WORLD BANK DISCUSSION PAPER NO. 352 Work in progress for public discussion I|C China's Urban Transport Development Strategy Prorefeings of a Sylniposuwin in Bejing, Novembelr 8-10, 1995 w A \~~~~ Edited b)" LS,phn / rs, R Lill Zlil Recent World Bank Discussion Papers No. 282 Trade Laws and Institutions: Good Practices and the World Trade Organization. Bernard M. Hoekman No. 283 Meeting the Challenge of Chinese Enterprise Reform. Harry G. Broadman No. 284 Desert Locust Management: A Timefor Change. Steen R. Joffe No. 285 Sharing the Wealth: Privatization through Broad-based Ownership Strategies. Stuart W. Bell No. 286 Credit Policies and the Industrialization of Korea. Yoon Je Cho and Joon-Kyung Kim No. 287 East Asia's Environment: Principles and Prioritiesfor Action. Jeffrey S. Hammer and Sudhir Shetty No. 288 Africa's Experience with Structural Adjustment: Proceedings of the Harare Seminar, May 23-24, 1994. Edited by Kapil Kapoor No. 289 Rethinking Research on Land Degradation in Developing Countries. Yvan Biot, Piers Macleod Blaikie, Cecile Jackson, and Richard Palmer-Jones No. 290 Decentralizing Infrastructure: Advantages and Limitations. Edited by Antonio Estache No. 291 Transforming Payment Systems: Meeting the Needs of Emerging Market Economies. Setsuya Sato and David Burras Humphrey No. 292 Regulated Deregulation of the Financial System in Korea. Ismail Dalla and Deena Khatkhate No. 293 Design Issues in Rural Finance. Orlando J. Sacay and Bikki K. Randhawa No. 294 Financing Health Services Through User Fees and Insurance: Case Studiesfrom Sub-Saharan Africa. R. Paul Shaw and Martha Ainsworth No. 295 The Participation of Nongovernmental Organizations in Poverty Alleviation: The Case Study of the Honduras Social Investment Fund Project. Anna Kathryn Vandever Webb, Kye Woo Lee, and Anna Maria Sant'Anna No. 296 Reforming the Energy Sector in Transition Economies: Selected Experience and Lessons. Dale Gray No. 297 Assessing Sector Institutions: Lessons of Experiencefrom Zambia's Education Sector. Rogerio F. Pinto and Angelous J. Mrope No. 298 Uganda's AIDS Crisis: Its Implicationsfor Development. Jill Armstrong No. 299 Towards a Payments System Lawfor Developing and Transition Economies. Raj Bhala No. 300 Africa Can Compete! Export Opportunities and Challengesfor Garments and Home Products in the European Market. Tyler Biggs, Margaret Miller, Caroline Otto, and Gerald Tyler No. 301 Review and Outlookfor the World Oil Market. Shane S. Streifel No. 302 The Broad Sector Approach to Investment Lending: Sector Investment Programs. Peter Harrold and Associates No. 303 Institutional Adjustment and Adjusting to Institutions. Robert Klitgaard No. 304 Putting Institutional Economics to Work: From Participation to Governance. Robert Picciotto No. 305 Pakistan's Public Agricultural Enterprises: Inefficiencies, Market Distortions, and Proposalsfor Reform. Rashid Faruqee, Ridwan Ali, and Yusuf Choudhry No. 306 Grameen Bank: Performance and Stability. Shahidur R. Khandker, Baqui Khalily, and Zahed Khan No. 307 The Uruguay Round and the Developing Economies. Edited by Will Martin and L. Alan Winters No. 308 Bank Governance Contracts: Establishing Goals and Accountability in Bank Restructuring. Richard P. Roulier No. 309 Public and Private Secondary Education in Developing Countries: A Comparative Study. Emmanuel Jimenez and Marlaine E. Lockheed with contributions by Donald Cox, Eduardo Luna, Vicente Paqueo, M. L. de Vera, and Nongnuch Wattanawaha No. 310 Practical Lessonsfor Africafrom East Asia in Industrial and Trade Policies. Peter Harrold, Malathi Jayawickrama, and Deepak Bhattasali No. 311 The Impact of the Uruguay Round on Africa. Peter Harrold No. 312 Procurement and Disbursement Manualfor Projects with Community Participation. Gita Gopal No. 313 Harnessing Informationfor Development: A Proposalfor a World Bank Group Strategy. Eduardo Talero and Philip Gaudette No. 314 Colombia's Pension Reform: Fiscal and Macroeconomic Effects. Klaus Schmidt-Hebbel No. 315 Land Quality Indicators. Christian Pieri, Julian Dumanski, Ann Hamblin, and Anthony Young No. 316 Sustainability of a Government Targeted Credit Program: Evidence from Bangladesh. Shahidur R. Khandker, Zahed Khan, and Baqui Khalily No. 317 Selected Social Safety Net Programs in the Philippines: Targeting, Cost-Effectiveness, and Optionsfor Reform. Kalanidhi Subbarao, Akhter U. Ahmed, and Tesfaye Teklu (Continued on the inside back cover) WORLD BANK DISCUSSION PAPER NO. 352 EastAsia and Pacific Region Series China's Urban Transport Development Strategy Proceedings of a Symposium in Beiing, November 8-10, 1995 Edited by Stephen Stares Liu Zhi The World Bank Washington, D.C. Copyright © 1996 The International 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 America First printing January 1997 Discussion Papers present results of country analysis or research that are 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. Some sources cited in this paper may be informal documents that are not readily available. 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. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission 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 work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. For a copy of Update describing new publications, contact the Distribution Unit, Office of the Publisher, TXhe World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or Publications, The World Bank, 66, avenue d'Ina, 75116 Paris, France. A catalog and ordering information are also available on the Internet at http: / /www.worldbank.org. Cover photograph by Kevin Fehon. ISSN: 0259-210X Stephen Stares, until his death in 1996, was senior transport specialist in the Environment and Municipal Development Operations Division of the World Bank's China and Mongolia Department. Liu Zhi is a transport economist in the Transport Division of the Bank's Transportation, Water, and Urban Development Department. Library of Congress Cataloging-in-Publication Data China's urban transport development strategy: proceedings of a symposium in Beijing, November 8-10, 1995 / edited by Stephen Stares, Zhi Liu. p. cm. - (World Bank discussion paper; no. 352) Includes bibliographical references. (p. ). Papers presented at the China Urban Transport Symposium. ISBN 0-8213-3841-2 1. Urban transportation - China - Congresses. 2. Urban transportation - China - Passenger traffic - Congresses. 3. Urban transportation - Environmental aspects - China - Congresses. 4. Urban transportation policy - China - Congresses. 5. Local transit - China- -Congresses. I. Stares, Stephen, 1944-1996. II. Liu, Zhi, 1961- . III. China Urban Transport Symposium (1995: Peking, China) IV. Series: World Bank discussion papers ; no. 352. HE311.C6C48 1996 388.4'0951-dc2l 96-49469 CIP CONTENTS Foreword ..................................................................v Abstract ................................................................. vi Currency Equivalents ................................................................. vii Abbreviations and Acronyms Used ................................................................. ix Introduction and Overview ................................................................. xi Liu Zhi and Stephen Stares Concluding Statement of the Symposium ................................................................. xix Katherine Sierra Opening Remarks ..................................................................1 Hou Jie Opening Remarks .................................................................7 Anthony J: Pellegrini Opening Remarks ................................................................. 11 Jin Liqun Opening Remarks ................................................................. 15 A. Timothy Peterson Keynote Address 1: Reforming China's Urban Transport Sector ........................................ 21 Li Zhendong Keynote Address 2: The Urban Transportation Problem in a Context of Economic Development ................................................................. 27 John R. Meyer Keynote Address 3: Urban Transport Problems in Chinese Cities: Causes, Trends and Options ................................................................. 35 Zhou Ganshi Theme Paper 1: Motorization in Chinese Cities: Issues and Actions ................................... 43 Stephen Stares and Liu Zhi Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge .......... 105 Michael P. Walsh Note: All Chinese names in this document follow Chinese convention, which is family name first and given name last. - iv - Theme Paper 3: Municipal Transport Management: A Domestic View ............................ 153 Wu Yong, Wang Jianqing, and Yao Zukang Theme Paper 4: Municipal Transport Management: Overseas Experience ..................... 183 Richard Meakin Theme Paper 5: Bicycles in Cities .............................................................. 213 Anton G. Welleman, Cees J. Louisse and Dirk M Ligtermoet Theme Paper 6: Investment in Mass Rapid Transit .............................................................. 253 Roger Allport Theme Paper 7: The Reform and Development of China's Urban Public Transportation Enterprises .............................................................. 311 Wang Jinxia, Zhang Kuifu and Qiao Junshan Theme Paper 8: Options for Bus Transport: Overseas Experience ................................... 339 John Flora Theme Paper 9: Private Funding of Transport: Infrastructure ......................................... 357 Kwok Hung Lee Theme Paper 10: Shaping the Future: Getting Prices Right ............................................... 389 Kenneth Gwilliam Theme Paper 11: Land Use and Transport Planning in China ............................................ 413 Li Xiaojiang and Yu Li Theme Paper 12: Shaping the Future: The Role of Urban Transport Planning ............... 459 Ralph Gakenheimer Closing Remarks: Toward the New Century .............................................................. 475 Mao Rubai Annex: Urban Transport Statistics .............................................................. 477 Li Yaming List of Participants .............................................................. 495 - v - FOREWORD The China Urban Transport Symposium, jointly sponsored by the Ministry of Construction, the Ministry of Finance, the People's Bank of China, the World Bank, and the Asian Development Bank, was successfully held in Beijing during November 8-10, 1995. The symposium and the publication of its proceedings is very timely, as China's economy is poised to grow rapidly into the twenty-first century, and yet Chinese cities are grappling with increasingly serious transport problems posed by rapid economic growth. The symposium brought together senior policymakers from the Central Government and a number of major cities, prominent international and domestic urban transport experts, and staff of the World Bank and Asian Development Bank, to discuss and clarify the complex issues facing China's urban transport sector and to debate future directions for urban transport policy and strategy. Many important themes were covered, they included motorization, motor vehicle pollution, urban transport management, bicycles in cities, mass rapid transit, public transit reform, the role of the private sector, transport pricing, and urban transport planning. We hope that the success of the symposium will serve as an auspicious beginning to a new era of sustainable urban transport development in China. We would like to dedicate the publication of these proceedings to the memory of Mr. Stephen Stares, Senior Transport Specialist with the World Bank's China and Mongolia Department at the time of his death. His vision in originating the call for this nationwide symposium and his constant attention to detail in its development and presentation lay at the heart of its success. Li Zhendong Nicholas C. Hope Vice Minister Director Ministry of Construction China and Mongolia Department East Asia and Pacific Region Jin Liqun The World Bank Assistant Minister Ministry of Finance A. Timothy Peterson Deputy Director Yin Jieyan Infrastructure, Energy and Financial Sectors Department Deputy Governor Asian Development Bank People's Bank of China - vi - ABSTRACT This Discussion Paper presents the proceedings of China Urban Transport Symposium that was held in Beijing during November 8-10, 1995. The symposium was jointly sponsored by the Ministry of Construction, the Ministry of Finance, the People's Bank of China, the World Bank, and the Asian Development Bank, and was attended by senior policymakers from the Central Government and a number of major cities, prominent international and domestic urban transport experts, and staff of the World Bank and Asian Development Bank. The major purposes of the symposium were to provide a forum for the discussion of increasingly serious urban transport issues and provide a basis for clarifying future directions for urban transport policy and strategy in China. The symposium covered a wide range of important themes, including motorization, motor vehicle pollution, urban transport management, bicycles in cities, mass rapid transit, public transit reform, the role of private sector, transport pricing, and urban transport planning. - vii - CURRENCY EQUIVALENTS (As of November 1995) Currency name = Renminibi Currency unlit = Yuan $1.00 = Y 8.5 Y 1.00 = $0.12 FISCAL YEAR January I - December 31 WEIGHTS AND MEASURES Metric System - rx - ABBREVIATIONS AND ACRONYMS USED 8FYP - Eighth Five-Year Plan (1991-1995) 9FYP - Ninth Five-Year Plan (1996-2000) l.g/dl - Micrograms per Deciliter glm - Micrometer or Micron ARF - Additional Registration Fee BOT - Build-Operate-Transfer CAA - Clean Air Act CARB - California Air Resources Board CAUPD - China Academy of Urban Planning and Design CFCs - Chlorofluorocarbons CNAIC - China National Automotive Industry Corporation CNG - Compressed Natural Gas Co - Carbon Monoxide CO2 - Carbon Dioxide COE - Certificate of Entitlement CRD - Central Retail District ECE - United Nations Economic Commission for Europe EIRR - Economic Internal Rate of Return EPA - Environmental Protection Agency ERP - Electronic Road Pricing ETBE - Ethyl Tertiary Butyl Ether FAR - Floor Area Ratio FRT - First Registration Tax GDP - Gross Domestic Product GNP - Gross National Product HC - Hydrocarbon HEI - Health Effects Institute HOV - High-Occupancy Vehicle IARC - International Agency For Research on Cancer I/M - Inspection and Maintenance IQ - Intelligence Quotient ISO - International Standards Organization KCR - Kowloon-Canton Railway kPa - Kilipascal LEV - Low-Emission Vehicle LPG - Liquefied Petroleum Gas LRT - Light Rapid Transit LTA - Land Transport Agency LTD - Land Transport Department LTE - London Transport Executive m 3 - Cubic Meter MRT - Mass Rapid Transit MTBE - Methyl Tertiary Butyl Ether MTR - Mass Transit Railway MV - Motor Vehicle NGO - Nongovernmental Organization NH3 - Ammonia NMHC - Nonmethane Hydrocarbons NO2 - Nitrogen Dioxide NO, - Nitrogen Oxide O&M - Operation and Maintenance O-D - Origin-Destination OMV - Open-Market Value PARF - Preferential Additional Registration Fee PM - Particulate Matter PM10 - Particulate Matter of 10 Microns or Less in Size POM - Polycyclic Organic Matter pphpd - passengers per hour per direction psi - per square inch PTA - Passenger Transport Authority PTE - Passenger Transport Executive RON - Research Octane Number RVP - Reid Vapor Pressure SACI - State Administration of Import and Export Commodity Inspection S02 - Sulfur Dioxide SOF - Soluble Organic Fraction SPC-ITE - Institute of Techno-Economics of the State Planning Commission SPM - Suspended Particulates TBA - Tertiary Butyl Alcohol TOG - Total Organic Gases TSP - Total Suspended Particulates TVE - Township and Village Enterprise UK - United Kingdom US - United States VMT - Vehicles Miles Traveled VOCs - Volatile Organic Compounds WHO - World Health Organization Liu Zhi and Stephen Stares xi INTRODUCTION AND OVERVIEW LIU ZHI AND STEPHEN STARES' INTRODUCTION The papers in this volume were presented at the China Urban Transport Symposium, held November 8-10, 1995, in Beijing. The symposium was jointly sponsored by the Ministry of Construction (MCon), the Ministry of Finance (MOF), the People's Bank of China (PBC), the World Bank (IBRD), and the Asian Development Bank (ADB). The symposium provided a forum for the discussion of increasingly serious urban transport issues and provided a basis for clarifying future directions for urban transport policy and strategy in China. The symposium was attended by over 180 delegates and twelve theme papers were presented covering a wide range of themes, including motorization, motor vehicle pollution, urban transport management, bicycles in cities, mass rapid transit, public transit reform, the role of the private sector, transport pricing, and urban transport planning. This was followed by two parallel open discussions on two of the key themes of the symposium: Dealing with Motorization and Developing Public Transport. There was a remarkable degree of agreement reached at the end of the symposium between domestic and international delegates on the key urban transport issues facing Chinese cities. In particular, there was general recognition that a comprehensive approach to urban transport was required which balanced transport infrastructure construction with transport policy measures. The Concluding Statement of the Symposium summarized this consensus, and set out five principles, four criteria, and eight actions to underpin the urban transport strategy for China, and to guide future assistance from IBRD and ADB. The full text of this statement is placed immediately following the Overview below. OVERVIEW Opening Remarks and Keynote Addresses The opening remarks were made by leaders or representatives of some of the sponsoring institutions. They were Hou Jie (MCon), Anthony Pellegrini (IBRD), Jin Liqun (MOF), and Timothy Peterson (ADB), respectively. All their remarks touched upon the major urban Liu Zhi is a Transport Economist in the Transport Division; Transportation, Water and Urban Development Department; The World Bank. Stephen Stares (deceased) was a Senior Transport Specialist in the Environment and Municipal Development Operations Division, China and Mongolia Department. Useful comments from Peter Midgley of the World Bank are gratefully acknowledged. xii Introduction and Overview transport issues that China's cities are currently facing and the opportunities that China has in addressing these issues. Hou Jie, Minister of Construction, first provided an overview of China's urban transport development in recent years, and highlighted the Ninth Five-Year Plan for urban transport development and the longer-term development targets set by the Central Government. Despite considerable efforts made in recent years, he observed, urban transport infrastructure and services are still deficient, and the problem will likely become worse. The Central Government clearly has made a strong commitment aimed at major improvements in urban transport by the year 2010. Achieving this, however, as Hou pointed out, will be especially challenging, and would not be possible without further policy and management reforms in the sector. Sustainable transport development was the central topic of Anthony Pellegrini's opening remarks. Indeed, this is also IBRD's long-standing stance toward urban transport assistance in developing countries. Noting the unprecedented pace of economic growth and the severity of adverse impacts of rapid motorization, Pellegrini stressed the need for effective policies to guide sustainable urban transport development. This, Pellegrini further emphasized, requires that urban transport policies and actions be evaluated with three complementary criteria-economic and financial sustainability, social sustainability, and environmental sustainability. Jin Liqun's remarks touched upon the efficiency issue in urban transport investment. Achieving higher efficiency, he observed, requires careful planning, realistic implementation, and better management. He pointed out that while every effort should be made to increase urban transport investment, greater attention should also be paid to the efficient use of available funds. Three key issues in China's urban transport-rapid motorization, the role of public transit, and the role of bicycle-were discussed in Timothy Peterson's remarks. He compared China's motorization with similar development in the Asia-Pacific region, and pointed out that the symposium was at the right time where policies and actions could be sorted out before motorization was too far advanced. He urged China to take early action to avoid repeating the errors made by other countries. The opening remarks were followed by three keynote addresses. Li Zheng Dong, Vice Minister of Construction, focused on urban transport sector reform. He emphasized the role of the market mechanism in the efficient provision of urban transport, and sketched several reform schemes. These include competitive bidding in urban transport infrastructure project design and implementation, municipal transport management reform, fuel taxation, public transit franchising, and the separation of government ownership from the operation of transport enterprises. In the second keynote address, Professor John Meyer put China's urban transport in both historical and economic perspectives, by comparing the process with the one experienced in the West. He discussed the complicated nature of urban transport problems and the major dilemmas that would be faced by policymakers and planners in China. He emphasized the role of pricing in the containment of negative externalities generated by rapid motorization. He also drew the Western experience of public transit decline and predicted that a major challenge for China is to develop a public transport system that can increasingly compete with improved individualized Liu Zhi and Stephen Stares xiii transportation. In terms of urban transport infrastructure provision, Professor Meyer warned of the danger of "thinking too big" without sufficient prior planning. Finally, in the third keynote address, Professor Zhou Ganshi offered a detailed assessment of the extent of current transport problems in large Chinese cities. The future growth of the vehicle fleet at the current rate and the required expansion of urban road space, he predicted, will pose serious land-use problems to large cities where the land constraint is most severe. Therefore, he called for government actions to control private cars in cities. Theme Papers Among the twelve theme papers, three were written by domestic authors and nine by international authors. All authors had substantial experience relating to the subjects they discussed in their papers. Most of the international authors, moreover, had extensive experience in China's urban transport and were familiar with the issues involved. During the preparation of these papers, a joint MCon, CAUPD, IBRD, and ADB sector mission, comprising most of the theme paper authors, visited five selected cities (Beijing, Shanghai, Guangzhou, Chengdu, and Jinan) and obtained useful first-hand information on the recent development of urban transport in China's large cities. The key messages from these theme papers are summarized as follows. Motorization Motorization in China is at an early stage but is rapidly gathering momentum. Moreover, the Central Government has recently promulgated an auto industry development policy that aims at promoting the auto industry through the promotion of the domestic market. Examining the issues associated with motorization in cities, therefore, is especially urgent at this time. Stephen Stares and Liu Zhi dealt with the broad issues of motorization in cities, with particular attention to traffic congestion. They first reviewed the trends in motor vehicle ownership in China and predicted continuing rapid growth in the fleet in line with China's strong economic growth. They noted that considerable benefits from increasing motorization would be rapidly dissipated without proper policies to control pollution and congestion. Then, they moved on to discuss in length two broad approaches to dealing with traffic congestion: expanding street and road capacity to absorb the growth in traffic, and managing the growth in traffic to match the available capacity. Much can be done on the first approach, through improved management of the existing street system, allocation of priority road space to public transit, and construction of new streets and roads. However, this approach on its own would not be sufficient to accommodate rapid growth in demand for vehicle transport. Hence, the second approach, which includes a wide range of pricing and nonpricing options, is necessary and should be applied in Chinese cities. To provide an international perspective, they also examined the urban transport experience in other Southeast Asian cities that have been coping with motorization for longer than China. They noted that the relative degree of success or failure in controlling traffic congestion seems to depend on the extent of control and intervention exercised by Government in the areas of land-use control, vehicle ownership and/or vehicle usage controls, and provision of good public transport services. Finally, in commenting on the auto industry development xiv Introduction and Overview policy, they noted that cities should be given the freedom to set limits to motor vehicle use consistent with sustainable growth, and should not be under pressure to accommodate faster growth to satisfy auto industry demands. Their analysis showed that a viable auto industry can be established on the basis of domestic demand, and controls on motor vehicle ownership in large cities will not endanger the viability of the auto industry. Motor Vehicle Pollution Environmental pollution is a very serious problem for China to cope with during rapid economic growth. Increasing motorization is likely to make the situation worse. Michael Walsh identified the nature of motor vehicle pollution and discussed in great detail the national and local strategies to reduce or ameliorate adverse impacts. He noted that China's motor vehicle pollution is increasingly serious in both absolute terms and relative to nonmobile sources of pollution. He further noted that China's motor vehicle fleet (including the motorcycle) has a large proportion of high-polluting vehicles using outdated engine technology, and there are few programs to detect illegal emission levels and to enforce standards. Moreover, gasoline is still almost entirely leaded, adding a particularly dangerous component to air pollution. With motorization still at an early stage in China, Walsh pointed out, there are opportunities to introduce pro-environment policies with some initial cost but with relatively high long-term benefits. He drew on international experience to show that motor vehicle pollution control strategies can be very effective. These include policies on cleaner fuels (in particular, a move to unleaded gasoline) and mandates on motor vehicle emissions. Municipal Transport Management Rapid economic growth is putting China's urban transport management system under pressure. The current institutional responsibilities for many urban transport functions are not clear-cut and are overlapping in many cases. Obviously, a more effective municipal transport management system is much needed for effectively tackling the growing urban transport problems. A pair of papers discussed this issue. Wu Yong, Wang Jianqing and Yao Zukang provided an overview of urban transport institutional development in China. They first reviewed various municipal responsibilities ranging from legal framework, to policy-making, pricing, infrastructure and service management, traffic management, and financing. They noted that the current system is the legacy of the command economy and has become inefficient during the transition to a market economy. They identified the strengths and weaknesses of the existing municipal administrative frameworks, and reported on the current thinking about possibilities for reform. In particular, they stressed that municipal transport policies should be made primarily at the municipal level, and that urban transport commercial functions should be separated from government planning and regulatory functions. Richard Meakin commented on current Chinese municipal transport management from an overseas perspective. Modern urban transport, he observed, is a complex process that requires a coordinated approach and requires institutions to implement policies that cut across the current subdivisions of the urban transport sector in most municipalities, but the current multitude of agencies has little horizontal integration and this impedes consistent transport Liu Zhi and Stephen Stares xv planning. Moreover, municipal governments in China are responsible for nearly all aspects of urban transport including planning, financing, construction, operation, and regulation, and an efficient market economy requires that the cities separate these functions and shed some of them. Appropriate roles for the municipality, Meakin suggested, are in financial planning and management, transport planning, construction management, traffic management planning and implementation, formulation of demand management policy, transport pricing policy, and transport regulation. Appropriate roles to be shed to corporatized entities, and perhaps eventually to the private sector, are transport operations, infrastructure decign and construction, and perhaps infrastructure maintenance. The Role of the Bicycle Chinese cities are well known for bicycle use. Half of all nonwalking urban journeys in major cities use the bicycle. Amid growing motorization, however, some city managers see the bicycle as a significant cause of traffic congestion and traffic accidents, and want to discourage its use in city centers. Others see the bicycle as a cheap, readily available, and nonpolluting mode of transport that should be further encouraged. Clearly, there is an urgent need for clarifying the role of the bicycle in cities. A.G. Welleman, C.J. Louisse and D.M. Ligtermoet used a comparative approach to examine the role of the bicycle. They first examined bicycle policies in the Netherlands and other countries. Their analysis of the revival of bicycle use in the Netherlands after a period of decline indicated strong grounds for increasing bicycle use in developed countries because of growing environmental awareness. They then reviewed development in bicycle use in Chinese cities and noted many similarities with the Netherlands. They urged that for reasons of equity and economic efficiency, the role of bicycles be retained in city transport policies. Even with rapid economic growth, bicycles will remain the only easily affordable means of transport for a large segment of the population for many years to come; it makes little sense to dismantle the currently impressive bicycle facilities in Chinese cities. Finally, they offered several recommendations to Chinese municipal governments for better bicycle policy. Mass Rapid Transit Currently, only a few Chinese large cities already have, or are building, metro systems, but a dozen other cities are planning them. Is metro the solution to urban transport problems? This issue was examined by Roger Allport. He first reported the recent findings on mass rapid transit in other countries and then related their applicability to the needs of Chinese cities. He noted that metros are mega-investments and are most unlikely ever to be fully viable financially, covering all costs of construction and operation, even in the largest cities. Therefore, the financial consequences of metros, both for initial capital investment and for operating budgets thereafter, must be thought through very carefully. A common mistake, Allport poilted out, is that most cities tend to take a narrow approach to planning metros with major decisions taken mostly on engineering rather than transport planning grounds, and with many key decisions made before looking for financing schemes. Metros on their own, Allport noted further, cannot be the solution, or even in most cities a major part of the solution, to mass transit needs. This is partly because the cost and complexity of the systems will postpone construction for many cities, and partly because they can serve only relatively narrow corridors. Allport then stressed that bus xvi Introduction and Overview transit for most cities will be the only affordable form of mass transit for many years, and it must be improved to meet immediate needs before conditions are mature enough for metro systems. Allport also examined the lower-cost alternatives to metro systems. In particular, he pointed out, busways are a viable alternative to metro and light rail, and have some key advantages over rail- based systems. An approach to providing mass transit facilities might be to install busways as a municipally-funded investment, to be replaced by rail facilities at a later date if sufficiently high demand develops. Public Transit Reform Recognizing the social consequences of uncontrolled motorization and the difficulties of treating bicycle traffic, public transit appears to be the obvious solution to much of the urban transport problem in China. However, many Chinese city public transit operators are experiencing declining patronage at a time when urban mobility is increasing with greater affluence. Wang Jinxia, Zhang Kuifu and Qiao Junshan provided an overview of public transit development in China, and reported on the recent development in public transit reform. They first discussed the major factors that are affecting public transit development in China. They noted that public transit is currently confronting a number of internal and external difficulties, and to overcome these, public transit reform must be continued and deepened. They also maintained that while continuing to encourage market competition, municipal governments should also continue to play an important role in public transit provision. Bus transit reform is also a major topic in urban communities worldwide, and there is a large body of international experience in dealing with the underlying issues. The key issues relate to the management of bus operations and the appropriate role of municipal government. In presenting options of relevance to China, John Flora outlined lessons learned and experiences, both good and bad, that other countries have had in developing their bus transit systems. Bus services in many developing countries deliver a range of product types at no cost to the public sector, but bus transit does not play its due role in China for various historical reasons. There is considerable potential for revitalizing the bus industry. Bus companies should be given responsibility for operations and finance. This requires disentangling municipal bus operations from municipal bus regulation. Municipal governments will still play an important role, Flora stressed, in determining the level of standards, fares, and operational procedures most appropriate and acceptable to their communities. He further discussed steps of transition to more efficient bus operations. These include corporatization and private-sector involvement. Finally, Flora pointed out that establishing more efficient bus companies will not be sufficient, and for buses to deliver adequate service, road space and traffic priorities must be allocated to buses. The Role of the Private Sector The shortage of funds for transport infrastructure financing has become a common problem to most municipalities. Private-sector involvement thus provides an alternative. The private sector, mostly in the form of foreign investors, is already active in China's urban transport, particularly in bus operations and infrastructure investments. K.H. Lee discussed various facets of private-sector participation in urban transport projects, the impacts on government policies and practices, and the problems confronting the investors. He noted that private-sector investments involve a number of legal and operational issues, and Government Liu Zhi and Stephen Stares xvii must set transport strategy and the project financing policy to guide the process. While private- sector involvement is a promising alternative to public funding, Lee cautioned, the private sector is unlikely to make a large contribution to urban transport infrastructure funding in the short term. However, any promising projects for private funding would likely be large in scale and would require careful definition and management. Transport Pricing Transport prices in Chinese cities have evolved over time and are bound up in complex systems of transport subsidy, earmarking, general municipal financing, and national taxation. Market policies and good governance require that prices be rationalized, and transport policy requirements should influence this process. In particular, there is an opportunity to introduce pricing mechanisms to help manage the demand for, and usage of, the private motor car before usage levels overwhelm city street systems. Kenneth Gwilliam's theme paper provided an primer on the role and impact of prices in urban transport. He showed how incorrect pricing leads to distortions in behavior and creates financing burdens for municipalities. He also discussed the role of pricing in managing the growth of motor vehicles and the use of scarce road space, and in infrastructure financing. These discussions offered general guidelines for the formulation of transport pricing policy. Gwilliam particularly emphasized the role of road user charges, or their proxies such as parking charges and taxation, on car ownership and use. The road user charges, Gwilliam pointed out, can help finance the provision of urban infrastructure, thus reducing financial dependence on land leasing. If the use of road infrastructure is correctly priced, moreover, bus companies can operate commercially and be totally self-financing, thus also reducing the burden on municipalities. Urban Transport Planning There is a widely recognized link between urban land use and transport. Most city authorities appreciate the need to consider the transport implications of land-use plans and regulations. Chinese cities formalize this in the master planning process. The master planning process, however, is currently under intense pressure to cope with rapid economic growth, the evolution of land markets, and the growing demand for better quality transport. Li Xiaojiang and Yu Li assessed the strengths and weaknesses of the current city master planning and transport planning process. They observed that the current planning process is seriously challenged by rapid urban growth, emerging land markets, the floating population, and decentralization of government decisions for transport investment. They noted that urban land use will be more and more guided by the market mechanism, but they argued that government interventions are needed to control the negative spillover effects. In finding solutions to the urban transport problem, Li and Yu noted that public transit and bicycles will continue to be the major modes of urban passenger transport for an extended period of time, and therefore, transport planninlg and management should give priority to them. Finally, they argued that policies for private car development should be based on objective and careful analysis. Infrastructure