Transport development in southern China

WDP 151 FILE COPY T jJ 151 |=| World Bank Discussion Papers China and Mongolia Department Series Transp ort Developnment in Southern China Clell G. Harral, editor Peter Cook and Edward Holland, principal contributors  Receint World Bank Discussion Papers No. 93 Assessment of the Private Sector: A Case Study and Its Methodological Implications. Samuel Paul No. 94 Reaching the Poor through Rural Public Employment: A Survey of Theory and Evidence. Martin Ravallion No. 95 Education and Development: Evidencefor New Priorities. Wadi D. Haddad and others No. 96 Household Food Security and the Role of Women. J. Price Gitfinger and others No. 97 Problems of Developing Countries in the 1990s. Volume 1: General Topics. F. Desmond McCarthy, editor No. 98 Problems of Developing Countries in the 1990s. Volume ii: Country Studies. F. Desmond McCarthy, editor No. 99 Public Sector Management Issues in Structural Adjustment Lending. Barbara Nunberg No. 100 The European Communities' Single Market: The Challenge of 1992for Sub-Saharan Africa. Alfred Tovias No. 101 International Migration and Development in Sub-Saharan Africa. Volume I: Overview. Sharon Stanton Russell, Karen Jacobsen, and William Deane Stanley No. 102 International Migration and EDevelopment in Sub-Saharan Africa. Volume II: Country Analyses. Sharon Stanton Russell, Karen Jacobsen, and William Deane Stanley No. 103 Agricultural Extensionfor Women Farmers in Africa. Katrine Saito and C. Jean Weidemann No. 104 Enterprise Reform and Privitization in Socialist Economies. Barbara Lee and John Nellis No. 105 Redefining the Role of Government in Agriculturefor the 1990s. Odin Knudsen, John Nash, and others No. 106 Social Spending in Latin America: The Story of the 1980s. Margaret E. Grosh No. 107 Kenya at the Demographic Tuming Point? Hypotheses and a Proposed Research Agenda. Allen C. Kelley and Charles E. Nobbe No. 108 Debt Management Systems. Debt and International Finance Division No. 109 Indian Women: Their Health and Economic Productivity. Meera Chatterjee No. 110 Social Security in Latin America: Issues and Optionsfor the World Bank. William McGreevey No. 111 Household Consequences of High Fertility in Pakistan. Susan Cochrane, Valerie Kozel, and Harold Alderrnan No. 112 Strengthening Protection of Intellectual Property in Developing Countries: A Survey of the Literature. Wolfgang Siebeck, editor, with Robert E. Evenson, William Lesser, and Carlos A. Primo Braga No. 113 World Bank Lendingfor Small and Medium Enterprises. Leila Webster No. 114 Using Knowledgefrom Social Science in Development Projects. Michael M. Cernea No. 115 Designing Major Policy Reform: Lessonsfrom the Transport Sector. Ian G. Heggie No. 116 Women's Work, Education, and Family Welfare in Peru. Barbara K. Herz and Shahidur R. Khandker, editors No. 117 Developing Financial Institutionrsfor the Poor and Reducing Barriers to Accessfor Women. Sharon L. Holt and Helena Ribe No. 118 Improving the Pe!formance of Soviet Enterprises. John Nellis No. 119 Public Enterprise Reform: Lessonsfrom the Past and Issuesfor the Future. Ahmed Galal No. 120 The Information Technology Revolution and Economic Development. Nagy K. Hanna No. 121 Promoting Rural Cooperatives in Developing Countries: The Case of Sub-Saharan Africa. Avishay Braverman, J. Luis Guasch, Monika Huppi, and Lorenz Pohlmeier (Continued on the inside back cover.)  Transp ort Development in Southern China  China and Mongolia Department Series Patterns of Diret Foreign Investment in China, Zafar Shah Khan, September 1991 Coal Pricing in China: Issues and Refortn Strategies, Yves Albouy, October 1991 Reforming Prices: The Experience of China, Hungary, and Poland, Anand Rajaram, January 1992 Developing Mongolia, Shahid Yusuf an, Shahidjaved Burki, January 1992 Sino-Japanese Economic Relationships: Trade, Direct Investment, and Future Strategy, Shuichi Ono, January 1992 The Sectoral Foundations of China's Development. Shahid javed Burki and Shahid Yusuf, editors, January 1992  151~ - = World Bank Discussion Papers China and Mongolia Department Series Transport Development in Southern China Clell G. Harral, editor Peter Cook and Edward Holland, principal contributors The World Bank Washington, D.C.  Copyright C 1992 The Intemational Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of Aimerica First printing February 1992 Discussion Papers present results of country analysis or research that is circulated to encourage discussion and comment within the development community. To present these results with the least possible delay, the typescript of this paper has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. Any mapc that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation. The material in this publication is copyrighted. Requests for permnission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its workz and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to photocopy portions for classroom use is not required, though notification of such use having been made will be appreciated. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from the Distribution Unit, Office of the Publisher, Department F, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'1ena, 75116 Paris, France. ISSN: 0259-21 OX At the time this paper was written, Clel1 G. Harral was principal transport economist in the World Bank's China Department, Asia Regional Office. He is currently senior econominst in the European Bank for Reconstruction and Development. Peter Cook and Edward Holland were consultants to the China Department. Library of Congress Cataloging-in-Publication Data Transport development in southern China / Clell G. Harral, editor; Peter Cook and Edward Holland, prinicipal contributors. p. cm.-((World Bank discussion papers; 151) Includes bibliographical references. ISBN 0-8213-2050-5 1. Transportation-China-Kwa ngtung Province-Forecasting. 2. Kwangtung Province (China)-E]conomic conditions. I. Harral, Clell G. II. Cook, Peter, 1942- . III. Holland, Edward, 1917- IV. Series. HE278.Z7K948 1992 338'.0951'27-dc2O 92-2872 CIP  v Foreword The World Bank's economic and sector work program in China is a very active one ranging over a wide spectrum of topics from macroeconomics to health and education. Each year we publish a handful of our formal studies, but thus far most of the background papers and informal reports, many of them containing valuable analysis and information, have remained outside the public domain. Through the China and Mongolia Department Working Paper Series, we hope to make available to a broad readership among the China watchers and development communities a few of the papers which can contribute to a better understanding of China's modernization. The study by Messrs. Cook, et. al. offers a detailed examination of Guangdong's transport system in the late eighties, highlighting the principal constraints, the likely pattern of modal development and the investment priorities. Guangdong is one of China's most dynamic provinces. Its future economic successes will depend to a significant extent on the ability to improve infrastructure, especially transport facilities. We believe that this paper not only identifies the scale of the problem but also indicates how it might be solved. Shahid Javed Burki Director China and Mongolia Department Asia Region  - vi - CURRENCY EOUIVALENTS (as of February 1989) Currency name = Renminbi Currency unit = Yuan (Y) = 100 Fen July 1986 - December 1989 $1.00 = Y 3.71 $0.27 = Y 1.0 $1.0 million = Y 3.7 million $270,270 = Y 1.0 million From December 1989 $1.00 = Y 4.71 $0.21 = Y 1.00 FISCAL YEAR (FY) January 1 to December 31 PRINCIPAL ABBREVIATIONS AND ACRONYMS USED AADT Average Annual Daily Traffic CAAC China Civil Aviation Administration CJCAC China Joint Civil Aviation Company COSCo China Overseas Shipping Company DOC Guangdong Department of Communications dwt Deadweight Ton EOZ Economic Open Zone ERR Economic Rate of Return GMS Guangzhou-Meixian-Shantou (Railroad) GNP Gross National Product GPPC Guangdong Provincial Planning Commission GVAIO Gross Value of Agricultural and Industrial Output ICT Institute for Comprehensive Transport LCL less-than-car-load LRIFC Long Run Incremental Financial Cost LRISC Long Run Incremental Social Cost LTL less-than-trailer-load MR Ministry of Railways MT million tons NPV Net Present Value NPV/CC Ratio of Net Present Value to Capital Cost OCC Opportunity Cost of Capital pcu passenger car unit SEZ Special Economic Zone SPC State Planning Commission TEU Twenty-foot Equivalent Unit (container) voc vehicle operating cost  - vii - Preface This Study on issues of comprehensive transport in Guangdong Prov- ince was, in line with suggestions by the World Bank, approved by the Chinese Government, under the leadership of the State Planning Commission. The Study was conducted jointly by the Institute of Comprehensive Transport of SPC, Guangdong Provincial Planning Commission, and the China Department of the World Bank. The long-term objective of this Study was set for 2000. The Study started in April 1986. Based on joint reconnaissance, a conceptual outline including key points and grounding was produced by a Chinese team of experts in September 1986. In November 1986, teams of experts from both sides met in Washington D.C. and exchanged views on the outline. A draft of the Study was written by the Chinese team in February 1987. The draft was discussed and modified by both teams and the Chinese team produced the Study Report with suggestions of proposals for the World Bank lending. Both teams of experts further exchanged views on the Study Report in September 1988, followed by a field survey for proposed projects in Guangdong Province. Based on analysis of transport long-run marginal financial cost and economic cost, the World Bank produced a final draft of the Study in May 1989. Views were exchanged again in Beijing and Guangzhou by both teams of experts in October 1989, followed by a field trip which was related to proposed proj- ects. The Study has been issued in green cover, and now is presented as a departmental working paper. This is the fourth edition of the Study. Earlier editions of the report received limited circulation, primarily within Guangdong Province, cen- tral government agencies in Beijing, and the World Bank. However, the report has attracted widespread interest, not only for its findings concerning Guangdong Province, but also for its possible relevance to the transportation problems of China more generally. It demonstrates an improved methodology for examining the potential economic role of the different modes of transport and indicates how China might better exploit the full spectrum of modern transport technologies. Consequently, it was decided to publish this new edition--des- pite many obvious limitations--in order to disseminate the information more broadly to all parties who have an interest in the important role that trans- port will play in the future development of the Chinese economy and to stimu- late further improvements in comprehensive transport planning throughout China. This edition updates preceding editions to take account of the major revisions in transport tariffs which occurred in September 1989 for passengers and March 1990 for freight (as reflected particularly in Chapter IV). Improved and updated information on the costs of major investment projects in Guangdong Province is also included (Chapter VI). Price series have been adjusted to March 1990. It has not been possible however, to update much other information, and, in particular, primary information on most traffic flows was not available beyond 1987, and consequently the original traffic projections beyond 1987 have been retained in many instances.  - viii - In order to present the material of the Study in a single volume, this edition includes only one of the technical annexes presented in earlier editions. The tables associated with this technical annex are not included but remain available to readers upon request. The leaders involved in this Study were Wang Derong, former Director of ICT and Liu Liru, current Director of ICT, Liu Yiling, Deputy Director of GPPC and Zheng Fanji, Inspector of GPPC. The Chinese team of experts, includ- ing He Tianziang, Wang Yan, Zhou Nan, Fu He, Zhang Haipo, He Zengrong, Go Wenlong and Zhou Lixing, was directed by Chai Bencheng and Huang Yuqing. Yang Hongnie took part in the field trip. Guo Yun took part in the discussion of the final version. Wang Xueling also took the job as an interpreter. Zhuang Wanlong, Zhang Baoxian, and Lou Yongchu of the Transport Division of GPPC, and Lao Yu, Ma Youliang, Xu Bangda, Tang Guangzheng, Liang Shuwen, Cai Kaiji, Deng Wenbang and Chen Weigiang and Kuang Shaozheng, etc. took part in the field trip, editing, and discussion of the Study. Ali A. Sabeti, Bernard Montfort and Jose M. Veniard of the World Bank have supported the Study. The World Bank team was directed in the early stages by Jacques Yenny and subsequently by Clell G. Harral. David Hawes and James F. Gruver and Li Lufeng provided assistance in the early formulation of the Study. The major burden of analysis in two of the editions fell on Peter Cook with the able assistance of Surhid Gautam and Liu Zhi. Edward Holland assisted in the formulation ais well as writing of Chapter IV, which is a true joint product of Cook, Harral and Holland. Helpful criticisms and advice were received from the Bank's peer review group consisting of Frida Johansen, Thawat Watanatada, Jacques Yenny, and Stephen Stares. Like earlier versions, this edition is the result of a joint effort by the Guangdong Provincial Planning Commission, the Institute of Comprehen- sive Transport of the Statea Planning Commission and the China Department of the World Bank, who therefore would appreciate any criticism and comments.  - ix - Suimmary The Guangdong Economy and Its Transportation Problems i. After the economic reforms commencing in 1978 Guangdong Province entered a sustained period of rapid economic growth which is fundamentally altering the economic structure of the province. Over the period 1980-86 the Gross Value of Agriculture and Industrial Output (GVAIO) in constant prices increased at an annual rate of 13.5 percent, and in 1987 reportedly surged by an extraordinary 27.2 percent. Industry, fueled by an exports boom, provided the primary thrust for the growth (18.4 percent annual growth over 1980-87) while agriculture grew at a comparatively slower rate (7.6 percent), and by the end of 1987 industry had increased to 67 percent of total GVAIO. Light industry in turn constitutes 65 percent of industrial output, and this share has remained largely unchanged, as heavy industry, led by a construction boom, has so far kept pace with light industry. The establishment of three of China's four Special Economic Zones (at Shenzhen, Zhuhai, and Shantou) in Guangdong and the proximity of the province to the financial and trading cen- ter of Hong Kong have facilitated major foreign capital inflows and helped Guangdong displace Shanghai as China's leading exporter. Guangdong may thus be viewed as the leading edge of China's so far successful coastal development strategy. ii. A number of factors will, however, increasingly impinge upon the Guangdong economy and restrain its future growth. Two critical bottlenecks are the interrelated shortages of transport and energy. Hydropower develop- ment in the province has about reached its full potential, and access to coal, either in northern China or eastern Guangdong, is limited by constraints in railway and ports capacity; as a result existing industrial capacity is already being idled for want of coal to provide power. And transport con- straints are an economic problem well beyond their effects on the availability of coal. Light manufacturing and high valued agricultural exports, which demand flexible, fast, door-to-door services to compete in world markets are constrained by the poor quality of transport services resulting from shortages of equipment, outmoded equipment and substandard, often congested infrastruc- ture. The development of mineral based activities in resource rich, densely populated, but otherwise impoverished eastern Guangdong has been blocked for lack of low cost access, denying both the growth centers access to potentially lower-cost resources and the eastern region participation in the economic miracle taking place elsewhere in the province. iii. The shortages of transport capacity are due partly to underinvest- ment in the transportation sector: recent economic growth has outstripped expansion of transportation capacity by a wide margin, and any previous slack in the system has by now largely been absorbed. Shortages of transportation (and energy) are also at least partly due to pricing distortions. Prices apparently well below costs for certain services have served to generate non- economic excess demands which clog the transport arteries, distort the choice of modes, and result in less than optimal use of the limited available facili- ties and services. Measures to help Guangdong Province meet its future trans- port demands most economically will include both massive investments to provide new capacities at lower user cost and substantial revisions in present  pricing and/or administrative traffic allocation rules to ensure more effi- cient use of existing facilities. The Study iv. The present study is primarily focused on delineating the nature of the future transport system which would best serve Guangdong's economic requirements and on identifying some of the measures needed to achieve such a system. A central question is the economic role of the different modes of transport: rail, road, water, and air. v. In addressing this question the study seeks to lay a foundation for systematic comparison of the costs (and to a limited extent at this stage the service characteristics) of the different modes for four major traffics: (a) passengers, (b) coal, (c) heavy freight, and (d) less-than-carload ship- ments. It estimates financial and economic (or social) cost functions for each of the modes on the assumption that, over the longer run, investments in the network (either for renewal or expansion) would employ the best of exist- ing modern technology as appropriate to China's resource endowments. This results in absolute costs and relative costs of the various modes which differ significantly in most cases from the average costs of the present technology mix. When these long-run ccosts are compared with present prices (or transport tariffs), major differences are also found. Comparative Cost Findings vi. The principal findings of these comparative cost and price analyses may be summarized as follows: (a) Highway economic costs are likely to fall significantly--possibly by half--as China and Guangdong Province move toward a modern road transport system, if prices and user charges are set for economic efficiency. Railway economic costs maV also fall, but by much less, since the railways are currently much closer to the efficiency fron- tier; (b) Recent tariff increases for coal were not sufficient to close the gap between tariffs and economic costs: rail tariffs would have to be increased by another 30 to 50 percent to bring them in line with long-run economic costs; (c) Recent increases in the costs of coastal shipping appear to have substantially eroded the comparative advantage of that mode in eco- nomic terms vis-a-vis all-rail movements of coal (and similar bulk commodities). This tentative conclusion is surprising, and the matter should receLve closer scrutiny--more detailed information is required on the prices of bulk-carrier vessels of different sizes and configurations, on the economic costs of transshipment and port handling, and on the potential economies from technological modern- ization of vessels and port-handling operations; (d) Recent tariff increases for rail passengers have eliminated subsi- dies and should encourage passengers to shift to bus transport,  - xi - which would be more economically efficient at least for short dis- tances; (e) Airline passenger fares vary markedly (relative to either carriers' financial costs or society's economic costs) across different routes; some routes are subsidized and others would appear to be highly profitable--some economic rationalization would appear desir- able; and (f) Comparisons of carrier's long-run financial costs under the present tax/price regime with society's economic costs show substantial net subsidies (20-45 percent), for all modes except highways and, to a lesser extent, inland waterways. For road transport, carriers' costs exceed society's costs by some 10 to 35 percent, reflecting a tax on road transport beyond the long-run costs of providing improved infrastructure and thereby exacerbating price distortions vis-a-vis the other modes. Economic Role of the Transport Modes vii. Several conclusions may be drawn from these cost comparisons: (a) Given the changing structure of the Guangdong economy, highways would be expected to play an increasingly larger role in the future even at present road transport costs, because the fastest growing segments of the Guangdong economy are industries which are typically prepared to pay premium prices for the speed and flexibility of road transport. The magnitude of the potential reduction in highway transport costs would further expand its scope beyond the presently limited usages and extremely short haul distances. The highway could provide more economic services for short to medium distance passenger transport and small lot size or door-to-door freight up to 1,000 km haul distance, thus relieving railways congestion and improving overall transport service quality. Improved interprovin- cial road haulage is a possible future priority once the intrapro- vincial road links are upgraded;l/ (b) The railway should concentrate on its most economic transport roles, which include medium distance freight and passengers and high-volume bulk commodities and heavy freight at all distances. The railroad should coordinate with coastal shipping for transport of coal from the north to Guangdong Province; (c) For coastal shipping, rapidly changing costs and technological inno- vation require more detailed studies, including a reevaluation of 1/ It should be noted that there is not unanimity of views among the co- authors of the report on this conclusion. In particular, the Guangdong Province Planning Commission and the Institute of Comprehensive Transport wish to record their view that constraints on the development of road transport are likely to limit its role to haul distances much below 1,000 km at least up to the year 2000.  - xii - the most efficient vessel sizes and configurations, in order to determine the appropriate economic role for that mode; and (d) Air transport should expand its long distance and medium distance services and price them in accordance with incremental costs to society. These services are expected to grow rapidly with economic growth and increasing value of time for business travellers in Guangdong Province and the rest of China. Aviation can make an enormous contribution to the economic efficiency of Chinese business and public administration, as well as continuing to earn important foreign exchange from the growing tourism industry. Investment Priorities viii. Massive investments will be required to approach an economically efficient transport system in Guangdong Province. A total program of Y 24-30 billion over the next five years (1991-95) has been tentatively identified: billion Yuan Z (1990 prices) Highways 18-21 70-75 Railways 3-4 12-16 Inland Water Less than 0.5 Less than 3 Seaports 2.5-3 8-10 Airports 0.5-1 1-4 Intermodal Less than 0.5 Less than 3 TOTAL 24-30 100 ix. For the two subsectors--highways and railways--for which available information permitted preliminary benefit-cost analyses, these show quite high economic rates of return (ERR) and net present values (NPV). Present poor conditions and the rapidly expanding role of highway transport result in the highest economic returns in that mode. Just under Y 1 billion of the proposed road investments have ERRs in excess of 100 percent; another Y 4.5 billion have ERRs ranging from 30 to 95 percent, and a further Y 15.5 billion have ERRs between 15 and 30 percent. The highest-yield components are for pavement strengthening on highly-trafficked roads with weak and/or aging pavements. x. In the railways subsector, two segments of the proposed Guangzhou- Meixian-Shantou railway, designed to support expanded mining and cement pro- duction in the Lianping-Meixian area, appear to have priority: Huizhou- Xiantangxi (Y 327 million) and Xiantangxi-Meixian (Y 664 million) with ERRs of 34 percent and 39 percent respectively. The third section, Shekeng-Shantou (Y 399 million) would have a somewhat lower ERR of 27 percent. The recently opened double-tracked sections of the main north-south (Guangzhou-Beijing) mainline are currently projected to reach their new capacity (30 million tons each direction) by 1992; however, the conclusions of this study concerning prospective diversion of coal traffic to coastal shipping routes should be fully evaluated before determining what additional improvements may be war- ranted in the north-south mainline.  - xiii - xi. For aviation and water transport available information did not per- mit similar levels of analysis of prospective investments. Nonetheless, anal- yses of the comparative costs and service characteristics of the different modes suggest a much larger role for these modes would be highly beneficial in economic terms. In addition, a regional study of shipping service examining hub-and-spoke vs. direct services and alternative sites for load center loca- tion (for coal, petroleum products, and containers) is recommended, with explicit analysis of the future role of Hong Kong ports. Further multimodal transport investments to support the recent surge in containerization should also be examined. The Planning Process xii. The present planning process in Guangdong Province (for transporta- tion as well as other sections) has undergone substantial restructuring to cope with the policy changes favoring decentralized management. More trans- portation planning must now be done on the local and provincial levels and port planning (except for new ports) is largely in the hands of port managers. This new system, however, places increased demands on provincial and local planners with few resources in terms of trained manpower or data oriented to the needs of decisionmakers, who have little experience in using analytic tools for assessing transportation priorities and policy impacts, especially in the broader context of a partial market economy. xiii. This process has benefited by the experience gained from comprehen- sive study; however, further steps are needed to reinforce and support provin- cial, local and modal planners and create a more open and continuous process that involves carriers and user groups to make the planning as responsive as possible to present and future needs. Increased coordination in project iden- tification and evaluation is required to avoid potential misinvestment. Spe- cific areas of improvement in planning approaches and associated data systems are recommended, as well as strategies for improving each transportation sub- sector and intermodal transport in response to the findings of the analysis. Multimodal transport investments, for example, will require a concerted effort by several agencies and private organizations to ensure a coordinated strat- egye  I  - xv - Transport Development in Southern China Table of Contents Paae No. I. GUANGDONG PROVINCE ECONOMIC CHARACTERISTICS AND PERSPECTIVES 1 A. Geographic Setting .. 1 B. Administrative Jurisdictions . . . . . . . . . . . . . . 1 C. Population and Labor Force . . . . . . . . . . . . . . . 3 D. Economic Activity ... . . . . . . . . . . . . . . . . 4 Income Levels and Growth . . . . . . . . . . . . . 4 Economic Geography . . . . . . . . . . . . . . . . 6 Light Industry . . . . . . . . . . . . . . . . . . 6 Heavy Industrial Production . . . . . . . . . . . . 7 Agricultural Production . . . . . . . . . . . . . . a Interprovincial and Foreign Trade . . . . . . . . . 8 E. Present Constraints on Development . . . . . . . . . . . 10 F. Economic Forecasts . . . . . . . . . . . . . . . . . . . 10 G. Location of Future Growth . . . . . . . . . . . . . . . 12 II. TRANSPORT INFRASTRUCTURE. SERVICES AND MANAGEMENT . . . . . . 14 A. Transport Network Overview . . . . . . . . . . . . . . . 14 B. Transportation Infrastructure by Mode . . . . . . . . . 15 Railways . . . . . . . . . . . . . . . . . . . . . 15 Roads . . . . . . . . . . . . . . . . . . . . . . . 15 Coastal and Marine . . . . . . . . . . . . . . . . 16 Inland Waterways ... . . . ..... . . . . . . 16 Air .... . . . . . . ..... . . . . . . . . . 17 C. Transportation Services ... . . . .... . . . . . . 17 Rail Services ... . . . . ..... . . . . . . . 17 Road Services ... . . . . ..... . . . . . . . 19 Road Vehicle Fleet ... . . . .... . . . . . . 20 Inland Waterway Services . . . . . . . . . . . . . 20 Coastal and Ocean Shipping Services . . . . . . . . 21 Air Services ... . . . . ..... . . . . . . . 21 Containerization ... . . . ..... . . . . . . 22 D. Management of Transport Modes . . . . . . . . . . . . . 23 Railways . . . . . . . . . . . . . . . . . . . . . 23 Ports . . . . . . . . . . . . . . . . . . . . . . . 24 Highways ... . . . . . .. 25 Waterways ... . . . . . ..... . . . . . . . . 26 Aviation . . . . . . . . . . . . . . . . . . . . . 26 Maintenance of Transport Modes . . . . . . . . . . 26 E. Transport Investment ... . . . . .... . . . . . . . 27  - xvi - Page No. III. PRESENT TRAFFIC AND SERVICE LEVELS . . . . . . . . . . . . . 29 A. Overview of Aggsregate Freight and Passenger Traffic Trends 29 B. Present Traffic! by Corridor and Type . . . . . . . . . . 31 Corridor F'reight Traffic . . . . . . . . . . . . . 32 Total Provincial Freight Traffic . . . . . . . . . 32 Passenger Traffic . . . . . . . . . . . . . . . . . 33 Geographic Patterns ... . . . . . . . . . . . . . 33 C. Modal Shares of Traffic by Corridor . . . . . . . . . . 34 Determinants of Modal Shares . . . . . . . . . . . 34 Freight Modal Shares . . . . . . . . . . . . . . . 35 Passenger Modal Shares . . . . . . . . . . . . . . 35 Total Provincial Modal Share Trends . . . . . . . . 36 D. Assessment of Existing Capacity Constraints . . . . . . 37 E. Assessment of Technology ... . . . . . . . . . . . . . 38 IV. COMPARATIVE ROLES OF THE TRANSPORT MODES . . . . . . . . . . 41 A. Comparative Transport Costs and Pricing Among Modes . . 42 B. Comparative Cost Analyses . . . . . . . . . . . . . . . 44 Passenger Transport Costs and Prices . . . . . . . 44 Bulk Commodity (Coal) Transport Costs . . . . . . . 46 Heavy Freight Transport Costs and Tariffs . . . . . 47 Less-than-Carload General Cargo . . . . . . . . . . 48 C. Impacts of Inefficient Conditions . . . . . . . . . . . 48 D. Conclusions . . . . . . . . . . . . . . . . . . . . . . 49 V. FUTURE TRAFFIC AND CAPACITY CONSTRAINTS . . . . . . . . . . . 52 A. Methods of Traffic Forecasting . . . . . . . . . . . . . 52 B. Freight Traffic Forecasts . . . . . . . . . . . . . . . 53 International Freight Traffic . . . . . . . . . . . 53 Interprovincial Freight . . . . . . . . . . . . . . 53 Intraprovincial Freight . . . . . . . . . . . . . . 53 Induced Freight Traffic . . . . . . . . . . . . . . 54 Network Provincial Total Freight Traffic . . . . . 55 Rail CapaciLty Constraints . . . . . . . . . . . . . 56 Freight Haul Distance . . . . . . . . . . . . . . . 57 C. Passenger Traffic Forecasts . . . . . . . . . . . . . . 57 International Passengers . . . . . . . . . . . . . 57 Interproviricial Passengers . . . . . . . . . . . . 58 Intraprovincial Travel . . . . . . . . . . . . . . 59 D. Vehicle Traffic . . . . . . . . . . . . . . . . . . . . 61 E. Summary of Corridor Forecasts . . . . . . . . . . . . . 61 Freight ... . . . . . . . . . . . . . . . . . . . 61 Passenger Trips ... . . . . . . . . . . . . . . . 62 F. ImplicatLons for Capacity Constraints . . . . . . . . . 62  - xvii - Paae No. VI. POTENTIAL TRANSPORTATION INVESTMENTS . . . . . . . . . . . . 64 Investment Evaluation Techniques . . . . . . . . . . . . . . 64 Highway Investments . . . . . . . . . . . . . . . . . . . . . 65 Railway Investments . . . . . . . . . . . . . . . . . . . . . 68 Seaports . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Inland Water Investments . . . . . . . . . . . . . . . . . . 70 Airport Investments . . . . . . . . . . . . . . . . . . . . . 71 Multimodal and Intermodal Investments . . . . . . . . . . . . 71 Total Transport Investments . . . . . . . . . . . . . . . . . 72 Investment Financing . . . . . . . . . . . . . . . . . . . . 72 VII. CONCLUSIONS AND RECOMMENDATIONS . . . . . . . . . . . . . . . 74 Economic Role of the Transport Modes . . . . . . . . . . . . 75 Investment Priorities ... . . . . . . . . . . . . . . . . . 76 Planning Process ... . . . . . . . . . . . . . . . . . . . 77 LIST OF TABLES IN TEXT 1.1 Provincial GVAIO . . . . . . . . . . . . . . . . . . . . . . 4 1.2 Growth in GVAIO by Transport Corridor . . . . . . . . . . . . 7 1.3 Forecast Economic Growth Rates (GVAIO) . . . . . . . . . . . 12 3.1 Freight Traffic . . . . . . . . . . . . . . . . . . . . . . . 30 3.2 Passenger Traffic . . . . . . . . . . . . . . . . . . . . . . 31 3.3 1987 Corridor Freight Traffic .32 3.4 1987 Total Provincial Freight Traffic . . . . . . . . . . . . 33 3.5 1987 Provincial and Corridor Passenger Flows . . . . . . . . 34 3.6 1987 Freight Modal Shares in Corridors . . . . . . . . . . . 36 3.7 1987 Passenger Modal Shares in Corridors . . . . . . . . . . 36 4.1 Inefficiency Ratios ... . . . . . . . . . . . . . . . . . .. 49 5.1 International Traffic in Major Corridors . . . . . . . . . . 53 5.2 Interprovincial Traffic in Major Corridors . . . . . . . . . 54 5.3 Intraprovincial Freight in Major Corridors . . . . . . . . . 54 5.4 Total Major Corridor Freight Traffic . . . . . . . . . . . . 55 5.5 Network Total Provincial Freight Traffic: Unconstrained . . 56 5.6 Network Provincial Freight Traffic: Constrained . . . . . . 57 5.7 Average Haul Distance . . . . . . . . . . . . . . . . . . . . 58 5.8 Forecast Growth in Passenger Trips by Transport Corridor: Base Case ... . . . . . . . . . . . . . . . . . . . . 59 5.9 Total Provincial Passenger Traffic Forecasts by Mode - Unconstrained . . . . . . . . . . . . . . . . . . . . . 60 6.1 Minimum Traffic Thresholds (AADT) for Typical Road Improvements 67 6.2 Prospective Transport Investment Program . . . . . . . . . . 73  - xviii - Page No. LIST OF TABLES 1 Guangdong Province - Population and Land Area by Corridor 80 2 Labor Force Employed by Sector . . . . . . . . . . . . . . 81 3 Production of Selected Major Commodities . . . . . . . . . 82 4 Inflow and Outflow of Commodities in 1987 . . . . . . . . . 83 5 Guangdong Province Foreign Trade with Hong Kong . . . . . . 84 6 Hong Kong Trade with Guangdong by Mode and by Type for 1987 85 7 Counties and Cities with High Growth Potential . . . . . . 86 8 Capital Investment by Sector . . . . . . . . . . . . . . . 87 9 Growth Indicators for Guangdong Province . . . . . . . . . 88 10 Comparison of Corridor Growth for Three Transport Systems - Base Case Forecas: .89 11 Comparison of Corridor Growth for Three Transport Systems - Optimistic Forecast . . . . . . . . . . . . . . . . . . 90 12 Comparison of Corridor Growth for Three Transport Systems - Pessimistic Forecast . . . . . . . . . . . . . . . . . . 91 13 Length of Internal Provincial Transport Routes by Mode--1986 . . . . . . . . . . . . . . . . . . . . . . 92 14 Key Railway Section Characteristics . . . . . . . . . . . . 93 15 Length of the Road Network by Class, Surface and Function . 94 16 Density of Road Network by Region, 1986 . . . . . . . . . . 95 17 Characteristics of Major Seaports . . . . . . . . . . . . . 96 18 Inland Waterway Transport System . . . . . . . . . . . . . 97 19 Airports by Class and Aircraft Accommodated in 1987 . . . . 