on an equivalent annual Samoa and Tonga have had new runways and discounted flow basis, i.e., a uniform amount passenger terminals constructed at their per year. Actual routine and periodic respective gateway airports with grant aid during maintenance needs are not uniform and increase the 1980s. However, on the basis that these with asset age. Unless a "sinking fund" is new facilities are warranted, a substantial established based on equivalent annual require- increase in maintenance expenditure is implied to ments, adequate financial resources will not be sustain the infrastructure over the long term. readily available. Of course, the need for - 44 - Fiour 22 . PACIFIC ISLANDS-ACTUAL VS A853553D MuINTINANCc 100- _ _ . _ - _ - Sol 40d 06 Kiribati Islands Tonga Vanuatu W. Samoa | Roads MV Marine Aviation| I 5 -45 - Maintenance Expenditure In Perspective C. IMPLICATIONS OF INADEQUATE MrAnENANCE 2.26 The ratio of actual to assessed maintenance expenditure for the five PMCs as a 2.29 The first priority for maintenance is whole is very similar for each mode (31, 28 and ensuring that it is only applied to warranted 28 percent respectively, for roads, marine and assets. The second priority is to ensure that aviation). However, there are significant adequate (assessed) maintenance is effectively differences between the countries, as shown in utilized and efficiently priced at the specified Figure 2.2. Actual expenditure relative to standard. Inadequate maintenance results in assessed is clearly lowest (18 percent) in more rapid deterioration of infrastructure assets, Vanuatu. Actual maintenance outlays in the and poorer quality of the services they provide. Solomon Islands, Western Samoa and Tonga are Deteriorated infrastructure reduces the efficiency less than one-half the assessed level, while of transport operations and increases total expenditure in Kiribati is 61 percent of that transport system costs through: assessed. For the five PMCs in aggregate, current expenditure on maintenance of transport * higher capital replacement costs because infrastructure is 4.2 percent of recurrent of the shorter economic life, government outlays. Maintenance expenditure at the assessed level for all infrastructure would * higher maintenance costs because, in most require a 10 percent increase in total government cases, infrastructure is more difficult to current expenditure (though a considerably maintain once it has deteriorated, and higher increase would be required in Vanuatu). This increase is unrealistic. . higher transport user costs as deteriorated infrastructure increases damage to 2.27 Limited expenditure on maintenance in vehicles, vessels and freight and increases the past has been made possible by substantial transit times. external assistance to rehabilitate neglected existing infrastructure, and possibly by a decline Higher total transport costs reduce overall in its standard. It is unlikely that the PMCs will economic performance; increases in the resource be able to depend on sufficient assistance being costs of collecting, handling and distributing available in the future for early goods lead to general price increases, weaker rehabilitation/reconstruction of all existing export competitiveness and a reduced share of infrastructure. export income to producers. (The general form of the relationships between capital, maintenance 2.28 Itisnoteconomicallyfeasibletomaintain and user costs is indicated in Box 2.1 earlier.) the entire existing stock of infrastructure in "good" condition; some infrastructure is Lire-Cycle Infrastructure Costs unwarranted and will need to be rationalized and future maintenance concentrated on the most 2.30 Estimating The Maintenance - Capital important assets. New investments, Cost Trade-Off. The implications of inadequate rehabilitation, and maintenance need to be maintenance of transport infrastructure in the considered jointly, together with institutional PMCs, are examined by estimating total requirements, in setting expenditure/assistance transport system costs (life cycle costs for priorities. existing infrastructure plus user costs) with current actual maintenance and those that would be associated with assessed maintenance. Changes in infrastructure life cycle costs and transport user costs are considered separately. -46- Data limitations constrain the quantitative p.a. to US$53.6 M p.a. (a saving of US$10.7 M examination of changes in transport user costs to p.a.) and from US$53.6 M p.a. to US$26.3 M the road subsector. p.a. (a saving of US$26.7 M p.a.). Each increment of maintenance expenditure secures a Savings in Infrastructure Costs corresponding net saving in capital cost. 2.31 The relationship between maintenance 2.34 Benefit - Cost Ratio of Improved and capital costs in the five PMCs is examined Maintenance. The benefit-cost ratios tend tc be by comparing the equivalent annual capital cost highest in the aviation subsector bit are also of infrastructure (based on a discount rate of high for other modes in certain countries. 7 percent and estimated economic lives of For each mode, in all countries the incremental infrastructure) under actual and assessed benefit-cost ratios are well above 1.0, with the maintenance with the cost of actual and assessed exception of roads in Vanuatu. The benefit-cost maintenance. The results of the analysis are set ratio of shifting to the assessed level road out in Table 2.8. The analysis indicates that the maintenance in the Solomon Islands *nd Vanuatu total capital cost of transport infrastructure is is low compared with the ratio for the other 46 percent higher under actual maintenance, than countries.M6 under assessed maintenance, i.e., US$79.7 million compared with US$54.5 million. 2.35 There is considerable heterogeneity in transport infrastructure, and the marginal value 2.32 The estimates of life-cycle infrastructure of additional maintenance expenditure will differ costs under different levels of maintenance with the type. In the marine sector, for generally confirm that the total costs,2' for all example, wharves, buildings and equipment have existing infrastructure, are reduced by shifting different economic lives and maintenance needs, from the actual level o, maintenance to the and hence there are different implications of assessed level. (It should be possible to reduce inadequate maintenance. Similarly, differences total costs further by increasing the productivity in the mix of fixed and renewable infrastructure of maintenance activity itself.) As stressed influences the total cost implications of different above, these results indicate the scope for levels of maintenance. For example, inadequate infrastructv-e cost savings; they are relevant maintenance of a sealed road should not, at least only for warranted assets. initially, result in deterioration of the road base and drainage components of the road. By 2.33 The estimated replacement capital costs contrast, the lack of a protective sealed surface and assessed maintenance levels of existing on other engineered and earth roads results in infrastructure are brought together in Table 2.9. more rapid deterioration of the road base in the The table illustrates (for a zero discount to event of insufficient maintenance. simplify interpretation) how the level of maintenance affects the economic life of 2.36 Unsealed roads predominate in the infrastructure, and thereby, in turn, its annual PMCs. The construction cost of these roads is capital cr;:. Shifting from the actual to the relatively low, but maintenance costs are high assessed level of maintenance doubles economic and depend primarily on environmental life. Savings in annual capital cost exceed conditions rather than traffic movement. These substantially the corresponding increase in conditions raise the possibility that there may be maintenance. Thus, increasing the level of instances where a cycle of road construction maintenance for all infrastructure from zero to with limited maintenance, followed by relatively actual (US$6.1 M p.a.), and from actual to early reconstruction, might result in lower ltfe- assessed (US$14.1 M p.a.), reduces the cycle costs.27 corresponding capital costs from US$64.3 M - 47 - Table 2.8: PACIFC ISLANDS-CAPITAL, MAINTENANCE AND USER COSTS OF INFASTRUCrURE Solomon Western Total or Kiribati Islands Tonga Vanuatu _a Samoa Average Lj Equivalont Annual Capital Cost of Infrastructure with Actual Maintenance (US$ irillion) /b Roads 2.1 7.3 9.8 12.4 14.9 46.5 Marine 0.9 2.7 3.4 3.7 2.6 13.4 Aviation 0.7 2.1 4.3 8.3 4.3 19.8 12111 3. i2L.I ild aL 2i2 19L Equivalent Annual Capital Cost of Infrsstructure with Assessed Maintenance (USS million) Roads 1.5 4.6 6.4 8.0 10.1 30.7 Marine 0.7 2.0 2.4 2.9 2.1 10.0 Aviation 0.6 1.4 3.0 5.9 3.0 13.8 il Uz u 11.8 i 2 54. Additional Annual Capital Cost of Infrastructure Resulting from Inadequate Maintenance (US$ rnillion) /c Roads 0.6 2.7 3.4 4.5 4.8 15.9 Marine 0.2 0.8 1.0 0.8 0.6 3.4 Aviation 0.2 0.7 1.3 2.4 1.3 5.9 l a Li 8 7 7 6.7 X Incremental Cost of Shifting from Actual to Assessed Maintenance (JS$ rnillion) td Roads 0.10 1.8 0.7 7.1 [2.3] 1.5 11.2 Marine 0.08 0.2 0.6 0.4 ^ 1.5 Aviation 0.08 0.1 0.3 0.5 .3 1.4 Benefit/Cost Ratio L2 Roads 6.4 1.5 6.0 0.6 [1.9) 3.1 1.4 Marine 6.1 3.3 1.6 1.9 2.6 2.2 Aviation 7.7 4.7 4.2 4.4 3.9 4.3 Tota 6 1. 9 & 1.0 f2 31 321 1.8 Roads: Vehicle Operating Cost (US$ per kn) Lf 0.26 0.25 0.23 0.21 0.2' 0.23 Vehicle Operating Cost Penalty L0 Annual Cost (US$ million) 0.4 2.1 ].2 2.1 1.9 7.7 17.61 % of GNP 0.8 1.4 1.3 1.5 1.6 1.4 [1.6] La The figures shown [ ] thus, represent road maintenance costs for Vanuatu, based upon the unit maintenance costs bv road typo for Solomon Islands. See para. 2.22 in the text. Lb Equivalent Annual Cost calculated at 7 percent discount rate and economnic lives based on nature of infrastructre for roads, rnarine and aviation. For details, see individual country Maintenance Survey Annexea, Volume Two of this report. Difference between annual capital cost of infrastructure with current and assessed maintenance. Ed Difference between current and assessed annual maintenance cost. k Ratio of saving in equivalent annual capital cost resulting from assessed maintenance to the incremental annual cost of-the improved maintenance. lf Average vehicle operating cost (net of taxes and duties) on paved roads in good condition. (See Country Maintenance Survey Annexes, Volume Two of this report). ag Indicative differences in economic road vehicle operattng costs between current road conditions and roads in good condition. Details of calculations are presented in the Maintenance Annex to each country sector survey in Volume Two. Source: Mission estimates. - 48 - Table 2.9: PACIFIC ISLANDS-EFFECT OF MAINTENANCE ON INFRASTRUCTURE CAPffAL COST AND ECONOMIC LIFE La Maintenance US$ million (USS million D.a.) /b Economic Life (vears) /c Level p.a. Road Marine Aviation Total Road Marine Aviation Average Assessed 20.3 [15.4]Ld 15.0 3.6 7.6 26.3 21 35 18 22 Actiual 6.1 31.6 7.9 13.7 53.6 10 16 10 11 Zero 0.0 45.1 10.5 15.2 64.3 7 12 9 9 Total Replacement Cost (USS million) Stock Value 316 126 137 579 /a For all five PMCs: Kiribati, Solomon Islands, Tonga, Vanuatu and Western Samoa. /b At zero discount rate. Under zero maintenance, annual capital cost is replacement coAt divided by economic life. /c Based on judgement for zerc maintenance. Ld Figure shown [ ] thus, is for adjusted road maintenance costs in Vanuatu (see pan. 2.20). Source: Mission estimates and Table 2.8. Savings in User Costs rougher roads. These conditions may result in a further decline in economic efficiency througn 2.37 In addition to the savings in life-cycle reduced accessibility and increased spoilage of infrastructure costs which may accrue to freight, and loss in net economic output (if Governments and donors from shifting to increased costs make a potential economic assessed maintenance, users of the transport activity non-viable).' system can also secure lower costs. Transport user operating costs decrease under better 2.39 Vehicle operating resource costs have infrastructure quality and more effective been derived ° for each of the PMCs; the maintenance; with improved maintenance the average vehicle operating cost on paved roads in costs of wear and tear (and possible delays) to good condition is described in Table 2.10. vehicles, vessels and equipment, decline.' Differences between the countries is largely These savings can be estimated most readily for attributable to differences in fleet composition the road subsector. and data quality. Vehicle operating costs were calculated for sealed and unsealed roads in good, 2.38 Road Condition and User Costs. Road fair and poor condition; judgements were made vehicle operating eosts rise with inadequate road on the current volume of traffic on each road maintenance because of the increased fuel type by condition. In indicative terms, it is consumption, more rapid vehicle depreciation, estimated that vehicle operating economic costs and greater need for repairs, which result fronm are between 8 percent and 14 percent higher at .49 - Table 2.10: PACIFC ISLANDS-COSTS AND BENEFITS OF IMPROVED ROAD MAINTENANCE, 1991 Solomon Western Total (or Kiribati Islands Tongs Vanuatu / Samoa Average) /c Boad Lenth km La 821 1,301 1,874 1,760 2,072 7,828 Se loda roatid8 (percent of total length) 4 8 6 7 14 (8) Reiplacemeont Cs lst million 18.7 45.7 64.9 79.8 106.7 315.9 USS'000/km 22.8 35.1 34.66 45.3 51.5 40.4 Masintenanco Cost Annual, si rnuillion Assessed 0.3 2.7 1.5 8.6 3. 8; 3.0 16.2 [11.4) Actual 0.2 0.9 0.8 1.5 1.5 5.0 Difference 0.1 1.8 0.7 7.1 (2.31 1.5 11.2 MAaintonanc Cost Annuai, 0347kni Awssed 370 2,150 8W0 4,880 [2,1601 1,450 2,069 Actual 240 690 430 850 725 640 sa[lviet mann JUSS million 2.1 7.3 9.8 12.4 14.9 46.5 (USS,000/km) (2.6) (5.6) (5.2) (7.0) (7.2) (5.9) JUSS million 1.5 4.6 6.4 8.0 10.1 30.7 (US '000/kcmt) (1.8) (3.S) (3.4) (4.S) (4.9) (3.9) ALvinQ iR Replacemont )MIMlt ASSeS M,alitenance USS mnillion 0.6 2.7 3.4 4.4 4.8 15.8 (US$'000/km) (0. 8) (2.1) (1.8) (2.5) (2.3) (2.0) Intimstr clrCost USS rmillion 0.5 0.9 2.7 -2.7 [2.1] 3.3 4.6 Vehiclo Opettinyr Cost /b Annual bamswt US3 rnmilion 0.4 2.1 1.2 2.1 1.9 7.7 NtAnnuial ais Iotal I nlzrstrcue and mI ion 0.9 3.0 3.9 .0.6 [0.01 5.2 12.3 L All roads; saed, ensineered and earth. Since the mix of.road type varies across the countries (see Table 2.1 above), the averge figures presented here need to be interpred with caution. l Rel ative to roads in ood condition. ISe text p 2.22 ror discuion of the inordinaly high level of road maintenance cost in Vanuatu; the levels inbown [ thus, are based on the unit maintennce costs, by road type, for Solomon Islands soo pam. 2.22 in ihe text. Source: Table 2.1, 2.7 and mission estimates. - 50 - present, than would occur if all roads were in these subsectors are far less developed than for good condition (see Table 2.8). This roads. Increased port operation costs result incremental cost is equivalent to an average of from inadequately maintained port facilities. 1.4 percent of GNP over the five PMCs-which For example, the load-carrying capacity of is high. Foreign costs are almost four-fifths of wharves and jetties may be reduced by vehicle operating economic costs, and the higher insufficient deck and pile maintenance, and vehicle operating costs add between 1 and equipment may be in disrepair. Increased port 2 percent to the value of imports in the PMCs. costs and delays add to the cost of imported goods and make exports, most of which pass 2.40 Total transport system costs can be through sea ports, less competitive on estimated only for the roads subsector and these international markets. Safety for port workers are shown in Table 2.10. Increasing may also be reduced. Quantification of these maintenance to the assessed level across all five implications has not been possible. PMCs (an increase of US$11.2 million per annum), is estimated to reduce total road 2.43 While major airports are not generally at infrastructure costs by US$4.6 million per risk of closure because of inadequate annum, and vehicle operating costs by US$7.7 maintenance, such closure would have a severe million per annum. The total net savings across impact on travel and, in particular, tourism. the five PNiCs is US$12.3 million per annum, Deterioration of airport terminals and other and the incremental benefit-cost ratio is public facilities will have an adverse impact on estimated to be 2.1. user perception and marketing credibility. Provincial and local airfields are more prone to 2.41 Savings in annual vehicle operating costs closure. In the Solomon Islands, for example, under assessed maintenance of roads are equal to about two airfields are closed each month, often about half of the savings in equivalent annual because of maintenance problems which can be capital replacement costs. The savings are easily remedied.3' relatively higher in Kiribati and the Solomon Islands. In general, though, in the PMCs the Maintenance Incidence and Responsibility savings in vehicle operating costs from improved maintenance are low in comparison with savings 2.44 The incidence of costs and financial in infrastructure capital cost because of low savings from improved maintenance will differ. traffic volumes involved. However, comparison A high proportion of transport infrastructure of vehicle operating costs and road maintenance investment in the PMCs, inclading costs, which both involve a full domestic impact rehabilitation, is funded through overseas aid. is revealing. For all PMCs, with the exception The external contribution to maintenance is of Vanuatu, the annual savings in vehicle generally limited to the provision of equipment operating costs exceed the additional and technical assistance. In the absence of maintenance cost to achieve them. It follows additional donor grant assistance for that it is in the direct economic intreest of the maintenance, improved maintenance requires PMCs to increase their maintenance of additional outlays by the Governments of the warranted road facilities. PMCs. The savings in rehabilitation costs to external donors would be about 80 percent 2.42 Marine and Aviation Infrastructure greater than the increaeed maintenance outlays. Condition and User Costs. It is more difficult Infrastructure user cost savings will accrue to quantify the cost penalty of poor condition directly to vehicle operators; where these infrastructure to transport users in the marine operators are providing a transport service for and aviation subsectors. Quantitative guidelines others (passengers and shippers), the savings of the implications of inadequate maintenance in should be passed on to these individuals, since -51- the structure of the transport s--vice markets in improved maintenance. These costs include the the PMCs is effectively competitive. additional management, administration supervision, and training involved. Since the 2.45 While the need for aid funds for transport number of personnel that would be required is infrastructure in the long-term will be reduced if small (possibly 2-3 semi-professionals for each maintenance is improved, there will remain a modal area, or a total of less than 10), the need for assistance,particularly in the near term. additional staffing costs should be modest. Attention should be focused on the prevailing Moreover, efficiency gains can be anticipated maintenance overhang (i.e., the amount of from contracting out, where possible, investment required to rehabilitate warranted maintenance delivery "on the ground" to the existing infrastructure to good condition), private sector. Nevertheless, several alternative particularly in the road sector. approaches are available for establishing these new management and implementation 2.46 Totaltransportinfrastructuremaintenance tasks-%arying from using expatriate needs to be considered against competing needs professionals ("imported services") to indigenous for external assistance. Table 2.11 indicates staff and training ("import substitution"). A transport infrastructure replacement and assessed cost-effective choice across these alternatives maintenance costs in comparison to total needs to be taken. A perspective on the net government and development expenditure, and gains of improved maintenance (i.e., after external assistance. The expenditure data plainly implementation costs) in the roads subsector demonstrate the large and unrealistic demands alone, cam, be obtained by considering the order that sustainable existing transport infrastructure of magnitude of the estimated net equivalent would place on existing Government expenditure annual savings as -' -vn in Table 2.10. In 1991 and ODA for each PMC. prices, these vary from US$900,000 in Kiribati to US$5,200,000 in Western Samoa. (Vanuatu 2.47 These findings are reinforced by is the exception, with no net gains given the examining the primary sources of ODA. Major inordinately high reported road maintenance bilateral assis.ance to the PMCs is provided by costs.) Notwithstanding the likely low accuracy Australia, New Zealand and Japan. Table 2.12 of these estimates, they are orders of magnitude reveals the order of magnitude of transport above the likely range of implemenmation costs. infrastructure assessed maintenance in A similar conclusion can be drawn for the comparison to prevailing levels of assistance marine and aviation subsectors; annual net from Australia and New Zealand. Although the savings in life-cycle infrastructure costs vary difference between existing and assessed from US$500,000 for marine in Tonga, Vanuatu maintenance represents a small percentage of and Western Samoa, to US$1 million and existing capital project assistance, total US$1.9 million for aviation in Tonga and infrastructure (replacement and assessed Vanuatu, respectively. These estimates do not maintenance) costs represent an inordinately high include the additional savings accruing to users proportion of total assistance. This brings into of the facilities.2 question the long-term sustainability by donors of the entire accumulated stock of existing 2.49 In brief, the stock of infrastructure is transport infrastructure. substantial, significant savings in its long-term costs are possiblk (following rationalization), and 2.48 Organizational Costs of Implementing the modest personnel costs in securing these Effective Maintenance. The benefits and costs savings would yield a high rate of return. associated with shifting to "assessed" maintenance indicated here do not take into account the costs of implementing and securing - 52 - Table 2.11: PACIFIC ISLANDS-TRNSPORT INFRARUCTuRE AND DEVELOPMENT EXPENDITURE, 1985-1988, 1991 Solomon Western Total or Kiribati Islands Tonga Vanuatu Samoa Average Population ('000) 67 304 101 151 158 158 GNP US$ million 46.3 15.1 92.6 138.9 115.7 81.7 per capita 650 430 800 820 580 517 Gov't Exp)enditure lb USS million TAnnual) 18.0 5.9 40.7 64.3 53.6 182.5 Development Expenditure lb USe million (Annual) 17.5 1.9 20.3 24.7 36.1 100.5 Overseas Devel. Expend. Ia USS million (Annual) 16.3 5.8 18.8 39.3 30.6 163.3 % Gov't expend. 90.5 84.9 49.1 61.2 55.5 55.1 per capita 243 192 186 260 182 207 Infrastructure L Replacement Value US$ million 31.0 83.5 123.3 174.6 166.1 578.6 US$'000 (Equiv.Annual) 2.7 8.0 11.8 16.8 15.2 54.5 % Dev. Exp. Lk %D GNP 73 55 132 122 142 165 per capita 438 248 1,207 1,061 980 685 Maintenance Assessed US$'000 (Annual) 300 2,800 1,500 8,600 3,000 16,200 % Gov't expend. 1.7 47.5 3.7 6.2 4.7 8.9 % ODA 1.8 4.8 7.0 21.9 9.8 9.9 Maintenance Actual (Annual) US$S'OO 200 900 800 1,500 1,'00 5,000 % Gov't expend. 1.1 15.2 2.0 2.3 2.8 2.7 % ODA 1.2 1.5 4.2 3.8 4.9 3.1 ToJ41 lpfrastruwcSes Cost US$th U (Eqaq uilv. Anen.) m 8.000 11 80 16.80 15.20 54.500 % Gov't expend. 15.0 136 28.9 26.1 28.4 29.9 % ODA 16.6 137 62.8 42.7 61.5 54.2 a 1988. 1985-88, Tonga 1985/86-1988/89; averaged as percent of GNP over corresponding years. InfrastructureTigures are as estimated for 1991, in 1991 prices; ODA and Government expenditure figures are based on average percentages of GNP over the indicated periods, generally 1985-88. Source: Table 2.8, and World Bank (1991a). - 53. Table 2.12: INRAsTucruRE AND ODA-AUSVRALA AND NEW ZEALAND, 1988 Solomon Western Kiribati Islands Tonga Vanuatu Samoa Total - -- ---- (USS million) Total Assistance Australia 2.03 7.57 7.44 7.68 6.63 46.35 New Zealand - 1.69 2.43 1.89 3.79 13.45 lQa 92 10 1 68 S 10 4S8 Economic Infrastructure Australia 0.29 0.15 3.37 0.74 0.74 4.12 New Zealand - - 0.14 0.34 0.07 0.54 Vll gm 0.15 Lii O 84 (% of Total ODA) (14.3) (1.6) (32.5) (12.6) (7.8) (7.8) Transport Infrastructure Assessed Maintenance 0.3 2.8 1.5 8.6 3.0 16.2 Assessed Replacement Costs (Equivalent Annual) 2.7 8.9 11.8 16.8 15.2 54.5 Total Infrastructure Costs (Equivalent Annual) Q2I 1 132 1 70.7 Source: Table 2.8; Table 2.10; South Facific Economic and Social Data Base, National Centre for Development Studies (Canberra); and World Bank (1991a). Maintenance Impact and Accountability direct link between income collected by infrastructure supply agencies and the capital and 2.50 Government agencies responsible for the recurrent funding provided to them, weakens the maintenance of roads are largely insulated from obligation of the agencies to justify the price and the transport user cost implications of inadequate quality of their infrastructure to users. This maintenance, but are faced with the need to separation of revenue and cost functions reduces sustain deteriorated roads with limited funds. more broadly the incentive for efficient Their funding is sourced almost entirely from development, use and maintenance of transport aid funds and government consolidated revenue, infrastructure. Improved accountability between with the scale of the latter being largely agencies and their users "clients' should be independent of road conditions. There are direct established; greater autor.omy and user fees in the maritime ar., aviation sectors, commercialization of the agencies should be but the income from them is usually funneled to introduced. go -nment consolidated revenue. The lmited - 54 - Implications for Future Planning Activities responsible for development and maintenance of its assets, in Tonga where port income in excess 2.51 Need for Information Based on of a designated amount may be retained by the Conditions in the PMCs. The current difficulty Ports Administration Department for its own use in adequately maintaining existing infrastructure to supplement its usual budget appropriations, results from funding constraints, management and in Kiribati where income from a toll on the capacity, delivery limitations, and excessive past Bairiki to Betio causeway is retained for investment in projects with inadequate economic maintenance of the causeway. rates of return. Development of a strategy for improved maintenance requires an understanding 2.54 The level and structure of user charges of the cost of infrastructure to PMC reflects government objectives and revenue Governments, to donors, and to transport users, needs to support general government activities and alternative approaches to maintenance. (including social obligations) and to promote This, in turn, requires: economically efficient use of transport assets. It will be rare that fees levels associated with each * The engineering and cost information of these objectives will be consistent. Ideally, needed for effective maintenance the pursuit of economic efficiency should take management. precedence as this expands scope for the pursuit of other objectives. However, the need for an * An understanding of the effects of assured, stable flow of funds for maintenance of inadequate maintenance and of alternative transport assets is of key importance and should design standards on asset life and be given greater priority in the five PMCs as a transport user costs. tool of economic and infrastructure management. Statutory obligations, such as with the Solomon - An understanding of the social and Islands Port Authority, are a means for economic opportunities foregone because promoting a change in emphasis and of deteriorated infrastructure. accountability. But improved management can also achieve much of this without new 2.52 An improved database and more detailed organizations. These circumstances may justify study than is available in this present study is the use of formal "earmarking" of revenue from required to specify in operational detail the best taxes on users for warranted maintenance.33 maintenance practices for transport Such modification alone is insufficient to secure infrastructure, and the full implications of optimal maintenance, and must be supported inadequate maintenance. There is, however, with developments in other areas, in particular sufficient evidence that the total cost of institutional capacity, management and sustaining satisfactory services from the stock of effectiveness in delivery. warranted existing transport infrastructure can be reduced by improved basic maintenance. 2.55 There is little knowledge of the level of cost recovery in each transport subsector at present. Cost-recovery has been estimated for D. MNrAIEANCE FUNDiNG AND CosT each mode, in each of the five PMCs, using RECOVERY available data on revenue, and data on infrastructure capital and maintenance costs as 2.53 There is generally no hypothecation of presented in Table 2.9. The estimates of cost- revenue from user charges for maintenance of recovery are based on assessed levels of transport infrastructure in the five PMCs. expenditure oii maintenance, not actual levels. Exceptions are in the Solomon Islands where the In only a few instances is current income greater Port Authority retains its revenue and is than actual maintenance expenditure. In no case - 55 - is revenue sufficient to recover both maintenance system (including maintenance at the assessed and capital charges (see Table 2.13). The total level) in Western Samoa, but an average of only cost of all existing infrastructure with actual 8 percent for the five PMCs in aggregate. Even maintenance is greater than would be the case with the inclusion of all revenues from import with maintenance at the assessed level (with the duties on vehicles and fuel attributed to road cost possible exception of roads in Vanuatu). In recovery, revenue from road users is addition, existing infrastructure netds to be significantly less than that required for recovery rationalized and maintenance allocated to the of capital and maintenance costs. Across all highest priority assets. Therefore, levels of PMCs, the implied level of cost recovery in the cost-recovery indicated below are lower than roads subsector is 34 percent (Table 2.13). Cost those that would result from rationalization, but recovery is highest in Western Samoa higher than those that actually prevail at present, (52 percent); Tonga and Vanuatu are very similar (42 percent and 44 percent, Roads respectively), if unit maintenance costs in Vanuatu are based on Solomon Island levels (see 2.56 With the exception of the toll on the para. 2.20 above). The higher unit road Bairiki-Betio causeway in Kiribati, there are no maintenance costs reported for Vanuatu would direct charges for road use in any of the five increase the maintenance expenditure for that PMCs. Road users contribute to the cost of country and reduce the level of cost-recovery (to development, maintenance and operation of the 31 percent). Duties on road vehicles and fuel road system indirectly through vehicle for these three countries are, on average, double registration fees, and annual vehicle and driver the average rate of duty on other imports. It has license fees. The fee for annual vehicle road- not been possible to estimate revenue from worthiness inspections required in some duties in Kiribati and the Solomon Islands. In countries is considered a user charge to recoup any event, duties on vehicles imported into the the cost of the inspection, rather than a Solomon Islands have recently been reduced contribution to the cost of the road system. substantially, and current revenue will be much Import duties are usually imposed on vehicles at less than that indicated by historic data. About a higher rate than other commodities; a sales tax two-thirds of income in Kiribati is derived from on vehicles in Tonga applies generally to goods. the toll for use of the Bairiki-Bet.o causeway. A petroleum import commission is also imposed Revenue from this source is only available for in Tonga. These duties are economy-wide fiscal maintenance of the causeway; cost recovery for measures, and part of the overall design of each the remainder of the road system would be less country's taxation structure; they are not than the average 8 percent estimated as shown in normally regarded as user charges for transport Table 2.13. infrastructure. However, they can be considered contributions to transport cost recovery, and 2.58 It is improbable that total revenue from they may be viewed as having a similar effect to road users can be increased sufficiently to equivalent user charges. An estimate has been achieve full cost-recovery, even in Tonga and made of the approximate scale of these duties Western Samoa, where implied cost recovery is regarded as charges, and they are identified relatively high. This circumstance parallels separately (see Table 2.12). The revenue over-investment in road infrastructure, and its generated from them is five and a half times that non-sustainability, at least at original design from registration and lice,ase fees. standards. 2.57 Revenue from vehicle registration and 2.59 Two key equity issues relate to cost license fees, and driver license fees, is equal to recovery in the land transport sector: (a) the 17 percent of the cost of operating the road geographical distribution of sources of revenue - 56 - Table 2.13: PACIFIC ISLANDS-TRANsPORT SECrOR COST RECOVERY, 1991 La Solomon Western Total or Kiribati Islands Tonga Vanuatu L Samoa Average Roads Income Registration & License Fees 0. 1lf 0.2 0.2 0.5 0.5 1.4 Import Duties /b .. .. 3.2 4.7 6.4 14.3 Tota1 0. 3.4 A 6.9 15.7 Operating Expenditure tc 0.3 2.8 1.5 8.6 [3.8] 3.0 16.2 Operating Surplus -0.1 -2.6 1.8 -3.4 [1.8] 3.9 -0.5 Capital Charge /d 1.5 4.6 6.4 8.0 10.1 30.6 Total SuIWus (Deficit) -1.6 -7.2 A£ -1. j21 63 --.1 Cost Recovery /e 89% 3 % 42% 31% [44%] 52% 34% Marine Income Ls 0.0 2.0 0.5 0.5 0.7 3.8 Operating Expenditure 0.1 1.4 1.2 0.9 0.7 4.2 Operating Surplus -0.0 0.6 -0.7 -0.3 0.0 -0.4 Capital Charge 0.7 2.0 2.4 2.9 2.0 10.0 Ttal Sumlus (Deficit) -0.7 U4 -3.1 -1a 4 Cost Recovery 5% 60% 15% 15% 26% 27% Aviation Income a 0.1 0.2 0.6 1.1 1.1 3.0 Operating Expenditure 0.1 0.5 1.1 1.3 1.1 4.1 Operating Surplus -0.1 -0.3 -0.5 -0.1 -0.0 -1. I Capital Charge 0.6 1.4 3.0 5.9 3.1 13.9 Total Su olus (Deficit) 06 _6 :.U -14.9 Cost Recovery 10% 9% 14% 16% 26% 17% Total Income Lg 0.2 2.4 4.5 6.8 8.6 22.6 Operating Expenditure 0.5 4.7 3.9 10.7 4.8 24.5 Operating Surplus -0.2 -2.3 -0.6 -3.9 3.8 -2.0 Capital Charge 2.7 8.0 11.8 16.8 15.2 54.5 Deficit: (US$ million) -2.9 -10.3 -11.2 -20.7 -11.4 -56.4 (% GNP) 7% 12% 12% 14% 10% 10% (% Government revenue) 16% 22% 37% 49% 21% 30% (US$/capita) 41 31 109 126 67 67 Cost Recovery 7% 19% 29% 25% 43% 29% Lg In USS million, 1991 prices. /b Revenue from import duties on road vehicles and fuel. / Includes allowance for maintenance at assessed level rather than current expenditure. La Equivalent annual capital cost with maintenance at the assessed level. La Revenue as a share of the sum of the maintenance and capital replacement expenditure. Lf Includes revenue from an earmarked toll on the Bairiki-Betio causeway. Lg Income from import duties for the marine and aviation sectors is not available. Import duties in these subsectors are lower than for the road subsector. The figures shown [ I thus, represent road maintenance costs for Vanuatu, based upon unit maintenance costs, by road type, for Solomon Islands. See para. 2.22 in text for explanation. Source: Mission estimates. - 57 - and disbursement of expenditure; and (b) the Tonga. The urgency of this issue is lessened by apportionment of costs and revenue by vehicle the small number of large vehicles in use in category. In relation to the first issue, it is most of the countries at present; most trucks are likely that the proportion of vehicles and road small in size (3 tonnes or less). use is more geographically concentrated than the replacement value of the road network. Thus, Marine cost recovery can be expected to be higher on the main islands. This requires more detailed 2.61 Data in Table 2.13 indicate that only in assessment to ensure social equity without loss the Solomon Islands and Western Samoa is of economic efficiency. In all PMCs a major current revenue sufficient to meet operating proportion of the road network involves low costs inclusive of an increased allowance for traffic volumes (less than 100 vehicles per day). assessed maintenance. Che data for the Such roads, typically gravel and earth formed, Solomon Islands combine the income and provide basic reliable accessibility as a social expenditure for the Solemon Islands Port service. Fixed costs for maintaining these roads Authority with other Government expenditure on are high-over 75 percent. In such maritime facilities.) Across the PMCs, port fees circumstances, a lower share of revenue for cost may be relatively low in some locations, recovery will derive from specific vehicle-related however, the almost fourfold increase in fees; a higher share will need to be supported by revenue, associated with meeting the long-term more indirect charges and general taxes.' The cost of sustaining the present maritime issue of providing reliable access to small, often infrastructure, indicates that there has been over- remote, communities also arises in maritime and investment in PMC port infrastructure.' It is aviation with small wharves and airfields. Local not likely that the five PMCs will be able to "ownership" and responsibility for maintenance sustain all of their marine infrastructure. may be more feasible in these subsectors, as the facilities are localized. However, they may still Aviation serve a wide "catchment area", involving many separate communities/villages. Cost recovery in 2.62 Current income would exceed operating this area needs to be considered in a much expenditure for airport facilities with assessed broader context. The issue of provision of basic maintenance only in Western Samoa. A fivefold transport requires a specification of threshold increase in revenue is required to meet the long- service and the alternatives for its provision. term minimum cost of sustaining aviation The alternatives include direct transport options infrastructure in the five PMCs. As with the (infrastructure standards and "vehicle" marine sector, over-investment in aviation combinations) and indirect approaches such as infrastructure is indicated, and the Governments the location of services as a partial substitute for of the five PMCs will not be able to sustain all transport. of their aviation infrastructure from their own resources. 2.60 The second equity issue pertains to the allocation of costs and revenue to vehicle types. 2.63 User charges and levels of cost Trucks (heavier axle loads) cause more damage recovery should be consistent across to the road system than smaller vehicles. While subsectors to promote edcient choice of registration fees are higher for trucks in most mode. In the PMCs at present, modal PMCs, the differential is small and makes only competition is limited, but occurs for domestic a minor contribution to the higher costs that transport, principally between road and marine. trucks generally impose on the road system. Therefore, user charges, in relation to costs, Moreover, import duties on trucks are lower should be consistent between these subsectors. than those on cars, in all couatries except - 58 - E. CONCLUSIONS users is only about 29 percent of the capital and operating cost of sustaining all existing transport 2.64 Infrastructure. Each of the PMCs has infrastructure. This level would be higher with a substantial stock of transport infrastructure. rationalization. The aggregate replacement value of this stock across all PMCs is US$579 million (1991 prices) 2.68 Targets corresponding to higher levels of or US$685 per capita. This stcck of cost-recovery for warranted assets need to be infrastructure, has been accumulated over time, established, and new efficient user charges and represents a major resource for the PMCs. introduced on a consistent basis across However, estimated assessed maintenance and subsectors. An initial target should be to cover replace.nent costs for the existing stock of warranted maintenance costs from user charges; transport infrastructure is US$70.7 million per this requires only modest increases in revenue in annum; this amounts to 17 percent of regional the marine and aviation sectors. agregate GNP for the five PMCs (US$408.6 million) and over one-third of external 2.69 There is merit in the provision of direct assistance. external grant support for maintenance activities. Ihis involves the risk of weakening 2.65 Maintenance Expenditure. Actual rationalization of existing infrastructure. It expenditure on maintenance is about 30 percent therefore needs to be matched by commitments of that assessed to keep the existing stock in by Governments to the funding of maintenance good condition. To date, considerable and a timetable for cost recovery. premature and avoidable rehabilitation has been applied to unwarranted investments, and to 2.70 Rationalization of Infrastructure and compensate for maintenance deficiencies. This Justified Maintenance. Major changes are is highly inefficient; infrastructure and user necessary in transport infrastructure resources costs are substantially higher than management. First, there is an urgent need for necessary. Maintenance priorities needs to be PMCs to proceed with a program of set for those assets that are warranted. rationalization. This rationalization should be undertaken systematically by establishing 2.66 Incidence of Costs. Aid donors provide priorities for maintenance based upon explicit substantial assistance for rehabilitation of evaluation of the economic and social merit of transport assets and for development of new individual infrastructure facilities. Second, assets, primarily through grant aid. Only existing infrastructure should be subjected to limited assistance is provided directly for stringent economic/social evaluation; maintenance. Governments of the PMCs must reconstruction and new investment should not be fund maintenance almost entirely from their own undertaken unless the economic/social resources, or seek donor assistance for the justification is demonstrably strong. Third, rehabilitation of infrastructure let deteriorate. PMCs need to increase mobilization oftheir own Undermaintenance of facilities significantly domestic resources to maintain the high priority impairs domestic economic performance and assets. growth; most of the consequences impact directly on users of the transport system and the community at large. 2.67 Cost-Recovery and Maintenance Funding. Across all five PMCs, for the whole transport sector (roads, marine and aviation), current total tax and fee revenues from transport -59 - CHAPTER 3 MANTENANCE DEVELOPMENT A. COMPONESm OF A estimated costs of current and projected MAIJTANCE SYSrM maintenance and rehabilitation needs. 