development should allow some room for future expansion of private car ownership, but managing demanid to prevent growing congestion will be more urgent and important. In another theme paper, Ralph Gakenheimer discussed the role of planninig in transport investment and management at both the strategic and micro levels. He described in detail the xviii Introduction and Overview hierarchical structure and the elements of a comprehensive urban transport planning system. He emphasized that urban transport planning should be a process to achieve agreement and commitment to a consistent and affordable plan of action. All plans need to be set in the context of quantitative justification and realistic budget envelopes. For China, he suggested, the current procedures, which treat transport components as a residual of land use-oriented master planning, must be evolved further to meet the increasingly challenging urban transport planning problems. * * * A large number of individuals and institutions have provided assistance to the symposium and the publication of the proceedings. Without their efforts these would not have been possible. Particular acknowledgments are owed to Wang Guantao, Lin Jialing, Wu Yong, Dong Li, Lu Ying Fan, Qiao Junshan, Wang Jianqing, and Ning Kai of MCon; Li Xiaojiang, Yu Li, and Li Yaming of CAUPD; Charles Melhuish of ADB; and Daud Ahmad, Zong Yan, and Li Chunyan of the World Bank Resident Mission in China, for their strong support and devotion to the work. During the preparation of the theme papers, John Meyer of Harvard University, Nick Anderson, Stephen Dice, Paul Guitink, Andrew Hamer, Slobodan Mitric, Richard Newfarmer, Alfred Nickesen, Jorge Rebelo, Richard Scurfield and Katherine Sierra of the World Bank provided valuable comments. The Municipal Governments of Beijing, Shanghai, Guangzhou, Chengdu, and Jinan provided great assistance to the joint MCon, CAUPD, IBRD, and ADB sector mission. The Institute of Techno-Economics of the State Planning Commission, the Ministry of Machinery Industry, and the China National Automotive Industry General Corporation provided useful information to the theme paper authors. The symposium also received financial support from the Governments of Canada, Switzerland, and the Netherlands. At the World Bank headquarters in Washington, Susan Xinsheng Zhen edited most of the English translations of the domestic papers and speeches, and in some cases, translated the Chinese text into English. Meredith Dearborn, Jan Ellison, Barbara Gregory and Vellet Fernandes provided editorial and secretarial assistance. Katherine Sierra xix CONCLUDING STATEMENT OF THE SYMPOSIUM KATHERINE SIERRA J On November 8-10, 1995, a Symposium was held in Beijing on China's urban transport development strategy sponsored by the Ministry of Construction, the Ministry of Finance, the People's Bank of China, the World Bank and the Asian Development Bank. This statement closed the Symposium proceedings. Two to three years ago, only the very largest municipalities were concerned about urban transport. Water supply and sewage disposal seemed more urgent problems. This has changed for one very clear reason: the rising number of motor vehicles-cars, trucks, taxis and motorcycles-has resulted in a sharp increase in traffic congestion. What can be done about it? Urban transport is complex. Action is needed in many areas at the same time-policy, institutions, management, prices, infrastructure construction, and other investments. It does not help to solve one problem if others are ignored. There was considerable agreement among the delegates on the general lines of actions to be taken. These have been brought together below classified under: Five Principles Four Criteria Eight Actions The five principles represent themes that should guide urban transport planning and operations at this time in China's development: 1. Transport is about moving people and goods, not vehicles 2. Transport prices should reflect full social costs 3. Transport reforms should be deepened to align with socialist market principles so as to increase efficiency 4. The role of government should be to guide transport development 5. The role of the private sector in providing transport services should be encouraged At the time of the symposium, Katherine Sierra was Chief, Environment and Municipal Development Operations Division, China and Mongolia Department, The World Bank; she is now Operations Advisor, Quality Assurance Group, Executive Offices, The World Bank. This statement was drafted jointly by a team from the Ministry of Construction (MCon), the World Bank (IBRD), the Asian Development Bank (ADB) and the China Academy of Urban Planning and Design (CAUPD), headed by Katherine Sierra, who also delivered the statement at the closing of the symposium. Team members included Wu Yong, Dong Li, Lu Yingfang and Lin Kai of MCon; Stephen Stares, John Flora and Liu Zhi of IBRD; Charles Melhuish of ADB; and Li Xiaojiang and Li Yaming of CAUPD. l va: Concluding Statement of the Symposium City plans and policies for managing transport should meetfour criteria: 1. Environmental sustainability 2. Economic viability 3. Financial affordability 4. Social acceptability Consistent with these principals and criteria, eight actions are recommended: 1. Reform urban transport administration 2. Upgrade the status of traffic management 3. Prepare a strategy to mitigate motor vehicle air and noise pollution 4. Develop policies to manage traffic demand 5. Develop a strategy for mass transit 6. Reform public transport management and operations 7. Develop a financing strategy for the transport sector 8. Strengthen the framework for transport planning and capacity building Principle 1: Transport is about Moving People and Goods, not Vehicles Priorities for the use of road space should be allocated to transport operations according to their relative efficiency in moving people and goods. Where appropriate, this will mean giving priority to public transport, bicycles, and pedestrians. Principle 2: Transport Prices Should Reflect Full Social Costs Social costs represent the total costs of transport, including those costs imposed on society through transport operations, in particular: * by environmental pollution (health, medical, productivity loss) * by congestion Principle 3: Transport Reforms should be Deepened to Align with Socialist Market Principles so as to Increase Efficiency The move to the market has started in the transport sector, but the process now needs to be extended, particularly in the fields of: * introducing competition * the pricing of public transport services * the ownership, operation, and regulation of public transport enterprises * wider use of user fees Principle 4: The Role of Government is to Guide Transport Development This is done through: Katherine Sierra xxi * establishing a stable and transparent legal and regulatory framework * setting and enforcing technical standards * preparing transport plans and strategies * financing strategies for infrastructure development * setting pricing policies, including: O determining need for subsidies to disadvantaged sectors of society 0 imposing tolls (taxes, fees) to ensure transport prices reflect social costs 0 mitigating monopoly profits and pricing * strengthening the role of municipal government while retaining the role of National Government in providing macroeconomic guidelines Principle 5: The Private Sector should have a Major Role in Providing Transport Services There is considerable scope for private sector services to supplement or replace current government operations: * Bus service provision and operation * Car parking provision and operation * Consulting services including infrastructure planning and design * Construction contracting * Funding of major infrastructure projects Criteria 1: Environmental Sustainability Actions and initiatives should be included to mitigate adverse impacts on the health and welfare of citizens and the depletion of natural resources. Criteria 2: Economic Viability Priority should be given to projects showing the highest economic returns measured in terms of the full resource costs of inputs. Criteria 3: Financial Affordability Transport systems and projects should be planned only in the context of realistic financing strategies for investment and operations. Criteria 4: Social Acceptability The transport needs of all sectors of society should be catered for, and particular attention paid to the needs of the poor and otherwise disadvantaged members of society. Social impacts of transport initiatives should be mitigated, in particular for those whose housing and businesses must be re-located to make way for new transport construction. xxii Concluding Statement of the Symposium Action 1: Reform Urban Transport Administration At the National Level * Consider the roles of each National Ministry or Bureau involved in urban transport, and their links to counterpart Municipal agencies, and the need for, and content of, enabling legislation. At the Municipal Level * Separate Government planning and regulatory functions from the commercial functions of municipal enterprises * Establish working groups to coordinate the planning and implementation of transport policies and plans * To the extent possible, open the market to the private sector Action 2: Upgrade the Status of Traffic Management * Establish a higher municipal authority for traffic management programs * Integrate the functions of different agencies: O Initially, promote interagency coordination O Later, restructure the responsible agencies * Develop a traffic management strategy O establish functional hierarchy of roads O allocate road space to priority users, including bicycles and buses O implement action plans * Develop programs to enhance road safety Action 3: Prepare a Strategy to Mitigate Motor Vehicle Air and Noise Pollution At the National Level: * establish a schedule to reduce or eliminate lead in gasoline * set emission standards for new vehicles At the Municipal Level: * implement vehicle inspection and maintenance programs * investigate use of alternative (cleaner) fuels * mitigate the adverse noise and emission impacts of new transport infrastructure Action 4: Develop Policies to Manage Traffic Demand * Establish a strategy for the development, control and pricing of parking in city centers * Evaluate other policies for controlling the use of motor vehicles in city centers, including quotas, access restrictions, and road use pricing. * Based on their efficiency and effectiveness, review current restrictions on motorcycles and goods vehicles * Establish a coherent transport pricing policy that: O eliminates untargeted subsidies O provides for cost recovery to the extent possible 0 fully reflects social costs Katherine Sierra xxiii Action 5: Develop a Strategy for Mass Transit * Identify corridors of high passenger demand * Evaluate and prioritize appropriate technologies for mass transit provision considering: 0 likely costs and available financial resources 0 environmental impacts 0 staging of investments and technology levels * Develop/implement experimental programs of busways * Research the impacts and effectiveness of existing metros * Integrate mass transit with the rest of the transport system Action 6: Reform Public Transport Management and Operations * Separate Government planning and regulatory functions from the commercial functions of municipal public transport enterprises * Pursue the reforms of municipal-owned transport enterprises, including the franchising or concessioning of services * Actively seek the involvement of the private sector in public transport operations * Strengthen the regulatory role to deal with enhanced private sector involvement Action 7: Develop a Financing Strategy for the Transport Sector At the National level: * Consider the development, collection and allocation of appropriate road user charges, taxes and fees, including those on fuel, vehicle purchase, and vehicle operation (including the Road Maintenance Fee). * Improve the regulatory and legal framework for private sector investment At the Municipal level: * Set within the context of an overall municipal financing policy * Determine the needs for municipal involvement in investment, operations, and maintenance * To the extent possible, implement a program of cost recovery from each transport user * Determine the extent to which investments can be financed from debt or current revenues * Determine the potential scope for private sector funding, either to replace or to supplement existing municipal transport budgets, including international sources xxiv Concluding Statement of the Symposium Action 8: Strengthen the Framework for Transport Planning and Capacity Building At the National Level: X Establish guidelines for the transport planning process * Expand institutions for the education and training of transport professionals and officials At the Municipal Level: X Coordinate land development with transport development * Integrate transport planning with land use planning * Reflect the special characteristics of China's urban land use and resources in formulating urban transport development strategy and plans * Ensure rigorous evaluation of all plans and policies, based on the four criteria defined above * Exploit the use of consultants (both domestic and foreign) to supplement municipal agencies with specialized skills. Hou Jie I OPENING REMARKS Hou JIE' Ladies and Gentlemen, and Friends, The International Symposium on China's Urban Transport Development Strategy is open today. At a time when China is striding toward the twenty-first century with sustained rapid economic development, the convening of this symposium is of great importance to explore strategies for China's urban transport development, to allev iate the urban transport "bottleneck" problem under the new circumstance of a socialist market economy, and to achieve sustainable urban development. The symposium is cosponsored by the Ministry of Construction, the Ministry of Finance, the People's Bank of China, the World Bank and the Asian Development Bank. The participants include international and domestic transport experts, city mayors, and leaders of the state ministries and commissions concernied. Here, I would like, on behalf of the Ministry of Construction, to take the opportunity to extend our warmest welcome to all participants. I would also like to express our heartfelt thanks to the ministries and commissions concerned, the World Bank and the Asian Developmeiit Bank for their continuing assistance to us. The Ministry of Construction is in charge of overall management of China's urban roads and public passenger transport, including public buses, taxis, metros, ferries, and so on. Our work focuses on the provision of policy guidelines, formulation of laws and regulations, comprehensive planning, human resource development, interniationial cooperation, and macro control, as well as sectoral and regional balance. The Ministry is committed to China's urban development in a sustainable and coordinated way. I would like to make the following points concerning China's urban transport situation, its development goals and the related policy orientation for your reference. The Development of China's Urban Transport and Its Challenge Cities are the vehicle for China's social and economic development and the center for industries and population. Urban transport is considered the basis and prerequisite for urban development, an indispensable social facility for urban production and people's living, and a basic infrastructure for urban investment and living environment. It is pointed out in A Resolution to Pace up Tertiary Industry Development, promulgated by the Central Party Committee and the State Council in 1992, that the urban public utility sector is a basic sector that has overall and guiding impact on national economic development. The urban transport sector has been given priority for, and will continuously receive, support in the field of capital investment for the present and in the relatively longer-run period. Moreover, China's Agenda in I Hou Jie is Minister, Ministry of Construction. 2 Opening Remarks the 21st Century outlined an ambitious blueprint for urban transport required: development of transport will be vigorously promoted in order to meet the demand for transport by the national economic development and the improvement of people's living and to contribute to sustainable social and economic development. Since the beginning of economic reform and the opening-up policy and with the rapid development of the national economy and the dramatic improvement of people's living, materially and culturally, China's urban transport has entered a period that witnessed its most rapid construction and development in history, impressive even by world standards. Since 1990, total investment in urban infrastructure, including urban road-building, has reached Y 163.68 billion. In 1994, the total urban road length was 116,000 kilometers (km) and per capita road space reached 6.6 square meters in China's 622 cities. The implementation of the Eighth Five- Year Plan (8FYP) indicated that the increased road length during this period was 8,400 km; that is 156 percent of the planned length. By the end of 1994, about 100,000 vehicles were in operation for urban public transport, the total length of service routes was 200,000 km long, and total ridership of urban public transport was equivalent to one third of the total volume of passengers carried by civil aviation, railway and roads combined. There were nearly 400,000 taxis. On average, there were 6.1 standard bus equivalent units per 10,000 people in cities. Since 1990, the modes of urban transport and the structure of the road networks have been experiencing dramatic changes. In megacities, rapid transport facilities are starting to play a major role in urban transport and the single-level system is increasingly evolving to a multilevel system. Many cities, like Beijing, Tianjin, Shanghai, Guangzhou, Jinan, Chengdu and so on, have built and continue to improve their expressway systems. Following the operation and improvement of the metro system in Beijing, Tianjin and Shanghai, Guangzhou, Shenyang, Dalian, Chongqing, Wuhan and Qingdao also started to develop rapid rail transit system. Meanwhile, the large- and medium-size cities have formed public transport systems, in which bus and trolleybus play the primary role, and are supplemented by various other transport modes such as public minibuses and taxis. China's urban transport development has effectively ensured urban economic development and social progress. Over the past 15 years, it has basically met the tremendously increased demand for transport, generated from rapid economic growth at a 10 percent annual average rate, continuous improvement of people's living standard and constant expansion of cities; it also has contributed to urbanization. However, we clearly understand that we are now facing the following challenges: construction of urban transport has lagged behind; the structure of urban land use is not rational and some land for transport purposes is being taken for other uses; the urban transport system is incomplete, the function of the road transport system needs to be improved, the structure of road transport networks and that of passenger transport are irrational; the funding sources for investments are not stable; there is no clear and consistent policy support for investments with high levels of benefit spillover; and urban transport management lacks coordination and the level of management is low. Urban transport, as the artery of cities, plays a role in promoting economic development and facilitating mobility. However, traffic congestion now causes over Y 10 billion loss each year. The "bottleneck" of urban transport has become a predominant constraint on economic and social development. The development of a socialist market economy needs a healthy and well-developed urban transport system to connect socialized Hou Jie 3 production and circulation, and to establish an integrated open market. The underdevelopment of urban transport has adversely affected the establishment of a socialist market-economy system. Often, the urban transport system offers low-level services. Especially in some of the megacities, the travel duration of residents is prolonged, the operating efficiency of the road networks decreases every year, hindering effective improvement in the quality of life and environment in urban areas. Urban Transport Development Objectives and Policy Orientation for the Ninth Five-Year Plan and Toward the Year 2010 The next five years are the last period for achieving the second-phase strategic objective of China's socialist modernization; and the years 2000 to 2010 are crucial for realizing the third- phase strategic objective. As a basic sector with overall and guiding impact on national economic development, urban transport is being given due attention wheni the Ninth Five-Year Plan (9FYP) and the Strategic Plan for National Economic and Social Development to the Year 2010 are prepared. In the guideline for the formulation of these plans, the Central Party Committee stated that one of the primary tasks for social development should be "to enrich consumption goods, improve the consumption structure, and particularly, solve the housing and commuting problems." The 9FYP for Urban Construction and the Development Strategy to the Year 2000, drawn up by the Ministry of Construction, reflect the main idea that during this period, the functions and quality of service of urban infrastructure and public utilities, including road transport, water supply and sanitation, residential gas and heating, environment and waste management and public green area, should meet the needs of the national economy and social development. The overall service capacity and level should approach those of a middle-income country in the 1980s. A preliminary comprehensive socialized service system would be established. Both the urban investment environment and the people's living environment would be markedly improved. It is predicted that three urban agglomerations will be formed at their preliminary stage around Pearl River Delta, Yangtze Delta and Bohai Bay Rim by the first decade of the next century. And the pace of urbanization will accelerate. About 40 percent of the total population will live in urban areas by the year 2010. China will enter into an historical transition from an agricultural society into an urban society. The government policy for the auto industry will also induce a rapid increase in motor vehicles. In addition, the implementation of China 's Agenda for the 21st Century will bring about a series of strategic changes in population growth and environment protection, as well as strategic reforms in urban transport planning, construction and the management system. The pace of overall development of urban transport should be slightly higher than the average growth rate of national economic and social development. The overall development strategy for urban transport should be in line with the national economic and social development strategy. By the year 2010, China's urban transport should reach or near the level of a middle-income country of the early 1990s. During the 9FYP period, cities in China will build up urban road network systems of rational layout and structure. The megacities with a population over I million will form urban expressway systems step by step, of which some should gradually build up rapid rail transit facilities. In urban public transit, megacities will explore the transport modes with large carrying capacities as the major means, such as metros, buses and trolleybuses, supplemented by various 4 Opening Remarks other public transport means. Medium-size and small cities will mainly depend on buses and trolleybuses. The dispatch system, station facilities and overall efficiency of urban transport need significant improvement. By the year 2010, cities will build up an integrated urban-rural road network that is mainly composed of expressways and characterized by appropriate planning and a rational structure. The integrated urban-rural public passenger transport system will be composed of rail transport, buses, trolleybuses and taxis. In this system, the high-speed and large carrying-capacity facilities will be the backbones, and the small- and medium-capacity modes will be the popular modes. It is rather difficult to attain the objectives mentioned above. However, they can be achieved through deepening of urban transport reform, implementation of development strategies tailored to the specific situation of the country and the cities, and introduction of competition mechanisms by reforming the existing urban transport management system. Urban transport will be further developed by expanding cooperation with the outside world; improving efficiency in road construction, operation and management; establishing stable channels for investment and financing in road construction; giving priority to development of public transport means while controlling the development of cars in city areas; and strengthening the formulation and enforcement of urban transport laws and regulations. - Under unified planning and management by the state, the Government should continue to take primary responsibility for urban transport while introducing a competition mechanism to accelerate its development. Urban transport is a basic sector whose services bring social benefits to the public. The Government therefore should be responsible for its planning, construction and management to realize optimal allocation of resources and to establish a satisfactory and comprehensive social service system. Under these principles, in order to improve the operating efficiency of transport facilities, we should first apply a bidding system to urban road construction projects; second, establish a franchise system in public transport, open the taxi and minibus market and encourage competition on an equal footing. During the 9FYP period, the task of establishing a standard financial and management system in the state-owned urban transport companies should be accomplished to improve their operating efficiency and quality of services. The legal and regulatory framework should also be set up to ensure the legal rights of the general public, as well as those of the state-owned public transport enterprises. * To ensure the development of road transport and establishment of an integrated, highly efficient urban road transport management system, we should reform the existing urban transport management system and strengthen the ability of government, especially that of municipal government, in planning, construction and management, while effectively changing the function of government. Under guidelines of the central government, the development of urban transport and establishment of an efficient management system will mainly rely on municipal government. The municipal government's role in urban transport financing should be strengthened. Under the central government's macro guidance, the municipal government should be given sufficient autonomy to carry out its duties. Hou Jie 5 * The setup of a legal and regulatory framework in urban transport should be accelerated and urban transport management should be strengthened. The main body of the legal and regulatory framework comprises "Laws for Urban Public Utilities," "Regulations on Urban Road Management" and "Regulations on Urban Transport Management," which serve as the basis for the whole legal and regulatory system. A series of supporting codes and regulations will also be promulgated accordingly at the local level. In addition, the law and regulations should be effectively enforced to promote urban transport development. * Priority will be given to the development of urban public transport. According to China's current level of economic development, the uncontrolled development of private transport modes, in particular, private cars, should be contained in large- and medium-size cities. The emphasis should be on the development of public transport. Therefore, it is very important to undertake timely studies on the strategy and policy for promoting public transport as a development priority and to effectively carry out this policy. v A reliable investment and financial system for urban transport should be established and all financial channels should be explored. A user charge scheme for urban transport facilities should be gradually implemented. Meanwhile, comprehensive urban development should be vigorously promoted. Funds needed for construction of municipal infrastructure, including urban road facilities, can also be mobilized through paid term leases and transfer of state-owned land. External funds will be actively introduced through further promotion of exchange with the international community and exploration of various financial means so as to accelerate the construction of urban roads and public transport facilities. To implement the 9FYP and materialize the long-term objectives in the year 2010 for the national economy and social development, the urban transport sector faces big challenges. Together, we have organized this International Symposium for China's Urban Transport Development to help fine-tune the 9FYP and the medium- and long-term strategies for urban transport development in a more scientific way. It is hoped that this international symposium will guide and promote the development of China's urban transport and lay solid foundations for the formulation of an overall strategy for China's urban transport development. I thank the Chairman and every participant. Anthony J Pellegrini 7 OPENING REMARKS ANTHONY J. PELLEGRINI ' On behalf of the World Bank, I would like to congratulate Mr. Hou and the Ministry of Construction for arranging this important Symposium. I would also like to thanik the Ministry of Finance and the People's Bank of China for their valuable sponsorship. We particularly appreciate the opportunity to work in partnership with you and the Asian Development Bank in this important emerging arena. I would also like to thank the Governments of Switzerland and the Netherlands for their assistance in making this symposium possible. Finally, we welcome the presence of city leaders from across China who are grappling with the issues that will be discussed in the next days. China's Urban Transport Challenge China, probably more than any country in the world, is facing enormous demands for improved urban transport. Two factors are at play at the same time. First, China's cities are growing rapidly, and second, the rate of motorization is increasing at an unprecedented pace. Many view these changes with great optimism since they are occurring in tandem with rapid economic growth and ma be seen as a consequence of increased economic wealth. The World Bank's experience, however, shows that without effective urban transport policy, and without good planning many economies have failed to realize their true potential. In fact, the quality of life, particularly for the poor, can actually become worse. The cost of inaction is high. With congestion, deliveries take longer, and fewer trips are made. Prices increase as a result of inefficient distribution. It is estimated that over one-third of the gross city product of Bangkok, Thailand, about $4 million per day, is lost due to inefficient urban transport in that city. Poorly managed urban transport systems are a threat to people's health. Air pollution is reaching crisis proportions in many cities of the world. Children suffer the most. In Mexico City, schools must close some days, because it is not safe for children to play outside. The intelligence of Bangkok's children has been permanently reduced by exposure to lead. In many cities around the world, traffic accounts for over 60 percent of urban air pollution. Emissions from gasoline engines are a primary source of lead in urban areas. People's lives are affected in many other ways. Anthony J. Pellegrini is Director, Transportation, Water and Urban Development Department, The World Bank. 8 Opening Remarks Rapid motorization has brought with it an equally rapid increase in traffic accidents and deaths. In some parts of eastern Europe where there has been a rapid transition to a market economy but where there has not been prior traffic safety planning, the accident rate has surged over 10 times that of western Europe. Other problems: noise pollution is a growilg citizen complaint. People's communities can be cut off from the life of the city when highway design is poor. And major infrastructure works can mean that more people must move from their homes than would be necessary with more careful planning. Motorization: Not Good-Not Bad Motorization, that is, increased use of motor vehicles, may be ilevitable, but these problems are not inevitable. We must redirect our thinking. Instead of concentrating on how best to move vehicles, we must think of how to move people and also freight. We know from experience in many countries that a plan that is only based on building more roads will not work-cities cannot build their way to a solution. To find a solution, we must look closely at many parts of the puzzle. How can mass transit, be it an improved bus and or metro system, help? How can we maximize the use of existing infrastructure. Are we getting best use of existing roads? How can we plan for the best use of the bicycle? Do better pricing of transport services and improved land use planning hold important keys? As we consider these questions, we must remember that for transport policy to be successfiuI, it must be sustainable over the long term. The Bank has learned that the evaluation of policies and actions should consider three complementary criteria-economic and financial sustainability; social sustainability; and environmental sustainability. Economic Sustainability Economic andfinancial sustainabilily requires high levels of efficiency in the provision of transport services and high levels of efficiency in the use of infrastructure. * Revenues must be collected that are sufficient to properly operate and to expand the systems. * Pricing services so that costs are recovered is critical. Introducing competition between modes and among service providers is also important. Transport services can be disaggregated and unbundled and some moved to the market by allowing commercialization and private sector participation through franchising and concessions. However, to get the benefits from competition, we must create regulatory institutions and performance standards that will ensure that the population in fact gains the benefits that are possible. Anthony J Pellegrini 9 Social Sustainability To achieve social sustainability, the population must perceive that the transport plans are fair and that the transport requirements of all parts of society are being addressed. The goal is to give people access to jobs, education, and social services. If subsidies are needed, efforts should be made to target the subsidies on families most in need. Subsidies should not remove incenitives for efficient operation. Mechanisms should be developed that provide reliable revenues at the local level to finance these subsidies. Land use planning and transportation should be integrated. Both public mass transport and nonmotorized modes of transport should be integral parts of the planninig process. And pedestrian access and the appropriate role of bicycles in urbanl transportationi should be planned for and protected according to the plan. Environmental Sustainability Finally, policies that seek to improve the environment so as to protect the health of people should have high priority. To combat air pollution, cleaner fuels that eliminate sulfur and lead are needed. Plans to promote safety, particularly for pedestrians and bicyclists, must be developed. And charging users for the environmental damage they cause, for the infrastructure they use and for the congestion they impose on others, would help to improve the enviroilneit. Sustainability, Synergy, and Balance Evaluation of alternative actions and investments within this triangle of Economic, Social, and Environmental Sustainability goals enables you who are responsible for urban transport policy to identify the conflicts in trying to achieve these goals and exploit synergy. This will contribute to development of a properly balanced program that supports community values and contributes to sustainable urban development. The World Bank in Partnership with China How can the World Bank help? Our partnership with China on urban transport issues is promoting these ideas in Shanghai where two projects are improving the operational and economic efficiency of that city's urban transport system. We have traffic improvement components in a project in Tianjin and we are also working closely with the Guangzhou municipal authorities who have completed an important strategic study that takes a comprehensive approach to urban transport development. We hope that this will lead to a project that will implement innovative solutions in the area of public transport development, environmental protection, institutional reform and traffic management. We are also cooperating with the National Environmental Protection Agency and the State Science and Techilology Commission on the development of strategies for improved motor vehicle pollution control. More broadly, our contacts with many cities in China have highlighted the growing sense of urgency on the part of city leaders as they face sharp increases in traffic congestioll and environmental degradation. Our purpose in being with you at this SymposiUm is thus to work together: in order to understand the nature of the urban transportation problem in Chilia; to draw insights from international experience; and to see whether there is an emerging consensus on the 10 Opening Remarks directions city leaders and the national government can take to meet these challenges. We hope that these insights will provide direction for future cooperation with China. Closing In closing, I would again like to express the appreciation of the World Bank for the opportunity to participate in this important event. The real value of a conference such as this is the opportunity for all of us to share experiences, and to learn from one another. I look forward to the next few days with great anticipation. Thank you. Jin Liqun Ii OPENING REMARKS JIN LIQUN' Honorable Guests, Ladies, Gentlemen, and Friends, First of all, I would like to extend, on behalf of the Ministry of Finance, my warm welcome to our friends from the World Bank, the Asian Development Bank, participants from municipalities and experts from both China and abroad. Hearty gratitude also goes to our colleagues of the Ministry of Construction who host this symposium, and colleagues from other commissions and ministries who have saved no effort in supporting this symposium. This year is the end of the Eighth Five-Year Plan period and also the critical year for drafting the Ninth Five-Year Plan. The last five years have witnessed a matured and fruitful cooperation between China and the World Bank and the Asian Development Bank. Completion of the Shanghai Inner-Ring Road and the two Huangpu River Bridges and the ongoing preparation of the Guangzhou City Center Transport Project and the Liaoning Urban Transport Project represent a new and a more extensive cooperation in the area of urban transport. Urban transport, as one of the major infrastructures for modern cities, is of particular strategic significance. China, which is undergoing its transition from a planned economy to a socialist market economy, needs to make the most of market mechanisms in resource allocation, and therefore should strengthen the pivotal and central role of cities in economic development and market functioning. Along with the deepening of reform, further opening-up to the outside world and rapid development of the national economy, our cities have been growing both in their numbers and in their size. Although significant improvements have been made in vehicles and urban roads, many cities find it increasingly difficult to cope with the continuing rapid traffic growth, a phenomenon closely associated with economic growth. In some areas where intercity roads have been improved, urban transport has become an enormous traffic bottleneck that hampers economic development. Many large and medium cities are experiencing increasing road congestion, travel inconvenience, and increasing pollution from auto emissions. Meanwhile, these cities are further pressured by the rapid growth of the urban population, especially the floating population. Very often, the completion of new transport projects is quickly followed by the emergence of new traffic problems. Urban residents are longing for greater improvements in urban road conditions and transport modes, especially public transit. Approaching the twenty-first century, great changes will take place in the national economic structure and urban-rural population ratio. Modernization of the national economy and urban economy will surely stimulate modernization in urban transport. Facing the opportunities l Jin Liqun is Assistant Minister, Ministry of Finance. 