98 20 Rail Passenger Services, 1988 . . . . . . . . . . . . . . . 99 21 Amount of Railway Rolling Stock by Type and Year . . . . . 100 22 Guangdong State-ow;ned, Specialized Road Transport Services 101 23 Motor Vehicle Ownership in Guangdong Province . . . . . . . 102 24 Inland Waterway Services, 1986 . . . . . . . . . . . . . . 103 25 Maritime Shipping Services, 1986 and 1987 . . . . . . . . . 104 26 Water Freight Tonnage by Port in 1985 and 1987 . . . . . . 105 27 Guangdong Air Services, 1986 and 1987 . . . . . . . . . . . 106 28 Summary of Past Transport Investments in New Fixed Assets . 107 29 Guangdong Provincial Freight Traffic . . . . . . . . . . . 108 30 Guangdong Provincial Passenger Traffic . . . . . . . . . . 109 31 Estimated Total Flow of Commodities by Corridor in Province in 1987 . . . . . . . . . . . . . . . . . . . . 110 32 Estimated Passenger Traffic for 1987 . . . . . . . . . . . 111 33 Average Daily Road Traffic . . . . . . . . . . . . . . . . 112 34 Modal Capacities by Corridor . . . . . . . . . . . . . . . 113 35 Guangdong Province Comprehensive Transport Study Summary of Key Capacity Constraints . . . . . . . . . . 114 36 Passenger Perceivecd Costs by Mode Compared with LRIFC and LRISC . . . . . . . . . . . . . . . . . . . . 115 37 Coal Shipper's Costs by Mode Compared with LRIFC and LRISC 116 38 Heavy Freight Shipper's Costs by Mode Compared with LRIFC and LRISC . . . . . . . . . . . . . . . . . . . . 117 39 General Cargo Shipper's Costs by Mode Compared with LRIFC and LRISC . . . . . . . . . . . . . . . . . . . . 118 40 Freight Traffic Forecasts by Corridor and Mode . . . . . . 119  - xix - Page No. 41 Potential Induced Freight Forecasts in the East Corridor 120 42 Total Provincial Freight Forecasts by Mode . . . . . . . . 121-2 43 Northern Corridor Potential Diverted Freight Traffic . . . 123-4 44 Total Corridor Freight Forecasts by Mode . . . . . . . . . 125 45 Forecast Growth in Passenger Trips by Transport Corridor Base Case System .126 46 Total Corridor Passenger Forecasts by Mode . . . . . . . . 127 47 Total Provincial Passenger Forecasts by Mode . . . . . . . 128 48 Northern Corridor Potential Diverted Passenger Trips Passenger Traffic - Base Case System . . . . . . . . . . . . . . . 129 49a Summary of Proposed Investment Projects: Roads . . . . . . 130 49b Summary of Proposed Investment Projects: Railways . . . . 131 49c Summary of Proposed Investment Projects: Riverports . . . 132 49d Summary of Proposed Investment Projects: Inland Waterways 133 49e Summary of Proposed Investment Projects: Seaports . . . . 134 49f Summary of Proposed Investment Projects: Airports . . . . 135 50 Preliminary Project Economic Efficiency Ranking: Roads . 136 LIST OF FIGURES 1 Transportation Organization in Guangdong Province .137 2 Average Daily Road Traffic - Guangdong Province . . . . . . 138 3 Transport Demand Forecasting Task Flow Chart . . . . . . . 139 4 Passenger Fare with Distance . . . . . . . . . . . . . . . 140 5 Passenger Cost with Distance - LRISC . . . . . . . . . . . 141 6 Passenger Trip Cost by Value of Time - Guangzhou - Shantou 142 7 Passenger Trip Cost by Value of Time - Guangzhou - Shanghai 143 8a Coal Transport Cost - UPC .144 8b Coal Transport Cost with Distance - LRISC . . . . . . . . . 145 9a Heavy Freight Transport Cost - UPC . . . . . . . . . . . . 146 9b Heavy Freight Transport Cost - LRISC . . . . . . . . . . . 147 10a LCL Freight Transport Cost - UPC . . . . . . . . . . . . . 148 10b LCL Freight Transport Cost - LRISC . . . . . . . . . . . . 149 TECHNICAL ANNEX Calculation of Long-Run Incremental Social Costs by Mode . 151-96   I. GUANGDONG PROVINCE ECONOMIC CHARACTERISTICS AND PERSPECTIVES A. Geographic Setting 1.1 Guangdong Province lies along the South China Sea at the southeas- tern corner of the People's Republic of China (see Map IBRD 20592). This position makes it the point of closest maritime contact between China and Southeast Asia, Africa, and Europe. Its small but important borders with Hong Kong and Macao make Guangdong Province one of the most important gateways between China and the rest of the world. 1.2 The province covers an area of 178,000 km2, excluding Hainan Island I/ and many small islands along the rugged coast (see Map IBRD 20592). More than three-fourths of the total land area of Guangdong Province is mountainous or hilly. The only large areas of flatland are the Pearl River Delta and the region around the port city of Shantou, in the east. Long dis- tance transport is obstructed by the Nanling Mountains in the north of Guangdong, the Yunkaida Mountains in the west, and the Lianhua Mountains in the east. The South China Sea creates a coastline of 4,300 km, the largest of any Chinese province, providing coastal shipping links with the other prov- inces as well as international trade. 1.3 Guangdong is drained by three main rivers, the Xi Jiang, Bei Jiang, and Dong Jiang, located in the western, northern, and eastern parts of the province, respectively. All three rivers are navigable to some degree. The three rivers unite to form the Pearl River (Zhu Jiang) and its delta, which plays an important geographic and socioeconomic role in the Province. The Pearl River Delta commences at the intersection of the Xi and Bei Rivers, 45 km northwest of Guangzhou (formerly known as Canton). The Delta contains a network of channels and canals stretching 110 km from north to south and 90 km from east to west. 1.4 Most of Guangdong province has a subtropical climate with tropical monsoons. The monsoon climate includes a rainy season from April to Septem- ber, when floods and typhoons frequently disrupt ports and interfere with land transport and coastal shipping activities. No frost or snow occurs except in the highest elevations of northern Guangdong. B. Administrative Jurisdictions 1.5 Administratively. Guangdong Province is divided into three prefec- tures and nine prefectural-level municipalities; on the lower level, it con- tains 76 counties and six county-level municipalities. Guangzhou is the pro- vincial capital, containing 3.36 million inhabitants. It is the largest city in Guangdong and one of the largest cities in China. Guangdong Province also 1/ This study focuses on the mainland, and wherever possible the data pre- sented here exclude Hainan Island. Hainan Island has an area of 33,900 km2 and is located at the southernmost point of the province. Formerly it was an administrative district of Guangdong Province; in 1988 it became a separate province.  -2- contains three of China's four special economic zones (SEZs): Shenzhen, Zhuhai, and Shantou (see Map IBRD 20592). 1.6 The Special Economic Zones in Guangdong were established in 1980-82. They were authorized to offer foreign investors preferential terms on land costs, labor, and taxes. Other market-oriented policies were instituted in these zones on an experimental basis. As a result, an enormous amount of investment has poured into the SEZs (Y 17 billion from 1985 to 1987) from both domestic and foreign sources to fuel light industry, commerce, property devel- opment, and tourism. The Shenzhen SEZ is especially well placed next to Hong Kong and has grown by far the fastest, with emphasis on the assembly of elec- tronic products and household electrical goods. The population of the Shenzhen zone jumped from about 30,000 in 1978 to 500,000 in 1985 and to 600,000 in 1987 (of which 313,000 is classified as temporary population). Shenzhen's currently most profitable industrial activity is assembly and sub- contracting work for Hong Kong manufacturers, but, aspiring to a larger role, Shenzhen has made high-technology industry a priority. In 1985, it set up its own Science and Technology Industrial Park to encourage further production of more sophisticated products such as quartz crystals and computer disk drives. Nearly all of China's provinces have built their own hotels in Shenzhen. There are now 2,000 domestic investors in the zone, and half as many foreign firms. Because of its proximity to Hong Kong and its relative freedom com- pared to other parts of China, Shenzhen remains a magnet for both foreign and domestic investment and economic expansion. 1.7 The Zhuhai zone is much smaller. It is managed as part of Zhuhai Municipality, which recently was expanded to include 1,266 km of land area and 145 islands off the co-st (shown in Map IBRD 20592). Zhuhai Municipality, which borders on Macao, is ]ocated 36 nautical miles from Hong Kong. Trans- port services between the two cities are available by hydrofoil, as wel]. as by more conventional modes. Since the SEZ was created in 1980, the Zhuhai zone has changed from a largely undeveloped countryside to a significant industrial and tourist center with a city population of 420,000 (including 180,000 tempo- rary workers). Foreign investment, primarily channeled through Hong Kong, contributed $500 million to 336 joint ventures as of 1987, and agreements for 2,215 more have been reached. with an anticipated investment of $1.5 billion. The foreign investment has been concentrated in light manufacturing (55 per- cent), commerce (21 percent), and tourism. Major industrial sectors include electronics, clothing, food, and furniture. Infrastructure in the zone has expanded dramatically in the form of new industrial parks, roads, power and utilities, and an improved port and heliport. Due to its "open door" policy and the attractive landscape along Zhuhai's coastline, tourism has developed rapidly. The Zhuhai authorities are considering a variety of development proposals for both island and mainland sites. 1.8 The Shantou SEZ is also relatively small compared to Shenzhen. It is more isolated and less well equipped in infrastructure than Zhuhai, perhaps because it was started two years later. in 1982. Nevertheless, Shantou is located in a highly populated area of the province with an abundant labor force. By January 1988, it had attracted nearly 500 industrial and agribusi- ness enterprises, including 259 foreign joint ventures and 211 domestic ven- tures, for a total investment of about $250 million. Commitments have been made for another $150 million, and improvements to major ports and other infrastructure are under way. The development to date has focused on aquacul-  - 3 - ture and light industry, with textiles, electronics, consumer goods and food processing taking the lead. The goal for the SEZ is to attain a growth rate of over 40 percent per year. 1.9 In addition to the Special Economic Zones, Guangdong Province has established two Economic and Technological Development Zones (Canton and Zhanjiang), and in 1986 it established an Economic Open Zone in the Pearl River (Zhujiang) Delta. These zones also will benefit from special incentives to encourage export-oriented production, and capital investment for infra- structure. The Economic Open Zone (EOZ) is a less structured concept than the SEZ, allowing the market to determine industrial prices as an experiment in price signals but with fewer administrative rules. The Pearl River Delta EOZ had by the end of 1987 designated 242 industrial satellite towns and 1,165 production and processing bases for agricultural exports. C. Population and Labor Force 1.10 Guangdong Province accounts for 59.7 million, or about 6 percent of China's total population.2/ Guangdong is one of the most densely populated provinces in the country, with an average population density of 295 persons/ km2. (The average population density for all of China is 107 persons/km2.) Population density in Guangdong varies greatly, with the highest density (880 persons/kin2) in Shantou prefecture which is located in the extreme east of the province, while Shaoguan prefecture in northernmost Guangdong has the lowest density (144 persons/kmi2). In general, the south is more heavily populated than the north and the east is more heavily populated than the west. About one fourth of the provincial inhabitants live in the 20 largest cities of 500,000 or above (see Map IBRD 20592.) Table 1 presents a breakdown of the population in each of the four economic regions of the province.3/ 1.11 Guangdong Province has long had more contact with overseas countries than any other Chinese province. The province is estimated to be the ances- tral home of nearly two-thirds of all overseas Chinese. Approximately 12 mil- lion persons, or 20 percent of the provincial population, have relatives liv- ing outside of China. These ties have contributed in recent years to external investment of major magnitude and important tourism in the province by over- seas Chinese. 1.12 Although the majority of Guangdong's labor force population remains agricultural, there has been a slow (but recently rapidly accelerating) shift in employment away from agriculture to industry over the past 30 years. Agri- culture currently accounts for 58 percent of the work force, but only 20 per- cent of total production,4/ although the numbers categorized as agriculture 2/ Excluding Hainan Island and the autonomous states. 3/ Tables 1 through 50, which are numbered consecutively throughout the report, appear following page 78. Summary tables appear in the text of each chapter and are prefaced by the Chapter number. 4/ Many of those classified administratively as "agricultural" actually work in factories and other town or city jobs. This has created a need for commuting services from rural homes to work locations.  - 4 - may include a large number of people actually employed in small-scale town or village manufacturing enterprises. In 1986, about 340,000 rural laborers in Guangdong Province switched to work in township enterprises, the focal point of light industrial and manufacturing activity. In the same year about 20 percent of the rural labor force was reported to have switched to employ- ment in the main provincial cities of Foshan, Jiangmen, Guangzhou, Shenzhen and Zhuhai. (All of these cities are located in the Pearl River Delta region). Industry is now the second largest employer, accounting for 16 per- cent of the labor force in Guangdong (Table 2); this percentage would be much higher if all those employed in village enterprises were included. The sector of commerce, trade, and marketing ranks third, with 7 percent of the labor force, and construction ranks fourth with 5 percent. Fourteen percent or less of the provincial labor fcrce is employed in the remaining sectors. D. Economic Activity 1.13 Income Levels and Growth. Economic activity in Guangdong Province has been growing rapidly in recent years. Between 1980 and 1985 the gross value of agricultural and industrial output (GVAIO) increased by 89 percent, from Y 36.5 to Y 69.1 billion in constant prices, or 13.6 percent per year (Table 1.1 below). In 1987, the rate of growth is tentatively reported to have increased to an extraordinary 27 percent. This acceleration was due primarily to industrial growth. Agricultural output appears to be fluctuat- ing--increasing to 1985, falling in 1986, and picking up again in 1987, as high-value-crop increases compensate for decreases in grain area sown. Table 1.1: PROVINCIAL GVAIO (Constant 1980 prices) Average growth Gross value Gross value Year GVAIO per year of industry of agriculture (Y billion) (Z) Y billion z Y billion z 1957 9.3 2.9 6.4 1970 19.1 5.7 10.2 10.2 8.9 2.6 1975 26.1 6.4 15.9 9.3 10.2 2.8 1980 36.5 6.9 24.3 8.9 12.2 3.6 1985 69.1 13.6 46.1 13.7 23.0 13.5 1986 78.2 13.2 59.5 29.1 18.7 -18.7 1987 99.5 27.2/a 79.1 32.9 20.4 9.1 /a Estimated. Source: Guangdong Province Yearbooks and GPPC for 1987.  1.14 According to sample surveys, the average annual wage of staff and workers in the province in 1986 was Y 1,461 ($409).5/ Total personal income has grown steadily since economic reforms were introduced in 1978, from Y 421 in 1978 to Y 965 in 1986. The average income of employees in state-owned enterprises in Guangdong was Y 1,645 ($460) in 1986; this represented a 204 percent increase over 1980 and a 22 percent increase over 1985. Employees of state-owned enterprises and farmers in Guangdong have incomes about 17- 20 percent above the national average, and these income differentials have been more accentuated in the Special Economic Zones where wages are set by the market. High incomes and rapid economic growth are major contributors to the growth in demand for travel and freight transportation in the province. This growth in demand is expected to continue with increasing economic activity and rising per capita income in the province. 1.:15 Past growth in agricultural output in Guangdong Province is mainly due to growth in certain subsectors: forestry, fish farming, and sideline production, which includes small-scale industrial production by farmers. These subsectors have expanded at the expense of plantation agriculture,6/ where the area sown for grain has decreased to make way for more profitable crops. 1.16 Although agricultural output in Guangdong has been increasing in absolute value, its share of total output value has been declining, from 39 percent in 1958 to 20 percent in 1987. Industrial production dominates the provincial economy, as is evidenced by its 79 percent share of total GVAIO in 1987. Of the industry total, light industry 7/ accounted for 65 percent, and this proportion has remained relatively stable over the last few years. Table 3 depicts the major commodities produced in the province in recent years. 1.17 According to sources both within and outside of the People's Republic of China, Guangdong Province may be fast becoming Asia's "fifth little dragon," alongside Singapore, South Korea, Hong Kong, and Taiwan China. Its easy access to both Hong Kong and Macao and its favorable investment poli- cies have spurred export-oriented production, which reached a value of $5.56 billion in 1987 (30 percent higher than in 1986). This growth was led by 5/l The official exchange rate was devalued in July 1986 from $1 = Y 3.45 to $1 = Y 3.70. For 1986 an average rate of $1 = Y 3.575 has been used. (In December 1989 the exchange rate was again devalued to $1 = Y 4.71.) 6/ Also called "farming", plantation agriculture covers the cultivation of grain crops, cotton, oil-bearing crops, sugar crops, bast-fiber plants, tobacco, vegetables, medicinal herbs, melon and gourd crops, and the cultivation and management of tea plantations, mulberry fields and orchards. 7/ "Light Industry" refers to the production of consumer and intermediate goods, for example, food, textiles, tools, electronics, and chemical pharmaceuticals. "Heavy Industry" refers to the production of primary goods or processes. It consists mainly of mining and lumbering, raw material processing, construction, and manufacturing of cement, ferti- lizer and heavy equipment.  - 6 - foreign and joint-venture enterprises, which experienced a 57 percent increase in output that year. Production is oriented to light industrial manufactured goods with food, textiles, and electronics accounting for 21 percent, 15 per- cent, and 11 percent of output value, respectively. The fastest growing activity is the manufacture of household electrical appliances. 1.18 Economic Geography. Guangdong Province can be divided into four regions according to shared economic and geographic characteristics: the Pearl River Delta, East Guangdong Corridor, West Guangdong Corridor, and North Guangdong Corridor (see Map IBRD 21438). The Pearl River Delta is the most economically developed region. Its combined industrial and agricultural out- put accounted for 60 percent of the provincial total in 1985, although it contains only 30 percent of t:he population and 16 percent of the arable land area in the province. The region includes two of the three SEZs: Shenzhen and Zhuhai, and supports more than 5,000 export-producing enterprises. The East Guangdong Corridor accounted for only 13 percent of the total output of the province in 1985, although it contains 33 percent of the total population and 41 percent of the cultivated land area. The corridor includes the resource-rich, but relatively little developed, Meixian Prefecture. The West Guangdong Corridor is primarily agricultural, except for the farthest western Prefecture (Maoming), which contains an oilshale and refinery complex. In 1985 the corridor accounted for 27 percent of provincial agricultural produc- tion and 10 percent of industrial production. It contains 30 percent of the population of the province and 34 percent of its cultivated land area. The North Guangdong Corridor is primarily mountainous. It provides a section of the major interprovincial transport corridor linking Guangzhou with Beijing. It is the most sparsely settled part of the province, with 9 percent of the cultivated land area and 12 percent of the population in the province, concen- trated in its mining towns (see Table 1). Further description of these corri- dors is found in Chapter III. 1.19 The four corridors havre experienced different rates of economic growth in recent years, as shown in Table 1.2. The 1985/86 growth pattern may be seen as typical of the period 1980-88, as it is close to the average for this period and reflects the effects of strong investment in the early 1980s in light industry. Also this growth reflects the broader development pattern in the Pearl River Delta outside the SEZs, typical of the late 1980s. These figures demonstrate the uneven growth within the Province and the favored position of the Pearl River Delta with respect to the other regions. Within the Delta the East Bank has the highest growth rate, but the West Bank has the highest share of growth. The Delta has absorbed Y 11 billion in infrastruc- ture investments over the last three years, representing fully 97 percent of the capital investment destined for economic zones. This establishes a stronger base for future growth in the Delta than in other parts of the Province. 1.20 Light Industry was concentrated in Guangzhou in the past with a lesser intensity in other population centers such as Shantou and some of the larger cities in the Delta. Since 1980 light industry has grown rapidly (33 percent in 1987) and became diffused throughout the Pearl River Delta with a special focus in the SEZs. Access to Hong Kong, especially by road or river, appears to be the principal criterion for the location of industry, and rural townshlips appear to be favored with higher growth in value of industrial output (47 percent in 1987) than cities.  - 7 - Table 1.2: GROWTH IN GVAIO BY TRANSPORT CORRIDOR (Y in constant 1980 prices) 1985/86 1985 1986 Growth share of Corridor GVAIO GVAIO rate (X) growth /a (Z) Pearl River Delta 3,491 3,939 12.8 52.8 East Guangdong 1,071 1,178 10.0 12.7 West Guangdong 1,325 1,679 19.2 29.8 North Guangdong 441 481 9.0 4.7 Total Province 6,328 7,177 13.4 100.0 /a The share is the difference in GVAIO for the corridor divided by the dif- ference for the Province as a whole. Source: Guangdong Yearbook, 1987. 1.21 Heavy Industrial Production in Guangdong Province historically has been located in relation to the resource base. Manufacturing has been concen- trated in Guangzhou, mining in the northern and eastern areas, and petroleum production from oil shale in Maoming in the West. Heavy industry has kept pace with light industry in recent years, mainly due to the construction boom in the province, which has generated large demands for cement, quarry prod- ucts, and other building materials. Iron and steel plants are located in Shaoguan and in Guangzhou, the latter is also a center for the production of heavy equipment. Cement plants are spread out across the province, although the greatest production takes place in Guangzhou and the Pearl River Delta (50 percent), and the remaining production is evenly divided among the East, West and Northern Corridors. While most cement plants are overcommitted, the plants in Shantou, Huiyang, and Meixian prefectures in the East operate below capacity due to a shortage of energy and/or transportation constraints. The Maoming refinery complex is the largest industry in the West and is used for processing oilshale and crude oil piped from the port at Zhanjiang. Oil exploration and development in the South China Sea also is becoming signifi- cant. 1.22 Coal production is the most important mining activity in the prov- ince. Iron ore production is the second most important mining activity. Most of the ore is produced within Shaoguan Prefecture. There are also nonmetal ores, of which ferrous-sulfate is considered to be the most important. Most minierals are produced in the northern part of the province. There are signi- ficant reserves of coal and iron in several areas of Guangdong. Other ores are found in the East Corridor, but their relative isolation has limited min- ing operations (see Map IBRD 21438). Expanding these activities would require costly investments in transportation facilities through difficult terrain. Oil shale production is less significant; shale mined in the west is converted to oil at the nearby Maoming refinery. Mining, petroleum and cement activi-  - 8 - ties generate much of the demand for transport in the Province, particularly for rail and water transport. 1.23 Agricultural Production. Despite the growth of industry in Guangdong Province in recent years, agriculture remains the major emplcyer and land user. Grains accounted for 70 percent of the land under cultivation in 1986, although total grain production fell 16 percent between 1980 and 1986. Cash crop production rose dur:ing this period. For example, sugarcane produc- tion increased by more than 140 percent, to 20.7 million tons in 1986. Most of the land under cultivation in Guangdong Province is in small farms, and transport demand is widely dispersed. 1.24 Information on the location of agricultural production according to region or corridor is limited. Nevertheless, several general observations can be made about each one. The Pearl River Delta Corridor is one of the most important farming areas in GuaEngdong. It produced more than half of Guangdong's agricultural exports in 1987. However, this production is decreasing for two major reasons. First, more and more farmers in the Delta are shifting from farming to sideline production and rural industry, so culti- vation is less widespread than before. Second, competition from other regions has been increasing rapidly in recent years. For example, the location of mulberry silkworm production has shifted since 1984 from the Delta to Shaoguan, Zhaoqing, and Zhanjiang. The East Guangdong Corridor produces sugarcane, jute, and rubber. Sugarcane is grown primarily on the Cao-Shan plain and jute is grown mainly in the coastal areas of East Guangdong. Rubber production is increasing in East Guangdong, and recently has been supported by international loans. In the West Guangdong Corridor, the Leizhou Peninsula also produces sugarcane, as well as tropical and subtropical fruits: bananas, citrus, lichi and pineapple. Finally, the sparsely populated central and northern hilly lands of the North Guangdong Corridor are known for their pro- duction of tea leaves and tree crops. 1.25 Forest products are important to Guangdong and forestry was one of the fastest growing "agricultural" industries between 1980 and 1985. In 1985 it contributed 11 percent to the total value of agricultural output. Wood production increased to 4.3 million cubic meters (Mm3) in 1985, but then declined to 3.4 Mn3 in 1986, due to concerns about deforestation. About 33 percent of wood production comes from the North, 25 percent from the East, 22 percent from the West, and 20 percent from the Delta. The past spurt in forest activities was supported by international lending for forestry develop- ment in southeast China which included the upgrading of forest roads. How- ever, production now appears to be stabilizing at past levels and imports are taking up the growth in demand. 1.26 Interprovincial and Foreign Trade. Although Guangdong Province is a significant producer of grain, coal, iron ore, iron and steel, petroleum prod- ucts and cement, it is still a net importer of these commodities, and more than half of the inflows to the province are made up of these six products. In fact, 80 percent more freight is shipped into the province than out of it,  -9- as shown in Table 4. The outflows are much more diverse, and include many of the inflow categories.8/ 1.27 The Pearl River Delta is China's largest trade outlet. The new open-door policy of China, the creation of two SEZs in the Delta, and the recently discovered oil reserves of the South China Sea have expanded the horizon for the Delta's development in foreign trade. The strategic location of Hong Kong as a focal point for imports and exports also has reinforced Delta trade. 1.28 In 1986 Guangdong Province became the largest foreign export earner in the country. Between 1980 and 1985, the total value of exports increased by 42 percent. Between 1985 and 1987, it increased by 92 percent, to $5.56 billion, outperforming all other export centers including Shanghai, which previously had been the leader. Textiles accounted for the largest percentage of export value (38 percent), followed by agriculture and special products (33 percent), industrial and mineral products (19 percent), and processed agricultural products (10 percent). In terms of tonnage, sand and gravel were also a major export, reaching over 4 million tons (Mt) in 1987. Total export tonnage is estimated at approximately 18 Mt. The main export markets are the entrepots of Hong Kong and Macao, which absorbed 69 percent of the exports. Direct exports to the US, Japan, Singapore, West Germany, Great Britain, Canada and Australia also are growing in importance. The value of Hong Kong and Macao exports grew by 36 percent between 1985 and 1986, and 46 percent between 1986 and 1987, rapidly increasing their share of the total. 1.29 The major foreign trade flows into Guangdong Province in 1987 included the following commodities: chemical fertilizer, steel products, cement, grain, edible oil, lumber, and paper. Total foreign imports are esti- mated at about 16 Mt. The volume of foreign imports into Guangdong decreased slightly in 1987, but this trend is expected to reverse. In the future growth in imports is anticipated to be significantly slower than growth in exports. 1.30 Hong Kong is by far the largest foreign trading partner with Guangdong Province; it also serves as a conduit for transshipped goods from other countries. Exports to Hong Kong have grown from 3.7 Mt in 1980 to 10.8 Mt in 1987, at a rate of 16.8 percent per year. This is more than twice the growth of exports to other countries. Imports from Hong Kong were only half the level of exports in 1987 (5.4 Mt) but have been growing at a rate of 43 percent per year from a very small base in 1980 (see Table 5). Imports and exports by commodity group are given in Table 6. The tonnages for exports are dominated by sand and gravel (taken by river barge to Hong Kong), but exports of textiles, food and metal products also are significant, while most imports into Guangdong are manufactured goods. Hong Kong's influence over imports and exports appears to be growing. However, port limitations may reduce future flows through Hong Kong in favor of direct import and export routes. 8/ The reasons for this overlap need to be investigated; it could indicate irrational transport patterns for the commodities involved; however, it more likely reflects (1) transshipment of commodities through Guangdong Province, (2) locational differences within the province or (3) the grouping of heterogeneous products under one classification.  - 10 - E. Present Constraints on Development 1.31 The two greatest bottlenecks to economic expansion of Guangdong Province are the interrelated shortages of energy and transport. Hydropower development in the province has about reached its maximum, and access to coal, either in northern China or in eastern Guangdong, is limited by constraints in transport capacity; both the Guangzhou-Beijing Railway and Huangpu Port, which are major conduits for low-cost coal from Shanxi Province, have been satu- rated, while eastern Guangdong is served only by costly narrow-gauge railway cum river or road services. Power production in the province soon will be supplemented by a nuclear pLant near Shenzhen, and a major new coal-fired plant in the Pearl River De:Lta recently has been completed with its own berth- ing facilities for coal barges. Hydropower from Guangxi also may be available for Guangdong. Offshore oiL is available but plans for its use in energy production are unclear. 1.32 Transport constraints are an economic problem well beyond their effects on the availability of coal. Inadequate transport constrains the development of mineral based activities, particularly in the East Corridor where potential limestone/cement, iron and other minerals, in addition to coal, are located. In other areas inadequate transport constrains the growth of light manufacturing and high valued agricultural exports whose demands for express door-to-door services are being met only partially and at relatively high cost, due to a combination of small, inefficient trucks, generally poor road conditions and increasiLng congestion on the highway network. An impor- tant constraint was temporarily lifted when double tracking of the railway in the Northern Corridor was completed in late 1988, and another will be eased when the divided expressway from Guangzhou to Hong Kong is completed (cur- rently scheduled for 1992). However, transport constraints will remain perva- sive throughout most of the province, as expansion of the transport system has lagged far behind the explosive economic growth of recent years. F. Economic Forecasts 1.33 Before 1980, the economy of Guangdong Province lagged behind that of the rest of the country. S:ince then it has grown faster than the national economy. The accelerated growth in the province is due to many factors, including the national economic reform policy which has made much of Guangdong a special economic incentive zone. Readily available land and labor and close ties to domestic and J-oreign investment sources in more than 20 coun- tries also have contributed to the province's rapid economic growth. 1.34 Three types of growth patterns are emerging: (1) export-oriented, light-industry-led growth; (2) resource-based, heavy-industry growth, partly linked to the construction iLndustry serving (1); and (3) heavy-industry pri- mary processing based on imported capital and resources (e.g., iron and steel plants). The first type depends on access to foreign markets and capital, which for the near future w111 be oriented to Hong Kong. The focus of this growth will clearly be the IPearl River Delta, but the growth could spread to other areas of the Province. The second type is tied to resource locations and accessibility to type 1 growth areas. The North and West will continue to produce metal and nonmetal ores, but the growth for coal mining (for home heating and power generation), iron mining, steel-making and cement plants is expected to be in the Northeast. The third type is dependent on deep-water  - 11 - ports combined with a good foreign investment climate. These features can be found in locations such as Zhanjiang or the Zhuhai islands. Managers of new factories of all three types are planning to use modern technology with a relatively high labor content, but they have identifiably different transport needs. 1.35 Official goals for the Province include the quadrupling of GNP between 1980 and 1995 (a compound growth rate of 9.7 percent), and provincial forecasts made in 1985 showed GVAIO increasing at a rate of 9.2 percent up to 1990. Actual growth rates, however, were much higher up till 1989 (GNP growth averaged 14 percent and GVAIO 15 percent from 1980-87), suggesting that future growth could also be higher. Short term growth rates for GVAIO range from 15 percent to the extraordinary 27 percent achieved in 1987. This accelerat- ing trend was fueled by external investment, and the province has obtained agreements for a further $15 billion in foreign investment in the near future. However, efforts by the central government to rein in investment growth nationwide in order to control inflation commencing in late 1988 took hold during 1989, dramatically slowing the growth of income as well as inflation. Inflation now having been brought under control, the government is now acting to resume a high growth pattern. 