3.1 A basic reappraisal is required of the 3.4 The Finance component provides manner in which each PMC manages its information on current levels of funding transport infrastructure assets, and the level of allocations and/or revenue which can be applied funding which is applied to their maintenance. to the maintenance function, and the options A key part of this reappraisal is assessment of available to change funding levels. Information all substantial existing infrastructure, and from the finance component determines the level identification of those assets which warrant at which maintenance and rehabilitation needs maintenance and are sustainable. Once this is can be satisfied, and the most effective achieved, it is important to direct attention to the maintenance strategy (including standards) effective provision of maintenance for warranted conditional on the feasible levels of funding. assets. In this Chapter attention is focused on this latter task, and specifically on the 3.5 The Delivery component contains the organization of an effective maintenance system. mechanisms through which maintenance services are provided. It involves the assignment of 3.2 There are three main com2onents of a maintenance responsibility and the means of well designed maintenance system: executing the required maintenance effort, including the private sector. * management * finance 3.6 These components are addressed in turn * delivery below. The roles of these components in the provision B. INsTITmoNAL DEVELOPMENT of maintenance are illustrated in Figure 3.1. Management Information 3.3 The Management component provides the strategic and operational decisionmaking 3.7 The development, implementation and framework to maintain the transport system in ongoing maintenance of decision-driven an effective condition. It comprises a data base management information is a fundamental containing essential information on the operating requirement for effective and efficient condition of the stock of assets, and current and management of a country's transport forecasted demand for use of those assets. Its infr&structure. Such information needs to be function is to produce priority programs and tailored to each PMC, the rature of its maintenance task, and the institutional, technical, managerial and financial constraints. - 60 - F1gure 3.1: MANAGEMENT, FINANCE AND DELIVERY ROLES FOR MAINTENANCE .v I F- --------- ~~ ~ Exatng - Inventory status Demand F Existing Financing | of of for Finance Options Assets Assets Transport Allocation , L…~~~~~~~~~Services for L evc Main°tenance Revised Allocation Reviev of Exieting Infrastructure 4I |Identification of Warranted Assets I' V~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ r- - - - - -- - - - - - - - - --- | Nintenance Needs Reeidual I .IMaintenance I Priority Program Deficiency and Costa -~~~~~~ I . .. ....... . ... i , Provin-ci-al-- Central or F Government Local Government I I -----1 1 1 Contract F Force I j Community Force | Contract Seiviceas Account | Based I Account S ! services I ~~~Labor I -61 - 3.8 At present there is little information Box 3.1: CONTINUrrY OF which can be used to understand and resolve the ORGANIZATIONAL ARRANGEMENTS problems of maintenance and rehabilitation of infrastructure associated with the road, maritime Because of the almost exclusive reliance an avato moe in eahons rp public sector force account operations and aviation modes In each PMC's transport (particularly in the roads subsector) the available sector. For example, only a rudimentary professional and mnanagement resources have been inventory currently exists for the various directed to the administration and nanagement of construction and maintenance workforce and the transport agencies and departments in each design and management of projects. The PMC.36 The capacity to develop a perceived priority for staffing assistance has been programmed approach to maintenance, from un these areas and selection of expatriate contract which the financial and institutional requirements officers has been based on the requirements. can be derived, is entirely lacking. The size of the day-t -day worldoads coupled with the short-term length of the contct (normally 2 years), and the varying backgrounds 3.9 In these circumstances, ministries are not and experience of contract officers, results in littlo able to advise Governments adequately of opportunity or motivation to implement or maintenance priorities and the efficient level of providn an onagement arangements dosignod to funding for maintenance and rehabilitation. approach. Angements whichareinitiated tend, Neither are they in a position to quantify the therefore, to be designed to meet immediate needs of leser bdgetry suportand to be base on selective experience. They are consequential effects of lesser budgetary support vulnerable to changes in senior naogement to naintenance, in order that a more rational personnel. consideration might be made by Governments of the allocation of its scarce budget resources. management of development of the transport 3.10 Government departments and agencies in system.3 some PMCs are subject to problems arising from frequent changes in executive and professional 3.12 Size, Government Responsibilities and personnel, particularly in the relatively large Management Development. The provision of number of expatriates employed through bilateral sound policy advice on transport matters, and staffing assistance schemes (see Box 3.1). The the management of the system, are primary roles development of consistent management policies of departments. Skilled management porsonnel and stable organizational practice and priorities. are scarce, although the requirements for is limited by these conditions. Management maintenance management are small. The information arrangements need to survive alternatives for securing management capacity frequent changes at management level, and be range from training indigenous staff to using tailored to the limited resources and needs of the expatriates-and phasing over time from the PMCs. External technical assistance needs to be latter to the former via counterpart training. provided in a way that supports this approach. While the cost-effectiveness of these alternatives needs to be examined, the question turns as 3.11 If the institutional strengthening needs of much on commitment to maintenance organizations are to be addressed, changes will management as a priority activity as availability be required in the emphasis of technical of personnel. assistance, and in the selection of personnel through staff assistance schemes. Effort needs Financing System to be directed to the development of organizational structures, systems and 3.13 In all PMCs, existing financial procedures which support the effective arrangements for the maintenance of road and aviation transport infrastructure assets are -62- unsatisfactory; in several PMCs the same applies within provincial governments. It appears to to marine infrastructure. The level of financial have resulted in a situation where the provision is low and unrelated to an assessment responsibility for some transport infrastructure of requirements. The stock and standards of (roads and wharves) is ill defined. Given the existing infrastructure are excessive and non- small size of the PMCs, the scarcity of sustainable, yet the tight maintenance budgets supervision and semi-skilled personnel, and the have not resulted in essential/priority lumpiness of mechanized maintenance, the maintenance requirements being identified and viability of effective provincial local funded. Agencies are unable to establish maintenance, in excess of routine tasks, is highly warranted maintenance financing needs nor the questionable. economic and financial implications of current or alternative funding levels. 3.17 The second major policy consideration is the extent to which private sector capabilities 3.14 Cost recovery for all modes is Iow in all can be developed and used in the deliv-ry of PMCs. Yet despite maintenance needs, little maintenance services through, for example, attention has been given, or incentives provided, period maintenance conttracts. The use of scarce to agencies to increase user charges. public sector professional and management resources in the day-to-day organization, 3.15 Financial and accounting systems within direction and supervision of maintenance road, aviation, and maritime transport activities is a significant misallocation. Those organizations do not provide comparability resources can be more effectively employed in between budget allocations and actual costs of the programming, scheduling, budget allocation program delivery. The present line budgeting, and cost control functions associated with the does not assist monitoring and cost control. overall management of maintenance needs. The relative efficiency of using the domestic private Maintenance Delivery Systems sector hinges to a large extent on scale; effective completion needs to prevail among private 3.16 A number of options are available to contractors. Where the domestic private secto. Governments for the way maintenance services is limited, overseas private contracts can be used are provided "on the ground". The cost- for major maintenance and rehabilitation; in this effective choice of approach relates to case, substantial fixed costs, such as plant and departmental organization and responsibilities, equipment relocation, are involved and the system and training needs, and financing contract period needs to be adequate, say three methods. There are two main dimensions which years. Opportunities exist to use comnvunify define options for delivery: assignment of based resources for undertaking routine, labor responsibility by level of government, and intensive, maintenance, as has already been assignment of tasks between the public and initiated in some PMCs. This work could be private sectors. The primary consideration is under the direction of either central or (where it the division of responsibility for asset exists) provincial government. Recognition of maintenance between the national government these requirements has important implications and, where applicable, provincial governments. for the targeting, and nature, of technical Government policy in several PMCs is to assistance, training, and organizational devolve responsibilities for the maintenance of strengthening in both the public and private local transport infrastructure (roads, wharves, sectors. airfields) to provincial administrations. In these countries, the piocess of devolution has been slow and constrained by the lack of financial, institutional and technical resources and skills - 63 - C. IMPLEMENTAnON REQU REMES 3.22 The primary role of government organizations should be to meet needs for 3.18 In this section conclusions and strategic planning and overall operational recommendations concerning maintenance of management which support efficiency of the infrastructure are considered under four transport system. Priority should be given to headings: management, finance, delivery and applying scarcc nianagerial and professional external assistance. resources in these areas, rather than to day-to- day construction and maintenance Management implementation tasks. 3.19 Improved management arrangemen"; are Finance needed for effective management of infrastructure. These arrangements must be 3.23 A major review of maintenance recurrent tailored in parallel with the development of budget funding requirements is necessary, based managerial and administrative capacities in upon those assets identified as warranted. This government departments/agencies. review must be undertaken after the recommended program of reassessment and 3.20 Given the general consistency of needs in rationalization of existing infrastructure. the PMCs, and their similar resource limitations, as far as possible, common approaches should be 3.24 The operation and maintenance costs of developed and implemented in each PMC. This all proposed new transport infrastructure would aid efficiency in the development of core investments should be explicitly identified as capacities, and internalize economies in the part of all project assessments. Provision should provision of support for their preservation and be made in user charges, and in recurrent extension, as required. It would also provide a budgets, for the net increases (or reductions) standard base against which to assess the resulting from accepted projects. requirements for donor assistance in the management of transport infrastructure in the 3.25 Clear financial targets should be set for region. In this regard, a "pilot" management cost recovery to cover maintenance and asset maintenance systum should be developed, tested, replacement. The mix of direct user charges, and its adaptability ai1d transferability assessed. taxes and duties for each mode to achieve these targets should be established. 3.21 In subsectors (such as aviation and maritime) where management functions include 3.26 Funding of a base level of maintenance revenue raising through e -ct user charges for expenditure through hypothecation of specified services provided, man: inent arrangements revenues should be assessed where there is should be commercial., oriented, and be persistent under-funding of warranted supported by cost center accouning. This does maintenance from recurrent budgets. Although not necessarily require the creation of "ear-marking" is not a first-best (efficient) government authorities, but will require the approach, since resources should be kept setting of commercial performance objectives fungible for allocation to their overall best use, and accountability procedures by Government. existing financing allocations and their stability For roads, expenditure, service standards, traffic are far from optimal. Under these imperfect volumes, and indirect user charges should be circumstances, "ear-marking" can ensure a reported by program to provide basic consistent base line level of funding, management information for monitoring co=;t circumscribe opportunities for ad hoc redirection recovery and planning. of resources to non-programmed purposes, and - 64 - provide a demonstrable commitment for should be performance based for each attracting donor assistance with maintenance. community area. In reducing public sector force account operations, Governments will need to 3.27 Financial and accounting systems within enunciate maintenance po!icy and intended transport organizations should be designed to arrangements, establish stable funding for provide information on the actual costs of maintenance activities, develop weli designed program delivery. The introduction of very contracting procedures, and be able to open-up basic prograrm budgeting systems should be a sufficient size of activities to attract effective considered in place of, or as an adjunct to, the competition among domestic (or external) private present line budgeting systems. contractors. Transition to contract-based maintenance will need to be phased-in over time, Delivery commensurate with the growth in domestic private and/or overseas involvement. 3.28 If devolution does proceed in some PMCs, there should be clear identification of the External Assistance Implications assets to be transferred, the financial arrangements for their transfer, the plans for 3.31 External assistance for transport strengthening provincial agencies, and the time infrastructure, in general, should be redirected frame within which transfer is to occur. from new investmeut to the maintenance and Appropriate arrangements may include the use rehabilitation of existing priority assets. This of agency agreements for the maintenance of will improve the productivity of assistance, and identified assets, as ap initial step to the full lower total transport system costs. transfer of responsibilities. Responsibilities should not be transferred without the concurrent 3.32 There is a need for donors to review implementation of these activities. policies regarding assistance for asset maintenance,^' in parallel with efforts by PMC 3.29 Given the small size of the PMCs, and Governments to rationalize their existing stocks the scarcity of management, technical skills, and of infrastructure, and to develop effective other resources in the central government (and in maintena e! practices. Incentive packages for many PMCs, in the private sector as well), recipient governments (such as co-pay or devolution of responsibility f3r such areas as matching arrangements) should be designcJ, maintenance should be seriously reconsidered. tying assistance for maintenance funding to the development of improved managemeiat, finance, 3.30 The ,ecommended emphasis on the and effective delivery of asset maintenance. higher level management functions in government transport organizations goes in hand 3.33 Present approaches to technical with government placing increased reliance on assistance, staff support and training, within the private sector (or local community) to public service organizations responsible for provide maintenance services. Three main transport infrastructure, require re-examination. options should be considered, depending on the The present priorities in training programs task and available capacities: domestic private appear to be directed to the development of contractors (where mechaniz-d equipment is professional, technical and in-the-field work required, for example, grading), community- skills to support the force account operations of based services, (low skilled labor routine Departments. A reassessment is required of operations) and overseas organizations (medium- these priorities, with emphasis on management, term management and rehabilitation). Given the planning, information systems operation, and vulnerability of community-based operations to contract management (including specifications, high cost and unreliability, payment for the work procurement and supervision). Stability and -65 - continuity in organization development, D. ACTION PLAN managerial arrangements, training, information processing and work practices, are crucial in 3.35 An action plan for each PMC establishing effective organizations. 'Twinning' Government to develop a program for arrangements' between competent mature maintenance management should include the overseas organizations in neighboring countries, following core steps: with training experience, and similar responsibilities to PMC public sector road, port (1) Develop a Transport Infrastructure and aviation agencies, should be adopted. In Maintenance Strategy addition, PMCs should be supported in the establishment and/or strengthening of regional * Reach agreement with donors on key road, maritime and aviation infrastructure deemes of a mantenace trategy professional associations, to enable exchange of covering management, finance and views and experience on common problems. delivery, based on key criteria (including The small size of the potential membership economic value and social accessibility) should, on balance, assist their usefulness. which anable facilities to be assigned a Several of these approaches have been priority; successfully adopted by national airlines in the South Pacific region. In addition, PMCs * Seek joint endorsement with donors of involvement and representation with well guiding principles for maintenance: established Pacific, Asian and international organizations should be expanded. For example, (a) Maintenance, rebs bilitation and PMC participation through area groups in such reconstruction of existing infrastructure organizations as the Road Engineering should be limited to those assets for Association of Asia and Australasia (REAAA), which they are warranted on a priority the Permanent International Association of Road basis. Congresses (PIARC), International Civil Aviation Organization (ICAO) and others should (b) External assistance should be conditional be supported. To some extent transport agencies on an agreed timetable of progress on of the PMCs do participate already in these improvements in management, finance associations, however, a more proactive role and delivery arrangements for asset should be taken. maintenance. 3.34 Donor assistance for establishing major (c) New investment in infrastructure should plant workshops and road inaintenance units only be undertaken if its economic merit within road authorities, and the direct provision exceeds identified priorities for of plant and equipment, should be reviewed. maintenance and rehabilitation of existing Ways should be deviseu to provide assistance to assets. local private enterprise to develop the financial capabilities, business management, and technical (d) Primary financial responsibility flor the skills necessary to take up maintenance and maintenance of infrastructure should be related functions. Similarly, for those PMCs assumed by each PMC, and be geared to who insist on pushing ahead with devolution of an agreed timetable of cost recovery maintenance, training schemes will need to be targets. available to provincial government and community based maintenance units. (2) Establish a Transport Infrastructure Maintenance Management Information Base - 66 -__ * Identify key data needs tied to priority (4) Review Maintenance Funding management tasks. Arrangements * Keep data collection to crucial core needs * Assess alternative direct and indirect user wnd commensurate with institutional charges for revenue potential and fiscal capacity. efficiency. (3) Review Public Sector Management * Establish feasible cost recovery targets Responsibilities and Expand by mode. Opportunities for Participation of Private Sector * Examine alternative financial arrangements with donors for joint * Establish requirements, including capital/maintenance expenditure contract and work specifications, and a programs. work program, to facilitate transition from force account to contract and community-based private sector arrangements. * Identify areas most attractive for competitive bidding. - 67 - Endnotes I These surveys are presented in Volume Two of this report. 2. Details of the maintenance situation in each country are provided in an Annex to each country sector survey. 3. Fiji has not been included in this present examination because of its special characteristics: Fiji is considerably larger than the other PMCs, and Fiji is currently addressing maintenance needs. Nevertheless, the issues considered in the other five PMCs are similar to those in Fiji, and many of the general findings for these countries are transferable. 4. Costing of the organizational and personnel components of implementing the strategy is not developed in detail, although orders of magnitude are indicated. This task is important and should be addressed in the design and selection of a specific implementation plan to assess cost- effectiveness of alternative approaches, and to ensure that positive net benefits are achieved. 5. Difficult pressures arise in the use of external assistance resources to maintain infrastructure. Premature and higher cost rehabilitation and reconstruction are often substituted for effective and systematic routine and periodic maintenance. Aspects of external financing, in effect, foster a form of 'moral hazard' for governments, that is, readily available support for major rehabilitation reduces (and may eliminate) the willingness of PMC Governments to commit an adequate share of their own resources to proper maintenance of their assets. 6. Substantial resources have been applied by the Bank and participating governments (especially since 1971) to the collection and analysis of road and vehicle operating cost data, culminating in the Highway Design and Maintenance Standards Model (World Bank, 1987) and the companion Expenditure Budgeting Model (World Bank, 1987). The effects of inappropriate and/or under expenditure on proper maintenance of roads on total road transport costs can be quantified, provided the required data are available and the application is within the calibrated range of the model. 7. Choice of a specific impl:mentation strategy will need to be based on an assessment of the relative costs of alternative organizational arrangements. The most cost effective approach (for example, training indigenous staff and/or use of expatriates) will turn on current organizational arrangements, the scarcity of skilled local management, and the availability and long-term costs of training and expatriate persornel. Nevertheless, in all PMCs the personnel requirements should be modest; substantial nat increase in public service employees is not needed. For the basic tasks of establishing and maintaining an inventory cf facilities, their condition and use status, functional expenditures, and the preparation of regular maintenance reports, staff requirements are modest; order of 1-2 semi-professionals for each mode should be adequate. S. Data collection and processing is costly, especially with the scarce professional and management skills available in the PMCs. The value of additional information needs to be weighed against the additional costs. At the same time the reliability/accuracy of data should be indicated. -68 - 9. The density of roads (length in relation to land area) should be interpreted with caution as the proportion of land which is inhabited and/or used in productive activity varies significantly across the countries. In addition, in the PMCs, a majority of population typically resides in small villages around the coast (circumference) of the islands. This geographical form tends to increase road length per capita in these countries. 10. This ratio is low in the Solomon Islands, reflecting the export of timber from private and informal facilities (no inventory or replacement values are readily available for private transport infrastructure), and possibly relatively efficient use of port infrastructure. The high ratio of marine assets to trade in Tonga results from the particularly large port investment at Nuku'alofa. The ratio is mnginally higher in Vamuatu, and is attributable there to two international ports, one at Santo, which is primarily used for exports, and one at Port Vila which is used mosly for imports. 11. The ratio of passenger movements to GNP in Western Samoa is relatively high as a result of the large social travel involving American Samoa, and links with Western Samoans resident in the USA, Now Zeald and Australia. Nationals from Tonga in these three countries also contribute to the similarly high ratio for Tonga. Tbe tourist industry in Vanuatu has recovered rapidly in recent years and the ratio of passenger movements to GNP is expected to become higher than indicated in Table 2.4. International aviation is less important to the economies of the Solomon Ilands and Kiribati. Domestic aviation is relatively more important in the Solomon Islands and Vanuatu. 12. This is exacerbated by the exclusion of two deteriorated (and unlikely to be redeveloped) facilities trom the inventory of marine assets in Western Samoa. 13. The increase in infrastructure with GNP across the PMCs is consistent with the normal observation that transport (and complementary infrastructure) services are income elastic (for example, taking a cross-sectoral view of the PMC region, if gross national product increases by 10 percent, transport activity increases by more than 10 percent, say 12 percent), where existing infrastructure capacity is warranted and satisfactorily utilized. 14. Where no data exist, estimates are based on evidence from discussions and limited field inspections. In some cases there is insufficient information to permit even broad estimates to be prepared. 15. The es:imates are based on assembled inventories which are likely to omit some infrastructure (especially in the marine subsector). Therefore the costs of rehabilitation of all existing infrastructure is likely to exceed these estimates. - 69 - 16. Strictly, the overall optimal maintenance level should be jointly determined with the corresponding optimal design standard; it is associated with the minimum total system costs of construction, maintenance and user costs (under prescribed user charge and vehicle/v'ssel/aircraft size regimes). In this study the focus is on maintenance per se; the question of 'best' standards, vehicle sizes and compliance are important ones in the PMCs, but are set aside here. It might be noted that it is not immediately transparent in the environment of the PMCs, whether undermaintenance, if persistent, should be matched by higher or lower standards relative to those associated with assessed effective mauntenance. Such choices should be made based on total transport system costs, i.e., infrastructure plus user/operator costs. For roads, with the low traffic volumes in the PMCs, the small and uncertain benefits, and the limited implementation effectiveness, it is likely to be safer to keep to lower (i.e., unpaved) standards. This issue is discussed more below. 17. The method used to determine assessed maintenance expenditure of road, marine and aviation assets should normally be applied to the replacement cost of assets net of once-off costs that are incurred for green field situations, such as land acquisition and site preparation. Data limitations do not permit such costs to be treated systematically in the present analysis. Land acquisition costs are generally not included in the replacement value of assets. The inclusion of costs such as site preparation are unlikely to result in substantial distortion, and needs to be seen in the context of order of accuracy of the existing data and analysis. 18. Derivation of these costs for each country, including vehicle operating costs for road use, is described in the Maintenance Annex to each Country Sector Survey presented in Volume Two. 19. The Solomon Islands has an environment (terrain, soil, weather and island geography) closest (among the PMCs) to Vanuatu, and hence provides the most suitable data at hand for reestimating road maintenance costs for Vanuatu. The application of unit maintenance costs from the Solomon Islands to Vanuatu reduces average road maintenance costs in Vanuatu to 44 percent of their reported level. 20. In thirty-four Sub-Saharan African couia-i. (over the period 1986-88) road maintenance expenditure was only about 0.2 percent of GNP; see World Bank and Economic Commission for Africa (1990). 21. The total length of sealed road in Kiribat, at 36 km, is very small in absolute terms, and relative to the other four PMCs. This illustrates the 'roblem of the different mix of road types (sealed vs. non-sealed) across the five PMCs. 22. The variation in the ratio of assessed maintenance to asset replacement value for marine assets in the PMCs is attributable to the different composition of marine assets, in particular the share of asset value attributable to equipment, for which maintenance costs are a much higher proportion of asset value, than for fixed assets (wharves, jetties, and buildings). - 70 - 23. Because of limitations in the quality and unknown reliability of data, aggregate data for the five PMCs may balance some of the irregularities in data for individual countries, and thus be a better basis ft -onclusions. Nevertheless, inter-country comparisons ate made where clear differences are ev ,nt. It should be noted that these findings are suggested as indicative; they are based on a limited database and are hampered by the uncertain applicability of relationships, such as those in the roads subsector, between maintenance and life-cycle costs, derived elsewhere for activity levels (such as traffic volumes) considerably higher than generally occurs in the PMCs. Thus, for example, the same economic lives for infrastructure under current and assessed maintenance expenditure have been adopted for all countries. It can be expected that the extension in economic life in moving from existing to assessed levels of maintenance expenditure would be less where current maintenance is a higher proportion of assessed maintenance (for example, Kiribati). 24. Estimates of changes in total transport system costs, including user costs, are only made for roads; for marine and aviation infrastructure costs alone are considered. 25. The only exception is in the road sector in Vanuatu where the data suggest that increased expenditure on maintenance exceeds the reduced equivaler.t annual capital cost of infrastructure. This anomaly stems from the particularly high maintenance costs reported in Vanuatu; use of unit maintenance costs levels established for the Solomon Islands would reduce the incremental annual cost of road maintenance in Vanuatu to US$3.1 million; the ratio of the saving in the equivalent annual Lapital cost to the incremental annual cost of improved maintenance rises from 1.0 to 1.4, which is similar to the ratio of 1.5 for the Solomon Islands. 26. The lower ratio may be attributable in part to adoption of a uniform differential in the economic life of infrastructure with current and improved maintenance; the lower proportion of assessed maintenance currently expended in these two countries would justify the use of a lesser differential in the economic life of road assets with current and good maintenance for Kiribati than the other countries. This would result in a lesser saving in the equivalent annual capital cost of infrastructure in Kiribati, and a lower benefit ratio than the stated value of 6.4. However, even with such an adjustment, the benefit-cost ratio for increased road maintenance in the Solomon Islands and Vanuatu are still likely to be lower than for the other PMCs. This is in contrast to the larger gap between actual and assessed road maintenance in these countries (35 percent in Solomon Islands and 17 percent in Vanuatu). Since it would be expected that the larger the gap, the higher the incremental benefit cost ratio, this finding reinforces the view that there is, in aggr gate, over- investment in these countries; traffic volumes appear low in Solomon Islands and road construction and maintenance costs high in Vanuatu. 27. Ibhis suggests the need for better information on the road maintenance-(re)construction cost trade off in conditions where road deterioration is dominated by the physical environment (rainfall, soil conditions and so on) and not traffic loads and volume. In addition, low traffic volumes also raise the question of appropriate treatment of user benefits; for low volume roads basic access at acceptable reliability may dominate conventional user cost savings (predominantly in vehicle operating costs); valuation of reli:!ble access by willingness to pay poses different empirical problems in the PMCs where surrogates such as land value differentials typically cannot be examined. -71 - 28. Transport user costs savings from improved maintenance imply that the level of warranted (optimal) maintenance is higher than would be the case if infrastructure capital and maintenance costs alone were considered. (See Box 2.1). 29. A further source of benefit from improved maintenance may stem from lowering user costs on feeder transport links. This may have a significant effect on the incentive to invest in crops and may increase producer surplus. However, reductions in the total cost of delivery of agricultural output overseas may be small unless there are avoidable constraints elsewhere. This requires analysis of the link between transport costs (including maintenance and user costs) and marginal agricultu.al production. 30. The estimates of vehicle operating costs in the PMCs should be regarded as indicative. Total vehicle operating costs for travel on sealed roads, in good condition are used as a datum from which vehicle operating costs are estimated for sealed roads in fair and poor condition and unsealel roads (in each of the three conditions) using road roughness indices and relationships in the RTIM model. These computations are set out in the Maintenance Annex to each country sector survey presented in Volume Two. Variations of vehicle operating costs with road roughness are estimated for fuel oil, tires and maintenance. The resultant vehicle operating costs are summarized in Box 2.1. Traffic volumes are low in all PMCs and representative volumes are used in estimating the total vehicle operating cost savings. More detailed work is needed to refine these estimates. 31. Examples include failure to mow airfield grass, at a cost cf US$220 per month, and the lack of a replacement wiiidsock, with a value of US$75. The economic loss associated with a foregone (or deferred) flight, due to airfield slosure, can be substantial. For example, in the Solomon Islands, the air fare for a one-hour flight is US$51 greater than the tariff for the alternative 24-hour ship journey. The (minimum) premium which air travellers are prepared to pay over the fare for sea travel is high in proportion to the cost of some critical elements of airfield maintenance. 32. The annual net savings in life cycle costs estimated here far exceed the guideline figure of 5- 15 percent savings in annual maintenance budgets that can be expected with the adoption of effective information data bases (for roads) suggested in World Bank (1989). Even if savings are limited to 10 percent of annual maintenance, the expected net gains from warranted maintenance become of the order of US$300,000 (roads), US$65,000 (marine) and US$50,000 (aviation). The fixed costs of information system development, implementation and initial support for the small sLze PMCs can be expected to be in the range of US$500,000 (or $35,000 p.a.) for roads, and US$150,000 (or US$10,500 p.a.) for ports and aviation. It follows that conservative estimates of the upper limit of justifiable additional annual personnel costs are US$265,000 for roads, US$54,500 for marine, and US$39,500 for aviation. These limits should be well witnin both the economic cost, and financial salary levels, for two additional local personnel. 33. The desirnhility of earmarking is open to question. It ensures a stable source of finance but may jeopardize supplementary government funding when expenditures in excess of the prevailing (earmarked) fund balance are warranted. 34. The structure of road user charges and other fees/taxes for road cost recovery is more complex in such situations. On benefit taxation principles, some form of property tax would be suitable, however, its practical application in the PMCs is very limited. 72- 35. Data on revenue from duty on the import of marine and aviation equipment and fuel are not readily available. The revenue is likely to be low, however, since vessels and aircraft, and fuel used for aviation and for interisland shipping, are usually subject to lower (sometimes zero) rates of duty than apply to the road subsector. 36. As noted earlier, there are important exceptions, for example, the Solomon Islands Port Authority. 37. These needs involve only small personnel requirements (of the order of 2 semi-professional staff in each modal area). The provision of professional and technical support for the day-to-day delivery of coilstruction and maintenance services is important, and needs to complement the management task. However, given existing institutional weaknesses, delivery needs should be considered a secondary priority. 38. Donor support for recurrent cost financing, including maintenance, has been advanced under certain circumstances, recently. See AIDAB and MERTNZ (1991). 39. See World Bank (1990). - 73 - PART m COMPARISON OF REGIONAL TRANSPORT COSTS _ 75 - CHAPTER 1 INTRODUCTION A. BACKGROUND AND OBJECTVES reflect the notion of remoteness. For the purposes of this study, remoteness is more 1.1 As described in Part I (Overview) above, usefully thought of in terms of accessibility (of the island States of the South Pacific region are the PMCs) to export and import markets. On very small and widely dispersed; they are also this basis, the distance to, and size of, these remote from their export markets and their markets become key factors. At the same time, sources of imports. It is often claimed that they remoteness and accessibiiity hinge on the cost, face inordinately high international transport time and ease of interaction, especia;.y trade. It costs which constrain their economic follows that generalized transport costs best performance and growth. reflect these concerns. In particular it is the set of basic factors that shape the structure, and 1.2 This Pa i III provides a comparative affect the ievel, of transport costs to countries, analysis of the costs and levels of service for sea such as the PMCs, that best capture the issue of and air transport to and from the six PMCs and remoteness. Papua New Guinea (PNG).' The objective of this analysis is to assess how transport costs 1.4 Factors Affecting Transport Costs. between the PMCs and their major trading Numerous factors influence the level of transport partners (including major sources of tourists) are costs and service. In the context of the PMCs, affected by the remoteness of their regional some of the most important are: location and the characteristics of their transport markets. The underlying hypothesis to be tested - distances to and from major markets; is that the PMCs experience intrinsically higher - route networks; international transport costs compared to similar - volume of trade (freight and passenger island States in other regions. movernents); - transport market structure and regulation; 1.3 Remoteness, Accessibility and - the efficiency of infrastructure; Transport Cost Structures. The rotion of - directional imbalances in traffic remoteness is not easy to operationally define. - transhipment A large distance from neighboring countries is - institutional arrangements and constraints; not sufficient. For example, Europe is some and 4,000 miles from North America, but neither - service frequencies and routes. continent is generally viewed as remote. On the other hand, there are islands (for example, 1.5 Several of these factors are interrelated; Chatham Islands and the Seychelles) reasonably notably traffic volumes and levels of service close to their adjacent mainland areas (New interact through vehicle scale economies (ship Zealand and East Africa) which adequately and aircraft size); low volumes of cargoes and -76 - passeng'r movements involve a higher level of cost model, supplemented by statistical analysis unit transport costs, but this level can be of actual rates. Air fares are assessed by a lowered by reducing service frequency and statistical comparison of international air fares increasing vehicle size, for the PMCs with air fares on similar routes in the Pacific region. 1.6 The principal independent factors that can be expected to shape transport costs for the Comparator Countries PMCs are distance, volume of traffic and, especially for shipping, route and voyage 1.11 Comparator countries were selected on networks. Therefore, the analysis of transport the basis of their similar size and geographic costs for shipping and air services relevant to the characteristics to the PMCs. Two groups of PMCs, is conducted with emphasis on these comparator islands are used: factors. (a) Non PMC islands in the Pacific Ocean (see Figure 1.1): B. APPROACH * Society Islands (Tahiti) 1.7 The approach to determining whether the * New Caledonia PMCs experience "abnormal" transport costs involves testing a series of hypotheses relating (b) Similar islands in the Indian Ocean (see transport costs to the volume of traffic and Figure 1.2): distance carried, using data that are readily available. * Mauritius *: Reunion 1.8 The associated analysis addresses two key * Seychelles questions: * Christmas Island * : :Cocos Islands (a) Are transport costs to and from the * Comore Islanids PMCs comparable wvith those to and * Maldiwes from other small Island States, and 1.12 The two Pacific Island comparator (b) Are they comparable with those to and countries are administered and significantly from other countries? supported by France and were included, in part, to determine whether this resulted in materially 1.9 The assessment covers both sea freight different transport conditions. In the Indian transport and air passenger transport. First, Ocean, Mauritius and Reunion have similarity in transport tariffs and services to/from the PMCs their economies to Fiji. Christmas Island and are compared with those for a set of similar Cocos Islands are supported by Australia, and "comparator" island countries. Next. transport are generally regarded as remote. The costs to/from all the island countries are Seychelles, Comoro Islands and the Maldives compared with transport costs to/from a larger have geographic and economic similarity to the sample of countries and regions. PMCs. 1.10 Different approaci;es are adopted for the 1.13 Socio-economic characteristics and trade shipping and aviation subsectors because of their flows for the PMCs, and comparator countries, different operational and institutional structures. are presented in Annex 1, in Table 1 and Sea freight rates are assessed using a shipping Table 2, respectively. The PMCs are highly FISme . IL PACIFIC OC3i IS&D-NcAMO hhZ FEDERATED STTES r . E * ' §,;OF KOCRONESIA % *.-- *S -*w', , *;** a~~~. * TWL * sKEA a,N,. LLANDS 'TLi CWEAU* . . ..* , : SOUIA1 P OCEAN *s t sof SEA _v rD GUINEN A,Al CORAL Pap~~~~~SUT PACIFI .5 PEMSS F~i~1 * 'CAN 10MW 0 aiic-ie Ccxmrie I'Cs I CWatoOK .:Iswd - 78 - dependent on imports for their domestic requirements and primary commodities dominate exports (for example, sugar, copra and coconut oil). As a result, the majority of freight flows are imbalanced; flows to the islands dominate, are non-bulk, and containerized; flows from the PMCs are primarily bulk. The major trade partners of the PMCs, and comparator countries, are shown in Annex 1, Table 3 and Table 4, respectively. With the exception of Fiji and Vanuatu, the tourism industries in the PMCs are relatively underdeveloped. The majcr tourist movements, for the PMCs and comparator countries, are shown in Annex 1, in Table 5 and Table 6, respectively. 1.14 Structure of the Assessment. The assessment of regional transport costs is presented below, first for shipping (Chapter 2), and then for aviation (Chapter 3). In each case a description of services, for the PMCs and comparator countries, precedes the analysis of costs and rates. (Descriptions of the various sources of data used in these assessments are provided in Annexes 1, 3 and 4). - 79 - IP!~~~~~~~~~~~~~~~~~~ .4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~04 I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ S. L.';U | | gne1A ji~~~~~~~~~~~~~~~~~~~~~~. - 80 - CHAPTER 2 ASSESSMENT OF SHIPPING SERVICES A. S}UPPING SERVICES TO mm Pacific Isblands ports close to their itineraries for SOUTH PACFC which the cost of deviation is low. Passing services are listed in Table 2.1 and shown on 2.1 The scheduled international shipping Figure 3 of Annex 2. Generally, passing services to the small island nations of the South services call at only one or two island ports; the Pacific can be divided into three categories:2 size and strategic location of Fiji and Tahiti make them the calls most commonly included. a passing services * dedicated extra-regional services 2.3 Since the vessel size on these routes is e regional services. determined by the demands of the major end-point markets, it is not surprising that these Passing Services services employ the largest vessels seen in the Pacific Islands trades: several of the services 2.2 These services are primarily on major (for example, OSCL, Euroceania, Barbican) use trans-oceanic routes, but include wayport calls at vessels with capacity in excess of 1,000 twenty- foot container equivalent units (TEU). Table 2.1: SOUTH PACIFIC ISLANDS "PASSING" SERVICES, 1991 Service To/From Island Calls Barbican S Africa/Australia to S America Fiji, Solomons, PNG, Tahiti Blue Star NZ/USA Fiji PACE (West) Aust/USA New Caledonia,Fiji SeaPac /a S.America/ Singapore Tahiti, Fiji, PNG Tasman Asia NZ/ Asia Fiji, New Caledonia Euroceania /b USA/Aust-Far East Tahiti, New Caledonia, PNG TransPacific Australia/ North Asia PNG Ocean Star Container Line USA/Australia Tahiti La Compania Chilena de Navegacion Interoceania (CCNI)/Compania Sud Americana de Vapores (CSAV) lb Compagnie Generale Maritime et Financiere (CGM). Source: Industry publications and interviews. - 81 - Dedicated Extra-Regional Services services are set out in Table 2.2 and shown in Figures 4 and 5 of Annex 2. 