12 Opening Remarks and pressure of historical development, urban transport must enter a new strategic phase. We need to work out a development strategy that is scientific, realistic, and future- and people- oriented, and that fits Chinese city characteristics. Urban transport is a key area of infrastructure investment, construction and management. It is closely linked with development of an urban economy and a national economy. By itself, it is also a precondition for development of a national economy and an urban economy. Further development of urban transport requires deepening reforms in the urban economy and in the management systems of related enterprises and institutions. Municipal governments should actively create opportunities for enterprises to operate independently. Meanwhile, urban transport development should be planned, not only in consistency with an urban master plan, but also with urban, even regional, socioeconomic development plans. The formulation of an urban transport development plan requires studies in the areas of institutional reforms, financing, management, cost saving, and efficiency improvement. Efforts also should be made to introduce advanced technology and international experience, as well as to make good use of domestic experience and lessons. To develop urban transport, we should not only increase investments through every effort, but also emphasize investment efficiency and scientific management, and make the transition from a long-adopted singled-minded approach to a more rigorous and diversified one. To this end, we should establish the right principles for urban transport development; we should formulate on a scientific basis the development plans and implementation schemes that are tailored for each city; we should ensure design, operation and management be carried out in a scientific way; we should make good efforts to save various resources including land, labor and capital; we should properly arrange resettlement work required for land requisition; we should protect urban air from auto pollution and reduce traffic noise; we should strengthen the management of urban road transport and the management of enterprises and institutions in the urban transport sector; and we should try our best to increase the overall benefits of urban transport investment. As to investment sources, not only should we improve the management of the infrastructure construction budget, urban construction and maintenance funds as well as other related sources, but also take initiatives to attract more investments from the general public, enterprises and foreign financial institutions. This can be done through further opening to the outside world, the deepening of reforms, as well as improvement of management. According to the local capability of raising counterpart funds and the capability of repaying debts, the Central Government will, through an integrated management system of borrowing, spending and repaying that unifies obligations, rights and benefits, continue to borrow, in the amount appropriate and with high efficiency, from international financial institutions for urban transport development. Urban transport is a key area of urban development. It requires not only the continuing efforts by government agencies within the sector, but also the support of municipal governments and agencies of other sectors. To follow the Central Government guideline for capturing development opportunity, deepening reforms, expanding openness, promoting development, and maintaining stability, all municipal governments and agencies will bear more important responsibilities. Yet, our past experience suggests that our cities will be able to achieve more in urban transport and socioeconomic development. Jin Liqun 13 Please allow me to wish the symposium success in helping us develop new thinking for urban transport development strategies. Thank you. A. Timothy Peterson 15 OPENING REMARKS A. TIMOTHY PETERSON' Introduction Honorable Minister Hou Jie, Mr. Chairman, distinguished guests, ladies and gentlemen. The Asian Development Bank is very pleased to take part in this Symposium on Urban Transport that, for the first titne in the People's Republic of China, brings together senior urban transport experts to discuss how different urban transport problems and issues can be tackled. It provides an excellent opportunity to share ideas of how other countries are addressing similar issues and how these might be applicable to the numerous cities in China. The Asian Development Bank Before moving on to the theme of the conference, let me give a brief background on the operations of ADB. ADB was established in 1966 and today it has 56 members, 36 of which are developing countries in the Asia-Pacific region. In almost 29 years of operations, the Bank has provided loans totaling more than $50 billion and technical assistance grants approaching $1 billion. The Bank has become a major catalyst in promoting the economic and social development of the Asia-Pacific region. Bank operations cover the entire spectrum of economic development-with particular emphasis on agriculture, energy, capital market development, transport and communications and social infrastructure. Its objectives are achieved through loans and equity investments to both the public and private sectors, through technical assistance grants and through policy dialogue. The Asian Development Bank has had extensive experience in funding transport projects in China. Given the strategic importance of the sector to the development and expansion of the economy, the Bank places high priority on relieving bottlenecks that impede efficiency and hold back the development of essential services. Our involvement in China covers a wide range of sector activities and includes support for roads, railways, ports, urban transport and urban development. Since 1986, when the People's Republic of China joined the Bank as a developing member country, ADB has made 17 loans for transport and communications infrastructure totaling more than $1.9 billion and has provided 44 technical assistance grants amounting to $17.8 million. A. Timothy Peterson is Deputy Director, Infrastructure, Energy and Financial Sectors Department (East). Asian Development Bank. 16 Opening Remarks Urban Growth in Asia This morning I would like to say a few words on the growth of urbail areas and urbanization in the Asia-Pacific region and how it is expected to chanige in the future. The Asia-Pacific region has witnessed considerable econoinic growthl over the past three decades. Consistent expansion of individual economies across the regioni has had a pronounced impact upon urban communities of all sizes. As economies have expanded, their structure has changed the composition of economies, moving them away from dispcrsed rulal-based activities toward manufacturing and services. The strong growthi trends have diverted economic and social activities away from rural-based operations toward those miore suited to urban enviroilnents and this is reflected in the accelerated trend of urbanization across the region. Over the past 30 years, the number of people living in urban areas in developing Asia has increased almost fourfold from 270 million to 850 million people. This represents an increase in urbaniizationi from 18 to 30 percent of total population. Furthermore, over the next 30 years the regioni's urbanl population is expected to increase to 2,250 million, representing more than 54 percenit of total population. Consequent with increasing urbanization is the expansionl in size of urbani comIlmunities. By the end of this century there are expected to be 280 cities in developing Asia with populations greater than half a million, of which 160 will exceed I million and( 32 exceed 4 millionl. Thle rapid pace of urbanization across the region is a major contributor to a range ol urbhan issues incltidinig those pertaining to urban transport. Vehicle ownership across the Asia-Pacific region is also expandinig rapidly. Historically, national vehicle fleets have increased faster than economic growth and this trend is expected to continue as per capita incomes continue to rise and vehicle owinership becomiles more affordable. In many countries such as in India, Malaysia and Thailand, the vehicle fleets are increasing at more than 15 percent a year. At this growth rate, the vehicle population is doubling every five years. Despite these large increases in vehicle ownership, current vehlicle ownershilp remains low in comparison with developed countries. The continued expectation of increasing prosperity across the region is expected to result in substantial increases in vehlicle fleets in thle futtire. The very large expected urbanization explosion across the region, coupled with increasing vehicle ownership and incomes, will combine to exert heavy pressure upOIn urban infrastructure. It is to be expected that this will significantly increase demand for substantial expansionis in transport network capacity and services. The magnitude of the resources required to provide adequate capacity and maintain quality of services is expected to be a major- probleiim for t(he majority of cities in the region. The particular problems associated with urban transport are primarily a function of the interactive factors contributing to development. These include growing population levels and per capita income and progressive urbanization. In certain respects they are a Iresult of achieving progress in economic and social development. Together, these factors constitute the major parameters that influence the demand for movement of both passengers and freighit. Over the past decade, the problems and constraimis associated with urban transport in the region have increased rapidly in both scale and intensity; today they ar e widespread throughout the region. Attitudes toward supporting development of cities and towns are changing rapidly. It is A. Timothy Peterson 17 increasingly being realized that improving the efficiency of urban areas is extremely important if rates of national economic growth are to be maintained. The development of physical infrastructure and related services is of paramount importance if continued investment in industrial and commercial activities are to be sustained and employment created. It is also increasingly recognized that social conditions in urban areas, including measures to improve the environment, need to be urgently addressed. Problems associated with traffic growth, infrastructure bottlenecks and deteriorating environmental and social conditions are rapidly combining to reduce the benefits of urban living. As a result, the need to address urban issues, including urban transport, has moved up the developmental agenda from both an economic and social perspective. With few exceptions, investment in urban transport has generally received low priority, in terms of budget allocations, in the majority of developing countries. Deferred investment is now resulting in severe transport constraints impeding the efficiency of urban areas. In several of the region's major cities, the need to tackle urban transport problems has risen to the top of the political agenda. The demanid for fhianicial resources is, and will conitinue to be, considerable and is expected to remnain outside the range that can be provided by the majority of governments and municipal authorities. External assistance will be required to help identify urban transport needs and priorities, ascertain the most appropriate use of available resources, investigate sector policies as well as provide finanlcial support for individual projects. The demand for resources is so high that innovative approaches to resource mobilization will be needed if the sector is to erode the escalating capital cost requirements. While provision of public sector resources will remain important, it is expected that a large proportion of resources will need to be sourced from the nongoverniment sector, not only in traditional areas for the provision of services but also increasingly for the provisioni of infrastructure and its operation. Within the Asia-Pacific region, it is anticipated that the finanicial allocations to urban transport, from all sources, will increase markedly over the next decade and beyond. The role of the Asian Development Bank will not only support the goals and objectives of governments but, more importantly, act as a catalyst for the mobilization of' nontraditionial sources of finance through provision of technical and financial support to private sector initiatives. The timing of this Symposium is very appropriate as the problems associated with urban transport in China's cities have only begun to emerge in the past few years. Thle Symposium, therefore, provides a suitable fOruImI to discuss how various urban transport problems have been dealt with in other counitries and how their experience might be of benefit to urban managers and planners in Chinia. 1 see from the agenida of the Symposiumli that a umiber of important topics are to be discussed and I would like to make reference to those that we, at ADB. consider to be most important. The problems associated with motorization, as I have noted earlier, are widespread throughout the Asia-Pacific regioni. Many countries have niot yet succeeded in addressing these issues and perhaps the onily city that has managed to containi problems associated with motorization is Singapore. In most of the cities in the region, governmenit and municipal authorities have only just begun to tackle the problem but because vehicle fleets have already reached unmaniageable proportions many actions are characterized as being "too little, too late." Thle position in China is unique in that the current numliber of motorized vehicles is comparatively small; therefore, there is 18 Opening Remarks still time to address and introduce measures to manage the problems associated with motorization before they emerge. It is important that policies designed to reduce the use of private transport, particularly in central business districts, are promoted before traffic intensities growth to an unmanageable level. Such policies will need to be enacted together with improvements in public transport as unless suitable alternatives are available reductions in private transport utilization will be impossible. I look forward to listening to your discussions on this topic as I believe that China is in a good position to be able to introduce the necessary measures to minimize the adverse impacts of motorization that have occurred in most other parts of the world. A second topic of interest concerns the role of public transport and the possible introduction of mass transit in some cities. In the majority of cities in the region, public transport has an important role in providing the means of movement for the majority of trips purposes; more than half of the trips in regional cities use available public transport services. In many Chinese cities, modal split has very different characteristics, with public transport only accounting for a very small proportion of total trips. Within the region, many of the successful public transport services are provided wholly by private operators. They provide efficient, low-cost services that are affordable to the public and are not a burden upon municipal budgets. I believe that there are possibilities for introducing some of the experiences in the region to cities in China to enhance the provision of public transport services and reduce the burden on municipal budgets. It will also provide the framework for improving quality and reliability and also introduce services that are responsive to the demand and needs of the residents. The introduction of mass transit in the region has been relatively slow due to the enormous costs associated with constructing such infrastructure. Very few municipal authorities have the resources to finance such schemes, which require the largest investment a city is ever likely to meet. While transit schemes can provide good economic benefits, they frequently fail to generate sufficient financial returns to cover their construction and operating costs. Most mass transit projects, therefore, require operating subsidies that can be large if a low fare policy is adopted. The introduction of mass transit is often seen by municipal authorities to be a panacea that will resolve the transport problems. Unfortunately, transport problems are complex and many investments have not achieved their stated goals: several have added to the financial burden of the city and its residents through the need to raise local taxes to pay for large operating subsidies. I look forward to listening to your discussions on the public transport and mass transit papers as these topics are critical to the provision of good accessibility in urban areas. Efficient and affordable transport services are required not only to promote economic growth and social development but also to provide urban residents, particularly low-income groups, with an improved quality of life. Another area of particular interest relates to the future role and use of the bicycle in China. Your country has the largest bicycle fleet in the world and in many cities, there are as many bicycles as people. The bicycle is affordable and its current use is widespread. Given the relatively short trip distances currently being made, the bicycle is the preferred modal choice and is highly competitive with most public transport services. However, this might not necessarily be the case in the future as cities expand both in terms of population and area. This is likely to result in longer trip distances, particularly for journeys to work, and the bicycle will become less competitive relative to motorized modes. The role of the bicycle in the future is an important urban transport issue in China and is one that I am sure will simulate interesting discussion. A. Timothy Peterson 19 Mobilization of Resources The Asian Development Bank has observed, with great interest, the rapid strides that China has been making in attaining its growth objectives. We are proud to have played a part in this effort in recent years, and stand ready to continue to help where we can in the future. We are encouraged by the sense of urgency we see in the country's efforts to expand and improve its basic infrastructure. And, we are pleased to see widespread appreciation of the importance of attracting private capital and foreign investment in several infrastructure sectors including urban transport. Despite these developments, considerable scope remains for greater private-sector involvement and investment in providing infrastructure facilities. This is driven, in part, by the enormity of the challenge that faces the country-namely, to mobilize the necessary resources to provide basic infrastructure services in a rapidly growing economy. Such resources must come in large part from the private sector, as financial resources from the various levels of government, domestic banks and the international donor community are limited. While the financing of urban transport is not a major theme of this Symposium, it will have an important, if not predominant, bearing on all decisions taken in the transport sector by municipal authorities. To meet this financing challenge, it will be necessary to examine and adjust the role of state and local government in infrastructure development, to strengthen the partnership between state and nonstate sectors, and to attract more foreign investment in municipal development. Private-sector investment can be mobilized through joint ventures, build-operate-transfer (BOT) projects and from the domestic and international markets. Recommendations from this Symposium on ways in which mobilization of resources to facilitate development of urban transport infrastructure and services would be particularly relevant. Conclusions Distinguished participants, the Asian Development Bank is pleased to support this Symposium on Urban Transport to discuss important urban transport issues emerging in Chinese cities. Through this Symposium we expect that a better understanding of the urban transport problems and issues will be ascertained and that the experiences of other countries in the region, as well as outside the region, will be of use to resolving transport problems in China. I look forward to your deliberations over the next three days and hope that this Symposium will provide fruitful and useful information to government and municipal authorities, which will enable them to better serve urban communities in China. I also hope that it will provide a basis for further future support by ADB to assist with the development of urban transport in China. Thank you. Li Zhendong 21 KEYNOTE ADDRESS 1: REFORMING CHINA'S URBAN TRANSPORT SECTOR Li ZHENDONG' Respected Chairman, Ladies, Gentlemen and Friends: Along with rapid urban economic development and the deepening of economic reform in China, the urban transport industry is experiencing dramatic changes. On the one hand, the investment of urban transport infrastructure in large cities keeps growing, the share of urban transport investment in total fixed asset investment gradually increases, and urban roads and passenger transport vehicles continuously grow. On the other hand, the growth rate of motor vehicle ownership and passenger tra'nsport volume is higher than that of transport infrastructure and facilities, and thus serious problems emerge in urban public transport, such as traffic congestion and the low quality of public transport in large cities. These problems have hampered urban social and economic development, resulting in a loss of efficiency in urban transport and deterioration of people's living condition in urban areas. Problems in China's urban transport have caused concern. The central, provincial and municipal governments are paying increasing attention to these problems. Facing a big challenge in the development of urban transport, we realize the urgency of making a comprehensive strategy for urban transport development to guide the reform and development of the urban transport sector in large cities. The International Symposium on China's Urban Transport Development Strategy will have significant impact on the formulation of urban transport development strategy in China. It will also give a strong impetus to urban transport development in China's large cities. The concerns about China's urban transport problems from international financial institutions such as the World Bank and the Asian Development Bank are very important. I hope that China's urban transport industry will get more financial and technical assistance from the World Bank and the Asian Development Bank. The existing problems in Chinese urban transport and their causes are very complicated. Mr. Hou Jie, the Minister of Construction, has presented some important opinions concerning the current status of China's urban transport development and the policy orientation for the urban transport industry, which provides us with a general picture of China's urban transport industry. The fundamental solution to China's urban transport problems lies in reform. Therefore, I will make some comments about the reforms in China's urban transport industry. I Li Zhendong is Vice-Minister, Ministry of Construction. 22 Keynote Address 1: Reforming China's Urban Transport Sector The Urban Transport Industry Should Meet the Demands of a Socialist Market Economy The socialist market economy mechanism requires that regulations, policies, institutional setup and management for every subsector of urban transport be made on the principles of a market economy. It also requires the use of economic leverage to reach an equilibrium in demand and supply and formulate a sound investment cycle mechanism in urban infrastructure and facilities, that is, investment-construction-user charge and cost recovery-investment. In order to meet the requirement of a market economy, the commercial attribute of transport infrastructure should be recognized. It is especially important to keep this recognition in mind while we make policies influencing the provision of urban transport infrastructure. This is to say that the price of transport infrastructure should be in line with its value of providing basic social benefits to society. This principle should also be applied to urban public transport services. Although these services have strong public welfare implications, their demand and supply should be determined, to the highest possible extent, through the price mechanism. Urban roads and parking lots take up a large share of urban lands and absorb a significant amount of construction funds. The problem at present is the lack of linkage between the use of urban roads and the sources of financing of roads and other related infrastructure. Urban infrastructure has been treated as public facilities providing social services. Their provision and usage thus do no reflect market principles. As a result, road use is either free or badly underpriced for a large number of vehicles (or road users), directly contributing to inefficient use of roads and growing road traffic congestion. It is therefore important to follow socialist market economy principles and use economic leverage to balance the relationship between construction and usage to realize sustainable and balanced development of urban transport. Opening the market and introducing competition are the main avenue to improve internal management of the urban transport sector, increase its operating efficiency and economic return, and promote its development. The concrete measures include: the gradual opening of the construction market for urban transport projects and the public transport market; standardizing market activities through the establishment of laws and regulations; strengthening sector management; introducing competition to improve the quality of design, construction and services, reducing costs of transport infrastructure construction and services; and introducing competition to improve state-owned enterprise operation and management. Urban land development, especially the construction of large-scale commercial centers in old town areas of large cities, and unrestricted real state development, aiming at an increase in floor-area ratio, intensify the concentration of urban transport demands, which puts enormous pressure on urban transport and causes the drastic increase in infrastructure investment and construction in certain areas by the municipal government. Therefore, it is necessary to establish an evaluation system to analyze the impact of large real state development projects on the transport system, which will be taken into consideration in project planning and approval. This practice is to ensure balanced development of land and transport infrastructure, whichi should complement each other. Li Zhendong 23 Reforming Urban Transport Management System Urban transport is a complex system. It includes planning, construction and maintenance of transport infrastructure; administration and ownership of transport means; provision of transport services; management of transport safety; formulation and implementation of policies in government finance, investment and pricing; land development plans; urban residents' living and working patterns; environmental protection and so forth. All these aspects of an urban transport system should be well coordinated. At present, however, there are many serious problems in Chinese big cities' transport management systems, especially the lack of coordination among different agencies of the system, overlapping of organizations and not- clearly-defined purviews and responsibilities for each organization. As a result, the following phenomena are pervasive: difficulty in reaching consensus, lack of communication, mutual restriction, and often many different policies concerning one particular issue from different government agencies. Therefore, the administration and management of transport systems have very low efficiency. We should take action and deepen the reform of the management system to correct those shortcomings and increase efficiency. The guiding principles are "simplification, effectiveness and unification." Another important task in the reform of management system is to change the Government's management style. It requires change from the current style of mainly using administrative instruments to a combination of legal, administrative and economic means; change from direct management to indirect management; change from management by departments and ministries to management by industry sector and society; separation of ownership and management of state-owned enterprises; and separation of government functions and enterprise operation. Building up Rational and Stable Funding Channels for Transport Infrastructure Construction In China, financial resources for urban transport infrastructure mainly come from urban construction and maintenance funds, government budget allocation, bank loans, and surcharge on property for municipal infrastructure development. The existing system of financing has the following problems: funding channels are not stable; resources for transport investment are separated from demands for transport infrastructure; the road maintenance fees are not used for urban road construction, although 70 percent of all motor vehicles make their trips on urban roads; the existing system of tax and surcharges does not reflect the relationship between the benefits to road users and the cost of road construction. In building rational and stable funding channels, the first task is to reform the existing system of taxation governing urban transport, and levy taxes on motor vehicle fuels, tires and usage of nonmotorized vehicles. The charge for road use should be in line with the cost. The ''users pay their way" principle should be reflected in the use of infrastructure facilities. Economic leverage should be used to influence the supply to and demand for transport so as to establish rational and stable funding channels for transport infrastructure and facilities. In line with government regulation, many municipalities allocate a lion's share of revenue from the lease and transfer of land-use rights into the construction of urban infrastructure. This practice relieves the tension between supply and demand to a certain degree. 24 Keynote Address 1: Reforming China 's Urban Transport Sector Some municipalities have urban road construction as the forerunner of urban development and implement a comprehensive development plan, thereby solving the serious problem of insufficient funds in the renovation of transport infrastructure in old town areas. In addition, the following methods in fund-raising should be promoted and applied to a broad area in a stable fashion: using domestic and international loans on favorable terms, issuing construction bonds, and attracting foreign investment in construction and operation through the lease of operating rights. Effectively Implementing Public Transit Priority Policy In the light of increasingly crowded urban transport conditions in Chinese large cities, development of public transport is a rational strategy to effectively relieve the tension between supply and demand. Urban public transport is the one with the lowest social costs and highest comprehensive benefits among all means of urban transport. Not only does it offer the social benefits of urban transport, but it also provides low-price services to middle- and low-income urban households. Taking Chinese cities' specifics, such as land, population and economic capacity, into consideration, we should let the regular bus and trolleybus play a major role among all the public transport modes including bus, trolleybus, minibus, taxi, light rail and subway. A small number of megacities may gradually develop a rapid rail transit system. In recent years, the Ministry of Construction has formulated a series of policies encouraging the development of public transport. However, implementation of these policies has been slow and ineffective. The prioritization of public transport asks for strong government support in investment and price, including the implementation of franchising public transport operations and the efficient use of government subsidies. It also needs preferential treatment in road use, transport management, provision of facilities and equipment and so on1. In 1994, the central government launched "The Development Policy for the Automobile Industry." This policy encourages the production of automobiles, especially cars, and the private ownership of family cars. Prior to the rapid development of urban private cars, we should build up an urban public transport system offering high-quality services to enhance the attractiveness of public transit. The objective is to make China's urban transport structure more rational and efficient. Strengthening the Reform of State-Owned Public Transport Enterprises In China, the state-owned urban public transport is of a large scale. By 1994, the state- owned enterprises had 109,000 standard bus equivalent units. Carrying 75 percent of the total ridership, they are the major carriers in urban public transit. It is very import for the development of urban public transport to deepen state-owned enterprise reform and improve their efficiency and quality of services. At present, the Ministry of Construction is promoting the franchising system in public transport operations and strengthening the management of the public transport sector through legal and administrative means. These policies aim at the formation of an urban passenger transport market with fair competition. In the process of establishing advanced accounting and organizational systems, the state-owned enterprises should deepen the enterprise reform, LiZhendong 25 promote scientific management and technology innovation, improve the performance of operations, reduce and eventually eliminate the operational losses. They should continue to play the major role in urban passenger transport. Municipal governments should strengthen management of the urban public sector, rationalize the institutional framework of management and improve the system of laws and regulations. They should effectively carry out their administrative functions such as planning, coordination, inspection, supervision and so forth. Municipal governments should also help to create a conducive environment for the reform of state-owned public transport enterprises. They should delegate authority to the state-owned public transport enterprises in personnel management, finance and equipment. They should also take resolute measures to separate the government function from enterprises, the ownership rights from the rights of management and operation. The idea is to develop and improve the public transport enterprises through market competition. Enhancing the Research and Planning of Urban Transport The development of urban transport follows its owIn law. Depending on the levels of social and economic development and institutional arrangements, each country's urban transport will have different characteristics. China is now experiencing rapid urban economic development and transition of the economic system. We therefore must pay close attention to the research of urban transport theories to discover the laws and patterns of urban transport development in Chinese big cities. The research should include many fields such as social and economic development, government finance, environment, culture, administrative system, etc. Our objective is to gradually establish a system of theories, taking China's country specifics into consideration, guiding the formation of urban transport development strategies and policies, and the practice in construction, maintenance and management of urban transport. Urban transport construction projects involve a wide spectrum of the society and entail large-scale investments. To minimize the mistakes in policy-making and investment decisions, we should strengthen and promote feasibility studies in urban transport planning and project identification, appraisal and approval, to improve the quality of and efficiency in government decision-making. Meanwhile, we should pay close attention to fundamental theory research in transport projects to improve urban transport planning and project design. At present, the Ministry of Construction is preparing priority research projects for urban transport, which are identified in China's Agenda in the 21st Century, and key research projects of urban transport science and technology identified for the Ninth Five-Year Plan period. We hope that these research projects, with international cooperation, will introduce advanced technology from abroad, help in raising the technical levels of China's urban transport sector and promote fundamental theory research in urban transport. In order to meet the demand of China's urban transport development, we should educate and foster a team of competent professionals for urban transport research and design. Up until now, the Ministry of Construction and some megacities already have a number of research and design institutes for urban transport. The preparation of the Urban Transport Engineerilg and Technology Research Center is now being undertaken by the Ministry of Construction. We 26 Keynote Address 1: Reforming China's Urban Transport Sector should deepen reform of these research and design institutes to meet the demand of economic system transition and better serve the rapid development of urban transport. Thank you! John R. Meyer 27 KEYNOTE ADDRESS 2: THE URBAN TRANSPORTATION PROBLEM IN A CONTEXT OF ECONOMIC DEVELOPMENT JOHN R. MEYER' Some centuries ago, our forefathers brought forth in this world some places called cities. On the whole, these have proven very good places to live-wealthier, more innovative, more productive than the surrounding hinterlands. Today, however, a certain blight or difficulty has settled in: urban dwellers around the world increasinglv find themselves in difficulties when moving about their cities. Even when they can move, they find that the process often generates all kinds of discomforts and harm: pollution, congestion, noise, dirt, dust, and delays. Furthermore, urban transport improvements often seem self-defeating, creating as many problems as they solve; they also commonly make the city a less pleasant place to live. This pervasive phenomenon has come to be known as the "urban transportation problem." What to do about it? Can it be solved? These are the issues, with an emphasis on the special characteristics of China, to be addressed here. But first perhaps we should ask why? What has caused this problem? The answer basically resides in urban growth processes (as several people suggest in their papers at this conference). As a city grows, it typically evolves from a centralized, high-density, compact mass into a more decentralized larger area. As this happens, the pattern of urban trip-making also changes dramatically; specifically, from a mainly radial pattern serving a centralized city to a mix of radial, circumferential and cross-city patterns serving an increasingly diffuse and decentralized city. Obviously, as a city's population increases, the volume of total urban trips also increases. Significantly, though, commuter trips per employed worker fall as an economy develops (because of a shorter work week or abandonment of long midday lunch or rest periods, etc.). By contrast, other types of trips (for shopping, visiting friends or family, recreation, etc.) usually grow more rapidly than the economy. The total number of urban trips thus increases disproportionately with economic development, but with the commuter share of total urban trips declining as incomes rise. Because of these disparate rates of development, utilization of the various parts of a city's transport network will also change with development. The typical network will experience "economies of fill" as trips are distributed relatively more to underutilized segments of the system. Disproportionate increases will occur in the use of facilities off-peak and at peripheral or less central locations. This overall improvement in the use of system capacity will lower the average capital or capacity costs per trip, all else equal (wlhich of course it will rarely be!). On 1 John R. Meyer is Professor of Capital Formation and Economic Growth, Harvard University, United States. 