1.36 In the medium term, energy and transport constraints will slow eco- nomic growth in Guangdong Province. The rapid growth has used transport capa- city faster than it could be expanded, even with the extraordinary measures being taken by provincial and local authorities (see Chapter III). Recent history shows that sustained growth by rapidly industrializing countries has not exceeded 20 percent over a 5-10 year period, even at their peak, and has generally fallen within 8-15 percent. However, Guangdong Province is better endowed than any of the "four tigers" in both human and natural resources, and it has the additional advantage of overseas investors of Chinese origin who have more than a passing interest in this specific location. Therefore, growth can be expected to be in the higher end of the range. 1.37 Overall, the dramatic growth of the provincial economy of recent years will soon run into major constraints. The present backlog of investment will lead to substantial growth in the near term in any case, but energy shortages, congested transportation corridors, and government policies to curb inflation can be expected to slow growth from the peak rates of 1986-88. Another element of uncertainty is the extent to which world markets will encourage or discourage rising Chinese exports. Recognizing these uncertain- ties, three possible scenarios of economic performance have been assumed to reflect pessimistic, base case and optimistic growth assumptions, as summa- rized in Table 1.3. 1.38 The pessimistic scenario assumes a fast drop off in investment due to worldwide recession and resistance to increased Chinese exports combined with major constraints on economic growth. The optimistic scenario assumes sustained growth at rates somewhat slower than in the past, but still higher than those achieved by other countries (including the "four tigers") over a sustained period. The base case represents a balance of optimistic and pessi- mistic factors, and provides the primary basis for the economic analysis of proposed investments in this report. These three scenarios are a first  - 12 - Table 1.3: FORECAST ECONOMIC GROWTH RATES (GVAIO) (average % per year) Medium term Long term 1990-95 1995-2000 Pessimistic 6 3 Base Case 9 6 Optimistic 15 12 approximation which is designed to cover the range of likely performance for transport planning purposes.9/ G. Location of Future Growth 1.39 The location of future growth will be influenced by the State and Provincial spatial development policies as well as by economic market forces. The government's policy is to try to balance growth among the regions and to support small city and town growth more than larger city growth. This policy will increase public investment in the least developed eastern and western sections of the province (particularly Meixian Prefecture), but the extent of private investment in these areas is conditioned upon the availability of transport, which is so far quite limited. Meanwhile a relatively high level of infrastructure and other investment will continue to be made in the Pearl River Delta based on initiatives from local authorities and the SEZs. This implies that high growth areas will continue to grow, and that other regions can expect to increase their growth to the extent that they become more closely integrated with the expanding markets and investment sources of the Pearl River Delta. Table 7 summarizes the counties and cities with the high- est growth potential for industrial development. 1.40 In the future, industrial growth in the province, particularly export-oriented industry based on electronics, food and textiles, is expected to increase faster than agricultural growth. Light industry investment is expected to accelerate with the introduction of new technology from Hong Kong and other sources, and may only be constrained by the development of infra- structure and limits on external financing imposed by the state. The major investments needed to support heavy industry, particularly coal and iron ore mining, cement and steel production are contingent on a number of regional 9/ These approximate forecast rates could be made more detailed in the future by using a dynamic model of the provincial economy based on a light industry-led growth pattern. The implications for different sec- tors of the economy could also be modeled to better forecast the mix of sectoral growth and the interactions between sectors. Such an analysis would also give a better basis for surplus/deficit forecasts of specific goods which is related to transportation demand in Chapter V.  - 13 - development and investment decisions which are interrelated with transport investment decisions.10/ 1.41 Agriculture is expected to experience much slower growth; grain production is actually expected to decrease as the cultivated area is reduced, and certain cash crops, sideline products, and higher value products provide the impetus for growth. Guangdong Province is expected to adopt certain preferential policies to stimulate grain production and fertilizer use. These policies should help to counteract the decrease in the land area devoted to grains. Emphasis also will be placed on the production of tropical fruits and vegetables, fish farming, and animal husbandry. The net result is expected to be an average growth in total output value of agriculture of approximately 3 percent per year, under all growth scenarios. 1.42 State funding for capital investment in industry has been reduced under the Seventh Five-Year Plan and the available funding concentrated in those industries with the greatest potential. More investments are also being made in infrastructure, including transport, power, and communications (see Table 8). However, the percentage of public investment in infrastructure was significantly lower in 1986/87 than in the Sixth Five-Year Plan period (1981-85). Policy statements from Chinese leaders suggest that transportation and energy are likely to be given higher priority during the Eighth Five-Year Plan (1991-95). 1.43 The population of Guangdong Province is expected to grow faster than the national average through the year 2000, and some migration into the prov- ince is expected from other, slower-growing provinces. The populations of the fast growing SEZs contain a high proportion of temporary workers. This pro- portion is expected to remain in effect until a government settlement policy is adopted to increase the number of permanent inhabitants of these areas. Meanwhile the population is projected to grow at 1.4 percent for the province as a whole, and internal migration should account for the shifts to the high growth areas. These population shifts are expected to affect the future demand for passenger trips. 1.44 The percentage of growth (see Table 9) that is attracted to each region or transport corridor in the Province is expected to shift, depending on the accessibility of the regions to Hong Kong. This shift cannot be pre- dicted with much accuracy at present due to the lack of information about how investment location decisions are made in the Province. However, the general direction of the shift can be estimated. Table 10 shows the Base Case fore- cast growth in GVAIO by Corridor, and the shift in share of growth as the expressways linking the Delta to Hong Kong are completed, and the West Bank and Guangzhou areas gain accessibility similar to that of the East Bank. This increased access to Hong Kong is expected to concentrate more growth in the Delta area unless other areas became more accessible. Optimistic and pessi- mistic scenarios are given in Tables 11 and 12, respectively. In addition to this shift in growth, a net increase in development may take place. The potential for this type of increase is described in Chapter V. 10/ Investment in these industries must be evaluated in conjunction with transport investments as a combined package for regional development. This is closely related to the calculation of benefits for induced traf- fic as discussed in Chapter V.  - 14 - II. TRANSPORT INFRASTRUCTURE, SERVICES, AND MANAGEMENT A. Transport Network Overview 2.1 The transport system of Guangdong Province consists of railways, roads, coastal shipping routes, inland waterways, civil aviation, and a pipe- line (Map IBRD 21438). Roads and inland waterways are the principal modes for intraprovincial or short-haul transport. Railways and coastal shipping are the principal modes for interprovincial transport; air movement of passengers is growing rapidly, but is still quite limited, and air freight is almost nonexistent. The length of internal provincial routes is shown in Table 13. Although the density of the road network and inland waterways in Guangdong is higher than average for the country, the province still has a relatively undeveloped transport system. This is because the transport density through- out China still is quite low, compared even to other developing countries such as India and Brazil. However, this situation is beginning to change as the government attempts to expand facilities to meet the demands created by rapid economic growth. Examples of these improvements include more expressways and Class I highways to strengt:hen the road system, linking of the rail system and increasing the capacity of main lines, expanding ports in several locations, and a number of new or expanded airports. 2.2 The spatial distribution of infrastructure in the province varies according to the level of economic development of the different areas. In general, however, the transport routes form four axes centering on Guangzhou. In shape they resemble two wings with a head and tail (Map IBRD 21438). This pattern is influenced by the prevalence of hilly and mountainous areas in between the north and east axes and north and west axes. 2.3 The major handicaps of the Guangdong transport system are its capa- city limits, reliance on outmoded technologies, high costs, and geographic sparsity. The existing ini-rastructure is overloaded because transport supply has not kept up with the rapid growth in demand. For example, the actual throughput of the Guangzhou North rail marshalling yard in 1985 exceeded its rated design capacity by 23 percent. Similar overcapacity conditions have been recorded for other ra:Llway facilities and for ports. Roads also are starting to experience significant congestion in the faster-growing areas. Moreover, many rural roads are far below the standards which would be economi- cally indicated by the traiFfic which they serve. Capacity limits and outmoded transfer and cargo handling technologies create bottlenecks that affect many parts of the transport system. 2.4 Much of the technology used for road, rail, and river transport resembles that used by developed countries in the 1940s and early 1950s. Although new rolling stock and road vehicle technology are being introduced in China, not all of it is weLl planned. For example, the recent introduction of modern heavy trucks is cauSing rapid deterioration of pavements which are too thin to support the heavier loadings. On the positive side, facilities at Huangpu Port and some of the smaller ports in the Pearl River Delta are being equipped with more modern handling equipment.  - 15 - B. Transportation Infrastructure by Mode 2.5 Railways. Compared to many provinces and regions, the length of railways in Guangdong is relatively short (see Table 14). The capacity of stations and yards is also limited relative to the demand, although the recent double-tracking and expansion of rail yards on the Henyang-Guangzhou and Guangzhou-Shenzhen sections will provide some temporary respite. The existing rail infrastructure is composed of five main lines, one junction station with marshalling yards,l/ and other major stations. The five main lines are equipped with semiautomatic block signaling. The railways connect the ports of Huangpu, Guangzhou, and Zhanjiang with the rest of the country, and also serve as an export link to Hong Kong. 2.6 Although Guangdong has nine local rail lines, the only significant one is the Meilong railway in the Eastern Corridor. This railroad was built to serve a coal mine at Shiwangjhang in 1971, but has since expanded its ser- vice to take general freight and passengers. It serves as the main link from the coal mine to river transshipment points (Map IBRD 20593). 2.7 Roads. Roads play an important role in Guangdong transportation, although most are of low technical standards, have poor surface conditions, and are often congested (Table 15). The average road density in the province is 29.7 km/100 km although plains areas have a higher density than hilly areas (Table 16). In recent years, however, road capacity has not expanded in line with the dramatic growth of traffic, and congestion has greatly reduced speeds and consequently increased vehicle operating costs. Congestion is par- ticularly high on many road sections which provide entrance and exit to cities and towns. The mixed use of roads by motor vehicles, tractors, bicycles and pedestrians without separation also contributes heavily to congestion when the different types of traffic interfere with each other, which is common where urban buildup has encroached on many roads. At present most trunk roads pass through cities and towns without separation from the streets of built-up areas, forming traffic bottlenecks, particularly in the Pearl River Delta. Outside of the Delta, poor road surface and geometric conditions are mostly responsible for the high transport costs. It is significant that over 2,300 km of existing gravel roads have traffic loadings in excess of 2,000 vehicles per day (vpd). This is at least four times the threshold at which paving becomes economic. 2.8 Ongoing improvements to road infrastructure in Guangdong include the construction of expressways, the upgrading of Class III and IV highways to Class II or Class I standards, and the paving of waterbound macadam roads with bituminous or cement surfaces. The major Guangzhou-Hong Kong expressway proj- ect is an innovative arrangement financed by a foreign bank consortium orga- 1/ Jiangchuan marshalling yard in Guangzhou is a newly completed second yard serving the junction, and Pingshikou is the main yard serving Hong Kong traffic.  - 16 - nized in Hong Kong. It is to be a six-lane expressway throughout its length where the traffic has already reached 20,000 vpd day on the existing road.2/ 2.9 Coastal and Marine. Guangdong's long coastline and favorable condi- tions for coastal and marine shipping are complemented by 13 seaports, four of which have been traditionally classified as major ports and three other SEZ ports have assumed a significant role in the Province (Table 17). These ports serve intraprovincial, interprovincial and international transport. Except for the deep water port at Zhanjiang and the new port being developed at Yantian (para. 2.11), the capacity of the major seaports is limited by the depth of approach channels (Huangpu and Guangzhou) or the water depth along side berths (Shantou). 2.10 Efforts have been made to improve facilities for handling, loading and unloading at ports. However, these efforts are falling behind the growth of traffic and ships frequently queue up to wait for berths. In Huangpu Port, ships waited an average of 5.8 days in 1986, causing a large compensation cost for demurrage. This problem will be eliminated temporarily when the second World Bank-financed expansion of Huangpu is completed, but growth in cargo flows again will strain capacity in the near future, especially if coal ship- ping increases. 2.11 Several new ports are being built to expand port capacity for the SEZs. The ports of Shekou, Chiwan and Jiouzhou (Zhuhai) were built over the past few years to satisfy traffic growth in Shenzhen and Zhuhai. Although these ports will satisfy demand in the short term, medium range growth will require further capacity. Funds have also been committed for projects to upgrade Shantou harbor and construct a new port at Yantian (Daqing Bay) near the Shenzhen SEZ and Hong Kong. The port at Yantian could become a major transshipment terminal in the region for coal and containers; Yantian has 6 km of shoreline available for development with depths of 10-14 m. The first phase is being financed by Japanese loans and will include container berths and a capacity of 2.8 million tons (200,000 TEU) per year. It will have a rail connection to Shenzhen and 650 m of sidings. However, the future role of this port should be planned taking account of ongoing port developments in Hong Kong. 2.12 Inland Waterways. Use of the waterway system in Guangdong Province has not grown rapidly, despite the system's 11,000 km of navigable inland waterways, 4,200 km, of which have water depths of more than one meter. Table 18 shows the depth of navigable sections and maximum ship size for each major river, the number of ports and their throughput. Exploitation of the water 2/ The toll level for the expressway on its first leg from Shenzhen to Guangzhou is expected to be set at the price of a rail passenger fare (currently about Y 20) for passenger cars and 2-3 times higher for trucks. This is substantially higher than the bridge tolls (Y 2 per car) in Guangdong and tolls on roads in other countries. However, the faci- lity will undoubtedly attract a major portion of traffic due to conges- tion and poor conditions on the parallel modes. The committed road improvement projects or projects under construction in 1988 are shown in Map IBRD 21438.  - 17 - resource for irrigation in recent years has reduced the length of navigable routes.3/ 2.13 Construction and improvement of inland waterway infrastructure has not received priority attention, and the condition of the water channels varies from good to poor, which limits long haul shipping, particularly up- stream. However, there is a major project to improve the Xijiang in a criti- cal 395 km stretch which will allow it to accommodate 1,000 dwt barges throughout its 1,200 km length into the center of Guangxi Province; this proj- ect is expected to be completed in 1990. Many river ports have not been equipped with modern handling equipment and loading and unloading often is done by manual labor. The technology has been upgraded in certain main river ports in the Pearl River Delta, but has severely limited road to river trans- shipment elsewhere in the province. 2.14 Air. Airports in the province serve civilian and military uses. Guangzhou (Baiyun), Zhanjiang, Shantou, Shaoguan (Guitou) Xingling and the newly opened Meixian airports are purely for civilian use (see Table 19). Foshan (Shati) and Huiyang airports are used for both military and civilian purposes.4/ In 1987, the Province had 54 intraprovincial, interprovincial, and international air route links. Because of the rapid increase in traffic, most airport terminals are overloaded. The small capacity of provincial air- port terminals limits the development of passenger services; the air traffic management system also needs to be modernized. Shenzhen, which has a large potential passenger demand, is seeking a proper site for a new airport. 2.15 Only one pipeline exists in the Province. It is used to transport crude oil from the oil berth of Zhanjiang Port to the refinery at Maoming (114 km). Oil products from the refinery are shipped out by rail and road. C. Transportation Services 2.16 Rail Services. Railroads provide a variety of services to passen- gers including express, fast, and slow trains with different types of coach cars: hard sitting coach, hard sleeping car, soft sitting coach, and soft sleeping car. Each train normally contains 13 to 20 coaches, although in the peak season more coaches are added and express trains sometimes pull as many as 25 coaches.5/ The present services are shown in detail in Table 20. 2.17 Fares charged by the Guangzhou-Beijing railroad were maintained at low levels from the 1950s until 1985, when they were increased for short trips (up to 100 km) by 37 percent. In September 1989, railway fares nationally 3/ For example, the navigable length of the Han River in the Shantou Region has decreased from 405 to 338 km, and the low water season is now four months long. 4/ Foshan military airport was opened in 1987 to serve civil domestic and long haul flights. 5/ The number of coaches generally is limited by the length of the platform. To receive better service, many small stations need to extend their plat- forms.  - 18 - were increased for short trips by 60.8 percent, and by 120 percent for trips over 100 km for most classes of service; at the same time, the premium paid by foreign nationals was reduced from 101.25 percent to 70 percent over domestic prices. 2.18 Local railways are permitted to charge higher fares to recover capi- tal costs. In the case of the Guangzhou-Shenzhen line, it is currently 50 percent above national fares. 2.19 The Beijing-Guangzhou railway is one of the two most important pas- senger routes in the country; it carries the heaviest passenger traffic. Dur- ing the holiday season in the spring, passenger trains are overloaded by at least 50 percent. Priority in time scheduling and seat reservation is given to long haul services, and some long haul trains do not serve short haul pas- sengers. The supply of sleeper coaches is far less than the demand, since the Ministry of Railways limits the number of sleeper coaches in order to maximize train capacity. Passenger traffic recently has grown along with the expansion of train sizes, and the double! tracking project completed in 1988 allowed pas- senger traffic to expand even more. 2.20 The railway provides services for the following kinds of freight: (a) Large volume freight allocated by the central government and the provincial authorities. This includes coal, oil, grain, iron and steel, ore, and cement. (b) Freight for foreign trade, through the ports of Guangdong, and freight for trade with Hong Kong. (c) General cargo originating from and received by enterprises inside and outside the province. 2.21 There are several types of freight wagons in service: covered, open, plain, tank and freezer. Different freight shipping requirements are met by wagons with capacitiies from 25 to 60 tons and by using containers in three sizes, 1, 5, and 10 tons. Table 21 presents a list of the available rolling stock in Guangdong Province. The table shows an increase of 33 per- cent in locomotives since 1978, a 55 percent increase in freight wagons, and a more than 100 percent increase in passenger coaches. Wagon utilization (or productivity) has also increased significantly in this period. 2.22 The railway serves predominantly long haul and medium haul traffic and discourages short haul traffic (less than 200 km). However, large or abnormally shaped freight is still handled by the railway even for short hauls. Rail transport has a much longer average haul distance (346 km) than roads and inland waterways. Nationally, several long haul rail routes run parallel to coastal shipping routes, but provide more direct services between inland origins and destinations. 2.23 Because demand exceeds supply on the railway, the system has adopted strict operating rules which allocate wagons to customers with lead times of 20 to 50 days. Rail freight: planners also direct nonpriority (out-of-plan) freight onto more circuitous routes to take advantage of rail capacity on parallel lines. This systerm has increased rail capacity at the expense of  - 19 - poor service standards, thereby inhibiting the growth of industries which demand higher transportation service standards. 2.24 Road Services. Guangdong Province has three major types of special- ized road transport enterprises: state-owned, collective, and privately- owned. In addition, a large number of nontransport enterprises and institu- tions own vehicle fleets to meet their own needs, and many also offer spare capacity for hire. Diversified passenger and freight services are offered, including day and night transport, nonstop and air-conditioned transport, and combined services with waterway transport. The major bus routes connect Guangzhou with Zhanjiang, Shantou, Meixian, Shenzhen, Zhuhai, Zhaoqing, and the cities and towns of the Pearl River Delta (see Table 22). Intercity bus service to other areas is limited (e.g., no service presently exists between Guangzhou and Shaoguan). 2.25 The supply and quality of passenger services is being improved. For example, in order to avoid daytime congestion on roads, many firms provide nighttime express services. The frequency is determined by the demand; on some high-traffic routes there is a scheduled run every 15 minutes. The state-owned bus companies provide scheduled services to most regions that are accessible by road. The fares of state-owned companies are kept to the mini- mum required to cover financial costs. Road passenger transport generally is preferred in rural areas where rail service is not available. In parts of the province where there is high mobility, an active market economy, and a high demand for quality, many collective- and privately-owned firms compete with each other to provide passenger services. 2.26 Numerous collective and privately-owned transport enterprises pro- vide road freight shipping services in Guangdong, and their number is increas- ing, as is also the number of own-account carriers. The state-owned, specia- lized freight transport companies (which are organized and managed at the pro- vincial, city, and county levels) played a central role in road transportation before the economic reforms in 1978, but since then their market share has steadily declined, amounting to only 5 percent by 1987. There are two main reasons for the decline in state-owned freight transportation. First, many new collective and private firms are entering the market; they have greater pricing freedom and operating flexibility and provide more flexible, reliable, and higher quality services.6/ Second, truck ownership by nontransport enterprises has been increasing rapidly, and part of the freight that was formerly shipped by specialized companies now is transported by the enter- prises themselves. 2.27 The development of road services and higher rail tariffs in 1985 shifted short-haul freight away from railways to roads. Now increasing con- 61 All types of freight transport enterprises in Guangdong have pricing flexibility up to a maximum regulated price (currently 23 fen per tkm plus Y 1/ton for normal cargo). Nonstate-owned enterprises sometimes charge more, but that is illegal. The high overhead costs of state-owned firms (including heavy pension obligations) reduces their pricing flexi- bility, and some experience problems with obtaining rationed gasoline and dealing with limitations on overtime and other operating rules that reduce their available operating hours and increase empty backhauls.  - 20 - gestion on roads is increasingly restraining traffic growth, especially on national highways. However, road transport is particularly suited to light industry and export agriculture such as that of Guangdong, which demand fast delivery, careful loading and unloading, and door-to-door transport services. Therefore there is a large and increasing market for these road services. 2.28 Tractors equippedl for local freight haulage and bicycles make up a substantial portion of roacl traffic in rural areas and towns. They can be an efficient means for short-distance local and farm-to-market transport, but they do absorb much scarce road capacity. Highway design with allowance for these vehicles is important. on secondary and urban roads. 2.29 Road Vehicle Fleet. In the past, the capacity of the road network was more than enough to meet demand, and the network did not begin to be fully used until vehicle ownership surged with an increase of 150 percent during the 1980-85 period of import liberalization (see Table 23). Over 1986/87, the passenger fleet increased more slowly than the truck fleet (9.2 percent com- pared to 12.4 percent per year). However, both have increased much more slowly than road traffic in the province. This implies that the average annual usage of vehicles is rising at a significant rate. The increase in accessibility due to bridge! construction in the Pearl River Delta could be stimulating fleet growth arnd vehicle usage, as some of the highest growth in traffic has occurred in are!as served by new bridges. The average distance of road freight haulage is increasing, especially for goods arriving from and being shipped to Hong Kong. This distance is expected to increase even more when expressways are constructed. In the short term, however, the effects of congestion will tend to counteract this trend. 2.30 The rate of motorization in Guangdong Province is higher than the average for China or Korea--42 trucks and passenger vehicles per 1,000 resi- dents, compared to 34 and 31 vehicles, respectively. However, it is still much lower than in some other developing countries such as Brazil, or in Japan (84 and 394 vehicles, respectively). The motorization rate for passenger vehicles is low partly because of the use of bicycles for short trips and vans for commuter travel. The motorization rate for trucks is low because trucks are not used for long distance shipments, and inefficient vehicles and poor roads have made trucking less attractive for short haul loads. In addition, import restrictions and foreign exchange restrictions have severely con- strained the growth of the road vehicle fleet. A shift toward larger, more efficient trucks and the use of container trucks has been encouraged over the last seven years, but large trucks are still not widespread, due in part to inadequate resources to improve road geometric conditions, constraining maneu- verability. 2.31 Inland Waterway Services. Inland river cargo services outside the Pearl River Delta are supplied mainly by small barges (20-110 tons) and pas- senger vessels for the rural population that lives near the river. Passenger services cover the West River, the Pearl River, parts of the North and East Rivers, and the coastal routes. Due to the slowness of river travel, the market for most routes is limited to low-valued bulk commodities or passenger trips which are hard to make by other modes, e.g., for passengers carrying large amounts of luggage or goods; available statistics are shown in Table 24. In some areas the river is the only mode of transport available; in other areas the rapid expansion of road services has taken over the market for  - 21 - waterway transport. In the Pearl River Delta larger barges (1,000 to 3,000 tons) and barge trains with tugs are used. These services have expanded dra- matically with new river ports serving Hong Kong from the West Bank using hydrofoils and container barges, as well as traditional vessels. Approxi- mately one third of inland water services are provided by privately owned and two-thirds by cooperative or state-owned vessels. The proportion of private service is higher in the Delta than other areas and joint venture firms with Hong Kong shipping companies are expanding. 2.32 There are few cases of congestion along the navigable rivers in Guangdong Province. However, inland waterway traffic is impeded by congested, inefficient ports. Due to outmoded technology and the limitations of port capacity, state-owned ships achieve a utilization rate only of about 20 per- cent. Most of these ships have outdated designs and are too slow and small for efficient transport, so they are sidelined much of the time. Some modern barge technology has been introduced in the Pearl River Delta, but not yet on the tributaries. (This may change with deepening of the Xijiang channel.) 2.33 The average cost of inland waterway transport increased between 1980 and 1985, as the prices of raw materials and fuel rose. A tariff increase was granted in 1985 but some enterprises, particularly the collective ones, still have suffered heavy losses. Consequently, the inland waterway shipping busi- ness has declined outside of the Pearl River Delta, where it appears to be profitable. 2.34 Coastal and Ocean Shipping Services. The major seaports in Guangdong Province provide different types of freight and passenger services (Table 25). Huangpu and Zhanjiang ports serve both international trade and domestic coastal shipping routes. Huangpu Port, with the largest volume, serves most of Southern China. Zhanjiang Port serves the southwestern prov- inces. Guangzhou Port serves the Guangzhou region and provides transfer ser- vices for inland water freight on the tributaries of the Pearl River. The smaller ports mostly handle domestic cargo; their foreign trade generally is transshipped via the larger ports. According to port statistics (see Table 26), foreign trade accounted for 38 percent of the major port traffic in 1985. Foreign trade dropped to 30 percent in 1987 as domestic cargo grew more rapidly. 2.35 The limited capacity of ports and outmoded vessel designs have pre- vented full exploitation of the good coastal shipping conditions. The prob- lems in freight service are manifested by the long average waiting times for vessels, particularly in Huangpu (6.8 days in 1986) and Shantou (7.6 days in 1986), and the high congestion costs incurred by shippers. Long travel times and delays also discourage passenger travel by ship, and as incomes have risen, fewer passengers have opted for coastal waterway transport. 2.36 Air Services. There are five international air routes through Guangzhou, 44 routes connecting Guangzhou with other large cities, and three intraprovincial routes covering Guangzhou-Zhanjiang, Guangzhou-Shantou, and Guangzhou-Meixian. The types of service and equipment used are described in Table 27. Air passenger traffic in the province has increased from an extremely small base at about 20 percent a year since 1978. The large reser- voir of suppressed demand and continuing growth of international trade and tourism provide a ready market for further expansion of air passenger trans-  - 22 - port. Recent efforts to encourage independent airline companies with compet- ing services have not been very successful, although there are prospects for future development. The developments planned could help to expand aviation services in the province and in China as a whole. Air services for freight transport are almost nonexistent in Guangdong Province, as in the rest of China. The present business is limited to special types of freight such as airmail and emergency service. 2.37 Containerization. There are 12 ports and four railway stations handling international standard containers. Container traffic, though still low, has recently begun to grow rapidly. In 1987 it reached 420,000 TEUs, compared with 1,800 TEUs in 1979. 200,000 TEUs of the 1987 container traffic were handled by waterway, 173,000 by highway, and 47,000 by railway. Conl- tainer penetration rates were approximately 6 percent of containerizable freight for waterways and much less for the other modes. The present level of container penetration in China's foreign trade (despite recent rapid growth) is the lowest of any of the world's major trading countries. Such low levels of adoption of containers (or other forms of cargo unitization, such as through pallets) can be explained partly by physical deficiencies in the infrastructure, particularly in railway and highway facilities connecting ports with their hinterlands. Institutional factors are also important, in particular the lack of competition in foreign shipping services, which are dominated by the China Overseas Shipping Corporation (COSCO), essentially a government monopoly, and also by similar circumstances in the railways. Freight stations and container transshipment depots are now established in central cities, and door-to-cloor service is beginning to develop despite the institutional constraints. 