2.4 These services are either provided exclusively, or significantly tailored to cater, for 2.5 Services are assigned to this category if the trading demands of the South Par fic, and South Pacific Island trades are judged to be link the countries of the region to Asia, Europe essential to the viability of the service. For and the Americas. Dedicated extra-regional nearly all dedicated services from Asia and the Table 2.2: PACIFIC ISLANDS-DEDICATED EXTRA-REGIONAL SERVICES, 1991 Service To/From Island Calls Bali Hai N Asia Fiji, Western Samoa, American Samoa, Kiribati, Tahiti, Tonga, New Caledonia, Vanuatu, Solomon Island Kyowa E Asia/N Asia Guam, Solomon Islands, Vanuatu, New Caledonia, Fiji, Western Samoa, American Samoa Paradise / N Asia PNG EPACSA L Europe Tahiti, Fiji, New Caledonia SOPAC /c N Europe/UK Tahiti, Fiji, New Caledonia, Solomon Islands, PNG South Pacific Line La Europe Tahiti, New Caledonia, Vanuatu, Solomon Islands, PNC New Guinea Pacific Line East Asia/S Asia PNG, Solomon Islands, Vanuatu, New Caledonia Blue Star Line (PCIS) USA Tahiti, Tonga Polynesia Line USA Tahiti South Seas Steamship Co USA Tahiti Hawaii-Pacific Lines USA Western Samoa, Tonga America- Niugini USA PNG La Nippon Yusen Kaisha (NYK)/China Navigation Co (CNCO)/Mitsui-OSK Lines (MOL). L2 NedLloyd: Westabout only. /c Bank Line/Columbus Line. La Polish Ocean Lines (POL). Source: Industry publications and interviews. - 82 - USA, the islands trade ^ rms the entire trade line to offer a far wider spread of destinations base. than the actual routes plied by its vessels would suggest. Services provided by the major lines 2.6 In the case of the European connections are supplemented by those provided by a number (SOPAC, EPACSA and South Pacific Line), of smaller operators using conventional vessels. matters are not quite so clear cut. In each case, the service provides only a southbound (that is, 2.9 Pacific Forum Line (PFL) provides the import) service to Lhe South Pacific. Return most comprehensive network of services, cargoes for Europe are obtained either by offering direct sailings to PNG, Solomons continuing onwards to SE Asia (South Pacific, Islands, Fiji, Tonga, and Western Samoa out of SOPAC), or returning to South America both Australia and New Zealand. PFL also (EPACSA). However, in each case: (a) there provides a transhipment service over Fiji to are few other southbound ports of call; (b) the Kiribati and Tuvalu using a dedicated feeder naming of the service suggests a South Pacific vessel. Chief Container Line focusses on marketing orientation; (c) the service operator services from Australia to the western Pacific, runs separate services covering the trades served offering direct services to a wide range of PNG by the homeward journey. These factors all ports and the Solomon Islands. The Australia suggest that the South Pacific trades are essential Pacific Islands Line service uses an extension of to the continued operation of each service. the Chief service to the eastern PNG ports (for example, Kavieng, Rabaul and Kieta) to directly 2.7 The range of ports covered by dedicated serve Vanuatu and Kiribati. APIL also charters services is much broader than those covered by space on PFL and Sofrana vessels. Sofrana passing services, with most Island States offers direct services to all major destinations receiving direct calls from at least one service. west of Fiji out of New Zealand. It also However, there is a marked disparity between operates vessels out oi Australia to Vanuatu and countries in service frequency: Fiji, Tahiti and Fiji, and serves the Australia-PNG trade by slot PNG are the best served, while Kiribati attracts chartering on the CGM Euroceania service. calls from only the comprehensive Bali Hai service, and even this call is sometimes omitted. B. STRUCTURE OF PACFC ISLANDS Vessel sizes for the main dedicated services are MARKET COMPETIION typically in the 250-500 TEU range, although larger vessels are used in the USA-Tahiti 2.10 Competitive pressure in the liner trades services. serving the South Pacific comes from three sources: competition between dedicated Rcgional Services operators; actual and potential competition for operators making wayport calls; and the 2.8 These comparatively short haul routes possibility of substitution between import link the South Pacific Island States to each other, sources. The balance of these forces differs and to Australia and New Zealand. These from trade to trade, and the intensity of services are shown in Table 2.3, and in Figures competition varies with the size of the market 6,7, 8 and 9 of Annex 2. The three principal concerned. However, no market has been regional operators are the Pacific Forum Line identified-taking all three sources into (owned by a consortium of Pacific Island account-in which a significant prospect of national Governments and New Zealand), Chief exercising monopoly/market power exists. Container Lines and Sofrana Unilines. There appears to be considerable cross-chartering of capacity between these operators, allowing each - 83 - Table . J: PACIC ISLANDS-REGIONAL SIPmNG OATORS, 1991 Service Description Pacific Forum Line Australia-South Pacific service: serves Fiji, Western Samoa, Tonga Paciflc Forum Line Australasia/PNG-South Pacific Service: serves PNG, Solomon Islands, FUi Pacific Forum Line New Zealand/South Pacific Service: servos Fiji, Ton"g, Western Samoa Paific Forum Line Fiji-Tuvalu/Kiribati feeder service Chief Container Lines Australia/PNG-Solomc.b Aust Pacific Wdands Line Australia/PNG, Vanuatu, Kiribati, Now (Chit) Caledonia Sofhna Unilines Australia/Vanuatu, Now Caledonia, Fiji Sofra Unilines West Pacific Service: New Zealand/PNG, New Caledonia, Solomon Islands, Vanuatu Sofrana Unilines New Zealand/Fiji Maasmond Express Lines Australia/New Caledonia, Fiji, Vanuatu Traunslink Pacific New Zealand/Fiji, Western Samoa, Tonga Translink Pacific New Zealand/New Caledonia, Fiji, Wallis & Futuna Islands W Idands Line New Zealand/ Fiji, Tonga Warner Pacific New Zealand/Fiji, Western Samoa, American Samoa, Tonga Warrer Pacific Australia/Tonga, American Samoa Source: Industry publications and interviews. Competition Between Dedicated Operators Kong Swire Group)-compete on the short-haul Australia/New Zealand routes. Competition 2.11 The degree of competition between between these operators is limited by differences dedicated operators depends on the market size in their operational focus: Chief, for instance, and the number of operators offering services. offers services only out of Australia and has But it is also influenced by the institutional strong PNG orientations, whereas Sofrana is framnework within which the services operate. strongest out of New Zealand. As a consequence, only the Brisbane-Lao sub-market 2.12 The 'dedicated' services provide a is at present served independently by all three comparatively stable core to the liner shipping carriers. This is unlikely, however, to be a services to the Suuth Pacific. Three major permanent division of spheres of influence: all lines-Pacific Forum Line, Sofrana Unilines parties are continually making adjustments to (owned by the French Delmas Group), and their operations that have impacts on the other Chief Container Lines (owned by the Hong operators. Chief Container Lines until recently -84 operated an Independent Fiji service; Sofrana has and leave a market, competition from cross- recently contracted its PNG operations, while traders plays a significant role in enforcing PFL has added a second vessel, and expanded its efficient performance, even though their size and PNG services. Competition between dedicated market share are small. operators out of Australia is further affected by the existence of a conference, called the 2.16 The activity of cross-traders has reduced Australia Pacific Islands Rate Agreement shipping rates in the major trades in recent (APIRA), of which all three major lines are years. This has been associated with a members. No corresponding agreement exists in succession of entries and/or exits. For example, the New Zealand trades. the recent entry of the PACE service into Australia-Fiji trade has seen rates on that route 2.13 A number of smaller operators compete drop by over $AI,000 per TEU or about with tl major lines for paticular nmarkets. Not 30 percent. Significantly, although the PACE surprising, entry and exit at this level is more line has subsequently joined the APIRA frequent-three smaller lines offering conference, its place as the 'threatening outsider' competitive regional services to the nominated bas been taken by Barbican Line, whicb countries (Translink Paciflc, Maasmond, and W continues to offer rates $200-$400 per TEU Islands Line) have entered since the mid 1980s. below those offered by conference members for the same service. 2.14 There is a surprisingly wide range of dedicated services from Europe into the South 2.17 The recent history of the Australia-PNG Pacific, with NedLloyd (EPACSA), trade is similar. Rates in that trade have Bank/Columbus (SOPAC), and Polish Ocean dropped by in the order of $1,000/rEU (again Lines (South Pacific Line) all providing about 30 percent), largely as a result of independent services whose primary focus is continued pressure from entrants. Transpacific serving Pacific Island destinations. Dedicated has recently left the trade, but pressure is services from Asia to the Pacific are dominated maintained through the activities of Barbican and by a consortium of shipping conference by the perception that there are a number of operators (NYK, CNCO and MOL), which lines sailing between Australia and Asia that are provide two services: the Paradise service, continuously on the lookout for entry serving PNG from Japan, New Guinea Pacific opportunities. Line (E Asia/SE Asia to PNG), and the Bali Hai service, serving the rest of the South Pacific. 2.18 Cross-traders are also active in a number The only other significant dedicated service is of the other trades. Blue Star's service from provided by Kyowa, which links East Asia and New Zealand to the Pacific calls at Fiji, Japan with a wide range of South Pacific providing an important source of additional destinations. competition on that route. Tasman Asia makes calls at Fiji and New Caledonia en route to Asia; Competition from Cross-Traders CGM offers slots between Australia and Papua New Guinea on its round the world Euroceania 2.15 Shipping markets are typically service. contestable. The actual and potential competition from cross-trading operators is Competition from Alternative Import Sources dictated by a combination of market size and the costs and benefits of the route deviations and 2.19 The structure of liner shipping services to port calls involved in serving a market. Because the South Pacific is invariably dictated by the of the ease with which cross-traders can enter needs of the import trades. Export volumes 8.5 - generally amount to a small percentage C. SHmPMNG SERVICES TO THE (frequently below 10 nercent) of ir orts. INDIAN OCEAN YSLANDS (MIauritlus, Madagascar, REunIon, 2.20 Sinceimportsconsistprincipallyofbasic (Mauritius , Comoros) consumables that are available from a number of Se,'cheles, Comoros) trading partners, the possibility of substitution 2.22 Psosing Services. The general pattern of between import sources adds an important services to Indian Ocean island countries is in dimension to the competitive discipline on liner many ways similar to that to South Pacific shipping operators. This applies with particular countries. However, the reliance on 'passing' force to services operating from the two major services is more marked and more focussed on import sources, Australia and New Zealand. the nodal port of Port Louis (Mauritius), which The market shares of these two supplier lies directly in the main sea lane from North and countries are reported to be highly volatile. For East Asia to South Africa. The MOL and SCI instance, the volume carried by PFL from services from Asia to East Africa also make Australia to PNG has halved over the period wayport calls at Suychelles. Passing services are 1990-1991, while PFL's volume out of New described in Table 2.4 and shown in Figures 10 Zealand to PNG has more than doubled. and 11 in Annex 2. Market ftructure Pand Contestability 2.23 Dedicated Extra-} egioo-l Service. eDedicated extra-regional services from Europe 2.21 The general picture of shipping services are dominated by the Capricorne service. This to the Soutt Pacific is one of a contestable service, provided by a consortium of European market. The stable core of the service network lines, serves two routes: the A-service, serving s provided by a small number of liner operators the main ports on a fortnightly basis with providing a comprehensive range of services. cellular tonnage; and the B-service, offering Competition between these operators varies in monthly sailings with multi-purpose tonnage intensity fromsumrt to sub-market i. covering some of the smaller ports. Dedicated Nevertheless, competition in each sub-market is extra-regional services are shown in Twable 2.5 brought together by two forces: the smaller and in Figure 12 of Annex 2. scale operators seeking opportunities to establish local services, and the larger cross-traders 2.24 Regonal Operators. Regional services seeking profitable wayport calls to top up are shown in Figure 13 of Annex 2. voyage revenue. These two forces, combined with the ever-present possibility that trades wili be lost because of changes in import sourcing, D. ASSESSMENT MEMODOLOGY provide a substantial degree of competitive pressure. The result is a well-articulated service 2.25 Factors Affecting Shipping Costs. The network which offers a quality of service as cost of transporting freight by sea is influenced high, and, as is documented below, freight rates by many factors, but predominantly depends on as low, as could reasonably be expected, given two variables: the distance goods are shipped, the low traffic density and extremely imbalanced and the traffic density on the route. In general, flows that characterize the demand for liner the longer and denser the route, the larger the shipping services in the region. vessels used. However, costs do not increase proportionally with distance, as a significant part of total costs, particularly for general freight, is associated with loading and unloading, and possibly transshipment. Nor do costs reduce -86- Ta"Ale 2.4: IPNDIAN OCEAN ISLANDS-"PASSING" SERVICES CALLING, 1991 Service From/To Island Calls Ahrenkiel Japan, Korea, Ht.ng Kong, Mauritius Singapore to South Africa NedLloyd (Asacon) WCNA, Japan/Korea, Hong Kong, Mauritius Singapore to South Africa & S America Barbican S America, Australia to South Africa Mauritius Kien Hung Far East to South Africa Mauritius Mitsui-OSK Line (MOL) ANZ to east coast of South America Mauritius Mitsui-OSK Line (MOL) Japan, Far East to East & South Africa, Mauritius, REunion South America Nantal Japan, Far East to South Africa Mauritius Nippon Yusen Kaisha (NYK) Japan/Far East to East Africa Seychelles Arrow L SE Asia to South Africa Comoros, Rdunion (on inducement) Shipping Corporation India to East Africa Mauritius of India (SCI) P&OCL India to East Africa Seychelles /A A subsidiary of NedLloyd. Source: Industry publications and interviews. Table 2.5: INDIAN OCEAN ISLANDS-DEDICATED EXTRA-REGIONAL SERVICES, 1991 Service Trade Region Countries Served Capricorne (A-Service) La UK,North Europe, Med Seycs , Reunion, Mauritius, Madagascar Capricorne (B-Service) North Europe, Med Reunion, Comoros, Madagascar Besta Line (Baltic Shipping) N Europe to East Africa Mauritius, Madagascar Marfret /b N Europe, Med Reunion, Madagascar Mediterranean Shipping Co Mediterranean R6union, Madagascar, Mauritius Nedlloyd (Eacon) Singapore (Japan/Far East cargo Mauritius, R4union, Seychelles tranship from Asacon service) Mitsui-OSK Line Japan, Far East Madagascar, Mauritius, Reunion SBN/SMN /c SE Asia R£union, Mauritius L A joint service of CGM, Navale & Commerciale Havraise Peninsulaire (NCHP), Societe de Navigation Cannaise (SNC), Scandinavian East Africa Line (SEAL), Hapag-Lloyd and Societe Nationale Malgache de Transports Maritimes (SMTM). P&OCL/Ellerrnan and Harrison Lines slot charter space for a UK- Mauritius service. /b Compagnie Maritime Marfret Ic Societe Bourbonnaise de Navigation (SBN); Societe Mauricenne de Navigation (SMN). Source: Industry publications. - 87 - uniformly with vessel size; for example, the Singapore and Fong Kong, as well as those savings in linehaul costs arising from scale from London and Rotterdam. economies in vessel size are partly offset by the Increased port time required for cargo ,.andling. 2.28 The comparison with shipping services A third factor, which is more important in for other countries focused on passing services determining the tariff charged, is the extent of to the Pacific Islands and the Indian Ocean back-loading; a route with a large proportion of Islands as follows: back-loading enables a lower average tariff to be charged for each leg of the voyage. The Pacific - Hong Kong - Chile Island routes have limited backloading and Singapore - Chile therefore involve higher inward rates than other - Hong Kong - South Africa routes which attract return cargo. In addition to - Singapore - South Africa these factors, ocean freight rates are shaped by - Australia - US west coast market atructure, especially ease of entry. - Fiji - US west coast. Approach and Methodology 2.29 Container Traffic Rates. Two-thirds of the total freight moved to/from the majority of 2.26 The assessment of the shipping services the South Pacific Islands is by container.' in the South Pacific and, in particular, to the Therefor, the analysis is confined to container- I'MCs, is approached in two ways: (a) by carrying vessels, and is concentrated on fill comparison of actual freighi rates on specific container load (FCL) traffic. Unless otherwise voyages for major services to the Pacific Islands specified, the freight rates used in the shipping from Australia and New Zealand and "expected" assessment are basic commodity rates, Freight freight rates estimated using a shipping cost A;l Kind rates (FAK), or general cargo rates for model, and (b) by statistical analysis of the Full Container Load (FCL) traffic, on a wharf- relationship between actual freight rates and, to-wharf basis, excluding terminal handling costs distances and volumes. The objective of these as levied by local port authorities.' Port analyses is to establish: operations and infrastructure also influence the total door-to-door ch.arges, but they reprasent a (i) whether rates to and from the small fraction of ocean freight costs; differences PMCs are different relative to the in these charges between ports are also small. selected comparator Island countries, taking into account 2.30 The freight rates used are as quoted by route distance and traffic agents; they do not take into account any volumes; and customer rebatcs that a shipping line might offer. Where these exist, actual rates will be up (ii) whether rates to and from all to 10-15 percent below the tariffs used herein. Island countries are different However, in some isolated cases (generally compared to other countries. promotional backload rates), discounts of up to 50 percent were quoted. A s:hedule of typical 2.27 Services Selected. In the case of the prevailing shipping tariffs is 1,iven in Annex 3. Pacific Island countries, the analysis of services to the PMCs and comparator islands centered on 2.31 Freight Rates in Relation to Distance. those to and from Australia and New Zealand, Figure 2.1 shows a plot of actual ocean freight and the regional shipping centers of Singapore rates for services ex Australia, New Zealand, and Hong Kong. In the case of the Indian Hong Kong, Singapore, Rotterdam and the UK Ocean countries, the services covered those from against distance. The plot indicates that there is Plaure 2.1: ACTUAL DISTANCE AND FIGHT AT_S_-PW M l * 8.- 9~~~~~ I ++a ++ + 4+ ~~~~~~~~~ ++ t+ + +t * sa ±; :~ a m 2 410 12 14 Distance (Nautical Mil cs Thousands) a Nominated Pacific Islands + Comparator Islands Otbor ca paratorc * 89 - no prima facie case that freight costs for the 2.32 Freight Rates In Relation to Frequency. PMCs as a group are any different to the Pacifi' Table 2.6 shows that the PMCs have service and Indlan Ocean comparator countries. Figure frequencies similar to those of the comparator 2.1 also shows that rates on the island routes are Islands. Mauritius (Port Louis), which lies similar to rates on many non-island routes. directly on a major route from East Asia to Non-island routes with the lowest rates are South Africa, receives the most cals per month between South Eqst Asia and South Africa, among the comparators. As a satellite of where backloading is substantial. Mauritius, Reunion be fits from relatively frequent services. However, the freight rates to Reunion are generally higher than the rates to Table 2.6: PACIFIC AND COMPARATOR ISLANDS AVERAGE FREQUENCY OF MAJOR CONTANER SERVICES, 1991 Average No. of Average No. of Destination Calls per Month Da Destination Calls per Morth La Pacific Islands C3mparator Islands Kieta 2 Mozothe 1 Popendetta 3 Dzaoudzi 1 Tarawa 3 Christmas Islarnd 1 Kavieng 5 Cocos Islands 1 Kimbe 7 Moroni 1 Alotau 7 Mutsamudu 2 Santo 8 Yap 2 Wewak 10 Truk 3 Nuku'alofa 11 Mayotte 4 Vila 11 Pago Pago 6 Madang 12 Male 7 Lautoka 12 Tamatave 9 Honiara 12 Mahe 10 Apia 13 Papeete 18 Rabaul 16 Reunion 18 Port Moresby 29 Noumea 27 Suva 29 Port Louis 35 Lae 29 La Calls offered by the forty shipping agencies and freight forwarders sampled. Actual service frequency often differs from the frequency "offered" by the shipping lines. For example, a service offered as a monthly service may call at a port as often as fourteen times per year or as few as four times per year, depending on the prevailing demand. Source: Industry Publications and Survey. 90- Port Louis, because of transhipment costs and E. ASSESSMERr OF SHPPING RATS lower volumes. Application of the Shipping Cost Model 2.33 Noumea and Papeete receive frequent calls as both countries have relatively large trade 2.35 Shipping services are assessed using a flows. New Caledonia and the Society Islands voyage cost simulation model (described in are amongst the largest merchandise lmpo ters in Annex 5) which calculates the expected" rates the South Pacific region (see Annex 1, Table 1). for a voyage, given nominated ports of call, Services to Noumea (ex Singapore, Hong Kong, volumes carried, distances travelled, and the size and Australia) are generally comparable to and age of vessels employed. In practice, services to Fiji and Papua New Guinea in terms commercial freight rates principally aim to of freiglt rates, volume and distance. However, recover total voyao costs and maximize proflts. it is claimed that rates to New Caledonia are For any one route, a range of rates wIll exist slightly higher to cover the higher port handling which is affected by local demand/supply costs in Noumea. factors, within the overall constraint that total voyage revenues at least equal total voyage 2.34 Of the PMCs (including Papua New avoidable costs. Guinea), Port Moresby, Lae and Suva receive the highest number of calls ver month (see Table 2.36 The cost model estimates the cost of 2.6) reflecting the volume of import cargo and operating container vessels of a specifled size greater backloading at these ports. Exports from over a fixed route, given a defined pattern of some of the smaller islands (for example, Tonga cargo uplift and discharge. The costs are a and Western Samoa) are reportedly so low that function oi three sets of data: vessels often carry return cargo back at highly reduced "development' rates. Giveni excess * Vessel characteristics backhaul capacity, relatively low outbound rates * Route definition are efficient and should prevail; -omparison of * Operating and cost parameters. outbounf. rates alone from the PMCs would reveal them to be low relative to oth3r countries. Cost Model Results The Solomon Islands receives fairly frequent calls as a result of its proximity to Papua New 2.37 A comparison of the freight rates Guinea (about 12 calls per month in the sample). generated by the shipping cost model ("expected Regarded as an extension of the PNG tr ;de, freight rates") and actual reported freight rates is shipping lines serve Honiara on a monthly basis, presented in Figure 2.2. On average, freight either direct, or by transhipment using smaller rates to the Pacific Ocean islands (including vessels. Services to Kiribati and the Comoros Noumea) are consistent with the expected costs are at the higher end of the rates scale. Both of serving them, given current market conditions these island countries receive relatively and traffic density. infrequent service reflecting their low traffic density. 2.38 At present, the Australian Pacific Islands Rate Agreement covers the trade areas of Australia to New Caledonia, Fiji and Vanuatu. To some extent, the similarity in actual rates reflects the existence of the conference since the selected lines all serve one or more of these countries. In addition, existing and potential competition outside the conference tends to keep JIl .2.2s2: PACIFIC ISLAUS-STI&TZSD SUIPPIM COTS AND ACTUAL 1131 T RATES 4 ..................................... .. .. ....... l~~~~~~~~~~d j 4~~~~~~~~~~~~~~~~~~~~~~~~~~~ .. E. .. ........... 'e' 3 |' . | ..-..... ''...... '--.. .... ................ .... ........ .... ...... .... '--..''*'-..----.*--'..''''| '-- | * z 2 z ............ ....... ...... ...... ... . .................. . .................... ......;;; * II I CHIEF CONTAINER SEWtIES 50FRANA PC N U O t1 I . x HE OTI/R E IE ORN PACIFIC FORUMV LINE 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Routes Actual LED E1pect.d 1. Umlbourne.Sydney.briabane.ou ea.Suva.United States 7. Uth.5Er Z*&laud.Stb.3w Zeaua.Xavm&a.Lme.Port Nbresby 2. Nelbour. .Sydney.Dribab n.Port Koreaby.La .0niara S. Sth.I. Zealaad.th .Bw ZelanL.S- a.Brlabm*.Port 3. Nelbourar.Syd1ey.3r1*ban..Port Iborezby.Lae .Honi1ra.Tarwva Uorelby.Lm.3Ilsbm.1kma.Usw Zmelnd 4. Hl.botirn.Syduey.Driabsat..ou a.Vila.Hoaiara.Tarrwa. 9. Sth.Z w Z m Ia.Ith.Esv Zealw d.ulebam.Port foreaby.Bwdara Port Noreeby.La. .Kav'ng.Rabaul.Kieta 10. Sth.B. Zealad.tb.ko, Zealead.S ridba.Loe.abawl 5. Neibourno.Sydney.Brlabanz. ouma.Vila.Santo.Lautoka.Suva 11. Sydey.arebasm.Sw a.Tarevn 6. Eth.'w Zea-lnd.3oumean.Vi1a.Santo.Honiara.Rab ul.La. 12 Sth. M W Zealnd.L4otoa.Sw a.*p1*.Ppg Pgp.uk-ualofa 13. Syds*y.Ir1.bam.Swva.&p1a.Pae Pap.mk-1a1ofa.Sva -92- rates down generally, as well as reduce the However, the Forum Papua New Guinea and the disparities between rates. For example, the Forum New Zealand II are involved in three entry of the PACE service into Australia-Fiji trades: Papua New Guinea, Fiji and Trans- trade has resulted in a cumulative reduction in Tasman (represented by Route No.8 in Figure freight rates in excess of AS1,000 per TEU. 2.2), and the freight rates for this latter route are generally in line with expected rates. 2.39 The range of actual rates for Acta-Pace and Sofrana Unilines fall entirely within the 2.42 The routes which involve transhipment range of expected rates. It is important to note (Route Nos. 9, 10 and 11 in Figure 2.2) have that the service by Acta-Pace is essentially a higher expected rates. In the model passing service to New Caledonia and Fiji en transhipment is treated as two separate voyages route to the US west coast, whereas Sofrana is whereas most shipping operators view historically the islands' trader. Both Acta-Pace transhipment on an incremental cost basis. and Sofrana have a service which calls Thus, the higher expected rates can be partly Melbourne/Sydney/lBrisbane/Noumea. It is explained by the model's approach to claimed that the Fiji and Noumea markets are transhipment. It is understood that the shipping static and that the trade areas are now less line "bears" the cost of transhipment to attractive with higher costs and lower rates. destinations including the Solomon Islands and Furthermore, the conference has set a Currency the minor ports of Papua New Guinea. Adjustment Factor (CAF) of 14.5 percent for Noumea. Noumea also has relatively high port 2.43 The expected rates for PFL's direct charges. Nevertheless, the shipping lines have services to Tonga, Western Samoa and continued to offer frequent services to these American Samoa (Route Nos. 12 and 13 in islands as indicated in Table 3.1. Figure 2.2) are lower than actual freight rates. The consistent overestimation of rates for routes 2.40 Figure 2.2 indicates that the expected involving transhipment and the underestimation rates for Chief Container Services (CCS) are of the rates for direct routes in Figure 3.2 consistently higher than the actual rates. This suggest the possibility of a tariff "equalization' reflects the competitive pressures in the Papua policy in the region where the full costs of New Guinea-Solomon Islands trade which is transhipment are not passed on. Although the CCS's core business. It is claimed that rates to present traffic to Samoa and Tonga is quite low, Papua New Guinea have fallen as much as inter-island traffics between Fiji, Western A$1,000 per TEU in the past ten years, and Samoa, American Samoa and Tonga are claimed there is evidence that there are currently to make contributions to voyage revenue. competitors charging as low as AS1,700 per TEU. CCS's other trades, New Caledonia, 2.44 Despite the low traffic growth in the Vanuatu and Kiribati face similar pressures. It Pacific market, the level of service in the area is alleged that Vanuatu market rates fluctuate has increased without an increase in freight because of its vulnerability to cross-traders. The rates. In fact, there has been no rate increase cost simulation model demonstrates that for New Caledonia, Fiji and Vanuatu since the servicing Kiribati is a relatively high cost conference was formed in July 1988. This again operation given the long distance and the low supports the view that the Pacific Islands market volume. is reasonably competitive. Evidently, the competitive pressures yield considerable benefits 2.41 There are considerable differences to the islands, mainly in the form of regular between the expected rates and the actual rates services at reasonably efficient freight rates. for the Pacific Forum Line for the island routes. This is particularly significant as the shipping -93 - operators believe that the current volume of is tested directly by statistical analysis. cargo does not warrant the number of vessels Although the density data were necessarily servicing the islands. partial, the results confirm the general conclusion reached from the results of the 2.45 The Impact of Traffic Density on voyage simulation cost model assessment, Freight Rates in the South Paciflc. The namely, rates to and from the Pacific Islands are impact of distance and traffic density on freight consistent with the distance travelled and the rates in the South Pacific is shown Figures 2.3 prevailing traffic density. and 2.4, respectively. In general, longer distances involve higher freight rates, and denser 2.48 The general shipping freight rate traffic routes involve lower freight rates. In this relationship estimated is: assessment, traffic density is measured as a volume/distance ratio, i.e., the number of TEU Y = A D' * Vi per 1000 nautical miles. For simplicity, the distances used are based on a straight port-to- where Y = freight rate charged per TEU port movement. The routes Melbourne to for the route Noumea (served by Acta-Pace and Sofrana), D = distance shipped Melbourne to US, and Suva to US (served by V = volume of traffic on the route Acta-Pace) are included for comparison. (TEU per annum) and A, a, b are calibration parameters. 2.46 The combination of long distances and low volumes for the PMCs characterize thin That is, shippirg rates are hypothesized to traffic routes for which, a priori, freight rates depend (multiplicatively) on distance and would be expected to be relatively high. Figure volume. It is expected that the calibration 2.4 indicates that thinner routes are indeed parameter a (for distance) will be positive and associated with relatively high rates, for the calibration parameter b (for volume) will be example, Sydney/Brisbane to Tarawa/Apia/Pago negative. The relationship was estimated and Pago. Conversely, dense routes are generally the Pacific Islands rates, as a regional group, associated with lower rates, as illustrated by the were also tested to see if they were statistically routes New Zealand to Fiji, Melbourne to Port different to the rates for comparator countries. Moresby, Melbourne to the US, and Brisbane to Estmation results are set out in Annex 5? Fiji. However, there are a number of low (The shipping data set used is provided in density routes which have relatively low freight Annex 1.) rates. These routes include Melbourne to Honiara, Melbourne to Vila, Melbourne to 2.49 The statistical analysis of shipping rates Noumea, and Melbourne to Suva. Some of shows that the effect of both distance and trade these 'thinner" legs connect to other legs which density on rates is significant. The results also are more dense. Some vessels call at several show that the ocean freight rates to the PMCs ports and traffic density also varies between and other island groups do not differ in the different sections of the voyage. manner in which they depend on distance and density-either among themselves as individual Statistical Analysis of Shipping Rates island groups, or as a group between other comparator countries. The results of the 2.47 Shipping cost structures and contestable statistical analysis are illustrated in Figure 2.5.' market structures predict that ocean freight rates The most extren;e outliers in Figure 2.5 are depend heavily on distance and traffic density. (with actual rates shown in parentheses): In this supplemental assessment, this relationship - 94 - Fiture 2.3: PACIFIC ISLANDS REGION-ACTUAL FREIGHT RATES AND DISTANCE 4000 3600 - 12 3000 - 14 3 *4 R 2600 ~~~~2, (05,17) tO; 1 *185 1600 ,c 1000 300 X 600 _ ol 0 1 2 3 4 6 a 7 Distance (Nautical Miles, Thousands) I Melbourne-Suva 10 Brisbane-Papua Now Guinea 2 Melbourne-Noumea 11 Sydney-Tarawa (via Brisbane) 3 Suva-USA 12 Sydney-Apia (via Brisbane) 4 Melbourne-USA 13 Sydney-Pago Pago (via Brisbane) 5 Melbourne-Honiara 14 Sydney-Honiara (via Brisbane) 6 Melbourne-Tarawa 15 North Now Zealand-Fiji 7 Melbourne-Port Moresby 16 Melbourne-Suva 8 Melbourne-Vila 17 Melbourne-Vll- 9 Brisbane-Fiji IS Melbourne-Noumea _ I 0- I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - 95 - figLe2.43 PACIFIC ISLANDS RION-ACTUAL MIGH lATn AND TRAMIC DENSITY 4000 X 3500 .12 uS .~~~~~~~4 3 3000 - 614 .10 p2500 - B 27 b 2000 - e :t 1600 ~~~~~~~~~~~~.16 j 1000 .a 500 0 20 40 80 80 100 120 140 160 180 200 Volume/Distance Ratio (TEU per 1,000 miles, Thousands) 1 Melbourne-Suva 10 Brisbane-Papua New Guinea 2 Melbourne-Noumea 11 Sydney-Tarava (via Brisbane) 3 Suva-USA 12 Sydney-Apia (via Brisbane) 4 Melbourne-USA 13 Sydney-Pago Pago (via Brisbane) 5 MNlbourne-Honiara 14 Sydney-Honiara (via Brisbane) 6 Melbourne-Tarawa 15 North New Zealand-Fiji 7 Melbourne-Port Moresby 16 Melbourne-Suva 8 Melbourne-Vila 17 Melbourne-Vila 9 Brisbane-Fiji 18 Melbourne-Noumea -96-o - Hong Kong - Santo (AS4,156) addition, the evidence reveals that oceanfrelght - Hong Kong - Port Vila (A$4,156) rates for the PMCs are not Intrinsically higher - Hong Kong - Nukualofa (A$4,156) than for thev.e comparator countries; traffic - Hong Kong - Tarawa (A$4,545) density and distance affect shipping costs to the - Singapore - Apia (A$3,896) PMCs in the same systematic manner that they - Singapore - Nukualofa (A$3,896) affect shipping costs for other countries. - Sydney/Brisbane - Tarawa (A$3,800). 2.52 Although the evidence reveals that Six of these seven routes involve transhipment. shipping rates for the PMCs are in line with Freight is transported from Hong Kon'g to Santo rates that apply for comparable routes for other and Port Vila with a transhipment at Sydney. countries, the ocean freight rates for the PMCs The routes from Hong Kong to Nukualofa and are nevrheless high-relativo to countries with Tarawa both involve transhipment at Kobe in high volumes of cargo flows and/or shorter Japan, which is reported to cost about A$900 distances to/from key markets. Specifically, a per TEU. The rates to Apia and Nukualofa route for other countries, with ton times the include an extra cost of A$l,000 per TEU for volume of PMC routes, can be expected to be transhipment at Suva. It is also reported that associated with freight rates of tho order of one transhipment can be double the direct freight 40 percent lower than rates on PMC routes. rate. Routes which involve half the shipping distance, typically involve rates 30 percent lower. 2.50 The actual freight rate from Sydney/Brisbane to Tarawa (in Kiribati) Is 2.53 Cargo volumes for the PMCs can be higher than estimated (on the basis of distance expected to remain low, in partit ilar, relative to and route density). This may be explained by an some countries with whom they compete for unusually nigh degree of uncertainty in service exports. Since the PMCs face international level and difficulties at the Tarawa port (Betio). prices for their exports, the higher transport Although the service to Tarawa is offered as a costs they confront presents a considerable monthly service, calls at Tarawa are made on constraint on their ability to compete on world demand rather than with a fixed frequency, and markets. Their higher international shipping in practice may be as infrequent as once every costs dictate lower f.o.b. prices. Thaes lower eight weeks. Furthermore, inclement weather f.o.b. prices, given internal transport costs, can cause considerable vessel delays in Kiribati. imply lower "farm-gate' prices for export While a vessel can clear Betio port in two to commodities. (Indeed, it is possible that for three days in good conditions, it is not some commodities implied farm-gate prices may uncommon for vessels to spend up to seven days be even below zero, i.e., local production of the in port. commodity is not commercially viable-at prevailing world prices and transport costs.) The main implication of the relatively high F. CONCLUSIONS transport costs faced by the PMCs is that local production costs must be correspondingly lower, 2.51 In sum, the results of the assessments of if commodities for export are to be competitive. shipping costs and freight rates confirm that For economic gro'vth in the PMCs, it is costs/rates to andfrom the PMCs are consistent especially Important that local production and with costs/rates to andfrom both the comparator internal tranport are efficient. Islands and other countries, when distance and traffc density are taken Into account. In - 97 - Figure 2.5: PACIFIC ISLANDS AND COMPARATORS-SHIPPING RATE ANALYSIS RESULTS 4 + a 1-8 ! 4 + -mato * ~+ 04 b0 0 05 1 1.5 2 2.6 3 3.6 4 4.6 6 Estimated Freight Rates (A$ per TEU, Thousands) U PMCs + Comparators - 98 - CHAPTER 3 ASSESSMENT OF AIR SERVICES A. Ant SERvicEs iN TlE 3.4 The larger Pacific Islands, such as Fiji PACFC REGION and Papua New Guinea, receive relatively frequent services. Fiji (Nadi) is both a major 3.1 The air services to the PMCs and hub in the regional network, and an comparator countries included in this assessment origin/destination in its own right. As a are described under the following main consequence, it has frequent services from both headings: Sydney and Auckland at relatively low air fares. As with shipping cost structures, increasing - Pacific Rim to Pacific Islands; distances lead to higher air service costs and * Pacific Island intra-regional services; and fares (and, with low travel demand, reduced * other routes. flight frequency). High volumes and traffic density (largely arising from the size of the 3.2 A total of 111 routes are included in the tourist industry) allow larger aircraft sizes, analysis. Details of the routes covering fares, especially over longer distances, and allow lower distances and traffic volumes are shown in unit seat costs and fares; this is notable tor Fiji. Annex 6. In addition to the Pacific Rim to Pacific Islands (34 routes) and the Pacific Island 3.5 Passenger volume also affects costs and intra-regional services (27 routes), other routes fares, especially for smaller Island States, such included are: as to Tonga and Kiribati. The incremental costs of flying to Pacific Island countries beyond - Australian International routes (9) Nadi, given the lower volumes of passengers, - Australian Domestic Tourist routes (11) are significantly higher on a per mile basis. For - New Zealand Domestic routes (7) example, the additional fare to Kiribati from Fiji - Papua New Guinea Domestic routes (8) represents around twice the Sydney/Auckland to - Western Australian Domestic routes (8) Fiji fare for one-half of the distance. Moreover, - Indian Ocean Islands routes (7). with small passenger volumes and higher seat costs, excursion or promotional fares are rarely Pacific Rim to Pacific Islands available on low density routes. 3.3 Pacific Island Countries. The major 3.6 Although services to the PMCs (other routes to and from the Pacific Islands, and major than Fiji and PNG) have increased significantly destinations on the Pacific Rim, are listed in over the past decade, the current levels of Table 1 of Annex 7 and shown by map in Figure demand do not warrant the use of wide-bodied 1 of Annex 8. aircraft. Instead, the airlines employ smaller aircraft (such as B727, 737, ATR42 and -99- HS748), which involve higher unit seat costs and Comparator Annex 7, Annex 8, hence higher fares. Route Groups Table No. Figure No. 3.7 Compartor Padfle Islands. Noumea Australian International 3 3 and Papeete are served by medium-sized aircraft Indian Ocean 4 4 and at higher levels of frequency, reflecting Australian Domestic Tourist 5 5 higher volumes of international visitors. In Western Australia Domestic 6 6 addition, Tahiti is a stopover for four of the six New Zealand Domestic 7 7 flights to Los Angeles from Sydney and Papua New Guinea Domestic 8 8 Aucldand. The fares from Sydney to Port Vila and Noumea are broadly comparable, although there is a slight disparity in the fares ex B. A_Maw METODOLOGY Auckland. In addition, the number of flights from Aucldand to Noumea is double that to Port 3.10 This section presents a two-stage Vila and involves more carriers (3 compared assessment of air services to the PMCs: with 1), nd (some) larger aircraft (DC10s and 767s). The focus for international services to (a) a qualitative aussment of the currt and from Noumea and Papeete is much more levels of service and aircraft type, given strongly centered on direct flight movements into the levels of air travel demand; and and out of the South Pacific region rather than within the region. For example, there is only (b) a quantitative assessment based upon a one flight a week between Noumea and Nadi statistical analysis of air fares for the and none between Papeete and Nadi, even PMCs, comparator Islands, and other though Nadi is a regional hub, (partly reflecting routes in the Pacific basin. the division between the Anglophone and Francophone countries in the region). QuaUtadve Assessment of Air Servlas for the PMCs7 Intra-Pacflc Island Regional Services 3.11 There are significant differences in the 3.8 Major intra-Pacific regional services are level and composition I of demand for air listed in Table 2 of Annex 7 and shown on services involving the PMCs, both in terms of Figure 2 of Annex 8. In general, the inter- movements into and out of the region, and island international services between the Pacific within the region. Consequently, different Island States are reasonably frequent given the aircraft type- and service frequencies are used traffic density. Fiji acts as a regional hub with on the various routes. This is illustrated in over 40 international flights to and from Nadi Table 3.1 which lists services from the Pacific each week. Rim to the PMCs.' For example, for Fiji the high level of tourist demand is reflected in the Other Routes use of wide-bodied jets such as B747a and B767s. As the major tourist destination in the 3.9 A list and map of the other air service region, Fiji also receives more frequent services. routes, used as comparators in the assessment of For example, non-stop flights to Fiji each weelk air fare and level of service for the PMCs, are include 7 from Sydney, 4 from Melbourne, 3 provided in Annex 7 and Annex 8, respectively from Brisbane, and 11 from Auckland utilizing as follows: the larger jet aircraft. In addition, there are 11 flights each week from Nadi to Honolulu. No other Idand State in Lhe South Pacific matches -100- the tourist trafric of Fiji. As a result, the other 3.14 Comparator islands with significant PMCs, with their lower levels of demand, are tourist industries also receive commensurate generally served by smaller aircraft, typically levels of service and frequer'cy, typically with with capacity ranging from 50 to 200 seats, wide-bodied aircraft. For example, Mauritius often from Nadi as a hub. Vanuatu, Western is served by B747, B767 and Airbus A310 Samoa, Solomon Islands and Tonga are all aircraft with 13 flights per week from London, generally served by medium-sized jets, such as Paris and Johannesburg. Similarly, there are B727s and B737s, with capacities around 144 nine flights from Sydney to Bali each week, and 109 seats, respectively. Furthermore, the using B747 and DC1O aircraft. The services to frequency of services to these islands, even with New Caledonia from Sydney and Auckl3nd smaller aircraft, is lower than to Fiji. Also, (seven and three per week, respectively) use direct services are often incorporated within a B737 and B767 aircraft, reflecting the medium multi-point route structure, for example, level of demand. In addition, given the demand Westen Samoa is served via Tonga, rather than ex-Japan, there are two DC1O services weekly by non-stop flights from Sydney and Auckland. from Tokyo. These services from Sydney and Auckland to Noumea are more frequent than 3.12 Papua New Guinea receives relatively services to Tahiti, despite Noumea's relatively frequent services from Australia and New lower international arrivals. But, Tahiti is both Zealand. There are currently five services per a tourist destination in its own right, as well as week from Sydney to Port Moresby via Brisbane being a stopover on trans-Pacific flights. Also using B767 and A310 aircraft. Auckland to Port Tahiti is served by on-connection over Noumea. Moresby is not served directly but is via Sydney or Cairns. At present, there are eight services 3.15 In sum, with the exception of Fiji and, to from Auckland to Cairns, seven of which are via some extent, Vanuatu, the aviation markets to Sydney. Small aircraft such as F28s fly the andfrom the PMCs are thin (low traffic volumes route between Cairns and Port Moresby. Unlike over relatively long distances). As a result, Fiji, the demand for air travel to and from aircraft sizes are small andfrequencies are low; Papua New Guinea is not 'riven by tourist as a consequcnce cost structures for air services traffic but by the commerciE and government operating in these markets can be expected to be sectors. In 1988, approximate'y 62 percent of relatively high. total arrivals from South Asia and Oceania to Papua New Guinea were arrivals for Statistical Analysis of Air Fares business.'° 3.16 Specification of Air Fare Relationship. 