28 Keynote Address 2: The Urban Transportation Problem in a Context of Economic Development the other hand, if and when there are no more underutilized segments of which to take advantage, incremental costs of added use can rise sharply: this can become a particularly vexatious problem when a society becomes relatively advanced in its development. To a rough approximation, all this growth and change translates into a city typically progressing through three stages of transport development. These can be very broadly characterized as that of (a) animal or human power (bullock cart, ricksha, etc.); (b) mechanized public transportation (vans, buses, rail systems...); and (c) mechanized private or individualized transportation (mopeds, motorcycles, automobiles...). The evolution from one stage to another does not imply uniform progression from bad to good to better. Each stage has its distinctive features, some advantageous and some not. Nirvana does not wait at the end of the development process-just different problems and possibilities. Some urban dwellers, I am sure, will long for the "good old days" of stages past. Others will want to hasten the inevitable processes of change. Deciding how to compromise these conflicting views will be a major challenge for political leadership. Also, like most taxonomies of human activities, this one has its complications and definitional ambiguities. In particular, what is to be made of the bicycle, especially in its modern and increasingly "high-tech" versions? Are bicycles part of the past, Stage 1, or part of the future, Stage 3's individual transport? Obviously, this is a question of some importance in the Chinese context. As discussed by Welleman in Theme Paper 5, the prevalence and acceptance of bicycles as a mode of transport in China's urban areas is well established. The bicycle's convenience (door-to-door service, scheduling flexibility, ability to circumvent congestion...), combined with its relatively low capital cost, has made it a dominant mode in China. By contrast, the bus is often considered an inferior mode and is used primarily by those who cannot afford the investment or otherwise do not have access to a bicycle (the very young and old, the physically handicapped, etc.). At least in part, the explanation of the bicycle's dominance probably resides in other special characteristics of the Chinese urban scene: very compact and dense cities with relatively short average trip lengths. Nevertheless, as incomes rise, demand for more sophisticated bicycles, both pedal-powered and motorized, as well as larger motorized vehicles such as mopeds, vespas and subcompact autos, will increase. On the other hand, as cities decentralize, and urban trips become longer, the bus will likely be more competitive with the bicycle for at least some trips. Bus transport may also improve its position by making service, quality and comfort improvements. Still, the possibility cannot be ruled out that Chinese cities, to a considerable extent, may skip over much of Stage 2, going almost directly from Stage I to 3. This could be rather traumatic, even chaotic. The environmental damage, for example, could be sudden and extensive, as Walsh points out in Theme Paper 2. Less time would be available for adaptation. Fewer options for coping with the peculiarly difficult problems of Stage 3 would be in place. In China, where leaded fuels are still in widespread use, and most motorbikes have two-stroke engines, increased motorization could be particularly harmful to the environment. On the other hand, because of the possible continued popularity of the bicycle mode, Stage 3 may also be John R. Meyer 29 more different in China than elsewhere: less polluting and environmentally more benign but more difficult to manage because of widely different modal speeds. The substitution of individual (Stage 3) for public (Stage 2) transport is not only driven by apparent consumer preferences but also by the rising labor costs that normally accompany economic development. Importantly, any substitution of individual for public transport reduces the need for hired labor to operate vehicles. This can make even the automobile, let alone the motorbike or moped or "mountain bike," a relatively cost-efficient mode of transport, especially in providing the higher quality of transportation that people seek as their incomes rise. It is almost impossible for public transport, at any cost, to match the schedule convenience, flexibility (e.g., combining shopping or personal trips with commuting), privacy and comfort of the private individual modes. Noncommuter trips, especially, are often most easily (and sometimes even most efficiently) accommodated by these individual modes. For better or worse, as incomes rise in China, urban residents will buy more and better private demand improved vehicles (mopeds, motorized bicycles, autos, etc.). For transit to compete, as Meakin points out in Theme Paper 4, ever-greater improvements in public transport services and operations will be needed. In addition, as incomes rise, the operators of the public transit vehicles will demand higher wages. Labor-intensive activities like transit tend to become relatively more expensive, and therefore less used, as an economy develops. In the worst-case scenario, these trends could undermine the underlying economics of public transport before it is even established. In at least some Chinese cities, it may simply be cheaper and easier, not to mention more convenient, to bicycle or go by moped than to ride a bus. As incomes rise in China, and consumer capital is substituted for ever more costly labor, the bicycle will be traded in for new and improved versions of two-wheel vehicles, ultimately arriving at the four-wheel variety. Consequently, China will not only face the Western world's problem of relatively poor productivity trends in public transport, but will also need to develop a public transport system that can increasingly compete with improved individual transportation on all levels. Transit, as Flora so eloquently argues in Theme Paper 8, can best meet this competitive challenge if allowed to respond to the needs of the market. To do this, the transit system must be flexible and innovative. Experience suggests that privately-owned systems do this better than the publicly-owned. In addition, the existence of some degree of competition seems to help, suggesting that exclusive franchises for an entire urban area should be avoided. As incomes rise and more people own motorbikes and automobiles, living outside the central urban area (where land is more available and cheaper) will also become increasingly common. Public transportation tends to be less frequent (or otherwise available) and more costly (per unit used) at the fringes of urban development. By contrast, the private automobile becomes a relatively more cost-efficient means of transport at low densities (though still absolutely quite expensive). When the share of total urban trips made by motorbike, moped or automobile reaches a certain threshold, say about 50 percent, a city can be said to be well along on the transition into Stage 3. At this point, the city's population will have already sunk substantial personal capital into transport equipment. And the labor costs of operating the individualized private modes can be perceived to be low given their do-it-yourself nature. Individuals provide their own driving, 30 Keynote Address 2: The Urban Transportation Problem in a Context of Economic Development management, routing and even some maintenance services, at no explicit wage, as opposed to paying others, at increasingly higher wages, to provide those functions when using transit. Initially, too, the transition to mechanized public transport (Stage 2) may not be characterized by excessive congestion. Effective road capacity may actually increase as mechanized modes that take up less road space replace unmechanized modes. (For example, it takes a good deal less square footage of street space to accommodate 100 horsepower of diesel engine than the real thing!) However, mixing modes of different speeds may impede traffic flow, as the slower modes obstruct the movement of the more rapid; similarly, the faster modes (in attempts to circumvent the slower) may create safety hazards so that traffic, particularly pedestrian traffic, is unnecessarily endangered. Of particular importance is establishing separate rights-of-way for the different modes-powered vehicles, bicycles, pedestrians, animal-drawn carts, etc. If properly done, the immediate need and demand for more investment in road infrastructure may be relatively low since streets provided for commercial and public needs may meet emerging private mode requirements reasonably well-if not in all of China's highly dense cities, at least in many. However, as time proceeds and auto ownership and use increase, serious bottlenecks (points of severe congestion) will appear, especially during the hours when people are commuting to and from work. In China, the volume of bicycles during peak periods already has resulted in congestion problems, especially at intersections. Although bicycles are small and require little road space in a static (at rest) sense, they are slower than motorized vehicles, requiring more time per vehicle to complete a trip and in particular to clear an intersection. Consequently, as more motorbikes are introduced and substituted for the pedal-powered, traffic flows may initially improve (because motorbikes may make better dynamic use of available street space than conventional bikes). This "free lunch" is not likely to last long in crowded Chinese cities. It appears that urban transport development is affected not just by a city's population density, but more specifically by the relationship between road length per capita and population density. This relationship, as developed by Stares and Liu in Theme Paper 1, may provide important insights into a city's future road infrastructure needs. For example, Chinese cities have a very low ratio of roads per capita and very high population densities compared to cities in most developed countries and many developing countries. This suggests that as the cities in China continue to grow and develop, lack of road infrastructure will be a particularly aggravating and increasingly difficult problem. Stage 3 in China could be a nightmare of congestion in many cities. The increased congestion accompanying city growth and rising incomes may only increase many of the advantages of using bicycles. A transportation planning challenge in China could then be that of attracting riders away from bicycles to buses (thus more efficiently using scarce road space) by improving transit's quality and service. Eventually, though, increased trip making in mechanized private modes will lead to demands to provide more urban transport capacity. Unfortunately, eliminating bottlenecks by providing more highways or high performance public transit alternatives can be very expensive. That's the bad news. The good news is that there are some less expensive solutions to these problems. Furthermore, to some extent, congestion problems may be self-correcting, because of changes in urban trip patterns that occur as an economy develops. As an experienced traffic engineer once observed: "People do somehow avoid infinite queues." John R. Meyer 31 While people do avoid infinite queues, they may nevertheless often find themselves in uncomfortably long traffic queues or delays as an economy develops. As Gwilliam points out in Theme Paper 10, serious urban transportation problems are usually symptoms of a larger or more basic difficulty: this is the inability of simple market mechanisms to optimize urban transportation choices. A driver, or cyclist, when considering whether or not to use a particular road, will only consider the costs that are directly perceived or charged to him (namely, all direct costs such as those for gasoline, tires, direct wear and tear on his vehicle, etc.). As long as the road's capacity is sufficient to accommodate his use without slowing or otherwise interfering with anyone else's use, no particular problems are created. Traffic, however, sometimes increases to the point where an additional user interferes with other users by creating congestion on the road. At this point (and for all traffic volumes greater), the new or additional user would underestimate the costs of his using the road because he would fail to take into account delay or congestion costs he imposes on others. In other words, since the individual driver only perceives his own congestion delays and not those he imposes on others, private costs fall below social costs. This, in turn means that too many (that is, an inefficiently large number of) travelers will use the road (because the perceived price, or cost, of usage is too low). To rectify this situation and bring private costs back into balance with social costs, a toll should be imposed for using the congested road. In essence, where there is congestion, market forces can result in an optimal level of usage only if a toll is imposed on users. However, it is usually not optimal to eliminate all congestion. Eliminating all congestion can be expensive, with costs exceeding benefits-say, by building highways almost ad infinitum. Of course, the existence of a large number of users willing to pay congestion tolls also strongly suggests that expansion of a facility may be desirable; that is, more capacity may be justified if the sum of congestion tolls exceeds the cost of expanding the facility. The proper toll will obviously vary from place to place and by time of day. The highest tolls would be needed on the most centrally located facilities at the most congested peak commuter periods. Determining optimal tolls for all relevant times and places, and then implementing them, are not easy tasks. Full implementation would require sophisticated surveillance and recording equipment capable of logging the movements of every vehicle in the urban area. Clearly, a system of this sort could be costly, even if technologically feasible. Such a high degree of control and surveillance is also thought by many to be unacceptable on political or social grounds. Fortunately, simpler pricing schemes will often suffice or at least greatly alleviate congestion problems. For example, when access to a congested area is limited to a few major expressways, tunnels or bridges (such as in Manhattan or Hong Kong's Victoria Island), congestion tolls can be simply and economically assessed by placing toll booths at these entry points. China, moreover, already has had some experience with charging fees for use of new urban access roads, originally built as toll roads. Furthermore, even if a congested area's entry points are not "naturally" limited, it is still possible to assess congestion costs by using an area licensing scheme, such as that developed in Singapore. By increasing the costs of automobile trips to congested areas, area restraint should encourage people to carpool and switch from auto to public transit or perhaps to bicycles or walking. In the process, air pollution and other environmental damages from auto use should also be attenuated. More generally, as Walsh 32 Keynote Address 2: The Urban Transportation Problem in a Context of Economic Development points out, more proactive pricing policies, if properly designed, could also help with environmental problems by inducing use of cleaner fuels, four-stroke engines, etc. Several papers at this conference, that of Stares and Liu in particular, discuss not only the role of better pricing but a wide variety of other measures to reduce congestion, and often pollution as well. For example, one of the least expensive alternatives for increasing transportation capacity, and thereby relieving congestion, is improved traffic management. For example, conflicting traffic movements can be avoided or reduced by banning certain turning movements, by introducing one-way routing, and by traffic signals at intersections and pedestrian crossings. Benefits can also be derived by restricting stops along critical sections of the road network and loading and unloading of goods vehicles at certain times of the day. These measures require good enforcement as well as extensive use of traffic signs and road marking, but the costs generally will be relatively low. Eventually, though, there may be little alternative but to provide new facilities and infrastructure, or at the very least to substantially modify those that exist. To start, substantial benefits often can be achieved by minor, low-cost improvements, such as road widening and pedestrian foot bridges. Better maintenance of existing roads also often enhances effective capacity quite inexpensively. Of course, transit systems, like roads, can also become incapable of coping with demand. A need thus also arises to examine the options available for transit improvements and augmentation. Generally, as Allport points out, bus systems involve the lowest costs, are able to meet demands from the lowest to the highest levels and are flexible enough to meet the changing needs of a growing city. While the very highest demands (say. over 50,000 passengers per corridor per hour) may only be well met by heavy rail systems, the need for such systems is likely to arise only in the largest and densest of cities. Even then, priorities for existing public transit modes (namely, buses) should be aggressively pursued in the period before rail can be afforded or implemented. The great danger, in fact, at the transition into Stage 3, is to "think too big" without sufficient prior planning and thought. The temptation may be strong to effect a "quick fix" by making a major investment, say in rail transit or expressways. If not carefully planned in relation to the urban area's underlying trends, however, both may prove counterproductive. The rail transit system, for instance, may see its market disappear as decentralization reduces densities and moped or auto ownership opens up other travel alternatives. The expressway solution, by its very institution, may facilitate and hasten decentralization so that it creates almost as many new traffic demands as it satisfies. At some point, however, heavy and costly investments in transport infrastructure probably cannot be avoided. In fact, these may be so large as to require tapping into now private sector sources, as K.H. Lee suggests in Theme Paper 9. It is thus important, even if construction can be delayed, to make allowances (rights-of-way, land banking, zoning) for the physical expansion of the existing network in both short- and long-term development plans. Some foresight and planning for future construction helps to lower costs by minimizing disruption to traffic and other urban activities, unnecessary demolition of buildings, the diversion and reworking of utility services, and costly land acquisitions. In general, the increased complexity Jolhn R. Meyer 33 of the urban transportation problem as development proceeds is almost certainly better handled with more rather than less prior analysis. To summarize, solutions to the urban transportation problem vary in cost, derived benefits and ease of implementation. The types and magnitude of transport problems will vary considerably from country to country and from city to city around the world. Fundamentally, though, the problems have much the same origins in every c:ase. The major urban transport problem in most cities is congestion, closely followed by pollutioni. In particular, the rapid and uncontrolled development of personal motorized transport is likely to lead to severe congestion to the detriment of urban form and amenities. Congestion is a problem that can be alleviated, but usually should not be totally eliminated. The proper policy goal is attenuation, and that goal is well within the realm of the possible. The most direct and economically efficient way to achieve that goal would be to place tolls on use of urban transport tacilities so as to bring privately perceived costs into closer alignment with social costs. Simple calculations of economic efficiency will rarely suffice, however, as a public policy for urban transportation. Because of the externalities and public good characteristics of urban transportation, governments are quite properly and inevitably involved. As Gakenhleimer points out in Theme Paper 12, an important role for transport planining remains. Unlfortunately, government intervention often leads to an intensificationi of the very worst proclivities and difficulties commonly encountered in providing urban transport services. For example, the desire to serve certain social or political purposes can lead to the introduction of governmenit controls, which in turn induce inefficient use of resources and undermi-ne managerial and labor incentives. Almost universally, an important policy issue is defining and establishing the respective roles of the public and private sectors. The public sector is well suited for playing a role in situations where market imperfections exist, such as assigning and determining congestion costs or establishing environmental standards. In addition, the public sector is traditionally responsible, and usually needed, for enforcement of traffic laws, establishing and implementing traffic management measures, defining licensing and safety standards, etc. The private sector, on the other hand, has shown a greater ability than the public sector for developing and operating services efficiently, apparently because of the private sector heing profit motivated and better able to react to market changes. In short, the urban transportation problem is inherently difficult and complex. There are no simple market mechanisms that can be used in lieu of good public policy and intelligent public administration. On the other hand, the absence of any simple market solution can also lead to an overreliance on and indulgence of government "solutions" that are ephemeral or even counterproductive. The only apparent "escape" is to find some middle course-to define more precisely what can and cannot, what should and should not, what must and must not be done by the public and private sectors in solving the complexities of what has come to be knowni as "the urban transportation problem." Zhlou Ganshi 35 KEYNOTE ADDRESS 3: URBAN TRANSPORT PROBLEMS IN CHINESE CITIES: CAUSES, TRENDS AND OPTIONS ZHOU GANSHI ' The rapid changes of urban transport in China are stimulated by rapid economic development. Since the early 1980s, most cities have increased their efforts in road construction and the size of investments is unprecedented. With continuing economic development, however, urban transport problems become more and more serious. To use an old Chinese saying, the boat goes up when the river rises. CHARACTERISTICS AND PROBLEMS Several characteristics are prominent in China's urban transport. They are discussed as follows. The motor vehicle fleet has grown rapidly and bicycle ownership has reached a high level. In 1983, the total number of motor vehicles in China's cities was about 2 million, two times as many as that of 1977. The total number of urban bicycle in 1988 was about 4 million. In 1994, urban motor vehicles numbered 5 million. In the last few years, the annual growth rate of motor vehicles was above 15 percent. In some cities, the rate reached 30 percent. The total number of bicycles in all cities accounts for more than 180 million. Bicycle ownership in cities and towns reaches 198 bicycles per 100 households. The total number of motor vehicles in the cities accounts for about 50 percent of the national total; and that of bicycles, 40 percent. The last decade also has seen the doubling of private motor vehicle ownership. The total number of private cars in the country was 598,000 in 1993. In 1994, private motor vehicles accounted for 17 percent of the motor vehicle fleet in Beijing. Road construction, though impressive, fails to keep up with the growth of vehicles. With an annual growth rate of 10.2 percent, the total length of urban roads increased from 29,485 km to 104,897 km from 1980 to 1993, a 5,800 km addition per year. The length of urban roads per 10,000 population reached 6.3 km, from a low 3.3 km. During the same period, the total 2 ~~~~~~~~2 road area increased to 1,075 million m2 from 253 million m . The road area per capita increased 2 2 from 2.8 m to 6.5 m , which represents an impressive 11.8 percent annual average growth. Fixed-route public transit is declining. During 1978-93, the public transit vehicle fleet and service routes in China increased 2.5 times and 2.8 times, respectively. The number of buses reached 0.6 per 1,000 population. However, the operating speed of buses reduced to 5 to 10 km/hour from 12 to 14 km/hour. Therefore, the increased capacity was largely offset by the declining efficiency. For many years, the public transit operations have had to rely upon I Zhou Ganshi is a Full Member, China Academy of Sciences, and formier Vice-Minister, Ministry of Construction. 36 Keynote Address 3: Urban Transport Problems in Chinese Cities Causes, Trends and Options government subsidies. The loss-making operations accounted for 70 percent of total public passenger transport operations in 1993; and government subsidies in Beijing and Shanghai alone amounted to Y 1.6 billion. Because of the declining operating efficiency and service quality, public transit ridership has been declining each year; bicycle ridership has increased. In Beijing, for example, the modal split between public transit and bicycle became 4:6 in the 1990s, sharply contrasting to the 6:4 ratio of the 1980s. In Tianjin, the ratio was 19:81 in the 1980s, but the share of public transit declined to less than 10 percent in the 1990s. Taxis have grown rapidly and rail transit is starting to develop. With the increase of the urban residents' living standards, people are no longer satisfied with a single public transit mode. This is illustrated by the rapid increases in taxi services. In 1978, there were only 1,628 taxis in China. In 1993, there were 286,207 taxis, of which about 60,000 taxis operated in Beijing, 18,000 in Shanghai and 12,000 in Guangzhou. By August 1995, the number of taxis in Beijing reached 80,000, and in Shanghai, 32,000. At present, only three cities in China have a metro system. They are Beijing (43.5 km), Tianjin (7.4 km) and Shanghai (14.6 km). The total route length accounts for 65.5 km. However, more and more cities are now planning, preparing and constructing metros. By the beginning of the twenty-first century, Guangzhou, Chongqing, Shenyang, Qingdao and some other cities will have their metro or light rail transit system. The total length of urban rail in China by then will reach about 200 km. Traffic management is poor and the rate of traffic accidents is high. Because of various historical reasons, Chinese cities lack modern facilities for traffic management and safety. Here we compare Beijing with Tokyo to give an example. Both cities have a traffic control center, but the number of junctions under the center's control in Beijing is only 3 percent of that in Tokyo. The use of traffic marks in Beijing is only 70 percent of that in Tokyo; pedestrian crossings, 4.8 percent; pedestrian overpasses, 3.6 percent; and underpasses, 5 percent of those in Tokyo, respectively. We should point out that Beijing has the best traffic management facilities in China. It is thus easier to understand the similar problems in other Chinese cities. In recent years, a few large cities introduced some advanced traffic signal control systems from abroad. Due to the traffic's special characteristics in China, however, their implementation has been unsatisfactory. The shortage of traffic control facilities and the low level of management skills contribute to a high rate of traffic accidents. In 1991, there were 0.52 traffic fatalities per 10,000 persons, or 31.3 fatalities per 10,000 motor vehicles. This increased to 49.4 fatalities per 10,000 motor vehicles in 1994. In the last few years, the deaths from traffic accidents in Beijing were about 500 persons per year, that is 60 fatalities per 10,000 motor vehicles, a rate higher than the national average. In contrast, the fatality rate is only 1.6 in Tokyo, 2.6 in the United States and in Australia, and 2.7 in the United Kingdom in 1985. Urban vehicle operating speeds have been declining each year. Traffic volumes in many cities have been growing more than 20 percent per year, causing increasingly serious traffic congestion. As a result, vehicle speeds decline. On many arterial roads, vehicle speeds are only 15 to 20 km/hour. In city centers of large cities, vehicle speeds decline to 10 to 15 km/hour. Zhou Ganshi 37 To summarize, the key problem of urban transport in China is the lack of a comprehensive transport development strategy, lack of coherent transport policies, and disorder in transport management. All these contribute to growing traffic jams, declining vehicle speeds, and low efficiency-a typical case of downward spiral and vicious cycle. China is now in transition to a socialist market economy from a centrally planned economy. During this critical transition, potential dangers will arise that improper decision- making may lead to chronic traffic congestion for many years to come. The first danger comes from short-sighted activities. These include overzealousness in the construction of large-scale, high-standard transport facilities, such as elaborate interchanges, viaducts, large bridges and urban ring roads. The construction of certain projects are needed to accommodate the traffic growth. But in some cases, they are the outcome of improper decision- making or absence of systematic planning and overall consideration. Some cities tend to prefer projects that are larger in size and higher in standard over those that are actually needed. Such projects may help alleviate transport problems in the short run. However, they could not solve the congestion problem. Furthermore, they even generate some new problems. It should be pointed out that urban transport is an integrated and dynamic system. Any changes at one point will affect the whole system. It is impossible to solve all transport problems with a single large project. Without a good understanding of the nature of transport problems, it is certainly impossible to tackle these problems in an efficient and effective manner. The second danger arises from the fact that little attention is paid to public transport. What mode should be relied upon in urban transport? Should it be private transport or public transport? These questions have been debated for many years. In fact, public transit is the most efficient transport means. It has a large carrying capacity, and occupies relatively smaller road space. Many countries in the world have chosen the policy of giving public transit priority after a long, painful experience. Even the United States is not an exception. Because of the shortage of land and high density of population in China's cities, the Chinese government made the policy giving priority to the public transit development many years ago. However, because the economic benefits of public transit are largely reflected as social benefits and the modern mass rail transport is characterized by large amounts of investment and long investment cycle, the modem mass rail transport construction has been largely ignored by officials who are interested only in short-term economic benefits. The phenomenon of emphasizing road construction at the expense of slowing down rail transit development will adversely affect urban transport development in the early years of the next century. PROSPECTS AND OBJECTIVES Prospects Forecast of urban person-trips. It is predicted that rapid economic development in China will sustain for at least another 15 years, and urbanization will also continue. It is estimated that by the year 2010, total urban population in China will reach 360 million. If each urban resident makes 2.7 trips per day, total annual person trips in all cities will be about 250 billion. If public transit is developed as the major mode of urban transport, buses and trolleybuses will takes 20 to 30 percent of all person-trips; rail transit, 10 to 15 percent; company vehicles and private cars, 10 to 15 percent; and bicycles and walking will take the rest. 38 Keynote Address 3: Urban Transport Probtems in Chinese Cities: Causes, Trends and Options Forecast of urban public transport vehicles. Historical trends suggest that the number of public transport vehicles will grow by 7 percent per year. By 2010, the total number of vehicles will be 281,000, which is 192,000 more than that of 1993. That is an annual increase of 10,000 vehicles. By 2010, there will be 7.7 vehicles per 10,000 persons in China's cities, which means one vehicle will serve 1,300 persons. In recent years, the number of taxis has been increasing very fast. In some megacities, the number of taxis is near the saturation level. According to 1993 statistics, each taxi served 578 passengers on average. Using this figure, we can estimate that there will be 620,000 taxis by 2010, which is 340,000 taxis more than that of 1993 and an increase of 20,000 vehicles per year. Road facilities. Because of urban land constraints, the growth of length and area of urban roads will not be higher than that of the previous years. According to "The Public Utilities Development Projection for the Eighth Five-Year Plan and the Ten-Year Plan," a report published by the Ministry of Construction, by the year 2000 the urban road length will be 2 between 174,418 and 197,674 km and total area of urban roads will be 16.4 to 18.5 billion m . If these can be realized and the trends continue, by the year 2010, the length of urban roads will be 2 388,120 km and the area of urban roads will be 3.4 billion m . This means a total increase of M2 in2- 2,320 km in land use for urban roads, or 136 km increase per year. The number of motor vehicles. According to "A Study of China's Household Car Development Strategy," the objective of automobile industry development is to realize two key strategic changes. The first is the change of production from truck dominance to car dominance by the year 2000 or a bit later. The second is the change of production focus from business car to household car. This will be realized by 2005 or later. Household car production will reach 0.4, 1.2, and 2.3 million by the years 2000, 2005, and 2010, respectively; private car ownership will reach 1.2, 4.6 and 13.2 millioni, respectively. In 1993, private car ownership in China was 50,000. According to the automobile industry development objectives, by the year 2010. private car ownership in China will be 13.2 million. This is a tremendous change. By that time, private car ownership in China will be 15.8 per 1,000 persons. Cities are the center of the national economy. The use of cars naturally concentrates in the city, especially the large cities. The rate of car ownership in large cities therefore will be higher than that of the national average. It is estimated to be more than 50 vehicle per 1,000 persons by 2010. According to this calculation, by 2010, Chinese large cities will have an additional 15 million cars. It is then necessary to build another 300 km2 area of road and parking spaces. As the land has been almost used up in large cities, this is no doubt a serious problem. How cities and urban transport meet automobile development will then become the focus of the problem. Objectives The overall objective should be to establish a modern multilevel and comprehensive transport system in large cities. The implementation of this objective can be divided into two stages. The first stage will be completed by the end of this century. It includes the basic establishment of a road network suitable for economic and urban development, the strengthening and revitalization of public transport, appropriate development of paratransit, and the alleviation Zhou Ganshi 39 of current serious transport problems in the large cities. The second stage will be from the 2001 to 2010. During this period, large cities will fundamentally improve the quality of their urban transport network, speed up the construction of rail transport, set up the urban transport control and guiding system, and reach equilibrium between the total supply and demand of transport. The realization of this objective will be the turning point in China's urban transport history. By that time, the situation of the transport will be as follows: * There will be a prototype transport structure for large cities, which will meet Chinese characteristics. Public transport will take up about 20 to 30 percent of urban total person-trips. Others will be trips of bicycles and pedestrians. Obviously, bicycle transport still accounts for quite a proportion. * The multilevel transport network system will be implemented. The system will include a rapid road system for motor vehicles and a slow transport system for bicycles. The separation of motor vehicle transport and nonmotor vehicle transport will be implemented in most of the areas. - A mass rail transit system will be actively developed in some large cities. - The proportion of high-standard roads will be increased, to gradually allow the growth of household private car ownership. After this period, with the rapid development of urban modernization, transport in China's large cities will enter a modem transport era of high efficiency and speed by the year 2030. MEASURES AND SUGGESTIONS It is necessary to follow the principle of "comprehensive and coordinated development" to tackle the urban transport problem, and to sustain the urban transport development trend along with economic development. With urban development and the introduction of a land leasing system, urban transport development now faces more constraints. Modern urban transport in large cities involves nearly all the departments within a city. Only with all parties' attention on urban transport, dealing with the problems comprehensively, combining resources, and supporting each other, are we able to build a modern transport system. Developing Overall Guidelines for Urban Transport And Establishing a Transport Commission in Large Cities The key issue is to strengthen the leadership of municipal governments in large cities, and to set up a high-level steering body-the Urban Transport Commission. The commission will formulate a unified urban transport development strategy; manage and guide the urban transport construction; and plan the fund-raising and the system reform in a unified way. It will also coordinate the existing urban transport management organizations and other related departments, such as municipal finance, planning, pricing, land and taxation to ensure the implementation of key transport policies. 