2.38 Container vessels used in Guangdong include full container carriers, roll-on, roll-off (Ro-Ro) ships, and container barges. The province has 15 full container carriers for ocean shipping totaling 202,000 dwt, four Ro-Ro ships totaling 54,000 dwt, and more than 50 container barges totaling 30,000 dwt. Guangzhou port now serves regular routes to Japan, Europe and North America. Three quarter-s of the waterborne container traffic of the province is destined to or fr-om Hong Kong. 2.39 The Huangpu contairLer terminal of Guangzhou port is one of the largest in the country with two berths and a designed annual capacity of 200,000 TEUs. Second and third generation container carriers can be berthed without lightering. The actual throughput of container traffic was 57,500 TEUs in 1987, representing an approximate container penetration rate of 10 percent of general cargo. The potential for containerization at Huangpu is estimated to be between 40 percent and 60 percent of general cargo, or 4-6 times the present proportion. 2.40 Guangzhou has four rail stations, including Xiayuan, which handle international containers and deliver these boxes to the major cities of China. Of the 47,000 TEUs dispatched or received by rail, Xiayuan station accounted for 40,000 TEUs. However, about 95 percent of the present containers were opened at the station and only 5 percent were transported door-to-door. This is a major obstacle to the growth of intermodal services. 2.41 In 1987, 173,000 TEhUs were shipped on highways in the province. This represented an increase of 10 percent over the previous year. The devel-  - 23 - opment of door-to-door highway container services has helped relieve conges- tion at ports and stations caused by intraport stripping and stuffing. It also has created favorable conditions for door-to-door service between Guangzhou, Shenzhen and Hong Kong, without passing through a port. These services are now growing fast and can be expected to increase more rapidly than other freight services in the province. D. Management of Transport Modes 2.42 Transport modes in Guangdong Province are managed either by the State or by the Provincial Department of Communications (DOC). Infrastructure owned and operated by the State include railways, pipelines, ports, and air- ports, and these may be operated by an authority at either the national or local level. Modes managed by the Provincial Department of Communications include roads and waterways. The multiplicity of transport enterprises serv- ing these two modes--state-owned, collective, private-owned, and co-owned (mixed public and private)--are regulated by several levels of government. The large state-owned enterprises and institutions responsible for central railways, maritime shipping, and air transport were formerly managed directly by the corresponding central ministries. However, recent reforms have split off the operating enterprises from the government ministries. These reforms apply to the rail ways and some major ports and are intended to provide more autonomy to the operating enterprise, improve efficiency, and strengthen financial accountability. Local railway and coastal shipping services are organized as licensed local enterprises under the jurisdiction of the local authorities. The organization of both the central and provincial transport agencies in Guangdong Province is given in Figure 1. 2.43 The financing of infrastructure investment and fleet purchase varies from mode to mode. In general, the financial responsibility for highway infrastructure rests with the provincial and local governments; limited finan- cial contributions are made by the central government. Limited subsidies for construction of local railways and port facilities are provided by the central and Provincial authorities. The construction of small port facilities by individual enterprises for their own use also is encouraged. 2.44 Railways. The management of the central railways has recently been removed from the central ministry and decentralized into 12 administrations. Two of the administrations operate in Guangdong Province: Guangzhou and Liuzhou. Guangzhou handles the Beijing-Guangzhou-Shenzhen line and Liuzhou handles the Litang-Zhanjiang line. The development of local railways also is being encouraged, and they are now growing in Guangdong as in other parts of China. The local railways serve primarily as extensions of the existing cen- tral railways, rather than as competitors. The latest local rail extension projects--Sanshui-Maoming and Guangzhou-Meixian-Shantou--are being organized as :Local railway companies rather than traditional administrations. They are given separate financial management and operating budgets, which are designed, with the consent of the central government, to cover operating costs and allow tariffs to be set high enough to repay capital investment funds borrowed for construction. 2.45 Rail freight is treated in two major groups: in-plan and out-of- plan. In-plan freight is allocated to specific lines by the central Ministry of Railroads with the help of the Provincial Railway Administrations. Out-of-  - 24 - plan freight is treated as optional freight that the railway can attract. by offering its service in competition with other modes. At present most freight is allocated, but the proportion will be decreasing so that most freight except coal is expected to be out-of-plan in the near future. Within the in- plan freight the central government has set the following priorities: (1) Coal and Petroleum Products; (2) Foreign Trade; (3) Material Supporting Agriculture (by season); ancd (4) Perishable Fruits, etc. Out-of-plan freight at this time consists primarily of nonpriority manufactures and sideline agri- cultural products. 2.46 Provincial rail managers organize traffic by railroad section and allocate their rolling stock, each month to the shippers who have requested service. The managers must first allocate service to coal designated by the central government (about 30 percent of capacity); then they try to meet guideline quotas for total freight, total coal, and passengers. Any capacity that is left over can be used for out-of-plan freight and may be charged at higher prices. Certain managers, such as Guangzhou South Station managers, use computerized freight load planning and actively contact shippers for addi- tional freight when they have excess capacity. Guangzhou South also has opened up subsidiary offices in freight generating centers away from the rail line at Shantou, Zhongshan, Jiangmen, Shende and Zhuhai. 2.47 Provincial rail managers have no control over the general availabil- ity of rolling stock; this :is controlled at the central level for the trunk rail lines. Rolling stock :Is considered to be rented by day to the operating administration or company, although local railroad companies own their own locomotives and passenger coaches. The wait for freight wagons averages 25 days for in-plan traffic and 3-15 days for out-of-plan traffic, depending on the month. Less-than-carload (LCL) traffic waited longer than carload in-plan traffic, but out-of-plan LCI. traffic (which is fitted into space available) had shorter waiting times than carload lots. 2.48 Ports. As a resu:Lt of the recent reforms, the four major ports in Guangdong--Huangpu, Guangzhou, Zhanjiang, and Shantou--are now managed by their respective municipalities; Huangpu and Guangzhou were in 1988 consoli- dated under one administration. The smaller ports have also been made the responsibility of local authorities, but they are subject to some regulation by the Provincial Department of Communications. The recent reforms are based on the principles that all costs of port operation and maintenance will be covered by operating revenues, that most new capital costs will be financed from revenues, and that ports should compete for cargo by keeping their costs low. Prior to the reforms, the state (through the provincial authorities) set tariffs, allocated all cargo, provided grant financing for major investments, covered any deficits, and received all surpluses generated by the ports. 2.49 Under the reforms, provincial ports operate similarly to railroads in that they are required to first handle in-plan cargo. However, presently about 20 percent of the total cargo is out-of-plan, and tariffs for this cargo are allowed to be set competitively within a narrow range. Profits are now retained by the ports and are subject to taxes. These new arrangements vary in scope from port to port, reflecting a deliberate experimentation with dif- ferent models. The tax arrangements tend to be ad hoc (usually fixed lump sum payments) and are apparently based on the needs of individual ports to retain surpluses. In the future, the government plans to progressively reduce the  - 25 - controlled allocation of traffic, reform tariffs and taxation policies, and eliminate grant financing of investments for ports. 2.50 Port authorities can apply for equipment loans if needed. For port expansion a feasibility study usually is prepared; this study allows the authorities to apply for a combination loan and government subsidy (the sub- sidy normally applies to the underwater structures). Investments under Y 10 million can be approved by local authorities. Provincial-level approval is required for projects costing from Y 10 to Y 200 million (higher if no other sectors such as energy are involved), and central government approval for projects over Y 200 million. 2.51 The main objectives of the decentralization program are to develop the capability and initiative of local port management, and to increase coor- dination between port development and the development plans of the region served by a port. Further, it is expected that the need to be financially self-sufficient and to compete for cargo will encourage efficiency in the ports subsector. Comments by Chinese leaders have indicated satisfaction with the program in general and, while it is still too early to evaluate the dif- ferent models adopted (which favor certain ports), it appears that future policy will tend to be more even-handed and uniform in its treatment of vari- ous ports. Key policy elements will relate to taxation, tariffs, and the financing of investments.7/ 2.52 Highways. Regulation of the road transport industry is the respon- sibility of the Provincial DOC, which controls licensing of vehicles and drivers. Bus routes are regulated by DOC for intercity routes and county leve:L authorities for local routes. The road freight industry has largely been deregulated, allowing essentially free entry and pricing. However, bus passenger transport is still closely regulated by the local Bureau of Communi- cations or the Provincial Department of Communications (DOC). Licenses are issued by the Commerce Bureau but approval is based on market demand and the technology of the firm, and no firm is allowed to monopolize a service if others are interested in providing it. The vehicle type and mechanical skills of the personnel are examined as part of the firm's technology. 2.53 The provision of highway infrastructure is primarily the responsi- bility of provincial and local governments. The central Ministry of Communi- cations is responsible only for establishing overall design standards and for making a minor financial contribution for the national highways. In Guangdong Province an itnovation has been introduced through an agreement by foreign financiers to finance 100 percent of the new Guangzhou-Hong Kong Expressway (para. 2.19). Financing costs will be recovered by operating the road as a toll concession. Other toll roads under construction in Guangdong are being financed by the provincial and local governments, and the provincial govern- 7/ In order to develop efficient policies in these areas, information on port operations, cost and finances must be improved. To support the Government's initiatives in this area, the Bank is assisting MOC in con- ducting research on alternative organizational forms employed in other countries and the development of port costing and management information systems.  - 26 - ment has constructed five major toll bridges in the province using funding from foreign sources. 2.54 In Dongguan, local authorities are also experimenting with a new measure to raise funds for r-oad construction. The measure involves the crea- tion of an Energy and CommurLications Company empowered to raise money from state, collective, and privelte sources to fund road and power plant construc- tion. The company pays interest on these funds (14 percent per year) and will eventually repay the capital through user fees and tolls. The company in the experiment plans to upgrade 16 principal county roads over the next eight years, and is well on its WELy to collecting the Y 6.33 million in investment funds needed for the road pr-ojects. 2.55 Waterways. Recent. deregulation of the waterway cargo industry by DOC has increased competition in this mode. Apparently, market entry is not difficult, but there are some regulatory requirements that must be met by collectives and the larger private transporters, particularly for safety pur- poses. Most cooperatives that apply are well-prepared and have no problem with registering their service. Some own-account vessels have entered the industry unofficially but the extent of this practice is not documented. 2.56 The Guangzhou Ocean Shipping Company (a branch of China Overseas Shipping Company (COSCO)) and Guangzhou Maritime Shipping Bureau both operate under the Ministry of Communications. The former is the predominant supplier of ocean shipping services. The latter is the major supplier of coastal ship- ping services. Foreign shipping companies also serve the ocean ports. Compe- tition among public carriers in the ocean and coastal trade is limited, but enterprises are permitted tco own and operate their own vessels. Joint venture shipping companies have been established in recent years between Hong Kong shippers and the Special Economic Zone bureaus. Groups of communities can also owin ships as cooperatives and make investments in small port handling equipment. 2.57 Aviation. Most airport infrastructure and aviation services are still integrated under the China Civil Aviation Administration (CAAC), a gov- ernment monopoly. However, the military is now operating some airports and China United Airlines, a military subsidiary, is operating charter flights from military fields. CAAC is responsible for maintaining civilian airfields and the Air Force for maintaining military fields. 2.58 CAAC as an airline! is highly profitable. Its total after tax profit in 1986 was Y 604 million, 21 percent of operating income. Its profitability is partly due to high fares (see Chapter IV). In 1986 domestic revenues increased by 36 percent a year and international revenues increased by 42 per- cent. Costs at this time were climbing only slightly faster (48 percent). Even with substantially increased service levels CAAC seems likely to remain profitable because of its monopoly position and high fare levels. 2.59 Maintenance of Transport Modes. Maintenance of infrastructure is treated differently for different modes. Where the mode is both owned and managed by the state, infrastructure maintenance is the responsibility of the operating authority. When the mode serves collective and privately-owned transport, maintenance is handled by the Provincial Department or Local Bureau of Communications. Data on maintenance expenditures by railroads are included  - 27 - in general operating data and are difficult to separate out. Data on port maintenance were kept by the Provincial Department of Communications until 1986, when local authorities took over the responsibility for port maintenance in Guangdong Province. Recent data are now kept by the individual port authorities. Road maintenance data are the best known of all infrastructure maintenance data, as budgets are made and research is available for this mode. Waterway maintenance is a much more sporadic activity and is treated as a capital expense. 2.60 The Guangdong Department of Communications spends approximately Y 300 million per year to maintain 18,000 km of national and provincial roads and to finance new road construction. This revenue comes from three sources: (a) a Highway Maintenance Fee from taxes on road transport services, which, despite its name, 30 percent is applied for new construction, and 70 percent for maintenance; (b) the Provincial budget for the Department; and (c) the MOC contribution from motor vehicle tax revenue (purchase sur- charge, 10 percent for domestic and 15 percent for imported vehi- cles). Some of these funds are passed on to local authorities for the maintenance of local roads. These funds are supplemented by bank loans and foreign sources which together yield another Y 100 million per year. The DOC considers these funds to be adequate for minimum maintenance but not sufficient support for new construction.8/ Local authorities also collect approximately Y 100 mil- lion per year for road maintenance and construction from taxes on motorcycles and tractors. In addition, they receive a subsidy from the Provincial budget, which accounts for 80 percent of the local maintenance and construction bud- get. Road maintenance costs in Guangdong Province are relatively high, espe- cially for waterbound macadam (unclassified) roads which require replenishing of surface materials every year and a major maintenance every two years. This high level of expenditure is due to very high traffic levels; considering also the much larger vehicle operating costs being paid by road users, it would be far more economical to provide higher standard pavements, but delays have occurred in the backlogged paving program. E. Transport Investment 2.61 The absolute value of transport investment has been increasing in current value in Guangdong Province since 1980, although as a percent of total capital investment the investment has been declining. During the Sixth Five- Year Plan period, transport was accorded 30 percent of total capital invest- ment; this dropped to 19 percent in 1986 and 20 percent in 1987. The decline 8/ Field analysis and sample data on surface condition for Huiyang Region high ways showing that 40-50 percent of these highways have serious defects suggests that periodic "heavy" maintenance such as bituminous overlays are also suffering from insufficient funds, although good main- tenance management has limited the defects to a lower level on the highest trafficked roads in this region.  - 28 - illustrates the problem of high growth and its impact on infrastructure, where adjustment seems to be limited by implementation ability as well as budgetary constraints. Investment in road construction has headed the list of transport expenditures, but rail and port investments have also been important (see Table 28).  - 29 - III. PRESENT TRAFFIC AND SERVICE LEVELS A. Overview of Aggregate Freight and Passenger Traffic Trends 3.1 Since 1978, freight traffic in the province has grown more slowly than passenger travel (11 percent vs. 16 percent from 1978 to 1985 in terms of freight tons and passenger trips, respectively), but since 1985 freight ton- nage appears to be growing at a rate closer to that of passenger trips (18 percent vs. 21 percent). River traffic, primarily low valued bulk commo- dities, which grew slowly in earlier years, has been growing more rapidly in recent years (see summary Table 3.1 below). Highway freight is growing the fastest and had overtaken rail traffic (in ton-km) in 1987. Longer term trends also seem to favor road traffic due to the increasing proportion of higher valued light manufactures and exports in the Guangdong economy. 3.2 The highest growth rates since 1978 in both freight tonnage and ton- km were experienced by the road and air modes, although rail ton-km also grew dramatically in the period 1985-87 as short distance traffic was replaced by longer distance flows. On the whole freight ton-km in Guangdong Province grew more slowly than freight tonnage (see Table 29) from 1978 to 2985. This was due in the case of road and water transport to shorter haul traffic growing faster than longer haul traffic, due largely to the concentrated growth of freight traffic in the Pearl River Delta where average travel distances are short. 3.3 Passenger traffic for Guangdong Province as a whole is growing very rapidly, as shown in the following summary Table 3.2 (see Table 30 for more detail). Total traffic increased much faster than "social" traffic, which is carried by state and collectively-owned specialized transport companies. The total volume of passenger trips by "social" carriers traffic dropped in recent years as the share of those transporters decreased dramatically.l/ 3.4 Contrary to the pattern for freight, total traffic in terms of pas- senger-km in Guangdong Province has grown even more rapidly than passenger trips. Road pass-km grew fastest in the last two year period, while air pass- km grew the fastest in the period 1978-85, slowing slightly as shorter air routes were inaugurated in recent years. Average passenger trip length for all modes, however, grew up to 1987. 1/ Official (social) figures may underestimate total traffic and traffic growth because traffic statistics are collected for only state-owned and large collective specialized transport enterprises, and traffic by own- account, small collective and privately-owned carriers is only estimated. The privately-owned or collective transport services (both specialized and nonspecialized) appear to be growing far more rapidly than the state- owned and collective sector in road and river transport. Total road and water traffic was estimated based on the judgment of knowledgeable DOC officials and past traffic growth data as well as a 1985 estimate by ICT, which are consistent with vehicle counts on the roads for which data are available.  - 30 - Table 3.1: FREIGHT TRAFFIC /a Average Annual Growth (Z) 1987 1978-85 1985-87 Total Social/b Total Social Total Social Freight Tonnage (million) All Freight /c 713.5 162.5 11 -6 18 9 Rail 40.6 40.6 1 1 8 8 Road 531.3 19.8 14 -20 21 -4 River 100.3 60.9 6 0 9 7 Coastal ship 30.6 30.6 8 8 8 8 Pipeline 10.6 10.6 n.a. n.a. 3 3 Air 0.1 0.1 17 17 29 29 Freight ton-km (billion) All Freight /c 108.8 86.0 4 1 12 8 Rail 14.0 14.0 6 6 23 23 Road 16.0 0.6 11 -25 29 -2 River /c 19.5 13.1 -6 -10 1 -3 Coastal ship 58.2 58.2 9 9 8 8 Air 0.1 0.1 n.a. n.a. 16 16 n.a. = not available /a Freight estimates are based on growth rates for vehicle-km in Guangdong and the judgments of DOC officials concerning the market share of state, collective, and privately-owned carriers in both road and water modes. Total freight includes estimated volumes of private specialized transport firms and own-account hauling which are not recorded in the official sta- tistics, but excludes local railways. More data collection is recommended to increase the accuracy of these figures. /b State-owned and collective specialized transport only (i.e., excluding transport by own-account, small collective, and privately-owned carriers). /c Excluding foreign ocean shipping but including transit traffic. Source: Table 29.  - 31 - Table 3.2: PASSENGER TRAFFIC Average Annual Growth (Z) 1987 1978-85 1985-87 Total Social/a Total Social/a Total Social/a (Est.) (Est.) (Est.) Passenger Trips (million) All passengers 709.3 327.8 16 14 21 -8 Rail 36.1 36.1 5 5 5 5 Road 642.2 266.0 20 17 23 -10 Water 27.8 22.5 1 0 0 -8 Air 3.1 3.1 20 20 48 48 Passenger-km (billion) All passengers 41.7 26.4 20 18 24 2 Rail 6.0 6.0 16 16 15 15 Road 30.0 15.5 22 19 28 -1 Water 2.7 1.9 6 6 14 -4 Air 3.0 3.0 36 36 21 21 la State-owned and collective specialized transport companies only (i.e., excluding transport by own-account, small collective, and privately-owned carriers). Source: Table 30. B. Present Traffic by Corridor and Type 3.5 Information concerning aggregate transportation trends such as that presented above helps provide an overview but is not adequate as a basis for transport planning, which requires information disaggregated by geographic area and by trip purpose. Modern techniques of transport planning normally rely on comprehensive information concerning the origins and destinations (0-D) of trips and shipments in analyzing transportation demand to understand the factors determining choice of mode, routing and hence traffic on individ- ual segments of the transport system. Such a detailed understanding of pres- ent traffic patterns lays the basis for rational forecasting of future traffic demands. Unfortunately, only partial O-D information was available for this study, and less satisfactory approaches have had to be devised to provide a rough approximation of demand patterns. 3.6 The principal device employed for this purpose involves grouping of the major categories of traffic into principal corridor flows and local traf-  - 32 - fic. Present traffic in Guangdong Province was also grouped into interna- tional, interprovincial and intraprovincial traffic for both freight and passengers. Corridor traffic encompasses virtually all international and intraprovincial traffic, but only about 15 percent of intraprovincial traffic; roughly 85 percent of intraprovincial traffic is local traffic outside the principal corridors. Total provincial traffic is the sum of corridor traffic plus local traffic. 3.7 Corridor Freight Traffic. International freight flows, which are concentrated in the South Corridor, include foreign trade flows through ports and road and rail traffic to and from Hong Kong, some of which is also included as transit traffic in the North Corridor. Almost 23 million tons, or 57.5 percent, of international flows are Hong Kong traffic flows. Interpro- vincial flows are concentrated in the North Corridor, including coastal ship- ping arriving in Guangzhou/Huangpu port, while more than 55 percent of intra- provincial corridor flows are found in the South Corridor (Table 3.3). Table 3.3: 1987 CORRIDOR FREIGHT TRAFFIC (million tons) International Interprovincial Intraprovincial Total North Corridor (5.86)/a 52.55 5.00 63.41 West Corridor 5.44 8.78 25.76 39.98 South Corridor 33.85 /b 53.29 87.14 East Corridor 0.66tc 2.29 12.66 15.61/c Total 39.95/b 63.62 96.91 200.28/b /a In order to avoid double counting on the provincial level, the column total excludes transit traffic from the north which also passes through the South Corridor (5.86 million tons). /b Interprovincial flows into Guangzhou/Huangpu port are considered North Corridor flows. /c Some East Corridor traffic exits by road to Hong Kong and therefore is shown only for the South Corridor. Note: See Table 31 for more detail. 3.8 Total Provincial Freight Traffic. Provincial freight traffic other than corridor flows is intraprovincial traffic composed of local traffic or intercity traffic within the province but not oriented along the main corri- dors. Therefore total freight flows by type exhibit a much higher proportion of intraprovincial traffic '85 percent) compared with corridor flows (48 per- cent intraprovincial). TabLe 3.4 illustrates this composition for 1987 traf- fic. 3.9 When internationa] port traffic is excluded to avoid double count- ing, the main corridors of (;uangdong Province carry a total of 160 million tons of shipped goods (see Table 31 for details). The Pearl River Delta route, or South Corridor, accommodates the largest amount of freight traffic  - 33 - Table 3.4: 1987 TOTAL PROVINCIAL FREIGHT TRAFFIC (million tons) International Interprovincial Intraprovincial Total North (6) 53 39 98 West 5 9 80 94 South 34 /a 400 434/b East 1 2 105 108/b Total 40 64 624 728/c /a Interprovincial flows into Guangzhou/Huangpu port are considered North Corridor flows. /b Some East Corridor traffic exits by road to Hong Kong and therefore is shown only for the South Corridor. /c The total excludes transit traffic from the north which also passes through the South Corridor, as well as air and pipeline flows, but includes 31.1 million tons of foreign trade not shown in Table 29. (53 million tons) as it links Guangzhou, Shenzhen, and Hong Kong. The North- ern Corridor carries the second largest amount of traffic serving as the main link between Guangdong Province and the rest of China. Coal, ores, iron and steel products, and grain dominate long distance rail and water flows. Break- bulk and higher value cargoes for destinations within the province travel by road. 3.1() Passenger Traffic. Less detail is available on passenger traffic by type than on freight flows. Consequently total corridor flows and provincial flows are provided below without the greater detail that is useful for more accurate transport analysis. Estimates of the distribution of these flows by corridor based on traffic levels for passenger vehicles (Table 3.5) and trains shows that the South Corridor accounts for 47 percent of the flows and flows in the other corridors vary from about 13 percent to 23 percent. 3.11 Geographic Patterns. Passenger and freight traffic along both sides of the Pearl River Delta corridor have risen faster than in the other corri- dors, and development in the Delta is the driving force for traffic in the province. The high mobility of Hong Kong and Macao residents leads to high levels of passenger traffic across the borders; the corridor also serves as an entrance and exit point for a growing number of foreign tourists. The active market economies of the SEZs and the Zhujiang Economic Open Zone also increase the mobility of residents around the Pearl River Delta.2/ Road traffic cur- 2/ Passenger traffic in 1985 accounted for 11 million trips on the Guangzhou-Shenzhen railway, 1.11 million trips by inland waterways, and 0.12 million air trips between Guangzhou and Hong Kong. Passenger trips on the Guangzhou-Shenzhen highway were estimated at about 18 million, which suggests that road trips may account for about 60 percent of the trips in the corridor.  - 34 - Table 3.5: 1987 PROVINCIAL AND CORRIDOR PASSENGER FLOWS (million trips) Corridor Flows Provincial Flows /a North 30.6 93 West 53.4 160 South 110.5 334 East 40.8 123 Total 235.3 709 /a Due to lack of detailed data, the geographic distribu- tion assumed to be the same as for corridor flows. For details, see Table 32. rently accounts for nearly 60 percent of freight flows on the East Bank and an estimated 70-80 percent of the freight flows on the West Bank, and even higher proportions of passenger flows in the Delta due to the large number of origins and destinations and short distances involved. Road and river transshipment is significant and is increasing. 3.12 The volume of daily vehicle traffic on the road network in Guangdong has fluctuated since 1980. Between 1980 and 1985 it increased by about 20 percent per year on national, provincial and county roads, and even faster (26-31 percent) on rural roads. In 1986 it slowed to 7-10 percent, then in 1987 increased to almost 30 percent on all roads (see Figure 2 and Table 33). C. Modal Shares of Traffic by Corridor 3.13 Determinants of Modal Shares. In Guangdong Province, as in much of China today, there is such a severe shortage of transport capacity in relation to demand at existing prices that neither goods shippers nor passengers in reality have much choice among modes; they are fortunate if transport by any mode is available. This situation stems partly from underinvestment in trans- port in the context of a rapidly growing economy, and partly also from an apparently inflexible and distorted price structure for transport which, at least until the tariff increases of September 1989 (passengers) and March 1990 (freight), has served to generate excess demands, particularly for railways. This means that in many cases, particularly longer distance transport, capa- city rationing rather than c-ompetitive choice has been the basis for modal traffic allocation decisions. 3.14 The modal split for freight transport for most line-haul movements of major commodities is still determined by administrative allocation at the central government level, just as the prices for most transport services are still centrally administerecd. The role of markets, wherein shippers are free to choose among freely competing suppliers according to price and quality of services offered, is still limited and largely confined to the highways freight mode. Administrative allocations of major commodity movements by mode  - 35 - are understood to be governed by two primary factors: availability of exist- ing capacity and the financial prices of alternative modes. No attempt is made to correct for distortions between financial prices and economic costs in the allocation of current traffic to existing capacity, although SPC has introduced economic shadow pricing in the evaluation of proposed investments, which will affect capacity and hence modal shares in the future. 3.15 Passengers are free to choose among alternative modes, but also in a market where severe capacity constraints limit their choices over many routes. Rail services in particular are in short supply, and while earlier passenger fare hikes were designed to provide an incentive for rail passengers to shift to other modes for trips under 300 km, the September 1989 fare raises affected long-distance more than short-distance passengers (see Chapter IV for more discussion on pricing). 3.16 The major safety valve for pent-up demand in this context has been the freedom of the individual enterprise or collective (and more recently pri- vate individuals) to provide their own motor vehicle and move their own and others' traffic over the public roads. This is one important factor contri- buting to the extremely rapid growth of road transport and modest growth of waterway traffic in recent years. Additional factors include the flexibility and other superior service characteristics of road transport, for which ship- pers and travelers worldwide have been willing to pay premium prices. 3.17 Modal choice by shippers or passengers is normally specific to a particular origin and destination, and varies by trip length (see Chapter IV). Since O-D data were not available, the analyses have focused on the modal share within each major corridor. 3.18 Freight Modal Shares. The North and West Corridors show a heavy rail influence, which will increase for the West when the Yaogu-Maoming link is completed (Table 3.6). The South shows a much higher proportion of road traffic due to the shorter haul distances and the more dispersed pattern of development. The East shows the most river-oriented pattern, due to its rela- tive isolation on the rail and road networks. 3.19 River transport tends to specialize in construction materials and ores outside of the Delta, and serves general cargo primarily for longer trips and in locations where road transport is limited. In the Delta this traffic is more diversified, and containers have gained a foothold as well. Rail traffic focuses on heavy commodities and longer haul traffic, while roads have been relegated to very short haul traffic in the past, that is only recently starting to include longer hauls, especially for light industry products and high value imports. 3.20 Passenger Modal Shares. Road modal share for passengers is signifi- cantly higher than road freight shares in all corridors, as shown in Table 3.7. Only the North Corridor has less than 75 percent share by road (although in the West, rail share may increase on completion of the Yaogu-Maoming link). River travel, on the other hand, is much less popular for passengers than for freight. Little is known about passenger modal choice except that road and air travel appear to be increasing faster than other modes and river passenger travel appears to be declining according to the approximate estimates below.  - 36 - Table 3.6: 1987 FREIGHT MODAL SHARES IN CORRIDORS (Z of corridor tonnage) North West South East Rail 58.3 45.0/a 16.2 n.a./c Road 16.8 19.6 47.9 49.8 River 2.6 10.2 35.9 37.9 Coastal Ship 22.4/b 25.2 /b 12.3 Total 100.0 100.0 100.0 100.0 /a Litang-Zhanjiang railroad which only covers a small part of the corridor. Ib Interprovincial ship traffic is considered a North Corridor flow. /c Excludes traffic of Ma;ixian-Xipu-Xingning local railway and other local railways. Note: Air freight traffic and local rail traffic outside the connected system are not included. Source: Table 31. Table 3.7: 1987 PASSENGER MODAL SHARES IN CORRIDORS (%) North West South East Rail 58.6 6.6 13.3 n.a. Road 34.9 76.4 85.1 75.2 River 6.5 3.7 1.5 - Coastal Ship - 13.1 - 24.5/a Air /b 0.2 0.1 0.3 Total 100.0 100.0 100.0 100.0 /a Service from Guangzhou to Shantou canceled in 1988. /b Data on interprovinciaL flights not available. Source: Table 32. 3.21 Total Provincial Modal Share Trends. Road and air shares of origi- nating traffic (tons of freight or passenger trips), have been increasing rapidly over the last five years, while the relative shares (but not absolute volumes) of rail and water have been declining. The same is true for ton-km modal shares, except that the rail and water shares are declining more slowly for passenger transport and faster for freight. Road transport now accounts  - 37 - for 90 percent of passengers and nearly 75 percent of freight tonnage (exclud- ing foreign trade), but under 75 percent of pass-km and 15 percent of ton-km. The former dominance of the rail transport system is still evident in the case of heavy commodities, but other transport categories are becoming more impor- tant in the Province. Coastal water transport seems to be gaining for bulk cargoes, and trucking seems to be gaining for light cargoes where road trans- port is available and road conditions are adequate. 3.22 The rapid increase in nonurban passenger traffic by road and air modes is consistent with rapid growth in business and tourist travel, which requires more convenience, flexibility, and speed than other types of travel. For long distance travel, rail transport is still the dominant carrier, but is limited in the amount and quality of service it can provide. The slower and less frequent service characteristics of coastal ship transport have caused it to lose its long distance travel market to other modes, despite its availabil- ity. The same is true for river passenger service in the areas where river travel is substantially more time-consuming than other modes. Short-distance international passenger travel by hydrofoil has gained in importance in the Pearl River Delta due to its speed and convenience, including new passenger terminals. D. Assessment of Existing Capacity Constraints 3.23 As in other parts of China, transport supply in Guangdong has not kept up with demand. Each corridor has its capacity limits which define the present transport shortage. Table 34 summarizes the modal capacities within each corridor. 3.24 Construction to expand the railway sections in the Beijing-Guangzhou corridor to a rated capacity of 30 million tons in each direction was com- pleted at the end of 1988. This will temporarily release a severe constraint on demand, which was held to 18 million tons in 1987. The marshalling yard at Jiancun also has been rebuilt and now handles 32 million tons of freight per year. These improvements will allow rail traffic to increase at normal rates, which are forecast at 16-19 percent per year as long as the railroad maintains its share of interprovincial freight traffic. This means that the capacity of this rail link will again be reached in the 1990s unless a strat- egy is adopted to shift freight to other modes and/or add still more capacity. The railways account for 21-33 percent of total freight capacity in the North corridor and 8-11 percent of total passenger capacity. The difference reflects the preference given to freight service over passenger service in allocating rail capacity. 3.25 The strain on the port system is almost as severe as that on the railroad. For maritime shipping, deep-water berths for vessels in excess of 35,000 dwt are available only at Zhanjiang, and the capacity of handling equipment is limited at most ports. This leads to long waiting times for ships--6-7 days in Huangpu and Shantou. Shenzhen and Zhuhai ports will pick up some of the additional cargo but they are limited to 10,000 dwt ships. In addition, there are bottlenecks on land connections to ports, especially sec- ondary ports and major river ports. 3.26 In 1987, the total throughput of river ports was 40.2 million tons, or 42 percent of the combined throughput of river ports and seaports. This  - 38 - level is down somewhat from the 50 percent recorded for 1985, but the demand is increasing for modern service, especially in the Pearl River Delta, where SEZ ports are increasing their throughput rapidly. 3.27 Traffic on all the trunk roads entering Guangzhou now exceeds rated capacity. Entrances to other large cities also are congested. Some sections carry traffic at very low levels of service, as indicated in Table 34. In addition, many highways pass through highly congested, urban centers, and most highways are carrying a wide range of slow and fast vehicles and pedestrians. 3.28 Civil aviation airports are not designed for their present levels of traffic. For example, GPPC estimates that the Baiyun Airport terminal meets only 50 percent of the demand for air travel to and from Guangzhou. It recently has become the busiest airport in China. The military airport at Foshan has been used to relieve some of the demand. Demand is clearly growing in Shenzhen for both domestic and international flights where airport capacity does not exist except via Guangzhou or Hong Kong. 3.29 The Zhanjiang-Maoming pipeline carries crude oil to the refinery. It has no return capacity for refined products, and this limits the use of coastal shipping. 3.30 Corridor capacity is limited in many cases by specific bottlenecks. Table 35 presents data on each network link that is a potential bottleneck operating at capacity. Railway and port capacities have been greatly exceeded by recent volumes of freight traffic. This implies that the capacity calcula- tions are normative prescriptions related as much to personnel performance standards or to the specific equipment in hand as to actual infrastructure constraints. Better capacity calculations for each link in the Guangdong Province transport network a;re critical for improved project identification and evaluation. E. Assessment of Technology 3.31 Although there are footholds of modern technology in a few places for every mode, much of the itransport technology used in Guangdong Province dates from an earlier generation. The extremely low levels of containeriza- tion and other more modern forms of cargo unitization have already been noted in Chapter II. 3.32 In the railroad subsector the rail grades in the Northern Corridor have been improved with the double-tracking project, but the mountainous ter- rain still places limits on locomotive capacity. Stations are spaced too close together for high speeds. Signaling systems may need further updating, and the operating information system should be computerized in most locations to increase efficiency. Conmmiunications technology is also still a weak point. 3.33 Technical standards for roads are extremely poor and surface condi- tions are lower even with more frequent maintenance and high maintenance costs than the national average. There are too few class 1 roads, and class 2 roads and expressways are only now being built on the trunk routes in the Pearl River Delta and many unpaved roads exceed 2,000 vehicles per day. Only 13.4 percent of roads are well surfaced, with cement concrete or asphalt. Pavement design in the past was for very low axle loads, which are now  - 39 - increasing dramatically with larger vehicles. Consequently road pavement deterioration has accelerated leading to rapidly increasing maintenance costs. There are many broken-stone roads (which would not meet the normal standards of water-bound macadams) with extremely rough riding characteristics akin to those of cobblestone pavements. Bridge weight limits on rural roads also are often low. 3.34 The trucks owned by specialized, state-owned transportation compa- nies have an average capacity of only 3.7 tons, while the average capacity on state-owned buses is 40.7 seats. New vehicles are entering the nonstate-owned fleet, and there is a large percentage of foreign-made vehicles, especially for passenger traffic and for co-owned enterprises that have entered the Guangdong-Hong Kong road transport business. There is an increasing share of imported pickup and larger trucks (including container trucks) in the road fleets, and both truck and bus fleets will benefit from the planned importa- tion of more fuel efficient engine technology for domestically manufactured vehicles. Due to inadequate pavement and geometric design standards, and the consequent poor condition of the road surfaces, the costs of vehicle ownership and operation are unnecessarily high. The role of highway transport therefore has been more limited than it would be with improved technology and an improved infrastructure. 3.35 Inland river port equipment and outmoded (50-110 ton) barge technol- ogies are not competitive with the increasingly efficient road and future rail systems. This is already changing in the Pearl River Delta where larger barges, barge trains and container barges have increased their services and efficient port handling equipment has been installed in a few locations. The deepening of the Xijiang to handle 1,000-ton barges its whole length will encourage more modern barge technology in the future. The greater use of even larger integrated barge trains using equipment recently imported with up to 5000 horsepower tugboats is starting in other provinces but not yet available in Guangdong Province. 3.36 Major port technology is being upgraded in Huangpu, as well as in the SEZ ports (Shantou, Shekou, Chiwan and Jiouzhou) and other Pearl River Delta Ports. The Huangpu port project is designed to increase port capacity from 17.5 to 22.3 million tons/yr. This will include expansion of coal termi- nals and general cargo terminals, with rail cornections. However the channel depth limitation to 20,000 dwt vessels is a major constraint on the use of larger vessels for both coastal shipping and foreign trade. 3.37 Coastal shipping technology is also presently limited by shallow port approach channels. As one particularly important example, it should be noted that the coal berths of the new power generation plants in the Delta are only designed for barge depths, with service from the Xi-Ji coal terminal at Huangpu. This raises the question of what feeder ship/barge technology and load center is appropriate for coal service, for petroleum products and con- tainers as well. Recent rapid technological progress in development of shal- low draft ocean-going barges in other countries could provide vessels which would avoid the need for transshipmernt altogether, thus reducing demand for new transshipment hubs. It should also be noted that the high technology Hong Kong ports are gaining cargo, but may be saturated in the 1990s. Conse- quently, the whole regional location and technology mix is a question that must be resolved for each major cargo category.  - 40 - 3.38 Finally, air transport technology is relatively modern with jet service, but the airports are not up to standards, and the management of the sector is diffused. Communications technology is not up to the increasing level of demand in this sector. The future role of all airports when Hong Kong is integrated into the system is also important and should be included in ongoing planning for the Shenzhen airport.  - 41 - IV. COMPARATIVE ROLES OF THE TRANSPORT MODES 4.1 In China as a whole, past concentration of development strategy on heavy industry and reliance on coal as the primary source of energy have been coupled with a transportation strategy focused to an unusual degree on the rail mode. There has been relatively limited exploitation of the two most important 20th Century innovations in transportation technology: modern high- way and aviation systems. Consequently, the railway has played a dominant role in transportation in China, enjoying a virtual monopoly in many respects but at the same time burdened with much traffic for which it is inherently ill-suited, and the service quality concerns of transportation users have been largely neglected in a situation of chronic shortage of transportation ser- vices. 4.2 Guangdong Province suffers like the rest of China from an overall shortage of transportation services and infrastructure, particularly for flows to and from other provinces and fast-growing areas. Until now it has had inadequate and substandard highways and only limited service of railways, pri- mari:Ly through the critical rail linkage to the north in the Guangzhou-Beijing corridor. Including the Yaogu-Maoming line now under construction, the den- sity of the railway network in Guangdong would be just equal to the national average (Table 13) despite the greater than average economic production of the Province. This situation may reflect the fact that the economy of Guangdong Province is concentrated to a much lesser degree on heavy industry than other provinces. In deciding on directions for future development of transporta- tion, it is important to consider the needs of a development pattern that appears to be primarily, but not exclusively, oriented to the growth of light manufacturing and high value agricultural processing with a strong emphasis on foreign trade. This means that service quality concerns, including the requirements of foreign trading partners for fast, flexible transport for light-industry products assume great importance. 4.3 This chapter examines present modal shares, tariff structures, and costs, analyzes what the costs would be under more economically ideal circum stances, and draws some conclusions about the effects of present price distor- tions and the most economic roles of different modes in the future. For this analysis, a set of long-run cost functions was developed, based on the assump- tion that future traffic demands would be met with adequate infrastructure and the most efficient vehicles for each mode. In Chapter V, the analysis will be carried further to project future traffic demands and the expansion of differ- ent modes that would handle them in an economically efficient manner. 4.4 Unfortunately, rigorous economic analysis of transportation costs and service quality is not yet well developed in China, and firm quantitative evidence is limited. The cost functions developed here are based on a variety of data, not only from Guangdong Province but also from national sources, sup- plemented by a series of assumptions based on experienced judgment where empirical evidence is lacking. This analytical framework is spelled out in the annex where quantitative cost estimates are also developed in detail. No claim is made as to the precision of the estimates; they are intended rather to delineate a logical framework for examining the issue of the economic roles of the different modes of transport, and to elicit a debate that hopefully will lead to better data and an improved understanding of the options for  - 42 - future development of transportation. Nonetheless, the results even at this stage suggest some important practical conclusions which would appear reason- ably robust over a wide range of assumptions. These conclusions, although directed particularly to Guangdong Province, are thought to be generally applicable to much of the rest of China. A. Comparative Transport Costs and Pricing Among Modes 4.5 Three concepts of cost for transport service are used in the follow- ing analysis: (a) users' (shippers' or passengers') perceived costs, (b) carriers' long-run, incremental financial costs, and (c) society's long-run incremental economic costs. The users' perceived costs are based on present fares and tariffs and other costs incurred by the user (e.g., time costs, packaging), while the long-run costs (either financial or economic) represent the potential future situation based on the assumption that best existing technologies will be efficiently exploited. Each of these cost concepts is defined below, and detailed calcu- lations for each mode are found in the annex. 4.6 Users' Perceived Costs (UPC) are defined for passengers as total out-of-pocket expenses including fares and other travel costs such as meals and hotels on long bus trips, as well as time costs, and for freight as tar- iffs, storage, and handling costs.l/ To the perceived costs of the princi- pal mode are added the extra door-to-door costs for rail, water and air modes where truck or bus transport is needed to get from or to the ultimate origin or destination. Costs for unreliable service or infrequent service are not included here for the sake of simplicity (although these may assume great importance in the case of passengers and some high-valued freight services). In past analyses in Guangdong, the value of time was assumed to be the same for all trip purposes for all travelers and was estimated by dividing nonagri- cultural GNP by nonagricultural work hours.2/ In this analysis, however, costs were calculated for a range of time values, allowing the analyst to consider differences between passenger groups. Senior civil servants and national businessmen, for example, have a greater opportunity cost for their time than unskilled workers, and foreign businessmen and tourists likely use 1/ In addition to the costs included here, a comprehensive approach would also include packaging costs, time costs, and unreliability costs for freight which now result in very large stockpiles of coal and other com- modities waiting for transport. Time costs can be critical in certain types of freight services, particularly for manufacturing process indus- tries, where "just-in-time" (JIT) logistics and inventory management techniques have given rise to a whole new standard of transport services in modern industrial countries. A true generalized transport cost would include the costs of excessive inventories associated with poor transport service, as well as packaging for shipment. 2/ This yielded a value of Y 0.488 ($0.14) per hour based on 1985 data.  - 43 - an even higher opportunity cost in assessing the options for the mode of travel. In addition the value of time for all groups will increase in the future as their real per capita incomes increase. 4.7 Long-Run Incremental Financial Cost (LRIFC) is defined as the aver- age unit transport cost incurred under long-run conditions 3/ by each car- rier (road, rail, water, or air) in order to carry an additional passenger-km per year or an additional freight ton-km per year. It includes variable costs and other operating costs, depreciation of additional rolling stock, vehicles, or vessels (at replacement value) and average incremental overhead costs, where these are borne by the carrier and vary with traffic volumes. These costs are after subsidies and include all taxes and fees charged by the gov- ernment (on the assumption that the present tax/subsidy regime remains unchanged). Transshipment costs and local pickup and delivery costs are also included in LRIFC for the rail, water and air modes (as for perceived costs above) to compute comparable door-to-door carriers' financial costs for each mode. 4.8 Long-Run Incremental Social Cost (LRISC) is defined as the economic resource cost to the society for an additional ton-km or passenger-km trans- ported. These costs are identical to the LRIFC except for three adjustments. First, the LRISC calculation omits government taxes, fees and subsidies and replaces thein with the actual incremental costs to the government or enter- prise for providing the infrastructure or other capital goods needed to pro- duce the transport service. This includes the opportunity cost of capital that is invested in infrastructure, vehicles, vessels, or rolling stock. Second, financial costs are converted to economic efficiency costs by using conversion factors or shadow prices which represent the costs to the economy of each resource category; this is necessary because financial prices paid in China for many resources do not reflect the actual cost to society in terms of forgone alternative uses ("opportunity costs"). Third, in the case of passen- gers, the economic value of time in transit is included. 4.9 Four Tests, comparing these costs for different modes and services, are significant for assessing economic problems in the transport system: Test 1 LRISC vs. UPC: If LRISC is greater, noneconomic excess demand is being encouraged for the transport service; if the UPC is greater, demand is being suppressed below levels which would be economically desirable over the long run; Test 2 LRIFC vs. UPC (excluding time values): If LRIFC is greater, there is definitely a problem of financial viability of the service. If UPC is greater the service would be financially viable with best existing technology and other conditions assumed under LRIFC; under actual conditions a higher UPC may reflect either high profits for efficient carriers in a market with chronic shortage of supply ("sellers' market") and/or 3/ "Long-run conditions" or "in the long run" means with appropriate expan- sion of infrastructure, equipment, and facilities to keep an economical balance between demand and capacity.  - 44 - inefficiencies (e.g., poor roads, congestion) which raise present operaLting costs above long-run costs; Test 3 LRIFC vs. LRISC (excluding time values): If these are unequal, it indicates that the present regime of taxes and subsidies, if applied to long-run costs, would distort the cost structure, thereby creating uneconomic incentives for the carrier; and Test 4 the ranking of modes by LRISC vs their ranking by UPC: If the rank orders cliffer, users are getting the wrong price signals for their choice of mode. 4.10 It should be noted that while the basic aim of the study is to delineate the potential future situation in the long run, Tests 1, 2, and 4 involve comparisons of the long-run potential costs with actual present prices, while Tests 2, 3, ancd 4 are based on the assumption that the present regime of taxes and subsidies (which enter into the carriers' costs--LRIFC) would be maintained in the future. These tests are intended to give some indication of how the costs of the future transport system might be different from those viewed today, and how current pricing and taxation/subsidy poli- cies, if maintained, might interfere with long-run economic efficiency. It should also be recognized, however, that in a situation with chronic shortage of capacity and attendant congestion, prices at least in the short run should be set above long-run costs both to encourage efficient utilization of exist- ing capacity and to generate revenues to finance capacity expansion. The emphasis here is on long-run planning options, rather than on current pricing policies per se. B. Comparative Cost Analyses Passenger Transport Costs and Prices 4.11 The costs according to the three concepts defined above are shown in Table 36 for passenger trips by different modes between selected points. Fig- ures 4 and 5 show, respectively, costs to the passenger (not including the value of time) and economic costs to society (LRISC) as functions of trip dis- tance. In using these graphs, it should be noted that the distance between two points sometimes differs significantly according to the mode used. From Guangzhou to Shantou, for example, Table 36 shows the distance to be 380 km by air, 481 by road, and 543 by ship. 4.12 Examination of the table and graphs permits applying the four tests prescribed above for analyzing potential distortions or maladjustments. Test 1, comparing actual users' cost (UPC) with economic resource cost under improved conditions (LRISC), and Test 2, comparing UPC with financial cost under those conditions (LRIFC), both show trips by road (bus trips) apparently overpriced, even without including any cost for the value of passengers' time. Table 36 shows UPC to be from 3 to 4 times as high as the long-run incremental costs. Test 3 further indicates that bus carriers' financial costs (LRIFC) would exceed society's economic costs (LRISC) by about 7 to 14 percent, reflecting a tax on road transport beyond the long-run costs of providing improved infrastructure and contributing to the distortion vis-a-vis the other modes, which are subsidized. A large part of the apparent overpricing of bus  - 45 - fares is probably due to existing poor roads and congestion, which may increase vehicle operating costs 130 to 150 percent above normal. The remain- der represents some unidentified combination of operator and vehicle ineffi- ciencies and possibly (for the more efficient enterprises) significant prof- its. 4.13 Rail passenger fares are now substantially above either LRISC (Test 1) or LRIFC (Test 2) for all trips, with the margin substantially greater for shorter distances (over 130 percent) than for longer trips (60 to 80 percent). Comparing carriers' costs with social costs (Test 3) indicates that LRIFC is just under 90 percent of LRISC, reflecting primarily lower financial charges than economic costs for amortization of assets (depreciation plus interest) and fuel. This, of course, enhances the substantial financial profits now being made by the railways on passenger services, which can help to meet capital requirements for capacity expansion and assets renewal. 4.14 For aviation, society's economic costs (LRISC) and carriers' costs (LRIFC) are almost identical (Test 3), but when either (Tests 1 and 2) is compared with user costs (UPC) at zero-time values, it appears that some routes are substantially underpriced (e.g., Guangzhou-Shantou, where fares cover only about 55-60 percent of costs), others are approximately break-even (Guangzhou-Shanghai), and still others show a significant profit (about 15 percent for Guangzhou-Beijing). Travel by ship appears to be overpriced for all trip distances relative to either economic costs (Test 1) or carriers' costs (Test 2), and particularly so for longer trips (e.g., for Guangzhou- Dalian, where user costs exceed LRISC and LRIFC by 77 and 155 percent, respec- tively). Carriers' costs appear to be substantially below society's costs (Test 3) for ship passenger services, especially for longer distances. 4.15 The ranking of modes (Test 4) is affected by the length of the trip. In terms of user cost (i.e., present fares and transfer costs) without allow- ing for any value of time, Figure 4 shows air travel by far the most costly mode at any distance, with road the cheapest over short distances and rail and ships very close for longer distances. These comparisons are strongly influ- enced by tax differences for different modes and by arbitrary fare structures. A radically different picture results when the comparison is made in terms of long-run incremental economic cost (LRISC), still omitting time value, as shown in Figure 5. In these terms, travel by bus turns out to be less expen- sive than by rail or ship at any distance. 4.16 For passengers whose time has a high value, of course, the faster modes are attractive. Figure 6 shows, for a trip from Guangzhou to Shantou (380 to 543 km, depending on the mode), how the traveler's perceived cost (fares plus time cost) and the long-run incremental social cost vary when dif- ferent time values are included. Even at very low values of time, ship travel loses out to rail and road modes because of their superior speed, for either perceived cost or long-run social cost. For this short distance, air travel becomes competitive at values around Y 12 per hour in perceived costs and at Y 20 per hour for LRISC. A second comparison is made in Figure 7 for the longer trip from Guangzhou to Shanghai, comparing air (1,235 km), rail (1,811 km), and ship (1,689 km). In this case also ship travel is not  - 46 - selected at any value of time above zero.4/ Air travel becomes less expen- sive than rail for passengers with a value of time of Y 7 per hour or more. 4.17 Regardless of time values, the modal choice distortion shown by com- paring ranking of modes by LRISC and by user charges (Test 4) appears to be the greatest in terms of bus passenger transport, which in an economically optimum situation in China would be less expensive than rail or any other mode, not only for short to medium distances, but also for longer routes, par- ticularly where road distances are shorter than or equal to rail distances. The air mode would also be much more competitive for long distance passengers with high time value than the comparison of fares alone would indicate. Bulk Commodity (Coal) Transport Costs 4.18 Table 37 shows the three different costs for shipping coal, China's principal bulk commodity, over various distances by various modes. Figures 8a and 8b show users' cost (UE'C) and the idealized economic cost (LRISC) graphic- ally as functions of distance, which helps to see how the costs of the differ- ent modes rank and which one offers the lowest cost at any distance. 4.19 Test 1, comparing LRISC with UPC in Table 37, shows that, relative to economic costs under ideal conditions, transport by road and large barge are overpriced, while rail transport (the most-used mode at present) and long- haul coastal shipping (also extensively used) and short-haul barges are under priced. For the typical long hauls involved, present railway tariffs appear to recover only about 65 to 75 percent of LRISC; for coastal ship (with rail ways bringing the coal to shipping ports) the percentage is about 80 percent for the 30,000 dwt vessel, and even less for the smaller ship. Such severe underpricing of the transportation of coal, a commodity in which transporta- tion is a major component of the delivered price, invites excess demand and uneconomic use of both transportation and energy, two of the scarcest goods in the Chinese economy. 4.20 Test 2, comparing LRIFC with UPC, yields similar results except for the particularly important case of railways, where tariffs paid by shippers are adequate to cover the railways' financial costs, which are less than eco- nomic costs (LRISC). Test 3 shows all modes except road transport having financial costs lower than economic costs, indicating some degree and form of subsidy. For road transport, carriers' financial costs (LRIFC) would exceed society's economic costs (1,RISC) by over 40 percent, reflecting a tax on road transport well beyond the long-run costs of providing improved infrastructure; this distortion has less significance in the case of coal than for other traf- fic, since road transport would not be competitive even in the absence of the tax. 4.21 The modal ranking test (Test 4) may be most easily grasped from the graphs of Figure 8. Either under existing prices or prices based on long-run economic costs (which woulcd mean cutting road transport prices by approxi- mately half), neither road transport nor small barges would be a preferred 4/ Persons traveling for tourist purposes to enjoy the voyage would have negative time costs (e.g., benefits) while very low income persons will still prefer this mode on the basis of lowest fares.  - 47 - mode (except for the shortest hauls or for local pickup and delivery services for those origins and destinations not directly served by rail or large ships). Rail is preferred under either cost criterion for all but the short- est hauls. 4.22 The finding that rail shipping is more economical than coastal ship ping (plus rail) for long distances conflicts with earlier findings of Chinese authorities, which led to major investments in large-scale coal loading facil- ities in northern China ports and in railways connecting them to the mines. It is also contrary to the results of earlier editions of this study based on prices of 1988 or earlier years. The new findings result from sharply higher cost increases in the case of coastal shipping than for railways, due prima- rily to large increases in the prices of ships, reflecting not only recovery of previously depressed world markets for ships, but also possibly some under- valuation of ship costs assumed in previous analyses. As the issue is an extremely important one--the huge tonnages involved dominate domestic shipping in China--it requires more detailed investigation. Given the current capacity constraints in the railways, coastal shipping (the costs of which are not that different from railways) must, in any case, remain a major coal carrier in China for the foreseeable future. Heavy Freight Transport Costs and Tariffs 4.23 The costs for transport of heavy freight, such as iron, steel, and timber, by different modes are presented in Table 38 and Figures 9A and 9B. Relative to long-run economic costs (Test 1), transport by road is greatly overpriced, short-distance barge movements are significantly underpriced (par- ticularly in the West River), and coastal shipping and rail transport are slightly underpriced, at least for longer hauls. In financial terms (Test 2), all modes 5/ except barges over shorter distances appear to cover costs under the long-run assumptions. Comparisons involving rail and water trans- port are strongly affected by the assumption of transfers by truck at each end of the trip, greatly increasing the cost over what it would be on a siding-to- siding or wharf-to-wharf basis for shorter hauls. Thus, Test 3 appears to show that economic and financial costs are approximately equal for each mode (except for road), because the significant line-haul subsidies are small per- centages of the total delivered cost. 4.24 Test 4, the ranking of modes, shows road transport the best mode for short hauls under either existing users' costs, UPC, or economic costs, LRISC. By either criterion rail would become the preferred mode for longer distances. As Figure 9 shows, the change-over distance based on present UPC is much shorter than if based on LRISC because of the current relatively high price of truck transport, due in part to bad roads and congestion. In future, with roads designed for modern, large semi-trailer or trailer trucks and with ade- quate capacity for the traffic, road transport can be competitive with rail- ways for heavy freight, even for haul distances up to 1,100 km. The compari- son assumes that freight hauled by rail must be picked up and delivered by truck at both ends of the trip. This conclusion would be invalid if both 5/ The 25-ton semi-trailer truck assumed in Table 38 would not normally be used over haul distances as short as 50 km; a smaller truck, as normally used, is more economical over such distances.  - 48 - shipper and receiver have rail sidings, so that pickup and delivery costs (at least for large shippers) could be lower. In the extreme case where local pickup and delivery costs were essentially nil the rail would always be the least costly mode (in terms of either UPC or LRISC), provided also that rail- way wagons were available when needed, so that additional costs for extra inventory holdings would nol: be incurred; such inventory costs have not been included in the present calculations. Waterborne transport is not competitive under either cost criterion under the assumptions employed in Table 36, but could be under different circumstances, particularly where both origin and destination lie on the waterways. Less-than-Carload General Cargo 4.25 As with the other classes of freight, Table 39 and Figure 10 show the variation with distance of each of the three cost concepts for each mode, for shipment of general freight in less than carload lots. Again, Test 1 shows road transport priced far above LRISC, the economic cost under improved future conditions. Rail tariffs are also substantially higher than LRISC, particularly for longer haul. distances, but less so than for trucks. Water- borne transport, on the other hand (both barges and coastal shipping), appears to be priced well below economic costs, thus potentially generating noneco- nomic demands. Test 2 shows similar results: road and rail priced well above long-run financial costs, while water modes in most cases, particularly barges, are priced below financial costs, suggesting LCL cargo services on those modes would not be financially viable over the longer run. 4.26 Test 3, comparing economic and financial costs in the idealized conditions assumed (LRISC and LRIFC), shows (i) the two costs approximately equal for railways; (ii) road transport overtaxed; and (iii) waterborne modes subsidized. In this case these distortions are of significance, since highway transport should play a larger role. 4.27 The rank order of modes by UPC and by LRISC depends on the haul distance, as seen in Figure 10. By either criterion, road transport is least costly for short hauls, as would be expected when the other modes require pickup and delivery by truck at the ends of the haul. The most efficient and lowest priced land-based mocde for longer trips is the railway, but the dis- tance at which this becomes the preferred mode is much longer (1,150 km) on the basis of LRISC than at present user costs, UPC (400 km), just as it is for heavy freight. C. Impacts of Inefficient Conditions 4.28 Apart from the for-egoing comparisons of costs and prices, it is useful to examine any factor-s that are known to raise costs, in order that policies and programs may be designed to deal with such factors. Only by eliminating these factors can the assumed long-run efficient conditions be achieved and costs reduced to the level of LRISC. Some of these efficiency- reducing factors have alreacly been mentioned in passing; others have not. Table 4.1 brings them together and summarizes them for each mode.  - 49 - Table 4.1: INEFFICIENCY FACTORS Cost Increase Mode Primary Factors Carrier Effects to Society Road too small vehicles inability to compete up to 90% road capacity constraint congestion reduces up to 100% /a speeds and km/yr poor surface condition reduces speeds and km/yr up to 30% short haul distance reduces annual km up to 20% Rail too many short hauls limits long hauls up to 20X too many pass. trains limits freight trains noneconomic diversion restrictions on shippers none reduced economic interaction River too small barge size inability to compete up to 100% too little port capacity fewer travel days, due 60% cost to waiting time /b increase for 2 days increase/trip. Coastal slow unloading equipment fewer travel days " " Ship too little port capacity fewer travel days " " restrictions on ships none reduced inter- action Air too large aircraft size/c low load factors /c up to 100% overpricing of service lower utilization rates up to 50% restrictions on passengers none reduced inter- action and fragmented markets /a If congestion reduces the number of trips the economic impact is more significant. /b Efficient port waiting time is 1 day or less. Present average is 4-7 days for major ports and less for smaller ports. /c May apply only to trips under 1,000 km. D. Conclusions 4.29 Comparing actual present prices against potential long-run costs can be helpful in understanding the potential for change in the future, certainly as to the direction of those changes and, to a degree, their expected magni- tudes. These may be summarized:  - 50 - (a) Highway transport costs are likely to fall significantly--possibly by half--as China and Guangdong Province move toward a modern road transport system, if prices and user charges are set for economic efficiency. Railway costs may also fall, but by much less, since the railways are much closer to the efficiency frontier, while rail- ways' tariffs should rise further for such key commodities as coal and heavy freight; (b) Long distance coal transport by rail is still severely underpriced, even after the tariff increases of March 1990, continuing to encour- age excess demand for both energy and rail transport; (c) Although recent tariff increases were not sufficient to close the previous gap and cover recent cost increases for coal, the gap for heavy freight has been substantially reduced, while passenger tar- iffs now exceed either financial or social costs by a wide margin for all distances; (d) Potential long-distance passengers are being uneconomically discour- aged by high rail fares, relative to carriers' economic or financial costs, while air fares appear substantially higher than economic costs for some routes and below costs for other routes; (e) Recent increases in the costs of coastal shipping appear to have substantially erodedi the comparative advantage of that mode vis-a- vis railways for ,-arrying coal (and similar bulk commodities). Given the importance of this traffic, and the prospects for coastal shipping relieving congested railway corridors, this issue should receive more detailed study; and (f) Comparison of LRI]TC and LRISC suggests that the present regime of prices and taxes dliscriminates against road transport and to a lesser degree inland water transport, inflating the costs of carri- ers in those modes relative to other modes. 4.30 Several conclusiorLs concerning the future roles of the transport modes can be drawn from the above analysis, despite possible uncertainties in the cost data. These are: (a) Road transport should be made more efficient by expanding the capa- city and upgrading the quality of highways and by using modern large buses and heavy semi-trailer trucks for longer hauls. Fares and tariffs over the long run should be brought down accordingly, and service expanded to much more than local delivery, especially in the areas of short to medium distance passenger transport, and LCL freight up to 1,000 km;6/ 6/ It should be noted that there is not unanimity of views among the co- authors of the report on this conclusion. In particular, the Guangdong Province Planning Commission and the Institute of Comprehensive Transport wish to record their view that constraints on the development of road transport are likely to limit its role to haul distances much below 1,000 km at least up to the year 2000.  - 51 - (b) Air transport should expand its long distance and medium distance services and price them to attract long distance travelers from the rail system. These services are expected to grow rapidly with eco- nomic growth and increasing recognition of the value of time for business travelers. (c) The railway should concentrate on its most economic transport roles which include medium distance freight and passengers, and, at all distances, coal, other bulk commodities, and heavy freight; and (d) More detailed studies of the technological options and costs of coastal shipping are needed to better define the comparative advan- tage of that mode for bulk commodities (particularly coal). 4.31 It should be emphasized that, as explained in para. 4.10, these analyses and conclusions are primarily intended to define the economic role of the modes, delineate long-run planning options, and indicate how future costs would compare with present prices, rather than to define pricing policies. However, the chronic shortages and attendant congestion in the transportation network in Guangdong Province, like China as a whole, suggest that transport prices on average should be raised further to improve efficiency of utiliza- tion of existing assets and to generate revenues for expansion. Certainly, for economic efficiency, prices should be increased for those services which are currently priced below even long-run social costs (LRISC). The present "overtaxing" of road transport may be appropriate at least in the short run where road congestion is significant and funding for road improvements is limited. In the long run, supply of road infrastructure should be increased to meet demand, taxes reduced to bring road transport financial costs (LRIFC) in line with social costs (LRISC). For the other modes subsidies should be reduced also in order to bring financial costs in line with social costs.  - 52 - V. FUTURE TRAFFIC AND CAPACITY CONSTRAINTS A. Methods of Traffic Forecasting 5.1 Traffic forecasting was carried out as part of the transport demand analysis depicted in Figure 3. Forecasts were made for three primary types of traffic: foreign trade flows, domestic freight and passenger flows, and potential induced traffic flows. Foreign trade flows are based on forecasts of foreign trade for each major commodity group and are calculated for both Guangdong Province and for China as a whole. Domestic freight and passenger flows were forecast using economic indicators for regional growth in each transport corridor and subcorridor (see Tables 10, 11 and 12), and using fore- casts of interprovincial traffic for each major commodity group originating or ending in Guangdong Province. Potential induced traffic can be found where an industry would not be established or expanded if a specific transport improve- ment were not made, but could be created if the investment were made. Spe- cific cases of induced traffic were identified in the Eastern Corridor, and more general types of induced traffic were identified in other areas. 5.2 In principle, the traffic volumes associated with each corridor should be forecast by origin and destination (0-D) for each significant com- modity group. The choice of mode then should be forecast for each O-D pair using the modal planning for that specific flow or the shipper's modal prefer- ence and comparative modal service characteristics for the trip or shipment. However, the absence of traffic data by origin and destination eliminates the possibility of using this method. Instead, a hybrid approach was taken in which commodities with large volumes and limited destinations were forecast independently by origin and destination, and the remaining traffic was fore- cast using a more general relationship to the growth in each corridor. This growth in commodity flows and income levels starts from the three alternative economic forecasts established in Chapter I and results in three alternative forecasts of total demand for freight and passenger transport initially on the assumption of no capacity constraints. Each of these forecasts reflects nor- mal growth in traffic, without the induced traffic associated with system improvements. The forecasts were then modified to reflect the expected impact of the Pearl River Delta-Hong Kong expressway on traffic volumes in 1995 and 2000, and in a second stage capacity constraints were introduced for the Northern Corridor railway. 5.3 The forecast of future traffic to be carried by each mode and link in the transport system was based on two factors: the allocation of planned commodities to each mode under the state allocation rules, and the choice of mode by user preference for out-of-plan commodities and passengers. The choice for out-of-plan commodities is modified by congestion and price consi- derations as described in Chapters III and IV. The initial result is a base case traffic forecast by mode, at this stage assuming no improvements in the transport system other than ongoing construction (including the Pearl River Expressway System), which is used as a reference point to identify future bottlenecks in the Guangdong transportation system and subsequently to assess the impact on traffic of potential improvements. 5.4 The forecasts carried out using these techniques are clearly a first approximation that must be refined for applications more specific than general  - 53 - corridor and area growth rates. However, they can give the general trend and range of likely traffic volumes at a sufficient level of accuracy to identify overall transportation requirements. B. Freight Traffic Forecasts 5.5 International Freight Traffic. The international commodity freight forecasts summarized in Table 5.1 reveal substantial differences in traffic growth among the four corridors. As expected, the Pearl River Delta exhibits the highest growth. The Eastern Corridor will experience significant growth but from a very small base volume. Table 5.1: INTERNATIONAL TRAFFIC IN MAJOR CORRIDORS (million tons) Pessimistic Base case Optimistic Actual forecast forecast forecast Growth Growth Growth 1987 2000 rate (Z) 2000 rate (%) 2000 rate (Z) North Corridor ja (5.9) (11.3) 5.2 (11.3) 5.2 (11.3) 5.2 West Corridor 5.4 14.7 7.9 19.2 10.2 38.3 16.2 South Corridor 33.9 115.7 9.9 159.2 12.6 373.5 20.3 East Corridor Lb 0.7 2.0 8.9 2.7 11.3 6.0 186.5 Total la 40.0 132.4 9.6 181.1 12.3 417.8 19.8 /a The total for the province excludes all transit traffic in the North Corridor which is also counted in the South Corridor. /b Some East Corridor traffic exits by road to Hong Kong and therefore is shown only for the South Corridor. If foreign trade expands through the improved port in Shantou the growth could be higher. Source: Study estimates. 5.6 Interprovincial Freight. Domestic inflows and outflows of Guangdong Province will increase at about the same rate as foreign imports and exports, but here the North Corridor is expected to play the largest role (Table 5.2). 5.7 Intraprovincial Freight. There are two types of intraprovincial freight flows: (i) medium distance intercity flows and (ii) local traffic. The west and south are expected to have the highest growth rates. The South Corridor will continue to handle over 60 percent of intraprovincial flows in the future (Table 5.3). 5.8 Total major corridor freight flows will increase fastest in the South Corridor with the West and North close behind (as summarized in Table 5.4 and given in more detail in Table 40). The average growth in the South is expected to be 9 percent to 21 percent between 1987 and the year 2000, while the slowest growth is forecast for the East Corridor at about 60 percent of  - 54 - Table 5.2: INTERPROVINCIAL TRAFFIC IN MAJOR CORRIDORS (million tons) Pessimistic Base case Optimistic Actual forecast forecast forecast Growth Growth Growth 1987 2000 rate (Z) 2000 rate (%) 2000 rate (%) North Corridor 52.6 130.8 7.3 200.1 10.8 476.3 18.5 West Corridor 8.6 L9.0 6.3 27.8 9.5 59.5 16.1 South Corridor /a a n.a. /a n.a. /a n.a. East Corridor 2.3 7.0 8.9 11.4 13.1 29.8 21.8 Total 63.5 156.8 7.2 239.3 10.8 565.6 18.3 n.a. = not applicable L Interprovincial flows irLto Guangzhou/Huangpu are considered north corridor flows, even if they arrive by ship via the Pearl River Delta, since North China is the dominant source. Table 5.3: INTRAPROVINCIAL FREIGHT IN MAJOR CORRIDORS (million tons) Pessimistic Base case Optimistic Actual forecast forecast forecast Growth Growth Growth 1987 2000 rate (%) 2000 rate (%) 2000 rate (Z) North Corridor 5.00 11.2 6.4 15.0 8.8 29.4 14.6 West Corridor 25.96 74.4 8.4 115.7 12.2 271.8 19.8 South Corridor 53.29 160.2 8.8 273.0 13.4 688.6 21.8 East Corridor 12.66 23.7 4.9 33.9 7.9 66.8 13.7 Total 96.91 269.6 8.2 437.6 12.3 1,056.6 20.2 those rates. Growth in the East Corridor will range from 6 percent to 16 per- cent; without new resource development (i.e., without new rail service) it will drop from 7.5 percent to 5.5 percent of total flows. The provincial average growth of freight traffic is expected to be between 8 percent and 19 percent from 1987 to 2000, depending on the forecast used. 5.9 Induced Freight Traffic. The reduction in transport costs resulting from the Guangzhou-Meixian-Shantou (GMS) Railway, if realized, would be expected to generate a significant amount of induced freight traffic. Road improvements and better access in all parts of Guangdong have been found to  - 55 - Table 5.4: TOTAL MAJOR CORRIDOR FREIGHT TRAFFIC (million tons) Pessimistic Base case Optimistic Actual forecast forecast forecast Growth Growth Growth 1987 2000 rate (Z) 2000 rate (%) 2000 rate (Z) North Corridor 63.5 153.3 7.0 226.4 10.3 517.0 17.5 West Corridor 39.9 108.2 8.0 162.7 11.4 369.7 18.7 South Corridor 87.2 275.9 9.3 432.1 13.1 1,062.1 21.2 East Corridor 15.6 32.7 5.9 48.0 9.0 102.6 15.6 Total /a 200.3 558.8 8.2 857.9 11.8 2,040.1 19.5 /a The total for the province excludes international traffic for the North Corridor (5.9 million tons in 1987 and 11.29 million tons in 2000) which also passes through the South Corridor. Source: Table 40. increase traffic. However, structural factors in the economy have limited the rate of increase in these transport flows. Nevertheless, specific induced traffic flows can be identified in the East Corridor. There transport cost savings of 50 percent would lead to the development of coal mining, iron min- ing with pelletization, and cement plants. These induced flows would increase the volume of major freight in the East Corridor by 6.5 million tons, or 150 percent, if the project were built (see Table 41 for details). This induced traffic would add 1 percent to total provincial flows. Normally, a 50 percent reduction in transport costs would generate induced traffic which is less than 50 percent of normal traffic in a given commodity. However, for certain commodities such as mineral resources and perishable fruits and vege- tables, induced traffic rates are much higher. This is the case for the East Corridor. 5.10 Network Provincial Total Freight Traffic. Traffic along the four major corridors is presently about 28 percent of all provincial traffic, and this proportion is expected to decline in the future. Local traffic is likely to increase faster than long distance traffic, as the economic structure of the economy shifts toward more light manufacturing and per capita incomes rise. Modal shares would also be expected to alter with water (river, coastal, and ocean shipping) and particularly highways, increasing their shares as railways grow more slowly, even when it is assumed that additional rail capacity is built as needed to accommodate increased demand. Railways and waterways are expected to be more specialized in bulk, long-haul trans- port, according to their comparative advantage. The unconstrained forecast is summarized in Table 5.5 below and given in more detail at Table 42A. 5.11 The shift to the highway mode can be expected to increase as a result of several factors in both supply and demand: on the supply side, the  - 56 - Table 5.5: NETWOR: TOTAL PROVINCIAL FREIGHT TRAFFIC: UNCONSTRAINED (million tons) Actual Pessimistic forecast Base case forecast Optimistic forecast Growth Growth Growth 1987 2000 rate (2) 2000 rate (C) 2000 rate (%) Rail 41 (06) 1.09 (05) 8 164 (04) 11 397 (04) 24 Road 532 (72) 1,815 (77) 10 3,192 (81) 15 8,411 (85) 19 River 100 (14) 253 (11) 7 320 (08) 9 628 (06) 15 Coastal ship /a 30 (04) 62 (03) 6 77 (01) 7 141 (01) 12 Ocean ship /b 32 (04) 106 (05) 10 145 (03) 12 308 (03) 19 Total 735 (100) 2.345 (100) 9 3.898 (100) 12 9,885 (100) 22 La This category does not include international shipping. /b International shipping only. Note: Figures in (parentheses) are percent of column totals; air and pipeline are excluded as minor flows. Source: Table 42. 2,000 km of highways already built during the Sixth and Seventh Five-Year Plans, the replacement of many ferries with bridges in the province, the rapid growth of trucks and tractors owned by private enterprises, and the reduction of costs as vehicles and infrastructure are improved, and, on the demand side, the increasing demand for h:Lgh quality, fast, door-to-door service by light manufacturing and export industries, which are expected to be the leading growth sectors. 5.12 Major commodity tonnage shipped by coastal vessels would more than triple by the year 2000 in the base case corridor forecast (although the coastal share of these commodities, such as coal from the north, would decrease from 4 percent to 2 percent of total flows). Ocean shipping would grow even faster and maintain its share of total flows. The share of river traffic is expected to increase, except where parallel rail capacity or road capacity is built. River transport would be used more for bulk commodities such as construction materials, cement, and fertilizer, except in the Pearl River Delta where other commodities would begin to dominate future flows. Rail would grow the slowest of the modes. The lower growth of rail traffic reflects the fact that the types of traffic for which rail is best suited are not expected to grow as rapidly as other traffic better served by other modes, including an increasing share of coal traffic which would move more economic- ally by coastal and inland shipping. Transit traffic through the province is expected to increase at a slower rate than Guangdong traffic, and this will affect primarily the rail mode. 5.13 Rail Capacity Constraints. Capacity constraints will further retard rail traffic growth and the amount of traffic diverted from the railways in the North Corridor (Tables 5.6 and 43) will have a significant impact on the other modes. Under the Base Case forecast the double-track rail from Beijing to Guangzhou in the south bound direction will reach capacity by 1990. North bound trains also will reach capacity by 1995. Without any further increase in rail capacity or any reduction in passenger service, 23 million tons of southbound traffic and 1.7 m:Lllion tons of northbound traffic would be  - 57 - diverted to road and coastal shipping by 1995 under the Base Case forecast. More than twice these volumes would be diverted under the high forecast. This diversion will put pressure on port capacity for bulk commodities and road capacity in the North Corridor for interprovincial traffic. Table 5.6: NETWORK PROVINCIAL FREIGHT TRAFFIC: CONSTRAINED (million tons) Actual Pessimistic forecast Base case forecast Optimistic forecast Growth Growth Growth 1987 2000 rate (%) 2000 rate (Z) 2000 rate (Z) Rail 41 (0.06) 82 (0.03) 6 101 (0.03) 9 164 (0.02) 14 Road 532 (0.72) 1,830 (0.78) 11 3,237 (0.83) 16 8,600 (0.87) 26 River 100 (0.14) 253 (0.11) 7 320 (0.08) 10 628 (0.06) 16 Coastal ship /a 30 (0.04) 74 (0.03) 6 95 (0.02) 9 185 (0.02) 15 Ocean ship /b 32 (0.04) 106 (0.05) 10 145 (0.04) 13 308 (0.03) 20 Total 735 (100) 2,345 (100) 10 3.898 (100) 15 9.885 (100) 24 la This category does not include international shipping. /b International shipping only. Note: Figures in (parentheses) are percent of column totals; air and pipeline are excluded as minor flows. Source: Table 42. 5.14 Freight Haul Distance. The average rail freight haul distance has increased slowly in the past, but over the past ten years it has grown by an average of 6 km each year due to a shift of shorter distance traffic to high ways. Road and water freight transport also show an increase in average haul distance, which can be attributed partly to the shift of the railways shorter distance traffic (which is nonetheless much longer than average road dis- tance) 1/ and partly to faster growth in corridor traffic and interprovin- cial coastal shipping rather than to short distance flows. This trend is assumed to accelerate for rail and road, and continue at present rates for coastal shipping (Table 5.7). For river traffic, the increase in short haul traffic is expected to be about equal to the increase in long haul cargo, due to faster growth in Pearl River Delta traffic. Freight traffic in terms of ton-km therefore will increase faster than total tonnage, as shown in Table 42 for the provincial total and Table 44 for the corridors. C. Passenger Traffic Forecasts 5.15 International Passengers. International passengers can be classi- fied into three groups for forecasting purposes: tourists, business travel- lers, and commuters to the SEZs from Hong Kong and Macao. Available statis- tics do not separate tourists and business travellers, consequently both are 1/ In 1987 the average haul distance by railway in Guangdong was 346 km and by road 30 km. Nationally, it was 662 for rail and 50 for roads.  - 58 - Table 5.7: AVERAGE HAUL DISTANCE (km) Year Rail Road River Coastal 1987 346 30 1951a 1,900/a 1990 400 35 180 2,000 2000 565 40 170 2,200 L/ Approximate estimate. forecast at the same rate here.2/ Future tourism and business trips are forecast at the recent growth rates of 25 percent per year, although a lower long term rate can be expected as foreign exposure to China becomes less of a novelty. The number of Hong Kong commuter passengers is expected to grow with the growth of the SEZ economies, and the Pearl River Delta. This means short- term growth of 13-17 percent and medium-term to long-term growth of 7-13 per- cent, depending on the scenaLrio (low or high) and destination. 5.16 Interprovincial Passengers. The demand for interprovincial travel in Guangdong Province in the past has been suppressed by limiting the number of available passenger trains and air flights. The policy behind this limita- tion is changing, and individuals are taking long trips, as shown by the faster growth in pass-km than in passenger trips. This effect is accentuated by the development of the Special Economic Zones which attract business tra- vellers and temporary workers from other provinces with much higher trip rates per capita than other provincial residents. This means that past travel rates per capita are not a good indicator of future rates, which are likely to increase faster. Now that the double-track line is completed the rail capa- city constraint is temporarily released, and pent-up demand for interprovin- cial passenger travel may bring up interprovincial trip rates to a level 50 percent higher than in 1988 although the effect of the 115 percent average increase in passenger tariffs in September 1989 will dampen this somewhat.3/ If aviation services are also made more available, increases of 30 percent to 50 percent in the use of air travel are foreseen.4/ 2/ More data on these passengers should be collected to allow better fore- casting in the future. 3/ A careful analysis of the future rail traffic using origin-destination studies is recommended to aid in future forecasts. Also the present trade-off between rail freight and passenger trains is not directly linked to demand, but is a decision of the railway administration that could be changed. 4/ Based on an assumed elasticity of 1.9 with respect to per capita income growth, which is similar to the elasticities observed in other countries for long distance trips. Air travel elasticities with income are expec- ted to be even higher than rail elasticities due to higher values of passenger time.  - 59 - 5.17 Intraprovincial Travel. The same level of suppressed demand does not apply to intraprovincial travel; there the rapid spread of new road trans- port services has already led to increasing travel rates. Growth rates on some roads have no doubt been pushed up to some extent due to capacity limits on the railroad, but they also can be related to increasing incomes. As Guangdong travellers place a high value on time and favor faster, more conve- nient transport in the future, the relative share of water transport in pas- senger travel is assumed to decline or stagnate. The only exception to this are high speed services such as hydrofoil, which already have increased trip rates between the West Bank of the Delta and Hong Kong. The effects of the SEZs on population movements within the province (e.g., temporary workers) will, moreover, tend to increase local trip rates over and above the income effects. In forecasting passenger travel within the province, the same income elasticities (1.925) as for interprovincial travel were used, although these represent somewhat higher levels than those obtained by the use of regression analysis for past trends. These assumptions produce total trip growth rates which vary by corridor as summarized in Table 5.8 below (and shown in more detail in Table 45). Table 5.8: FORECAST GROWTH IN PASSENGER TRIPS BY TRANSPORT CORRIDOR: BASE CASE (millions) Pessimistic Base case Optimistic Actual forecast forecast forecast 1987 2000 Growth 2000 Growth 2000 Growth Trips Trips rate (%) Trips rate (%) Trips rate (%) North Corridor 30.6 69.4 7 102.2 10 247.7 18 West Corridor 53.4 165.1 9 265.5 13 781.6 23 South Corridor 110.5 473.9 11 797.0 16 2,427.3 27 East Corridor 40.8 97.8 7 158.7 11 407.6 19 Total 235.3 770.2 10 1,343.4 14 3,865.2 24 Source: Table 45. 5.18 Based on 1978-87 data, passenger traffic over the medium term is expected to grow much faster for the road and air modes than for other modes (Table 46). This is true despite the effect of pent-up rail demand for long distance travel, which is expected to expand passenger traffic on the North and South Corridor rail lines after the recent double tracking. The elasti- city of passenger trips with respect to income appears, on the basis of avail-  - 60 - able data, to be about 1.6. Increasing use of the road and air modes as described in Chapter III indicates a higher elasticity for these modes.5/ 5.19 The corresponding (unconstrained) forecasts for province-wide pas- senger trips (Table 5.9) show annual growth of close to 10 percent in the pessimistic case, 15 percent in the base case forecast, and 25 percent in the optimistic forecast. Total provincial traffic in terms of pass-km is expected to increase slightly faster than total trips (Table 47). This is due to the greater proportion of interprovincial trips which will result from the North Corridor rail capacity expansion (double tracking) completed in 1988, and to a greater proportion of longer trips, particularly by road, taken within the province as a result of changing patterns of employment. However, the growth of traffic (freight and passengers) will soon overtake the 1988 capacity expansion, and it is expected that nearly 6 million passenger trips will be diverted from rail to road in the North Corridor by 1995, and nearly three times that number by the year 2000 if rail capacity is not further expanded (Table 48). Table 5.9: TOTAL PROVINCIAL PASSENGER TRAFFIC FORECASTS 13Y MODE - UNCONSTRAINED Pessimistic Base case Optimistic Actual forecast forecast forecast 1987 2000 Growth 2000 Growth 2000 Growth Trips Trips rate (%) Trips rate (Z) Trips rate (Z) Total trips (million) Road 642 2,297 10.3 4,171 15.5 12,556 25.7 Rail 36 94 7.6 133 10.6 293 17.5 Water 28 54 5.2 75 7.9 137 13.1 Air 3 18 14.5 36 20.9 119 32.4 Total 709 2,463 10.1 4,415 15.1 13,105 25.1 Total Pass-km (billion) Road 30 138 12.4 250 17.6 753 28.0 Rail 6 24 11.4 35 14.4 76 21.6 Water 2 4 5.2 6 7.9 11 13.1 Air 3 16 14.5 32 20.9 106 32.4 Total 41 182 12.1 323 17.1 946 27.2 Source: Table 47. 5/ This preference is rela,ted to the relative price of the modes, as dis- cussed in Chapter IV, aLnd also may be influenced by rail capacity con- straints.  - 61 - D. Vehicle Traffic 5.20 Passenger trips and freight shipments by road are directly related to vehicle traffic. The total pass-km growth is expected to be the same as growth in bus and passenger car-km. Similarly freight ton-km by road will be converted to truck-km. The rate of growth in vehicle traffic depends, of course, on the future vehicle mix. For example, larger trucks are expected to increase their share of total traffic in ton-km, and therefore truck traffic growth is expected to be slightly slower than growth in road transport ton-km in Guangdong Province as the average loading increases. E. Summary of Corridor Forecasts 5.21 Freight. The freight forecasts by corridor show a narrow range of growth rates, from 15 percent to 20 percent in the optimistic forecast and 9 percent to 13 percent in the Base Case forecast. However, north-south coastal shipping will be growing faster than the other modes except roads (see Table 40) due to the expected diversion of bulk commodities from rail to ship. Coastal shipping will overtake the Guangzhou-Beijing Railway in total inflow tonnage between 1995 and 2000 in the Base Case forecast and between 1990 and 1995 in the optimistic forecast if there is no further increase in rail capa- city. North-south flows will continue to dominate traffic patterns in the province despite relatively high growth in the West Corridor; the North and the South Corridors are expected to maintain about 30 percent and 40 percent, respectively, of total corridor traffic, while the West increases and the East reduces its share of total traffic. Within the Pearl River Delta, the East Bank will grow fastest. This subcorridor must deal with the growth of the Shenzhen economic zone and the increasing demands on transport by light indus- try. Local road traffic probably will be growing even faster than corridor traffic. 5.22 There are four possible strategies for dealing with capacity con- straints in the North Corridor. One is to keep the newly established limit on rail capacity (30 million tons in each direction), expand shipping capacity to absorb the major bulk commodity movements, and expand road capacity to absorb the other commodities. A second strategy would be to increase rail capacity to 50 million tons each way, e.g., through measures to increase train size, electrification, etc. A third strategy would sacrifice passenger train ser- vice in favor of more freight trains and increased freight capacity. The selected strategy should be consistent with the conclusions of Chapter IV, that a combined rail-ship movement of coal and other bulk commodities to Guangdong is less costly than the all-rail movement. The forecast assumes that port capacity can be expanded over time as needed so that port congestion will not be a problem and that decisionmakers involved in traffic allocation for these commodities will change their allocations in response to economic costs even if prices are not changed accordingly.6/ 6/ The outflow (northbound) from Guangdong by coastal ship in this corridor shows a high growth rate starting from a very low base. This could be facilitated by having ship transport enterprises use marginal-cost pricing (to reflect empty backhauls). Similar pricing policies would encourage better utilization of empty backhauls by rail.  - 62 - 5.23 The slow growth of rail transport in the West Corridor is offset by much higher growth in river transport. Rail and river transport do not actu- ally compete in this corridor as rail serves primarily as an entry and exit point for other provinces to the port city of Zhanjiang.7/ The growth of road transport in this corridor will probably be much higher for intraprovin- cial flows. The planned rLew local rail section from Yaogu to Maoming will change this relationship by adding more rail capacity in this corridor, in competition with road and coastal shipping. The new rail section is to be financed by the Asian Development Bank. 5.24 The East Corridcor shows moderately high growth in road traffic, resulting from the development of light industry on the periphery of the Hong Kong-led development in th,e province. This reflects a continuation of the present trends. The growth of the Meixian subcorridor is expected to be the slowest in the Province in the base case forecasts. 5.25 Passenger TriDs. Passenger trip forecasts by corridor show a dif- ferent pattern from freight forecasts, as the South Corridor accounts for almost half of the passenger trips and is growing faster than the average for the province. The West also is growing rapidly and is expected to expand its share of passenger trips. Passenger traffic in the North Corridor is the smallest and slowest growing of the four. Road traffic there is expected to dominate, but air traffic also will increase if capacity is available. Rail passenger traffic is expected to expand rapidly with track capacity and there- after remain constrained in the absence of any further capacity expansion. F. Imp:Lications for CaPacity Constraints 5.26 Capacity constraints are already evident throughout much of the net- work. Growth rates in freight and passenger transport demand over the next ten years could result in a 250-500 percent increase in total demand on the transport system. Short term projects to meet demand by increasing road and rail capacity in major corridors are under way, but are not sufficient to accommodate the amount of growth anticipated. Therefore, the present conges- tion will certainly get worse. Higher growth for road transport in particular will lead to increasingly severe and more pervasive congestion throughout the network since most slack capacity already has been absorbed by the rapid growth in recent years. 