3.13 Volumes of traffic on routes involving As arguded above and as with sea freight tariffs, the other PMCs are much smaller; international air fares are predominantly influenced by visitor arrivals are 20 percent (Western Samoa) distance and traffic Jensity. However, there are or less than 10 percent, of the level to Fiji. some added complications because airlines offer Thus, total available seat capacity, aircraft size, a range offare types reflecting different service and frequency are all substantially lower. There conditions. For the purposes of this analysis, air are no direct services to Kiribati from Sydney fares are grouped into two types: (a) fidl and Auckland. The main route is through Fiji, economy fares'2; (b) excursion fares." with a connecting flight to Tarawa (over Tuvalu en route to the Marshall Islands) by turboprop 3.17 The large difference between full aircraft (HS 748) twice a week." economy and excursion fares has a major influence on any comparative analysis of air fares. On some routes, excursion fares are Table 3.1: PACIFIC ISLANDS FROM THE PACIFIC RIM-INTERNATIONAL AIRLINE SERVICES, 1991 Todl No. of Idernmio Origin Sydney Melbourne Bridm Cairn Anckm Sigpopre Hon Kong Tokyo Visitors 00) (1988) Destion _t 250.6 FDI - Nadi /c FCghtr per week 7 4 2 11 3 Aircraft type 747 767 767 767 747 23.9 VANUATU - Port Vila Flights per week 2 1 2 2 Aircraft type 727 | 7 727 737 51.7 WESTERN SAMOA - Apia Flights per week 2 5 Aircraft type 727 737/721 9.9 SOLOMON ISLAND, - Honiara Flights per week 1 2 1 2 Aircraft type 767 737 737 737 20.3 TONGA - Tongatapu Flights per week 1 6 Aircraft type 727 737 48.9 PNG - Pot Moresby Flights per week 5 5 4 7 2 1 3 Aircraft type 767 767/A310 A3101F2S 767 h F2Uo A310 A310 A310 3.0 KIRUBATI - Tarawa Flighte per week 2 2 Aircraft type 747 & 767 HS748/a HS74Sh _ 82.0 NEW CALEDONIA - Nounee Figh per week 7 2 3 3 2 2 Aircraft type 737r767 7371767 737/767 737/7671 747 DCIO DCIO 139.7 TAHM-Papeete/c Fligh per week 3 3 _ Aircft type 747M7 747DC10 a HS748 friom Nadi to Tauwa. F28 font Cain. to Port Moreasby. /c Alro srvieu ex USA. Soue: Aidim Tineabies and Wodd Bank (1991a). -102- limited, generally either because of low demand comparable promotional fares were used. T .o or because of the regulatory environment. This measures of distance were used. Full economy is the case on a number of intra-Pacific Island fares were related to maximum permitted routes; however, there are many excursion fares mileage (MPM) and excursion fares were related available to and from the region. to shortest operated mileage (SOM). MPM takes into account the alternative routings 3.18 The statistical analysis comparing air availaMie to economy fare trivellers on longer fares was undertaken using data on eight groups journeys. A number of proxies for traffic of routes: density were identified, including total visitor numbers and flight seats capacity. Although seat (a) to and from the Pacific Islands; capacity is a measure of supply, information on (b) between the Pacific Islands; visitors is not consistently available across the (c) international routes from Australia; routes selected. Seat capaciy is adopted as a (d) domestic routes in Western Australia; measure of traffic density and, hence, a proxy (e) tourist routes within Australia; for traffic demand on the basis that, in the (f) New Zealand domestic routes; medium to long run, airlines systematically (g) Papua New Guinea domestic routes; and allocate aircraft types and frequencies, and (b) to and from Indian Ocean Islands. therefore seats, to particular routes based on market demand. Seat capacity per annum, was 3.19 Air fares typically vary systematically estimated as the product of the number of flights with distauce; there is an implicit flagfall per year and the typical seat capacity on the component and a component variable with aircraft serving the route. Passenger seat distance. An objective of the present analysis is capacity for typical aircraft types and to determine if there are differences in air fares configuration in the Pacific and comparator across regions that cannot be explained simply in regions is shown in Table 3.2. Where the route terms of distance and traffic density. is served by different types of aircraft, a weighted average of the typical number of seats 3.20 The general relationship tested is: available war calculated. Y = A lD * Vb C. ASSESSmENT OF AIR FARES where Y = air fare charged D = distance travelled 3.23 The data set represents 111 routes with V = volume (seat capacity per annum) the current economy and excursion fares, the and A, a, b are calibration parameters. shortest operated distance and the maximum permitted distance, and the number of seats 3.21 This relationship was examined to test offered on the route per year. Table 3.3 whether any of the eight groups (including the summarizes these data for each of the eight route PMCs) were statistically different from the groups. The full data set is provided in remainder. " Anne;. 6. 3.22 Definition of Fares, Distance and 3.24 The international routes receive the Density. Given the wide range ot excursion greatest discounts on full economy fares (see fares available, an average of the low season Table 3.3). Higher volumes allow larger excursion fare (designated YLE) and high season aircraft, lower average seat cost, and a wider excursion fare (designated YHE) were used. range of excursion fares. Considerable Where these fares were not available, discounts are available for the high volume - 103 - Table 3.2: AIRCRAff TYPES AND TYPICAL SEAT CAPACITY Typical Passenger Aircraft Types Seating Capacity Wide-bodied Jets B747-300/400 380-440 B747-100/200 370-430 DC-10 250 A300 250 A310 210 B767-200 210 Narrow-bodied jets B727 144 A320 i32 B737-300 109 DC-9 92 British Aerospace BA3 (BA3 146) 75 F28-4000 72 Large turbo-prop Fokker 50 50 F27 36 De Havilland Canada Dash 8 (Dash 8) 36 Aerospatiale ATR 42 44 Hawker Siddeley 748 30 Light twin-prop Embraer Bandeirante 18 De Havilland Canada Twin Otter (Twin Otter) 20 Source: Aviation industry. Australian international routes. In addition, density is higher. Seat capacity for intra-Pacific highly trafficked tourist routes are generally Island routes is relatively low. These routes are associated with lower fares. This is thin and served by small aircraft with higher per demonstrated by the large difference between the unit costs, and tourist traffic is small and average economy fare and the average excursion oriented to niche markets. As a result, fare for the Indian Ocean island routes. discounting on intra-Pacific Island routes is Table 3.3 also indicates that there is some relatively modest and limited.15 discounting available on fares to and from the Pacific Islands, although the extent of 3.25 The average economy fares and excursion discounting is not as great as the other two fares per seat mile are plotted in Figures 3.1 and groups of international fares, where traffic 3.2, respectively. As is to be expected, the Table 3.3: PAcinc AD CoMPARATOR REGIONS-AIR FARs, SECO DisTANcrs AND CAPAcIT, 1991 a Discount Return Retun Between Maxium godese AIma Eonomy Excutsion Economy Excursion Economy & Paemited Oprated Seat Fae per Fare per Route Group Fae Fare Excursion Mileae Mideae Cpcity Seat Mile Seat Mie (AS) (AS) Fae (.AS) (miles) (mies) (000) (AS per m) (A$ per m) 1. Pacific Rim to Pacific Islands 1,816 1,049 767 2,277 2,061 27.3 0.399 0.254 2. Intr-Pacific Islands 959 745 214 1,349 1,162 27.0 0.355 0.321 3. Australian International 3,555 1,344 2,211 5,193 4,252 401.5 0.342 0.IS8 4. Westen Australia Domestic 635 419 216 720 720 53.7 0.441 0.291 5. Austdalian Domestic Tourist 688 505 183 857 857 221.8 0.401 0.294 6. New Zealand Domestic 365 253 112 356 356 246.1 0.512 0.355 7. Papua New Guinea Domestic 342 280 62 297 297 55.9 0.575 0.471 8. Indian Ocean Islands 3,434 1,576 1,858 4,785 4,011 57.9 0.359 0.196 /f Details are provided in Annex 6. Source: Industry surveys. - 105 - comparatively shorter domestic routes of Papua 3.29 In Figure 3.3. there are only two Pacific New Guinea and New Zealand have the highest Island routes where the actual economy fare is average fares per seat mile. Furthermore, Papua substantially different from the estimated fare, New Guinea domestic routes are not highly namely, Sydney-Papeete and, Apia-Honolulu and trafficked routes, and are serviced mainly by neither of these routes involve the Pacific small aircraft. Economy and excursion fares per Islands. Similarly, Figure 3.4 shows that the mile to and from the Pacific Islands, and majority of the Pacific Islands' excursion air between the Pacific Islands, are lower than those fares fall in the middle of the sample; only five of domestic flights in Papua New Guinea, New Pacific Island routes are identified where the Zealand and Australia. The lower economy excursion fare is not in line with expectations: fares per mile for the Indian Ocean Island routes Noumea-Papeete, Singapore-Noumea, Apia- and Australian international routes result from Honolulu, Sydney-Tarawa. and Auckland- longer average distances and the use of wide- Tarawa. Only three of tlese excursion fares bodied aircraft. involve PMCs; two involve Tarawa, where the fare is a sum of sector fares constructed over Statistical Analysis Results Nadi. Excursion fares are available from Australia and New Zealand to Nadi, but not 3.26 Two sets of statistical analyses were Nadi to T,rawa, which is a very low density, undertaken: one tested economy air fares and relatively high cost route. the other tested excursion fares. Detailed results are provided in Annex 10. The general funding of the statistical analyses is that the pattern of air D. CONCLUSIONS fares is indeed largely explained by distance and traffic density. Excursion fares, where available 3.30 In summary, the evidence reveals that air are used by most travellers, are strongly related fares in the Pacific Islands and comparator to distance and inversely related to density. regions are shaped by distance and traffic Economy fares are even more strongly related to density. The evidence does not support the distance, but show no statistical dependence on hypoth.sis that air fares for the PMCs are traffic density. Both these results confirm intrin.ucally higher than for other comparable general experience."6 States; distance and traffic density exert L systematic and uniform influence on air fares. 3.27 The tests for regional differences in the fare relationship were negative, except for 3.31 At the same time air service routes to and economy fares on intra-Pacific Island routes. from the PMCs are thin; distances are relatively On these routes economy fares are lower for long and traffic densities are low. Thus, air other regions/routes, at the same levels of fares are high-relative to shorter hau!s and distance and density. If anything, the results relative to routes with high volumes. The suggest that the intra-Pacific Island routes are evidence indicates that a four-fold increase in experiencing a modest relative fare advantage."7 airline passenger movements (which corresponds to a doubling of total arrivals) corresponds to air 3.28 Figures 3.3 and 3.4 show a plot of actual fares that are 28 percent lower. against estimated fares for economy and excursion air fares, respectively. Both figures 3.32 Notwithstanding the potential for securing illustrate the result of the statistical analysis; the larger volumes of traffic in concert with lower Pacific Islands, including the PMCs, are not fares, since existing fares are high, the attraction disadvantaged in comparison to other groups. of more visitors hinges on the total visitor costs and the intrinsic attraction of PMCs relative to - 106 - Figure 3.Is PACIFIC AND COMPARATOR REGIONS-AVERAGE ECONOMY FAPFS, 1991 0- ,e 0.6 .1.4 [ * 7.PN0 DOMESTIO " F.6 . e. NZ DOMESTIC 14 4. DOMESTIC a 0 t PACIFIC RIM TO fl0FIC ISLANDS 10.6 0.4 -Oa. AUST DOMESTIC TOURIST 4* S. INDIAN OCEAN ISLANDS . 4 2. INTRA-RX0IF0 RE1ON S. AUST INTERNATIONAL 0.1 914 02 * 0 0.5 1 t. 2 2.5 8 8.5 4 4.6 6 6.5 S Average Distance (Miles '000) Fimure 3.2: PACIFIC AND COMPARATOR REGIONS-AVERAGE EXCURSION FARES, 1991 0) 0.5 V) 0 7. PNO DOMESTIO 0.4 - ¢ * . NZ DOMESTIC U 2. INTRA-PACIFIO REGION * 0.3 - * 0. AUST DOMESTIC TOUR18T 14. V DOMESTIO *. PA*OFiO Rim TO AIIFiO ISLAN0D hI g 0.2 F. INDIAN OCEAN ISLANDS 4 S.1 0. AUST INTERNATIONAL '10.1 0 4 <<0 0.5 1 tS 2 2.6 8 8.5 4 4.5 6 5.5 S Avorage Distance (Miloe '000) - 107 - ligure 3.3s iPACIFIC AND CONPARATOR REGIONS-ESTIMATED AND ACTUAL o ECONOMY AIR FARES o 4* + 4 ~~~~~~~~+ 5 ~~~~~~+ U + *iA r tS 9 DS s ^ . ^ & ^ Afl 4- ~ ~ ~ ~ + | ~~~Estfimated Return Economy Air Fares (A$'000) Fiture 3.4s PACIFIC AND COMPARATOR REGIONS-ESTIMATED AND ACTUAL EXCURSION AIR FARES 2_ -, o* + 2- U-- _ 02 2 0.4 0.0 0.I ' $ + + Estimated Return Excursion Air+* a ,..4 0 02 Ia 0 0 f in th sample ~~ ~Estimted Return Exconomy Air Fares (A$'000) Finu* 3.4:fl PAIFICd AN COtPhAT r REGIOS-SIAE AND ACTUAL - 108 - alternative destinations. For tourists, the local cost of key facilities-hotes, resort/isure activities-noods to be that much lower relative to competing destinations for which volumes and proximity to source markets allow lower air fares. -109- Endnotes 1. In Part III it is taken as understood that Papua New Guinea (PNG) is included in the set of PMCs. 2. Local and coastal shipping services are outside the scope of this study. 3. Touche Ross (1984). 4. "FAK rates" is a general term for freight rates offered to freight forwarders for consolidated cargo. General cargo rates refer to rates for cargo not otherwise specified on the freight schedule. In the case of the Pacific Islands, both rates can be viewed as "typical" rates. 5. This test employed the use of regional "dummy" variables in the regression analysis of the shipping rate relationship (see Annex 5). 6. The size of the effects are reflected in the values of the calibration parameters (coefficients) a and b, which represent elasticities. The estimates of these (0.15 and -0.04, respectively) seem low but are consistent with those found in other studies. In particular, a study of rates for 8,100 pairs of ports to and from the US in 1979 gave elasticities of 0.22 and -0.056. See Jansson and Shneerson (1987). The elasticity levels indicate that doubling shipping distance increases unit rates by 30 percent, while doubling route volume decreases rates by 8 percent. 7. A more detailed examination of air services for the PMCs is presented in Part IV of this report, in the context of cooperative arrangements among the airlines. 8. Composition by trip purpose, namely business, tourism and visiting friends and relatives (VFR), requires analysis of immigration arrival records and is not consistently available. 9. Passenger volume data on individual routes or sectors-on an uplift-discharge or true origin destination basis-are not readily available. 10. World Tourism Organization (1989). 11. Other connections to Tarawa are available over Nauru, and Air Tungaru (of Kiribati) may introduce jet (737) service between Tarawa and Nadi, possibly under a lease arrangement with Air Nauru. 12. Economy fares have almost no restrictions. There is no minimum booking period, the validity of the ticket is generally for up to one year, they can be used on all services, flight changes are allowed and cancellation carries no penalty. For international services, there is the added advantage that unlimited transit stops can be made, and a variety of alternative routings are often available for those routes with low direct service frequencies. 13. Excursion fares offer significant discounts but have tight restrictions on booking period, advance purchase, cancellation penalty, changes in itinerary, length of stay and intermediate stops. Nevertheless, they are by far the most common ticket in general use, especially for tourism, and they represent an estimated three-quarters of all travel from Australia to the Pacific Islands. -110- 14. In terms of statistical methodology, the airfare relationship is a simple log linear model. Regional differences are tested for by the use of region specific dummy variables. 15. Table 3.3 also shows that there is the difference between economy fares and excursion fares for domestic routes in Australia and New Zealand, in comparison to domestic routes in Papua New Guinea. This reflects the heavy discounting which has resulted from airline deregulation in Australia and New Zealand. Although Western Australia and Papua New Guinea routes have comparable seat capacity, the average excursion fare within Western Australia is about 66 percent of the full economy fare, whereas the average excursion fare for Papua New Guinea domestic routes is over 80 percent of the economy fare. 16. The responsiveness ("elasticity") of air fares to distance and traffic density are estimated to be 0.68 and -0.07, respectively for excursion fares and 0.79 and 0, respectively for economy fares (economy fares show no statistical dependence on traffic density). These figures mean, for example, that a 100 percent increase in traffic density, is associated, on average, with a 7 percent decrease in excursion air fares. It should be noted, however, that as traffic density increases, the fare mix between economy and excursion changes. Much of the reduced unit seat costs from larger aircraft on high density rc tes is captured by substantial increase in the share of excursion and other discount fare types. aznce fares and costs are treated equivalently, the results need to be interpreted with caution. More rigorous econometric analysis is required to exaniine market demand and supply conditions. 17. Availability of access to excursion fares on specific services, however, may be limited. - III - PART IV REGIONAL AIRLINE COOPERATION - 113 - CHAPTER 1 INTRODUCTION AND BACKGROUND A. THE PROBLEMS OF SMLLNE be low, but frequency is low and so passengers AND REMTESS are inconvenienced. Alternativdy, if small aircraft are used, frequency is high, but so are 1.1 Cooperation in transport, and especially airline seat costs. Apart from some routes to aviation, is a long-standing issue in the South and from Fiji, which are moderately dense, the Pacific. The island States of the region are all routes in the South Pacific are quite thin. Thus, small, many consist of widely dispersed islands, even with the most efficient aviation possible, and they are separated from each other by long generalized costs are bound to be relatively high. distances. For their size and income, they have a high dependence on aviation. Air services are 1.3 A related problem is that of small traffic necessary for internal communication, for travel volumes both for individual PMCs, and for the between the States, and for international region as a whole. This gives rise to tourism. An efficient aviation sector is diseconomies of low scale. It is often stated that important to economic growth; in particular, it there are few economies in air transport, at least is essential for external competitiveness that the so far as operational costs are concerned.' total 'generalized' cost of air service, including However, this is true only when a minimum the cost to airlines, and the time costs to users, efficient scale is reached. There may be few be kept as low as possible. The air services economies in moving from fifty to one hundred networks for the Pacific Island Member aircraft, but there are considerable economies in Countries in the Pacific Rim region are shown in moving from one to two aircraft or from a small Annex 1.1 to a larger aircraft. For the small airlines of the South Pacific scale economies matter. This has 1.2 Air Service costs in the South Pacific implications for the importance of region are shaped heavily by market conditions, cooperation-a small airline will need to contract in particular size and distance.2 The island out, or cooperate with other airlines, for many States, including the Pacific Member Countries functions that larger airlines would undertake in- (PMCs) of the Bank, are small in size and their house. In addition, smallness brings along air travel markets are thin (i.e., low traffic problems of indivisibilities. Some airlines do volumes over relatively long distances).' These not require four or five aircraft of an appropriate factors give rise to higher cost structures. type-rather they require 3/7 or 4n of an Thinness of markets leads to high costs since the aircraft. This puts greater demands on cost per seat or per passenger of air travel falls cooperative arrangements. with the size of the aircraft used. There is a trade off between airline costs (and hence, air 1.4 A further problem in the South Pacific is fares) and the cost of convenience (time) to the the atypicality of air routes. The ideal aircraft passenger. If large aircraft are used, fares can would be small with relatively long ranges. By and large, these do not exist, or if they do, they - 114 - are special aircraft with high operating costs, if the form of cooperation can result in There are scale economies in building aircraft, inefficiencies being removed, the overall cost and manufacturers concentrate on large aircraft disadvantages would be lessened. for long hauls, and small aircraft for short hauls. The larger aircraft can operate South Pacific 1.8 A Regional Airline. The common routes (though sometimes airport facilities are a response to these problems has been to suggest constraint), but they are difficult to fill; smaller various institutional forms within which the size aircraft often are inappropriate for the long stage problem could be addressed. The most lengths. This cost penalty is intrinsic to the important of these is the concept of a regional geography of the region. airline (which at one stage, there was the possibility that Air Pacific might become). The 1.5 Limited experience of some South Pacific idea is to have one airline serve a group of airlines is another factor affecting cost Pacific States, owned jointly by them.' structures. Moreover, all airlines operate from developing countries with limited skilled labor. 1.9 Aircraft Leasing. Another sugested (Indeed, some Governments are keen to arrangement is a Pacific aircraft leasing encourage their airline as a way of building up company which would own aircraft and lease skills in a technologically advanced, tradable, them to individual airlines on the basis that scale industry.) economies would be achieved in the ownership and maintenance of aircraft, Tourism Council of 1.6 In spite of the importance of keeping the South Pacific (1989). Yet since only about costs low, all PMCs have their own airline, in 10 - 20 percent of an airline's total costs are one form or another, many of which are very related to the ownership and maintenance of small, and some incur significant losses. aircraft, potential gains are modest. Moreover, Several airlines need to fly sectors without individual airlines have different requirements uaffic rights, resulting in poor seat occupancy for aircraft and coordinating these requirements (oad factor), and some airlines find it difficult would be difficult. Thus, there has been no to obtain good utilization from the aircraft. For clear move towards operationalizing the Pacific the traveller, movement around the South Pacific leasing idea. can be slow due to low flight frequencies combined with many indirect services. Several 1.10 Ad hoc Cooperative Arrangements. of these problems appear to be amenable to The more ambitious proposals, for regional alleviation by expanded cooperation. airlines or leasing companies have been overtaken by ad hoc developments. Aircraft have been a major problem for the South Pacific B. THE COOPERATIVE RESPoNSE: airlines, which want access to modern, efficient PROPOSALSAD ACTUAL ARRANGEMENTS aircraft, but do not have the markets to make effective full time use of them. To resolve the 1.7 Cooperative arrangements are seen as a problem, the individual airlines have made way to lessen some of the regional air transport agreements to share aircraft with a number of problems. To deal with smallness, cooperation Australian and New Zealand airlines, and is a way to directly reduce the smilJ-scale cost between themselves (see Box 1.1 for a list of disadvantages. However, even if the airlines current arrangements). The airlines have shown cooperated to the point of becoming one airline, considerable ingenuity in devising arrangements they would not, as a whole, constitute a large or to make effective use of 'lumpy' aircraft even a medium size airline. Other problems capacity. This is perhaps the most obvious impose cost penalties which cannot be avoided manifestation of cooperation, but cooperative (for example, airport suitability problems). But, agreements also cover a number of other aspects - 115 - Box 1.1: AIRCAF SHARNG ARRANGEMENTS IN THE SOUTH PACIFRC Country Airline Partner Airline Aircaft Fji Air Pacific Qantas (Australia) Boeing 747 Vanuatu Air Vanuatu Australian (Australia) Boeing 727 Westem Samoa Polynesian Ansett (Australia) Boeiny 727 ToDP Royal Tongan Solomons Islands Boeing 727 Solomons Islands Solomon Islands Royal Tongan Boeing 737 Kiribati Air Tongaru Air Nauru Boeing 737 of operations, such as maintenance, and C. STRuenRE OF THE ANALYSIS managerial support. In addition, there are seat purchase and code share agreements which make 1.13 Airline cooperation should be facilitated for bettr utilization of aircraft. by Governments ensuring that the operational, policy and regulatory environment in which 1.11 There have been problems with these aviation takes place enables gains from agreements. In some cases, the parties to them cooperation, consistent with economic have become dissatisfled and ended them, but efficiency, to be achieved. then typically sought replacement agreements with the other parties. Dependence can also 1.14 The analysis of cooperation proceeds by become a problem." Most cooperation is on a examining the following matters in turn. The bilateral, airline to airline basis. While these objectives of individual States and constraints arrangemeSnt may appear unsystematic, they facing them (Chapters 2 and 3), actual have the very real advantage that they are in cooperation that has taken place (Chapter 5), place; it has not been necessary to wait for gains from further cooperation (Chapter 6), and multilateral institutions to be established. finally, interventions which enable cooperation (Chapter 7). The forms of cooperation that can 1.12 Approaches such as the regional airline work most effectively are best determined by and regional aircraft leasing company involve an those actively involved in aviation in the region. institutional solution to cooperation; the solution At the same time, there can be risks of market would be imposed rather than emerge. The power with cooperation and therefore airlines problem with these approaches is that they need also need to have incentives to pursue efficiency not be in the interests, actual or perceived, of all and, where appropriate, through cooperative States whose agreement is necessary for its means. success. While these approaches offer reduced airline operating costs and better coordinated schedules, they may not meet the requirements of the individual States.! - 116- CHAPTER 2 ISSUES IN AIRLINE COOPERATION A. COOPERATION AS A B. COSMS OF COOPERATION MANAGENT DECON 2.3 Direct familiarity with the airline industry 2.1 Cooperation is best viewed, not as a set is necessary to assess the various transaction of specific prescribed foi .s, but as a possible costs and risks associated with cooperative option for airline management in the manner it actions. For example, low aircraft utilization chooses to obtain and employ some of its inputs rates may suggest that airlines could cooperate (such as aircraft, training, maintenance) in by sharing aircraft and thereby keep capital costs producing air services. down. But, cooperation may increase some costs of doing business (for example, insurance), 2.2 There may be good reasons why expand risks and reduce gains from cooperation often does not take place, for specialization. Highly specific external example, government imposed objectives may be recommendations about forms of cooperation broader than, and in conflict with, simply low are, in general, inappropriate. Airlines are the cost air transport. Also, there may be parties to cooperation and they need the constraints on, and lack of incentives in relation discretion to adopt or reject particular to, cooperation. Furthermore, where there is cooperative arrangements. potential for various types of market failures (for example, monopoly airlines using their market 2.4 At the same time, airline managements power, and external benefits or costs from have their own objectives. Therefore, tourism and unemployment), some forms of governments should structure incentives to cooperation may not be in the interest of some airlines and their managements, to pursue countries. For example, cooperation may offer government objectives. It follows that lower air transport costs, lower fares, more governments need to monitor the overall tourism, and gains to the region as a whole, but performance of their airline (for example, by an individual country may lose, with tourists comparing leading indicators with industry bypassing it to reach others. In this case, it is norms). not in the interest of the country (which may lose) to cooperate-unless there are ways for other countries to provide compensation (see C. EFFIcIENT COOPERATION Box 2.1). 2.5 Cooperation is a way of efficiently achieving objectives, it is not an objective in its own right. Thus, if an airline is given incentives to pursue minimum costs it will seek out those cooperative arrangements which lessen - 117 - Box 2.1: THE "FREEDOMS" OF THE AIR The rights that an airline may or may not possoss to fly between one country and another are often summaized on terms of the Freedoms of the Air". There are five formal freedoms, and an informal, sixth freedom. They are illustrated as follows. First Freedom The right to overfly a territory without stopping |RHomStete |- [Foreign State Second freedom The right to land for tochnical and non-traffic reoms s om Sta7te | |FrlnSa Third Freedom The right to set down traffic from the Home State How t Foeg 80m State Fourth ^reedom The right to pick up traffic bound for the home state. |Bomtate | Foreign State. Fifth freedom The right to carry traffic between the Foreign State and another Foreign State. |Hoe State Foreign State Foreign State Sixth State The (informal) right to carry traffic between Foreign States via the Home State. Foreign State I Bom State I Foreign State Cabotage Carrying traffic within the ove State. I Rome Sta-te j Individual countries negotiate, usually bilaterally, to determine which rights will be granted to airlines from the various countries to fly on intemational routes. Only first and second freedoms are generally available-other freedoms must be established through negotiation. - 118 - costs. However, cooperation is just one of enter agreements which are contrary to the set many ways in which an airline may be able to objectives. Thus, the factors underlying reduce its costs. decisions as to whether cooperation takes place or not need to be identified. To do this, the 2.6 Setting up an environment which objectives of governments and airlines (and facilitates cooperation where it enhances whether the airlines have incentives to pursue efficiency, and giving airline management cooperation where it is efficient) need to be incentives to pursue efficiency, is a difficult and considered' and constraints that race airlines and demanding task; it is not achieved perfectly governments identified and assessed. Where anywhere. Where objectives for an enterprise cooperation means gains to the group (of are simple (such as maximizing profits), an countries or airlines) as a whole, but losses to incentive structure can be created which serves some, options need to be devised to provide for this quite effectively (for example, within the compensation arrangements. These issues are airline). However, where objectives are more examined in detail below. complex (as those given to aviation regulators or chief executives of government-owned airlines), incentive structures often break down. Yet, if airlines do not have direct incentives to pursue set objectives, they will not pursue cooperative agreements which advance them, and they may - 119 - CHAPTER 3 CONSTRAINTS TO AIRLINE COOPERATION A. REASONS FOR NON-COOPERATION but losing tourists through higher fares) but other countries are very likely to lose. One way 3.1 In examining the scope for more for a country to guarantee that it is not losing cooperation between airlines, it is useful to out is for that country to have its own airline. explore the reasons for non-cooperation. In In principle, it is possible that it will be to this some cases of non-cooperation, there may be country's advantage to have an airline, even if real barriers, while in others barriers may be this airline operates a loss, since the loss it artificial and relatively easily lowered. incurs is more than made up for by tL - oenefits of competition in the form of lower fares. 3.2 There are several distinct reasons why However, this may be an expensive way of cooperation, which could potentially reduce dealing with the perceived problems of a costs, does not take place. These may be potential monopoly, since it may be possible for grouped under four headings: the country to use its power over aviation rights to force fares down or share in the profits of the 3 market imperfections at the airline level; monopoly. * broader objectives of government; Employment and Training - incentive structures; 3.4 Where there is unemployment or underemployment in a country, the social * negotiation costs. opportunity cost of labor (the "shadow" wage) is likely to be less than the market wage. If so, These are discussed in turn below. the country will prefer that some services be produced domestically rather than imported. Airline services are a type of service that the B. MARKEr FAILURES AT THE home country can produce directly, and hence it ARLINE LEVEL may prefer to have a home based airline and employ nationals, rather than rely on another Market Presence and Concern with Monopoly country's airline, even if this would be cheaper financially. If this consideration is relevant, it 3.3 In a free trade situation, where any would be applicable to all industries, not just airline could serve routes in the Pacific, a airlines, and the best approach would be to monopoly may emerge. This monopoly might tackie the problem directly (for example, be able to use its market power, and raise through subsidies to employment), if this is prices. The country which owns the monopoly possible. may gain or lose, (gaining through higher profits -120 - 3.5 In aviatior, specialized skilled labor is cooperation. However, tourism is an area where involved and this may be in short supply, not the benefits are prone to exaggeration (see excess supply, in the South Pacific Island States. Box 3.1); policy intervention in this sector The shadow wage for trained personnel could be requires careful assessment. Some of the above, not below, the market wage. If so, care possible external benefits are noted below in the is needed in encouraging an industry which uses context of the use of airline services to support some labor that is scarce. them. 3.6 At the same time, airlines (along with Foreign Exchange other industries) may create useful external benefits through training labor.' By employing 3.10 Tourism is popular as an earner of people in areas where they can develop skills in foreign exchange. As foreign tourists arrive, high demand throughout the local economy, an they spend money on domestic goods and airline may be creating benefits for which it is services, and add to foreign exchange receipts. not paid (when employees leave, it loses the Net receipts are significantly less than the gross benefits of training them). This would be receipts, especially for South Pacific countries, another reason why the shadow wage could be as many of the goods sold to tourists must be less than the market wage. imported.'1 Any gains in foreign exchange from tourists (and from providing the air 3.7 Quantification of these externalities and transport they use) is appropriately valued at the how they can be best taken into account are shadow price of foreign exchange. Only if this difficult to gauge. However, Governments may differb from the market price will there be any take these human resource factors into account net external benefit. The two prices might differ in having a home based airline. if exchange markets are not free to adjust, or there are significant trade distortions such as Prestige and Defence Benefits tariffs, quotas and input subsidies on other goods. 3.8 Sovereign governments sometimes operate airlines for prestige or defence reasons. 3.11 Adjustment problems can arise if there However, it should be noted that benefits from are fixed exchange rates, and there is either an airiine as a national flag carrier, or as a excess demand for foreign currency, or excess defence resource, may have high opportunity demand for the local currency. But this is not a costs in terms of achievement of the country's major issue in the PMCs. Trade distortions other objectives. arise when imports are subject to tariffs. If more foreign exchange is earned from tourism, then the exchange rate rises, and imports of C. GOVERNmENT OBJECTmES: other goods and services also rise. The value THE INTERACTION OF TouRm that consumers put on these, their domestic AND AVIATION market price, includes the cost to the country, the trnded price, and the government obtains the 3.9 The tourism industry is a major potential difference as tariff revenues. The value of the solurce of economic growth for the PMCs and foreign exchange, in terms of goods and services air transport is a major factor in influencing the at domestic prices, is underestimated by the flow of tourists.'" Moreover since there can be market exchange rate. Trade distortions, positive and negative externalities, associated especially as a result of import tariffs, do apply with tourism, and it may be easier to affect to the PMCs. However, estimates of their full tourism flows through aviation than directly, this impact for individual countries are difficult to link is of policy interest and can constrain air ine establish. - 121 - Box 3.1: SOMI MYTHS AOUT TOURISM Foreign tourism exponditure is geaily to the benefit of a country, however, the benefits it creatoa can be grossly exaggerated. Some myths are: "Tourism benefits are measured by the size of the expenditure by foreign tourists.' This is not correct because it fails to take into account the costs of serving the tourists. Imports are required, goods and services must be diverted from other uses, and labor must be divert from other industries. Normally therm ar not benefits, but they will be a fraction of the gross benefits. 'Foreign exchange earnings of tourism a the sam as its impact on the current account. Quite apart from th fact tht impos will be necemay to ervo touris, other faco entor. More foreign exchange earnings wiU push up the exchange ratoe, encourage more imports, and discourage exports. The net impact on the currnt account is likely to be much smaller, and possibly negligible. 'Tourism results in significant increses in overall employment, especially through multiplier effects.- Tourism can contribute to increased employment, especially whwe there is unemloyment However its not contribution is reduced by its indirect impacts. A foreign tourism boom wil raise the exchange rate, leading to lower employment in other tradeable industries. The not mpcis difficurt toesimae,bgut it ishlkely to bemuch low than the immediate impat. 3.12 A premium on foreign exchange receipts contribute towards increasing overall (and encouragement of industries, like tourism, employment and if the shadow wage is less than that attract them) may be sound. However, the market wage in the formal sector, additional current account deficit exposure, as experienced tourism will create net benefits. by several PMCs, does not constitute a valid reason for a premium on foreign exchange. The Environmental and Social Impacts source of deficits is more in domestic savings and investment behavior, than in overvalued 3.14 Tourism has many environmental and exchange rates. social impacts; some are generaliy regarded as desirable, others as undesirable. Tourists add to Employment congestion of facilities, but make extensions to facilities viable, giving benefits to residents. 3.13 Many PMCs do perceive tourism as Tourists can over-use fragile parts of the being desirable as a way of increasing physical environment, and thus degrade it. As employment. Expansion in tourism usually against this, they can make preservation, which results in a direct increase in demand for labor. may not have occurred otherwise, a viable The impact it has on overall employment is not proposition. Tourists may weaken the local clear, since some workers will be hired away cultural identity, but they can provide a market from other industries, and rises in exchange for local crafts. In some cases, tourism is rates will result in contraction of other export regarded as a threat to the way of life. Workers and import competing industries. With the are attracted into the market sector, leaving their underemployment in several PMCs, tourism will villages and their traditional industries; - 122 - communities may perceive a loss in their cultural frequencies. The airline industry is more identity. General assessments of whether these concentrated than the tourism industry, which impacts are positive or negative are not possible; consists of many independent suppliers, and it the net outcome depends on individual value may be administratively simpler to grant judgments and circumstances. subsidies or impose taxes at the airline level. The problem is that it is less direct (non-tourists Prices, Taxes, and Terms of Trade Effects will be affected as well as tourists) and any benefits may be dissipated amongst others, such 3.15 An increase in overseas tourism as foreign airlines. For example, indirectly expenditure in a country leads to a change "taxing' aviation through tight capacity controls similar to a terms of trade improvement, and the and high fares may be to the benefit of foreign country gains. An increase in demand for local airlines on the routes, as well as the home resources, such as beaches, historical places or airline. areas of natural beauty, increases their value. Since each country has an element of 3.18 Implications for Airline Cooperation. uniqueness, the demand for tourism increases as If a PMC values tourism highly, then this places its price falls. Changes in complementary a premium on convenient and frequent airline inputs, such as those induced by better aviation services which connect effectively with others; links, shift the demand for tourism, and with this encourages tourism across the region. If a tourism being in less than perfectly elastic PMC is keen to maximize its share of tourism supply, prices rise. The owners of tourism expenditure, facilitating travel to substitute facilities gain. Thus, even if there are no true destinations, may dilute its share of tourism. externalities associated with tourism, a country may wish to increase demand for its tourism services through improving its air service links. D. INSTwrTTIONS AND INCENiVEs A country also has the scope to tax or subsidize tourism. By taxing it, it can use its market 3.19 In the aviation subsector, cooperation power, and increase receipts received, at the cost takes place at the national (Government) level, of some reductions in flow. and the airline level. The key organizations directly involved are the airlines and ministries 3.16 All countries, and especially PMCs, face of civil aviation, or their equivalent. dilemmas in their tourism/aviation polices. Two such dilemmas are: (a) whether to tax or 3.20 The outcome of cooperative efforts subsidize tourism (b) whether to do so at the depends upon the objectives of these tourism stage or the transport/aviation stage. If organizations, and the incentives structures that a country does not consider that there are they face. A private airline will normally have substantial gains to be made from foreign profit as its core, or only, objective. If airline exchange and employment,it may be more owners structure the incentives for management willing to take advantage of its market power, such that they seeks profit, cooperation will be and tax tourism. Alternatively, if it considers entered into where it lowers costs. In this that there are important externalities, it may respect, the private airline will perform well. encourage tourism by subsidies. However, it will also seek to cooperate with other airlines to increase profits in other ways, 3.17 Once a country has decided whether to for example by raising international fares. tax or subsidize, it can exercise its policy at the Normally, this is undesirable behavior, since it tourism or aviation level. Keeping air fares reduces benefits overall. However, in the South down may be an effective way of encouraging Pacific context, the owners of the airlines are the tourists-so too may be keeping up flight Governments/nationals of the PMCs, whereas, in - 123 - the main, travelers are not. Thus, by raising airline management to be independent, and fares a PMC airline is taxing tourists; this might pursue organizational objectives, such as or might not be in the interest of the airline's expanding the size of the airline. If this occurs, home country. the airline will often prefer to undertake tasks in house, possibly at higher cost, rather than 3.21 Airlines of the PMC are Government- contract out or cooperate with other airlines. owned and there are several ways in which broader government objectives may differ from 3.24 k.isting evidence indicates that the the narrow profit objective of a private, or government-owned airlines of the PMCs are commercially-oriented, airline. The most typical of public enterprises around the world. important conflict of interest is likely to arise Some are given explicit objectives, some are from tourism: a government may want low air not. For some, commercial performance is an fares to encourage tourism, while the airline can important objective, for others not. Some PMC profit most from a high fire. airlines publish financial statements, others do not. All have been set objectives broader than 3.22 Managerial Discretion and profit alone, such as encouraging tourism. Organization Objectives. Airlines, like other Explicit mechanisms, linking performance to companies, whether privately or government managerial rewards, are rare, though with some, owned, are under the day to day control of their informal mechanisms may be in place. For managements, and it has long been recognized some cooperative agreements, it is clear that that the objectives of owners and managers airline managers are unlikely to be rewardedfor diverge. Managers may seek greater size, putting them in place, even though they would be growth or internal satisfaction, at the expense of valuable. If the institutions which can make profit. The scope for this tends to be greater cooperation work have inadequate incentives to with a government-owned enterprise where the do so, it will not happen. discipline of the shareholders through the capital market cannot emerge. To address this issue, 3.25 Many countries are addressing the the rewards of ma.