40 Keynote Address 3: Urban Transport Problems in Chinese Cities: Causes, Trends and Options Increasing the Density of the Road Network and Improving the Quality of Decision Making in Transport Construction Because of the large amount of "unpaid debts" in urban transport infrastructure, the current state of urban transport is very poor and unable to meet the demands of rapid economic development. It is then necessary to increase the density of the road network and to maintain a high volume of roads in the coming years. The development of the infrastructure should avoid blindness. Under the guidance of careful planning, we should speed up the construction of urban arterial and distributor roads and high-speed roads. The layouts of interchanges, pedestrian crossings, parking and bicycle lanes should be arranged in a rational way. In the redevelopment of the old cities, the construction of large interchanges occupying large land areas should be kept to a minimum. Channeling road junction traffic should be emphasized. The "bottleneck" sections of roads and the road structures should be renovated and improved. It is necessary to promote construction in a systematic way, to emphasize land saving, and to avoid the blind pursuit of high standards. By doing so, we will be able to save on the total transport cost and increase the economic efficiency of transport construction. Adjusting Transit Fares and Reforming the Public Transport System A low bus fare policy, in fact lower than the cost, has long been implemented in the large cities. The impact of bus fare adjustment on social stability has been considered as the fundamental principle guiding the bus fare adjustment. Because of the wide social implications, most municipal governments are not willing to adjust the bus fares as the adjustment will likely influence the residents' perception of municipal government performance. Due to a lack of economic resources, the enterprises are unable to develop, and the passengers could not have satisfactory services from public transport. They thus pay much higher fares on other transport means for better services. This situation turns out to be contrary to the intended purpose of the low fare policy. Therefore, to solve the existing urban transport problems, we first have to adjust bus fares and reform the subsidy policy. The objective is to provide satisfactory services while recovering the costs and even realizing a small profit margin. However, we should notice that public transport enterprises do not have profit-making as their single purpose. Not only do they realize economic benefits through provision of passenger transport, but also their value is reflected in other social economic activities. It is thus necessary to have government support for transport development, including favorable tax and fuel supply policies, as well as necessary financial subsidies. In the near future, it is imperative to adjust the irrational relative prices, to include operating revenues in the domain of price management, and to improve the subsidy system for policy-induced losses. The strict separation of policy-induced losses and losses due to mismanagement should be emphasized. The existing way of giving subsidies should be changed and subsidy quotas should be introduced based on realistic norms and indices. For instance, to quantify the performance of transport enterprises, the annual ridership, number of employees, and passenger vehicles should be used for evaluation. Fulfillment of production targets will be rewarded and enterprises with unsatisfactory performance in production will be punished. The deadline for improvement should be specified. Only by doing so can the enterprises be forced into independent operation. Zhou Ganshi 41 Adopting Traffic Restriction Measures to Guide the Development of Private Car Ownership The development of private car ownership will mainly be managed through total transport demand control. Municipal governments can adopt direct or indirect control measures depending upon the road capacity of the city. It is important to coordinate well the development of private cars and road capacity. First, overall transport monitoring and analysis should be effectively implemented. Different adjustment and control measures should be applied to different areas; especially, car traffic should be restrained in the central city areas. At the same time, it is necessary to further reform government finance and taxation policies, to design and implement an appropriate tax system, to collect fees from automobiles for road cost recovery, and therefore enable the economic means to play the macro adjustment role. Motorcycles belong to the category of motor vehicle management. Since motorcycles bring more serious environmental pollution and transport accidents than automobiles, large cities must have a strict restraint policy for motorcycles. Various Channels of Investment China needs a considerable amount of investment for its urban transport construction. Besides the central and municipal government grants, it is very important to explore various financial channels to raise funds under favorable policies from governments. This practice has proved to be a feasible and effective way to raise funds for urban transport construction. At the macro level, we should first increase the ratio of investment in urban transport infrastructure over the GNP. The United Nations' Division for Social Policy and Development Social Development estimates that total investment in urban infrastructure should take 3 to 5 percent of GNP for developing countries. Based on this estimate and China's investment in the last few years, China's investment in urban road transport should take at least I to 2 percent of GNP each year. However, even the highest investment ratio (in 1993) was only 0.6 percent. We recommend that China's annual investment in urban transport construction be no less than I percent of GNP in the future. Second, the Government should pay more attention to the reform of urban investment system. For major urban transport projects, the Government should formulate appropriate strategies and plans to raise funds, and give favorable policies to introduce foreign investments. In many countries, a certain percentage (I to 3 percent) of a fuel surcharge is imposed and the tax revenue is used for road construction. This is a reasonable system to raise revenue for road construction. It has been implemented in Hainan Province since the beginning of 1995. Up to now, the Province has favorable opinions about this practice. We recommend this system be established in the whole country. In addition, many cities have implemented the development policy described as "road construction stimulates real estate development, and real estate development in turn supports road maintenance." The revenues from urban land transactions and real estate development are used for infrastructure construction. Urban development and road transport could progress at the same time. Although these practices are still at a preliminary stage, China has already accumulated some experience, and is capable of carrying out the gradual reform of the investment system through further studies, experiments, and careful operations. 42 Keynote Address 3: Urban Transport Problems in Chinese Cities: Causes, Trends and Options Third, for urban infrastructure, we should allow large cities to formulate their own fee collection policies fitting their city specifics. The fees will supplement local government revenues and be used for road transport development. The central government and the related departments should improve the fee collection mechanism through fiscal and taxation instruments, which should be made into clearly defined national policies. Strengthening Research in Science and Technology, and Improving Public Awareness of Transport To know the directions of urban transport science and key technology development, we must strengthen research in urban transport science and educate urban residents. To attain this goal, we must successfully carry out the following six tasks: * Emphasize research on economic policies for and frontier technology development in urban transport, and achieve economic efficiency in these researches to better serve urban transport. The newly published research proposal, "Modern Urban Road Transport: High-System-Efficiency, High-Speed and Congestion-Free Technologies," is the result of many experts' collective efforts under the auspices of the Ministry of Construction. It points out the direction for China's urban transport research in the twenty-first century. We suggest that this research proposal be incorporated into the national "Ninth Five-Year" key science research project plan. * The Government should formulate related technical standards, regulations and laws; and strengthen the education of professionals and skilled workers, and the on-job training. The system of professional qualification certification in transport engineering and transport planning should be introduced to improve the quality of urban transport planning, design, construction and management. * A national urban transport engineering and technology study center should be established under the guidance of the State Commission of Sciences and Technology and the Ministry of Construction. The Center will organize the fundamental theory research on urban transport, and promote the applied technical research and application. * The Ministry of Construction should sponsor a National Urban Transport Expert Committee to guide the healthy development of the national transport science and technology, and appraise and evaluate key urban transport construction projects. * The long-term transport plan and the short-term transport comprehensive improvement plan in large cities should be incorporated into city plans. When arranging urban transport construction projects, the planning agencies should make sure that those projects meet the purpose and requirements of the urban transport plans. * Urban transport education should be incorporated into primary compulsory education. We should also strengthen transport safety education for workers of all sectors, enhance transport awareness of the population, and manage well urban transport with the cooperation of all urban residents. Stephen Stares and Liu Zhi 43 THEME PAPER 1: MOTORIZATION IN CHINESE CITIES: ISSUES AND ACTIONS STEPHEN STARES AND Liu ZHI' A. INTRODUCTION This paper looks at the issues connected with motorization in urban China. Section B starts with the numbers, looking at the current stage and pace of motorization, and going on to assess the likely scale of motorization in future years. Section C considers the likely consequences of motorization in urban areas, with particular emphasis on the issue of traffic congestion. Section D discusses ways of combating traffic congestion by better management and use of existing road space, and by the construction of new roads, but finally concluding that these approaches provide only a partial answer to congestion. Section E then reviews alternative approaches to combating congestion through the management of transport demand. Section F reviews and compares experience of motorization in several Southeast Asian localities. Finally, Section G summarizes the conclusions of the paper for urban transport strategy in China. B. TRENDS AND PROJECTIONS Recent Trends Driven by rapid economic growth, China's motorization has been gathering momentum during recent years (Figure 1). The growth of the country's entire motor vehicle fleet (excluding motorcycles) averaged an unprecedented 15 percent per year between 1984 and 1994*2 As in other countries at a similar stage of economic development, the passenger vehicle fleet in China has grown much more rapidly (at 20 percent per year) than the goods vehicle fleet; this led to a decline in the share of goods vehicles from 77 percent in 1984 to 62 percent in 1994. Rapid motorization in China also bears the mark of the country's transition from a centrally planned economy to a more market-oriented one. During 1984-94, the number of privately owned motor vehicles increased by 28 percent per year, and the number of privately owned passenger vehicles increased by 64 percent per year. Private car data are not available, but available statistics for privately owned passenger vehicles3 indicate a decline in the average I Stephen Stares (deceased) was a Senior Transport Specialist, Environment and Municipal Development Operations Division, China and Mongolia Department, The World Bank. Liu Zhi is a Transport Economist in the Transport Division, Transportation, Water and Urban Development Department. Helpful comments provided by Ralph Gakenheimer, Gregory Ingram, and John Meyer are gratefully acknowledged. 2 Motorcycle ownership data are not available at the national level. For convenience, we exclude motorcycles when using the term "motor vehicles" hereafter. 3 Passenger vehicles in China are classified as small, medium and large, with approximately I to 8, 9 to 20, and over 20 seats, respectively. 44 Theme Paper 1. Motorization in Chinese Cities: Issues and Actions passenger seats per vehicle, from 22.2 seats in 1985 to 12.9 seats in 1994. This suggests that while most of these passenger vehicles are still for commercial use, the share of private cars in the fleet is rapidly increasing. FIGURE 1: MOTOR VEHICLE GROWTH IN CHINA 12.00 10.00 / 0~~~~~~~~~~~~~~~~~~~~~ 8.00 Passenger * 6.00 - . Goods IE 4 00 j. _- , ,' . All MVs 4 4.00 2.00 t- o° W c r- (7 t-o X o oo o 0~~~~~~10 Year Despite impressive growth in recent years, China's per capita motor vehicle ownership remains among the lowest in the world at eight vehicles per 1,000 population, of which only one is a passenger car. China's motor vehicle ownership is not only much lower than richer e.ountries, but also lower than countries of comparable incomes. All this suggests a very low base and therefore a huge potential for continuing rapid motorization in the years to come. The growth of motorization in some of the more developed Cbinese cities during the last several years is even more impressive. Table I shows that Beijing, Guangzhou, Chengdu, and Shantou have seen more rapid growth in motor vehicle ownership than the country as a whole. Even in the most densely populated city of Shanghai, where physical conditions for motor vehicle use are clearly unfavorable, annual growth has still reached 10 percent. Moreover, these cities have substantially bigher levels of per capita ownership than the national average. Another aspect of the growing motorization in urban China is the dramatic growth of motorcycle ownership. This can be illustrated by the cases of Guangzhou and Shantou, shown in Figure 2. Both cities experienced a sustained 35 to 40 percent annual growth in the number of motorcycles during the 13-year period 1980-93, a much higher rate than for passenger vehicles and goods vehicles. In response to this growth and following similar measures elsewhere, Guangzhou capped the growth rate in 1992 by imposing an annual quota on new motorcycle licenses. By 1993, motorcycle ownership per 1,000 population reached 61 units in Guangzhou and 94 units in Shantou. Stephen Stares and Liu Zhi 45 TABLE 1: GROWTH OF MOTOR VEHICLE OWNERSHIP, SELECTED CITIES Motor Vehicles (excl. Motorcycles) Population Motor Vehicles Per City and Number ('000) Annual % (millions) 1,000 Population Statistical Area 1990 1993 Change 1990 1993 1990 1993 Beijing, city proper 258 402 16 6.99 8.41 37.0 47.8 Shanghai, city proper 172 230 10 7.83 9.48 21.9 24.3 Guangzhou, city proper 94 152 17 3.58 3.70 26.2 41.0 Chengdu, municipality /a 54 124 23 9.19 9.47 5.9 13.1 Shantou, municipality 12 26 30 0.86 0.92 13.8 28.3 China 5,836 8,776 15 1,134 1,185 5.1 7.4 /a The data for Chengdu are 1990 and 1994 data. Source: (1) Population data are from China Statistical Yearbooks; and (2) vehicle data are provided by local governments. FIGURE 2: GROWTH OF MOTOR VEHICLE OWNERSHIP IN GUANGZHOU AND SHANTOU Guangzhou Shantou 25 .. . .... . .. 10 . .. . .. E. - -e l~ l 200 8 .~150 6 100 4 Motorization .0 00 at t N Lev0l Year Year .....Car & Van - - -- Other MV Motorcycle .. Cr&Va F, e-rOte MV Motorcy-cle Motorization Forecasts at the National Level There are many uncertainties associated with projections of motorization levels for China, given its low starting levels, its current high rate of growth, and the recently announced Government policy for promoting the auto industry and its domestic market. Even so, it is worthwhile to project future motorization levels, to give a sense of the possible magnitude of future growth. International comparisons of motor vehicle growth in many countries indicate that while the levels of motorization are affected by various economic, cultural, and geographical factors, per capita income growth remains the single most important driving force for the growth of motor vehicle ownership. This causal relationship, which is clearly demonstrated in Figure 3, 46 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions has been extensively studied by economists employing cross-section or time-series regression methods to explain the variations in motor vehicle ownership between countries or cities, or within a single country or city over a period of time. Thus, a large body of consistent empirical evidence has accumulated showing a high statistical correlation between motor vehicle ownership and income growth. FIGURE 3: PER CAPITA INCOME AND MOTOR VEHICLE OWNERSHIP IN 52 COUNTRIES OR REGIONS 800.0 _ _ _ _ _ _ _ _ _ _ _ _ 700.0 C 600.0 c 500.0 0: 400.0 Z 300.0 .5 * c 200.0 100.0 0 5,000 10,000 15,000 20,000 25,000 Per capita GDP in 1990 lISS Studies of the motor vehicle growth typically estimate the income elasticity of motor vehicle ownership, which measures the percentage change in ownership for a 1 percent change in income. Table 2 lists the income elasticity estimates reported in several representative cross- section studies.4 These estimates range between 1.02 and 1.95, implying that a I percent increase in incomes would lead to 1.02 to 1.95 percent increase in vehicle ownership. It should be noted that these estimates were obtained at different times from different cross-country samples, which include both developed and developing countries, and both market and command economies. The narrow range of these estimates demonstrates that in terms of motorization these countries, rich or poor, are only at different points of the spectrum. 4 Aubrey Silberston (1970), "Automobile Use and the Standard of Living in East and West," Journal of Transport Economics and Policy, Vol. 4, No. I; William C. Wheaton (1982), "The Long-Run Structure of Transportation and Gasoline Demand," The Bell Journal of Economics, Vol. 13; John F. Kain (1983), "Impacts of Higher Petroleum Prices on Transportation Patterns and Urban Development," in Theodore E. Keeler (ed.), Research in Transportation Economics, Vol. 1, JAI Press Inc.; and John F, Kain and Liu Zhi (1994), "Efficiency and Locational Consequences of Government Transport Policies and Spending in Chile," Harvard Project on Urbanization in Chile, Harvard Institute for International Development. Stephen Stares and Liu Zhi 47 Data in Table 2 also indicate that income elasticities of passenger cars are consistently greater than the income elasticities of commercial vehicles and total motor vehicles. This suggests that passenger cars will grow faster than other types of motor vehicles as per capita incomes grow, and recent experience in China supports this. Because passenger car ownership and use in developing countries tend to concentrate in their more developed urban areas,5 these estimates imply that with the growth of income, urban road systems will bear more rapidly growing demands than intercity highway systems. TABLE 2: INCOME ELASTICITIES OF MOTOR VEHICLE OWNERSHIP Vehicle Study Sample Ownership Silberston (1970) 38 free-market countries, 1965, automobiles 1.14 38 free-market countries, 1965, total vehicles 1.09 46 countries including USSR and East European 1.21 countries, 1965, automobiles Wheaton (1980) 25 countries, early 1970s, automobile fleet 1.38 25 countries, early 1970s, total vehicle fleet 1.19 42 countries, early 1970s, automobile fleet 1.43 Kain (1983) 23 OECD countries, 1958 1.95 23 OECD countries, 1968 1.59 98 non-Communist countries, 1977 1.30 Kain and Liu (1994) 52 countries, 1990, passenger cars 1.58 52 countries, 1990, commercial vehicles 1.15 52 countries, 1990, total motor vehicles 1.44 60 world cities, 1980, passenger cars 1.02 It must be noted that other social and geographical factors are also important determinants of motor vehicle ownership, and their effects could be substantial in the short run. However, partly because these factors are difficult to quantify, their quantitative relationships with vehicle ownership are not well-established empirically. Since income is the predominant determinant of long-term vehicle ownership growth, we rely on income elasticity estimates to project the likely growth of vehicle fleet for the next 25 years.6 It also should be noted that the use of per capita incomes to project car ownership may mask the effect of income distribution. More precisely, car ownership is highly correlated with the upper end of income distribution; this is particularly true for the early stage of motorization. However, we are not able to pursue this further due to the lack of empirical data and to the high degree of uncertainty that exists in the projection of future changes in income distribution. 5 For more discussion on the concentration of motor vehicle ownership and use in urban areas in the developing world, see Wilfred Owen (1978), "Automobiles and Cities: Strategies for Developing Countries," in Ralph Gakenheimer (ed.), The Automobile and the Environment: An International Perspective, The MIT Press. 6 Methodologically, it is proper to use the income elasticity estimates obtained from cross-section data, because these estimates capture the long-run effect. 48 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Forecasts of car, truck, and total motor vehicle ownership for China as a whole for the period 1995-2020 were prepared using income elasticities (1.58 for cars and 1.15 for trucks) estimated by Kain and Liu (1994), the most recent cross-country study that includes China in the sample.7 In projecting the future levels of motorization, we assume that China's future per capita gross domestic product (GDP) growth rate declines gradually from the current rate of 10 percent per year to a sustainable 8.0 percent per year by 2000.8 Considering that the current high growth rate of motor vehicle ownership results partly from rapidly growing incomes, and partly from "catching up" on the enormous pent-up demands caused by a tightly controlled supply of motor vehicles in the past,9 we further assume two scenarios for the changes in vehicle ownership growth rate in the next five to ten years. In the lower-growth scenario, the current high rate of vehicle ownership growth is assumed to decline gradually to the level determined by income elasticity (that is, equal to income elasticity times the assumed income growth rate) by 2000; and in the higher-growth scenario, we assume current high rate of vehicle ownership growth continues through 1996-2000, then gradually declines to the level determined by income elasticity by 2005. The projections are summarized for several key years in Table 3. TABLE 3: CHINA: PROJECTIONS OF VEHICLE GROWTH Lower Growth Higher Growth Year Car Truck Total/a Car Truck Total/a Vehicles per 1,000 population 1995 2 6 9 2 6 9 2000 5 10 16 6 10 19 2010 16 24 45 25 26 57 2020 53 57 124 83 62 162 Percent growth per year 1995-2000 16 10 10 23 11 15 2000-2010 12 9 11 15 10 12 2010-2020 13 9 9 13 9 11 /a Total motor vehicles, excluding motorcycles. 7 Kain and Liu (1994) used cross-country data to estimate regression models for passenger car, commercial vehicle, and total motor vehicle ownership. The sample included 52 countries, ranging from the very poor (such as Tanzania and Malawi) to the very rich (United States and Japan), and from some former centrally planned economies (China, Hungary and Poland) to countries that once implemented or continue to implement policies to control private motor vehicle ownership (Korea and Singapore). 8 A country study by the World Bank (1993), China: Managing Rapid Growth and Transition, predicted that the current very high growth rate would fall to a sustainable rate of about 8.9 percent if the Government moves swiftly and actively in its reform, or to a rate of 7.1 percent if the Government continues its reform in stop-go cycles. 9 Kain and Liu (1994) provided empirical evidence of pent-up demands for motor vehicles in China. The analysis estimated a vehicle ownership regression model, using international observations including China, to predict vehicle ownership. It was found that China's actual motor vehicle ownership was substantially lower than the predicted level. The finding suggested China's motorization level was lower than the average level for the countries of comparable incomes. Stephen Stares and Liu Zhi 49 Motorization Projections at the City Level Projecting motorization at the city level is more difficult, simply because the physical conditions for car use vary widely city by city and potential government controls on ownership and usage at the local level are very hard to foresee. Again, we turn to international experience. There are very few cross-world city studies available. A recent study by Kain and Liu (1994) provided some useful estimates that would help our projections of motorization at the city level. This analysis used data from 60 world cities to estimate cross-section regressions of passenger car ownership, and found that about 80 percent of the variance in passenger car ownership at the city level could be explained by two variables: per capita GDP and city population density. The role of per capita GDP in explaining levels of urban motorization is similar to the cross-country case. City population density is used as a proxy to reflect the impact of urban physical conditions on car ownership. The study estimated elasticities of 1.02 and -0.21 for these two variables, implying that a I percent increase in per capita incomes would lead to 1.02 percent increase in car ownership, and a 1 percent decrease in city population density would be accompanied by a 0.21 percent increase in car ownership.'" We used the elasticities for income (1.02) and density (-0.21) to forecast the future level of car ownership for a hypothetical large Chinese city prototype." We focused on passenger cars only, for cars are going to be the major problem at the city level. For a 1995 base year, the city prototype was assumed to have 3 million population, with a gross population density of 20,000 persons per square kilometer-km (a level that is comparable to the actual levels found in Guangzhou and Chengdu), and a passenger car ownership level of 10 per 1,000 population. A range of assumptions were made on future income growth, car ownership growth, and population density decline, as follows: * Income Growth. Two possible scenarios for future per capita income growth rate, both plausibly higher than those assumed for the whole country in the earlier projections: 0 a gradual decline from the current 15 percent per year to a sustainable 9 percent per year by 2000; and 0 maintaining the current growth rate through 1996-2000, and then gradually declining to a sustainable 9 percent per year by 2005. * Vehicle Ownership Growth. Corresponding to the two income growth scenarios above, two alternative scenarios for future car ownership growth were assumed: 10 The city car ownership regression used 1980 data from 60 large world cities (in 40 countries). Because income data at the city level were not available, the data of per capita GDP at the country level were used as a proxy. Given the greater measurement errors for the income variable used, the income elasticity estimate obtained for the regression is considerably smaller than those obtained in the cross-country regressions. ' There is a methodological problem here that per capita incomes and, to a lesser extent, car ownership, are also determinants of city population density. However, we consider this a miner problem on the assumption that city population densities in China will be largely determined by land use policy. 5Q Theme Paper 1: Motorization in Chinese Cities: Issues and Actions 0 a steady decline from the current 20 percent per year to the level determined by income elasticity by 2000; and 0 continuation of the current rate through 1996-2000, and then declining steadily to the level determined by income elasticity by 2005. Population Density Decline. Most large Chinese cities currently have extraordinarily high population densities, but many are now undergoing a dedensification process. Therefore, we anticipated a decline of population density in the future, and assumed two possible scenarios for the planned population density for the year 2020 (the long-range urban master planning target year for many Chinese cities): 0 higher density: 15,000 persons/kM2 (comparable to Seoul in 1990, and the target density in Shanghai's long-range urban master plan); and 0 lower density: 10,000 persons/km2 (comparable to Amsterdam in 1980). Projections of car ownership based on four sets of combined assumptions for income growth and population density decline, are summarized for several key years in Table 4. TABLE 4: CHINA: PROJECTIONS OF CAR OWNERSHIP Higher Density Lower Density Lower Higher Lower Higher Income Income Income Income Year Growth Growth Growth Growth Cars per 1,000 population 1995 10 10 10 10 2000 21 25 21 25 2010 51 83 54 86 2020 127 204 138 222 Percent growth per year 1995-2000 15.9 20.3 16.1 20.5 2000-2010 9.4 12.6 9.7 12.9 2010-2020 9.5 9.5 9.9 9.9 1995-2020 10.7 12.8 11.1 13.2 The projections show that car ownership growth in urban areas could be sustained at high rates for many years if incomes continue to grow. In addition, urban dedensification will contribute, to a lesser extent, to the increases in car ownership. The projected levels under higher income growth scenario for 2010 are similar to current Singapore levels, and for 2020, about 30 percent higher than current Seoul levels. Hence over the next 25 years with continuing growth and expansion, Chinese cities could see the car fleet increase by some 13 to 22 times, or an average growth of 11 to 13 percent Stephen Stares and Liu Zhi 51 per year. The estimates, moreover, can be regarded as conservative because we use per capita incomes, instead of upper end of income distribution, as the major determinant. As will be seen later, such rapid growth has been sustained in other SE Asian countries over past years, and is clearly attainable by China. The Unknown: Motorcycles The trends of motorcycle ownership in Guangzhou and Shantou shown earlier clearly indicate that motorcycles are likely to be the fastest-growing mode of urban transport in the near future, unless controls on ownership are imposed. It is not so clear, however, where the trends are heading to in the longer run. Experiences outside mainland China, as illustrated by the following cases of Italy and Taiwan (China), suggest two possible trajectories after a period of fast growth: stabilization at a moderate level or continuing rapid growth to near saturation.'2 The historical trends of motorcycle ownership in Italy and Taiwan are shown in Figure 4. In Italy, motorcycle ownership remained higher than car ownership from 1950 until around 1963, the time when the subcompact economy cars (Fiat) bccame available at affordable prices. After this, car ownership continued to grow while motorcycle ownership remained at a quite stable level. By 1990, Italy had 475 cars and 96 motorcycles per 1,000 population. Taiwan (China) followed a different path. In 1972, there were 63 motorcycles (comparable to the current Guangzhou level) and 5 cars per 1,000 population; but in 1992 motorcycle ownership reached 485, still substantially higher than car ownership (146). It took merely two decades for Taiwan's motorcycle ownership to grow from a moderate level to near saturation. Why did Italy switch to car while Taiwan continued with motorcycle? Several factors probably intervened. In the case of Italy, the motorcycle image probably deterred older candidates from this form of motorization, public transport was probably better, and cheap motorcycles (Vespas and Lambrettas) arrived not long before the affordable car and therefore had less time to get established. Very importantly, incomes in Italy were quite high in the mid- 1960s (Figure 5) when the affordable car arrived, and urban densities were lower, so the switch to the car took place quite readily. In Taiwan, cheap motorcycles were available from the start of the period of rapid economic growth in the mid-1960s, and incomes did not reach the Italian mid-1960s level until 25 years later. Hence. motorcycles had a long period of dominance before the small car became widely affordable. There could be other reasons too, such as climate and cultural differences. But no matter how complicated the reasons, there is one common message from both cases: motorcycle ownership can grow rapidly if there is physically little room for passenger car use or no policy to control motorcycle ownership. Mainland Chinese cities are typically densely developed. By all indications, motorcycle ownership growth in these cities would tend to follow the Taiwan (China) pattern instead of Italy's. Given continuing rapid growth in per capita incomes, motorcycle ownership in some cities could reach saturation level in a fairly short period of time. For example, if Guangzhou 12 The saturation level for motorcycle ownership is detined as 500 motorcycles per 1.000 population, which is based on Taiwan's current motorcycle ownership. the highest found in the world. Actually, whether this level can be said as saturation remains a question, as motorcycle ownership in Taikan still grows (though at about 5 percent per year in recent years, a much lower rate than earlier years). 52 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions relaxed its cap on new motorcycle licenses, it would take only 20 years for motorcycle ownership to reach near saturation at a conservative growth rate of 10 percent per year (just one quarter of the actual growth rate before the cap was imposed). For Shantou, it would take just 16 years. FIGURE 4: TRENDS OF CAR AND MOTORCYCLE OWNERSHIP IN ITALY AND TAIWAN (CHINA) 500 450 / 400 _ 350 . 300 _ Italy cars o. 25/... Italy motorcycles 84> 200° / __ T aivan cars 200 / SO / / < _ " ' , t _ _ _ _Taiwaln motocycles 150 /_ ioo 50 0 It 00 > V 0 C 0 T 0 O < ' t 0 '0 t- r- X 00 0a O N ON ON 0N O N ON O N O Year FIGURE 5: COMPARISON OF PER CAPITA GDP: ITALY VS. TAIWAN (CHINA) 18000 16000 w 14000 f 12000 e 10000 Italy 8000 .,.,. Taiwan 6000 4000 ^ 2000 o I 0if I? I 0I Ii Ii III on Vr W '0 \0 '0 tor- os 00 00 00 00 o0 Year Stephen Stares and Liu Zhi 53 Goods Vehicle Ownership in Cities Currently, goods vehicle, occupies a large share of the total motor vehicle fleet in Chinese cities. In Beijing, for example, the number of goods vehicles accounts for 59 percent of total motor vehicle fleet; in Guangzhou, 64 percent; and in Chengdu, 54 percent (in all cases excluding motorcycles). According to recent motor vehicle surveys conducted in Jinan, Fushun, Nanning, Fuzhou, and Shantou, total vehicle-kilometers traveled by goods vehicles account for roughly half of the total vehicle-kilometers by all motor vehicles. While absolute numbers of goods vehicles will increase, their proportion in the vehicle fleet will decline, perhaps by 2020, reversing the current approximately one-third/two-thirds split in favor of goods vehicles. As discussed earlier, the passenger vehicle fleet is growing much faster than the goods vehicle fleet in urban China. This is consistent with international experience that passenger vehicle demand is more elastic to income growth than goods vehicle demand. In general, the goods vehicle fleet grows proportionally to growth in total GDP, while passenger vehicle ownership grows more than proportionally to per capita GDP growth. Although goods vehicles will continue to be a major component of total motor vehicle traffic for many years to come, it is the rapid and uncontrolled growth of car ownership in cities that will cause much of the urban transport problem in the future, and this is the main focus of this paper. China's Motor Vehicle Inidustry Development Policy The central government of China has recently promulgated an auto industry development policy"3 that aims at promoting the motor vehicle industry to lead the development of the national industrial sector. The central focus of the policy is on the formation of a domestic market, particularly a market for household cars, to ensure economies of scale for the domestic industry. The policy sets a domestic car production capacity target of 1.2 million units per year by 2000 and 3.5 million units per year by 2010, with 90 percent of the products sold domestically. The policy aims to encourages private car ownership, and calls for the elimination of government controls on vehicle purchase, for car prices to be determined by the market, and for taxes on cars to be reduced. The foundation for this policy is an official study of China's future motor vehicle industry prepared by the Institute of Techno-Economics of the State Planning Commission (SPC- ITE). The study indicated that future car ownership in China, particularly household car ownership, depends on three key factors: (a) household affordability; (b) levels of need for car use; and (c) physical conditions for car use.'4 Household affordability is measured as the ratio of standard car prices over per capita gross national product (GNP). Based on an evaluation of international experience, the study estimated that household car ownership would break through the benchmark level of five cars per hundred households when the affordability ratio falls into the range between 1.4 and 4.0. The study suggests a likely affordability ratio of 2.0 to 3.0 for 13 Institute of Techno-Economics of the State Planning Commission of China (SPC-ITE): "Household Automobile Development Strategies for China," The Economy Doily (in Chinese), No. 4774, October 24, 1994. 