5.27 The growth rates of total road traffic have been 20-30 percent per year and have been accelerating throughout the province since 1980. If the supply of vehicles remains relatively plentiful and additional road capacity is provided, these extraordinary growth rates could continue unabated even if there were substantial increases in railway capacity. Growth will continue due to three factors: (1) the apparent magnitude of pent-up demand, (2) the greater flexibility of road transport compared to other modes, and (3) the changing economic structure of Guangdong Province, particularly the rapid growth of light manufacturing. These factors imply a major need for expanding highway capacity, particularly in high growth areas such as Shenzhen. There a 7/ Completion of the Yaogu-Maoming line will probably increase transit traf- fic in the future, but this has not been explicitly forecast in this report.  - 63 - 25 percent annual average growth rate in road traffic will double demand in just over three years. The effect of this increase on traffic congestion is exacerbated by the mix of different speed vehicles in the traffic flow; extremely high vehicle operating costs are anticipated until more traffic separation is achieved. 5.28 The share of transit traffic by mode will change in the future as provincial rail flows crowd out transit flows from other provinces, particu- larly those which are not in the natural hinterland of Guangdong Ports unless rail managers consciously increase interprovincial import/export traffic in the face of increasing local demand. This will reduce the share of transit traffic tonnage in rail flows from 15-20 percent at present to 10-15 percent by 1990, and 5-10 percent by the year 2000. Even though the transit flows will increase in terms of absolute tonnage, they will be lower in proportion to the rapidly growing Guangdong import and export flows. Highway transit traffic in the North Corridor will increase from virtually nothing to a sig- nificant tonnage even if only part of the excess rail traffic diverts to road. It would increase even faster if an interprovince expressway system were introduced, a prospect which is not envisioned in these forecasts.8/ 5.29 International traffic flows are expected to increase, especially for road and river traffic to Hong Kong. Port congestion is expected to be sig- nificant, especially since coastal shipping will increase dramatically under all forecasts. This congestion is temporarily being addressed by the con- struction of small ports in Shenzhen and Zhuhai, but these ports will not be able to cope with major flows of bulk commodities and containers. The role of Hong Kong and other existing and potential ports in the region must be exa- mined as possible international and/or domestic load centers for (1) petroleum products, (2) coal and (3) containers. This may require a major structural change in the economic roles of ports of the South China region. 8/ If the proposals for an interprovincial highway system are accepted by the State, then the modal shares will need to be restructured and further analysis of the consequences for port access and intermodal transfers will be required.  - 64 - VI. POTENTIAL TRANSPORTATION INVESTMENTS 6.1 This chapter broadly identifies investment implications of the con- clusions of Chapter IV on the future role of the transport modes and Chapter V on transport capacity in comparison with future transport demand in Guangdong Province. It also presents conclusions drawn from preliminary economic evalu- ation of those projects or classes of projects--in the roads sector and for the GMS railway--for which more detailed information was available and sug- gests strategies that can lead to further clarification of investment priori- ties. Although these conclusions were intended to be comprehensive, certain weaknesses and gaps in the available data have led to only partial results and more detailed study of the individual components will be required to more definitely establish priorities. Investment Evaluation Techn:iques 6.2 In the absence of budget or other constraints, economic rate of return (ERR) and net preseni: value (NPV) are the criteria normally used to establish project feasibilit:y, while first year benefits (FYB) are frequently used to establish the optimum year for making the investment. However, an extension of these simple methods is required where budget and/or other con- straints restrict the size or composition of the investment program from the optimum level or structure. 6.3 The methodology used should ideally have the capability to recognize both multiple types of constraints (not only financial constraints but also physical constraints, e.g., shortages of material or equipment in the con- struction industry) and multiple time periods (so as to incorporate allowances in decisions made in the present budget period for the fact that such con- straints will probably exist., and may even worsen, in the future). Formal methods for dealing with multiconstraint, multiperiod transport investment problems have been developed, in the World Bank's Expenditure Budgeting Model (EBM).1/ However, due to data and time constraints it has not been possible to apply such formal methods in the present study. 6.4 Simpler, approximative methods involving two criteria have been applied here. First, major projects or project classes have been ranked by their ratios of net present va:Lue to capital costs (NPV/CC), since the highest ranking projects under this measure would normally be included in the group which as a whole maximizes NPV within a limited capital budget. However, the ratio NPV/CC is not by itself a sufficient ranking indicator, since particu- larly in a high growth situation such as Guangdong Province, low benefits in the early years of a project may be compensated by the rapid growth of bene- fits in the subsequent years to such an extent that the project yields a high NPV/CC ratio, when nonetheless low benefits in the initial years indicate that it should be postponed in order to maximize return on the limited investment 1/ Although originally developed as a planning tool for the highways sector, EBM is a general purpose tool applicable to similar constrained-optimiza- tion problems in any sector. See Chapter 8 of Thawat Watanatada, et al., The Highway Design and Maintenance Standards Model, Vol. I, Description of the HDM-III Model (Johns Hopkins University Press, 1987).  - 65 - budget. A second test has therefore also been applied. The second test requires that a project also meet the criterion that its opening year benefits exceed the social rate of discount (assumed to be 10 percent) of the initial investment (i.e. FYB > 10 percent). Highway Investments 6.5 The role of the highway sector must expand in order to satisfy the burgeoning demand for faster, more flexible transport stemming from the rapid growth of exports and light industry, and associated rising incomes. Also (as shown in Chapter IV) upgrading of the highway system and introduction of the full array of modern vehicles (including very large trucks) could be expected to lower highway transport costs to about half their present levels, so that substantial new traffic would be generated or diverted from other modes to highways. Additional efforts to decongest the railways by diverting traffics more economically served by highways--particularly less-than-carload (LCL) freight and short-distance passengers and freight--will further increase high way demands. 6.6 Meeting these expanding demands will require extension of the net- work for both trunk highways and local roads as well as upgraded capacity in the existing network. In addition, the introduction in recent years of heav- ier trucks from abroad (although still well below the upper weight limits of modern vehicle technology employed in other countries) has given rise to greatly accelerated pavement deterioration, for bituminous pavements in Guangdong Province (as elsewhere in China) were not generally designed to withstand the heavier loadings of modern trucks. If timely action is taken, this deterioration can be arrested by pavement strengthening measures for only one-fourth to one-third the cost of the reconstruction which would otherwise be required; in many cases that will be the preferred solution, and provision for a major pavement strengthening program in Guangdong Province should defi- nitely be made. However, in certain segments of the network where existing roads suffer not only from weak pavements but also from poor geometric stan- dards (width, horizontal, and vertical alignment) and congestion due to heavy volumes of slow-moving local traffic (pedestrians, bicycles, and tractors) the preferred solution may be to construct a new road on an entirely new alignment to serve fast-moving, through traffic, and to relegate the existing road to local traffic needs. These and other strategic options could be comprehen- sively evaluated, in a more extensive future study, using tools such as the World Bank's Highway Design and Maintenance Standards Model (HDM-III) in order to establish highway program priorities on a more rigorous basis. While the present study has applied simpler methods, several important conclusions none- theless have emerged. 6.7 A variety of potential highway improvement projects to meet future transport demand or solve existing bottlenecks were identified by the GPPC and the ICT Interim Report as shown in Table 49a, based on input from several sources. These projects were in some cases identified initially by local authorities or by the Guangzhou Department of Communications, and forwarded to the GPPC, which plays a coordinating role for investment projects in the Prov- ince. The Special Economic Zone offices (now part of the municipalities in Shenzhen, Zhuhai and Shantou) also identified transport projects in their zones. Strategic analysis at the province level identified the need for expressways linking Hong Kong to the Pearl River Delta, and linking the Delta  - 66 - to the other Provincial centers of Zhangjiang and Shantou. Other components were jointly identified by the World Bank, ICT, and GPPC. 6.8 Economic benefit-cost analysis shows that many of the proposed high way improvements have a very high ERR and NPV/CC ratio in Guangdong Province (see Table 50). Based on 10 percent as the social rate of discount for dis- counting future benefits and costs, pavement improvements for relatively high traffic volumes rank highest, followed by a variety of capacity improvements and extensions of the road network. This preliminary economic benefit-cost analysis was carried out based on preliminary engineering estimates of con- struction costs (or where specific cost estimates were not available, general cost functions related to terrain, type and class of road), vehicle operating costs, road maintenance costs and traffic composition and volume. 6.9 These calculations also yield results in terms of economic traffic thresholds 2' for typical improvements as summarized in Table 6.1 below. It must be understood that these thresholds are approximate: they are based upon assumed typical values for t:he costs of the indicated improvements and will vary as actual costs vary. They constitute necessary but not sufficient con- ditions for economic justification. In particular it must be recognized that given the shortage of transportation today, coupled with current budgetary constraints, the economic opportunity costs of capital in the transportation sector in Guangdong may well exceed 30 percent per annum. While using such high discount rates to establish NPV/CC rankings is not an appropriate method for establishing priorities under budget constraints (see Watanatada et al., 1987), it does give a rough indication of the impact of those constraints. This can be seen by comparing Table 6.1 part A (calculated at 10 percent social rate of discount) with part B (30 percent discount reflecting more the opportunity cost of capital, i.e., the expected rate of return on capital from competing investment opportunities which must be forgone). 6.10 It is clear from the fact that there are roads in the Guangdong Province that can be classified in the top categories of this threshold matrix, that pavements are not now near an optimum configuration, and that a major effort is needed to develop a lower-user-cost and less-maintenance- intensive system.3/ In the short term, a priority ranking of existing road improvements can be established either by calculating NPV/CC as described 2/ Traffic level where vehicle operating cost savings and maintenance cost savings (if any) and induced traffic benefits equal 10 percent of invest- ment costs for the year of opening (FYB > 10 percent). See the annex for an estimate of construction costs, road maintenance costs and vehicle operating costs by terrain type and class of road. 3/ In fact there is a major potential deterioration problem in the province as roads designed for much lighter vehicles are now subjected to traffic by modern larger trucks. This will be exacerbated as more imported trucks with heavier axle loads are used on the roads. Unless the pave- merits of major roads are strengthened (as is happening on some highways with cement pavements), or entirely new roads are constructed to serve the main traffic streams, thus freeing existing roads to serve predomi- nantly lighter local traffics, the potential for a disaster in pavement management is present.  * 67 - Table 6.1: MINIMUM TRAFFIC THRESHOLDS (AADT) FOR TYPICAL ROAD IMPROVEMENTS Terrain Type Project type From Flat-rolling Hilly Mountainous A. At 10% (Social Rate of Discount) Expressway /a Class II 9,700 7,100 3,200 Class I road Class II 7,600 6,600 9,200 Class II road Class III 5,200 5,800 3,900 Class III road Unpaved 470 630 720 Pavement overlay /b Class I 330 330 330 Pavement overlay lb Class II 240 240 240 Pavement overlay /b Class III 200 150 150 B. At 30% (Opportunity Cost of Capital) Expressway /a Class II 10,600 8,500 n.a./c Class I road Class II 7,600 6,300 n.a. Class II road Class III 16,600 5,800 n.a. Class III road Unpaved 820 1,050 2,690 Pavement overlay /b Class I 640 640 640 Pavement overlay /b Class II 250 470 500 Pavement overlay /b Class III 300 300 300 /a Threshold in this case refers to the part of the traffic potentially divertible to expressway (e.g., 60 percent of motor vehicle traffic on present roads for Guangzhou-Zhangjiang). /b Assuming existing pavement in poor condition (other thresholds assume fair existing condition). Ic Due to capacity restriction-, mountain roads do not achieve positive NPV at 30 percent at values of parameters assumed in these analyses. Note: Threshold calculations exclude safety benefits to cyclists and pedes- trians that would lower the class I and class II threshold which incor- porates bicycle lanes. above or by calculating the ratios of traffic levels to these thresholds. However, in the medium and longer term, the optimum relationship between road construction and maintenance (pavement management) demands a more comprehen- sive analysis of alternate strategies that can best be accomplished through a formal modeling process, such as application of HDM-III. 6.11 Further secondary road construction is also required in fast growing areas to provide access for new industries and residential or commercial areas and for commercial agriculture areas. The priority for this type of construc- tion is best determined by local priorities and area planning for future land use with related traffic forecasts. Innovative fund raising for these improvements is already progressing through Special Economic Zones and special authorities to create toll roads. Nevertheless the Road Maintenance Tax  - 68 - levels may need to increase to handle the future needs. The role of bicycles and tractors needs to be especially examined in this context as efficient means for local transport. 6.12 Economic priority between highway investments in the major transport corridors and investments in fast growing areas is an issue that still needs to be resolved, as these are interdependent investments. Also the strategic linking of the different regions of the province is an equity issue in addi- tion to an economic efficiency issue, as the Shantou and Meixian areas have not benefitted from economic growth to the same extent as the rest of the province. Provincial decisionmakers could therefore raise the priority of these investments, especially if the potential has been identified for induced economic development, as long as they meet a minimum economic return crite- rion. 6.13 A further issue that was not resolved in the context of the present provincial-level study is the strategic value of improved interprovincial highway connections. The corridor analysis suggests that there is an increas- ing capacity constraint in the Northern Corridor which handles a substantial portion of interprovincial traffic, some of which would prefer the flexibility of road transport if good road transport conditions were created. Also the structure of the economy in adjacent regions could change as a result of improved access to Guangdong Province. However, this is a regional issue, and given the high return on other transport investments in Guangdong, it can be addressed in the future when other high priority issues have been resolved. 6.14 Total highway investment that is high priority for the Eighth Five- Year Plan (i.e., with optimum opening year of 1995 or before) could amount to about Y 8.3 billion as shown in Table 50. This includes approximately Y 2.7 billion for paving 5,000 kin of high traffic, broken-stone roads, and Y 1.2 billion for 4,000 km of high priority pavement overlays, and about Y 4.4 bil- lion for main arteries shown by name in Table 50. Another Y 12 billion could be justified for the Ninth Five-Year Plan (1996-2000). Secondary road invest- ments are not included in this total, but could amount to Y 1 billion over the next five years. Railway Investments 6.15 The main railway investments under consideration by Guangdong Prov- ince (now that double-tracking of the Hunan-Guangzhou-Shenzhen lines is com- plete and the Yaogu-Maoming Local Railroad is under construction), are the construction of the Guangzhou-Meixian-Shantou Local Railroad, the electrifica- tion of the Hunan-Guangzhou-Shenzhen lines, the construction of more passenger stations, and the expansion of the freight yards in the present system (see Table 49b for details). The GMS rail line appears to be capable of generating substantial induced traffic, particularly in iron ore, as discussed in Chapter V. Consequently this line has a relatively high rate of return as shown in Table 50, although not as high as some road improvements, and given its poten- tial contribution to more equal growth in the Province, it can be considered a relatively high social priority, particularly for the first two links between Huizhou and Meixian. Further economic evaluation is needed of practical links to the local economy for the remaining sections of the GMS Railway in order to maximize their economic return.  - 69 - 6.16 The priority and timing of the electrification investments which would increase the capacity of the North and South Corridors depends on the prospects for reducing rail transport demand by shifting coal and other bulk freight to coastal shipping, and the development of parallel road and water transport systems. The faster the development of these parallel systems, the later this rail investment need be made. The optimum timing for this invest- ment, therefore should be investigated in a combined analysis of the least- cost development of all these modes in the Corridor. In addition the shortage of electrical power in the province will require some analysis of the opportu- nity cost of taking power from other uses. 6.17 The freight yard improvements and passenger station investments are at too micro a level to be evaluated in this report. They require some simu- lation planning of rail operations (such as that being developed under the Fourth IBRD Railway Project in the Beijing-Shanghai Corridor) in order to determine the least-cost solutions, including alternate management rules, that would govern the optimum timing and location of these facilities. 6.18 Total railroad investments in Guangdong Province in the next five years will vary considerably, depending on whether or not the electrification projects are included. Without electrification Y 3 to 4 billion would be invested, including GMS, Yaogu-Maoming and Litang-Zhanjiang Railroads. Elec- trification of the Guangzhou-Beijing Line would require an estimated Y 326 million for the portion in Guangdong Province and far larger sums for the sections beyond Guangdong. Less expensive means of increasing capacity would have major benefits if they could be identified. Seaports 6.19 Seaport development is a major ongoing area of investment, with the development of the SEZ ports and the enlargement of Huangpu port. This devel- opment will continue as exports and coastal movements of coal and other bulk commodities continue to grow; this is recognized in the number of port expan- sion projects foreseen by the GPPC as shown in Table 49d. The relative prior- ity of these projects depends on the amount of congestion and ship waiting time in each port. The development of coordinated linear ports along the Zhujiang (Guangzhou/Huangpu and the Shenzhen ports of Chiwan and the future Mawan) and the other SEZ ports are a means of reducing concentration of freight traffic entering the province. Port management reforms may also help increase capacity with a minimum of investment. 6.20 The World Bank members of the study team suggest that two interre- lated issues of great strategic importance for the future are (i) the integra- tion of the Hong Kong and Guangdong port systems and (ii) the selection of the future load center ports for both bulk goods and containers. The port of Hong Kong, one of the largest container handling ports in the world, is now reach- ing the limits of its capacity and future expansion depends on the availabil- ity of alternative sites. Several port expansion possibilities are being dis- cussed in Hong Kong and Guangdong. The construction of the port of Yantian, the expansion of Hong Kong port and the possible construction of a new port for bulk cargoes (coal, ores, grain and oil) in the islands under the Zhuhai SEZ should all be coordinated in order to avoid duplication of capacity and to achieve the economic benefits of specialization. This will require a signifi-  cant study effort in coordination with the ongoing long term Port and Airport Development Study (PADS) in Hong Kong. 6.21 There is also a question as to the extent to which they very large domestic movements (especialLy coal) should be concentrated in hub and spoke operations or through services. Domestic coal and oil shipments by coastal vessels could be focussed pr:imarily in one deep-water port in order to use larger vessels for the sea voyage and then transshipped to smaller feeder ves- sels to serve inland ports. The alternate strategy, particularly for coal, would be to introduce modern ocean-going barges which could traverse both the coastal seas and the inland rivers, thus avoiding the need for transshipment altogether. Further study is required to determine which of these alternative strategies would be most economic. 6.22 Total sea port investment needed in the next five years is substan- tial. A range of Y 2.5-3 billion was estimated for the ports listed in table 49e, dominated by Huangpu, Yantian and perhaps Shantou, although the high cost of dealing with difficult silting conditions there indicate the need for a detailed review. On a risk analysis basis port expansion may be the best type of project, since waiting time costs have major effects on total shipping cost, but the cost per ton of port capacity is relatively small. Conse- quently, port projects have a higher ranking when the expected value of NPV with traffic variation is tak;en into account. The indirect benefits of port projects, especially coal ports, are also high in the present circumstances in Guangdong Province.4/ Inland Water Investments 6.23 Inland water transport investments include port development, channel dredging, and fleet purchase. River port development is now in the hands of local authorities on a self-financed basis, and each port has an expansion plan (see Table 49c). River ports are reputed to have deteriorated in the past as some traffic shifted to other modes (e.g., road) along the tributaries of the Zhujiang. However, in the Pearl River Delta, there are several new ports, and there appears to be joint venture (Hong Kong-Guangdong Province) capital available for river vessels, including container barges and hydrofoil passenger services linking the Delta to Hong Kong. These investments should be encouraged as they seem to fit the specific needs of Delta shippers and passengers and create efficient multimode transfer points. The greater spe- cialization of inland waterways in the transport of bulk commodities and con- struction materials, rather than general cargo, is also a trend that needs specific recognition in future river port development outside the Pearl River Delta. Care must be taken in, determining the need for port development and related benefits, to separate oiat the growth cargo from the declining cargo types. 6.24 Channel dredging has been identified as a potential investment in several river stretches (see Table 49d), in order to increase the size of vessels using the stretch or to extend the navigation season. In the case of the Xijiang, deepening to allow up to 1,000 dwt barges is in progress. The 4/ This conclusion does not necessarily hold for new port construction with major infrastructure costs in addition to berth construction costs.  - 71 - main Pearl River channel to Huangpu Port is also proposed for deepening to allow fully laden 35,000 dwt ships into the port on a regular basis. In other cases (e.g., East River) the season has been shortened by up river dams with water releases uncoordinated with transportation needs. An analysis of the economic advantage of channel dredging (e.g., savings of shipping costs), and related maintenance dredging, is necessary to establish the relative priority of these projects, and this has not yet been carried out. In addition the possibility of multipurpose dam management, as practiced in other countries, may be able to create improved navigation conditions without new investment. Further analysis of this type should be used to establish the relative prior- ity of these investments,5/ and the size of the investments needed. Total inland river investments, however, for the next five years should be less than Y 0.5 billion. Airport Investments 6.25 Six airport projects have been identified in the Province, with one new airport (Shenzhen) and several expansions. It is clear from the traffic forecasts that long distance air travel and possibly medium distance air travel should grow rapidly. In particular the results of the analysis of the role of the modes in Chapter IV showed that there should be some significant shift of longer distance rail passengers to air both to permit more efficient use of the rail system and to facilitate economic interaction and market inte- gration in China. In turn this will allow larger aircraft and lower cost air travel. Consequently airport expansion for passenger travel will have a high priority whenever congestion becomes evident. Air freight will also grow as an adjunct to passenger travel. 6.26 The World Bank members of the study team suggest that the main stra- tegic consideration for airport development in the Province is the future integration of Hong Kong airports into the provincial system. In particular, some effort is needed to coordinate the planning of the new Shenzhen airport with the development of new Hong Kong airports that is now being planned in the PADS program. 6.27 Total airport investment in the next 5 years could amount to Y 0.5 to 1 billion, of which half is destined for the new Shenzhen airport. High benefits will accrue to means of expanding capacity without new runway con- struction. Multimodal and Intermodal Investments 6.28 Integrated intermodal services are still in their infancy in Guangdong province, but there is already a recognition of their importance in the export industries, and the growth of light industries is likely to give rise to demand for domestic integrated services in the future. Some organiza- tional investment is needed to facilitate the use of single integrated bills of lading for through freight using more than one mode. This may require the development of a coordinated communications system; the Ministry of Railways has begun some efforts in this direction and the Guangzhou South Station has 5/ A feasibility study of the Huangpu channel deepening is being undertaken under the ongoing Huangpu Port Project financed by the World Bank.  - 72 - developed a system of substations in cities away from the rail line to coordi- nate freight shipments. Similar experiments such as the multimodal service of the Jiangmen Bureau of Commlnications, and individual freight forwarding com- panies and SINOTRANS have aLso been made, but there is a long way to go to achieve the ability to ship anywhere in the country with one bill of lading. 6.29 The encouragement of freight general cargo unitization in standard containers, through pallets, or other more recent cargo handling innovations is another aspect of intermodal management which could benefit the province and the country as a whole. China has been slow to introduce cargo unitiza- tion compared to other countries and the present extremely low levels of con- tainer penetration (6 percent of containerizable cargoes on the waterways and much less in the other modes) is an indicator of savings in door-to-door costs that are not now being realized by shippers in China, and is also a measure of unnecessary barrier to integration of China's exportlimport industries into the world markets. Although partly due to institutional factors,6/ it is also partly due to physical deficiencies in infrastructure and equipment, par- ticularly limitation in raiLways and highways connecting ports with their hinterlands. Railways lack the necessary rolling stock and station handling equipment; tunnel and bridge clearances are also a problem on particular routes. Movement by highways is restricted mainly to primary routes where road width and alignment permit access to the necessarily large vehicles. Design of future infrastructure and services must make provision for a sub- stantially greater role for cargo unitization in the future. Total Transport Investment 6.30 The total investment in high priority needs for the next five years described above, amounts to Y 19-23 billion in 1988 Yuan, composed by mode as shown in Table 6.2. Clearly the majority of investment is needed for road construction, although rail and port needs are also important. The proportion of investments would change substantially if rail electrification is included, and this places major import:ance on considering lower-cost alternatives for increasing system capacity, both on rail and other modes, for interprovincial traffic. Investment Financing 6.31 At the past levels of transport investment funding (Y 1.2 to 1.4 billion per year according to Table 30) less than 40 percent of high priority needs will be met over the next five years. Part of the additional resource requirements have already been filled by resorting to toll roads with Hong Kong and other financing sources for expressway construction, which account for an important part (Y 3-4 billion) of the highway funding requirements. Private financing from Singapore will also finance port development at Mawan (about Y 400 million). This still leaves a financing shortfall of Y 7-12 billion. 6.32 Additional taxes will be needed in Guangdong Province to help finance the priority road program. The present major sources of revenue for 6/ See Chapter II, para. 2.37 for a broader discussion of containerization issues.  - 73 - Table 6.2: PROSPECTIVE TRANSPORT INVESTMENT PROGRAM billion Yuan z Road /a 18 - 21 70 - 75 Rail /b 3 - 4 12 - 16 Inland Water less than 0.5 less than 3 Seaports 2.5 - 3 8 - 10 Airports 0.5 - 1 1- 4 Intermodal less than 0.5 less than 3 Total 24 - 30 100 /a Not including local roads. /b Not including electrification of the Guangzhou-Beijing line. highways are the road maintenance fee and the motor vehicle purchase surcharge which raise approximately Y 300 million per year. However only 30 percent of these funds are available for new construction, amounting to approximately Y 500 million over the next five years or 10 percent of the nontoll road needs. These revenues are increasing with the vehicle fleet, but will not increase funding to the levels needed for financing the above road program unless the basic tax rates are increased. 6.33 Local railway and port project funding, on the other hand, can be substantially self-financing, with private sector and Bank loans supported by user charges. The recent increase in rail tariffs will provide a source of funds for investment, and the rates to be charged by the new Guangdong local railways, as approved by the central government, are designed to recover all costs and pay back capital over 10 years. It is likely that external finan- cing can be raised if these organizations are set up on a financially autono- mous basis. Similarly, seaport and inland port expansion are being carried out on a cost recovery basis with port charges increased to achieve a reason- able payback period. Singapore financing of Mawan Port and the World Bank loan for Huangpu/Guangzhou improvements are cases in point. These loans will increase the overall funds available for port expansion in the next five years by about Y 1 billion.  - 74 - VII. CONCLUSIONS AND RECOMMENDATIONS 7.1 Several important and interrelated problems confront Guangdong prov- ince transport planners today. The rapid economic growth experienced by the province since 1978 is expected to be slowed and possibly brought to a halt in some locations by rising congestion on the transport system and associated shortages of energy and key materials. Investment in transport infrastructure has not kept pace with inveistment in industry and commerce in the last five years, despite some substantial investments in rail, highway and port improve- ments. Flows in all corridors have suffered, but especially North-South flows and secondary or primary road traffic in fast-growing regions. 7.2 At the same time lprices for certain transport services well below short-run or even long-run incremental social costs have served to generate noneconomic excess demands which clog the transport arteries, and to distort the choice of modes, thus resulting in less than optimal use of the limited available facilities and services. The ability of the transport sector to cope with further economic growth in Guangdong will require well thought-out strategies on the part of GPPC and other provincial operating departments, as well as a major effort (including also central government planning and price authorities) to coordinate transport investment and tariff policy. Without such coordination existing assets will not be used in the most efficient man- ner and badly needed new investments will not be placed in the locations and modes that have the highest social priorities. 7.3 The major revisions of tariffs for freight services in March 1990 constitute a significant improvement in this respect. The gap between freight tariffs and society's long-run incremental costs has been reduced for all the commodities examined here, although a large gap still remains for some, par- ticularly for the important commodity of coal, as tariff increases were not sufficient to close the large gap that previously existed. Moreover, an addi- tional premium may be needed in the short run to facilitate allocation of limited transport capacities to the highest economic uses until increases in transport capacity over the long run establish an equilibrium with demand. 