±agers need to be structured so problems of public enterprises. One option is to that they have a keen incentive to pursue profit. privatize, and another is to institute explicit Ways of doing this include relating senior objectives and incentive structures based on management remuneration to profits, performance relative to objectives. Another performance targets, and share ownership approach is to "corporatize' the enterprise: give schemes. Airlines which possess market power, it a corporate form, give prominence to profit as and are less subject to the discipline of an objective, but constrain (for example, through competition, have more scope to pursue price regulation) inefficient behavior that this "managerial objectives". might lead to. With corporatization of airlines by PMCs it is necessary to establish how 3.23 Government-ownedenterprises,including broader government objectives (for example, airlines, tend to be given very broad objectives tourism and employment) can be included within which may be in partial conflict; typically little a corporatized framework. The same issues explicit guidance is provided by Governments as need to be resolved with privatization. to how the conflicting objectives are to be resolved. Incentive mechanisms are usually not 3.26 To an important extent, airline specified, and may depend more on informal cooperation also depends on the regulatory 'hidden agendas', than on explicit performance. authorities, such as the ministries of civil These circumstances pose problems for airline aviation (or their equivalent). These authorities managers, and for Governments in evaluating have the power to allow or disallow various performance. It follows that scope may exist for forms of cooperation such as agreement to third - 124 - country seat purchases. The attitude of airlines E. NzGOTIATION COMS and aviation authorities Is shaped by the objectives that are set. For cooperative 3.27 Bargaining problems are often a reason agreements to be adopted, there needs to be why gains from cooperation are not achieved, mechanisms for facilitating them in a way that especially in the aviation sector. Countries may those entering the agreement gain, or be seen to agree on what is the best course of cooperative have performed well. For example, an aviation action, but do not move to it because of authority may be willing to cooperate by disagreement about how the gains are to be granting fifth freedom rights but choose not to shared. In the short term, refusing to agree, and do so as there may not be any way it can secure holding out for more in a bargaining context, payment, from those who gain from this can be perfectly rational. In the long term, such cooperation. As a result, potentially valuable a situation should not prevail if the parties are cooperation may not come about. rational, because some compromise is better thn none. 125 - CHAPTER 4 EFFICIENCY, COOPERATION AND COMPETITION A. GIVG MEANANG TO EnIcJENcY (a) Many users of the airlines are foreign visitors, whereas the airlines and those 4.1 Efficiency concerns the extent to which who enjoy the profits, or finance the objectives are achieved subject to prevailing losses, are nationals (mainly national constraints; the simplest approach to economic governments). These groups may be efficiency involves maximization of the total of viewed differently by the countries net benefits to consumers (travellers), producers concerned. (airline profits) and governments (revenues).12 (b) Foreign tourists are valued for the benefits they may bring and therefore, a 4.2 Two aspects of efficiency are particula:ly country prefers lower fares which relevant to the issues facing aviation in the encourage more tourists (to that country). PMCs. These are: allocative efficiency-which involves whether the types, levels and prices of (c) Costs faced by an airline may not be the air services that best serve travellers are being true opportunity social costs to the supplied (in terms of what travellers are country. If there is unemployment of prepared to pay) and productive labor, the shadow wage (opportunity cost efficiency-which involves whether the costs of of labor) may be less than the market air services are as low as possible. Cooperative wage actually paid by the airline. If the arrangements have the potential to affect both skilled labor required by the airline is in aspects. excess demand, the reverse could be the case. 4.3 EMciency at the Industry ¶nd National Level. Economic efficiency can be examined (d) Gains in net benefits by the PMCs as a with respect to a narrow or broad base. Taking whole may coincide with a loss for some a narrow base, efficiency is considered at the individual countries. It may not be easy level of the airline industry; interactions with for the gainers to compensate the losers. other parts of the economy are set aside. Improving the efficiency of air services is 4.4 In order to addWiss net benefits over the regarded as improving a country's overall whole South Pacific region, it is necessary to economic position. However, for the aviation take a broad view of efficiency, rather than one subsector in the South Pacific, simply ensuring limited to the airline industry. The different that the airlines (of one or more countries) countries in the region see the aviation problem perform efficiently may not lead to efficiency on as more than one of simply ensuring cooperation the broader base of the overall economy. This between airlines, or efficiency of the airlines, or difference between industry level and economy of just aviation and tourism. Unless the broader wide efficiency arises from the following: concerns of individual countries are taken 126 - account of, some valuable cooperation will not are justified-although this will reduce airline take place. Steps toward operationalizing this performance and traffic. Airlines may not use broader approach to economic efficiency are this scope; this depends on the objectives and outlined below. incentives they face. 4.8 There is also value in cooperation. B. COOPERATlON AND EmciENcY Airlines may be able to take advantage of economies of scale in certain functions (for 4.5 Risks from Cooperation. Cooperation example, maintenance) and achieve greater is not always desirable since competition usually aircraft utilization and better trained staff promotes efficiency." Indeed, in many through cooperation. In addition to reducing countries, for many areas of economic activity, costs, cooperation may also enable better cooperation is prohibited, or at the least products to be offered to the traveller, through discouraged." more convenient schedules, connections, and transfers. 4.6 In some cases, there is cooperation in aviation between countries to establish traffic 4.9 Just as cooperation may undermine rights, regulate safety, or develop operating efficiency (for example, collusion to raise prices standards. This type of cooperation need have above costs), it is true also that competition is no major impact on competition. Other cases of not always conducive to efficiency. When cooperation may have the effect of enhancing numbers of competitors are small, competition competition (whether or not that is the object), can lead to inappropriate product mixes, such as for example, when two countries allow 5th parallel scheduling of flights, which is Freedom operators (see Box 2.1) on a route inconvenient for passengers."5 between them. Airlines may also cooperate in ways which do not diminish competition and 4.10 Trade off between Competition and may indirectly increase it. For example, two Cooperation. Cooperative arrangements may airlines which serve different routes may enable costs savings to be made, or better cooperate in joint use of an aircraft, or arrange services to be provided, but reduce competition their schedules so as to make interlining and weaken the pressure to perform. As a result convenient for passengers. But in cases where cost savings may be dissipated. This trade off is cooperation is directed at supply conditions, for a particularly delicate one for the PMCs. The example, fares or capacity, it can be directly small size of the airlines, and the thinness of anti-competitive. Thus, any given cooperative their routes, makes inter-airline cooperation to agreement can have both positive and negative reduce costs and offer better schedules an aspects-it may reduce costs and also reduce obvious option. At the same time, competition competition. between airlines is weak, and the internal incentives to pursue cost efficiency are not 4.7 The Value of Competition. Competition strong. Lessening competition through inter- provides pressure on airlines to keep prices close airline agreements will weaken the already to costs, and to provide the types of services that limited pressure to perform well on the part of travellers want. Competition also provides some of the airlines. Some possible trade-offs pressure on airlines to push costs down as low are indicated in Box 4.1. as feasible-if necessary, by cooperating with other firms rather than doing everything "in- 4.11 A practical approach to the problem of house". If competition is weak, airlines have cooperation involves setting out the various the scope to allow costs to rise. After all, they possible forms of cooperation, looking at the can pass on cost increases, whether or not they benefits, and assessing whether they are likely to - 127 - Box 4.1: COMPETITION AND COOPERATION: SOME TRAFC TRADE-OFFs Three aspects of performance are considered here: cost efficiency, price/cost ratios and passenger convenience. Positive Desirable (+) and Negative Undesirable (-) impacts are suggested. Some common forms of co-operation are listed here: Form of Impact on Impact on Impact on Co-operation Costs Pnce/Cost Passenger Margins Convenience Baggage Handling + + ve Revenue Pooling - ve + ve (possibly) Seat Purchaes + ? Possibly - ve + ve Aircrft Sharing + ? Possibly - ve - ve (possibly) Code Sharing ? Possibly - ve + ve Maintenance + Staff Training + (possibly) + ve (possibly) involve problems of lessening competition or time, one cost reducing solution is to lease it out not. On this basis, guidelines can be established part time to another airline. Other areas have as to what forms of cooperation are desirable, less obvious solutions-for example, should an undesirable, or problematic. Again, given airline train staff "in-house", or in cooperation information limitations, external assessment can with others? The many different forms of only provide a general indication of the benefits potential cooperation are discussed in Chapter 5 and costs. Design of a framework under which below. desirable forms of cooperation are encouraged, and undesirable forms are not, is set out in 4.13 Where an airline cooperates with other Chapter 7 below. enterprises, including airlines, some costs may be reduced but other costs are incurred, principally: C. THE DEISION TO COOPERATE (a) Transaction Costs. These include the 4.12 It is rarely possible to state costs of negotiations, costs of travel, the unambiguously whether cooperation of one form costs of additional administration, etc. or another among firms will actually reduce costs. Managers are employed to work out and (b) Coordination Costs. In cooperating, it activate the best ways of getting tasks done,"6 is necessary to standardize some products in particular whether to have a task performed and procedures; this may lead to higher "in house", to contract out to non-airlines, or to costs for individual airlines. cooperate with another airline (which may or may not be a competitor). The most cost (c) Control Costs.' When one airline (the efficient choice depends on many factors. Some principal) gets another airline (the agent) areas have more obvious solutions than to undertake a task, it cannot be sure that others-for example, where an airline has an the latter will do it as well as it would do aircraft which is not used for 50 percent of the it itself. If an aircraft is shared, the - 128 - other airline may not always make it 4.15 Management Incentives. The way in available in time (there are always which an airline is structured, and the incentives excuses for lateness), or if an aircraft is faced by the managers, are critical for cost maintained by another, 'unavoidable cost effective cooperation. It is essential that airline overruns" may occur. Cooperation managers have the freedom, incentive and always involves risk which is normally responsbility, to choose forms that ure cost lower by doing things 'in house", where effective and reject those that are not. If managers can keep things more closely managers are actively seeking efficient under surveillance. performance of the airline as a whole, they will make good decisions concerning cooperative 4.14 A key role of airline managers is to ventures. Thus, addressing the structure of, and weigh up the benefits of cooperation with the incentives within, airlines, to ensure that they costs and the risks. Ideas which look promising are consistent with generally pursuing efficiency, ftom outside the airline may not prove cost Is the key factor in establishing a framework in effective; caution must be exercised when which efficient cooperation can come about. sugesting what forms of cooperation will be cost effective. -129- CHAPTER 5 COOPERATION BETWEEN AIRLINES IN THE SOUTH PACIFIC A. COOPERAnoN BETWEE AaJEs pervasive forms of cooperation in the region has been the sharing of aircraft-this is not 5.1 One of the most evident worldwide trends necessary, nor common, among larger airlines. in aviation has been increasing cooperation between individual airlines, especially among 5.3 The types of cooperation arrangements international carriers. Within the US, currently in place 21 can be divided into three cooperation between separate airlines is limited main groupings: by antitrust policy, but there has been a sharp reduction in the number of airlines, often by * use of aircraft, seat purchases, and merger, and the formation of "mega-carriers'. revenue pooling; Undoubtedly, this is in part a reflection of underlying economic pressures." The pattern * coordination of schedules and cooperative of changes amongst the international airlines has advertising; been different, even though it reflects the same economic forces. Ownership (many airlines are * ground handling, catering, staff training, still government owned) and the different and technical standards. regulatory environments at tlie international level are possible reasons for this. Though mergers Apart from these, other forms of cooperation, have been few, there have been several cases of such as management contracts and cross cross-shareholding arrangements. At the shareholdings, are present. operational/scheduling level, there has been integration of schedules and code sharing,'9 and 5.4 Cooperation takes place at two different joint developments of reservations systems. levels: between the airlines of the South Pacific themselves, and between the South Pacific airlines and larger airlines (mainly based in the B. COOPERAnON IN THE SOUTH PACIC Pacific Rim, in particular Australia and New Zealand). 5.2 In the South Pacific region, the forms of cooperation tried have been numerous, and the 5.5 Patterns of airline cooperation between airlines hawe shown considerable ingenuity in the South Pacific countries are summarised in resolving the difficulties they face. However, Table 5.1, which indicates whether cooperation with different challenges, especially the small of a particular form does, or might, take size of markets and airlines, the airlines of the place-it does not indicate its likely significance. PMCs have developed quite different forms of cooperation, generally at the operational 5.6 In Table 5.2 cooperation between the level.3 As an example, one of the more South Pacific airlines and other airlines is indicated. - 130- - Table 5.1: COOPERATION BluWEEN SOUTH PACIFIC AIRLINES, 1991 Royal Air Air Air Air Polynesian Solomon Tonga Air Caledonie Nauru Niugini Pacific Airlines Airlines Airlines Vanuatu Operational Joint Ownership A/c Full Time lemss a/c P Part Timlease a/c P X X X Co-operation in cboice of Airaft N MaintemUc_ X X Spar.. Holding Aircraft Scheduling X P X X Crew Scheduling P x x SeoaPurchaw X P X X X X X X Pooling P Sals and Marketing Pa.mger Scheduling p X AcceOS toSth Froedom P N X Reservation Systems Fares Sotting X X Promotion Advertising Ifrastructural Corporate Seminars X X X X X X Ground Handling X X X X X Catering X X x Staff Training X X Accounting Systems Multilateral Negotiations Technical Standards X Negotiation of Traffic Rights X X Other Cross Shareholdings P N X X N Code: X = Curent P - Possible N - No longer present Source: Survey of Airlines, Annual Reports. C. PATrERNS OF COOPERATON with airlines of the Pacific Rim (Polynesian with Ansett), especially in connection with 5.7 Several airlines have some form of maintenance and spare parts stockholding. cooperation in place which affects the provision There are some examples of aircraft and crew of aircraft. They may cooperate between each scheduling cooperation between airlines of both other (Solomon Airlines and Royal Tongan) or groups. Seat purchases, by one airline on - 131 - Table 5.2: CooIATIoN BEWEvN SOUWH PACIFIC AND PACINC RIM AIRIANES, 1991 Royal Form of Co-opeation Air Air Air Air Polynesian Solomons Tongan Air Caledonia Nauru Niugini Pacific Airlines Airlines Airlines Vanuatu Operational Joint Ownership A/c Full Time lease A/c X Part Time leame X X X X Co-operation in Choice of Airaaft X X X X X Maintenance X X X X X X SparesHolding X X X X X X Aircrft Scheduling X X X Crew Scheduling X X X Sat Purchas X X Pooling X X Sals and Market Pa_gOr Schoduing X P Acco to 5th Freedom X Reservations Systems X X X X X X X X Fare SStting X X X X Promotion X P Advertising infrutruchu Coorate Seminas X X X Ground Handling X X X X X X X Catering X X X X X X Staff Training X X X X X Accounting Systems Multilateral Negotiating X X Technical Standards X Negotiation of Traffic Rights X Other Shareholdings N X N N N Managerial Assistance N X X N Code: X = Current P = Possible in Future N = No longer Present Souce: Survey of Airlines, Annual Reports. another's flight, are quite common. There are South Pacific airlines rely quite heavily on the some attempts to coordinate individual airlines' Australian airlines for computer reservations schedules for passenger convenience. (For systems. They cooperate a little between example, Air Caledonie International links its themselves, but more with the Pacific Rim Noumea-Vila flights with UTA's flights.) The airlines, on fare setting. There is modest - 132 - cooperation on promotion and advertising. The 5.8 Several forms of cooperation are airlines take advantage of ASPA (Association of conspicuous by their absence, or at least, their South Pacific Airlines), to a lesser extent, the rarity. Cross-shareholdings do exist, but they OAA (Orient Airlines Association) (Air are minor and of limited importance. Pooling Niugini), and IATA (International Air Transport appears to be of minor importance, especially on Association) in handling industry wide issues. the routes internal to the region. This is perhaps Cooperation between airlines of both groups, in because many routes are operated by only one ground handling and catering, is very common. airline (and other airlines gain a presence by seat There is some cooperation between the South purchases). There is less joint advertising and Pacific airlines on staff training, but most rely, promotion than one might expect. The same is to some extent, on Facific Rim airlines for true of staff training, and the development of assistance. At a more general level, several rely accounting systems. Code sharing is a form of on the Rim airlines for the provision of some of cooperation often practiced by large airlines, but their management expertise (see Box S.1). it is not that common in the South Paciflc, possibly because the airlines wish to market themselves as specialists, rather than airlines with comprehensive networks. Box 5.1: CHOICE OF PARTNER: SOUTH PACIFC AIRLINE OR PAcnlc RIM ARLINE? 1. Aircraft Leasing: who has aircraft available at the right time? Rim airlines have larger fleets and more flexibility in making aircraft available part of the week. Two South Pacific airlines may have sufficient demand for one aircraft, and may be able to share. Co-ordination of aircraft schedules may be easier for them than if a South Pacific Airline co- operates with a Rim airline. 2. Managerial Assistance: who has the greatest expertise? The larger size, and greater experience of operation, means that managerial assistance is more readily available from Rim airlines. 3. Which airlines have simnilar tasks? South Pacific airlines face similar tasks and problems (small size, thin routes) and may be more adept at meeting some of the challenges than larger airlines from the Rim. 4. Which airlines are competing in the market place? Airlines which are actually competing in tbe market may find it difficult to co-operate, e.g. on sharing an aircraft. While the South Pacific airlines do compete with Rim airlines (especially Air New Zealand and Qantas) they do not compete very much with domestic airlines like Australian. They may feel that there are fewer conflicts of interest with Rim airlines. 5. Do some airlines have greater power in negotiations because of size? The Rim airlines are much bigger than t' South Pacific airlines, and they may use their advantages to gain a lion's share of the benetits from co-operation. At different stages, however, they have been quite competitive in offering agreements to the South Pacific airlines. 6. Which airlines can get tasks perforned cheaply? By virtue of their size, the Rim airlines may be able to undertake tasks (e.g., maintenance) relatively cheaply. They can take advantage of scale economies whereas even if all the South Pacific airlines co-operated, they would not be able to achieve the same scale. - 133 - 5.9 There appears to be more cooperatIon 5.13 There are also arrangements which can between airlines of the South Pacific States and be undertaken with airlines anywhere. Although those of the Pacific Rim, than among the South they are not common in the South Pacific, cross Pacific airlines themselves. The larger Rim shareholdings could be with airlines of any part airlines have a wider range of services available, of the world. Managerial agreements are more and are able to take advantage of scale common, and while any airline could supply economies in them. The South Pacific airlines expertise, existing cases only involve airlines of are small, and in some cases, inexperienced, the Pacific Rim countries. while those of the Rim are significantly larger and more experienced. 5.14 Beyond these aspects, there do not seem to be any dominant patterns of cooperation. The S. 10 There are some types of services that are ties that come about appear to be somewhat ad- best supplied by larger airlines. These include hoc, and depend on which airline has spare maintenance (there are scale economies in aircraft capacity at the time, or which airline is maintenance of any particular type of aircraft) able to offer a better management contract. The and reservations systems (for which there are contractual ties do appear fluid, if not fragile. large set-up costs). Proximity is moderately While there have been longstanding relationships important for these services, though it is not (for example, between Air Pacific and Qantas), critical, and the small airline can shop around the nature of these relationships has tended to amongst a few airlines for the best partner. This change. There have been some major changes applies also for staff training. of partner, for example, Air Vanuatu was originally operated by Ansett, though now it has 5.11 Other forms of cooperation are more a lease agreement with Australian. A one stage, dependent on proximity. This is especially true in the 1980's, it looked as though Ansett of aircraft sharing arrangements; for these to Airlines might have taken a dominant role in the make sense, they must be made with airlines South Pacific; its involvement is now very which have operations close by. Other limited (see Box 5.2). arrangements are specific to location, such as ground handling and catering. Thus, some 5.15 As indicated, there are several reasons arrangements are made between South Pacific for these cooperative efforts. Size is one; the airlines, and others are made with airlines of small South Pacific airlines can reduce costs by Rim countries to which the South Pacific airlines contracting out to larger airlines. Inexperience operate. (For example, Air Vanuatu shares an is another; managerial expertise may be in short aircraft with Australian Airlines, and it operates supply, and staff training schemes take time to between Vanuatu and Australia.) With these develop. Cooperation to assist in marketing and arrangements, there is some, though limited, developing an attractive network with convenient scope for shopping around. scheduled for passengers takes place, though it does not appear to be a dominant consideration. 5.12 When cooperative arrangements involve Finally, cooperation to limit competition and use aircraft scheduling, pooling and seat purchases, market power sometimes takes place, but it does the possible partners are defined by routes and not appear to be significant in the South Pacific. networks. For examaple, if Air Vanuatu is to This is perhaps because several of the thin routes purchase seat on a specified route, the only of region do not have more than one airline it can purchase from might be Solomons independent operator. Airlines. - 134 - Box S.2: A HISTORY OF COOPERATIVE INVOLVEMENTS: THE CASE OF ANSETT Ansett has boen involved in a variety of co-operative and other arrangements for the provision of air servicoo in the South Pacific. The arrangements are listed in chronological order by country. Papua New Guinea i. Operatea scheduled service to and within Papua New Guinea (1961 to 1973). ii. Operated Australia to New Guinea scheduled service on their own 'ebalf and on behalf of Air Niugini (1973 to 1976). iii. Become a 16% shareholder in 'r Niugini. iv. Now holds no shares in Air Niugini. Republic of Vanuatu i. Operated nd assumed economic risk of Air Vanuatu Ltd (a joint venture company 60 % owned by the Republic of Vanuatu and 40%6 by Ansett (through a subsidiary company). (1981 to 1986). ii. Operated by agremnt from Qantas (ANR .eg 201) the Australian capacity entitlement to Vanuatu. These services were operated with 'AN' designatik n but have ceased to operate. Independent State of Western Samoa D. PROBLEMS wrrI COOPERAnON performance of the other accurately; there is always an element of uncertainty, each .vill 5.16 While numerous examples of cooperation claim to be upholding its part of the bargain. exist amongst the South Pacific airlines, several This type of problem is more likely to arise with examples of rooperation have broken down. non-standard, as compared to standard, (See Box 5.3) There are several reasons for agreements. For example, if ground handling this. by one airline for another is not efficient, this will be obvious: negotiation or separation will 5.17 Non-viability. In a few cases, a quickly take place. Some of the cooperative cooperative venture has been tried, and later arrangements in the South Pacific are unique and discontinued, essentially because the venture was complex, and it is not easy for one party to not a financial or economic success. The check all aspects of performance by the other. cooperation between the Cook Islands and Ansett Thus, disagreements develop and cooperation (in the Cook Islands International venture) did breaks down. This type of problem may not persist, basic. lly because the uperation itself underlie the recent disagreements between was not economic. It is inevitable that some of Ansett, Polynesian Airlines and the Western the ventures tried will not be successful; this Samoan Government. need not be because of a failure of the cooperative arrangements per se. 5.19 Changing Cost Structures and Networks. A cooperative arrangement may 5.18 Moral Hazard or Control Problems. work well for a time, and then circumstances When two parties come to an agreement about a may change, leading to its demise. For joint venture, or where one party undertakes to example, an airline may have spare aircraft do certain tasks for another, problems can arise capacity which it leases to another. It may because of the lack of information. One party change its aircraft, or its own network, and its may not pull its weight, and not do all it can to spare capacity may cease to be available, make the venture succeed, and even may take necessitating the end of the agreement. This is actions which put the venture in jeopardy. One quite likely in the South Pacific where aircraft party may not be able to monitor the constitute a significant 'indivisibility", and - 135 - Box 5.3: AIRLiNE COOPERATION IN THE SOUTH PACIFIC: SOME VENTURES THAT NO LONGER ExIsT i. AnsettiAustralian (Trans Australian as it then was) and Air Niugini. Shareholdings and operations on behalf of Air Niugiai. Ceased because of independence of the country and the political desire for the airline to become independent. ii. Ansett and Air Vanuatu. A shareholding and economic risk-takin vonture by Ansett. It wu not possible for Ansett or The Republic of Vanuatu to agree on terms for renewal of the first contract. Among the matters in dispute were the failure of Ansett to introduce ni-Vanuatu as cabin crew (Vanuatu complaint) and the wish of Ansett to have Vanuatu participate in the risk sharing of the operation whilst Vanuatu wished to retain the royalty basis of its income. iii. UTA and Air Nauru. UTA charter-d B.737-200 capacity (wet) from Air Nauru to operate services between Noumea and Port Vila and Noumea and Walis Island via Nadi. The arrngement ceased when the newly established Air Caledonie International obtained its own aircraft, a Caravelle SE. 12. iv. Polynesian Airlines and Air Vanuatu. Polynesian Airlines, having recognzed that in order to provide a fully comprehensive service to the Cook Islands, it should serve the destination twice a week from Sydney via Apia, came to an arrangement with Air Vanuatu to part charter its aircraft to Air Vanuatu for the Sydney to Port Vila leg of the SYD-VLI-APW-RAR vw service. This arrangement provided several benefits: a. it gave extra capacity to Port Vila from Sydney at a day time slot when Air Vanuatu could not itself operate. b. it gave Polynesian Airlines additional revenues on the SYD-APW services c. it established a new air route Port Vila to Samoa. d. it enabled new, morm comprehensive South Pacific tour itineraries to be developed. where excess or under capacity are much more two competitors to cooperate, even if likely than in a large, dense airline system. The cooperation takes place on one route and airline which has been purchasing the excess competition on another. It is significant that capacity of another may well regard itself as several of the more complex cooperative having been let down. arrangements are between airlines which, under current national policies and air service 5.20 Two parties may set up a cooperative agreements, cannot compete (Australian and Air venture which is successful. However, they Vanuatu), or airlines which, in practice, are not may disagree about how the gains are to be likely to compete strongly (Solomons Airlines shared; one party may regard the original and Royal Tongan). agreement as unfair. (Collapse of the agreement between Ansett and the original Air Vanuatu 5.22 Costs of Cooperation. Cooperation may have involved this perception.) usually involves some costs that could be avoided if the tasks were all done in-house. 5.21 Cooperating with Competitors. There are costs of negotiation, of drawing up Airlines often cooperate with competitors; for agreements, and of policing them. Such costs example, one airline may provide ground do not seem to have been a major problem in the handling for its competitors. When this South Pacific. cooperation takes a standard form, with agreed prices, it usually works smoothly. However, 5.23 Cultural Differences. Differences in the where the form of cooperation is less standard, ways business is done, or the languages that are and uncertainty is greater, it may be difficult for spoken, can make cooperation more difficult and - 136 - costly. For example, one airline may require its They face a number of problems, of which small crews to have a knowledge of French, while scale, thinness of markets and, in some cases, others may not; this would lessen the scope for lack of experience, are severe. All these cooperation in staff training. These problems do problems are addressed through the cooperative exist in the South Pacific, but they are agreements in which they rarticipate. Most understood, and do not appear to have enter aircraft sharing arrangements which lessen constituted malor difficulties. the cost disability of small scale. Many have code sharing and seat purchase agreements, 5.24 Only some of the problems encountered which help them keep a presence in low density with cooperative arrangements come about markets. They have maintenance and because of cooperation per se. If a joint venture managerial assistance agreements which help ends because it is not viable, or because costs them overcome problems of inexperience and and circumstances change, it is not a failure of small scale. Most of the cooperation which cooperation as such, since much the same would exists in the South Pacific is primarily directed happen if one airline had responsibility for all towards cost reduction and product services. The moral hazard problem is inherent improvement; it is not simply to lessen with cooperation, as is disagreement about the competition and raise prices. sharing of gains, the transaction costs of cooperation, and the problems of cooperating 5.26 Each country and airline has its own with competitors. Cultural and language solutions to the problems it faces, and the problems could pose a problem no matter what solutions adopted differ. As a result, there is institutional arrangements are adopted. not the same homogeneity as would be the case with an overall, multilateral solution, such as a regional airline. Nevertheless, the present E. GENEAL ASsEsSME% T patchwork of arrangements is viable and appears to be serving the region quite well. 5.25 The airlines of the South Pacific participate in a wide variety of cooperative agreements possibly more than most other airlines. No doubt, this is because they need to. - 137 - CHAPTER 6 ASSESSING THE GAINS FROM IMPROVED COOPERATION A. NDICATORS OF INADEQUATE proceeds as follows: first, a general measure of COOPERATION costs and fares is considered; next, various indicators of performance such as a',craft 6.1 Considerable airline cooperation already utilization, load factors and lack of traffic rights, exists for the PMCs. Two outstanding questions are examined, then, indicators of convenience to are what is the extent of potential gains from passengers (parallel schedules and schedule uarther cooperation and, if these ae significant, coordination); finally, forms of cooperation what are the barriers that are preventing their which can indirectly improve performance are achievement? noted. 6.2 Assessment of possible gains from better cooperation in aviation matters is a difficult task. B. ARFARES AND GENERAL Part of the problem lies in the sketchy data on COST LEVEILS the finances and operations of the South Pacific airlines.22 This limits quantitative Airline Operating Costs assessment;' only the order of magnitude of potential gains can be indicated. Moreover, for 6.4 An overall indicator of whether there some forms of cooperation measures of might be gains from .ooperation would be performance are possible; for others, they are higher than expected costs of operation and not. Some forms of cooperation reduce costs fares. Airline costs are affected by a number of (for example, better aircraft utilization), others characteristics, such as route densities and stage do not lower airline costs, but improve the lengths, and any meaningful comparison of costs quality of air service (for example, better needs to adjust for these. However, relevant schedule coordination). Further, some forms of cost data do not exist for most South Pacific cooperation may im.iprove some indicators (for airlines, so it is necessary to compromise and to example, lower costs) while impairing others examine whether the prices of the final output, (for example, passenger convenience). Aircraft i.e., air fares, are higher than should be subleasing may reduce costs but at the expense expected. of lower schedule reliability. 6.5 Air Fares. An examination of whether 6.3 Approach. Notwithstanding these air fares for the South Pacific Islands are higher difficulties, it is possible to put together a than dtose available in other regions is presented picture of whether significant net gains can be in Part III of this report.' South Pacific air expected from (further) cooperation or not. The fares are found to be essentially similar in analysis which follows indicates that costs can be comparison with air fares in Australia, New reduced by better cooperation, although the Zealand, internal Papua New Guinea, and some evidence suggests modest gains. The analysis other small Islands.2Y These findings confirm - 138 - the view that the smallness and remoteness of may not be fully included in operating costs. As the South Paciflc States creates problems for against this, some airlines over-depreciate, and aviation which may be alleviated, but only in underestimate operating profits, only to record part, by further cooperation. "extraordinary" profits on the sale of capital items. 6.6 In considering the case for cooperation, however; other possibilities also need to be Table 6.1: CosT RECOVERY IN evaluated: SOUrH PACIFC AIRLINES, 1990 (a) Input prices may be higher; wages for skilled staff (pilots, managers) may be Airline Ratio-operating higher but wage rates of unskilled and revenues/operating costs semi-skilled labor, may be lower. Remoteness may build in cost penalties, such as with aircraft maintenance. Air Niugini 1.04 (1989) Overall, input costs should not be much Air Pacific 1.03 higher than elsewhere in the Pacific Air Nauru 0.43 region. Air Caledonie 0.89 Air Vanuatu 0.83 (b) Airilne efflcency may be less; it is Polynesian Airlines <1.00 possible that higher fares and costs might Solomons Island Airlines < 1.00 be reduced, not by cooperation, but by Royal Tongan 0.80 more efficient performance of individual airlines. Sources: Field surveys and Airline Annual (c) Small scale of operations may impose a Reports. cost disadvantage on the South Pacific airlines; even when they cooperate to the fillest feasible extent, partly because 6.8 On balance, it is likely that the South their scale will still be small in Pacific airlines are optimistic in reporting their aggregate, and because cooperation itself financial position. While some are making is not costless. profits (for example, Air Pacific), others are not, although they are trying to move towards (d) Market Power may be available to full recovery of costs. Thus, there is not much airlines; airline costs may be at the same evidence of pricing above cost.26 level as elsewhere, but airlines may be able to raise fares above costs; it is also Summary of the Ljmited Evidence on Costs possible that subsidies to airlines make air fare differences understate cost 6.9 Airfares, and most likely airline costs, differences. are higher in the South Pacific and in other similar regions, due, at least in part to low 6.7 Financial Performance. The financial traffic densities. There Is only limited scopefor position of many South Pacific airlines is cooperation in addressing this market difficult to document. Ratios of operating circumstance. revenues to operating costs are available and presented in Table 6.1. These ratios and their comparison need to be treated with caution. Capital cost of airline assets, including aircraft, - 139 - C. AIRCRAFr UT=IZATION earning routes. (For example, at existing traffic demand, Air Pacific must leave its Boeing 747 6.10 Low aircraft utilization may be an in Tokyo for an extended period). Cooperation indicator of potential gains from cooperation. In can lessen this problem, for example, when one the South Paciflc, there is a problem for airlines airline leases another's aircraft during its in achieving high utilization of equipment. This "unused" time. is partly because of indivisibilities-several have the routes and/or traffic densities to utilize only 6.11 The actual position of aircraft utilization a fraction of an aircraft of a technically for airlines in the region is summarized in appropriate type. One way to achieve Table 6.2. Utilization for the main type of "fractional operation" is to share aircraft with aircraft in each airline's fleet is given, along another airline. Apart from this, routes may be with some indicators of "normal" utilization, structured such that it is difficult for a single taken fom U.S. utilization rates; for the smaller airline to obtain high utilization because aircraft aircraft types, (for example, DHC 6) an are in the wrong place at the wrong time, and arbitrary level of 6.0 hours per day is taken as they may not be able to fly them on income- the "norm". Table 6.2: SOUTH PAcmc AIRLINEs-AIRcRmT UTILUZATION RATES (HOUR/DAY), 1990 Airline Type Hours Norm (US) Air Niugini A 310 11.06 11.0 F 28 6.14 6.2 Air Pacific La B 747 9.31 11.6 B 767 11.8 10.4 ATR 42 7.2 6.0 AirNauruLa B737 11.8 7.3 Air Caledonie B 737 9.3 7.3 International DHC 6 1.20 6.0 Air Vanuatu La B 727 12.0 7.6 Polynesian Airlines La B 727 DHC 6 5.1 6.0 Solomon Islands Airlines La B 737 9.9 7.3 Royal Tongan La DHC 6 3.8 6.0 La Undertakes part time leasing of large aircraft. Source: Field surveys, Annual Reports and Avmark; Quarterly Aircraft Operating Costs and Statistics, March 1990. -140 - 6.12 Utilization rates for the large aircraft are Table 6.3: SouTH PACIFIC AiRLINES-LOAD satisfactorily high. The airlines appear to have FACTORS, 1989 OR 1990 assessed their needs accurately, and undertaken arrangements to secure the appropriate availability level of these aircraft. Utilization Airline Load Factor (%) rates for the smaller aircraft, which most airlines operate as sole users, are, for some airlines, low." Air Vanuatu 66 6.13 Aircraft utilization data need to be Polynesian Airlines 65 interpreted with care. The forms of cooperation (aircraft sharing) relied upon at present to Royal Tongan 65 improve utilization are useful but there may be more cost effective forms of cooperation. High Air Niugini 64 utilization may be being achieved by flying aicraft on unprofitable routes. These routes Air Pacific 60 might be unprofitable because fares are too low, or because load factors are too low. The Air Caledonie International 58 evidence discussed above does not indicate that fares are low, however, load factors for some Solomon Island Airlines 55 routes may be.21 Air Nauru 52 D. LOAD FACrORS29 AVERAGE 61 6.14 The evidence on load factors, given in Table 6.3, reveals that the levels are typical for Source: Field visits and Airline Annual scheduled airlines. Air Nauru's load factor of Reports. 52 percent is fairly low, but all of the others are acceptable. There is little evidence, in load factors, of problems of lack of cooperation. aspect of cooperation at the country level is on Indeed, the airlines, through the cooperative the granting of traffic rights. Countries B and C arrangements, have been able to substantially could grant traffic rights between B and C for lessen the problems of flying aircraft with airlines of country A, and the airline system as unremunerative loads. a whole would be more efficient, in the sense that there would be fewer wasted seats. If airlines are flying unremunerative sectors, better E. UNREMUNATIVE SECTORS cooperation by countries could enable improved airline efficiency. 