4 Ibid. It is the only document we obtained on projected car ownership. It does not specify how the levels of need and the physical conditions for car use are considered in the projections. 54 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions China to reach the benchmark. This implies a household with annual income of Y 50,000 would be able to afford a subcompact car that costs Y 60,000 to buy and Y 6,000 a year to operate. Based on a random household survey conducted in 14 provinces for a total of nearly 10,000 households with household income over Y 20,000 per year, the study estimated the number of households that would be able to afford, and also willing to own, a private car in future years. Based on these demand projections, the study provided the lower-bound estimates for the domestic car production capacity required. All these results and projections are summarized in Table 5. TABLE 5: AUTO OWNERSHIP AND AUTO PRODUCTION PROJECTIONS BY SPC-ITE Projections Item 2000 2005 2010 Households able and willing to buy (million) 4.0-4.7 15.5-16.5 37.5-40.5 Annual total demands for cars (million) 1.3-1.6 2.2-2.7 3.5-4.4 of which: household demands (million) 0.4-0.6 1.2-1.6 2.3-3.0 Total car ownership (million) 6 12 22 of which: share of household cars (%) 20% 40% 60% Cars per 1,000 population 4.7 9 15.8 Annual car production capacity (million) 1.2 2.2 3.5 of which: household car production (million) 0.4 1.2 2.3 Household car gas consumption (million ton) 0.8 2.8 6.5 Percent of national total gas consumption 2% 6% 11% Average gas consumption (ton/household car) 0.7 0.6 0.5 Source: SPC-ITE (1994). Reactions to this policy vary from highly skeptical to full acceptance. The skeptical view emphasizes the possible consequences on city air pollution, loss of agricultural land to road construction and low-density urban development, and traffic congestion in city centers. Those who accept the study emphasize the potentially strong role of the motor vehicle industry in leading long-term economic development, and regard growing traffic congestion as part of the price to pay for, and even as a symbol of economic growth. It is fair to say that the policy is based on quite realistic demand forecasts. The car production capacity target does not appear to be overstated for a country with such large potential demands. There are strong reasons for China to develop its motor vehicle industry, not only from a long-term perspective of export-led development, but also from a short-term perspective: developing an internationally competitive motor vehicle industry requires economies of scale, which are not easy to achieve without a domestic market to absorb its products during the early stage. These issues, and the implications that they might have for the development of the domestic vehicle manufacturing industry in China, are the subject of the rest of this paper. However, at this stage it can be noted that the SPC-ITE estimates of future car ownership come very close to our low-growth scenario estimates and well below our high-growth scenario. This suggests that there is a sufficiently strong base of demand to develop a viable motor vehicle industry while at the same time permitting each city to control the use of motor vehicles so that Stephen Stares and Liu Zhi 55 they do not overwhelm the capacity of the city transport system. This will be returned to in the final Section G of this paper on China's future strategy. C. CONSEQUENCES OF MOTORIZATION Motorization is already a powerful force in the economic growth of China. The Guangdong Province Transport Study'5 (mostly completed in 1989) reviewed transport costs and operations in that province, and concluded that motorized highway transport was ripe for development, partly to reduce transport costs, and partly to remove pressure on the overburdened rail system: "In China as a whole, past concentrations of development strategy on heavy industry and reliance on coal as 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 highways 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 the transportation users have been largely neglected in a situation of chronic shortage of transportation services.... ....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 highway 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 and medium distance passenger transport and small lot size or door-to-door freight up to 1,000 km haul distance, thus relieving railway congestion and improving overall service quality." These analyses were confirmed in the 1992 Yangtze Economic Zone Study."6 The findings of both these two studies are validated by the explosive growth in highway traffic in the past six years. In particular, the availability of, and control over, motorized transport in the form of trucks and small vans have probably been one of the keys to success of the nonstate-owned industry sector, in particular the Town and Village Enterprises (TVEs), which has been so central to the recent economic development of China (see Figure 6). Motorization of freight transport is by no means the only reason for nonstate-owned industry success, but they probably could not have succeeded so widely without it. 5 Guangdong Province Transport Study, World Bank, June 7, 1991. 16 Yangtze Economic Zone Transport Study, Yangtze Economic Zone Comprehensive Transport Study Group, April 1992. 56 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions As with freight traffic, motorized FIGURE 6: GROWTH IN CHINESE INDUSTRY personal transport offers greatly improved mobility and expanded geographical horizons, ? and this has the power to radically change X 6,000 lifestyles. While it clearly costs more to buy 5.000 o state and use a car than to take a bus or bicycle, o 4,000 o Nonstate timeliness, convenience, comfort, and 3,000 reliability of transport are real factors that 2,000 translate into personal satisfaction for Q 1,000 _ individuals. Motorization in any case does o 0 not come at a fixed unit price. Individuals 0 8 o § T can choose between a new Mercedes a limousine or a second-hand moped. What counts is access to mobility at an affordable price, and the fact that demand for personal motorized transport in China is soaring is proof enough that real benefits are perceived. The freedom and flexibility of transport afforded by motorization also gives rise to its greatest problem-traffic congestion. Access to the road system is generally freely available to those with vehicles, but clearly some parts of the road system attract more traffic than others. While traffic congestion on rural roads can generally be addressed by building more road capacity, traffic congestion in cities is a much more intractable problem. The nature and treatment of urban traffic congestion now becomes the main theme of this paper. The Nature and Costs of Traffic Congestion Every vehicle that joins the traffic stream interferes, to some extent, with the progress of other vehicles already on the road. In light traffic this interference is negligible; in heavy traffic it is severe. Hence, as more vehicles join the traffic stream, the speed of traffic falls until it reaches crawling speed with frequent and intermittent stops-the classic traffic jam. An obvious result of congestion is that journeys take more time to complete. Chronic traffic congestion can even result in fewer journeys being made because people are not prepared to do much more than the arduous commute to and from their workplace. Some suggest that we all have an unconscious, but more or less fixed, daily travel time budget-if this budget is used up on essential travel, there is nothing left (meaning no energy or inclination) to make additional journeys for leisure or other nonessential journeys. This can be interpreted as a reduction in the quality of life in a congested city. Congested traffic also means higher costs for the following reasons: Engines are less efficient at slower speeds, and consume more fuel. For example, dropping from 20 kilometers per hour (km/h) to 15 km/h will cause an average car to consume about 25 percent more fuel for every kilometer traveled. Stop-go conditions are even less efficient, with the engine idling during stops, and with frequent speed-change cycles. Stephen Stares and Liu Zhi 5 7 * More vehicles are required to do the same job. For example, slower traffic means that a wholesale delivery van can service less shops in a day. To deliver the same number of goods would require the distributor to invest in a larger vehicle fleet. * Time costs money, whether this is directly in the wages of goods vehicle, bus and taxi drivers, or in the lost time of business executives, school children and shoppers caught in traffic jams. Also, congested traffic conditions means that additional time has to be allowed for journeys to allow for the greater uncertainty of travel, wasting yet more time. An individual might weigh these costs and still decide that a journey through congested traffic is worth making for reasons of personal convenience. By doing so, however, the individual is unconsciously making travel more difficult-slower, more uncomfortable, and more costly-for everyone else already traveling. In fact, the costs and time penalties imposed on others in congested conditions can be larger than the costs incurred directly by the individual. This is particularly true for car travel which is the most inefficient user of road capacity. Some numbers are put to this Box 1. Economists believe this to be the crux of the traffic congestion problem. Motorists choose to travel on congested roads by car because they are responsible for only their own costs, and have no liability for the additional costs they impose on others. In other words, motorists do not pay the full price of their travel and this encourages excess traffic. More generally, economists believe that congestion is, to a considerable extent, created by the systematic underpricing (namely, prices well below cost) of all modes of urban transport services. This underpricing is evidenced in several forms, ranging from free or very low-cost parking for bicycles, motorcycles or cars, subsidies to public transit, and, as discussed here, free use of highly congested streets in central parts of cities. Hence, pricing policies are likely to be a key weapon in the fight against traffic congestion, as discussed in Section E of this paper. Impact of Traffic Congestion on Bus Services Traffic congestion affects all road users, but bus services are particularly badly affected. For example, take the case of a bus dispatcher at a terminus sending out buses on a route that takes one hour to complete. To maintain a 6-minute bus headway, 10 buses will be needed since the first bus will return in time to make the next journey along the route after the tenth bus has departed (give or take spares for fueling breaks and breakdowns). If traffic congestion increases so that it now takes half an hour longer to complete the route, the dispatcher has two choices; either send buses out less frequently -now every 9 minutes--or put more buses on the route- 15 buses are now needed to maintain a 6-minute headway. IThe second choice does not exist in the short term but, faced with generally increasing congestion, a bus company over the longer term can acquire more buses to maintain timetables. Either way, service to passengers will deteriorate with definitely longer travel times and, if extra buses are not provided, longer waiting times. If the bus companies do buy more buses to maintain headway, these must be paid for, along with the extra drivers, extra depot space and larger maintenance staff. This means either higher bus fares for passengers or an increasing deficit financed by the municipality. While the latter appears to avoid impacts on the passengers, in the long run bus company deficits cannot be 58 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions allowed to rise forever, so that lower service or high fares will likely prevail. (See Box 2 for some estimates of the impact of traffic congestion on bus operations.) Box 1: MOTORIST CONGESTION COSTS Analysis shows that in congested traffic conditions, the costs a motorist imposes on others far outweigh the costs incurred directly by the motorist. This is illustrated in the Box Figure, which shows the total cost of operating a car at different speeds, separating out the cost incurred by the car owner, and the cost imposed on other road users by the presence of that car. The computation was based on the methodology set out in the Smeed Report' but using road usage parameters from Hong Kong. Values are expressed in 1995 US dollars. Box FIGURE: TOTAL COST OF CAR OPERATIONS COST PER CAR-KM (US$) 14 I F 0 OWNER COST 12 - COST TO OTHERS - TOTAL COSTS 10 6 4- 2- 0 F- I+----++--- - T - | _ 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 TRAFFIC SPEED (KM/H) At 30 km/h, which would be an excellent speed in urban conditions, a motorists incurs a cost of about $0.5/km. At the same time, the motorist is imposing a total additional cost on all other road users of about $0.08 for every kilometer traveled. At 12 km/h, which corresponds to typical congested conditions, the situation changes significantly. The costs incurred by the motorist rise by about 50 percent to $0.8/km, but the cost imposed on all other road users increases nearly 20 times to $1.4/km. In chronic congestion at 5 km/h, costs incurred by the motorist almost double from the previous level to $1.5/km, while the costs imposed on others increases another sevenfold to $10.2/km. The calculations are less reliable at lower speeds, but the graphical representation of the steep increase in costs imposed on other traffic at low speeds is generally realistic. Road Pricing: The Economic and Technical Possibilities, Smeed, R., et al., London, 1964. Thus, increasing traffic congestion results in poorer service and rising fares, which in turn encourages passengers to find some other way to make their journeys, adding further to congestion. Thus develops the downward spiral of increasing congestion, rising costs, poorer service, higher fares, reduced patronage, and yet higher congestion that confronts urban bus operators the world over. Stephen Stares and Liu Zhi 59 Box 2: IMPACT ON CONGESTION ON HONG KONG BUSES Estimates below show Hong Kong bus operations under two conditions of traffic congestion in 2001, as simulated for the 1989 Second Comprehensive Transport Study. Congestion Level Lower Higher Av speed (km/h) 14.3 10.8 Daily bus-km 602,000 602,000 Daily bus-hours 42,000 55,700 Number of buses 4,120 6,160 Hence with a reduction in average operating speeds of 3.5 km/h, yet trying to maintain the same volume of bus operations measured in daily bus-km, operating hours would increase by one third and about 50 percent more buses would be required to maintain the same frequency of services. The total cost of operations would be increased by perhaps 30 percent which would have to be covered by a subsidy from Government (not accepted in Hong Kong) or an increase in fares. At the time of the study, bus and minibus transport together carried about two-thirds of all public transport passengers in Hong Kong, so the need to control traffic congestion was very apparent. Impact on the Environment Emissions from motor vehicle engines are a serious source of air pollution in motorized cities. Pollutants from vehicle engines include lead, carbon monoxide, nitrogen oxides, ozone, and particulates. These pollutants cause a wide range of adverse impacts on urban populations, including impaired intelligence and reduced learning ability (from lead poisoning) and a series of respiratory ailments including infections, asthma, and decreased lung efficiency. Besides adversely affecting individual welfare, these impacts impose health burdens and costs on the community that reduce the economic efficiency of urban areas. Air pollution from motor vehicles increases under congested traffic conditions. The generally lower traffic speeds, intermittent stop-go operations and, in chronic congestion, frequent periods when vehicles are halted with idling engines, not only increases the amount of fuel consumed, but the fuel is burned less efficiently by the engines. The net result is that the quantity of emissions per unit of distance traveled is sharply increased. In fully motorized societies, motor vehicle emissions comprise a high proportion of total urban air pollution. For example, in the countries of the European Union in 1990, motor vehicles caused 50 percent of all nitrogen oxide (NO,) pollution and over 60 percent of all carbon monoxide (CO) pollution. But motor vehicle pollution is perhaps an even greater problem in rapidly motorizing developing countries where emission controls are less extensive and effective, and engine technology is older. In Manila, for example, the contribution of motor vehicles to total NO, and CO emissions in 1990 was 80 percent and almost 100 percent, respectively. While China is at an early stage of motorization, the potential course of motorized pollution is plain to see. In Beijing, one of the centers of motorization, the contribution of motor 60 Theme Paper 1. Motorization in Chinese Cities: Issues and Actions vehicles to NO, and CO pollution has already reached 46 percent and 30 percent, respectively. For less-developed Chinese cities, industrial and domestic emissions of pollutants are probably the dominant concern at present. However, in the next few years as industrial and domestic pollution are reduced and motorization advances, motor vehicle emissions will become a much more significant source of air pollution. These issues are discussed in detail in Theme Paper 2 on "Motor Vehicle Pollution Control in China: An Urban Challenge." The paper also reports that one of the lessons from motorized countries is that motor vehicle emissions can be greatly reduced with an aggressive strategy for cleaner air. China currently has an opportunity to set such a motor vehicle emissions control strategy, which could significantly reduce the extent and costs of future air pollution in urban areas. Air pollution is not the only environmental problem posed by increased motorization of cities. Other problems include motor vehicle noise, especially from concentrated high-speed traffic on urban expressways, decreased local amenities when excessive vehicle flows dominate local streets, visual intrusion from poorly designed highway structures, and reduced traffic safety due to the mixture of intrinsically incompatible vehicles, bicycles and pedestrians. All of these adverse environmental impacts can be mitigated by better design, sensitive to the function and needs of the city. However, for better design to flourish requires both a commitment from the city to improving environmental conditions, and also the development of professional skills to exploit the techniques available for improved design. Impact on Nonmotorized Traffic The mixture of motorized and nonmotorized vehicles is inherently hazardous under any traffic condition. The relatively higher speeds of motorized vehicles, and the unprotected nature of the bicycle, means that cyclists are highly vulnerable in accidents. The effect of traffic congestion on bicycle traffic is complex. The competitiveness of the bicycle is generally increased when congestion slows motorized traffic to the speed of the bicycle. Also, riding conditions are possibly less hazardous, with the effects of higher volumes of motor traffic offset by the generally lower speeds in congestion. On the other hand, air quality is very much worse with greatly increased motor vehicle-generated pollution affecting cyclists more than most road users. Uncontrolled congestion also has a tendency to spread to all parts of the street network, so that motor vehicles will penetrate back lanes and side roads previously the sole domain of bicycle and pedestrian traffic. On balance, conditions for bicycles in congested traffic are probably worse, but their competitive advantage is probably increased. It must be recognized that bicycles are substantial users of road capacity, and that the massive bicycle flows in many Chinese cities are themselves a contributor to traffic congestion, especially at intersections. Until recent times, however, the mainly bicycle flows were manageable. It is the mixture of motor vehicles and bicycles that has sparked off serious congestion. More is said on the role of the bicycle in Section D of this paper. Stephen Stares and Liu Zhi 61 Impact of Motorization on Land Use and Urban Form The evolution of residential and employment location patterns commonly seen in Western cities provides a useful reference for projections of the likely impacts of motorization on land use and urban form in China. Although the details of Western land use patterns vary city by city, they possess some fundamental similarities. These can be summarized as a concentric urban form comprising a central business district (CBD), a central city, and outlying suburbs, with a radial transport network linking the three components. Basic businesses'7 mostly cluster in the CBD, residences and population-serving businesses mainly locate in the central city and suburbs, and most manufacturing businesses locate in suburban areas with convenient road transport. Both residential and employment densities decline with distance from the CBD. Forces for Suburbanization. Historically, Western cities were overcrowded with both population and employment located within close proximity. It has evolved into the modern form described above through decentralization or suburbanization, a process characterized by expansions of urban geographical boundaries and declines in both population and employment densities. Suburbanization is caused and affected by many socioeconomic factors, the most notable among them being (a) increasing urban populations, (b) rising real incomes, (c) falling real transport costs, and (d) the relative price of rural and urban land. Growing urban populations have direct impact on the expansion of urban boundaries. However, rising real incomes and falling real transport costs are the main driving forces for the declines of both population and employment densities, which also lead to urban spatial expansion. The pace and extent of suburbanization have been affected by the availability of cheap land as well. It is important not to misunderstand the role of private cars in this process. Suburbanization occurs in all parts of the world, developed and underdeveloped, with and without widespread car ownership. In the United States, suburbanization did not start with the emergence of the private car, but with streetcars. It is misleading to attribute the lower-density suburban development in the United States simply to widespread ownership of private cars. The post-World War II rapid urban sprawl in the United States was the consequence of a number of factors, including the federal housing policies that encouraged single-family detached housing construction. Private cars played a role to reinforce the trends, but only a small role. The generally lower densities of US cities than many cities in other parts of the world are also explained by the fact that land is more abundant and, hence, land cost is relatively lower in the United States. When land supply is limited or land prices are relatively high, availability of private cars does not lead to such widespread suburbanization. Motorization does play an important role in modern manufacturing business location, but it does not act alone. The suburbs provide cheaper and larger pieces of land for the single-story plants more suitable to the needs of many modern manufacturing businesses. These are most easily served by road transport, which provides cheap door-to-door services. 7 Urban businesses can be divided into two major categories. The so-called basic businesses are those exporting their goods and services to other areas. The "population-serving" businesses sell consumer goods and services to local households. 62 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Motorization and Land Use. To further understand how motorization affects urban land use patterns, we should understand how businesses and households respond to changes in transport in their chosen location. The modern urban form results from a market process where land prices are determined through demands for land by businesses and households, and typically decline with distance from the CBD, reflecting the comparative advantages of location. Businesses choose their location to maximize profits. The basic businesses cluster in the CBD because, by locating there, they are able to generate sufficient revenues to cover the high land prices and still make a profit. Households choose their residential location in the outer areas where the much lower housing prices (due to lower land prices) more than offset the higher commuting costs. Population-related businesses, of course, follow the residential locations. A transport improvement between the CBD and the suburbs lowers the cost of commuting, and makes it possible for households to move farther away from the center to enjoy larger housing. This decrease in commuting cost also allows businesses at the center to contain wages, thereby improving their position over competitors located in other areas, and maintaining the attractiveness of the CBD to new businesses as a center of employment. An increase in commuting costs on the other hand, as would be caused by increasing traffic congestion, initially has an opposite impact: centralizing residences and decentralizing jobs. As congestion continues to grow and the CBD is seen as an increasingly expensive and unattractive place to work, jobs would tend to decentralize, taking households with them. In summary, if land is not controlled, increasing motorization will likely promote decentralization of both population and employment locations, but this process will be greatly accelerated if severe congestion is allowed to dominate the city center. Clearly, decentralization caused by congestion is a very inefficient process. Motorization and the Development of Chinese Cities. The major Chinese cities currently have quite different land use patterns from those found in Western cities. Much of the difference is the legacy of several decades of urban land development under strict government controls in the absence of a land market, especially the widespread implementation of enterprise- based provision of housing and other services. However, the rapid changes now taking place, most notably residential dedensification and large-scale addition of office buildings in the city center, lower density residential development in the city fringe areas, and relocation of manufacturing businesses from city center to suburbs, are following the evolution of urban form similar to that seen in the West. These changes are associated with three major factors: (a) the emergence of an urban land market that determines land values to reflect locational advantages, (b) income growth that makes overcrowding increasingly unsatisfactory and dedensification possible, and (c) availability of faster transport modes, including cars. Motorization will facilitate such changes of urban form in Chinese cities. It is also probable that uncontrolled motorization would speed up the decentralization process further, but in an inefficient manner. While there are good reasons for guiding motorization to avoid excessive urban sprawl, the Government should be aware of three lessons from the Western experience: * Car ownership is not the single factor for urban sprawl, and excessive urban sprawl would not be avoided by simply restricting private car ownership. Stephen Stares and Liu Zhi 63 * The key to avoiding excessive urban sprawl is a firm land development policy; this is particularly important before the land market is fully developed to reflect the scarcity of land resource. If the Government loses control of land development, growing motorization will no doubt help lower the density of development in city fringes. * Growing traffic congestion will likely erode the locational advantages of city centers and encourage businesses to locate in outer areas. Impact on Economic Growth The negative impact of traffic congestion on economic growth is self-evident. Congestion increases transport costs for both freight and commuters. Higher cost for freight transport directly translate to lower productivity; and higher commuting cost requires employers to pay higher wages that will increase production costs and thus decrease productivity. Estimates of the scale of economic losses due to congestion are very uncertain. Most of these estimates were obtained by comparing the costs of transport in currently congested conditions with costs under assumed ideal conditions, but these ideal conditions are difficult to identify and can be quite unrealistic. After all, congestion is a consequence of more economic activities associated with economic growth. This is particularly true for urban areas, in which these activities tend to concentrate due partly to urbanization driven by economic structural change, and partly to economies of scale in production. This close proximity of activities, in location and timing, is therefore both a cause of congestion but also creates the climate for economic growth. Studies in several cities including Seoul, Bogota, Bangkok, and cities in northeastern India and Nigeria,"8 indicate that the city center acts as an incubator for the creation and growth of small- and medium-size businesses. They require close proximity to other similar businesses that they do not find in the suburbs. However, severely deteriorating traffic conditions in Bangkok have impeded that role of the city center, forcing medium and small businesses to move to less-productive locations. This supports the notion that mild congestion is part and parcel of urban life, but chronic congestion diminishes the economic life of a city. Conclusions on Consequences The benefits of motorization to China are likely to be substantial in terms of improving the quantity, quality and efficiency of transport services. The motorization process is integral to continuing national economic development. Despite some alarmist speculation, economic forces will likely control the depletion of both fuel and land resources. The best defense here is the continued move to a market-oriented economy with prices fully recognizing economic resource costs. Is The results of these studies are summarized in Cavelle D. Creightney (1993), Transport and Economic Performance, A Survey of Developing Countries, World Bank Technical Paper Number 232. 64 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Motorization poses two substantial threats to city life. The first is the increase in air pollution, which, if unchecked, could seriously damage health and impose significant costs on hle community in the form of increased health care and loss of productivity. This subject is taken up in Theme Paper 2 on "Motor Vehicle Pollution Control in China: An Urban Challenge." The other main threat is the development of chronic congestion in cities. This would dissipate many of the expected benefits of motorization, and destroy much of the amenities and quality of city life. Techniques for dealing with traffic congestion are discussed in tlie next two sections. D. DEALING WITH TRAFFIC CONGESTION-MANAGING ROAD SPACE SUPPLY The instinctive solution to traffic congestion for most city managers is to try to supply more street traffic capacity to absorb the excess traffic demand. This can be done in two ways: by managing existing street space more efficiently to maximize available capacity, and by constructing new streets and roads to add more capacity. These two approaches are discussed in 00is Section. .he Street Management Solution Available road space in most Chinese cities is not used to the limits of its capacity because this has not been necessary until relatively recently. With growing traffic pressures, cities are now in the process of developing traffic management skills to make more efficient street systems, but this will take a little time to mature. Elements of traffic management solutions comprise the following: * A logical road hierarchy, defining the traffic function of each street; * Improvement of street and junction layouts through placement of curbs, painting white lines, installation of proper signing; * Segregation of major traffic classes-bicycles, buses, and others-to the extent possible by: * horizontally-separate lanes or streets for each class of vehicle; * by time-selective priorities and restrictions by time of day; * vertically-grade separation of key junctions; * Extension, automation, and linking of traffic signal systems; * Removal of hawker traffic to side streets; and * Education and enforcement. With the exception of traffic signals and grade separations, traffic management is essentially low-technology and low-cost. However, it requires training and skills to develop and apply appropriate solutions and, above all, organization and determination to bring all the components together into a coherent traffic management strategy. The status and importance of Stephen Stares and Liu Zhi 65 traffic management in Chinese cities needs to be raised, and this could be assisted by the development of interagency teams to tackle traffic management in a comprehensive manner."' It is estimated that street capacity in currently congested cities, where some traffic managemelt measures have already been applied, could be boosted by 30 to 50 percent withl retined techniques. This is well worth while pursuing. Bicycle Solutions. Since the mixture of bicycles and motor vehicles appears to exacerbate congestion, two extreme bicycle "solutions" have been proposed to make mol-e efficient use of road space: either ban the use of bicycles altogether and rely entirely on motor vehicles (including public transport), or ban the motor vehicle and give the streets back to the cyclists. Neither is realistic. Banning bicycle use would be immensely disruptive to urban life and the urban economy since they remain the dominant mode of transport in even the largest cities. It is possibly the most efficient mode of transport for short journeys (2 to 5 kin) and this efficiency wouild be lost by a total ban. Even if the disruption of a ban could be accepted, it would put enormllous pressur-e on the public transport system, and this could not be expanded overnight. Hence, the out-iglt elimination of bicycles could not be supported. However, in the context of leveling the playing field for all modes of urban transport, there is a case for removing unreasonable subsidies to cyclists, and even imposing fees on cyclists commensurate with their use of road space (as long as all modes are treated evenhandedly). Banning motor vehicles and giving the streets back to bicycles is equally unrealistic. The economic forces outside cities driving the motorization process are unstoppable, and motor vehicles in the form of buses, vans, taxis, trucks (especially small trucks), mopeds, motorcycles and, increasingly, cars, will have needs to penetrate city centers. 'T'hey cannot realisticallv be relegated to the city fringe. Bicycles and motor vehicles must learn to live together. Positive provision for cycisis through segregated facilities is likely to be a more efficient solution than either outriglht bans on1 cyclists or pursuing dreams of turning the clock back to an earlier transport age. Positive provision for the bicycle would benefit both bicycles and motor vehicles as discussecd furtlher in Theme Paper 5, "Bicycles in Cities." The Smaller Vehicle Solution-the Role of Motorcycles. Of all indtdivualhiet) vehicles, motorcycles can be the most efficient users of road space in terms of the volumile aod speed of passengers transported. In urban areas, motorcycle speeds are comparable to cai speeds, but a motorcycle requires only about half of the road space needed by a car. Assumving both motorcycles and cars have similar average occupancies, as is the case in Guangzlhou, a road can carry twice as many travelers by motorcycle as by car. The efficiency of motorcycles relative to bicycles is not so obvious. Compared with a bicycle, motorcycles occupy more road space, but carry more passengers and travel faster. Probably bicycles are more efficient lor shoI journeys in the city center, but the balance of efficiency would swing to the motorcycle for longer-distance trips on relatively uncongested roads. 19 Discussed further in Theme Paper 4, "Municipal Transport Management: Overseas Experienice." 