7.4 The increases in passenger tariffs of September 1989 appear to place a high premium above long-run costs, particularly for the railways, and it remains to be seen whether this will establish a short-run equilibrium; over the long run, economic efficiency criteria suggest that rail and bus passenger tariffs should be lower. Aviation fares warrant more detailed study, and some restructuring may be desirable to align them more closely with long-run incre- mental costs to society, plus a premium for the short run on routes where capacity constraints cannot be quickly overcome. 7.5 Although transport planning in Guangdong Province by the GPPC in coordination with the operating departments and authorities must continue to further define the specific priorities and strategies that will be most eco- nomically efficient, certain general directions and recommendations can be identified at this stage in the process. These concern (i) the role of the modes, (ii) investment priorities, and (iii) the planning process.  75 - Economic Role of the Transport Modes 7.6 Several conclusions were drawn from the analysis in Chapter IV despite possible uncertainties in the cost data. These are: (a) Given the changing structure of the Guangdong economy, highways would be expected to play an increasingly large role in the future even at present road transport costs, because the fastest growing segments of the Guangdong economy are industries which are typically prepared to pay premium prices for the speed and flexibility of road transport. The magnitude of the potential reduction in highway transport costs would further expand its scope beyond the presently limited usages and extremely short haul distances. Highways could provide more economic services for short to medium distance passen- ger transport and small lot size or door-to-door freight up to 1,000 km haul distance,l/ thus relieving railways congestion and improving overall transport service quality. Improved interprovin- cial road haulage is a possible future priority once the intrapro- vincial road links are upgraded; (b) The railway should concentrate on its most economic transport roles, which include medium distance freight and passengers and high volume bulk commodities, and heavy freight at all distances; (c) Coastal shipping should coordinate with the railway for transport of coal from the north to Guangdong Province; although recent cost increases for ships appear to have eroded the economic comparative advantage of the rail-cum-coastal routing, the costs are not very different from the all-rail option and, at least over the short run, diversion of this traffic to the sea is needed to relieve congestion on the railway main-lines; (d) Coastal shipping should shift from heavy freight and general cargo to coal and other bulk commodities, except in locations where the sea distance is shorter than rail, where long distance coastal shipping is still competitive; and (e) Air transport should expand its long distance and medium distance services and price them in accordance with incremental costs to society. These services are expected to grow rapidly with economic growth and increasing value of time for business travelers in Guangdong Province and the rest of China. Aviation can make an enormous contribution to the economic efficiency of Chinese business and public administration, as well as continuing to earn important foreign exchange from the tourism industry. 1/ In the view of the World Bank members of the study team, based on the assumption that a modern highway system and competitive market conditions are developed.  - 76 - Investment Priorities 7.7 Total transport investment in high priority projects should reach Y 24-30 billion in Guangdong Province over the next five years in order to approach an economically efficient transport system. The economic return on these projects would average well over 20 percent per annum if they were all implemented immediately, and the returns are growing rapidly each year as con- gestion increases, and shortages in capacity remain. Some classes of projects have rates of return over 100 percent and several over 50 percent. These should be given a very high priority. 7.8 Based on economic efficiency analysis and long run social costs, the investment balance between modes would shift more in favor of roads, which would receive 60-70 percent of total transport investment on the provincial level in the near future. Paving of macadam roads with high traffic levels, pavement overlays and selected expressways and other highway improvements eco- nomically warrant immediate implementation. These should include several Class I and II arteries and the Guangzhou-Shenzhen expressway. Building of the first phase of expressways and special roads from Guangzhou to Shantou and Zhanjiang is also economically feasible at this time. 7.9 GMS railway investment to tie the Meixian area into the rest of the province and encourage mining and cement production has an economic return over 30 percent, and also economically warrants immediate implementation. Railway electrification projects should be deferred while less expensive means to increase capacity are expLored. The priority and timing of the electrifi- cation project investments which would increase the capacity of the North and South Corridors, depend on the possibilities for reducing rail transport demand by shifting coal and other bulk freight to coastal shipping and shift- ing passengers to road and air, which in turn depend on the development of parallel water, air and road and transport systems. The optimum timing for these investments, therefore depends on a combined analysis of the least-cost development of all modes in the Corridor. The expansion of rail container handling facilities and integrated intermodal services should also be closely evaluated as this is a near-term priority for the growing international and interprovincial traffic. 7.10 Seaport developmenit is a major ongoing area of investment, with the development of the SEZ ports and the enlargement of Huangpu port. Projects to expand existing seaports with additional berths should be implemented as soon as waiting times start growing over one day on average, since these projects have major benefits and relatively low risk in terms of transport costs. Port management improvements to gain capacity without major capital costs would have even greater rates of return. Also, the development of coordinated linear ports along the Zhuji.ng and the SEZ ports may be a cost-effective means of reducing port congestion due to the concentration of freight traffic entering the province. Finally, the development of major regional load cen- ters ports for dry bulks and liquid bulks, as well as containers, and asso- ciated feeder services, is a high priority that requires serious study in coordination with Hong Kong. 7.11 The conclusions reached concerning investment in the inland water transport sector are particuLarly tentative due to lack of actual cost-benefit data concerning proposed investments. However, it is clear that investment in  - 77 - the Pearl River Delta should continue to expand to meet the demand. Private and local government investments should be encouraged to meet the specific needs of Delta shippers and passengers and create efficient multimode transfer points. The priority of inland river channel dredging is not clear but may be feasible in the context of multipurpose dam management in some cases, and in the context of long hauls which do not compete with a railroad. 7.12 The air transport sector has not been examined in depth in this report, but the following observations can be ventured: (a) a major expansion of air services, including improvements in airports and air navigation systems for passenger travel, will have a high economic priority for the service of long distance business travelers, tourists and senior government officials; not only would this relieve railways' congestion it would greatly facilitate economic interaction and integration of markets within China and overseas; (b) an important strategic consideration for airport development in the Prov- ince is the future integration of Hong Kong airports into the provincial sys- tem. Particular effort is needed to integrate the planning of the new Shenzhen airport with the development of Hong Kong airports, which should be facilitated by the recent publication of the Hong Kong Port and Airport Devel- opment Study (PADS); and (c) expansion of air freight services, perhaps con- tinuing in the early stages as an adjunct to passenger travel, would appear to be warranted. 7.13 Investments in integrated intermodal services are needed to ensure that transfer of cargo and passengers between modes takes place efficiently. Intermodal facilities and services should receive far more attention in the design of new or expanded ports and rail stations. This is particularly important in Guangdong and other coastal provinces where foreign trade is significant, since the competitiveness of Chinese exports in foreign markets will be affected by the speed and quality of transport services. But it is also important in the domestic markets as well. An experiment to test the feasibility of through pallets to facilitate intermodal transfers of general cargo in the Hainan Island-Guangzhou Corridor has been proposed, and the results of this pilot service should be carefully evaluated as to its wider applicability. 7.14 In general, projects linking all areas of Guangdong to Hong Kong as the growth pole for the province would have benefits in terms of broadening the growth pattern and reducing the concentration of light industry growth in the Pearl River Delta, although further links within the Delta will help sus- tain present growth rates. Further secondary road construction in fast- growing areas is also required to provide access to new industries and resi- dential or commercial areas. The priority for this type of project is best determined at the local government level. Planning Process 7.15 Although the central government still retains a major voice in large capital projects, the increasingly decentralized planning process is moving rapidly in Guangdong Province to reinforce the role of local authorities and SEZs or municipalities, especially with regard to roads and ports. At the same time strategic planning for railways and expressways or arterial roads has devolved increasingly to the provincial level. These trends are intended to allow faster decisions that are more adapted to local needs, but they also  - 78 - require stronger coordination in order to avoid misinvestment in some areas where the national or provincial viewpoint would yield a different allocation of resources than a combination of local decisions. This is clearly the case with respect to new seaport development which requires both provincial coordi- nation with Hong Kong and provincial-level decisions concerning specialized load centers. 7.16 The local authorities have been especially dynamic in conceiving ideas for transport system development, and where there is a potential problem of competing investments among port proposals or between road and rail infra- structure, the capability for evaluating these multiple investments and pro- viding provincially coordinated data for planning at the local level is now needed in the GPPC. The local authorities themselves need enhanced capabili- ties to carry out project analysis and design. 7.17 A specific analysis framework at the provincial level is needed for all the modes so that critical data for planning decisions are identified in a comprehensive, multimodal context rather than strictly within each mode, as is now the case. This framework should be provided by the GPPC to the modal and local authorities and include (a) data on transport demands that relates to future prices of modal services, (b) expected changes in demand levels, (c) the specific potential induced economic growth effects of transport deci- sions, and (d) least-transport-cost analyses, in order to enable the authori- ties and GPPC to arrive at a net cost or benefit to society of each project or strategy. 7.18 Within this comprehensive transportation planning framework each modal planning unit should have the appropriate tools for evaluating projects. In particular, the following transport planning areas need reinforcement: (a) traffic surveying by origin and destination (b) traffic forecasting (c) benefit-cost ana]ysis (d) network analysis (e) project identification (f) priority planning (g) infrastructure management (h) data base coordination for planning and management. By expanding these analytic capabilities, more emphasis can be placed on improving the actual transportation services, and less on infrastructure con- struction by itself. 7.19 In order to implement a comprehensive approach at the provincial level, GPPC should also develop a set of transport strategies that can be articulated clearly. This will allow the local decisionmakers at different levels to support each provincial strategy with local-level decisions, and lead to actions at all levels that are more consistent with the goals and objectives of the GPPC. Examples of such strategies are: (a) the shifting of bulk coal shipments to coastal ship from rail; (b) the connection of growth centers by high quality highways as they become economically viable;  - 79 - (c) the elaboration of highway pavement management systems to achieve the least cost road network; and (d) the development of a system of integrated intermodal transport in the Pearl River Delta to promote export industries. With a careful formulation of strategies and policies consistent with long run social costs GPPC could guide development of the necessary transport infra- structure best suited to support for the continuing development of Guangdong Province.  Table 1: GUANGDONG PROVINCE POPULATION AND LAND AREA BY CORRIDOR Total Urban Total Cultivated Percent of Population Percent Population Percent/a Land Area Land Area Cultivated (ha) (ha) Area CORRIDOR Pearl River Delta 17,391,328 28.08 6,648,158 38.23 1,948,650 430,766 15.71 Guangzhou 8,089,963 3.359,183 400,050 97,052 East Bank 1,727,625 1,487,475 522,150 21,940 West Bank 7,573,740 1,801,500 1,026,450 312,774 East Guangdong 25,963,494 43.42 2,938,730 11.32 5,629,270 1,135,081 41.30 1 Coastal Route 16,741,458 2,001,336 1,656,220 263,948 Meixian Route 9,222,036 937,394 3,973,050 871,133 West Guangdong 9,483,784 15.86 1,590,658 16.76 5,921,100 935,560 34.04 Maoming Route 3,420,871 1,396,887 3,691,900 689,213 Yanan Route 6,062,913 193,771 2,229,200 246,347 North Guangdong 6,961,744 11.64 371,955 5.34 3,373,500 246,140 8.96 Total /b 59,800,350 100.00 11,549,501 100.00 16,872,520 2,748,547 100.00 /a Percent of Urban Population in each Corridor. /b Excluding population and land area of Hainan province and autonomous states. Source: Guangdong Yearbook, 1987.  - 81 - Table 2: LABOR FORCE EMPLOYED BY SECTOR ('000) Guangdong National Total X Total X Farm, Forestry, Fish and Water Cons. 18,247 58 313,110 61 Industry 4,942 16 89,800 17 Construction 1,694 5 22,710 4 Transportation, Port & Telecom. 953 3 13,050 3 Commerce, Catering, Trade, Supply & Marketing 2,074 7 24,850 5 Education, Art, Culture, Radio and TV 684 2 13,240 3 Government parties and organizations 581 2 8,730 2 Other 2,229 7 27,330 5 Total 31,404 100 512,820 100 Source: Statistical Yearbook of China, 1987.  -82 - Table 3: P'RODUCTION OF SELECTED MAJOR COMMODITIES IN GUANGDONG PROVINCE (million tons) Annual Annual growth growth 1978 1985 1986 1987 1988 rate of rate of 1978-85 1985-88 Coal 10.47 8.10 8.57 8.73 9.28 -3.6 4.6 Coke n.a. 0.45 0.46 0.46 n.a. n.a. n.a. Crude Oil a/ - 0.10 0.26 0.45 0.39 n.a. 57.4 Pig iron, steel and rolled steel 0.85* 2.10 2.34 2.67 2.80 13.8 10.1 Cement 3.70 11.20 12.80 14.76 17.46 17.1 16.0 Timber (0.8T/M3) 1.80 3.46 2.80 2.32 n.a. 9.8 n.a. Chemical fertilizer n.a. 0.60 0.64 0.72 n.a. n.a. n.a. Tires n.a. 0.84 1.05 n.a. n.a. n.a. n.a. Sulfuric acid n.a. 0.65 0.73 n.a. n.a. n.a. n.a. Sugar 0.96 1.44 1.93 1.56 1.23 6.0 -5.1 Paper/paper boards 0.28 0.53 0.58 0.70 0.81 9.5 15.2 Salt n.a. 0.58 0.65 0.82 n.a. n.a. n.a. Beverage n.a. 0.36 0.48 n.a. n.a. n.a. n.a. Grain <primarily rice) 15.10 16.04 17.18 17.02 16.37 0.9 0.7 Peanut 0.35 0.57 0.66 0.54 0.52 7.2 -3.0 Sugar cane 8.35 18.31 20.44 13.39 15.39 11.9 -5.6 Fruit 0.29 1.16 2.01 2.64 2.78 21.9 33.8 Pork Q0..48 0.98 1.19 1.14 1.25 10.7 8.4 Fish, seafood & agri. product 0.66 1.09 0.87 1.55 1.75 7.4 17.1 a/ Including oil from Soutlh China Sea in 1986. k/ 1985-1986 growth rate. * Rolled Steel not included. n.a. Not available. Note: Includes Hainan Island data. Sources: Statistical Year Book of China, 1986, 1987. Statistical Year Blook of Guangdong, 1987, 1989. Guangdong Provincial planning Commission.  - 83 - Table 4: INFLOW AND OUTFLOW OF COMMODITIES IN 1987 FOR GUANGDONG PROVINCE (million tons) Inflow Outflow Commodity Amount % Amount Z Coal 13.19 14.1 3.32 4.9 Metal. Ores 2.66 2.8 3.87 5.8 Nonmetal Ores and Sand 2.60 2.8 1.61 2.4 Grain 3.98 4.3 1.90 2.8 Cement 2.60 2.8 0.48 0.7 Iron and Steel 2.33 2.5 2.97 4.4 Crude Oil 9.07 9.7 0.32 0.5 Petroleum Products 7.76 8.3 2.92 4.4 Fertilizer 1.68 1.8 3.51 5.2 Timber 0.40 0.4 1.84 2.7 Others /a 47.27 50.5 44.36 66.1 Total 93.54 100.0 67.10 100.0 la Other category includes 20.74 million tons of outflow and 19.85 million tons of inflow traffic by road and river for which the commodity breakdown is unknown. Note: Estimate based on 1987 total tonnage by mode and 1985 detailed break down by commodity. Source: Study estimates.  - 84 - Table 5: GUANGDONG PROVINCE FOREIGN TRADE WITH HONG KONG Imports from Growth Exports to Growth Year Hong Kong rate Hong Kong rate ('000 tons) (Z) ('000 tons) (Z) 1980 423 3,649 - 1981 723 70.9 4,188 14.8 1982 982 35.8 4,633 10.6 1983 1,703 73.4 5,660 22.2 1984 2,397 40.8 6,237 10.2 1985 3,486 45.4 7,597 21.8 1986 3,988 14.4 9,840 29.5 1987 5,406 35.6 10,792 9.7 Note: These figures excludle any trade with Guangdong by ocean, but include river traffic. Source: Study on Rail Freight Development, Technical Paper No. 1, Mott, Hay and Anderson/Wilbur Smith Associates for the Hong Kong Government Secretariat Transport Branch, September. 5, 1988.  - 85 - Table 6: HONG KONG TRADE WITH GUANGDONG BY MODE AND BY TYPE FOR 1987 -------- Mode ---------- ('000 Tons) OUTBOUND Rail Road River Total I Share Commodity group Textiles, Clothing & Fibers 499.6 740.0 1,239.6 11.4% Food & Beverages 484.5 456.0 940.5 8.7% Metal Products 256.1 291.0 547.1 5.1% Chemicals 99.2 99.2 0.9% Petroleum Products 92.4 75.0 167.4 1.5% Paper Products 80.0 73.0 153.0 1.4% Chemicals & Pharmaculets 35.1 35.1 0.3% Mineral Ores 26.4 26.4 0.2% Machinery & Equipment 27.5 399.0 426.5 3.9% Timber & Wood Products 26.8 36.0 62.8 0.6% Household Appliances 12.8 12.8 0.1% Live Animals 195.8 195.8 1.8% Fertilizers 2.4 2.4 0.0% Sand & Gravel 4,628.8 4,628.8 42.7% Miscellaneous 239.2 897.0 1.157.2 2.293.4 21.2% Total : 2,077.8 2,967.0 5,786.0 10,830.8 100% Modal Shares : 19.18% 27.39% 53.42% INBOUND Textiles, Clothing & Fibers 87.8 596.0 683.8 12.7% Food & Beverages 17.7 38.0 55.7 1.0% Metal Products 36.7 367.0 403.7 7.5% Chemicals 22.3 22.3 0.4% Petroleum Products 23.1 154.0 177.1 3.3% Paper Products 14.5 209.0 223.5 4.1% Mineral Construction Materials 7.8 7.8 0.1% Mineral Ores 1.1 1.1 0.02% Machinery & Equipment 33.6 381.0 414.6 7.7% Timber & Wood Products 9.2 66.0 75.2 1.4% Household Appliances 12.5 12.5 0.2% Miscellaneous 108.8 629.0 2.581.0 3.318.8 61.5% Total : 375.1 2,440.0 2,581.0 5,396.1 100% Modal Shares : 6.95% 45.22% 47.83% Source: Technical Paper No. 1, Study on Rail Freight Development, Hong Kong Government Secretariat Transport Branch, 5/9/88  - 86 - Table 7: COUNTIES AND CITIES WITH HIGH GROWTH POTENTIAL IN GUANGDONG PROVINCE ------ Growth Indicators ----------- Recent Access to Special growth Hong Kong Economic Resource Names Rate (hours) Zone Potential Bao'an 39.25% 02 3 Shenzhen 31.85% 02 1 Dongguan 34.64% 02 to 4 3 Huizhou 45.04% 04 to 06 4 Goasming 46.22% 04 to 8 3 Shansui 36.69% 04 to 8 3 Xinhui 29.87% 04 to 8 3 Zhongshan 29.38% 04 to 8 3 Heshan 21.78% 04 to 8 3 Doumen 18.84% 04 to 8 3 Fengyu (Panpu) 12.81% 04 to 8 3 Nanhai 10.54% 04 to 8 3 Zhuhai 10.20% 04 to 8 1 Shunde 9.33% 04 to 8 3 Fushan 8.23% 04 to 8 3 Jiangmen 8.19% 04 to 8 3 Zhenchang 14.45% 06 to 8 4 Guangzhou 3.92% 06*to 8 2 Hua xian -9.70% 06 to 8 4 Lianping 6.96% 16 to 24 4 H Enping 33.37% 24 to 32 4 Guanging 22.93% 24 to 32 4 Shihui 21.75% 24 to 32 4 Gaouao 20.85% 24 to 32 4 Shantou 19.16% 24 to 32 1 Biqing 18.70% 24 to 32 4 Nan'ao 18.13% 24 to 32 4 Shaoguan 12.81% 24 to 32 4 H Meixian 11.22% 24 to 32 4 H Dapu 7.98% 24 to 32 4 H Wuhua 4.85% 24 to 32 4 H Xinging 1.38% 24 to 32 4 H Fengshun -0.53% 24 to 32 4 H Longchuan -6.67% 24 to 32 4 H Xinxing 26.51% 32 to 48 4 Liajiang 25.83% 32 to 48 4 Yangjiang 25.15% 32 to 48 4 Zhanjiang 23.53% 32 to 48 2 Luoding 22.43% 32 to 48 4 Xinyi 21.64% 32 to 48 4 Deging 21.38% 32 to 48 4 Fengkai 20.73% 32 to 48 4 Yunan 20.72% 32 to 48 4 Wuchuan 20.63% 32 to 48 4 Dianbai 19.51% 32 to 48 4 Huazhou 19.33% 32 to 48 4 Huaiji 19.27% 32 to 48 4 Gaozhou 16.34% 32 to 48 4 H MaominK 15.90% 32 to 48 4 H Notes: High Growth areas selected as having good access to Hong Kong, or high past growth (>18%), or having resource potential, or Special Economic Status. Special Economic Status: 1 - Special Economic Zones. 2 - Economic & Technological Development Zone. 3 - Economic Open Zone. 4 - No special status Resource Potential H - High Source: Guangdong Yearbook, 1987, GPPC and interview  Table 8: CAPITAL INVESTMENT BY SECTOR (Y million) 6FYP (1981-85) 1986 1987 1988 Total (1981-88) Current 1980 Current 1980 Current 1980 Current 1980 Current 1980 Prices Prices Prices Prices Prices Prices Prices Prices Prices Prices Total 14,350 10,920 6,344 3,433 6,950 5,194 9,077 5,960 36,721 25,507 Industry 7,407 5,637 4,467 2,417 5,019 3,751 6,001 3,940 22,894 15,745 Metallurgy 360 274 87 47 87 65 151 99 685 485 Electric Power 1,541 1,173 2,178 1,178 2,508 1,874 1,970 1,293 8,197 5,518 Coal 823 626 42 23 38 28 31 20 934 697 Petroleum 599 456 117 63 318 238 650 427 1,684 1,184 Chemical 594 452 187 101 265 198 967 635 2,013 1,386 Machinery 815 620 434 235 362 271 738 485 2,349 1,611 Construction Materials 542 412 409 221 444 332 517 339 1,912 1,304 1 Forest Industry 89 68 58 31 64 48 116 76 327 223 x Food 1,213 923 346 187 369 276 246 162 2,174 1,548 Textile 639 486 386 209 294 220 285 187 1,614 1,002 Paper 75 57 23 12 31 23 21 14 150 106 Construction and Geology 318 242 153 83 160 120 12 8 643 453 Agriculture, Forest, Water Resources and Weather 2,324 1,768 499 270 392 293 623 409 3,838 2,740 Transport and Telecom- munications 4,301 3,273 1,225 663 1,382 1,033 2,441 1,603 9,349 6,572 Z Transport Telecom. 30.0 30.0 19.3 19.3 19.9 19.9 26.9 26.9 25.5 25.5 Notes: (1) Including Hainan Island which represented 1-2 percent of the total. (2) 1980 prices calculated using average GVAIO deflator for China. Sources: Guangdong Provincial Planning Commission; Guangdong Statistical Yearbook, 1989.  - 88 - Table 9: GROWTH INDICATORS FOR GUANGOONG PROVINCE BY SECTOR BY CORRIDOR -- GVAIO Growth -- ---- AgricuLture ---- ---- ALL Industry ---- ---- Light Industry Pessi- Base Opti- Pessi- Base Opti- Pessi- Base Opti- Pessi- Base Opti- mistic Case mistic mistic Case mistic mistic Case mistic mistic Case mistic Province TotaL 6.6% 9.6% 15.3% 3.0% 3.0% 3.0% 7.3% 10.7% 16.9% 7.2% 12.5% 19.7% Pearl River Delta 6.8% 9.9% 15.8% 3.0% 3.0% 3.0% 7.0% 10.3% 16.4% 6.7% 11.8% 18.8% Guangzhou 3.1% 5.1% 9.5% 3.0% 3.0% 3.0% 3.1% 5.2% 9.7% -6.2% 2.2% 8.4% East Bank 10.3% 14.1% 20.7% 3.0% 3.0% 3.0% 11.0% 14.9% 21.7% 11.7% 16.1% 23.2% West Bank 7.9% 11.4% 17.1% 3.0% 3.0% 3.0% 8.4% 12.1% 18.6% 9.0% 1..0% 21.3% East Corridor 5.1% 7.6% 12.6% 3.0% 3.0% 3.0% 6.1% 9.3% 15.3% 5.3% 10.0% 16.7% Coastal subcorridor 5.9% 8.7% 14.0% 3.0% 3.0% 3.0% 7.0% 10.3% 16.4% 6.7% 11.3% 18.0% Meixian subcorridor 3.2% 5.0% 8.9% 3.0% 3.0% 3.0% 3.3% 6.3% 11.9% 0.0% 5.7% 12.4% West Corridor 7.6% 10.7% 16.6% 3.0% 3.0% 3.0% 9.5% 13.3% 20.0% 11.4% 17.4% 25.2% Maomnong subcorridor 7.5% 10.6% 16.4% 3.0% 3.0% 3.0% 9.1% 12.8% 19.4% 11.0% 17.4% 25.3% Yunan subcorridor 7.9% 11.1% 17.0% 3.0% 3.0% 3.0% 10.6% 14.7% 21.6% 12.2% 17.5% 25.1% North Corridor 4.8% 7.1% 12.0% 3.0% 3.0% 3.0% 5.6% 8.7% 14.6% 2.0% 10.1% 17.9% Source: Annex Tables 2.1, 2.2 and 2.3  Table 10: COMPARISION OF CORRIDOR GROWTH FOR THREE TRANSPORT SYSTEMS - BASE CASE FORECAST Without With With induced growth Hong Kong Expressway Shantou Expressway Growth Growth Growth 1986 1990 2000 Rate 1990 2000 Rate 1990 2000 Rate CORRIDOR Pearl River Delta 39,391 67,744 137,857 9.4Z 67,744 147,507 9.9% 67,744 133,548 9.1Z Guangzhou 16,746 20,349 29,257 4.1Z 20,349 33,643 5.1Z 20,349 30,984 4.5Z East bank 6,425 16,693 42,086 14.4Z 16,693 40,622 14.1Z 16,693 33,311 12.5% West bank 16,220 30,702 66,513 10.6Z 30,702 73,242 11.4% 30,702 69,254 10.9Z East Guandong 11,782 18,587 35,416 8.2% 18,587 32,678 7.6% 18,587 48,498 10.6% 0:0 Coastal route 7,700 13,200 26,801 9.3Z 13,200 24,633 8.7% 13,200 34,471 11.3% Meixian route 4,081 5,387 8,615 5.5% 5,387 8,045 5.0Z 5,387 14,028 9.2Z West Guangdong 15,789 31,876 71,657 11.4Z 31,876 65,642 10.7% 31,876 64,047 10.5% Maoming route 11,279 22,459 50,106 11.2% 22,459 45,989 10.6% 22,459 44,926 10.4Z Yunan route 4,510 9,417 21,551 11.8-Z 9,417 19,653 11.1% 9,417 19,121 10.9% North Guangdong 4,810 7,323 13,538 7.7Z 7,323 12,641 7.1% 7,323 12,375 7.0% TOTAL 71,772 125,529 -258,468 9.6% 125,529 258,468 9.6Z 125,529 258,468 9.6Z Note: Growth rate average from 1986 to 2000.  Table 11: COMPARISION OF CORRIDOR GROWTH FOR THREE TRANSPORT SYSTEMS - OPTIMISTIC FORECAST Without With With induced growth Hong Kong Expressway Shantou Expressway Growth Growth Growth 1986 1990 2000 Rate 1990 2000 Rate 1990 2000 Rate CORRIDOR Pearl River Delta 39,391 80,030 271,842 14.8% 80,030 307,261 15.8% 80,030 267,496 14.7% Guangzhou 16,746 21,910 47,288 7.7Z 21,910 59,782 9.5Z 21,910 52,207 8.5% East bank 6,425 21,143 87,930 20.5% 21,143 89,312 20.72 21,143 68,483 18.4% West bank 16,220 36,978 136,624 16.4% 36,978 158,167 17.7% 36,978 146,806 17.0% East Guandong 11,782 21,536 71,261 13.7Z 21,536 61,680 12.6% 21,536 106,748 17.0% Coastal route 7,700 15,584 54,330 15.0% 15,584 48,154 14.0% 15,584 76,179 17.8Z Meixian route 4,081 5,952 16,931 10.7% 5,952 13,527 8.9% 5,952 30,569 15.5% West Guangdong 15,789 38,847 150,621 17.5% 38,847 135,042 16.6% 38,847 130,497 16.3% Maoming route 11,279 27,304 106,065 17.4% 27,304 94,337 16.4% 27,304 91,307 16.1% Yunan route 4,510 11,543 44,556 17.8% 11,543 40,705 17.0% 11,543 39,190 16.7% North Guangdong 4,810 8,412 26,119 12.8% 8,412 23,561 12.0% 8,412 22,804 11.8% TOTAL 71,772 148,826 527,544 15.3% 148,826 527,544 15.3% 148,826 527,544 15.3X Note: Growth rate average from 1986 to 2000.  Table 12: COMPARISION OF CORRIDOR GROWTH FOR THREE TRANSPORT SYSTEMS - PESSIMISTIC FORECAST Without With With induced growth Hong Kong Expressway Shantou Expressway Growth Growth Growth 1986 1990 2000 Rate 1990 2000 Rate 1990 2000 Rate CORRIDOR Pearl River Delta 39,391 61,101 93,942 6.4Z 61,101 98,462 6.8Z 61,101 91,924 6.2Z Guangzhou 16,746 19,505 23,677 2.52 19,505 25,732 3.1Z 19,505 24,486 2.8Z East bank 6,425 14,287 26,182 10.6Z 14,287 25,496 10.3Z 14,287 22,071 9.2Z West bank 16,220 27,309 44,083 7.42 27,309 47,235 7.9Z 27,309 45,367 7.6Z East Guandong 11,782 16,993 24,875 5.52 16,993 23,593 5.1% 16,993 31,003 7.2% Coastal route 7,700 11,912 18,282 6.4Z 11,912 17,267 5.9Z 11,912 21,875 7.7% Meixian route 4,081 5,081 6,593 3.5Z 5,081 6,326 3.2Z 5,081 9,128 5.9Z West Guangdong 15,789 28,107 46,741 8.1Z 28,107 43,923 7.6Z 28,107 43,176 7.4Z Maoming route 11,279 19,840 32,790 7.9Z 19,840 30,861 7.5Z 19,840 30,363 7.3Z Yunan route 4,510 8,267 13,951 8.42 8,267 13,062 7.9Z 8,267 12,813 7.7Z North Guangdong 4,810 6,734 9,645 5.1Z 6,734 9,225 4.8Z 6,734 9,100 4.7Z TOTAL 71,772 112,934 175,203 6.6Z 112,934 175,203 6.6Z 112,934 175,203 6.6Z Note: Growth rate average from 1986 to 2000.  Table 13: LENGTH OF INTERNAL PROVINCIAL TRANSPORT ROUTES BY MODE - 1986 (including that of Hainan Island) Local Inland Coastal Civil Itemr Total Rail rail waterways shipping Road aviation Pipeline Length (km) 81,373 1,128 449 10,793 1,810/a 53,820 1,300/a 182 Z 1.13 0.57 13.65 2.22 80.60 1.60 0.22 Route Density (km/km2) Guangdong 0.0043 0.0022 0.0520 n.a. 0.3095 0.006 0.001 National 0.0054 0.0003 0.0114 n.a. 0.0982 n.a. n.a. /a 1987 data. 1 Notes: (1) Civil aviation includes Guangzhou-Shantou, Guangzhou-Zhanjiang and Guangzhou-Haikou. When interprovincial and international routes are added the total is 64,059 km. Hainan island routes represent 40 percent of the provincial total. (2) Coastal shipping includes Guangzhou-Shantou, Guangzhou-Zhanjiang, Guangzhou-Haikou and Zhanjiang-Haikou. Hainan island routes represent 40 percent of the total. n.a. = not applicable Source: 1987 Guangdong Statistics Yearbook, 1989. Guangdong Yearbook, China Yearbook, 1987 and Map of Guangdong Province.  - 93 - Table 14: KEY RAILWAY SECTION CHARACTERISTICS No. of Type of Maximum Gross trail- Type of Section Class Length tracks signaling grade ing tonnage locomotive (km) (Z) (tons) State-Level Railways (rail gauge 1.435 m) 2,200/ Pingshi-Guangzhou II 336.2 double /a Semiautoma- 9.1/11.6 /b 2,500 Dongfeng 4 tic block 2,200/ Guangzhou-Shenzhen II 146.5 double /a Semiautoma- 7.5/8.0 /b 2,500 Dongfeng 4 tic block Guangzhou-Sanshui III 48.7 Single Semiautoma- 7.8/6.2 1,600 Dongfeng 3 tic block Hechun-Zhanjiang II 82.3 Single Semiautoma- 6.0 3,000 Qianjin tic block Hechun-Maoming II 61.0 Single Semiautoma- 6.0 3,000 Qianjin tic block Total State-Level 674.7 Local Railways Name Gauge (mm) Length (km) Endpoints Nanlin 1,055 92.5 Pingshi-Muchong Quren 1,055 27.5 Huanggang-Chunding Meilong 702 172.5 Meixian-Xipu-Xingning Rulian 702 59.5 Loujiatong-Wanbei Yingde 702 47.0 Dongguapu-Jintan Longhe 702 7.0 Hetou-Chachang Yunfu 702 16.2 Yuncheng-Luidu Maan 702 12.0 Maan-Xijiangan Dalishan 702 32.0 Dalishan-Binglongpu Total Local 466.2 La The double-track project was completed in 1988. /b Grade should be lower due to double tracking project. Source: ICT Interim Report, May 1987, and Guangdong Province Planning Commission.  - 94 - Table 15: LENGTH OF T'HE ROAD NETWORK BY CLASS, SURFACE AND FUNCTION Length (km) Percent By Class 1985 1987 1988 1985 1987 1988 Class I 0 7.3 32 0.0 0.01 0.06 Class II 538 835.1 1,245 0.8 1.6 2.3 Class III 2,197 1,985.3 2,694 3.4 3.7 5.0 Class IV 36,224 31,359.2 31,255 56.5 58.9 58.1 Substandard (out of class) 25,111 19,031.5 18,594 39.2 35.8 34.5 By Surface Type Cement 1,272 1,986.2 1.9 3.7 Bituminous 7,291 6,675.4 11.5 12.5 Gravel 12,621 38,416.1 19.7 68.4 Gravel (low quality) 26,562 } 41.5 } Nonpaved 16,324 8,141.7 25.5 15.3 By Function Trunk (primary) 7,439 n.a. 11.6 n.a. (of which: national roads) (4,537) 4,067.1/a (7.1) (7.6) Trunk (secondary) 8,179 8,582.7/b 12.8 16.1 County road 17,635 14,799.1 27.5 27.8 Rural road 30,481/c 25,670.5 47.6 48.2 Special functioned 336 99.0 0.5 0.2 /a National roads. /b Provincial roads. /c In ICT Report (Chinese Version), p. 123, this number is 30,750. Notes: (1) Trunk (primary) consists of national roads and provincial trunk roads. (2) Number of bridges: 10,503 of which: dangerous bridges 519 total length (m): 296,001 total length (m) 17,514 Of the total permanent bridges 10,412 length 292,697 m semi-permanent bridges 66 length 2,269 m temporary bridges 25 length 1,035 m Data provided by Provincial Bureau of Highways. Sources: ICT Report, May 1987, Chinese version, p. 123; Department of Communi- cations, Bureau of Highway Administration for 1987 data; Guangdong Statistics Yearboolk, 1989.  - 95 - Table 16: DENSITY OF ROAD NETWORK BY REGION, 1986 Area Length of roads Density of road (km2) (km) (km/100 km2) Provincial total Ia 177,890 52,785 29.7 Guangzhou 16,200 5,897 36.4 Shantou 10,345 3,657 35.4 Foshan 5,476 2,056 37.5 Jiangmen 17,039 4,978 29.2 Shaoguan 30,740 7,653 24.9 Zhanjiang 12,102 7,653 24.9 Maoming 11,323 3,895 34.4 Shenzhen 1,863 870 43.5 Zhuhai 1,365 353 25.9 Meixian 15,843 3,090 19.5 Huiyang 31,855 9,167 28.8 Zhaoqing 22,619 5,851 25.9 la Hainan Island is excluded. Source: Guangdong Yearbook, 1987.  - 96 _ Table 17: CHARACTERISTICS OF MAJOR SEAPORTS IN GUANGDONG PROVINCE Number of berths for Actual ship size of Maximum Total throughput Total 10,000 tolns ship size capacity in 1987 Port berths or more ('000 dwt) (mln tons) (mln tons) Type of freight Huangpu /a 45 19 35/b 17.51 26.21 Coal, oil, ores, containers, general cargo Zhanjiang 31 15 50/b 12.70 14.14 Oil, coal, ores, general cargo Guangzhou /a 102 0 10 8.31 6.93 Dry-bulk, general cargo Shantou 9 0 5 1.50 2.51 Dry-bulk, general cargo Shekou and 41 4 10 3.01 5.08 Dry-bulk, Chiwan general cargo Jiouzhou 13 2 10/b 1.50 0.80 Dry-bulk, general cargo Total 241 40 44.53 55.67 /a Huangpu and Guangzhou are now operated as one port. lb Accessible with high tide orLly. Note: The definition of capacity for ports used in the Province clearly understates actual capacity. Source: ICT Interim Report and GPPC.  - 97 - Table 18: INLAND WATERWAY TRANSPORT SYSTEM A. Port Throughput by River (1987) Number of ports Throughput (mln tons) West river (Xijiang) 11 5.43 North river (Beijiang) 20 2.71 East river (Dongjiang) 15 2.52 Han river 14 1.43 The Pearl River Delta 37 20.47 Others 29 7.66 Total 126 40.22 B. Major Navigable Sections Maximum ship River Section Length (km) Depth (m) size (dwt) West river Fengkai-Sixianjiao 209 2.0 - 2.5 1,000 Sixianjiao-Daweijiao 68 2.0 300 North river Nanxiong-Shaoguan 123 <0.5 10-20 Shaoguan-Lianjiangkou 124 0.8 50 Lianjiangkou-Sanshui 134 1.0 100 East river Cuhe-Fenshu dam 32 0.4 <15 Fengshu dam-Heyuan 155 0.6 30 Heyuan-Shilong 200 0.8 50 Shilong-Dongjiangkou 42 1.5 50-100 Shilong-Niwei 42 1.5 50-100 Han river 310 <1.0 <50 The Pearl River Network >1.0 >100 Source: ICT Interim Report, May 1987 and GPPC.  - 98 - Table 19: AIRPORTS BY CLASS AND AIRCRAFT ACCOMMODATED IN 1987 (Civil passenger airports) Aircraft Manager Class /a accommodated Baiyun (Guangzhou) CAAC I Boeing 747 Zhanjiang CAAC II Boeing 737 Foshan /b CJCAC /c II Boeing 737 Shantou CAAC II Boeing 737 Meixian CAAC II Boeing 737 Huiyuan CJCAC II Boeing 737 Zhuhai CAAC Helicopter (Heliport) /a Class I = International Class II = Domestic lb Foshan airport was opened in March 1987 to civilian traffic. It serves the Pearl River Delta area. /c China Joint Civil Aviation Company, a military subsidiary company. Source: ICT Interim Report and GPPC.  Table 20: RAIL PASSENGER SERVICES, 1988 Beijing-Guangzhou Guangzhou-Shenzhen Litong-Zhanjiang Item Express Fast Slow Fast Slow Fast Slow hard seat hard seat hard seat air-conditioned hard seat hard seat hard seat Type of coach hard sleeper hard sleeper hard sleepter hard seat hard sleeper hard sleeper soft sleeper soft sleeper soft seat soft sleeper Number of coaches 20 24 maximum 13-20 12-20 13-20 13-20 13-20 per train Frequency 3 11 2 14 3 n.a. n.a. (pairs/day) Fare prices /a 0.022 0.018 0.014 0.027 0.021 0.018 0.014 (Yuan/person-kn) e /a Prices for hard seat only. Premium for soft seat is 0.4 fen/pkm, hard sleeper 1.3 fen, soft sleeper 3.3 fen, and express 0.8 fen. Source: Guangzhou Railway Administration.  Table 21: AMOUNT OF RAILWAY ROLLING STOCK BY TYPE AND YEAR 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Locomotives Diesel 108 121 122 125 119 122 146 184 189 197 Steam 56 44 22 22 22 18 18 20 22 21 Freight wagons 300 299 299 299 339 359 409 454 462 462 Passenger coaches 489 557 579 530 539 562 645 708 817 983 1 Avg. daily operating O freight wagons on line 2,566 2,603 2,672 2,659 2,631 2,837 3,029 3,459 3,72i 4,010 Note: Hainan excluded. Source: Guangzhou Railway Administration.  Table 22: STATE-OW'IED, SPECIALIZiED ROAD TRANSPORT SERVICES 1986 1987 Number of service offices 1,563 n.a. Passenger services: (nonurban) Number of routes 2,996 3,390 Areas served: - All counties and cities of the province - 95.74Z of all districts; (about the same as 1986) - 60.39Z of all rural units (Xiang) Frequency (vehicle trips/day) 11,898 11,512 Total passengers (millions/day) 330 297 Total pass-km (billions) /a 14.1 14.0 Freight services: (total)