6.15 Most international airlines are obliged to fly some sectors on which they have no traffic Exchange Devices rights. Thus, an airline from country A may fly from A to B and then on to C, but it may have 6.16 There are ways around constraints on only the right to carry passengers from A to B traffic rights. One option is for one airline to and A to C (so called "3rd Freedom rights") not sell seats to another which has the rights (for from B to C (so called "5th Freedom rights"). example, country A's airline could sell seats to Since it can expect to lose some passengers in B, country B's airline for the B to C sector). it will have some spare seats from B to C. One Another option would be to 'code share' with - 141 - an airline that has the traffic rights. Thus, and profitable routing, if it could carry traffic Flight 100 of Country A's airline from A to C between B and C. Thus, lack of traffic rights via B might also become Flight 300 of Country could constitute something of a problem for C, when it operates between B and C. These several airlines. Quantitatively, this problem is devices mean that the otherwise empty seats can not likely to be large, since the sectors flown be used productively. without rights would be a small proportion of the total. Nevertheless, greater flexibility in 6.17 A group of routes with sectors on which operations and sales would reduce costs. particular airlines do not have traffic rights is shown in Table 6.4. However, in many cases, seat purchases, or code share arrangements, Box 6.1: TRAmc RIGHTS MND enable airlines to obtain revenue from the sector SEAT PURCHASES (see Box 6.1). S A e nc ights betweom Port Vila (Vanuatu) and Nadi (Fiji) 6.18 Tramc Rights. Although airlines have and b n Port Vils anucktad Now successfully lessened the problem of Zealand). However, it is efficient for it to unremunerative sectors by cooperation amongst o peardme trs t of its Honiara (Solozones)-Nadi and Houiaira-Aucklandl services. themselves, the devices used are imperfect. Im solution hs been for Air Vanuatu, which Airlines may find it difficult to cooperate witi does have rights to carry traffic on these sectors airlines with whom they compete on other ThAu, Solomofs Airlinsflight I1E708 (whfichis sectors. Further, only sectors which are flown also Air Pacific flight FJS) from Honiam to without traffic rights are observable. There may oNadi aso becomos Air Vanuatu flight NF1OS be attractive but unflown sectors, because of the from Port Vi to Nadi. lack of rights. Thus, as illustrated in Figure 6.1, an airline of country A may fly to C via D, because it has traffic rights from D to C, rather than via B, which would be a more economic Table 6.4: SHARING ARRANGEMENTS ON SECTORS WITHOUr TRAFIC RIGHTS Airline Sector Form of Sharing Arrangement Air Vanuatu Brisbane - Sydney Domestic Australian Air Nauru Honiara - Sydney Polyniesian Airlines Noumea - Sydney Air Nuigini Brisbane - Sydney Domestic Australian Solomon Islands Airlines Cairns - Brisbane Domestic Australian Cairns - Brisbane Domestic Australian Port Villa - Nadi Seat Purchase Port Villa - Auckland Code share Nadi - Auckland Limited rights Source: Field surveys and airline timetables. - 142 - sector demand are taken into account; Nadi does Flgure 6.1: TRAMC RIGHTS AND function quite effectively as a hub for the region. NETwoRK EmcIENcY Constraints on Scheduling 6.21 When different airline systems interact, some poor connections are inevitable. There are C limitations on equipment (for example, their B speed and whether thty can use particular airports), there are limitations on available airport slots and times of flights (airport curfews in the surrounding countries), and there are other network demands, such as requirements for connections elsewhere. For most of the perceived poor connections in the region, there A are good reasons why they exist, and in the D main, greater cooperation between airlines is unlikely to significantly reduce their incidence. Parallel Scheduling 6.22 As indicated above, airlines acting F. SCHEuLE INCONVENENCE independently are often motivated to schedule flights at about the same time, usually the most 6.19 A serious inefficient ramification of lack popular time, rather than spread services out of cooperation would arises if the schedules of over time in a way more convenient for the different airlines are designed without any passengers. However, for the South Pacific reference to one another. The result is region, inspection of schedules suggests that this inconvenient delays to passengers.30 is not a serious problem. Such examples as do Independent airlines do have incentives to occur usually can be explained, like poor coordinate their schedules, since more people connections, by operational factors (see Box will use an airline's flights if they know they can 6.2). make connections easily. 6.20 Since market densities in the South G. OTHER GAiNS FROM COOPERATION Pacific are low, limited frequencies (some routes are served only once a week) are inevitable and 6.23 Many forms of cooperation can lead to air travel in the region is relatively inconvenient. small cost reductions, and it is virtually This in turn puts a premium on good impossible to ascribe particular performance connections since the wait imposed by a poor improvements to them. But, in aggregate, their connection is long. Poor connections are not cost impact could be significant. There may be unusual in the South Pacific. For example, economies in joint purchasing or leasing of flights from Nadi to Kiribati depart before the equipment, or at least, in harmonizing choice of flights from Sydney to Nadi arrive; an overnight equipment. There are inventory economies in stay in Nadi is required for the Sydney/Kiribati spares holding. It may be cheaper to coordinate trip. However, there is not strong evidence of maintenance, possibly with much of it being navoidable' poor connections in the South done at a hub, or there may be gains in group Pacific, when the low (weekly) frequencies and negotiations with prospective non South Pacific - 143 - negotiations every airline will wish to portray Box 6.2: PARALLEL SCHEDULES itself as the 'marginal' operator). Those ar not a major problem in the South Potential for Further Cooperation Pacific, but a few examples exist. It is possible that they come about for operational or market 6.25 Evidence from field surveys of airlines in demand reasons, and would not be subject to the region indicates that a good deal of elimination if more cooperation took place. cooperation already takes place. Scope for more Route Services cooperation appears modest. It is significant that Port Vila - Nadi S services per week no airline in the region considers that there are 2 on Sunday important untapped areas of potential Port Moresby 6 services per weel cooperation available. 2 on Sunday Cairns - Port Moresby 7 services per week H. CONCLUSION 2 on Friday and Saturday Port Moresby - SydneyS services per week 6.26 The PMC airlines already cooperate to a 2 on Sunday very significant extent, both between each other and between themselves and selected airlines of the developed Pacific Rim countries. In some cases, it is possible that better internal cooperation would lessen the need for airlines for maintenance. Coordination in cooperation with Rim country airlines, and marketing may result in better promotion of the induce cost savings. However, there is no region, and a more attractive product for visitors evidence for this, and in many respects the Rim (easier transfers between airlines). There may country airlines may have more be economies in training personnel and in complementarities with the South Pacific airlines developing accounting systems. than the South Pacific airlines have with one another. Costs of Cooperation 6.27 Most of the specific performance 6.24 Cooperation does have its costs. By indicators which would be affected by lack of coordinating the purchasing or leasing of cooperation, such as aircraft utilization and load equipment, incremental gains from specialization factors, give no evidence of major gains to be are lost. For example, harmonization of aircraft reaped from better cooperation. There is some choice will lead to savings in purchase/leasing, evidence of low utilization of smaller aircraft economies in spares and economies in training. and it may be possible to achieve gains from However, it may mean that the aircraft are not cooperation for them. Several flights are made the most appropriate for some airlines. Even over routes where the operating airlines do not though the group as a whole may gain, some individual airlines may lose. This need not happen if incentives are given to internalize with the group (for example, if an airline is offered a service such as maintenance, on a marginal rather than average cost basis). However, while possible, constructing incentives for individual airlines to coordinate can be difficult (in -144- have traffic rights. In some cases, devices such as seat purchases do already lessen the waste in capacity from this, and although there is scope for further savings. 6.28 In summary, there Is already conslderable cooperation amongst the airlines of the South Pacific which reduces the problem of their small scale. Further aspects of cooperation may be tapped, however, they cannot be expected to produce large reductions in overall costs and fares or important gains In overall service quality. Lack of coopration Is not a serious problen - 145 - CHAPTER 7 AN ENVIRONMENT AND INCENTIVES FOR EFFICIENT COOPERATION A. TE CONSRANS ON COOPERATION (c) Taxation Policy. Taxes can create artificial incentives to cooperate or not 7.1 Fostering cooperation in aviation needs to cooperate, lowering the cost of the be examined against potential constraints. airline but raising costs overall. At Constraints can exist at the airline level and at present, for the PMCs, there is no the country level. apparent evidence of tax influences on cooperation Constraints on Airlines (d) Technical Differences. Differences in 7.2 Several types of constraints on airlines technical regulation can circumscribe the may apply which limit cooperation: scope for cooperation. If standards differ from country to country, it may not be (a) Legal Limitations on Cooperation. An possible, or economically feasible, for an airline may not be permitted by its own airline to undertake a task on behalf of Government, or a potential partner's an other. Technical differences do exist Government, to enter into particular between countries of the South Pacific types of cooperative arrangements. however, they are recognized, and to a These do not appear to be significant for certain extent, are being addressed the PMC airlines. (Enari, 1991). (b) General Economic Polides. Economic (e) Cultural and Linguistic. These are policies, which are applied generally (and natural barriers and they are present in not particularly to airlines) can prove a the South Pacific. They are not easily constraint on cooperation. For example, removed, and they can lessen the scope foreign exchange restrictions may induce for cooperation. For example, language an airline to undertake tasks at home, at differences make cooperative training higher cost, rather than cooperatively ventures more costly and cumbersome, abroad. Economic restrictions can also and may result in less cooperation. encourage cooperative solutions, even More general cultural differences, for where these need not be the lowest cost. example, business practices make Thus, foreign borrowing restrictions may coordination less easy. These prevent an airline from owning an constraints, of which there are some in aircraft, but it may make arrangement the South Pacific, are not easily with others to lease one. Various removed. economic policies, on balance, appear to be neutral !n their impact on cooperation. -146 - (f) Government Pressures. A country's (a) Regional Gain vs. Individual Country airlir.e, especially if it is owned by the Loss. It may not be possible for a national Government, is subject to country to gain, even though the region Government influence. This pressure as a whole gains, from cooperation. need not be explicit or systematic. At This is a fundamental problem which times the airline may be expected to: often occurs worldwide with aviation, serve markets that are not commercial, and other trade negotiations. For modify its purchasing policy, and market example, an airline (and itS home tourism from a particular region. country) may gain from being granted Airlines around the world, including fifth freedom rights on a sector but one those in the PMCs, are subject to of the counries on the sector may lose government pressures which affect out through loss of profits to its airline whether they enter cooperative or loss of tourism. Overall, air service arrangements. costs may fall, and tourism among the countries may rise, and so the group as a (g) Limitations on Traffic Rights. Airlines whole gains. However, unless it can require traffic rights, of some form or secure other concessions, the country another, to operatte in certain markets. which loses will veto the grant of traffic These rights may be formal rights, such rights.a2 as third, fourth, or fifth freedom rights, or informal 'rights' such as sixth (b) Scope of Trade Negotiations. In freedom rights, or acceptance of seat aviation, negotiations are sometimes purchase or code share arrangements. undertaken on a 'tit-for-tat" basis, The denial of one form of operation (for trading like for like, and this limits the example, fifth freedom rights) can lead to scope for cooperation. But more often, substitute cooperative agreements, such negotiations are broader, and one form of as seat sales. This issue is of moderate aviation concession (fifth freedom rights) importance in the South Pacific, since is traded for another. This widens the there is a significant proportion of sectors scope for cooperation, but it does not do which are flown without traffic rights, so completely, as the loser country may and a variety of ways are used to fill the not be able to gain from the aviation available capacity. concessions. Trade may take place on the basis of aviation concessions for non- In short, several factors do constrain the ability aviation concessions-this further of PMC airlines to cooperate. Removal of these explains the scope for cooperation. would assist cooperation. Plainly, if countries are willing to trade aviation concessions for money, the Constraints on Countries scope will be as wide as possible; cooperation which is to the benefit of the 7.3 Government Limitations. There are group can take place and compensation few immutable constraints on the countries made to any individual countries that whici would stop then from cooperating on may lose." aviat.on matters, or permitting their airlines to cooperate.31 However, governments are not (c) Broad Objectives. As noted earlier, perfectly well informed, nor can they always act Governments have broader objectives unilaterally. Thus, they do face limitations than simply minimizing airline costs. In which circumscribe cooperation. particular, they may wish to encourage tourism and increase employment, and 147 - they may use their airline to advance country may be the most cost effective these objectives. Thus, they may prefer way of obtaining capacity, but the the airline to employ home residents, at Government may be under pressure to higher cost, rather than cooperate with induce the airline to purchase its own. other airlines. Broader objectives are not Domestic pressures will sometimes limit necessarily a constraint on efficient a government's ability to pursue cooperation, but they do mean that, to cooperation with its neighbors. achieve it, matters become very much more complex, especially when the 7.4 The various constraints on cooperation instruments available to advance the are best addressed by Government, for example, objectives are limited (for example, better information can be collected (though at a where employment is an objective but cost). Other constraints, such as domestic employment subsidies are not feasible). political constraints, are always present to Design of practical ways of recognizing through support of an environment in which broader objectives, but also facilitating cooperation is feasible, and decision makers efficient cooperation, is a difficult task. have an incentive to pursue it. (d) Information Limitations. Governments may not be well informed about how B. SUPPOR11NG AN ENVIRONmENT various policies, such as one which FOR COOPERATION affects airline cooperation, will affect government objectives. For example, a 7.5 Aviation in the Sc;:, .>acific will, for the 5th freedom arrangement between foreseeable future, take ce in an environment countries could affect tourism to a which is at least partly determined by individual country-but its affect would be difficult countries, and operations will be carried out by to quantify. The country may not have individual airlines. iLt is unlikely that any a good measure of the value of additional multilateral cooperative solutions, either with tourism, and the trade offs involved-for countries creating an environment in which example, would an additional 1,000 cooperation between airlines is unconstrained tourists be worth a reduction in airline ("full deregulation"), or airlines all agreeing on profit of $50,000? Non-airline effects an overall cooperative solution ("single regional (tourism, employment) tend to be much airline"), will come about. Cooperation will less concrete than airline effects (cost take place between individual countries, and reductions, profits) and it is difficult to between individual airlines. compare them. If anything, there may well be an exaggeration of the value of 7.6 Bilateral and Multilateral Cooperation. tourism and employment aspects, leading Aviation rights are such that they are determined to a downplaying of the gains from and exercised on a bilateral basis. Groups of airline cooperation (see Box 3.2 above). countries can agree to form a regional regulatory structure. This is happening in Europe, but in (e) PoliticalPressures. Governments, while the context of the much broader range of EEC sovereign, are also subject to internal and agreements affecting most economic activity. external pressures. A government may This could happen in the South Pacific, though be convinced of the desirability of a it is not likely in the near future. The various policy, but may be under piessure from individual countries could, however, agree to voters or interest groups to adopt a paricular aspects of coordination, for example, contrary policy. Thus, leasing an on technical regulation, even though they are aircraft from the airline of another unlikely to create a single aviation market. - 148 - 7.7 Individual countries enterir.g (mainly) If they are commercial they will cooperate where bilateral arrangements, along with individual their costs are reduced and their revenues airlines making arrangements between improved. themselves, is a situation which is consistent with the efficient degree of cooperation taking 7.10 An Environment for Cooperation vs. place. Where there are gains from cooperation, Specific Prescriptions. A popular means of individual countries or airlines trading can addressing the cooperation issue in South Pacific ensure that no individual losses, and all gain, so aviation is to suggest specific forms that it might that if self interest prevails, cooperation will take take. Thbese include particular forms of industry piace. This requires that agreements can take structure for the airline sector, such as a single place between countries and airlines where airline, or an aircraft leasing company. (See aviation concessions can be traded for more than Boxes 7.1, 7.2). It is also possible to suggest just aviation concessions-for example, where a regulatory structures for the region-for country or an airline can be paid money to example, there could be a deregulated cooperate. Within this environment, it is still environment within the region, or, alternatively, possible that negotiations may falter because of a restriction of traffic to a 3rd/4th fredom the bargaining stances of the parties. basis. These specific suggestions could have merit, and they could come about as the result of 7.8 Notwithstanding this, there is a chance cooperation between states or airlines. that countries will not cooperate even though it However, as discussed earlier, it is not likely to is in their interest to do so. This raises the be helpfld to advocate specific forms of question of what* in a country's in.erest. For cooperation as being "the ideal", since the best example, a country may refuse to concede 5th forms of cooperation depend on many specific Freedom rights to others, no matter how much details which, in general, are not known it is offered in return for them. Non sufficiently outside the airlines. The best cooperation by one country can impose costs on approach at a general level, is to support an others. With a group of individual countries and environment which will enable the best solution,s airlines, it is inevitable that cooperation will be less than that which is economically warranted. To the extent that costs are thereby imposed on Box 7.1: A REGIONAL AIRLINE: others, these will have to be borne, as there is ADVANTAGES AND DISADVANTAGES no way of enforcing cooperation. Advantes 7.9 The Cooperation Process. Achieving * Scale economies lead to cost effectiveness the gains from cooperation is a two stage * Schedules would be coordinated process. First, countries themselves must enter * More bargaining power when dealing with arrangements with other countries which enable other airlines and goverments their airlines to cooperate and secondly, where Disadvantaes one country may gain and another lose, there * Airline would be dominant and possea must be a compensation mechanism. If, for mArket power, using this to raise fares example, the technical standards of one country * Lack of competition would give it scope to are agreed upon, some costs will be imposed on be cost efficient are ageduo,sm ot ilb moe nFirms with market power, and so imperative others which must adjust to conform to the new to make profits tend to be unresponsive to standards. Somehow, they must be compensated market demands, and may ignore the iftheyaretoagree to the change. Once an trequirements of some countries, especially if they are :a agree totecag.Oc nthe remote ones. environment exists which enable cooperation, airlines must have the incentive to unde-take cooperation where it is economically justih -' - 149 - to be pursued by those able to take decisions. that airlines have incentives to minimize their At a more specific level, some of the costs and own costs. benefits of different types of cooperation can be collated and processes for their detailed 7.12 Such an environment need not involve examination suggested. complete commercial freedom to airlines, nor "open skies" within the region. Complete commercial freedom might mean airlines which Box 7.2: A PACIiC AIRCRAT LEASING have strong market power on some routes COMPANY: ADVATAGEs AD raising prices to well above costs, or airlines DISADVANTAGES Colluding with one another to create and use market power. The balance between allowing AdYantages airlines, whi;.c ia. have some market power, the * Could lemon the problem of undeirabilitis in commercial freedom to minimize cost (desirable) WN *for South Pacific airlines but also raise prices (undesirable) is always a • My enawble scale economies in related s, sensitive one. Moreover in addressing it, there e.E. nuintemance * may improve the sirlines barganing power are many complications in regulating price Disadvantages behavior, or setting codes of competitive conduct. To achieve the best practical solution, * Airlines woeds ar divers, and would not be it may be necessary to compromise in one mot very wel by tho smal flet postssed by direction or another-for example, allow the * Ther would be major proble airlines scope to raise prices above costs in order coordinating demand for aircraft and to preserve incentives to minimize costs. avalabiity at parcular tum-all would want the same aurcraft at the sam tim. * Either tho countries sad airlines jointly ago 7.13 Regulatory Regimes and Cooperation. to only lea from the Lessing Comany, and There are several regulatory environments which they will create mR" pow urino can be consistent with efficient cooperation and sectorlading to high fanot es and y ly,i cost efficiency. One might be a regional "open perfrmance, or if it not gven a monopoly, it will break down as individual airlines ae skies" policy, under which all airlines from the offered better deals elsewhere. region have rights to fly and sell seats on all routes under their own name. An alternative environment would be one which any airline could operate on any route, but only 3rd/4th 7.11 Effidency and Cooperation. Several freedom operators could sell seats on a route. characteristics of a desirable environment can be Other operators would find it cost effective to identified. Fundamentally, it should be possible sell seats, on routes for which they have no for airlines to cooperate where cooperation will traffic rights, to the airlines which do. The lead to lower overall costs. If an airline can 3rd/4th freedom operators, in turn, would be operate on a route as cost efficiently as any able to gain by buying capacity from others. In others on that route, it should be possible for it both environments, the airlines involved have an to operate, and sell seats, on it; technical and incentive to minimize costs. However, the legal conditions should facilitate this. It should distribution of the gains (or profits) from not be the case that airlines are forced, by operating the rontes will be different (also regulatory constraints, to operate at inefficiently profits, and fares, will tend to be higher where low load factors on a sector, or adopt higher 5th freedom rights are not present since cost routings than would otherwise be possible. con.petition between individual airlines will be In short, it should be possible for airlines to less). Under the 5th freedom ervironment, the provide services in the region, and to and from country, and also airline operating services on a it, at minimum cost. In addition, it is important - 1S0- 5th freedom basis, will gain a greater share of inefficient.3' Governments can control the gains than otherwise. competitive behavior to an extent-for example, they can prohibit forms of anti- 7.14 The environment outlined above is one competitive behavior, such as pooling. which is capable of facilitating low cost air transport. Even where this is the primary (c) Imposed Obligations. Governments objective, a number of specific issues, concerned may impose specific obligations on with the ownership and control of airlines, airlines. For example, a Govermment regulation of market behavior, and structure of may decide that certain uneconomic aviation negotiation still need to be resolved. services should be provided, and it may These are discussed next, below. negotiate separate 'contracts' for subsidies with the airline to provide them. Alternatively, a Government may C. ISSUS Tw BE RESOLVED require the airline to provide them, but offer compensation. 7.15 A Framework for Facilitating Emdent Cooperation. In general it is desirable for (d) Constraints on Airline Operations. A countries to support a framework within which Government may, for political or other effickincy in the provision of airline services is reasons, decide that its airlines meet made possible, and encouraged. One way in certain minimum conditions: the airline which this can happen is through cooperation may be required to be fully based in Lhe between different airlines, and between different country, possess, or at least have countries. Each country can choose whether or prescribed access to, a minimum number not it wishes to take advantage of the gains of aircraft and use a minimum number of possible from cooperation or not. local cabin crew. Each of these requirements limits cooperation with 7.16 In pursuing aviation cooperation and other countries, and each has a cost. efficiency more generally, each country must Governments must resolve whether Lhe resolve a number of issues: gains from these restrictions outweigh their costs. (a) The Corporate Form and Objectives of its Airline. A private, or public but (e) Non Aviation Objectives.3" corporatised, airline will most effectively Governments may need to make choices seek cooperation. Airline corporate between higher airline profits and more objectives should be kept limited, simple, tourism. If the aviation environment is explicit, consistent with good economic to work to a country's advantage, it is performance, anu be driven by clear clearly important that Governments have managerial reward incentives. some assessment of how important other objectives are, and how different aviation (b) Competition and Regulatory Policy. policies will affect them. The nature of Airlines in the South Pacific possess linkages between aviation and other some market power; few routes are sectocs, such as tourism needs to be highly competitive. The problem of established to guide aviation policy. market power does not appear to be a severe one at present (air fares are not (f) Scope of Trade in Aviation abnormally high), however, it could Negotiations. Each country needs at become more of a concern if airlines some point, to determine how wide the become more profit oriented or highly set of trades it is prepared to make when - 151 - entering aviation negotiation. Indeed, some countries appar to go againsttheir Cooperation and efficiency are greatest own self-interest in not cooperating. By not when this set is widely drawn. cooperating, these countries may be imposing greater costs on other countries. (g) Response to Non-Cooperative Behavior. As the difficulties being Countries could attempt to sanction those who experienced in the current (1991/92) do not cooperate, by imposing a cost on GA1T Uruguay Round of trade themselves, for example, through air traffic negotiations are an eloquent reminder. rights. This option is only in a country's Cooperation between nations (or interest if the sanction has some chance of companies) is not always easy to achieve. altering the other country's policy in the future. - 152 - CHAPTER 8 CONCLUSIONS 8.1 Case for Cooperation. The PMCs face Pacific Rim countries. There do not seem to be a number of transport disadvantages which stem many major problems which could be eliminated from the small size and long distances aociated or substantially reduced by adopting cooPative with their transport markets. Anything which agreements. The agreements In place are can help reduce transport costs is of considerable largely on a bi-lateral basis, sometimes ad hoc, value. The PMCs operate their independent and differ from case to case. They may appear airlines and not surprisingly, the Issue of unsystematic, but they are fairly efctec. cooperation between them is a contentious one. Nevertheless, there remain some problems which Many consider that the problems faced could be better cooperation could help solve, an there alleviated, and perhaps eliminated by better are likely to be gains from more extensive cooperation on aviation by both the countries of cooperation. However, it is difficult to measure the region and their airlines. Cooperation is these, and specify their details, since the costs sometimes perceived from outside to be less and benefits from specific forms of cooperation extensive than it might be; when some solutions are only known with accuracy by the airlines are sugested, they often do not come to themselves. fruition. Each country seems to be pursuing its own agenda, and cooperation appears difficult to 8.4 The traditional approach to promoting achieve. cooperation has been to suggest rarticular forms. A "regional airline" has been one approach, and 8.2 Bilateral vs Multilateral Arrangements. an "aircraft leasing company" has been another. The individual PMC Governments, and their These apnroaches have potential advantages, but individual airlines, have their own distinct also severe disadvantages. Many of them objactives. Cooperation is possible, but it does require multi-lateral action, which is very require the resolution of a number of issues. difficult to achieve. Perfect coordination on a multi-lateral basis is unlikely to be desirable or to come about. But 8.5 The Enabling Environmental Approach bi-lateral arrangements, between countries and to Cooperation. Rather than specify what airlines, can and do, alleviate a number of specific forns of cooperation are appropriate, problems. the Governments of the PMCs should concentrate on sapporring an environment In 8.3 Existing Cooperation. In fact, there is which forms of cooperation that lead to extensive cooperation in place between the Improved perfornance are facilitated and countries of the region on aviation matters, and encouraged. In such an environment, airlines especially between airlines. Tbe PMC airlines should be able to choose w*at types of have addressed a number of problems they face, agreement they wish to make, and with whom,* it as a result of small scale or inexperience, is they who will have to make cooperation work. through a wide range of cooperative agreements between themselves, and with airlines of the 153 - 8.6 AirUne Objectives and Managenent low cost aviation, is in the interest of the PMCs Incentives. Much of the burden for pursuing generally, and overall, it will encourage tourism. cooperative arrangements falls on airlines and Nevertheless, for ineividual countries there are their managements. It is critical that airline trade-offs between tourism and aviation management be given a framework and objectives, and the problem is how to resolve the incentives which are consistent with the pursuit conflicts (see Annex 2). of both cooperative arrangements and efficiency. Cooperation should not be seen as an end in 8.9 A first step, whi :h is essential, is for Itsef but as an option to improve efficiency. Governments to determine the value of tourism Efficiency involves keeping costs as low as (compared to other objectives, such as profits feasible for a given standard of service, and in from airlines). One approach is to address ensuring that the servic&s that the users are tourism issues directly (for exarmple, through prepared to pay for are provided. Airlines tourism promotion and tax treatment of the which have profit as an objective, and which tourism industry and broad-based aviation face competition in their product markets will negotiations). An alternative is to require an operate efficiently in this sense. airline to be profit-oriented but entice it to provide services deemed by Government to be of 8.7 Like many public enterprises around the importance to tourism (for example, flights to world, the airlines of the South Pacific have particular destinations), by directly contracting' often been given vague and conflicting with the airline to provide them services. In this objectives. This is not conducive to efficiency, way, airlines do not become overloaded with and there is a danger that an airline will try to conflicting objectives; airlines can be left obliged satisfy too many objectives and in the process, to pursue efficiency, including the use of do nothing well. It is being recognized cooperative arrangements which advance elsewhere that clarity and simplicity of efficiency. At the same time, the Governments' objectives for public entermrises is needed to more general objectives can be addressed. encourage efficient performance. In addition, airline managers must be given a direct incentive 8.10 In sum, a sound approach to airline to pursue such efficiency, notably by linking cooperation problems should involve several rewards to performance. Privatization is one distinct elements as follows: option, another is to restructure public enterprises so that they pursue profit and act * Simple commercial objectives established commercially, (corporatization). Such reforms by Governments. would increase the vigor with which airlines pursue cooperative arrangements that increase * Airline management incentives and efficiency. rewards tied to performance. 8.8 Governments do have broad objectives * Capability to monitor price/cost ratios and which their airlines are often expected to serve. market behavior. In particular, Governments want to increase employment and encourage tourism. The former * Separation of non-aviation Objectives. objective is unlikely to be pursued effectively through imposing hiring requirements on * Explicitcompensation for non-commercial airlines. However, tourism and aviation go services. naurally together, and it is not surprising that Governments see airli. s as devices through * Broad-based aviation negotiations. which tourism can be encouraged. Efficient, - 154 - Endnotes 1. Additional details of the international air service routes for individual PMCs are shown in each Country Transport Sector Survey presented in Volume Two. 2. Higher route traffic volumes are shown to be associated with lower air fares for the Pacific region, in Part III of this report. The high costs of low density air service markets, have also been demonstrated in several econometric studies. See Caves, Christensen and Tretheway (1984). 3. Data on air service routes for the PMCs and comparator countries are presented in Part III, Annex 6. 4. It is important to distinguish between economies of density, which refer to how costs vary as the traffic volume on individual route markets change, and economies of scale, which refer to how costs vary as the overall size of the airline increases. Economies of scale are not substantial, but economies of density are (see Srinivasaw (1986). 5. This would be similar to the successful Scandinavian Airlines System (SAS), jointly owned by Sweden, Norway and Deinark. Other examples of multi-national airlines include Air Afrique, formed by former French colonies in Africa, which has operated since 1961 and East African Airways which collapsed in the 1970's due to differences between the shareholder States. 6. See Kaspar (1991) on Air Pacific and Qantas. 7. The basis for Air Pacific was Fiji Airways, which was formerly owned by several larger British Commonwealth airlines. The various countries of the South Pacific were to take a shareholding in Air Pacific; a few still retain small shareholdings to this day. Air Pacific is now effectively a Fijian carrier. For a time it had a ro!e as a regional airline, though in the 1980s it devel3ped more as an airline operating between the region and the Pacific Rim countries. In addition, most of the States wished to develop their own airlines; these individual States did not perceive it in their interests to grant monopoly rights to an airline based in Fiji. 8. For a view of the problems faced in airline co-operation see Eiliku (1991). 9. Externalities occur when actions of one party impose costs or benefits on others, and these are not compensated for or paid for. 10. See Dwyer and Forsyth (1991). 11. See Dwyer (1988). 12. In some cases, costs and benefits may be 'weighted' according to who g-ins or loses; in this discussion all gainers and iosers are treated equally, regardless of who they are. In the simplest case, this involves maximizing the difference between gross benefits to consumers and gross costs to the producers or equivalently, maximizing the sum of consumers' and producers' surpluses. For a recent example, see Sherman (1989). - 155 - 13. Collusion between sellers to raise prices is a form of cooperation, but it is usually inefficient because the losses to consumers outweigh the gain in profits. 14. Every country faces problems of deciding between competition and cooperation, both internally and externally, for example, in national policy towards mergers and takeovers. Mergers offer the benefits of cooperation, but they lessen the domestic competition between firms, and the benefits that may bring. 15. As an example, consider a route with only two flights per week, and the most popular day for travel is Saturday. If only two carriers serve the route, such as under 3rd/4th Freedom traffic rights, it will be in the interest of both to schedule their flight on Saturday. One Saturday flight and (say) a Wednesday flight would be more convenient to passengers and more efficient. The airlines would have a profit incentive to cooperate and uperate this schedule, if the airline which carried the fewer passengers could be compsad by the other (for example, through seat purchase arrangements on each flight). This result of competition is observed when there are only two or very few competitors. This happened extensively in the past on Australian domestic trunk routes. It is most likely to occur when there are only two operators on the route, they use similar type (capacity, speed) equipment, and substitute routings are more costly and/or inconvenient. The phenomenon was first analyzed by Hotelling (1929) and applied to the airline industry by Hocking (1972) and Gannon (1979). 16. Recent summaries of developments in the understanding of enterprise behavior are given in Holmstron and Tirole (1989). 17. The control issues here relate to a class of problems denoted the principal/agent problem. For a discussion of these in the context of public enterprises, see Vickers and Yarrow (1988). 18. One of the more important of these was the increased dependence on reservations systems, for which the scale economies are very substantial. These have been significant in providing a degree of market power to the (airline) owners which first developed them. 19. Code sharing, where separate airlines operate flights that connect, and give each flight a number for each airline, enables the airlines to offer flights to a wide range of destinations. For example, British Airways is able to offer flights to hundreds of destinations in the USA through its code- share agreement with United. 20. Cooperation between the major world airlines has tended to be at the marketing and financial levels. 21. These arrangements were identified, in part, from a field survey of PMC airlines undertaken in August 1991. 22. Some broad statistical indicators for the South Pacific airlines and selected Pacific Rim airlines are given in Annex. 1. The financial data for the South Pacific airlines is subject to limitations. 23. As noted earlier, airline management (and one with good internal accounting) is best placed to assess the transaction costs, risks,and synergies available. 24. See "Assessment of Air Services", Chapter 3 of Part Im. - 156 - 25. This is based on a statistical analysis, which relates air fares to sector distance and market density (proxied by seats available) and test for regional differences in the air fare relationship. 26. It is possible, of course, that other airlines in the region, producing air services against which the South Pacific services were compared, are themselves incurring systematic losses. While it is true that several airlines in Australia and New Zealand have recently reported losses, these losses have amounted to a few percent of total revenue, and there is no evidence of widespread subsidization. Thus, air fares in adjacent regions can be considered as unsubsidized, 'normal" fares. Among the South Pacific routes, it is likely that some cross subsidization takes place-however it is unlikely to be so extensive as to distort the overall picture. 27. Special circumstances may account for these, for example, Air Caledonie International uses two DHC 6 aircraft to operate 'essential services" for the Government between Wallis and Futuna. 28. Low utilization may also be due to poor investment decisions. An airline ma- over-invest in aircraft, or adopt poor scheduling, and have aircraft underutilized. At different stages in the past, some South Pacific airlines appear to have been in this position. More recently, they have cut back on fleet size. Indeed, most of the airlines in the region operate arrangements whereby they do not have access to large aircraft all of the time; they lease in or lease out on a part time basis. T-his form of cooperation enables higher utilization to be achieved. 29. Load factors might be low for several reasons-airlines may be trying to get good utilization from their aircraft, or they might have inappropriate fare structures (performance includes yield, as well as load factor), or they might be operating larger aircraft than needed for their routes, given the frequency. Low load factors may also result from a "need' to maintain acceptable frequencies with the aircraft available. 30. Even some parallel scheduling (where two flights to the same place leave at about the same day/time) can arise. Given the low overall frequencies in the region, this is an inefficient outcome. Cooperation in scheduling can facilitate both efficiency and commercial performance. 31. In some cases, countries are constrained by international treaty obligations-however these should not prove serious constraints to airline cooperation. 32. The problem is thus one of lack of instruments for trade, since the countries gaining could win the support of the country which loses with appropriate inducements. 33. Quite often, countries declare that they are only willing to trade aviation concessions for other aviation concessions. But it is not uncommon for aviation negotiation deadlocks to be broken at the same time as other trade problems are sorted out, perhaps at the highest political level. 34. Regulation of airline pricing is a difficult task-few countries have done it effectively for long periods. Nor is it sufficient to rely on potential competition to discipline price behavior. The largest air transport market, that with the US, has been revealed to be not very contestable [See Morrison and Winston (1987)1, and the markets in the South Pacific are likely to be far less contestable. 35. Further discussion on the relationship between broader (government) objectives an aviation (cooperation) is presented in Annex 2. - 157 - BIBLIOGRAPHY ABC. 1991. World Airways Guide. United Kingdom. Air Tariff. 1991. Worldwide Fares Book 1. United Kingdom. Air Tariff. 1991. Worldwide Rules, Add-Ons, Routings and Mileage Book 1. TUnited Kingdom. Asian Development Bank (1981), Solomon Islands Inter-Island Shipping Problems and Potential, Mission Report. Asian Development Bank (1987), Road Maintenance Report. Western Samoa, prepared by Beca Worley International, 1987. Asian Development Bank (1989), National Transport Development Plan, prepared for the Republic of Vanuatu by Wilbur Smith and Associates. Asian Development Bank (1989a), Road User Revenues and Road Expendtures in Fyf (Draf Final Report, November. Asian Development Bank (1989b), Road Traffic Safety In Fyi, November. Asian Development Bank (1989c), Commercialization of the Government Shipyard - Interim Report, November. Asian Development Bank (1989d), Divestiture of the Government Shipping Fleet and Rationalization of the Inter-island Shipping Services - Interim Report, November. Asian Development Bank (1990), Privatization and Commercialization Policies and Strategies Project in Solomon Islands, Ernst & Young. Australian International Assistance Bureau (1987), Fua'amotu Airport, Tonga - Airport Development Study, prepared by Airport Consulting and Construction Australia Pty. Ltd. Australian International Development Assistance Bureau (1988a), Outer Islands Airfields, Project Identtfication, Vanuatu, prepared by McIntyre & Associates. Australian International Development Assistance Bureau (1988b), Review of Public Works Departm, Vanuatu, prepared for the Department of Public Works by Cameron McNamara Pty. ltd. . 