66 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions There are, of course, many disadvantages to motorcycles. They are generally much more polluting than cars in terms of noise and engine emissions. Also, the combination of higher car-like speeds with bicycle-like lack of personal protection makes motorcyclists vulnerable to traffic accidents, with adverse consequences for both individuals and the community. It is also clear from Section B earlier that the demand for motorcycles is immense and, if uncontrolled, could quickly swamp the capacity of the street systems of most cities. Some of these issues could be dealt with by specifying higher standards for motorcycles (particularly on engine emissions and mufflers); this would alleviate (but not eliminate) some of the problems at source and, because of the higher prices associated with the higher standards, would restrict the rate of growth. Other controls on motorcycle growth might be deemed necessary, but these should be considered carefully in the light of the clear benefits that motorcycles bring to a segment of the population. They are not inherently inefficient users of urban road space, and the potential role of motorcycles should not be left to go by default. The Larger Vehicle Solution-More Use of Public Transport. Public transport buses are potentially the most efficient users of road space per person transported, and greater use of buses in place of either bicycles or cars could improve overall road transport efficiency. In the busier corridors, removing passenger flows from roads onto metros would have even greater impact, but high expense and complexity are likely to limit metros for the foreseeable future to just a few corridors in the major cities. Development of road-based bus mass transit is a much more affordable solution for most Chinese cities. In particular, the development of bus transit along important radial routes between the city center and the suburbs could serve as a "premetro." Such schemes could both satisfy existing demand at an affordable price, and help concentrate passenger flows and strengthen the viability of subsequent upgrading to rail-based transit systems. These aspects are discussed in detail in Theme Paper 6, "Investment in Mass Rapid Transit." For buses to be effective, they need to be protected from the impacts of road congestion by allocation of street priorities. These could include the premetro busways in major traffic corridors as discussed above, but also city-center street priorities including bus-only lanes, bus- only streets, bus "gates" (short sections of streets accessible only by buses), priority for buses at traffic signals (through automatic detection of bus arrivals), and so on. Such priority measures are in common use throughout the world, but are still rare in China. One objection heard is that dedicated bus lanes often appear lightly used relative to adjacent general-traffic lanes. While this will be true in terms of the number of vehicles, bus lanes are potentially much higher carriers of people, and this is the key measure to efficient use of street space in urban areas. The Role of Taxis. Taxi service can also increase efficiency of city-street usage. Combined with other policies and services, taxis can substitute for the use of private cars in city centers, providing higher-quality and more timely transport for those disinclined to use public transport or bicycles. The role of taxi service needs special attention because of its close relationship with the development of public transit. Taxi transport is a growing industry in urban China. In Guangzhou, the number of licensed taxis increased from approximately 8,000 at the end of 1991 to 12,800 in 1995, an average growth of 16 percent per year. In Beijing in 1994, the taxi fleet of nearly 50,000 Stephen Stares and Liu Zhi 67 vehicles carried more passengers than the subway, and accounted for 13 percent of all public transit journeys (3.9 billion passengers). By all accounts, current taxi services are financially viable. For example, Beijing taxi vans cost around Y 50,000 and are rented out by the taxicab companies to drivers for about Y 4,500 per month; drivers are reported to generate sufficient revenues to cover operating costs and provide a wage. Although rapid growth in the taxi industry has resulted primarily from rapid income growth and the resulting demands for faster and better passenger transport services, other contributing factors should not be ignored. In particular, underdeveloped public transit systems prevalent in many Chinese cities leave many tripmakers, particularly business visitors, few choices but the taxi. In other words, poor public transit systems have provided an opportunity for taxis to penetrate the market. As a result, the number of taxis per 1,000 resident population is high in many large cities: 8.0 in Beijing; 3.4 in Shanghai; 4.2 in Guangzhou; 3.5 in Qingdao; 3.7 in Urumqi; and 3.2 in Fuzhou. This contrasts with 3.0 taxis per 1,000 population in Hong Kong, a city noted for its relatively high taxi usage. Rapid growth in the taxi industry also appears to be a mixed blessing for at least some large Chinese cities. While taxis have met fast-growing demands for better service, large volumes of taxi traffic are adding significantly to traffic congestion. It was reported in Beijing that taxi flows comprised 60 percent of the peak-hour traffic volume in 1994, much of it due to the cheap and popular minivan taxicabs (nicknamed "yellow bread loaves" for the vehicle's shape and color) introduced in 1992. The municipal authority recently decided to restrict new registration and licensing for this type of taxi vehicle. Given the limited data available, it is difficult to make a definitive statement on the proper role and extent of taxicab services in Chinese cities. This is worth more detailed study. Clearly, taxis have proved very popular in filling a void left by inadequate public transport services, but their massive presence in some cities is now possibly inhibiting the development of good public transport. The taxi policy should be linked with public transport policy, and relative pricing of taxi and public transport services should be a key issue. Both services should be able to cover all their costs of provision, and these costs should include the costs of congestion imposed on other traffic. Since taxis impose greater congestion costs per passenger than do buses, taxi fares should probably increase in the future relative to public transport fares. This would promote bus usage, helping to restore the balance between the two types of transport service. Other Efficiency Options. A variety of options are available to spread peak transport demand or use road space more efficiently. The following are a few examples: * Staggered Working Hours. Since the main transport problems occur in the morning and evening peak hours, widespread implementation of staggered working hours, including flextime, could improve the peak-hour commuting problem and relax the need for traffic control measures. However, such policies are not popular with employers and positive efforts are needed to gain wider acceptance. * Park and Ride. Construction of bicycle and car parks at mass transit terminals and stations promotes the use of public transport for access to the city center. 68 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions * Vehicle Sharing. Filling cars with passengers makes them more efficient users of road space, and this can be encouraged by giving advantages to high-occupancy vehicles (HOVs). It is common practice in the United States to allow HOVs to use designated bus lanes on freeways. A common practice in South America, also followed in Washington, DC, is to promote sharing of taxis by allowing drivers to pick up additional passengers along the route. While this does lead to higher occupancy of taxis, which is more efficient for the road space used, taxis as a result tend to keep to the main corridors that are already served by public transport, which reduces benefits. While all these methods are worth exploring, they operate at the margin and must be regarded as supplementary to other policies. The Road-Building Solution In response to the sharp increase in motor vehicle traffic, many cities in China are developing plans for extensive new road construction. This section looks at the background to these plans, their scope, and their likely effectiveness in dealing with road traffic congestion. Urban Road Infrastructure Deficiencies Many Chinese cities were developed long before the advent of the motor vehicle and their street systems now struggle to cope with modern traffic flows. Much can be done to improve the efficiency of existing streets by traffic engineering measures, as discussed above. However, many city planners also believe that Chinese cities are underprovided with roads as compared with modern city prototypes. A much-quoted statistic in China is the percentage of the total urban area given over to roads. Low figures for China (for example, 5, 6 and 11 percent for Shanghai, Guangzhou and Beijing, respectively) are compared unfavorably with much higher figures elsewhere (for example, 18 percent for Seoul, 23 percent for Tokyo, 35 percent for London, and an incredible 45 percent for Washington, DC). With grave suspicions about the contrast between Shanghai at 5 percent and Washington, DC at 45 percent, we conducted and solicited actual measurements of selected streets in Washington DC and several Chinese cities, with the following results:20 City and District: % Road Space Washington Downtown 16-18 Central Guangzhou 8 Shanghai Downtown 17 Shanghai Luwan District 12 Shanghai Southwest Suburb 6-8 Beijing Within 2nd Ring 10-12 20 Stephen Stares, unpublished note, February 1995. Stephen Stares and Liu Zhi 69 The striking finding is that downtown Shanghai has approximately the same road space percentage as downtown Washington. This is hardly surprising because downtown Shanghai is really a European street layout, but how could the quoted statistics have diverged so much? Probably for several reasons having to do with the definition of the measurements. The earlier- quoted Washington figure of 45 percent is almost certainly not the share of road space. It is more probably the total percentage of the urban area not occupied by buildings, and so includes the wide sidewalks and possibly even the park areas of its famous Mall. It is also possible that the quoted Chinese cities' statistics include much undeveloped land in the measure of the total urban area, thereby depressing the percentage of street space. Quoting a single figure for an entire city, in any case, gives a misleading impression since considerable variation exists between districts. Whatever the reasons, such simplistic comparisons are dangerously misleading Based on the measurements cited above, though, it can be concluded that road space in many Chinese cities probably is low, although not as low as commonly quoted. The comparative underprovision of roads is probably due to a street layout common to many Chinese cities where very large city blocks are bordered by 4- to 6-lane streets, but penetrated only by narrow alleys scarcely one vehicle wide. How Much Road Construction Do We Need? It seems clear that many major cities in China could benefit from additional road space. However, the needs for road space can be quite different among cities of similar level of motorization, because there are many other interrelated factors. These include the levels of public transit service, the spatial distribution of jobs and residences, urban topography, government policies that affect the relative prices of using different modes and allocation of urban road capacity among them, and so on. Hence, there is no single formula applicable to all cities. For a particular city, how much road space is needed and where it should be located require careful evaluation in order to gain the maximum benefit with limited financial resources. With these cautions in mind, we have nonetheless attempted to quantify, in very broad terms, the scale of future road construction needs in typical Chinese cities. This was done by making comparisons of international data on two aspects: (a) the relationship of per capita road length vs. population density, and (b) the proportion of high-grade urban roads. These comparisons strongly suggest that there are certain regularities in the relationships, which could be helpful in assessing the needs for urban roads in Chinese cities. Per Capita Road Length vs. Population Density. It should be acknowledged that a more precise measure of physical amount of roads is lane-kilometers of roads by functional type. These data, however, are generally unavailable. Therefore, we have to use total length of urban roads as a proxy for urban road space. This will be supplemented later with an examination of the variations in the composition of urban functional roads among different cities. Figure 7 plots per capita urban road length against population density for 65 world cities, including several Chinese cities. Some of the cities appear twice or three times in the sample because data for different years (mostly 10 years apart) are available. All plots lie fairly closely around a smooth curve similar to a hyperbola function, strongly suggesting that denser cities have lower per capita road length than less dense cities. When we fitted the two variables into a log-log function, we 70 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions obtain a striking -1.0 elasticity estimate, which could be interpreted that a 1 percent decrease in urban population density would be accompanied by roughly a I percent increase in per capita length of urban roads.21 FIGURE 7: RELATIONSHIP BETWEEN URBAN POPULATION DENSITY AND PER CAPITA LENGTH OF URBAN ROADS (137 observations for 65 world cities) 16.00 . . . . . . . ..... . .... ..... . . . . 14.00 * 12.00 10.00 8 8.00 PediU treiced 6.00 - 4.00 A.. 2.00 0 5,000 10,000 15,000 20,000 25,000 30,000 Density (persons/kiM2) The underlying reasons for the simple relationship between per capita road length and population density are complex, but reveal a simple fact: per capita length of urban roads is, by and large, determined by the population density. In Figure 7, several US sunbelt cities lie at the low-density extreme, while Hong Kong is at the high-density extreme. Hong Kong is much richer than many other high-density cities in the sample, but it has the lowest per capita road length due to its extremely high density. Taking this international regularity as a yardstick, it is safe for us to say that urban road development in China's major cities is unlikely to go anywhere near the level of Los Angeles or Phoenix, as long as densities in Chinese cities remain high.22 21 Liu Zhi (1996), "Urban Form and Transportation: An International Perspective," unpublished research note. 22 Even the post-World War 11 US urban highway construction has been criticized for providing excessive and uneconomic capacity. The available evidence suggests that for most of the very costly centrally located facilities, the costs of added capacity exceed the incremental benefits of reduced congestion. For more discussion, see John R. Meyer and Jose A. Gomez-lbanez (1981), Auto, Transit, and Cities, Harvard University Press, p. 228. Stephen Stares and Liu Zhi 71 It should be indicated, moreover, that the plots for Guangzhou, Chengdu and Jinan all lie far below the curve (or the international average) and the plot for Beijing is roughly at the curve. This suggests that per capita road lengths for these cities, except Beijing, are substantially low by international standards. Proportion of High-Grade Roads. The previous comparison of per capita road length masks the variation in composition of urban road types among world cities. The character of urban road systems is very much determined by the combination of expressways, major arterials, minor arterials, collectors and local roads, and investment requirements for different types of road are obviously different. Does urban road composition by functional type vary significantly city by city? To answer this question, we examined the proportion of road space dedicated to urban expressways and arterial roads for 16 world cities and for 113 US urban areas.23 The results shown in Table 6 strongly suggest that the share of road space given over to high-class roads, and particularly expressways, is quite small and varies within a narrow range. Several East Asian cities with higher levels of motorization have similar shares: 2.4 percent in Tokyo, 2.6 percent in Seoul, and 3.4 percent in Singapore. The small standard deviations for the larger US sample are striking, considering that the sample embraces a wide range of city types, from New York to Los Angeles. TABLE 6: COMPOSITION OF URBAN ROAD FACILITIES (Percent of total urban road length) 16 World Cities 12 with higher 113 US All 16 cities auto ownership Urban Areas in 1990 Standard Standard Standard Road Type Mean Deviation Mean Deviation Mean Deviation Expressways 1.71% 1.65% 2.40% 1.53% 2.93% 1.15% Arterial Roads 16.31% 7.44% 15.53% 5.88% 15.67% 4.00% Primary Arterials n/a n/a n/a n/a 6.35% 2.13% Secondary Arterials n/a n/a n/a n/a 9.32% 3.08% Collectors n/a n/a n/a n/a 9.53% 2.51% Local Access Roads n/a n/a n/a n/a 71.87% 5.39% It is not hard to understand the small share of high-grade urban roads. These roads are provided for higher-speed motorized travels between urban districts, and any traffic using them comes from and ends up on the local street network. One should not build more high-capacity roads than the local streets can bear in various parts of the city. It should be noted, moreover, that most of the developing cities in the world still do not have high-grade urban roads such as urban expressways because of their very low levels of motor vehicle ownership and use. 23 These 16 cities include Washington, DC, Denver, Toronto, London, Paris, West Berlin, Tokyo, Osaka, Singapore, Seoul, Kuala Lumpur, Bangkok, Jakarta, Warsaw, Krakow, and Gdansk. These cities were selected due to availability of comparable data. 72 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Increasing motorization will require the construction of some high-grade roads to complete the urban functional road hierarchy needed to serve modem traffic. Estimated Road Needs for Chinese City Prototypes Based on the relationship between per capita road length and population density, and the proportion of high-grade roads, we projected the urban road needs for three large Chinese city prototypes. They were assumed to have the general characteristics of Chinese cities with population over I million. In general, larger Chinese cities have higher population density, lower per capita road length, and slower population growth due to government control. These three prototypes are: 2 * Prototype 1. A city of I million population at 100 persons/km , 0.39 meters (m) of roads per capita, population growth of 5 percent per year, and no change in density. 2 * Prototype 2. A city of 2.5 million population at 150 persons/km , 0.33 m of roads per capita, population growth of 2 percent per year, and a declining density to 100 persons/km by 2020. 2 * Prototype 3. A city of 5 million population at 200 persons/km , 0.26 m of roads per capita, population growth of I percent per year, and a declining density to 150 persons/km by 2020. Based on the assumed population density for the future years, we estimated two sets of total "target" road length for each city prototype. The results are shown in Table 7. The first set of estimates, termed the "status quo" estimates, was obtained simply applying the growth rate of per capita road length determined by the -0.1 density elasticity, to the base year per capita road length. The estimates give the additional road space needed to accommodate population growth and dedensification, assuming the same character of road network (or more precisely, the road network density, that is, kilometers of roads per squared kilometer urban built-up area) is maintained as previously. The second set of estimates, termed "toward the world average" estimates, assumed these city prototypes make extra efforts to increase per capita road length to the world average level (represented by the regression curve shown previously in Figure 7) during urban expansion and dedensification. The difference between these two sets of estimates of total road length is illustrated in Figure 8. The extra efforts would require both higher road density in the new development area and a modification of road network in the existing built-up area, particularly the crowded central area. Adding more roads in the city-center areas would require massive redevelopment that makes room for new roads, or building the expensive viaducts. But these could be done only in a small part of the center area one at a time. In other words, it is nearly impossible to upgrade per capita road length to the world average level in a few years. This would take a long time, probably half a century, to achieve. Therefore, we assumed by the year 2020, or 25 years from now, per capita road length for these prototypes achieve the mid-point level between the status-quo estimate and the world-average estimate. We further used the mean share of urban expressways for 12 world cities with higher auto ownership (2.4 percent) to estimate the length of expressways for these Chinese city Stephen Stares and Liu Zhi 73 prototypes for each future year. Again, two sets of estimates were obtained. These results are also shown in Table 7. TABLE 7: PROJECTED URBAN ROAD NEEDS FOR LARGE CITY PROTOTYPES BY SIZE Urban Assumed Urban Total Urban Roads (km) Expressways (km) Urban Density Land Toward Toward Population (persons/ Area Status the World Status the World Year (million) k2) (km2) Quo Average Quo Average 1.0 Million Pop. 1995 1.0 10,000 100 390 390 8 8 2000 1.3 10,000 128 500 580 11 12 2005 1.6 10,000 163 640 830 14 18 2010 2.1 10,000 208 810 1,190 17 26 2015 2.7 10,000 265 1,040 1,680 22 36 2020 3.4 10,000 339 1,320 2,350 28 50 2.5 Million Pop. 1995 2.5 15,000 167 810 810 17 17 2000 2.8 14,000 197 960 1,060 21 23 2005 3.1 13,000 235 1,150 1,380 25 30 2010 3.4 12,000 280 1,370 1,790 29 38 2015 3.7 11,000 337 1,650 2,330 35 50 2020 4.1 10,000 410 2,010 3,030 43 65 5.0 Million Pop. 1995 5.0 20,000 250 1,300 1,300 28 28 2000 5.3 19,000 277 1,440 1,570 31 34 2005 5.5 18,000 307 1,600 1,880 34 40 2010 5.8 17,000 341 1,780 2,250 38 48 2015 6.1 16,000 381 1,990 2,690 43 58 2020 6.4 15,000 427 2,230 3,220 48 69 Note: The Scenario "Toward the World Average" assumes that per capita road length in large Chinese cities catch up with the world average level by 2045. FIGURE 8: ALTERNATIVE PROJECTIONS OF PER CAPITA ROAD LENGTH 0 .7 O ....... ............. ....................................................................................... ............. ................... . . . 0.60 0.50 t 0.40 20.30_. _-:---_____ _., ____._.._ 1 0.20 0.10 0.00 I l I 1995 2000 2005 2010 2015 2020 Year -..Status quo.....Towardtheworld average The rld aweage] 74 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions As the estimates in Table 7 indicate, total urban roads for the three city prototypes will increase dramatically during the next 25 years. The total increase in urban road length for each prototype can be decomposed into three contributing factors: (a) urban spatial expansion associated with population growth, (b) urban spatial expansion associated with dedensification, and (c) the joint effect of population growth and dedensification. In the first prototype, all new roads are required by the urban expansion due to population growth, because of the assumed fixed density. In the second prototype, 44 percent of the new roads are required by population growth, 34 percent by dedensification, and 22 percent by the joint effect. In the third prototype, 40 percent by population growth, 47 percent by dedensification, and 13 percent by the joint effect. Clearly, many assumptions are made in obtaining the estimates shown in Table 7, and they cannot be taken as a guide to road construction in any specific city. However, they do provide a useful measure of the scale of road construction needed in Chinese cities in coming years, mostly required by the expansion and dedensification of the existing urban areas. With increased awareness of motorization needs, the designs for new road layouts, both in the existing city where sites are cleared for development as well as in the newly added urban areas, will follow modem design standards, so that the proportion of urban space allocated to roads will gradually increase. Urban Expressway Plans Of particular interest in Table 7 is the proportion of road length dedicated to expressway construction. If Chinese cities follow the pattern of the rest of the world, the entire expressway "allocation" for most cities might be taken up by just one ring road and one or two radial routes, with perhaps an additional ring in the largest cities. Yet, it is not uncommon now to see Chinese masterplans for major networks of high-performance urban expressways comprising two to four ring roads, linked by radial routes extended to the planned new development areas, and even networks of expressways criss-crossing the central areas within the planned Inner Ring, all superimposed on the existing urban area street network. Such plans go well beyond the scale of expressway construction in existing developed motorized cities across the world. Why have existing motorized cities stopped short of such elaborate expressway plans? In part, it is because of the huge costs of construction of urban expressways, especially in the city centers where land is costly. It is also because the intrusion of these roads disrupts communities and builds up local resistance. Few new expressways have been built in Western city centers in recent years, largely for this reason. Both of these factors will probably contribute to a scaling back of the Chinese city expressway masterplans. In addition, though, developed cities no longer favor urban expressway systems because of increasing doubts on their efficiency to relieve traffic congestion. In other words, the notion that we can build our way out of congestion is increasingly questioned. Can We Build out of Congestion? Plans for extensive urban expressway networks are conceived in a determination to eliminate congestion. City leaders envisage high-speed roads passing over the old congested street system, ensuring quick and easy access to all parts of the city. Does this work? Can we Stephen Stares and Liu Zhi 75 build out of congestion? If international experience of urban road construction spanning more than half a century can be any indication, the answer to these questions is a quite definite "no." Building roads is only part of the solution to the complicated urban transport problem. Even in US cities with the most advanced highway system, such as Los Angeles, traffic congestion in the city-center area and some suburban expressways is chronic and still growing. The underlying reason for this is best stated by what is known as the Downs Law,24 which can be summarized as follows: new highway construction along a traffic corridor reduces travel impedance, but it also induces traffic from other corridors and other modes, often resulting in a return to previous congestion levels. Moreover, latent demand created by the previous shortage of road capacity also surfaces and swamps the new highway capacity. Downs made this observation in the early 1960s and part of the reason for the phenomenon also could be attributed to the rapid growth of car ownership in the US urban areas during that time. Nonetheless, for many places to this day, his interpretation that new capacity attracts new demand still holds true, and the equilibrium between capacity and traffic is still far from reality. Downs Law will certainly hold for many Chinese cities for an extended period of time. Even if the municipal governments substantially increase their efforts in new road construction, traffic congestion will remain as long as incomes, population, and vehicle ownership continue to grow. Under this dynamic circumstance, building more roads makes sense only in preventing traffic congestion from getting worse. Theoretically, continuing road construction would eventually satisfy all demands including those attracted by the new capacity, but achieving this goal may not be economically wise because the cost of providing roads may far exceed the benefit of eliminating congestion. This is also the reason why we have to look for other more economically viable options, such as traffic demand management. Conclusions on Urban Road Construction It is concluded that road construction will be an important feature in the development of Chinese cities, in part to match the expansion and dedensification of the existing urban areas, and in part to make good the current underprovision of road space. However, some of the more elaborate plans for urban expressway networks seem unrealistic, both in terms of what can be afforded (in cost and resettlement), and in terms of their impact on traffic congestion. International experience clearly shows that road construction on its own provides no solution to traffic congestion except in the very short term. Further, the need for expressway network should be analyzed in the context of a sustainable and budgeted transport strategy. Each separate section of the network should be subject to detailed economic, technical and environmental feasibility analyses. This is further discussed in Theme Paper 12, "Shaping the Future: The Role of Urban Transport Planning." 24 Anthony Downs (1962), "The Law of Peak Hour Expressway Congestion," Traffic Quarterly, Vol. 16. 76 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Conclusions on the Supply Management Approach Several ways of improving the efficiency of the street system have been discussed, and the need for some new road construction identified. However, Downs Law-that the demand for transport will often exceed supply of road space under a dynamic situation-intervenes. It was introduced in the context of new highway construction, but in reality applies to all methods of improving the efficiency of the road system. The overwhelming conclusion from all experience of motorization to date is that if the road system is made more efficient, or of greater capacity, yet more motorists will be attracted to using it until congestion again intervenes. Since most transport improvements are incremental, the effect is that any new capacity gained is swallowed up immediately by new traffic, seeming to make all efforts (and particularly small-scale efforts) in vain. Hence, the conclusion is that road transport demand will always exceed the supply of road space, no matter how well managed and how efficiently used. Therefore, municipalities have no alternative but to consider ways of controlling and managing the demand for road space. This is discussed in the next section. E. DEALING WITH TRAFFIC CONGESTION-MANAGING TRANSPORT DEMAND This section looks at policies for managing the growth in transport demand. Available policies are listed in Figure 9, grouped into two main classes. One set of policies attempts to deal with the problem at source and discourages or limits the ownership of vehicles. These options are classified under Vehicle Ownership Controls. The main alternative is to permit wider vehicle ownership but then control the use made of vehicles; these policies are classified under Vehicle Use Controls. Within these two approaches, both pricing measures and regulatory measures can be contemplated.25 Before discussing these policies in detail, a comment is made on the potential function of prices in controlling congestion. Congestion Pricing The price of transport, whether in the form of costs for operating a vehicle, or the fare paid to use a public carrier, profoundly affect the volume of transport demand. Higher costs and fares result in lower demand. Since the fundamental cause of traffic congestion is that drivers are responsible for only a fraction of the costs that they actually cause (see previous Box 1), thus encouraging wasteful journeys, pricing as a means of managing traffic demand has many merits. Economists argue that if drivers could be confronted with the full costs of their journey-their own costs plus the costs they impose on other vehicle operations-their decision to make the journey could well be modified. Thus has arisen the concept of a congestion tax imposed on traffic. If the tax is set at the right level, marginal journeys would be deterred while essential journeys would continue to be made. 25 Three documents were consulted in preparing this section: Hong Kong Second Comprehensive Transport Study, op cit.; Methods of Traffic Limitation in Urban Areas, J. Michael Thomson, OECD, Paris, September 1972; and Congestion Charging Mechanisms for Roads: An Evaluation of Current Practice, Timothy D. Hau, World Bank Transport Policy Research Working Paper WPS 1071, Washington, December 1992. Stephen Stares and Liu Zhi 77 What determines whether a journey is marginal or essential? Economists argue that this must be decided by the price an individual is willing to pay. Those who are willing to pay at least the full cost of a journey-their own costs, plus the costs imposed on others-can be deemed essential. Those who are unwilling to pay the full costs must be considered marginal. In theory, it is possible to estimate a congestion tax where the resulting traffic demand is in balance with the capacity of the road system. Set too high and roads will be wastefully underutilized; set too low and roads will remain clogged with traffic. FIGURE 9: MANAGEMENT OF TRANSPORT DEMAND | POLICY MEASURES VEHICLE VEHICLE OWNERSHIP USAGE CONTROLS CONTROLS NONPRICING ACCESS BASIC PARKING AREA CONTROL PRICING CONTROLS PRICING Vehicle Taxation Traffic Management Fuel/Tire Public Tolled -Vehicle MTaxes -On-Street -Cordons Quotas -Areawide Bans Trunk -Public -Supplementary Garaging -Odd-Even Road Tolls Off-Street Licenses Requirements Plate Bans _-Route Off-tPrivate Road SVehicle Restricts/Design. Off-Street Pricing Standards -Planned Delays Operating -Permits In practice, determination of such a tax is difficult. To be effective, the tax must vary by time of day-higher during the morning peak, much lower for a quiet Sunday afternoon-and by location-central business districts require a higher tax than rural areas. Only the most sophisticated transport demand management policies seek to achieve such elaborate road use pricing. However, many policies utilize pricing devices to achieve the objectives of controlling excess traffic demand, usually complementing nonpricing traffic allocation measures such as quotas or road space allocation to specific modes. Congestion pricing is only one aspect of transport pricing. For further details, and a discussion about the relevance of pricing policies in China, see Theme Paper 10, "Shaping the Future: Getting Prices Right." 78 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Controls On Vehicle Ownership Vehicle ownership controls seek to limit the size of the vehicle fleet by either fiscal or regulatory means, in this way restricting the number of vehicles available to use the road system. Several approaches are possible. Vehicle Ownership Taxation Taxes are commonly raised on vehicles both at the time of vehicle purchase, and thereafter annually or monthly for all vehicles maintainied in use. These taxes and fees are independent of the amounlt of usage-they apply equall) to all vehicles, whetlher they are used every day or once a month. The principal function of the taxes is to raise Government revenue, but they also have the effect of making the ownership of vehicles more expensive, thereby putting it out of the reach of some. Additional vehicle ownership-related purchase fees, annual licensing fees, and residential parking fees can be imposed with the objective of making the ownership of a vehicle even more expensive and thereby further restricting the total size of the vehicle fleet. This policy has been used very successfully in both Hong Kong (see Box 3) and Singapore. Vehicle ownership taxation is a fairly crude instrument for control of congestion. While they have the effect of restricting the total number of licensed vehicles, thereby reducing overall traffic levels, those who pay the necessary ownership fees can then use the licensed vehicles when and where they like, so that congested roads are not directly targeted. China, in common with other countries at an early stage of economic development, imposes relatively high vehicle taxes on imported vehicles to restrain the volume of imports, but quite low taxes on domestically manufactured vehicles in order to encourage local industry. There is strong pressure from the vehicle manufacturers in China to maintain this system into the future. Vehicle Quotas Vehicle quotas can be used to restrict the growth of the vehicle fleet to a fixed number of vehicles per year. Quotas are currently used widely in Chinese cities to control the number of motorcycles, taxis, and enterprise-owned vehicles. Strictly enforced quotas can be effective in limiting vehicle fleet size, but some care is needed in application. There are several problem areas. First, there is the danger of restricting a particular vehicle type unjustifiably, thereby forgoing the benefits of its wider use. This is possibly the case with motorcycles, which have been restricted recently in several Chinese cities, but with little evidence of serious analysis of the costs and benefits of wider motorcycle ownership. Second, quotas applied by area can be circumvented by registering the vehicle with a friend or relative living outside the area, and might have to be supplemented by some additional usage regulations to prevent vehicles registered outside the designated area from coming in. This has its own problems as discussed later under vehicle usage controls. Third, if demand is high but the quota is low, licenses within the annual quota become very valuable. Some cities auction licenses and so gain this increase in value, but this requires an effective administrative Stephen Stares and Liu Zhi 79 system. The alternative of allocating licenses to customers at the normal prequota prices could be quite arbitrary, with opportunities for favoritism or, worse, corruption. Almost inevitably, a black market in licenses at higher-than-official prices would become established. Singapore successfully combines quotas and pricing to restrict the growth of the motor vehicle fleet to 3 percent per year. Purchasers of new vehicles must first obtain a Certificate of Entitlement, of which a fixed number are auctioned to the highest bidder each month. This, combined with other duties, taxes and fees, increases the cost of cars to about five times their CIF value (all are imported). Box 3: IMPACT OF VEHICLE TAXES ON RATE OF OWNERSHIP IN HONG KONG The size of vehicle taxes can dramatically affect the size of the vehicle fleet, as illustrated by the history of tax changes on private cars and vans in Hong Kong shown in the Box Figure. A large increase was imposed in 1973 when vehicle growth appeared to be getting out of hand. When the effect of the increase wore off and growth resumed, even bigger increases were imposed in 1982 when the First Registration Tax (FRT, effectively the purchase tax) was doubled and annual license fees tripled. The 1982 increase started a decline in the car fleet that was reversed only in 1987. Currently, the FRT is approximately 90 to 120 percent of CIF value, and annual license fees range between HK$3,899 and HK$5,800 depending on engine size. Box FIGURE: VEHICLE OWNERSHIP GROWTH IN HONG KONG THOUSANDS (END-YEAR) 300 -_ _ _ _ _ ! - GOODS -PRIVATE CARS 250 -- 200 TAX INCREAS 150 TAX INCREASE TAX INCREASE 100 50 O t;L I - -- -L-L+, 1 -_4-i2| 1.L1 - 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 One impact of the 1982 tax increases on private cars was to increase demand for personal use of light goods vehicles-vans and pickups-which were then taxed much more lightly. The growth in the light vehicle fleet was stemmed in 1991 by increasing the FRT on light goods vehicles from 15 percent of CIF value to 90 percent. Since 1987, the vehicle fleet has again grown rapidly, despite adjustments to annual license fees to keep pace with inflation. It was recently suggested that a further vehicle tax increase is now required, but the size of increase would have to be very substantial to have any significant effect, and there are concerns about its acceptability. 