158 - Australian Intenational Development Asistance Bureau (1989), Reswlng of MainJ Rural HIghways, Western Samoa, prepred by Cardno & Davies Pty. Ltd. Australianlntentional AuistanceBureau (1991a),MInistryof Works, Tonga, lfrastrcturewDvelopment Project, Pacific Regional Team, Sydney, February. Australian Intenionda Dovdeopme Assistance Bureau (1991b), Reew of the Mrnsty of 7ranWorr Works and Utilities, Solomon Islands, SMEC and the Mayo Group. Autraliaon International Developmont Asstance Bureau (199lc), Feasibility Sne of ProvWnclAlpomr Dewlopment in the Solomon Islands. Airplan. Australin Intrtionl Assistanco Bureau and New Zealand Miisty of Extenal Relaions and Trade (1991), Joit Ausslafaw Zealand Study on Recurmnt Cost FSnance In Sowh Pacific Island Countries, Cabera/Wellington, October. Boca Worley Intewrnonal (1988), Road Evtuadton - Tonga, December. Bradfield (1986), Nadi hneaona Airp Master Plan, August. Boadway and Bruce (1984), We&re Economics. Oxford: Blackwel. Caves, Christensen, Tretheway (1984) 'Economies of Denity ad Ecoomies of Scale: Why Tnk and Local Service Airlines' Costs Differ', Rand Jounal of Economics, Vol. 15, Winter, pp. 471-89. CEMA (1988), A PrposalJfor Market Development in the Solomon Islands. Cole and Dorrance (ods) (1990) "Statistical Annex." Pacific Economic Bulletin, Vol. 5, No. 1, pp. 35- 56. Cole and Dorrance (eds) (1990) "Statistical Annex." Pacifc Economic Bulletin, Vol. 5, No. 2, pp. 45- 74. Cole and Parry (eds) (1986) Pacific Policy Papers No 2: Selected Issues In Pacific Islands Devlopment, National Centre for Development Studies (Mhe Australian National University), Canberra. Department of Transport and Communications (Australia) (1991), Air Transport Statistics: Monthly Provisional Statisdcs of Internaional Scheduled Air Transport, p. 16. April. Dwyer and Forsyth (1991), "Measuring the Benefits and Cosu of Forein Tourism", Center for Economic Policy Research, Australian National University Discussion Paper No 248 February. Dwyer (1986), "Tourism in The South Pacific" in Cole and Parry (Eds) (1986). Dwyer (1988), "Import Content of Hotel Food and Bverawo Purchases in The South Pacific", Pacific Economc Bulletin Vol 3 No 2, December. - 159 - Eiliku (1991) "NatIonal Needs and Economic Viability In Airline Operations : The Task for ASPA'. Address to the 9th AGM, Association of South Pacific Airlines, Tonga, June. Enari, K. L. The Need for Assistance in Legislative Drafting to Aviation in the South Pacific, Consulting Report, South Pacific Forum Secretariat, September, 1991. Europa World Year Book (1991), (31st Edition), Europa Publications Ltd. European Community (1988), 'Project Appraisal for Lupepau'u Airport Improvements, Vava'u Island'. Forsyth (1991), "The Economics of Thin Air Service Markets' Seminar Paper. Strategic Options Transport Development, Conference organized by The Department of Transport and Civil Aviation, Papua New Guinea, Port Moresby, September. Gannon (1978), "Paurallel Scheduling", in Domestic Air Transport Policy Review, Volume 2, Appendix A12-1, Canberra: Australian Government Publishing Service. GITEC Consult GMBH (1990), "Fourth Multiproject Appraisal and Implementation Study, Subproject B, Road Improvement Appraisal Report", prepared for the Asian Development Bank and the Kingdom of Tonga, September. Government of Solomon Islands (1984), "Economic and Engineering Studies for the Extension and Upgrading of the Lambi-Aola to Marau Sound Roads. Government of Fiji (1990), Socio-Economkc Development Strategies and External Assistance Priorities (Volume 1), Development Cooperaion and Aid Coordination (Volume 2) and Project Profles (Volume 3), Suva, May. Government of Kiribati (1987', Sixth National Development Plan 1987-1991. Government of Kiribati (1989), Second Kiribati Development Conference, Kiritimati 1989. Govermnent of Tonga (1986), Five Year Roads Programme, prepared by the Ministry of Works. Government of Western Samoa (1987), Western Samoa's Sixth Development Plan 1988-1990, Apia, Western Samoa. Government of Solomon Islands (1984). "Economic and Engineering Studies for the Extension and Upgrading of the Lambi-Aola to Marau Sound Roads. Gutteridge Haskins & Davey Pty. Ltd. (GHD) (1990), Survey of Domestic Aerodromes to Determine Sofety-Related Deflciencies, prepared for the Australian International Development Assistance Bureau, May. Hocking (1972), Some Economic Aspects of Australia.' 71o Airline Policy, Melbourne, Committee for the Economic Development of Australia. Hoteling (1929), "Stability in Competition" Economic Journal Vol 39 1929 pp 41-57. -160 - Holmstrom and Tirole (1989), 'The Theory of the Firm', D. Williamson "Transactional Cost Economies" in R. Schmalensee and R. Willig (eds) Handbook of Industrial Organization (Amsterdam: North Holland, 1989) International Civil Aviation Organization, 1990, Organization Study of the Civil Aviation Authority of FIji, Project TF/FU/89/201, March. International Maritime Organization (1988), Study of the Establishment of a Ports Authority in Western Samoa, Mission Report. Jansson and Schneerson (1987), Liner Shipping Economics, Chapman and Hall. Japanese International Cooperation Agency (1987), The Study on Development of the Ports in Western Samoa. Kasper (1991), "The Economies and Politiw of South Pacific Development: An Outsiders's View. Aid and Development In the South PacLflc. Pacific Papers 2, Center for Independent Studies, St. Leonards, NSW, 1991 p67. King (1984), The Air Traffic Market and Tourism: Some Thoughts on the South Pacific in C. Kissling, (ed). Transport and Communications for Pacific Microstates (Suva, University of the South Pacific, 1984). Morrison and Winston (1987), 'Empirical Implications and Tests of the Contestability Hypothesis Journal of Law and Economics, April, pp 53-66. New Zealand Ministry of Foreign Affairs (1988), Road Maintenance in Western Samoa, prepared by TGB Armitage. Pacific Area Travel Association (1990), Annual Statistical Report 1989, pp. 5-64. Perry (1989), 'Vertical Integration: Determinants and Effects", in R. Schmalensee and R. Willig (eds) Handbook of Industrial Orqanization (Amsterdam: North Holland). Port of Singapore Authority (1990), PSA Tarff 1990. Ports Authority of Fiji (1989), "Port Maintenance Policy', Report No. TR-64-16, Suva, Fiji. Ravenhorst (1991), Shipping Activity Between Rotterdam and Indian/Pacific Ocean Islands, Maritime Economic Research Centre, August. Redecon Australia (1987), Third Muliproject, Road Improvements Tongatapu, prepared for the Asian Development Bank, July. Roughton and Partners (1988), Guadalcanal Road Improvement Project, Maintenance Report. Ministry of Transport, Works and Utilities. - 161 - Roughton and Partners (1989), Road Upgrading Project - Final Report, prepared for the government of Fiji and the Asian Development Bank (TA No. 1046-FU), November. Sherman (1989), Fhe Regulation of Monopoly, Cambridge University Press, Cambridge. South Pacific Bureau for Economic Cooperation (1990), Review of Non-Directional Beacon Requirements, Vanuatu, prepared by the Civil Aviation Authority of Australia. Srinivasaw (1986), "The Costs and Benefits of Being a Small, Remote Island, Landlocked or Ministate Economy', World Bank Research Observer, Vol. No 2 July, pp 205-218. TecnEcon (1988), Solomon Islands Rural TransportProject. Commission of the European Communities. Touche Ross and Co. (1984a), Regional Transport Survey of the South Pactfic - Snping Study Anlsis Report - Ports, London. Touche Ross and Co. (1984b), Regional Transport Survey of the South Pacifc - Shipping Database Reports, London. Tourism Council of the South Pacific (1989), Air Transport - Tourism Linkage Study Final Report, April. Tourism Council of the South Pacific (1990), Solomon Islands Tourism Development Plan 1991-2000, Ministry of Tourism and Aviation. Tourism Council of the South Pacific (1990), Tourism Sector Report, Western Samoa. United Nations Development Program (1985), "Port Development - A Handbook for Planners in Developing Countries", TD/B/C.4/175.01. United Nations, (1990), Economic and Social Survey of Asia and the Pacific, New York. United Nations (1990), National Accounts Statistics Aggregates and Detailed Tables 1988. New York. United Nations (1990), Statistics Year Book for Asia and the Pacific 1989. New York. United Nations (1988), Survey of Economic and Social Conditions in Africa 1985-86, New York. Vickers and Yarrow (1988), Privatization: An Economy Analysis, M.I.T. Press, Canibridge, Mass. Westem Samoa Airport Authority (1990), Master Plan Update for Faleolo International Airport, prepared by Henry Fan (Singapore Airport Authority). Western Samoa Accident Compensation Board (1990), A Study of Accidents. World Bank (1987), The Highway Design and Maintenance Standards Model. World Bank (1988), "Road Deterioration in Developing Countries - Causes and Remedies", World Bank Policy Study, Washington (June). - 162 - World Bank (1989), South Pac(fic Islands Transport Review. Country Reports, prepared by Pak-Pok and Kneebone Pty Ltd., December. World Bank and Economic Commission for Africa (1990), 'The Road Maintenance Initiative - Building Capacity for Policy Reform", prepared by the Economic Development Institute and the World Bank (Africa Technical Department, Infrastructure Division), Washington, June. World Bank (1990), 'Information Systems for Road Management: Draft Guidelines on Systems Design and Data Issues", World Bank Technical Paper, Washington (September). World Bank (199 la), Toward Higher Growth in PacUic Island Economics : Lessons from the 1980s. Report No. 9059-ASIA, Wahington, September. World Bank, (1990), Western Swnoa Emergency Road RehabUitation Project. World Bank (1992), PacIfic Regional Post Secondary Educaton Study, Report No. 10522-EAP, Washington, May. World Tourism Organisation (1989), Compendium of Tourism Statistics, United Nations, New York. World Tourism Organisation (1991), International Tourism In East Asia and the Pac(fic 1970-1990, United Nations, New York. W.S. Atkins International, circa (1989), Vava u Regional Development Program prepared for the Ministry of Foreign Affairs, Tonga. i I I -165 - ANNEX Il. PACIFIC ISLANDS TRANSPORT SECTOR STUDY MANAGEMEN IN OCtMATON FOR ImASuTrm 1 Five primary functional levels or Ebment Aspects classifications, related to the level at which the system is viewed, may be defined. These Road Inventory Network/location/geometry, fucional levis are: oanvironmet * Budgeting: The sectoral level, viewed Pavement Pavement structure by government and the public, in terms of Pavement condition budgets and statistics of the extent and level of service provided. Structures Structues inventory Bridge condition i Network Planning: The planning, programming and budgeting of public Traffic Volume, loadings, accidents works and expenditures for road development and maintenance, Finance Unit costs, budget revenue encompasses strategic planning and work programming. Activity Projects, interventions conmmitments * Design Standards: Used in the functional and technical design of road Resources Institutional works, embraces the design of new Materials construction, betterment, rehabilitation Equipment and periodic maintenance works. 3. A further classification-that of * Maintenance: Information related to Information Quality and Detail-is us4 to facilities and operational management. indicate the level of detail for application to the Examples include routine maintenance various functional levels. Four levels of maiagement, traffic operations differentiation are used. These are set out management, equipment management and below. construction management. * Research and Development: Data are usually study specific and usually more detailed and precise than for project-level or operations. (Very limited in PMCs.) 2. These functional classifications are combined with technical groupings of information to form a systematic basis for classifying information needs. The technical groupings for road management comprise: -166- ANNEX 1.1 Infrastructure Information-Quality and Detail Classification Level I Comprehensive; In depth field investigation; Special project design; High skills and resources needed. Level II Standard design needs; Sample coverage for planning; Sufficient for evaluation; Reliable support and resources needed. Level m: Sufficient for network planning; Suitable for elementary project design; Combined manual and semiautomatic data collection. Level IV Basic only; Summary statistics of inventory, performance and utilization; Suitable only for standardized design of very low volume roads, simplest basic collection methods. Level IV information is most suited to the small agencies, with very limited resources, in the PMCs. 4. In the detailed e. 2mination of classifications are used in the marine and maintenance in the PMCs prsented in Part II of aviation subsectors to provide similar the text, the extent of existing management information however, no comparison standards information has been developed for each are available for these modes and subjective transport subsector in each PMC, based on this assessments have to be made. These should be classification system. The Management subject to further consideration and refinement, Information Inventory Table for roads uses the if development of appropriate data bases is to be classifications outlined above. Notional undertaken.' Endnotes I. For detailed discussion of the transport infrastructure maintenance situation in each PMC, see the maintenance annexes to each country sector survey, in Volume Two of this report. -167- ANNEX A .2 PACIC SLANDS TRANSPORT SECTOR STUDY DATA AND MODOLOGICAL ISSUES IN TE ANALYSIS OF INFMASMUCTRE MAIrANCE DATA ISSuis Exsting Asset Inventories. There are very limited data on transport inrastructure in Exbsng Data IUmitnado. The each of the five countries examined. Whi the estimated replacement cost of transport exception of the Solomon Islands Port Authority infrastructure, rehabilitation requirements, which maintains full commercial accounts, no assssed maintenace needs, and actual government miniistry or department in any of the manteace expenditure, are tablished in countrias holds a formal inventory of asts separate country surveys.' The etmate are which vwould serve management needs at the derived from existing databses that are severdy network:, project or operational levels. Pret limited in quality and coverage. Whil analyses inventory information, such as it does exist, is prepared separately for each PMC indicate generally limited to basic physical data, for moderate consistency, the results should be example, road length, the number (and treated with caution, cognizant of the quality of occasionally the length) of wharves, and the the data. The objective of the estimates is to number and lengdi of runways and surface provide an indicoae quar.ladwperatcdw of material.' No data are available on the present the present situation and to illustrate the nature condition of transport infrastructure, except as and scale of the transport infrastructure reflected in the accounting data of the Solomon maintenance issue. Islands Port Authority and a broad assessment of road quality prepared in a recent transport sector Lack of Data on Use of Infrsstnucture. development study for Vanuatu. Data on traffic throughput volumes and coirposition for transport facDfties are key METHODOLOGICAL ISSUES indicators of the value of an existing infrastructure; unfortunately in the PMCs these BasLi; for Estimating Replacement Cost of data are very limited. Available data and Transport Infrastructure observation for the road subsector indicate that traffic volumes are below 100 vehicles per day The replacement costs of transport assets outside the immediate vicinity of major towns, have been estimated using the prepared and are sometimes less than ten vehicles per day infrastructure inventory and estimates of current on much of the road network. Such low traffic unit construction costs. Transport infrastructure volumes, with apparent relatively low growth, assets exclude equipment used for maintenance, indicates a need for traffic data over the network but include some equipmen: and facilities wh.ch and for careful review of the incremental are an integral part of infrastructure serviaes. economic benefits gained from road expenditure For example, marine infrastructure includes and the appropriate standard at which roads buildings, hardstand areas, shipworking should be constructed and maintained.2 equipment, navigation aids and tugs and workboats in addition to wharves and jetties. Aviation facilities include runways, terminals, - 168 - ANNEX HI control towers, navigation equipment, and beacons and firefighting equipment. Replacement cost estimates exclude the opportunity cost of land used for transport infrastructure. Unit construction costs have been derived from recent construction costs and estimates of utypicalw costs where recent specific cost estimates are not available. Endnotu 1. See the Malntiace Annex to each individual country transport sector survey, prented in Volume Two of this report. 2. This is reinforced by the scarcity of maintenance supervision resources and the distance of much of the road network from primary centers. With such low traffic volumes, environmental factors are likely to exert more influence on road deterioration than road use. Technical knowledge of road maintenance and deterioration, for example, through the Highway Design Model (Wo,ld Bank, 1988), is of limited use for analysis in the PMCs because road conditions lie otitside the rang of detaild investigation and model calibration. 3. This information is not always readily available, but as complete an inventory as possible has been assembled for the present study. - 169 - ANNEX m.1 PACIFIC ISLANDS TRANSPORT SECTOR STUDY PART m: COMPARISON OF REGIONAL TRANSPORT COSTS STATCA DATA AND SOURCIS Socio-economic data and trade flows for Freight rates, cargo movements and the PMCs, ad the comparator countrios, are additional shipping charges were collected from available from a number of publicadons. shipping lines and ages in Australia, Now Zealand, Hong Kong, Singapore, Ronerdam and Mm most recent socio-economic data the United Kingdom. Although rebat are avaLble on all the islands is largely for 1988.1 common in the shipping industry and are Consequenty, the most recent year for any data generally regarded as commercially sensive, series varies from country to country. All costs their impact is relatively minor on tbo routes are convertd to US dollars (1988) for the included in this study and the data collected is a purpose of analysis. reliable indicator of current market rates. Additional data on Christmas hland and Air fares and hotel costs were obtained the Cocos Islands, and data on tourist from travel agents. Additional aviation data mov"ments in 1989 from Pacific Area Travel such as route structure and aircraft types were Ausocition and World Tourism Organization obtained from World Airways Guide ad Air were provided by the Federal Department of the Tariff Guide. Arts, Sport. the Environment, Tourism, and Territories (DASETI) in Australia. Endnotes 1. A more detailed assessment of the PMC economies is contained in World Jank (1991a). - 170 - ANNEX MA. Table 1: PACIFiC ISLANDS-Soclo-ECONOMIC CHAUACTERMSCS, 1984/1987 Solomon Wegem Year Flji Kiribati PNG kI-nds Tong, Vanuatu Sanoa Land Aes (Sq Km) 19U 18,272 710 452,u4 28,900 699 12,190 2,935 Population (000) 1988 719 67 3,560 304 99 150 162 Populationgr*owh (P.O.) 1980-88 1.6% 1.8% 2.2% 3.8% 0.8% 3.4% 0.6% ONP (USSm) 1983 1,130 40 2,920 130 80 120 100 GNP/Capita (USS) 1988 1,540 650 820 430 800 320 5U Av.realoNPgtowth(p.a.)t t 1980-88 -0.4% -5.9%Ik 2.6% 6.1% 3.0% 1.2%Lk -0.6% To lAd Flows (USSm) 1988 43 12 300 35 13 29 19 Ad as Prcent of GDP& 19805 3.0%& 740% *- 20.0f 28.0% 26.0% 25.0f Mardbadis Exels (USSni) 1988 304.2 4.6 1,4096 77.2 6.2[J 15.4 14.2 Maaehandfilabn Ivds(USSm) 198 460.7 22.1 1,170.0L1 74.8 90.0 67.7 68.5 Numhbwrof Viuorsh& 1989 250,565 3,000 48,918 9.380 20,348 23,86S Sl,740 Sea fSiht kladed Oonss)Md 1987 661,O0L0g 10,000& 2,623,000 39,795 64,000& S9,00(0h 43,00O& Sau eht oed( i 1987 807,000 2S,000 1,860,000 140,842 20,000 61,000 87,000 banoem Activty By So" (f of GDP)l! Agrlmr 1987 23.7 20.3 29.0 44.3 41.8 28.0 41.2 hthtiry 1987 20.9 10.8 35.1 8.0 17.3 12.9 15.4 laszyic t1987 55.4 68.9 35.9 47.7 40.9 59.1 43.4 Total: 1987 100.0 100.0 100.0 100.0 100.0 100.0 100.0 ajor Domesic Expos (% of Total E-x--.) B&MM19S8 - .X5 9a Bef 1988 14.9 - Cemen 1988 0.1 CoOa 19U - 3.2 4.6 7.3 4.2 Cocomat Oil 1988 0.8 - 18.3L- 39.3 Coffe 1988 - 9.3 - Copper 1988 36.6 - C00111 19U - 71.3 9.7 * 59.5 6.6 Desccated Coconut 1938 - - - - 6.35 - - Fio 19S8 * 27.5 - 48.8 ForeAt Prducts 1983 - - S.0 Gaold 1983 18.7 33.2 Molase 1988 2.6 * - - - - PalmOil t988 - - - 7.4 Sunr 19W 45.6 - - - Taro& Tamu 1988 - - - - 17.5 Timber 1983 - - 22.8 * 6.6 3.7 Vanill 1983 - - - - 8.1 Major Exports as % of Total Expons: 1938 67.9 99.3 90.8 93.4 48.611 88.2 71.3 Major Imporu (% of Total Impotos) Food & live animal 1937 17.3 31.0 16.9 15.0 21.6 13.7 23.0 Minals/Fualetc 1987 16.2 10.6 11.0 14.7 10.2 8.5 17.0 Cbemical 1987 3.6 5.1 8.4 6.7 6.5 6.0 6.0 Mschintry & Tramport quip 1987 19.4 14.5 33.5 29.1 18.3 25.7 20.0 Oher manufctuwnn 1937 31.0 24.2 26.5 29.8 27.6 35.0 27.0 bnports asa% ofTotal Imports: 1987 93.1 15.4 96.3 95.3 84.1 89.0 93.0 Euoa W.iid Year look. 1990. La Pak-Poy & Kneboee Pty Lad 9nd thd World Dank, South PadfJc IJ.Id 7hwupoe Sraor RnW: CoXuy Repoi December 1989. World Touuriam OrgnizAtion, 1991; Pcific Am Travel Ammocia ti,A1ua SLdscal Repe, 1989. Europa Woed Yea lpa, 1990; Pak Poy and Kneeb_oe and World Bank, 1989. ECroe World Year Book, 1990; Pik Poy and Kneebom nd Word Blank, 1989. For To*&: VUited Nation, UN&Eco.emk -d Soca5 Swy qfJAsir a,de Pac(R, 1990. KIq2 Lg - 1984 ds; & - 1985 daa; 5I - 197 dab; Li - 198045 dot; & - 193"7 dam. hWibsen: .. indicate data no available; - indiaes the valu of zero; o/a - s applicabb. Seuwc: PadRe Ecbomic AdL &anka Amux , June 1990 and Pmtk Eckmlc lMdw S&Oaredcal Annex, DIember 1990. New (1w * imqa Cll Yea;r (Calclui him& Mau" is 314... Se-iteirs =ITm' = ('umolam ma&3s I aod Area (S.I Ka) 399 16,750 4J211D 2.U 2512 454 ISS IJ 13.062 Ilm4n (IN 11/9 364 139 Ie.% 59 2' 3.35 33.5 3M hIqmh rgrnwsh(pF.) 111.U 1% 1.9% l4% I.% 0.9% . . 35 34% (i1w (Su) fin IU 3.D 133 3, t II U * 0 a (WPIC as 19 5.lqdIr 7.3' 3.3l 3.m1, / a 440 410 Au. real (INP *rrSh (p.a.) 3-MO-81 6. .1% 141%* 35% rnar/a 3(l1%6 3 3%" M1eefhamise P!xrwes(IISSM) 1967/1 2121 76 0l101 1.5562 169 9a/ a/a 116 446 1iereb_aad lrnqms f IKSI) 1967/1 5938 784.9 11570 1,115 339S a/a a/4 462 94S Numb_ I d V-W f W 112,r ll0 12S, 135.37 40.0 141,31 773nl W 5O 2.000 15S.15t Sbpin (seiB bud_ (Wl) O3m 1.13J1.111 .. 937 MAN4117 . .. . i sm 1.15J .%&kVk* ipr mbua (MI) 13 8311.10. 13.41M.m I56.3111 2 235S. 95.10 96,037 Pavvioneek Aeii3 fly SeC., (% o4 (CilW3 Agriea.e 317/U I 40 113 70 *3 a/ *f 419 I3I5 Mklury 337s/ 240 390 325 163 I. r/ */a 13.9 143 Sesrvc 117/US 742 770 5J2 76.3 73.6 */ / 44. 5J.7 lidat: s167/3311 Il100 3f(D0 1m00 1114)p 3og6 loop */ moo M'Ilapob" s (*% ad jmrI - -.-11 (bawes 331/U - 7i 345 (urea 33 U . . . 39 ..3 . 3I7/U - 36 - . . .9 * ol In Jt .u, lo1c1 - !14% NkleID mu'0|s. 137/U *1 - - - .4 N&Ld mask 9 WuLd.. 1got li 33 - - - - - - - 1"7^~~~~~~~~~~~ 3bah'dd.mseb 337/U - - 11m 337e/ - - a - - - - - 1 rn%de 31"7/ . 3 . 07 1_a 13/7U - - 7- - no V _riS I - J . . . ee7. - . 77 - Yhp-im 331/ - - - - . . . us 161arIhimek!kpadsa % dTlI Pdiqaa 1 376 732 34 7J14 119 7N 9. . 6 49J lilaborn I_d of Teld IMM)" Il r liveae 1"7/U 2.93 5 5S 45 171 .. . 337 111mrmfleht 1"7^el/ 9 2 a S0 6 2L2 ....* 1 13 (lI 9 1"7/P .. .. I? . . . M hireryA aO I S pme 331/ .. .. 102 3311 is .. 79 322 _ I Sher m_n _Nes 337/U 03 7.4 lie 417 212 .. .94 365 L M..wplmipus* sb a.a%. r.ual qr s.: 137/U IIS 13.17 15 Riol15 9. .. - 397 18" Sewe IJN, A__ It^,m a" IMF 1>ubu KS Ilk ,tAI. lmJka4.a.rv dies. ma 11. ;C a * t I7U dai;e I3e da. Ia - -991 ddo; g - 1971lI(, 1a - I!11SlI7.Ia l3ialmm . sim fad-esr, - indkes Khe v*ic 4 s.c rn/a - al i rgiciaWe - 172 - ANNEX M. Table 3s PAciric IZLms AnD MAot TRADING PART , 1983/1988 NOMINATED MAJOR TRADING PARTNERS COUNTRY Year Destination of Exports X Year Source of Imports % Fiji 1988 United Kingdom 29 1988 Australia 29 Now Zealand 24 New Zealand 19 Australia 22 Japan 10 Othdr 2S othen 42 lQO 1QQ Khibti 1988 Nethlands 69 1988 Ausrlia 44 F4i 23 Fiji 1S USA 4 Japan 11 Othen 4 Otes 30 Papua New Guinea 1988 Jn 41 1987 Austalia 43 Fedoral Republic of Gernmny 39 Jaan 18 Autlia 12 USA 8 Othes 98 Others 31 Solomon Islnds 1987 Japa 36 1987 Australia 41 Unitad Kingdom 12 Japan 19 Thailand 14 Singwore 9 Ohers 38 Others 31 Tonp 1986 New Zealand 39 1986 New Zoaland 39 Australia 30 Australia 30 Japan 9 Japan 9 Others 22 Others 22 12 122 Vanuatu 1986 Netherlands 33 1986 Australia 36 France 26 Japan 13 Japan 19 New Zoealnd 10 Others 22 Others 41 1lQ 2 Western Samoa 1983 USA 31 1983 New Zealnd 38 New Zealand 2S Australia 1S Austrlia 13 Japan 1S Othors 31 Others 32 Sowrn: Europa World Year Book, 1990. - 173 - ANNEX M. Tablo 4: CoeARAToa IsLAmD Couriz-MAjoR TEADINO PARmS Major Trading Partners Comparator Islands Year Destination of Exports % Year Source of Imports % The Society Island 1987 France 54 1987 France 52 USA 19 USA 10 Others 27 Australia S Others 33 loo 1Q2 Now Caledonia 1987 Franc 44 1987 France 53 Japan 19 Austrlia 8 Federal Republic of Germany 9 USA < Others 28 Others 3k 100 122 Mauitius 1988 United Kingdom 36 1988 Franc 12 France 23 USA 12 USA 13 South Africa 90 Others 28 Japan 7 United Kingdom 7 Others 53 Rounion 1988 France 70 1988 France 67 Italy 4 Others 32 Germany 3 Others 23 100 100 Seychelles 1985 Pakistan 35 1985 United Kingdom 15 Reunion 12 France 9 France 9 South Africa 9 Others 44 Others 67 100 122 The Comoros 1977 France 65 1977 France 41 USA 21 Madagascar 20 Madagascar 5 Pakistan 8 Others 9 Others 31 Source: Europa World Year Pook, 1990. Note: Exports and imports by majcr markets were not available for the Maldives, the Cocos Islands and Christmas IsLdd. - 174 - ANNEX m.A Table 5: PACIIC ISLANDS-VISITOR ARRIVALS BY ORIGIN, 1989 VISITOR ARRIVALS NOMINATED COUNTRIES BY ORIGIN (%) Solomon Wotemn Fiji Kiribati L_ PNG Islands Tonga Vanuatu Samoa The Americas 20.3 33.3 14.7 10.1 20.7 2.9 10.3 Europe 14.1 - 13.3 6.7 19.2 3.0 6.8 SouthAuia 0.0 0.6 - 0.4 - North Eat Asia '.5 S.4 S.S 11.9 2.9 South-East Asia 0.0 7.0 - 0.7 - Aunala/New Zealnd 50.0 66.7 52.8 48.3 32.8 66.8 39.7 South & Cual Pacific 7.2 - 4.0 20.4 14.2 17.9 40.7 Othr areas 2.9 - 2.2 9.0 0.1 6.5 2.5 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1* 1988 data. & Othr Areas include Africa, Middle East, and the USSR. Source: Pacific Area Travel Association, Annual Statistical Report, 1989. Table 6: COMPARATOR ISLAND COUNTRIES-VlSIrOR ARRIVALS BY ORIGIN, 1988 Vs rrivols Ivalw /a Society La Chrins Cocoe The & The by orin () Caledonis Islands Mauritius Rdunion Seychelles Wand Islands Comaro. Maldives The Aericas 2.4 44.4 0.8 2.6 - - 1.3 Europe 20.9 34.1 43.3 45.0 83.1 - - 77.8 SouthAsia - - 1.7 - - - - 11.1 Noth EastAsi 33.1 6.9 - South Est Asia 0.4 - - - -- - AugalialNew Zeeand 31.1 11.4 1.7 - 1.3 100.0 100.0 - 9.8 South & Cental Pacific 3.9 2.3 Other raa /c 8.6 0.5 52.5 550/d 13.0 - - - _ 100.0 aL100. 100.0 J0 J20.0 , 0 10. fA 1989 data. Them firquent visitors to the Comoro are the French snd the South Africans. & Other auea include Africa, Middle East, and the USSR. d Include viitors from the other regions listed above. Soures: Pacific Area Travel Asscistion, Annl StagiSoal Repon. 1989. Esimats for Chr_eis leand and Cocoa Ilands provided by Islands Liaison Office, Austrlia, 1991. - 175 - ANNEX M.2 PACIFIC ISLANDS TRANSPORT SECTOR STUDY MAS OF SK:NG SERVICES FOR PACIC IMANDS, COMPARTOR CouNTRES AND PACIFIC RIM nnU Is PACIFC ISV *ST mw MUO .__ ., ~~~~~_______ __ __ FEDERATED S-ATES , . . ************* * *~ \MICAOESIA e * * ..... ... _._. ..~~~~~~~~~~~~~~~~~~~~~~m mn .Uf TA #- . uuu '' B____ "_ .2 .- '.2 ..* t0 .wsB.S = ** * *t.wh _____________ i -mo t P _ _ * .~~~~pG * *~~~~~~Kowa* _ '0 .' */ *------eoo- tAMAWA-.i * 0__d_ _ _ __A_ _ _ _ _ IRl<;::~~~~~~~~~~~~~~~~~~~~OI 'T Quo* ~ ~ ~ ~~~~~ BI DS TUA O- TI d I I I[RALIA W .n ,.,.f ___ . 11 1fti POLNESI r... --**~ lOe*g* 2 .1 * * . . __ ls~~ * ! w :. *. * CALEDOf Ib A~" r AL1Je / - 0~~~~~~~~~~~~~~~~0 * ~~~~~ /' ~~~~~~~~ ~SOUTHf PACIFIC OCEAN SGidC a). AOOLO rUfisY AucMaid | |FlNomnInatod Pacific Boland Countries (3) Commaratcw Counirle In Pecill F~on [jg : PACIFIC ISA-*DICAflD" ZXTIA-3ZGIOUJM. I=SaVC .'cDs^ ~ ~ ~ - -- -- - FEDERATED STATES . . OF MICRoNESIA .0 = WEU -ew _ ' ' '*- *. .eW O* '-' 6SaAt.R-mu*p* Co - W Cd- 10 0 WAIn.Pac ' t - * ., *0 ** *q-- ,D * 0-0-- a in r0 "000 0 K. __,_________,. 112/~~~~~~~~~~a 9. . *: ******UAIGI-.nn'dv { a U S T R LI / ' S. - I ,-A ot., ISLAND ."J * Tb mbii UV_U I TCW!LAU * - ~ ~ I L $^LjtLj -.' FREWACI |-. *om'n-L POLY ESIA -FI Cn rC roAMERSCAN R glon le. ~~~~~~~~~~SAMoAk coK~fJ A STRALIA ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~uA**mwJ~, . : ~SoOLJm PACIFIC OCEAN OColmpaualoi Countries .n PacItic Region rLU3 PACIFIC ISUDW -NIIDICAO TDMMR-GCIA UP C l .FWArE F E he TEO E S r tt^A ;-. t "- d .AM p~~~~~~~~D co \ C^tEOOvg - z * * * -A tEVE"sb~~~~~~~~~~~~~~~~~~~~~~~~~~~~~% r6. 0-X | | F40n1nctcd Poclfic lel-nd CountR-e ~ ~ ~ ~ ~ ~ f, FEWC& 4 o"co onf nPdi o .~~~~A . .- F_zure 4: PACIFIC IStANDS-UCIOKAL SCMII UlVCIS FEDERATED STATES . .IES OF MICRONESIA .---- 10 12 dve .5. .. TA- o_ f z - | ~~~~~~~~~~~~~~~~KIRIBATI| \ l | /CA ANDS TUVAU - . -OEA .' 'U.Pap ^ **.. POLYNESIA r~ ~ ~ ~ ~ ~~~~~~FJ . . SAMfl .~O .:Mm A i A LLf~~~~i; IION93D ISLANOS~l~ o 'CALEDONI ,WLF SOUTH PACIFIC OCEAN LEGEND__ _ _I | ] |Nominated Pacitlc Island Countries O Comp-esto, Countiles In Pacilic Region Flhur Ss PACIFIC IS(MLDS-IICMLS ES S CES "SI FEDERATED STATES OPWU * * - OF MICRONESIA I a, .............. .. 111 vap ~ ~~~~~~~~~~~~~~- - -. - -*- - 3I v NAURU. dIAi SLAN TUVALUL TOKELAU A * x CX X \*K % * * IS. Z^N i FRENCR ______ COOK 7N. ~~~POLYNESIA I~~~~~E l, VmaUA ..* FIJIl. AUKAG1 USAQLt.|O 1': X * ~ B ......................AL/ COOK - r--- lJ~~ Nominated Pacific Isld iounts 8SOUTH PACIFIC OCEAN LEOEND s E:= Nominated P-cHlc Island CountNlos 0 ConWrnlor Countries in Poclelc nngton Fiture 6: PACIFIC ISLANDS-CIOL SNIPPISN ICMC3S __ ~ ~ ~ ~ ~ __ . _- _ __ _- , ,, .. . , -_ lUSl~ ~ US * - FEOERATEO STATES r A I* C FOM LE Of MWCRONESIA dove* _ _ _ _ - S dp laP .~ SX e* . ._, Seia a *r3 nS * *.~~~~~~~~~~~~~~~~~~~~~~~~~...... ..*.. Swuyp3-4vaeks 9._- ||Na_U' .1sTVALU TOKELAU KIRIBATI ~ ~ ~ ~ ~ ~ ~ FRfII l | i 8 . . @,OR POLYtIESIA j EA . rwwlI CALEDON TOKELAU | |Nomlnalod Po<:l#Ic Island Countrlo- 0 ComP--elor Coultrels In Peclilc Rolotwo -182 -i.Z. I~~~ Q W. z - zz J~~~~~~~~i W 0~~~~~~~~~~~~~~~~~~0 iii~~~~i 11imr. 8: INDIA OCEAN ISBalmSI=1 sUme 2SF 45*E 601 ~~~~~~~~~~~s-FS S.F 0* Z~~~~~~~~~~ _ MAI .-_i0 0 AAAV 0M000 /~~~~~~~~~~CA AUSTRALIA/ DAS-I-8DAA*M * S% *e IS .SI .-p I T ,--1 \ ~. * ...... SOUTH D. . ____________ __________\ ar~~~~~~~~~~~~~~~~~~~~~ -r - - - -@ - HstOLWVIAC014 3s p... ;;. SUKU 5N MI l Wl i NOPPQW VWCN KAMSHA LEGEND: ; ; fEE 151 3 . I0E [Z2J IndMan Ocean Comparator eslrnds. NOTE: EXACT ROUTE STRUCTURE NOT AVAILABLE YilUr. 9 INDIM OCIU ISLANDS-IPSSING uIus uavCs 31. I AALDr . ' A . o .*- | C,HELLES I OAR-66-4LAAA .Ud.......... ioe Dogroch;. A INDMN o(KeeCingIsland lnd| *CMOO * E N U t, W'sI' TAMATV 1maeIndL AUST AI OCEAN ~V S WOUTHST; lWZ es L:::::A ICAM tao"D h 1tE~~~~~~~~~~~~~~~~~~~10111111 SOEFNCLO ATSUM Pe kidian Oen Conwe_ror Ilaands. NOTE: EXACT ROUTE STRUCTURE NOT AVALASLE FiAure 10: INDIAN OCEN ISLANDS-"DEDIC&TED - CTRA-REIOSISL SEIMINl SERVICES l 0 t~~~E 4SE * E u iE WE ) lOr to 4 IS. (3~~~~~~~~~~~~~~~~~~~~~~~~~~~~I' :," 1l- ,''_*_ 1t~~~~~~~~~~~ 11 4--'1 J+nnt bo"J t>6)p"lo U OCEAN w AFRICA Durban _,T L ._b WUI_Pnu, \D xXxxxxxxxxxxmx xt1 ss a"t,~3V$ 45 1E OEI.E. -ISE LE.END . IIIndian Oc--v Comparator lelarf NJOTE: EXACT> ROUtE STRUCTDURE NOT AVAILAEILE ,ure II: INDIAX OCEUN ISLAUDS-ZGIOL ShIfMING SM CI W1 4S- -E 1t -F mi 4F -E / _ _ l / \flR(ONIOMR&IOASgINT IHE AN OCEAN ---= ioA / 5. v ___--.UNK / | U.DVE' _ CO,_,GNE WALGACHE ME NAVIGAI ION g EC1ELLES twtC7>i>sl ~~~~~~~INDIAN riC,.. KoItnQ) island-. /7 . * or ol I,., 5, *.: \tvA * RtsAL .l , , J 4OCEANl SCWT&t1/ 0*@1 oi 1_ ~~~C LEOZ^So3 n,4 -tE WEE I IIndian Ocean CowrelorI slands MM: EXACT ROUTE STRUCTURE NOT AVALA13LE _-~~~~~~p*OEN U AI -187 - ANNEX m PACIFIC ISLANDS TRANSPORT SECTOR STUDY SCHEDULE OF SHIPPING ROUTE TARIFFS AND VOLUmES -' 188 - inUZZL.Z TUANIpoI URCTOR ITDYT Schedule of UhutnEnS. Yawdif aad otie ACrUAL FRPHT VOLUWM RATE DISTANCE CT pe ROUTE amoup ROUTES (ASTU) (im) ) C euw SispeCos sands 4.W0 7 1 Id_ _ C 2.727 2.A 242 (ad. kui. SIA hpuie-Koms 3.117 3.43 24 ad baii HRe K _Ms+ . 2.727 3.000 O_M 1 ) M _ 6.371 4.201 Slap weMh 5.519 3,00, Hasa Kong-Mae 3.377 3.74. S!APPSMs-ujo 3.117 4,1# 24 S_Wau-MhI 2.247 I.U3. R_eusa-Mal. 1.313 6.491 Siwppare-mom 5.519 3.370 Readul-Memoi 64.925 6.543 Siappav.-Mo.e 4.286 3.370 Slapor.-Mutaudu 4.236 3.370 Sinppf-Nou.a 3.247 4. 17 1.209_C Re aa-Ne.r-N 4,191 11.376 M _ bo.m-Nmas S2.N10 1.597 1.09Q5 Roue,dum-Noum 3.633 I.S76 Siappo,-Nomwea 2.857 4.400 Roaerda-Nouam 4,114 I,376 Romrdam-Nouses 3.638 11.376 Romerdam-Ncm.a 4,605 11.376 Hoag KoarNoumes 2.7.7 4,103 !31 Haag Kong-Nouma 2.357 4.103 54t M.lbouras-Noums 2.900 1.597 1.149 Hong KonrNg our 3.396 4.112 Sydney/Brisbaa-PaSo Plsgo 3.800 2.660 363 Slappore-Peot Louis 2.073 3.324 Rnamdaa-Port Louis 3.224 6.997 Siappoe-Pont Louis 4.370 3.324 Romeraar-Port Louis 2.340 6.997 Shplgpo a-Por Louis 1.3t3 3.324 121 Rou,dm-Port Louis 2.533 6,997 Siahpte-Pot Louis 1.,13 3.324 Haag KongPoat Louis 2.208 4.747 BNe Keor-Pon Laub 2.143 4.747 730 Hoag KonrPont Louis 2.727 4,747 Hong Kong-Pon Louis 2.143 4.747 Haag Kong-Por Lui 2.208 4.747 SlagpoPoen Louis 1.333 3.324 Hang Kong-Port Louis 2.20t 4.747 Siaga p_-Pana 3.117 3,363 24 Roeu_dam-Pape 4.114 9.417 ShAgapev-PSp_P 3,377 6.549 RONSWdaM-PApanI 4,605 0.417 Loaede-Pspmit 3.390 15.123 - iJ9 - A2L ACTUJAL flUIG#7T VOLUM RATE DISTANCE CFU pw ROUTIE Ocup ROUT (ASIU) (£N) Guam) Rounda-Pp." 3.633 9.417 SIappws-915 3J06 6.549 605 Rounvdauh _pa 3.6Z3 9,417 Mes KasrP&m 4.026 7,13t Hoag Keag-Raslos 3.247 : 2372 Hefs K*-R .e 3.06 4.,72 1.123 Romwadus4uAIOU 3.701 '021 Ha Kong-Rms 3.117 4.r37 Rmu4sin-Ruueiiin 2.240 .21 Rs s_udm-auais 3,.6 7,021 8sgmpsus-Rn.sIss 4.70 3,450 .. osg Kaq-Rmumias 4.226 4.372 _o* KmrRiaM_ 3.117 4,37r2 1^t KmTrIC- 2.r. 23$33. S _pKsnTm2 2.93 2.f2 24 2sgp-Ta.m_" 3.117 3.575 Hag Kmg-TAMvs 3.306 3.375 Si_ pa- Taa_ 4.370 3.575 HMsKcomTammv 4.026 42738 SwpeT_ n" 2.922 357S ..5 Hea Koag-Tanave 3.96 4.73. SiappatChrul u d 2Ia 9 9 i S__Y _ : )351 3.150 :4 Ownr siagpwes-Valpu^ so 4.351 9.945 C.. phi Hong Kla.-Velp. ais* 4.675 10,212 ~?4as-1saZad isa) S _sVppa iVpuu 4.236 9.945 Hong Kng-Dut*m 1. .4S 6.O60 Hong Kont-Due'n 1.753 6.960- Siappo-Durba 14 5.6S0 Mulbous-US 3.ICO 6.60a 2!.90| Suva-'US 3.120 4,64.0 548 Nomluated Aaaeklaa-Apis 3,211 1.473 . Patie AwAad-Apia 2.758 14723 1slands Hong KnagApin 3.636 5.550 ?3 1 Slappt.-A* 4.026 5.635 Slna"OiuAVI 3.296 5.635 36 Saspew-Apls 3.396 5,635 Sydnsy/3Bh -Apim 3.500 2.550 3631 AwAklu-Apk 2.7',1 1.473 Hesl KIon-Apig 3.m 5.=50 121 AuekIad-Ffpi 2.302 1.140' Hoag KIa-:ji 2.J97 4.545 ll| BdIAsplji 2.35 1.545 2.427 Mesa Kng- Fiji 2.1 4.545 1.s25 ALusklui-PIjI 2.538 1.140 Slpais-7iji ; 3.117 4.736 Nsw Zalnd-PFij 1.765 1.140 3.610' AuIjid-P 2.S36 i.140 AmnladIj 1.471 1.140 " mouuwKH_ aa. 2. 2.200 363, Awk%od-H*WIt 3.350 1,3C0 .. I HRag Kag-Haesi 2.937 3.300 73 Slip lteasmnis 1.958 3.300 31oppoas-H4WAM 2.203 3.300 .. Haal KagrHalms 2.333 3,300 I^ I - 190 - .ZL ACTUAL FUJONT VOLUbEM RATE DISTANCE TEU VW ROITF GROUP ROUTES (ASIrU) (Sme) a !rToada-Hao'ms 3,442 113143 Syduy/Bri1ane-Homism 3.CCo 1.46 020 Auckabad-HonAn 3.346 1.300 3SngmpotKimbe 2.203 2.329 :Shpprs-La 2.:03 3.021 i I is-L4 3.247 11.23, Siappo-Lautoks 3.247 4,690 Kull-Lauoka 4,236 11.363 .. Sb_pe,-Laub 2.987 4.690 36 HuMl n 3.247 12.084 Si ppow-Madmu 3.247 2.439 Siappm-Madag 2.208 2.439 Ham K_gNub'sIof 4,156 4.935 73 sippao -Naldo 3.636 5.126 . . SIIappv,N4lkaW 3.3% S. 126 242 Uvlabm hPa Now Oaam 2.100 1.3IS 367 SbIq.pa-Pam Now a0mm 1.t33 2.540 .. Sipoa ePapua Now Guinn 1.94t 2.340 Muboum-Pet Momaby 2.400 2.303 10,950 AMWI &PoS 10resby 3.350 2.254 Hull-PonMormby 3.2:47 11, 39 AuckldanPortMorasby 3,346 2.2 54 Slaapou-PontMoresby 2.208 2.540 Hull-Raul 3.247 12.014 .. Slag)PoreSuao 4,026 4.,000 Hang Kong-Santo 4.156 3.920 73 Roaudam-Sao 6.486 12.476 Sl3apporSuva 2.987 4.736 Sinappoui-Suva 2.987 4.736 60 Slapporu-gSva 2.,57 4,736 4t4 S3appore-Suva 3.117 4,736 1.209 Hull-Suvs 4,216 11,295 Melbourne-Suva 2.500 2.250 1.209 Melbourne-Suva 2.600 2.250 1,643 Slappors-Tamaw 3.636 3.750 Syda.y/Brisbano-Taraws 3.300 2.220 367 Hong Kong-Tanwa 4,545 3,500 73 Auckland-Tar wa 3,624 2.270 Mdlbouwn-Taunwa 3.000 3,347 433 Medbourrn-Vila 2.500 1.990 423 S3ngmapr-Vils 3,247 4.100 60 A Uklad-VII 2.791 1.360 Siappors-VUI 4,026 4.100 Haug KoorVUa 3.636 4.100 121 Mdbournr-VDn 2.500 1.990 363 RoSwdaa-Vil 6.486 12.296 Hoag KonrVUn 4.156 4.100 73 S3lapors-Wowak 2.208 2.729 Si oWewak 3,247 2.729 Total: 152 Now: (I) .. Not vala. (2) T dumane. 1 London to Papn: is 15. 123 ilm via the Sum Cal. -191- ANNEX M.4 PACIIlC ISLANDS TRANSPORT SECTOR SURVEY THE SHPPING COST MODEL The shipping cost model used in the Route Information assessment of shipping rates is a spreadsheet model which calculates voyage costs as a A route is defined as a sequence of port function of a range of user-supplied parameters. calls. The volume of cargo loaded and This annex defines the main inputs and describes discharged at each port (including the first port the main options available. of call) must be also specified. MODEL STRUCTURE Sailing distances are from an internal cross-reference table, and the number cf calls The costing model estimates the cost ot mad at a port is also automatically retrieved. operating container vessels of a specified size. over a fixed route, given a defined patteiz of A route that has already been defined can cargo uplift and discharge. Voyage costs are be modified to include changes in port calls, calculated from vessel information, route loading patterns and sailing distances. information and operating and cost parameter information. Operating and Cost Parameters Vessel Information The model includes 'typical" unit cost data, and a set of relationships for estimating The minimum information that must be derived characteristics (such as crew size and supplied is the size (in TEU capacity) of the fuel consumption), which are used as the default vessel and its age. Using relationships derived values in route costing. Table 1 shows the from analysis of the current fleet, these two parameters which can be modified: parameters are then used to estimate the other vessel characteristics required in the costing COST STRUCTURES process. However, these characteristics such as crew size, consumption and deadweight can be The majority of costs incurred in subsequently modified if required. operating a liner shipping service are "direct" in the sense that they could be avoided if a Some of the cost elements-e.g., port particular service was not offered. charges-are estimated using step functions, with breaks at 100 TEU, 300 TEU, 700 TEU and Conventionally, shipping service costings 1,000 TEU. Varying vessel size within each divide costs into three cost categories: vessel segment of these functions will have no impact costs, voyage costs and cargo costs. on these cost components. However, most cost elements are estimated using functions that Cargo Costs respond to changes in vessel size in a continuous fashion. Cargo costs are those costs which vary directly with the volume of cargo carried, and include such item as container hire, stevedoring - 192 - ANNEX M.4 Table 1: OPERATING AND COST PARAMETER INFORMATION Parameters Examples 1. Capital Costs Building costs, interest rates, economic life of the asset for depreciation and scrap value of the vessel 2. Operating Costs Crew costs, maintenance costs and insurance administration costs 3. Voyage Costs Fuel consumption, fuel prices and speed for vossel size 4. Cargo Costs Costs of container hire, pre-tripping costs and agency fee 5. Ports Port to port distar,e, port charges and cargo handling rates by port costs and agency fees (usually paid on a abandonment of a service would lead directly to commission basis). reductions in administrative costs. In this sense, the bulk of administrative costs are also 'direct". Voyage Costs DATA SOURCES Voyage costs are those which are determined by the voyage pattern: principally The unit costs and relationships used in fuel costs and port charges. the shipping cost model as defaults are drawn from a variety of sources. All c.osts in either of these categories are avoidable costs assoc.ated with the making of a In general, the starting point for cost particular voyage. information was the ioint ESCAP/SEATAC study. Intra-Asean Shipping Study (ESCAP: Vessel Costs Bangkok, 1988). Vessel cost data, particularly relating to vessel capital costs, was supplemented Vessel costs are those costs which are by information drawn from articles and fixed by holding a vessel in service, irrespective schedules in various issues of Lloyd 's-Maritime of the route on which it is deployed. These Asia and Containerization International. include financing and depreciation costs directly Published and unpublished data are also drawn associated with the particular vessel, repair, from a number of studies of Australia/Asian maintenance and survey costs, essential stores shipping; (Gallagher and Meyrick, ASEAN- and victuals, insurance and crew wages. All of Australia Liner Shipping: A Cost-based these wosts are avoidable if the vessel is disposed Simulation Analysis (Canberra, 1984); of. Thomspon-Clarke, Japan-Korea Shipping Service Study for Transport Tasmania, (Hobart, In addition, it is conventional to include 1986); Center for Transport Policy Analysis, amongst the vessel costs an allocation of University of Wollongong (CTPA); Australian wadministration" costs-essertially corporate Flag Shipping and the Balance of Payments overheads. In the management structure of most (CTPA: Wollongong, 1988); CTPA, Stateships liner shipping companies, each service comprises External Benefits Study (CTPA: Wollongong, a separate profit center, and it is likely that the 1990); ESCAP, Regional Maritime Strategy - 193 - ANNEX I1.4 Study) (Liner Shipping Services to and from AUSTRALIA Australia (BTCE 1986); Trans-Tasman Shipping (BTCE 1987), A Model for Estimating Cargo Stevedoring costs are typically about Ship Costs (BTCE 1987). A$300 per TEU in Australia and the Pacific Island countries. At present (1991) stevedoring Port charges, handling rates and delays costs are relatively lower in New Zealand are estimates, primarily based on the rates used (approximately A$200 per TEU), Apia and for smaller ports in the ESCAP Regional Nuku'alofa (about A$250) and rather higher in Maritime Strategy Study. For the Pacific Island Pago Pago. ports, however, this was supplemented by information collated in the 1984, Regional Currently the Australian Pacific Islands Transport Survey of the South Pacific, and Rate Agreement has zero Bunker Adjustment information collected in the current study. Factor (BAF) and a 14.5 percent Currency Adjustment Factor (CAF) for the Noumea trade. Relationships between vessel size and age BAF and CAF for the Australia to Fiji trade are and characteristics such as crew size and presently set at zero. operating speed were obtained by analysis of the details of the current fleet operating Pacific HONG KONG Island routes provided in ContainerIoi2lon Internadonal Yearbook 1991 (National Magazine Terminal handling charges in Hong Kong Cc: London, 1991). are typically HK$500 per TEU. Container stuffing charge (CFS) average about HK$80 per Port to port distances were taken, revenue tonne or HK$62 per cubic meter. wherever possible, from the port to port distance tables presented in UJoyd's Maritime Atlas. Some shipping lines avoid terminal However, the coverage of these tables is limited. handling costs by loading and discharging cargo Where the distance between a pair of ports was mid-stream. not obtainable from this source, the port to port distance table from the ESCAP Regional There are currently no terminal charges Maritime Strategy Study was consulted. If the being levied on Indian Ocean services ex Hong distance sought was unavailaole from either of Kong because of competition. these sources-and this was commonly the case with the Pacific Island ports other than Suva and SINGAPORE Port Moresby-distances were estimated by scaling from maps; if possible, from Lloyds A current list of official port and Maritime Atlas, otherwise from the Encyclopedia handling tariffs in Singapore are listed in Table Britannica World Atlas. 2 and Table 3. These charges are indicative of the charges levied by major ports in the world. ADDMONAL SHIPPING CHARGES UNITED KINGDOM Indicative costs, which are additional to the net ocean freight rates (i.e., liner in liner out Currency Adjustment Factor (CAF) to rates), used in the shipping cost model analysis the Indian Ocean islands is currently are set out below. +7.13 percent. For services to New Caledonia, there is a terminal handling fee of £11.00 per tonne/cubic meter or part thereof. - 194 - ANNEX M.4 For services to Fiji, the terminal handling A bunker charge of -101 ECU (European fee is £125 per container load. Currency Units) is levied on rates for some services to New Caledonia, Tahiti and Vanuatu. ROlTERDAM Currently, a Bunker Adjustment Factor of + 8.1 percent is added on rates to the Comoros, Mauritius, Reunion and Seychelles ex Rotterdam. Table 2: MAIN CHARGES La FOR CONTAINER OPERATIONS, Lb 1990 Port of Singapore Authority: 20' box 40' box Stevedorage Charge: FCL-Loaded 155.00 220.00 FCL-Empty 85.00 123.00 LCL 325.00 452.00 Transhipment /c 100.00 145.00 Overheight/Overwidth FCL 300.00 500.00 LCL 400.00 600.00 Transhipment 200.00 300.00 Lift or./Lift off charge: Loaded 55.00 82.50 Empty 20.00 30.00 La All amounts in Singapore dollars. Lb Most shipping lines quote a terminal handling charge between S$150-165 per TEU in addition to the net ocean rates. Ic Excluding rebates. Source: Port of Singapore Authority, PSA Tariff, 1990. - 195 - ANNEX UJ.4 Table 3: OTHER CHARGES IN SINGAPORE, 1991 Local haulage (say 1 hour) S$140 per 20' box S$200 per 40' box Breakbulk freight S$12 per revenue tonne Minimum charge S$60 per bill of lading Documentation (Tradenet) S$40 per bill of lading Storage S$3 per TEU per day S$5 per m3 per week (breakbulk) Forklift S$2.50 per revenue tonne Stuffing/Unstuffing (LCL) S$100 per 20' box SS160 per 40' box S$12/metric tonne (one end) Laborer S$40 per manday Source: Singapore Freight Forwarders Association 1991. Table 4: CHARGES FOR SIPPING SERVICES EX UNITED KINGDOM, 1991 Pacific Port Islands Louis Male UK Terminal handling charge £75.00 £60 per 20' box £11 per tonne £95 per 40' box $11 per tonne/m3 Bunker Adjustment Factor + 15% +29% + 16% Bill of Lading (per shipment) £15.00 £15.00 £15.00 Freight agency commission 2.5% 2.5% 2.5% (% of the Basic rate) Source: Ariel Maritime, UK, 1991. -196 - ANNEX m.5 PACIFIC ISLANDS TRANSPORT SECTOR STUDY PART mI: COMPARISON OF REGIONAL TRANSPORT COSTS STATISTCAL ANALYSIS OF SHIPPING RATES-DETAED RESULTS Two models are estimated. The first the estimated coefficients for distance and model (Test 1) includes separate dummy density are different as between any of the island variables for the PMCs, the comparator Pacific groups. The second model (Test 2) includes a Ilands, and the Indian Ocean comparators. The single dummy variable for all islands as one results in Table 1 show that both distance and group; this model gives similar results to the volume have a statistically significant effect on first. Detailed results are given in Table 2. freight rates, but that there is no evidence that Table 1: SHIPPING RATE MODEL-REGRESSION REsULTS Bm ModW I. = 1099.17 * DO-5 * V4-00 (2.86) (-2.38) Test 1: Significance of Island Type = i584.59 * DI"' * V005 * (e N)007 * (ec)94I * (e)422 (2.14) (-2.59) (-0.46) (-0.59) (-1.25) where N = dummy variablc for the nominated Pacific Islands (1,0) C = dummy variable for the comparator Pacific Islands (1,0) I = dummy variable for the comparator Indian Ocean Islands (1,0) Test 2: Significance of Island Routes - 1252.77 * D0 14 * V40' * (eA1)4r (2.61) (.2.37) (-0.44) where Al = dummy variable for all islands (1,0) Notes: (1) T-statistics are in parentheses. (2) Independent variable is significant at 0.05 level when t-statistic is greater than + 1.96 o,r less than -196. - 197 - ANNEX MS Table 2: SHIPPING FREIGHT RATES-LOGLINEAR REGRESSION RESULTS BASE MODEL Regression Output: Constant 7.002309 A = 1099.168 Std Err of Y Est 0.190296 R Squared 0.240249 No. of Observations 51 Degdrees of Freedom 48 X Coefficient(s) 0.151304 -0.03888 Std Err of Coef. 0.052938 0.016315 T-Statistic& 2.86 -2.38 Ln Dist. Ln. Vol. TEST 1 Regression Output: Constant 7.368079 A = 1584.587 Std Err of Y Est 0.190646 R Squared 0.285116 No. of Observations 51 Degrees of Freedom 45 X Coefficient(s) 0.121913 -0.04618 -0.69334 -0.095266 -0.22411 Std Err of Coef. 0.056853 0.017823 0.1509336 0.1603096 0.179789 T-Statistics 2.14 -2.59 -0.46 -0.59 -1.25 Ln Dist. Ln Vol. Nominated Comparator Indian Pacific Pacific Ocean Islands Island Island Dummny Dummy Dummy (1.0) (1. 0 1. TEST 2 Constant Regression Output: 7.133108 A = 1252.765 Std Err of Y Est 0.191923 R Squared 0.243308 No. of Observations 51 Degrees of Freedom 47 X Coefficient(s) 0.144786 -0.04154 -0.066079 Std Err of Coef. 0.055445 0.01748 0.1516000 T-Statistics 2.61 -2.37 -0.44 Ln Dist. Ln Vol. Island/ Non-island Dummy (1.0) - 198 - ANNEX m.6 PACIFIC ISLANDS TRANSPORT SECTOR STUDY VOLUME ONE AiR FARES, DISTANCES AND TRAMC VOLUMS FOR PACIFIC ISLAND AND OTiaR RourEs, 1991. - 199 - IID.6 Tabl s AIR SEVICES ROUTE DATA ACTUAL ACTUAL AVERAGE DISTANCE: DISTANCE: ANNUAL. RETURN RETURN MAXIMUM SHORTEST SEAT, ECONOMY EXCURSION PERMIrTED OPERATED CAPACITY. FARES FARES MILEAGE MILEAGE (000)1 ROUTE TYPES ROUTES AS AS I 1. PACIFIC RIM Aucklan-Nad; 1,087 67 Ii.;4i ;.341 33.21 TO PACIFIC Au-aand-Port Vib 1."4 649 1.392 1.392 5.7' ISLANDS Avo ili-Apia 1,345 933 1.S01 1.S01 '..9j Awd_d-Ho 1.418 990 2.191 2.191 11.3i Aueklaad-ToagcApu 1.147 7I. 1.246 1.246 8.I Auclkand-PonMoresby 2.6 1.5J1 2.750 2.750 9.1 j Ausklad-Tanw 2.027 2.028 2.793 2.793 2.0 Au m_d-Noumm 1.066 54S 1.374 1.145 26.5 A ekl_-Papum 2.468 1.:47 3.0S2 2.544& 23.4 Brn_abe-Honiam 1.:33 514 -1.587 1.323 ::.3 Bri w-Nadi 1.3;2 764 1.686 168 21.3 ,isba-No,ouam 1.166 811 '.035 903 :4.9 Brisbmne-Port.omresoy 1.'68 796 1.S60 1.300 59.0 Brisb PatPonVila 1.422 625 !.10 !.1S0 '!° Cairn-Hoiaza 1.022 67 1.529 1.0J9 '.7 Cairs-PortMoresby 756 489 528 524 3'.9 ! Hong Kooq-Pon Moreaby 2.095 1.475 3.760 3.134 10.9: Mond&-Guam 1.121 614 !.915 1,596 143 | MinU-Port Moaey 2.337 1.239 2.930 2.J42 10.31 M lboume-Nadi 1.613 842 2.403 2.403 43.7 Melboumne-Nouines 1.716 952 2.004 1.669 16.6 Mdboune-Port Vila 1.922 914 1.981 1.981 7.5 Slappore-Noume 3.471 2.033 5.608 4.674 41.: Siappore-Port Moresby 3.063 1.743 3.686 3.072 21.3 Sydney-Nadi 1.440 702 1.971 1.971 124.1J Sydney-Port Vila 1,310 ,,4 1.544 .344 !5.0 Sydey-Apis 2.416 1.191 2.695 :.695 9.4 SydnerHonma 1.548 1,02? 2.145 1.785 10.9 Sydney-Tonagapu 2.423 1.07J 2.511 2,511 3.7 S,vdne-Port Moruby 1.556 904 2.100 1.712 54.6 Sydoey-Tarwa 2.781 2.053 3.423 3.423 3.7 Sydney-Noum 1,264 756 1.477 1.231 35.1 3ydnoy-Papeoe 3,396 1.712 3.302 3.802 32.5 Sydncy-Raronta 2.596 1,147 3.217 3.217 7.5 2. INTRA-PACIFIC Api-Hoooiuiu 1.161 676 4.r72 2.607 9.6 REGION Apio-No_... 1.045 724 2.035 1.541 7.5 Apis-Pago Pap 110 83 85 8i 68.9 Apisi-Ratonp 670 466 94 944 15.0 Apir-Suva 5:1 413 I00 '70 5.6 Apie-Tongetpu 461 321 557 557 .5 Fuotfud-Sva 560 560 661 661 1.9 Fmufsu*-Majuro 966 966 1.214 1.214 3.7 Funafw-Tsrwa 594 594 802 302 3.J GOaU-Poknpe 935 738 1.074 1.074 56.9 Gv_-Thak 616 517 635 635 56.9 Gmm-Ken 1.241 1.053 1.413 1.413 27.0 Mujwo-Hondulu 1.414 1.3 2.419 :.419 7:.4 Nsdi-Toinmu 419 419 540 S40 15.3 - 200 - h.Z.h AIR IKVICES ROUTK DATA cout'd. ACTUAL / CTUAL AVERAGE DISTANCE: DISTANCE: ANNUAL RETURN RETURN MAXIMUM SHORTEST SEAT' ECONOMY EXCURSION PERMTrrTED OPERATED CAPACrrY FARES FARES MILEAGE NILEAGE (0) I ROUTE TYPES ROUTES AS AS Nadi-Port Vila 643 4JJ 602 602 17.9 I Nedi-Tamwa 1.341 1.341 1.452 1.452 3;7I Nei-ApB 669 47S 761 761 16.2 I Ndiw-oeimm 1.334 184 1.306 1.30 ll.31 Nadi-HKAiIu 2.409 917 3.508 3.174 172.1 Nadi-Pout Moruby 1.563 1.095 2.177 2177 59.2 i NIN _am 7 56 756 1.12= 783 4.5 I Nud-Rumug 1.446 733 1.300 528 10.9 Ndi-WiUll lLed 723 527 527 Nommsa-Papme 2.341 2.341 2.949 2.949 19.2 Pott Memeby-Homaz 857 555 37 r73 9.4 |Port Moraky-jaypum 69S 693 S44 704 3.3 Por VDa-Nog.. 333 .91 399 333 25.3 3. AUSTRAUAN Sydny4angkok 4.X2 1.346 5.624 4.687 364.3 INTMENATIONAL Sydney-Honoulu 4.716 1.713 6.093 5.078 610.9 Sydney-AuekIand 1.042 576 1.343 1.343 705.0 Sydney-Hong Kong 4,062 1.530 5.503 4.586 313.0 Sydney-Los AgDeS 5.678 1.932 8.997 7.498 291.2 Sydncy-Bali 2.594 916 3.446 2.t72 140.4 Sydney-Tokyo 3.902 i.671 6.925 4.363 395.2 Sy. .y-Jakau 2554 985 4.105 3.421 172.41 Sy( ir-Singspore 3.144 1.323 4.699 3.916 620.9 1 4. WESTERN Penw. Albny 234 192 375 375 12.4 1 AUSTRALIA Prh-BDroome 906 596 1.042 1.042 103.2 ! DOMESTIC Perdu-Carnrvon 486 316 507 507 3.72 PeMh-Derby 836 575 1.115 l.11 53.3 ! Perth-eu dztos 3C2 200 230 230 44.8 Penh-KIcalgooi 394 265 334 334 69.3 Perth-K.anukh 712 478 777 m77 85.3 Perth-Kuauaim 1.110 730 1.376 1.376 42.3| 5. AUSTRALIAN Sydney-Gold Cams 4"S 332 422 422 455.4 1 DOMESTIC Sydney-Hamito,a Wand 760 560 949 949 147.41 TOURIST Sydney-Alice Springs 374 650 1.256 1.256 90.7 Sydne-Laummeon 51 386 s5 569 3 4.2, MaIbowns-Gold Comm 676 492 327 327 240.3 Molbou&su-Cuhm 1.055 792 3.440 1.440 26.4 i arc Toe-Thwnlwde 626 462 692 692 345.3 1 BrdeAer-aenky 516 3U4 497 497 77.4 Irisb Roe p 402 298 323 323 434.3 Alaide-Aliee Springs 672 437 t20 320 92.1' Adelid-Dawi 1.016 753 3.630 i.630 107.I 6. NEW ALAND Amcklmi-W"Iom 305 214 29 299 738.5 DOMESTIC AuekldCrImhu,rch 434 290 464 4S6 S' 4. AAelad-Dunodin 524 361 66 668 24:.1 Auskd_ed-Hain 133 86 66 66 22.5 _esklind- 1eeeua 230 161 113 113 52.01 WeUlnse-Queenmewn 524 36 406 406 45.5 waimnpe.-Invesarl 415 2.90 479 47. 3 - 201 - L AIR SnVICIS ROU DTA cont'd. ACTUAL ACTUAL AVERAGE DISTANCE: DISTANCE: ANNUAL RETURN RETURN MAXIMUM SHORTEST SEAT ECONOMY EXCURSION PERMIrTED OPERATED CAPACITY FARES FARIS MW.EAOE MI;EAOE (000) ROUTE TYPES ROUTES AS AS 7. PAPUA NEW GUINEA Pon Moeby-Las 256 205 202 '02 179.7 DOMESTIC Pon Moreby-Mt.Hapt 356 235 321 321 76.31 PgeMe,uy-Oomk 307 246 264 264 53.2j Pont Meby-Madmm 348 273 308 306 71.1 Pot Mom.by-W.w.k 485 388 474 474 41.2 Pat Moaby-Wa Papu 256 256 IS0 ISO 6.21 PonMaby-Tari 413 330 334 3U4 8.3 1 PM tM,s3v-D! 5 S 252 275 275 10.41 IS. INDIAN OCEAN Jo maubuq-Maawas 1.6C0 333 2.2S6 1.905 73.0; ISLANDS Lando_-Mawuus 5.385 2.216 7.32 6.074 62.4! Lcades-Sqiw.um 5.018 1.961 6.357 5.066 n2.s ! Pwb-Maudu 4.769 2.051 7.140 5.367 t1.9 1 Paru-Rusa 4.769 1.991 7.039 5.319 145.4 Penh-Cocos Lands 1."20 990 1.720 1,r70 3.7, Penh-Christ Island 1.20 990 1.623 1.62S :3 TOTAL 111 - 202 - ANNE( 1.7 PACIFMC ISLANDS TRANSPORT SECTOR STUDY Am SERvIcE RourEs FOR PACxFC ISLANDS, COMPARATOR COUNTRIES AND PACIC Rim Table 1: PACIFiC Rim TO PACIC ISLANDS-AIR SERVICE ORIGINS-DESTINATIONS, 1991 No. of Flights Origin-Destination AirlineaL per Week Flight Routingp 1. Aukand-Nadi FP 3 Auckland-Nadi NZ 1 Auckland-Nadi IE I Auckland-Nadi-Pot Vila-Homiar NZ/CP 2 Aucknd-Nadi-Honolulu-Vaucouver NZ 1 Auckland-Nadi-Rartoa NZ 1 Auckland-Nadi-Honolulu-Los Angeles NZ 2 (1) Brisbmne-Auckland-Nadi-Honolulu (2) Aucldand-Nadi-Honoluh-Lus Anees 3) Adelaide-Auckland-Nadi-Honolulu 2. Aucklad-Port Vila NF 2 Auckland-Port Vila 3. Aucldand-Apia PH 2 Auckland-Tongtapu-Apia PH 1 Sydney-Auckland-Tongatapu-Apia NZ 2 Auckland-Tongatapu-Apia 4. Auckland-Honiara IE 1 Auckland-Nadi-Port Vila-Honiara IE 1 Auckland-Port Vila-Honiara s. Auckland-Tongatapu NZ 3 Auckland-Tongatapu PH 1 Sydney-Aucklcand-Tongatapu-Apia PH 2 Auckland-Tongatapu-Apia 6. Auckland-Port Moresby Auckland-Cairns NZ/QF 7 Auckland-Sydney-Cais NZ 1 Auckland-Cairis Caims-Port Moresby PX/QF 6/7 Cairns-Port Moresby 7. Auckland-Tarawa Auckland-Nadi Nadi-Tarwa CW 2 Nadi-Funafuti-Tarawa-Majuro 8. Auckland-Nouniea NZ 1 Auckland-Noumea SB 1 Auckland-Noumea UT 1 Papeete-Auckland-Noumea 9. Auckland-Papeete UT 1 Noumea-Auckiand-Papeete NZ 2 Melbourno-Auckland-Papeete-Los Angeles 10. Brisbane-Honiam IE 1 Brisbane-Cairns-Honiara IE 1 Brisbane-Honiara QF 1 Sydney-Brisbane-Honiaua 11. Brisbane-Nadi FS/QF 2 Brisbane-Nadi PH 1 Sydney-Brisbane-Nadi-Apia-Pago Pago- Rarotonga 12. Brisbane-Noumes SB 3 Brisbane-Noumea - 203 - ANNEX M.7 Table 1: PACnFC RIM To PACIFC ISLANDs-AIR SERVICE ORIGINS-DEmNATIONS, 1991 No. of Fli hts Origin-Destination Airlines per Week Flight Routin' 13. Brisbane-Port Moresby PX/QF S Sydnoy-Brisbane-Port Moresby PX 1 Brisbane-Port Moresby 14. Brisbane-Port Vila NF 1 Brisbane-Port Vila NF 1 Sydney-Brisbane-Port Vila 1S. Cairns-Honiarm IE 1 Brisbane-Cairns-Honiara 16. Cairns-Port Morosby PX/QF 3/4 Cairns-Port Moroby 17. Houg Kong-Post MorosbyPX/CX 1 Hoag Kong-Port Mormsby 18. Mani!a-Guam CO 2 Manila-Guam PR 2 Manila-Guam ON 1 Manila-Guam-Nauru Island co 4 Manila-Guarn-Saipan co 3 Manila-Koror-Guam-Saipan co 2 Manila-Saipan-Guam 19. Manila-Port Moresby PX 1 Manila-Port Moresby 20. Melbourne-Nadi FJ/QF 4 Melboume-Nadi 21. Melbourne-Noumea SB 1 Melbourne-Noumea ON 1 Melbourne-Sydney-Noumea-Nauru Island 22. Melbourne-Port Vila NF 1 Melbourne-Port Vila 23. Singapore-Noumea VT 1 Paris-Singapore-Jakarta-Noumea VT 1 Paris-Singapore-Jakarta-Sydney-Noumea 24. Singapore-Port Moresby PX/SQ 2 Singapore-Pu L Moresby 25. Sydney-Nadi FJ/QF 3 Sydney-Nadi CP 1 Sydney-Nadi-Honolulu-Toronto PH 1 Sydney-Brisbane-Nadi-Apia-Pago Pago- Rarotonga FJ/QF 2 Sydney-Nadi-Honolulu-Los Angeles Ha'apai (Tonga)-Sydney-Nadi-Honolulu- Los Angeles 26. Sydney-Port Vila NF 1 Sydney-Brisbane-Port Vila NF 1 Sydney-Port Vila 27. Sydney-Apia PH 1 Sydney-Brisbane-Nadi-Apia-Pago Pago- Rarotonga PH 1 Sydney-Auckland-Tongatapu-Apia 28. Sydney-Honiarm QF l Sydney-Brisbane-Honiara 29. Sydney-Tongatapu PH 1 Sydney-Auckland-Tongatapu-Apia 30. Sydney-Port Morosby PX 2 Sydney-Brisbane-Port Moresby QF 3 Sydney-Brisbane-Port Moresby - 204 - ANNEX M.7 Table 1: PACIIC RIm TO PACIC IsLANDS-AIR SERVICE ORIGINS-DEfTINATONS, 1991 No. of Flights Origin-Deatination Airlines per Week Flight Routings 31. Sydney-Tarawa Sydney -Nadi Nadi-Turava CW 2 Nadi-Funafuti-Tarawa-Majuro 32. Sydney-Noume F 2 Sydney-Noumea UT I Paris-Singaporo-Jakarta-Sydnoy-Noumoa ON 1 Melbourne-Sydney-Noumea-NauruIsand UT I Sydney-Noumoa-Wallis Island-Papeoto SD 2 Sydney-Nouw. 33. Sydney-Papoete UT 1 Sydney-Noumee-Wallis Island-Papeeto QF 2 Melbourne-Sydney-Papeete-Los Angeles 34. Sydney-Rarotong, PH 1 Sydney-Brisbane-Nadi-Apia-Pago Pago- Rarotonga Ij Airline codoa ar given in Annex 9. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Ansett Timetable, 1991. - 205 - ANNEX M.7 Table 2: INTRA-PAcIFIc REGION-AVIATION ORIGINS-DESTiNATIONS, 1991 No. of Flights Origin-Destination Airlines L/ per Week Flight Routings 1. Apia-Honolulu HA I Apia-Honolulu 2. Apia-Noumea PH 1 Apia-Noumea 3. Apia-Pago Pago PH 32 Apia-Pago Pago TS 18 Apia-Pago Pago 4. Apia-Rarotonga PH 1 Sydney-Brisbane-Nadi-Apia-Pago Pago-Rarotonga S. Apia-Suva FJ 1 Apia-Suva FJ 2 Apia-Nadi-Suva 6. Apia-Tongatapu PH 2 Apia-Tongatapu-Aucklaad PH 1 Apia-Tongatapu-Auckland-Sydney WR Apri-Tongatapu NZ 2 Apii-Tongatapu 7. Funafuti-Suva PC 1 Funafuti-Nadi-Suva 8. Funafuti-Majuro CW 2 Nadi-Funafuti-Tarawa-Majuro 9. Funafuti-Tarawa CW 2 Nadi-Funafuti-Tarawa-Majuro 10. Guam-Pohnpei (Ponape) CO 4 Guam-Truk-Pohnpei-Kosrae (Kusaie)- Kwajalein (Marshall Islands)- Majuro-Johnston Isbands- Honolulu Guam-Truk-Pohnpei-Kosrae- Kwajalein- Majuro-Honolulu Co 3 Guam-Truk-Pohnpei ON I Guam-Truk-Pohnpei 11. Guam-Truk CO 4 G(uam-Truk-Pohnpei Gusm-Tntk-Pohnpei CO 3 Guam-Truk-Pohnpei ON 1 Guam-Truk-Pohnpei 12. Guam-Kosrae (Kusaie) CO 3/4 Guam-Truk-Pohnpei-Kosrae (Kusaie)- Kwajalein (Marshall Islands)- Majuro-Johnston Islands- Honolulu Guam-Truk-Pohnpei-Kosrae- Kwajalein-Majuro- Honolulu ON 1 Guam-Truk-Pohnpei-Kosrae 13. Majuro-Honolulu CW 2 Majuro-Kwajalein-Honolulu CW 2 Majuro-Honolulu CO 4 Guam-Truk-Pohnpei-Kosrae- Kwajalein-Majuro- Johnston Island- Honolulu Guam-Truk-Polmpei-Kosrae-Kwajalein- Majuro-Honolulu 14. Nadi-Tongatapu FJ 6 Suva-Nadi-Tongatapu Nadi-Suva-Tongatapu FJ 1 Suva-Nadi-Tongatapu 15. Nadi-Port Vila FJ 3 Suva-Nadi-Port Vila NF 1/2 Nadi-Port Vila - 206 - ANNEX III.7 Table 2: INTRA-PACIFIC REGION-AviATiON ORIGINS-DESTINATIONS, 1991 (cont'd) No. of Flights Origin-Destination Airlines per Week Flight Routings 16. Nadi-Tarawa CW 2 Nadi-Funafuti-Tarawa-Majuro 17. Nadi-Apia FJ 4 Suva-Nadi-Apia Nadi-Suva-Apia PH 1 Sydney-Brisbane-Nadi-Apia-Pago Pago- Rarotonga 18. Nadi-Honiara FJ 1 Nadi-Port Vila-Honiara IE 1 Nadi-Port Vila-Honiara 19. Nadi-Honolulu CP/NZ 2 Auckland-Nadi-Honolulu-Vancouver NZ 1 Brisbane-Auckland-Nadi-Honolulu 2 Auckland-Nadi-Honolulu-Los Angeles 1 Adeaide-Auckland-Nadi-Honolulu CP 1 SydnIy-Nadi-Honolulu-Vancouver QF 1 Sydney-Nadi-Honolulu-Los Angles 1 Hobart-Sydney-Nadi-Honolulu- Los Angeles 1 Melbourne-Nadi-Honolulu-Los Angeles VK 1 Nadi-Tarawa-Christras Island-Honolulu 20. Nadi-Port Moresby QF 2 Nadi-Sydney-Brisbane-Port Moresby QF 2 Nadi-Brisbane-Port Moresby IE 1 Nadi-Port Vila-Honiara 21. Nadi-Noumea SB 1 Wallis Island-Nadi-Noumea 22. Nadi-Rarotonga NZ/PH 1 Auckland-Nadi-Rarotonga I Sydney-Brisbane-Nadi-Apia-Pago Pago- Rarotonga 23. Nadi-Wallis Island SB 1 Noumea-Nadi-Wallis Island 24. Noumea-Papeete UT 1 Noumea-Auckland-Papeete SB/UT 1 Sydney-Noumea-Wallis Island-Papeete 25. Port Moresby-Honiara IE/PX 2 Port Moresby-Honiara 26. Port Moresby-Jayapura PX 1 Port Moresby-Madang-Wewak 27. Port Vila-Noumea SB 5 Port Vila-Noumea La Airline codes are listed in Annex 9. Source: World Airways Guide, 1991: Qantas Timetable, 1991, and Ansett Timetable, 1991. - 207 - ANNEX m.7 Table 3: AUmALAN INTERNATioNAL AIR ROUTES, 1991 No. of Flights Origin-Destination Airlines Lg per Week Flight Routings 1. Sydney-Bangkok QF 9 Sydney-Bangkok QF 1 Sydney-Melboume-Hong Kong-Bangkok TG 2 Sydney-Melbourne-Bangkok TG 3 Sydney-Bangkok AZ 2 Sydney-Melboume-Bangkok BA 2 Sydney-Bangkok LH 1 Sydney-Bangkok LH 1 Sydney-Melboume-Bangkok NG I Sydney-Bangkok 2. Sydney-Honolulu AA 4 Sydney-Honolulu CP 2 Sydney-Honolulu CP I Sydney-Nadi-Honolulu Co 2 Sydney-Auckland-Honolulu Co 7 Sydney-Honolulu UA 7 Sydney-Auckland-Honolulu QF 2 Sydney-Nadi-Honolulu QF 7 Sydney-Honolulu 3. Sydney-Auckland QF/NZ 26 Sydney-Auckland cO 5 Sydney-Auckland PH 1 Sydney-Auckland UA 7 Sydney-Auckland 4. Sydney-Hong Kong QF 5 Sydney-Hong Kong QF 2 Sydney-Brisbane-Hong Kong QF 2 Sydney-Melbourne-Hong Kong CX 5 Sydney-Hong Kong CX 2 Sydney-Melbourme-Hong Kong 5. Sydney-Los Angeles QF 7 Sydney-Los Angeles QF 3 Sydney-Honolulu-Los Angeles QF 2 Sydney-Papeete-Los Angeles QF 1 Sydney-Bnsbane-Honolulu-Los Angeles AA 4 Sydney-Honolulu-Los Angeles cO 2 Sydney-Auckland-Honolulu-Los Angeles UA 7 Sydney-Los Angeles NZ 2 Sydney-Auckland-Honolulu-Los Angeles 6. Sydney-Bali QF 1 Sydney-Bali QF 2 Sydney-Melbourne-Bali GA 6 Sydney-Bali 7. Sydney-Tokyo QF 8 Sydney-Tokyo QF 1 Sydney-Perh-Tokyo JL 7 Sydney-Tokyo NH 5 Sydney-Tokyo 8. Sydney-Jakarta QF 2 Sydney-Jakarta GA 6 Sydney-Denpasar-Jakarta UT 1 Sydney-Jakarta -208 - ANNEX Ml.7 Table 3: AuWiAuIAN INTERNAToNAL AIR ROUTES, 1991 cont'd No. of Flights Origin-Destination Airlines la Pat Wee Flight Routings 9. Sydney-Singapore QF 7 Sydney-Singapore QF 7 Sydney-Melboumne-Singapome SQ 7 Sydney-Singor JU 2 Sydney-Meu Sinapr BA 5 Sydney-Sinap~do OA 2 SynyM GF 2 Sydney-Singpr-aai LH 1 Sydney-MeloreSngpr-rnfurt UT 1 Nouame-SydyJaatSiapePra L& Airline codea are listed in Anne 9. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Aanet Timectable, 1991. - 209 - ANNEX M.7 Table 4: INDIAN OCEAN ISLANDS-AR SERVICES, 1991 No. of Flights Origin-Destination Airlines per Week Flight Routings 1. Johannesburg-Mauritius MK 2 Johannesburg-Mauritius 2 Johannesburg-Harare-Mauritius SA 4 Johannesburg-Mauritius 2. London-Mauritius MK 2 Johannesburg-Mauritius BA 2 London-Dubai-Seycheile-Mauritius 3. London-Mahe BA 2 London-Dubai-Mahe-Mauritius HM I London-Paris-Mahe HM 1 London-Frankfurt-Mahe HM 1 London-Rome-Mahe 4. Paris-Mauritius MK 2 Paris-Mauritius MK 1 Paris-Munich-Mauritius AF 2 Paris-Djibouti-Mauritius-R"union AF 2 Sydney-Paris-Nice-Jeddha-Seychelles- Mauritius-Reunion 5. Paris-Reunion AF 8 Paris-Djibouti-Antananarivo (Madagascar)-Rdunion Paris-Djibouti-Seychelles-Reunion- Mauritius Paris-Djibouti-Rdunion Paris-Djibouti-Mauritius-Rdunion Paris-Nairobi-Harare-Rounion OV 7 Paris-Lyon-Reunion Paris-Mauritius-Rdunion 6. Perth-Cocos Islands - 1 Perth-Cocos Islands 7. Perth-Christmas Island - 1 Perth-Cocos Islands-Christmas Island Li Airline codes are listed in Annex 9. /b Air charter service. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Ansott Timetable, 1991. - 210 - ANNEX M.7 Table 5: AuSTRALIAN DoMEsTIc TouRIST RoUTEs, 1991 No. of Flights Origin-Destination Airlines Li per Week Flight Routings 1. Sydney-Gold Coast TN 22 Sydney-Coolangatta AN 22 Sydney-Coolangatta EW 40 Sydney-Coolangatta EW 6 Sydney-Tamwortb-Coolangatta 2. Sydney-Hamilton Island AN 14 Sydney-Hamilton Island EW 12 Sydney-Hamilton Island EW 2 Sydney-Brisbane-Hamilton Island EW S Sydney-Coolangatta-Brsbano- Hamilton Island 3. Sydney-Alice Springs TN 6 Sydney-Alice Springs AN 7 Sydney-Alice Springs T 1 Sydney-Melbourne-Alice Springs EY\ 2 Sydney-Ayers Rock-Alice Springs 4. Sydney-LAuncoston TN 2 Sydney-Launceston TN 11 Sydney-Melbourne-Launceston AN 28 Sydney-Melboume-Launceston 5. Melbourne-Gold Coast AN 11 Melboume-Coolangatta AN 6 Melboume-Sydney-Coolangatta TN 9 Melbourne-Coolangatta TN 8 Melboume-Sydney-Coolangatta EW 7 Melboume-Coolangatta EW 7 Melboume-Sydney-Coolangatta 6. Melbourne-Cairns AN 7 Melboume-Sydney-Caims AN 8 Melbourne-Sydney-Brisbane-Cairns AN 1 Melboume-Brisbane-Cairns TN 6 Melbourne-Sydney-Brisbane-Cairns TN 7 Melboume-Brisbane-Cairns TN 1 Melboume-Brisbane-Townsville-Caims 7. Brisbane-Townsville OF 7 Brisbane-Townsville (with various stops) AN 21 Brisbane-Townsville AN 14 Brisbane-Gladstone-Rockhampton- Mackay-Townsville TN 19 Brisbane-Townsville YC 1 Brisbane-Townsville 8. Brisbane-Mackay OF 6 Brisbane-Mackay (with various stops) AN 14 Brisbane-Gladstone-Rockhampton- Mackay AN 14 Brisbane-Rockhampton-Mackay AN 6 Brisbane-Mackay TN 8 Brisbane-Mackay TN 7 Brisbane-Rockhampton-Macksy YC 7 Brisbane-4ladstone-Rockhampton- Mackay -211- ANNE( M.7 Table 5: Australian Domestic Tourist Routes, 1991 (cont.) No. of Flights Origin-Destination Airlines per Week Flight Routings 9. Brisbane-Rockhampton OF 11 Brisbaa-Rockhampton OF 5 Brisbane-Bundaberg-Gladstone- Rockampton AN 3t Briabn-Rockhampton AN 14 Brisbmn.-Oladstone-Rockhampton IN 17 BrisbaeRockhampton YC 7 Brisbano-Gladstone-Rockhampton 10. Adelaide-Alice Springs AN 7 Adeaid-Alco Sprinp TN 7 Adoide-Alice Springs 11. Alaidoe-Darwin TN 7 Adoeide-Alice Sprin-Darwin AN 1 Adelaide-Darwin AN 7 Alaidoe-Alie Sprin-Darwin MH 1 Adolide-Darwin-Kuala Luipur .i Airline codes are listed in Annex 9. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Ansett Timetable, 1991. -212 - ANNEX m.7 Table 6: WESTERN AUSTRALIA DoMEmC RourEs, 199. No. of Flights Origin-Destination Airlines /a per Week Flight Routings 1. Perth-Albany YT 12 Perth-Albany 2. Perth-Broomo MV 5 Perth-Broome MV 8 Perth-Port Hedland-Broomo MV 1 Perth-Newman-Port Hedland-Broome 3. Peth-Cmnarvon MV 3 Perth-Csrarvan MV 4 Perth-Geraldton-Cawnarvon 4. Perth-Derby MV 3 Perth-Derby MV 2 Perth-Brooms-Derby MV 8 Perth-Port Hedland-Brooms-Derby MV 1 Perth-Nowman-Port Hedland-Brooms- Derby 5. Perth-Geraldton MV 8 Perth-Geraldton YT 10 Perth-Geraldton 6. Perth-Kalgoolie MV 15 Perth-Kalgoolio YT 13 Perth-Kalgoolis 7. Perth-Kaffatha MV 18 Perth-Karratba MV 2 Perth-Paraburdoo-Karratha MV 2 Perth-Geraldton-Canarvon-Learmonth- Karratha 8. Perth-Kununurra MV 4 Perth-Kununurra MV 3 Perth-Derby-Kununurra MV 2 Perth-Broome-Derby-Kmunuurr MV 2 Perth-Newman-Port Headland-Derby- Kununurra /a Airline codes are listed in Annex 9. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Ansett Timetable, 1991. - 213 - ANNEX m.7 Table 7: NEw ZEALAND DoMEsTiC ROUrES, 1991 No. of Flights Origin-Destination Airlines /a per Week Flight Routings 1. Auckland-Wellington ZQ 76 Auckland-Wellington ZQ 7 Auckland-Palmerston-Wellington NZ 77 Auckland-Wellington PG 7 Auckland-Wanganui-Wellington 2. Auckland-Christchurch NZ 64 Auckland-Christchurch ZQ 55 Auckland-Christchurch ZQ 10 Auckland-Wdllington-Christchurch NM 7 Auckland-Rotorua-Chistchurch 3. Auckland-Dunedin NZ 26 Auckland-Cbristchurch-Dunedin NZ 7 Awckland-Wellington-Dunedin ZQ 13 Auckland-istchurch-Dunedin ZQ 5 Auckland-Wellington-Dunedin 4. Auckland-Hamilton EX 19 Auckland-Hamilton 5 Auckland-Tauranga-Hamilton 5. Auckland-Rotorua NM 20 Auckland-Rotorua 6. Wellington-Queenstown ZQ 7 Wellington-Christchurch-Queenstown NM 7 Wellington-Christchurch-Mount Cook- Queenstown 7. Wellington-Invercargill NZ 5 Wellington-Christchurch-Invercargill ZQ 5 Auckland-Wellington-Dunedin-Invercargill L Airline codes are listed in Annex 9. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Ansett Timetable, 1991. - 214 - ANNEX M.7 Table 8: PAPUA NEW GUINEA-DOMESTIC AIR RoUTES, 1991 No. of Flights Origin-Destination Airlines La per Week Flight Routings 1. Port Moresby-Lae PX 36 Port Moresby-Lae GV 6 Port Moresby-Lae GV 6 Port Moresby-Wau Papua-Bulolo-Lae 2. Port Moresby-Mount Hagen PX 13 Port Moresby-Mount Hagen PX 1 Port Moresby-Lae-Mount Hagen PX 1 Port Moresby-Lae-4oroka-Mount Hagen GV 1 Port Moresby-Mount Hagen GV 4 Port Moresby-Mendi-Tari-Mount Hagen GV 6 Port Moresby-Laoe-Goroka-Mount Hagen GV 1 Port Moresby-Kundiawa-Mount Hagen Moresby-Goroka PX 9 Port Moresby-Goroka PX 3 Port Moresby-Lae-Goroka GV 6 Port Moresby-Lae-Goroka GV 2 Port Morosby-Kundiawa-Goroka 4. Port Moresby-Madang PX 4 Port Moresby-Madang PX 2 Port Moresby-Mount Hagezi-Madang PX 9 Port Moresby-Lae-Madang PX 2 Port Moresby-Goroka-Madang GV 2 Port Moresby-Kundiawa-Madang S. Port Moresby-Wewak PX 3 Port Moresby-Madang-Wewak PX 1 Port Moresby-Mount Hagen-Wewak PX 5 Port Moresby-Lae-Madang-Wewak GV 1 Port Moresby-Mount Hagen-Wewak 6. Port Moresby-Wau Papua GV 6 Port Moresby-Wau Papua 7. Port Moresby-Tari PX 2 Port Moresby-Tari PX 2 Port Moresby-Mendi-Tari GV 4 Port Moresby-Mendi-Tari 8. Port Moresby-Daru PX 2 Port Moresby-Daru GV 8 Port Moresby-Daru /a Airline codes are listed in Annex 9. Source: World Airways Guide, 1991; Qantas Timetable, 1991, and Ansett Timetable, 1991. - 215 - ANNEX Mi8 PACIFIC ISLANDS TRANSPORT SECTOR STUDY MAPS OF AIR SERVICES FOR PACiFC ISLANDS, COMPARATOR COUNTRIES AND PACiFC RIM iLUnlst PACIFIC ISLAUDS-AVITIOU OIGII-DSTIUC MIM PACIIC RIN *~~~ - - _- *1 . , e .I _ - _, - _ - . - *4n t.u .ur * IW* ?I= -~~~~~~~E -- FEDERATED STATES .] ,,I C1A *U E _ OF MICRONESIA :! *. XM 1 \ .- * v - ,, oSX o_ .* '8 ^,-o * ~~~~~~~~~W (NJ,, *x - ' KM j x_ ... .Mal *. ..Ion r NAU* ; .erna E TU-rn--* Mmis TOKELAU 2- _ 1^ .; i*xb . . F* 1!J / . u'* ' I1LLA_I S] TUVAUATI'ini mu L* OENL F.E.C. I t~~~ i~~~ib1 I ~qq~ SSA U1DMIJ 1.*... -~~~~~~~~~~~~~~I * A .S W_ . . , NtXui& AOA NE U HE~~~E AUSTnALtA I SOUTH ~~~~~~~~~~PACIFIC OCEAN LEGEND - 1110AUOKL411) Nornieuslad Pacific island CountlIesm..... *r m masL Compeiato.r CountS..s In PacIicl Itoglon ssmsmin raw ILmzr 2: PACIFIC ISLAUDS-AYITIMI 031013-4MI IC FN PACIFIC RIM 146 % _V,~tsr a 1A'iA mv C_"5^}~~~~~~~~~~~~~~ ' 'm FEDERATED STATES ... / [I e i3 v OF MICRONESIA :- / I m _n 3 a olegAls ~ ~ Of' /* 3_ A )_ NAURU' o * VAWFAI .O ~~~~~~~~~v~~~~~~~~ 0s* r N A U A .f _ _ _ __._ _ _ _ _ _ _ _ _ _ q ',SY , SOLOMON Ul * -.ISLANDS | TUVI U.. - i / TOKELAtU Go I:I -h : '- Funat / ;,~N ;H i I WIEST * .. OItJSN FIJI - ISAMOAMO OK & CORAI~__ * P0IYIS AS 1 SO AMFnICA _ :PACIFIC OCEAN FVANUATti-z C - - Kr I4. . -.1 __St AMOS- I EGEND | _ FIW8D | | Nomineled Pacilic I-land Countries PROU PACIFIC OCEAN O__ oContie NIn ceiRE 0 Comnp-rslor Counbtries in P-citic RegUlon - 219 _ iiIII,t 0 9| 0 l . . § ffi5 j3: , ,] .4 .~~~~~~~ ' ,I. .' a.. l~~~~~ 0) FLuira 4 PACIUC ISLAWS-IUTA-RZOM AVIhfl OEIGZU-NSTIEfKS I..' * I.O.U 1W * IWw ~ ~ ~ ~ ~ ~ .. obuIISA.) ' 4 u~uo_,auauwu (~] Eum ,o.UWA FEDERATED STATES . . .B u 'opFN RD ma, rom ou"a * . * OF MICRONESIA *. . 1 WAN TO 1A* M2] BIA0 1WALLISMIA .. " . , , ,.17 w'. wowm [E] M rOwtOn *~ F "~ .. t ^ajuro . E3 w w [S lonTV TOIISHIa . * .... /. @gX° %rolwto 69 o:i 10 OUl ISV 10 PW kAWI10JVP NAURU . _ SDUEA t jiSOLOMON . 2 .1, - ^ | i t *ISLA TUVA * T9 EAU I % Wallis ~~~~~~~~~~~~FRENCIH -*'POt'NESIA SAMOA COOK SEA NIUEQW1MIY AUSTOIALIA SOUTH PACIFIC OCEAN 'IO E lorE _ flo*, _ TW .. LEGEND LEGEND FAMNAD Nominated P-cific Island Countries FUOUNOUMIA .....-1111111 O MOAEMY O Compmortor Couctiefs In Pacific Regilon - 221 - Zlr5U t AUSTRALIAN 13TnU3TIOUAL AIR ROUTK3-OUCGIS-DUSTIEATIOS x ' ~~~~~~~~~~~~~~~.w ...: Mmm. sYuu ......... M^ *vm.o To ,miu gym.., TowalA 1101IMAT90 $70*9 yUUy TO Mm 1AU 1Ym.Y TO TUiV f3 VONfy TO ACLA@ $b@Ow 'eA4JUU *~~~~~~~~WO~VT N"nw tfv Oslepr O %-" "- e_"* * , SEA ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ CA a. Sx Y-m.m"o'1 ZIEL I_.DIAN ,, OCIAN ISLAND IOTH PACIFICTOCA WI di eI d .JOWUJ TO M__ _A_ _ PA= TO.AI rr ONCONT@UMALMA UT TO CO LO CM Lodo To Ofam"s m Pop" TO C@O6U NA ~~ ija ~~ 47PM OTO WAmiLLSS INDIAN a 5/ ~~~~~~~~OCEAN -. Mm-~~~~~~~~~~~~~~~~~~~~AUTAI -U,. - ~~~~~~~~m. ~ ~ td I1ma s.1 USTU AUSTRALIA DOHSTIC AIR iumr :s s T TL1ST DUISTIC AlR RUIITS-C1IG, IS-1E1TI S RCDTNTIS It5-E 1201 E125 E 130-E tlO 135E If E t5E 150E IlsMSUW 200J ftm2OSm I 10So m Ima _EJ m _I.MCaLDarwin ( N 12] paIoemmw 12]..;nump9 1E __ w-__________________ (2 uwmseowmwou r)] \m i@nf I m_"1 mo < mer m" OD Is INDMN 1 KD nunurra m2 mueu m) i sCa's A \ D S m " C!B IDIn wCaine Broom-Ce , / T __D ED w__ mmown,v-lle INDAN f e / (2]m maA D w. | Aag °- room* -Mackay 20'S Alice Iand 1 l 0S s pring s Roc_ lamp oon Islan fIJ airralhe ~WESTERN 2' (j AUSTRALIA n1 olt cornervon ~~~~AUSTRALIA 0mm idt Adelaide r Sde Kelooolle ~~~~~~~Melbourne3S IiSE ~~~~~~~~~~~~~~~~~~~TASMANIAI~Ja5Launceston I 9511!120'E I 25'E 130'E _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ILlS 9sU ZILAN DCgSTIC AIR ROUTES [j=_l_ 10 PAPUA @flu& DOMSTIC AIR OuG1-IKST4ImATIoa 33s 5. TIms Io __=1 2-S t Wewak PACIFIC OCEAN \' . I ~~~~PAPUA AuckQe Madang Hamilton 9@\ } Taro S - M So~~ Rowua | N||NEW GUINEA TA4SMAN NE IDairu SEA Pr o Wellingion ARAFURA SEA PACIFIC Am r sChfIStchutch 7EALAND AT OCEAN |1!J I Ms w 0 uam$" To SueDE --f.#.Owon*t / OAWAIDMMClram >wowan"00o mTw"O CD &uMml*m / ram mo.w, wOes Snv..oSgOWg /j.,MMne C ram y 0 TAf A___________________________________________In__________ _ E J tam HEm,im T o -A ".0~~~~~~~~~~~~~~~~~~~~~~~~~~~C mt e"se_ To VAN .~~~~~~~~~~~~~~O -OFR E1OD°" - 224 - ANNEX MI.9 PACIFIC AIRLNS TRANSPORT SECTOR SURVEY ALIE CODES Code Airline Code Airline AA American Airlines OF Sunatat Airline. AP Air Fance ON Air Nauru AN Ansett Australia OV Aeromaritime AZ Alitalia PC Fiji Air BA British Airways PG Air Ndon CO Co50 iental Airlines PH Polynesian Airlines CP Canada Airlines PR Philippines Airlines CW Air Mars Islands PX Air Niugini CX Cathay Pacific Airways QF Qantag Airways EW Natwet Airlines SA South African Airways EX Eagle Airways SB Air Caledonia Intoational FJ Air Pacific International SQ Singapore Airlines GA Garuda Idoneia TG Thai Airways Intemational GF Gulf Air TN Australian Airlines GV Taair TS Samoan Aviation HA Hawaii Airlines UA United Airlines IE Solomon Islands Airlines UT Union de Transports Aeriens (UTA French JL Japan Airlines Airlines) LH Lufthas German Airlines VK Air Tungaru Corporation MK Air Mauritius WR Friendly Islands Airways Limited MV Ansett-Western Australia YC Flightwest Airlines NF Air Vanuatu YT Skywest Airlines NO Lauda Air ZQ Ansett-New Zealand NH All Nippon Airways NM Mount Cook Airlines NZ Air New Zealand Source: World Airways Guide, 1991. - 225 - ANNEX M.10 PACIFIC ISLANDS TRANSPORT SECTOR SURVEY RESULTS OF STATICAL ANALYSIS OF Am FAREs FOR PMCS AND COMPARATORS This annex presents the results of the log linear regression models estimates for economyc and excursion fares. Table 1: PACIFIC AND COMpARATOR REGIONS-R REGSIoN REs;ULTs FoR EROoY Am FARES Bas MNdd '9, = 3.35 * DI' * V4m (45.92) (-0.16) Test 1 = 4.82 * DO1)6 * V0 * (eN)m * (e1)o * (O')14 * (C")7 * (eAI)p17 * (eWA)4M e * (eZ)A * N (26.27) (-0.46) (0.66) (0.66) (1.20) (-3.27) (1.38) (-0.58) (-0.73) (0.01) Test 2 = 3.84 * DI'9 * V-0 * (eN)oM * (e")'S * (e" * (eM'4|5 (37.72) (-0.21) (0.90) (0.36) (1.00) (-2.86) where 9 = estimae return excursion air fare D = distance (shortest operated nleage) V = volume (seat capacity per amum) N = dummy variable for the nominat Pacific Islands (1,0) P = dummy variable for all the Pacific Islands (1,0) I = dummy variable for the comparator India Oceaen Islands (1,0) IP = dummy variable for the Intra-Pacific Island routes (1,0) Al = dummy variable for the Australian international routes (1,0) WA = dummy variable for the Western Australia domestic routes (1,0) ADT =dummy variable for the Ausralian domestic tourist outes (1,0) NZ = dummy variable for the New Zealand domestic outes (1,0) Notes: (I) t-statistics are in parentheses. (2) I t variable is significant at 0.05 level when t-statistic is greaer than + 1.96 or less than -1.96. C) Table 2: PAcnIC AND COMPARAToR REGIONS-REGREwON RESULTS FoR EXCusm AmR FARES Be Mode '9' = 7.13 * D * VJ07 (35.07) (-5.56) Test I 1' = 6.81 * DO1 * V407* (eN)-6 * * (el))A (e')` e (e)4-15 * (ewA)422 * (eC)4 * (21.56) (-3.64) (-1.10) (-0.55) (-0.66) (0.84) (-1.10) (-2.04) (-0.19) (.0.89) Test 2 '9' = 6.93 * D * , V4 O (eN)4 * (eI)s5 * (e)' * (cI (28.31) (-3.74) (-1.10) (0.72) (0.37) (0.65) whee 1' = estimated tum excursion air fare D = distanoe (shortest operated mileage) V = volume (satpcity per annum) N = dummy variable for the nominatd Pacific Islands (1,0) P = dummy variable for all the Pacific Islands (1,0) I = dummy variable for th comparator Indan Ocea Islands (1,0) IP = dummy varial for the Intra-Pacific Island rut (1,0) AI = dummy vaiable for the Austalia inteaonal mutes (1,0) WA = dummy variable for the Westan Ausralis domesfic utes (1,0) ADT = dummy vaiable for the Austnli domestic tou outes (1,0) NZ = dummy variable for the New Zaland domestic rutes (1,0) Nok&: (1) t-sueistics an in parenthes. (2) Indpendent variable is significant at 0.05 level when I-statistic is greater tha + 1.96 or less thn -1.96. - 228 - LoGLUA UGUZSSr IOU SULTS-UCOUOT1*1 BASE MOD-EL| Regrpswuo Output: ConJUuY 1.208450 A - 3.35 SW Err ofY Eat 0.IU147 R Squared 0.951303 No. of tObuivaiou 111 Dope. of F eo103 X CoeffihiA(a) 0.807369 -0.00187 SW Errol Cod. 0.017580 0.011993 T-emlghgg 45.92 -0.16 n Dis. LA VOd. Rapomlolsa Output: C4mwi" 1.573052 A- 4.32 SW EJs of Y Est 0.177793 R Squared 0.959736 No. of Obervuhoms III Degew of F 100 X Coesinsidms) 0.756533 -0.00773 0.033553 0.066J23 0.143360 -0.17-423 0.169773 -0.053110 -0.066456 0.0006573 Sad EwdCo f. 0.0C2300 0.017016 0.050U3 0.100997 0.119606 0.053348 0.1U351 0.0916383 0.091535J 0.094J29 T-mautido 26.27 -0.46 0.66 0.66 1.20 -3.27 1.33 -0.58 -0.73 0.01 AuIL Nominsted Indin hum- AMn. W.A. Domes N.Z. PwAic Pacific 0cm Pacific lcrl Domaic Towin Domeai ad, wand Iand lad & u in Dis. In Vol. dummy du my dummy dummy dummy dummy dummy damy (1.0) (1.0) (1.0) (;.0) (1.0) (1.0) (1.0) (1.0) Repin_ Output: Coaoata 1.345798 Au 3.34 Skd Ea of Y Est 0.180470 R 3quae 0.9535 No. of CObewvo III Depees of From 104 X Codflsilu) 0.739023 -0.00334 0.045605 0.024315 0.0I7017 -0.14662 Sid En of Cof. 0.00920 0.015921 0.050613 0.066730 0.0t7301 0.051184 T-aatudcs 37.72 -0.21 0.90 0.36 1.00 -2.36 No _mn Iin Irs- PaslAs Pa O_es Pa?fAg lmd Wand Ws1d 1 and In Dim. In Vol. dumy dummy dummy dumm (1.0) (1.0) (1.0I (1.0) - 229 - IZL.1Q LOCLIEZAR UGRISSIOI 3NSUTS-TICUSI WARNS ZMILMMRI Regrsso. Output: Coegan 1.964957 A- 7.13 Sad En of Y Es 0.197270 R Sqgd 0.920501 No. of Obsaso III Dgrs. of Fds. 101 X CosfFmimaws) 0.63706 -0.06994 Si an of Cos. 0.0194 0.012576 T _~miss 35.07 -5.56 laLDimt. inVIl. TEST i -' R Ouqm CAMoo_ 1.918014 Aa 6.31 Sid Er Y Em 0.196227 R Squared 0.927166 No. of Obamaioos 111 Depe of FrIedom 100 X Cos3isazs) 0.697696 -0.06777 -0.06120 -0.05931 -0.08606 0.049462 -0.14514 -0.206334 -0.019197 -0.093214 SidErofColf. 0.032353 0.012839 0.055490 0.107729 0.130112 0.058912 0.132467 0.1012217 0.100917 0.1042328 T-SUalues 21.56 -3.64 -1.10 -0.55 -0.66 0.84 -1.10 -2.04 0.19 O.89 Nomiated Indian Inra- Aust. W.A. DoMc N.Z. Pific Pfic Oma Pacifi 1.i'1 Domestic Touri Domeic lalgkd Isla Isld sland routes rout" rO rotes In Dist. IA Vol. dummy dummy dummy dummy dummy dummy dumy dummy (1.0) (1.0) (1.0) (1.0) (1.0) (1.0) (1.0) (1.0) TEST 2 Repsusi Ouu: COD UUG 1.935354 A 6.93 S4d Err of Y Eit 0.198149 R Sqiured 0.922762 No. of ObaertCnU 111 Dsp. of Fepiom 104 X Coaflmcaln) 0.679544 -0.06492 -0.06106 0.0511J9 0.035011 0.036990 Sid Er of Co.f. 0.024007 0.017374 0.055712 0.072323 0.095457 0.056740 T-amadtle 23.31 -3.74 -1.10 0.72 0.37 0.65 Noriased Indian mUm- PAcfc pael Oma Pc_ ac _and Wand Iand Wdod In Dit. La Vol. dummy dmy dummy dummy (1.0) (1.0) (1.0) (1.0) it-I AIR, SuvCi ROUTE UrIOIKS-?ACIVJC ISLA UM PACIFIC IDS. 1991 Is WHng Kong \ M0anlai _Guam \ Singapore &~~~~~nimo \C ^uli5 p o~~~~~~~~~~~~~~~~~~~~~~apeet 1 3 ~~~~~~~~~~~~~~~~~~~~Ppeete} a n~gs Rarotonga -MSClbourn Pacific Member Country, International Airport La Routes shown focus on Pacific ilember Countries# not all air service routes for the region are shpwn. - 231- ANNEX IV2 PACIFIC ISLANDS TRANSPORT SECTOR STUDY BROADER OBJEC'ES AND AVATION COOPERATION Multiple Objectives. Aviation is seen as by governments, it is best encouraged in other linked to other government objectives, such as ways. greater tourism and employment opportunities. Thus, some cooperative ventures which lower Airline objectives can and should be kept airlin, costs may be seen as lessening home simple. If airlines are required to concentrate on employment and tourism, and therefore not in commercial performance, this will lead to the interests of particular countries. Airlines airlines minimizing costs and serving all markets which are set broad and possibly conflicting that are commercially viable. If an airline objectives are likely to reject cooperative possesses and uses market power, it can be ventures because they are seen as being contrary regulated (at some cost in terms of incentives to some of these broad objectives. and efficiency). At the aviation negotiation level, airline efficiency, profits, and tourism, The problem is one of recognizing non- need to be taken into account. However, in aviation objectives and of creating a workable general, objectives such as employment and structure to facilitate pursuit of several tourism are best handled separately, and objectives. This requires specifying how airlines especially separate from the controls and are to operate, what constraints they are to be objectives set for airlines. subject to, and what incentives they are to be given. For countries, it is also important to Well Specified Objectives and determine what factors are relevant in aviation Incentives. For an airline (as for other regulation, their relative importance, and how enterprises), simple objectives and constraints broad aviation negotiations can be, in terms of produce the best performance. If commercial what rights and concessions can be traded.' performanice and profit is the objective, and airline management is given a clear incentive to Treating Separate Objectives advance it, the airline will try to minimize costs Separately. A practical approach of addressing and serve all viable markets.2 There is an the several key objectives of the South Pacific additional reason wihy special attention should be countries is to treat the differing objectives given to getting the internal incentives for independently. Thus, if underemployment is a efficiency right in the airlines of the South problem, an airline should not be expected to Pacific. This is because the external pressures, have any preference for local, rather than through competition, are not strong. Many of foreign, employees. Airlir.es should be the routes which these airlines fly are thin, and instructed to concentrate on commercial have only one or two operators (who may have performance; underemployment is best handled commercial agreements). The pressure of by addressing it directly. Likewise, airlines competition is not strong, and the airlines are should not be expected to be vehicles for not under strong pressure to keep costs at the encouraging tourism. If a route brings tourists, minimum feasible level consistent with the but is losing money, it should be dropped. If standard of service offered. In such an environ- tourism is deemed to warrant special treatment ment, it is necessary to rely more on internal - 232 - ANNEX IV.2 incentives to keep costs low,3 Of course, there Less home employment by an airline could mean are difficulties in corporatizing public lower costs, fares, and ultimately more tourism. enterprises-in setting proflt objectives for them, In establishing managerial incentives tied to Higher profits mean more revenue to the profit, and in eliminating political intervention, government, and thus lower taxes or greater Such firms may make decisions, such as to spending elsewhere-this could have an impact reduce home employment and cooperate with on employment. More purchases overseas by an foreign firms to reduce costs w'iich will be airline puts downward pressure on the country's politically unpopular-it is difficult for exchange rate, and thus it improves prospects government to resist the temptation to intervene for import competing and export industries. and override the enterprise. Quite apart from the fact that airlines tend to employ skilled people, who are more like!y to AMrlnes In Support of Non-Aviation be in excess demand rather than excess supply, Objedives. Notwithstanding the difficulties of employment policies at the airline level will be multiple objctives, they need not be ineffective ways of addressing a country's incompatible. Governments may advance overall employment objectives. specific objectives which are best handled at the airline level, for example, airlines may be an Tourism. Encouragement of tourism, efficient way of providing training in certain partly for its employment effects, is a common skills. If so, a government can offer training objective of governments. This is true for most, subsidies (to airlines and others). Airlines are in if not all, South Pacific countries. It can be a good position to market their countries as addressed directly, through tourism promotion, tourism destinations-tourism authorities can through the tax treatment of the tourism form joint ventures with airlines to do this. industry, and through specific grants or Govermnent may want certain commercially subsidies. It can also be effected through non-viable services provided for social lowering the cost of tourism infrastructure, reasons-it can offer the airline specific including perhaps aviation infrastructure, such as subsidies. Specific, targeted, measures, within airports. As noted above, aviation negotiations the context of a commercially oriented airline may need to take into account tourism aspects, can be used effectively to serve specific non- since the structure of traffic rights may have an aviation objectives; under these conditions the impact on tourism flow. There is no need to airline is not being expected to advance a range involve airlines directly in pursuing tourism of vague and conflicting objectives. objectives. Treatment of Other Objectives. Other In reality, airlines of many countries, and objectives can be handled separately. If there is especially those in the developing world, are a problem of unemployment or often regarded as means of encouraging tourism. underemployment, it is usually an economy- Clearly, air services are essential for tourism to wide, or regional, problem. There is little to be island economies. However, providing air said for expecting particular industries to face services does not guarantee tourists in sufficient higher costs through increasing their numbers to warrant the costs of an air service. employment beyond that which they we "d It is possible to create a framework in which otherwise choose on commercial grounds. corporatized public or private firms are free to Indeed, it does not follow that when a particular pursue profit objectives, but in which firm of industry employs more people, that governments use airlines as instruments to attract overall employment will rise. (This is because tourists. of indirect, and general equilibrium effects.) - 233 ANNEX IV2 One way is for the government to offer Low Air Fares and Other Objectives. support for subsidiaries (in principle, to any Low air transport costs are likely to be in the airline) to operate specific commercially non- interest of the region as a whole. This is not viable services that are regarded as important for self-evident, granted that other objectlves exist. tourism development purposes. If the subsidy is However, looking at the two other objectives sufficient, airlines will operate the route, and the which could be of importance, employment and cost of advancing the tourism objective will be tourism, it is un'ikely that these are harmed by explicit. Alternatively, tourism development can efficient aviation. While more direct be set as an objective for an airline, in addition employment in aviation may be improved at the to profit, and the airline rewarded for its expense of higher air transport costs, it need not performance in terms of this objective. This follow that overall employment is increased. would involve (a) determining the value of However, aviation costs within the region can additional tourism, and offering the airline a result In lower fares, so that the region as a subsidy (for example, $40 for each additional whole can act more tourist and more tourist), (b) estimating the contribution t'ie employment. Thus, the region as a whole can airline makes to tourism, and (c) ensuring that gain through having more efficient aviation. management incentives are linked to overall profit, including payments for (or numbers of) Negotiations to Secure Not Gains. tourists. This would be a complex procedure, Individual countries may lose under certain and it would not be easy to implement. arrangements, even though It should be possible However, it would be superior to the present ta compensate them. For example, low air fares situation, whereby airlines are often expected to could mean that tourists go to other countries in encourage tourism and make a profit, but they the region, and a particular country loses out. are given no guidance as to how to balance any Tbis is best handled, not by restrictive regulation conflict between these objectives. and high cost air transport, but by the country which loses negotiating for a share of the other Where airlines are given simple countries gains (for example, Jirough seat sales objectives, such as profit, they will seek to to its airline rather than granting 5th freedom operate cost efficiently. In doing this they will rights to other countries airlines). The country enter those cooperative arrangements which are will lose tourism, but gain other benefits.' likely to lead to cost reduction-however, they What is needed, most fundamentally, is will not pursue cooperation for its own sake. If knowledge, on the part of the countries all airlines operate in this way, costs of air involved, of the likely scale of gains and losses transport in the region will be as low as feasible. from a particular proposal, and a willingness to If some countries and airlines pursue other negotiate a settlement acceptable to all. In objectives and operate differently, cost of air practice, this will not always be the case. travel in the region will be higher, but the However, where it is, it is possible to have an remainder of the countries and airlines will be efficient air transport system, with independent acting so as to make air transport as cost but cooperative airlines, and no country lose out. efficient as possible, granted that cooperation is incomplete. - 234 - ANNEX IV2 Endnotes At a theoretical level, it is possible to suggest a solution based upon the concept of a country "social welfare function" determined by government which reflects trade-offs (shadow prices) between various objectives. Airlines would be instructed or induced to evaluate proposals on the basis of the appropriate shadow prices, not market prices; for example, airlines with commercial obligdtions would face market prices which are made equal to shadow prices by taxes and subsidies. Thus, if shadow wages were less than market wages, employment wou!d be subsidized. This solution is sketched e-t as a benchmark, not because it is workable. Shadow prices are sometimes used in the evaluation of projects, but broad adjustments of market prices by tax/subsidy mechanisms is rare, except in simple situations. Even countries with advanced price mechanisms and well developed information flows do not attempt very detailed corrections of market prices. Rather, they identify the imperfections in their economies, and usually address them one by one, rather than all at the same time. This approach is theoretically imperfect, but it is practical. (See Broadway and Bruce (1984). 2. Publicly owned airlines can be given corporate forms, and managers can be set incentives to seek profit and behave commercially; i.e., they can be corporatized. Where this happens in a competitive market, profit will be consistent with overall efficient market performance. Problems do arise if market power is significant, and regulation of firm behavior may be desirable. 3. There are alternatives, but they have not proved very satisfactory. Public enterprises around the world are often given broad objectives, rarely with explicit weighting, and it is impossible 'o either measure performance or to gear incentives to performance. The result is that none of the objectives is met. The use of public enterprises as broad political instruments (with vague and conflicting objectives, coupled with inappropriate incentives) is the major reason why cost efriciency of these enterprises is often so low. 4. If a country loses, but other countries gain more, yet that country's approval is required, then other countries should be prepared to grant concessions to gain its approval.