80 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Garaging Requirements A widespread practice in urban Japan, as well as in at least some Chinese cities, is to restrict vehicle ownership to those who have off-street parking places. In Japan, proof of parking place must be deposited at vehicle licensing time, consisting of maps and letters of authority. Thus, it is relatively simple for the authorities to make spot checks on such claims. The main aim of establishing garaging requirements is to ensure that residential streets are not cluttered with parked vehicles. The effective restriction on total vehicles that results is a by-product, but probably very effective in high-density city centers. Vehicle Standards and Classification High vehicle standards enforced by annual inspections are primarily aimed at enhancing vehicle safety and reducing motor vehicle emissions, and are practiced in many countries. However, higher vehicle standards also effectively influence vehicle ownership by raising the costs of vehicle maintenance. It was thought that the introduction of more rigorous vehicle inspections in Hong Kong in 1986 was responsible for extending for a further year the downward trend in licensed vehicles initially caused by the car taxation increases of 1982. However, in terms of overall control of the size of the vehicle fleet, this is essentially a supplementary measure. Conclusions on Vehicle Ownership Controls A major disadvantage of vehicle ownership controls as a means for limiting the use of motor vehicles is that it is administratively complex, perhaps impossibly so, to differentiate between urban and rural ownership. Without differentiating these ownership classes, rural vehicle ownership would have to be maintained at the same low levels necessary to prevent urban congestion, which is inefficient. For this reason, vehicle ownership restraint to limit urban congestion has been practiced only in the city states of Hong Kong and Singapore. Extensions of vehicle ownership controls in China could form part of an overall package of measures to control the growth in vehicle usage and generate revenues for infrastructure improvement, but could not be recommended as the sole policy. Controls on vehicle use seem more appropriate for China, as discussed in the next section. Controls on Vehicle Use Controls on vehicle use offer the opportunity to be more discriminating in the management of traffic demand. A large number of methods are available, which vary greatly in their sophistication. For ease of discussion, they are grouped into four overall classifications: nonpricing access controls, basic pricing, parking controls, and area pricing; but in reality there are many overlaps. Nonpricing Access Controls A widely used method of eliminating unwanted traffic is simply to ban vehicle use. The scope of such bans varies greatly, ranging from time-based restrictions on selected vehicles on Stephen Stares and Liu Zhi 81 parts of some streets, to complete and permanent areawide bans. A range of such measures are discussed below. Traffic Management Measures. In Section D of this paper, traffic management measures were discussed in the context of making the street network more efficient and increasing vehicle throughput. It is also possible to deliberately restrict vehicle usage by, for example, narrowing junctions, creating one-way traffic schemes, widening footpaths, and even creating pedestrian precincts out of streets previously open to traffic. This might be done to protect a residential area from through traffic, or to create a more pleasant environment in a central area. Such measures are now termed traffic-calming and are complementary to the earlier efficiency-based traffic management measures considered. Used in combination, traffic can move more efficiently on improved through routes while being restricted from inappropriate areas. Of course. by allocating road space to particular users-traffic lanes or entire roads designated for bicycle or bus use-even efficiency traffic management measures can have the effect of restraining the use of other motor vehicles. Traffic-calming schemes have been pursued vigorously in the older cities of Europe where the charm and scale of the sometimes ancient street pattern has been swamped by excessive use by cars for which they were not designed. However, they also make sense in any residential and shopping area dominated by pedestrians. One Chinese example is the set of restrictions imposed along Nanjing Road, the premier shopping street in Shanghai, where vehicle access is now physically restricted during periods of peak pedestrian activity. Traffic restrictions of this nature are often opposed by commercial forces, which equate restrictions on traffic access with restrictions in trade, even though the opposite effect can be demonstrated. It sometimes takes an unusual event to convince people of the validity of traffic restrictions. For example, a bombing incident in the heart of the City of London in the early 1990s resulted in widespread street closures for many months; these have proved so popular that they are now being made permanent. Areawide Bans. Bans on certain types of vehicles entering the city center during daytime hours are quite common. Many European cities impose such bans on private cars. For example, Milan in Italy bans private cars from the city center between 7:30 am and 4:30 pm. In Hong Kong, goods vehicles of any type are banned from the Mid-Levels, a major residential area, during the morning and evening peak hours when passenger vehicle flows are greatest. In China, access to the center of large cities by heavy goods vehicles is generally restricted to nighttime hours. There has also been some discussion of restricting the use of vehicles that are not registered in the city, and this is already practiced with motorcycles in some cities. This selective approach costs more to administer, and is not particularly logical; it is by no means clear that interaction with a neighboring city should have lower priority than the free use of roads by residents. Extensive use of such discriminatory methods against nonresidents could constitute a barrier to free trade and economic growth, and could also lead to retaliatory regulations in the neighboring cities. The particular issue of goods vehicle controls is discussed in more detail later. 82 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Odd-Even Number Plates. Several cities, including Athens, Lagos, and Mexico City, impose restrictions on the days on which selected vehicles can operate, based on the date and the number plate-vehicles with even-number plates operate on even dates, and with odd-number plates on odd dates. This system was also applied in Seoul, Korea during the period of the 1988 Olympic Games to ensure easier traffic conditions for visitors. It is currently applied in some larger Chinese cities as a means of limiting the access of light goods vehicles to the city center. While odd-even number plate control is a simple system to set up, it is essentially arbitrary in operation and open to fraud. In Mexico City, where its effectiveness in reducing the levels of motor vehicle pollution has been studied, it was concluded that the scheme's effectiveness diminished over time.26 As a short-term measure, such as the Seoul application, the scheme has merit, but other methods are preferable over the long term. Route Restrictions, or Route Designation. An alternative to areawide bans is either to prohibit certain types of vehicles (such as heavy goods vehicles) from some classes of road, or to designate specific routes for such vehicles and signpost them clearly. This recognizes both the economic value of such vehicles being able to penetrate to all parts of the city, but also their unsuitability for operation on all parts of the network. Planned Delays. Some schemes have been devised to impose regulated delays on traffic with the hoped-for advantage that some drivers would switch to public transport while others could plan journeys with more certainty. A major scheme of this nature, known as the Zones and Collar Scheme, was tried in Nottingham in the United Kingdom in the early 1970s. Traffic leaving some large housing estates, and traffic approaching the city center, were held up at traffic lights with a very low proportion of green time. Car parks were provided at the traffic-light locations on the main roads with free bus services to the city center. At all traffic-light barriers, buses could bypass the lights. The scheme did not succeed, partly because the main roads could not accommodate the long traffic queues that developed, and partly because many motorists refused to obey the long delays at traffic signals at the exit to the housing estates. The hoped-for switch to public transport did not take place and overall economic benefits were estimated to be negative. Operating Permits. Car operating permits issued on the basis of need have been suggested for use in London, but not implemented. In reality, it is very hard to define need adequately-doctors on call and permits for the handicapped are two categories that may be accepted by most but, for the most part, it is impossible to define watertight criteria that could be generally accepted or easily administered. More on Goods Vehicle Traffic Controls. Restricting daytime access of heavy goods vehicles to the city center is widely implemented in many Chinese cities. Because goods vehicle transport is highly associated with urban economic growth, this practice deserves careful evaluation. 26 Rationing Can Backfire: The "Day Without a Car" in Mexico City, Gunnar S. Eskeland and Tarhan Feyzioglu, World Bank Policy Research Working Paper No. 1554, Washington, DC, December 1995. Stephen Stares and Liu Zhi 83 In general, there are four types of road freight movements in cities: import, export, transit, and intraurban movements.27 Historically, most Chinese cities have long sought to minimize these freight movements through urban master planning. In particular, facilities that handle large volumes of freight, such as warehouses, heavy industrial plants, and rail marshaling yards, are required to locate at convenient locations near the city-fringe areas, with proper separation from residential and other business functions. The urban master planning principle also requires large cities to provide bypass roads for transit traffic. However, the rapid urban expansion that occurred in many large cities during the last 10 to 15 years has dramatically changed the relative locations of these facilities. Locations on1 the city fringe in the past are now part of the main urban area, and former bypasses now function as urban roads. Industrial relocation and new bypass construction often lag far behind the rapid growth of urban areas. As a result, large volumes of goods vehicle traffic remain in the much expanded city-center area. The current urban land-use changes, particularly residential dedensification and relocation of manufacturing firms to the outlying areas, if continued would have substantial impact on the future pattern of freight movements. This land-use decentralization process will be further facilitated by economic reforms (specifically corporatization and privatization) in the wholesale and retail service industries, which will give firms more freedom in the choice of location. It can be expected that manufacturing firm relocation will shift more import and export freight movements to the city fringe area. Part of the intracity freight, such as those for population-serving retail services, will also decentralize with residential dedensification. The freight relating to central-area retail services may remain. In sum, as current trends in land-use change continue, urban freight movements in Chinese cities will be increasingly removed from the city-center area, but these more decentralized freight movements will require suburban road facilities that provide faster mobility and more widespread access. These trends should be recognized in urban master planning and transport strategic planning. However, land-use changes will take a long time to be effective, so goods vehicle movements will likely remain an important part of city-center traffic for some time to come. It is understandable that many large cities currently choose to control heavy goods vehicle traffic in their central areas. However, improperly planned goods vehicle bans may do more harm than good, because goods vehicle traffic is so closely related to the functioning of urban economy. There are some opportunities for time-based restrictions on goods vehicles, since the peak hours for passenger traffic and goods traffic are normally different. For example in Jinan, peaks for passenger vehicle traffic are during 8:00 to 9:00 am and 16:00 to 17:00 pm, while those for goods vehicle traffic are during 9:00 to 10:00 am and 14:00 to 15:00 pm. The use of urban roads by goods vehicles during the off-peak hours for passenger vehicles makes economic use of otherwise underutilized road capacity, and therefore should be encouraged. There is also a good case for confining goods vehicle to selected designated routes as discussed above. The outright ban of goods vehicle traffic during the daytime, as implemented in many large cities, should be reevaluated. 27 The first two types involve the carriage of freights into or out of the city. Transit movements are those intercity freights passing through on urban roads. The intraurban freight movements are those both generated from and destined to locations within the city. 84 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Conclusions on Nonpricing Access Controls. Nonpricing access controls have their place in protecting sensitive areas, or in better allocating road space among road users. Their nonpricing nature means that they should only be imposed after careful study to be sure that all the costs and benefits are carefully weighed. Basic Pricing Fuel and Related Car Use Taxes. Virtually all countries impose duties and taxes on fuel as a general revenue-raising measure. Increasing these duties is an administratively simple traffic control measure, directly affecting the cost of using vehicles and thereby reducing usage. Other car consumables, such as spare parts and tires, can also be taxed for the same purpose, although parts and tire taxes can have adverse effects on vehicle safety by discouraging proper vehicle maintenance. The disadvantage of the fuel tax as a specifically urban traffic control measure to reduce congestion is that it acts across the board, hitting off-peak motoring in uncongested areas almost as much as motoring on the busiest roads. Users of congested roads pay rather more per kilometer, since fuel consumption goes up as speeds decline with congestion, but rural and off- peak motorists would still pay at least two-thirds of the tax paid by the urban peak-hour motorist. Differential fuel taxes, higher in urban areas, lower in rural areas, remains a possibility to curb specifically urban road use, but would be difficult to administer and would almost certainly encourage uneconomic journeys from urban to rural areas simply to obtain cheaper lower-taxed fuel. As an urban traffic control measure, therefore, fuel taxes are only a little better than vehicle ownership taxes. Higher fuel taxes tend to encourage the use of smaller, more fuel-efficient vehicles, which is useful from the resource conservation and environmental pollution standpoints. Higher taxes on gasoline can also encourage use of diesel-engine vehicles, which might not be so beneficial. This latter is an important consideration in China where diesel engines are widely used in agricultural vehicles and waterway vessels, complicating any decisions to raise diesel fuel taxes. Overall, however, as discussed in Theme Paper 10, "Shaping the Future: Getting Prices Right," fuel taxes are very low in China compared to most other countries in the world and there are excellent arguments for increasing them from current levels. The impact of curbing excessive vehicle usage would be a useful side benefit. Trunk Road Tolls. In many countries, including China, tolls are commonly charged on expressways, bridges and tunnels in both urban and rural areas in order to collect revenues for financing construction and maintenance. Tolls also affect traffic volumes using the tolled facilities, and can be used to control the volume of traffic demand, as discussed in Box 4. The toll principle can be extended to cover other trunk roads in order to control the traffic levels and ensure congestion-free conditions for longer journeys. Toll systems can take two forms: open or closed. Open toll systems require toll plazas at intervals along the length of the road, with the disadvantage that traffic could bypass tolled sections by returning to the street system for sections of the journey. Closed toll systems require toll barriers at all entrances and exits to the toll road, to ensure that all users of the road pay for the full length of the journey on Stephen Stares and Liu Zhi 85 the road; the disadvantage is greater land take and the generally higher cost to build and operate such a system. There are two major issues on urban trunk road tolling. First, land required for toll plazas in urban areas is normally very expensive, although use of prepaid tolls combined with automatic vehicle detection can greatly reduce the land required. Second, the traffic deterred by the tolls must continue to use the existing street system. To control congestion on existing streets, other traffic management policies are still required. Tolls as a means of controlling traffic demand in urban areas are most effective when the tolled facilities cross a natural barrier with no parallel nontolled routes. The cross-harbor toll tunnels in Hong Kong and the river bridges in Guangzhou are good examples. They are less effective for overall urban traffic control purposes where there are parallel routes of significant capacity, although they remain an option for raising financing for new trunk roads. Tolls are discussed further in a later section in connection with cordon tolling. Box 4: IMPACT OF TOLLS ON TRAFFIC DEMAND The use of tolls to influence traffic demand is shown in the Box Figure, which shows the impact of a doubling of tunnel tolls on the Hong Kong Cross-Harbor Tunnel in 1984 (in US prices, from about 70¢ to $1.30). Car and taxi traffic was reduced by about 15 percent and light goods vehicle traffic by about 13 percent-medium and heavy goods were almost unaffected. Traffic growth since then replaced those traffic reductions, but three years of improved operations were gained. Box FIGURE: CROSS-HARBOR TRAFFIC VOLUMES, HONG KONG VEHICLES PER DAY (THOUS) 115 - ~ --- ---- -- -- TUNNEL TOLL DOUBLED 1 10 \ 905 - -- - - - - - - i - 95 90 I I 1982 1983 1984 1985 1986 Guangzhou Municipality also imposed differential tolls in 1990 to adjust traffic levels between five river bridges. Two new bridges with tolls built outside the main city center in the 1980s supplemented three older bridges inside the city center that were untolled. Finding that traffic on the two new tolled bridges was very light, they imposed tolls on all five bridges, thereby shifting traffic to the new bridges. Overall, traffic on the old bridges reduced by 25 percent, and overall cross-river travel by 12 percent. 86 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Parking Controls The areas of highest traffic congestion tend to be major business and shopping districts. Therefore, controls on vehicle parking can be used to limit the number of trips made to these districts by car. Controls can be imposed in three main ways: by quotas on the number of parking places, by pricing the use of parking places, or by specifying rigorous criteria for parking garage provision that are difficult to meet.28 The three approaches all have weaknesses. Reliance on nonpriced parking quotas effectively provides a substantial subsidy to motorists, and forgoes the substantial revenues that can make parking provision self-financing. On the other hand, reliance on purely pricing controls is vulnerable to distortions in the market. This is particularly true where a true market economy is not yet fully established and where hidden subsidies could finance parking oversupply. The third method of defining and enforcing rigorous criteria for parking provision requires considerable sophistication on the part of city officials and technical officers, and could not work without this. Hence, many cities rely on a mixture of all three methods. (See Box 5 for examples.) BOx 5: PARKING POLICIES In Hong Kong, the quantity, location and prices for public parking provision are specified by the Government, with prices set to achieve an average 15 percent availability rate as measured in regular surveys. The quantity of private parking provision is guided by parking provision standards that vary by district and class of building. Developers of new buildings are required to demonstrate the traffic acceptability of their project by means of a detailed traffic impact statement, which considers parking provision among other things. This provides a basis for planning permission, including any departure from the parking provision standards. Bangkok is an example of a city with chronic traffic congestion and poor public transport, where building developers have been permitted to provide for massive vehicle usage with little or no control. New buildings typically provide 8 to 10 floors of parking, greatly increasing the traffic load on surrounding streets. It is reported that simply departing the parking lot during daytime can take up to an hour. Parking controls, as a means of limiting traffic demand, suffer from the key weakness that through traffic, which does not need to park in the targeted district, is unaffected. This has been summarized as "the inherent problem of the parking policy of attempting to control moving vehicles by charging stationary ones." Parking controls can also induce additional circulation travel searching for an available parking place, and delivery/collection travel where a driver will enter the controlled area to drop off or pick up passengers. For these reasons, many economists compare parking controls unfavorably to more sophisticated area pricing schemes (considered below). However, area pricing still remains a largely theoretical concept, while parking controls, imperfect though they 28 For example, Chicago did not permit parking garages to exit onto streets with flows greater than 25,000 vehicles per day, which effectively limited the number of such garages in the central area. For detail, see Methods of Traffic Limitation in Urban Areas, op. cit. Stephen Stares and Liu Zhi 87 are, are widely understood and accepted. Also, lack of parking controls can seriously undermine other traffic control or traffic provision measures. Singapore, the only city with any history of area pricing, regards parking controls as a necessary and complementary measure to their area pricing scheme. Public On-Street Parking. On the basis that streets are principally intended for moving traffic, on-street parking is normally restricted to short-duration visits, typically between 15 minutes and 2 hours. On-street parking bays are clearly marked and located where least intrusive on the traffic flow. Availability of parking places can vary throughout the day, with some places made unavailable during the peak traffic hours. On-street parking can be free or charged, the latter requiring either permanent attendants (as in most charged parking in China) or the installation of parking meters or parking ticket machines. Whether free or charged, on-street parking requires enforcement by traffic wardens. Generally, in developed cities with intensive demand for vehicle parking, the revenues from charged parking more than cover the costs of equipment and enforcement. Public Off-Street Parking. Public off-street parking provides for longer-duration parking. Fees should cover at least the costs of carpark provision, operation and maintenance. In fact, parking is generally a profitable business in developed city centers; the market can bear higher prices and potential demand generally exceeds supply. Hence, an increasing number of cities leave the management, and often the provision, of public off-street parking to the private sector, and have privatized previous Government-run carpark operations. Hong Kong privatized such operations in 1984. Many cities use the availability and price of off-street parking as the main means of controlling traffic access to the city center, applying different charging rates depending on parking duration and time of arrival to obtain a balance between long-term commuting parking, and short-term parking for other purposes (business, shopping, etc.). The Government pricing policy can still be reflected by privatized carpark operations as long as there are clear franchising agreements. New public parking provision can be accepted in places where the road system is of adequate capacity, although the necessary analysis to back up this policy is demanding. Private Off-Street Parking. Private off-street parking is not usually charged directly; staff or clients are simply allocated the use of parking spaces at the owner's discretion. The privileged users of these spaces, therefore, park free and are unaffected by a public parking policy. Authorities normally attempt to influence private off-street parking supply by specifying parking provision standards. In the early stages of motorization, developers of new commercial buildings are often reluctant to provide parking places, seeing this as an unprofitable use of valuable real estate. In an effort to ensure that motorists visiting such buildings do not clutter the surrounding road space with parked vehicles, authorities can insist on a minimum parking provision. This is currently the situation in most large Chinese cities, and remains the situation in Hong Kong where the private car is used for only about 15 percent of all journeys. After motorization takes hold, and especially if public transport services are not well developed or are curtailed through road congestion, building developers tend to switch, seeing provision of 88 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions parking as a key attraction for potential tenants. At this stage, many city authorities will then switch to specifying a maximum parking provision to discourage overprovision. Goods Vehicle Parking. In developing a parking policy for both on- and off-street parking, it is necessary to consider the needs of goods vehicles. Many businesses depend on regular deliveries from several suppliers, which requires access throughout most of the day. The nature of goods deliveries requires parking at the site since heavy loads must be moved. Off- street parking is desirable, but often not available in older and smaller establishments, so that on- street parking near the premises is a necessity. This needs careful planning and regulation, but is normally possible given that the peak period for goods vehicle use in cities normally does not coincide with peak car usage. Conclusions on Parking Controls. Control of parking, despite its imperfections, should be a key component of a policy to control vehicle usage. In China's mixed economy, pricing mechanisms need to be supplemented by public parking quotas and mandatory standards for private parking provision. There will always be pressures to relax parking provision standards in order to attract important developers to their area, but the Bangkok example demonstrates the need to place some limits. Probably the key to controlling private off-street parking is to require developers to provide traffic impact statements for all new developments to demonstrate the traffic feasibility of development proposals. Area Pricing Area pricing aims to charge motorists for the use of roads in selected districts, with the highest charges for the most congested districts. Frequent users of busy roads would accumulate high charges, but the more casual motorist, by careful selection of times and routes, could avoid road charges altogether. The practical effect should be less traffic on the busy roads as motorists avoid the higher-priced congested districts. The main difficulty with area pricing is in devising a practical method for collecting charges from the motorist. Three distinct classes of schemes can be identified (although modern methods of electronic vehicle detection and charging are tending to blur the distinctions): Tolled Cordons Supplementary Licenses Electronic Road Pricing Tolled Cordons. A cordon is set up around the area to be charged with toll gates installed at all entry points. Toll charges can be varied by time of day to make peak-hour entry more expensive, and charges can be abandoned altogether at off-peak times, on weekends or on public holidays. This approach has been introduce in several cities in Norway, starting with Bergen in 1986 and followed by Oslo in 1990 and Trondheim in 1991. The systems have been designed, and accepted by the public, as a means of collecting revenues for financing transport infrastructure construction. However, in all cases, traffic has dropped, by 12 percent in Bergen with an $0.80 toll, and by 5 percent in Oslo with a $1.60 toll, the difference in traffic impact having to do with the particular characteristics of each city. Stephen Stares and Liu Zhi 89 A practical disadvantage of such schemes is the land needed for toll booths. The practical problems of finding such land tends to dictate a larger controlled area than desirable for traffic control purposes, and even then would be very difficult to implement in high-density cities. A related disadvantage is that the operation of the toll barriers requires traffic to stop; resulting delays are dependent on the number of toll booths provided, and hence the size of the toll plaza. With high traffic flows, large toll plazas would be required, with up to six toll booths per high-capacity lane. As with toll roads discussed above, both these disadvantages can be reduced with prepayment of tolls combined with automatic vehicle detection technology to detect prepurchased electronic tags. Both Oslo and Trondheim use such methods, but still need to retain some manual toll booths for vehicles (perhaps from other cities) without the necessary tags. Some cities, by accident of geography, could find such schemes more practicable. For example, it has been proposed, but not yet implemented, to control traffic entering and leaving Manhattan Island in New York, taking advantage of the fact that entry to the Island is restricted to a limited number of bridges, tunnels or ferries. Supplementary Licenses. These schemes require cars to display a special prepurchased supplemental license to be able to enter or use roads in a designated area. Enforcement of the scheme depends a little on the criteria for vehicle use. If entry criteria are specified, checkpoints need to be set up on all entry roads. While this is similar to the tolled cordon discussed above, it differs in the key respect that it can be tightly defined around the main congested area of a city, making it more acceptable to absolutely ban all vehicles without a license. Hence, manual toll booths with their expensive land requirements can be completely eliminated. The Singapore area licensing scheme is such a scheme (see Box 6). The Singapore scheme currently does nothing to control movements within the control area after entry. More elaborate schemes have been devised, but not implemented, using different-colored licenses for areas with different congestion levels, thus allowing more subtlety of charging for different levels of congestion in different areas. For example, a high-priced red license may be required for the most congested areas, but cheaper blue and yellow licenses could permit use of roads in other, less critical areas. Many areas would not require supplementary licenses at all. The red, blue and yellow zones would be clearly designated by maps and roadside signs. Enforcement of such a supplementary license scheme would be the main problem, although not insuperable. Such a scheme for Central London, proposed in 1974 but never implemented, estimated that 400 wardens would be required to achieve 90 percent compliance at a cost equal to 6 percent of gross revenues. Electronic Road Pricing. This is the most sophisticated category of area pricing schemes and practical implementation is the goal of many traffic planners worldwide. Whereas previously discussed schemes levy charges by entry to, or presence within, a designated area, electronic road pricing (ERP) aims to charge for actual vehicle usage in congested areas. A key feature is its dependence on fully automatic vehicle detection and charging. By careful setting of detectors and varying charges by time of day, traffic planners could target the heaviest charges on the busiest roads, while leaving uncongested roads charge-free. By this flexibility, ERP offers the most effective way of countering traffic congestion yet conceived. 90 Theme Paper 1: Motorization in Chinese Cities: Issues andActions Box 6: SINGAPORE AREA LICENSING SCHEME The Singapore area licensing scheme introduced in June 1975 requires a special license to enter the central area during peak hours. Gantries have been set up at all entry points to the designated central area, with flashing lights to indicate times when the special license is required to pass the barrier. Traffic officers monitor the traffic stream; offenders are photographed and tape recordings are made noting time of offense. These photographs and recordings are admissible evidence in courts when offenders are prosecuted. One very interesting aspect of the Singapore scheme is the extent to which it has been modified from time to time, based on practical operating experience. The initial scheme covered just the morning peak hours, Monday to Saturday from 7:30 am to 9:30 am, but the period was extended after two months of operation to 10:15 am to reduce problems caused by large numbers of vehicles arriving immediately after the restraint period ended. In June 1989, licenses were introduced for the evening peak, initially set from 4:30 pm to 7:00 pm, but shortened six months later to end at 6:30 pm. In January 1994, the scheme was extended to cover the entire day, 7:30 am to 6:30 pm (3:00 pm on Saturdays), but with lower charges for off-peak use. Charges for the morning peak started out in June 1975 at S$3 per car (about US$2), but were increased to S$4 after six months of operation, and then to S$5 in March 1980. When the all-day scheme took over in 1994, the all-day charge was reduced to S$3, with a S$2 part-day fee for the period 10:15 am to 4:30 pm (10:15 am to 3:00 pm on Saturdays). Some vehicles were initially exempted, including taxis, motorcycles and goods vehicles, but are now charged. High-occupancy vehicles were also initially exempt but the ingenuity of local youth, who filled car seats for a fee to enter the restricted area, defeated this. Public buses and emergency vehicles are now the only exempt vehicles. Company cars are now charged at twice the rate of private cars. The physical extent of the protected area has also been increased from about 5.5 km2 in 1975 to about 7 km today. The Singapore Government has further plans to improve the scheme, most notably by replacing the paper licenses with electronic tags for automatic detection, and might even extend to a full electronic road pricing system. The Singapore scheme, as modified over time, has been very successful. In the simplest terms, traffic volumes in the central area have been reduced so that average speeds in the central area are around 30 km/h. However, the supplemental licensing scheme is only one element of urban transport policy in Singapore, and its success is in large part dependent upon the support of many other strong initiatives in urban transport planning-strong traffic management, good public transport including both buses and metros, effective infrastructure provision to a clear plan including bypasses around the designated area, and effective pricing policies. A further important element is the ability (both technical and legal) to identify and prosecute offenders. All recent proposals would require vehicles traveling in the designated zone to be provided with devices that can be recognized, or accessed, by devices mounted in or near the road. Several alternative schemes are under investigation, differing in both the technology and also the approach to pricing (see Box 7). While much (but by no means all) of the technology for ERP has been demonstrated, the only actual installation to date has been the 1985 ERP pilot project in Hong Kong. This demonstrated the feasibility of the technology, but was not deemed to be acceptable politically. This was partly because of the privacy concern mentioned in Box 7, but also because the public could not be convinced of the value or necessity of the scheme. While some of the objections have been overcome, in theory at least, by improved technology, actual implementation remains a plan in Hong Kong and elsewhere. Conclusions on Area Pricing. Singapore has demonstrated both the workability and benefits of area pricing. Singapore has also demonstrated the virtue of starting with a relatively simple scheme, and then refining and adjusting it to achieve better results. Further, what they have accomplished so far makes very little use of advanced technology, demonstrating that Stephen Stares and Liu Zhi 91 technology need not be the driving force behind effective area pricing, even if it has the potential to make it more efficient. All of this would augur well for initiating area pricing schemes in China. Box 7: ELECTRONIC ROAD PRICING TECHNIQUES A variety of electronic road pricing schemes have been proposed, making use of different technologies, and much experimentation is currently underway, The technical details (for example, should the link between vehicle and roadside monitoring device be by means of optical, infrared, radio, or microwave technology) need not concern us here, but will be critical in finally developing a workable system. Some differences are more crucial for public acceptability, as follows: Off-Vehicle or On-Vehicle Metering. Off-vehicle metering monitors the progress of a vehicle through the congested zone, accumulating road-use charges in a central computer. The motorist is then sent a monthly bill for congestion charges in much the same way as for regular utility bills. This raised objections on privacy when piloted in Hong Kong in the 1980s, since it leaves the possibility for investigators unrelated to the ERP scheme to raid the central computer for information on the whereabouts of any registered vehicle. More recent schemes therefore emphasize on-vehicle metering where a prepaid meter (perhaps an electronic smartcard) mounted on the vehicle is turned on and off by the roadside devices, and must be replaced, or replenished, when completely used up. Point Charging or Continuous Charging. The question here is whether a charge is accumulated for each passage past a fixed roadside point or whether the meter runs continuously in the charged areas with the roadside devices simply turning the meter on and off at entry and exit points. The latter method clearly could provide a better measure of actual usage but, for many motorists, would be alarmingly open ended- a bad traffic jam could be really expensive. The former method would be less rigorous in charging for congested road use, but would give the motorist a much clearer idea of the extent of financial commitment for a particular journey. Fixed Rates or Variable Rates. Some schemes propose fixed and posted rates, varying by zone, while others seek to vary the charging rate according to the actual level of congestion as measured by average road speed measured by detectors. As with point and continuous charging, the latter is more effective but the former is likely to be more understandable and therefore more acceptable. On the other hand, the Singapore scheme was established on a secure basis of strong urban transport management underpinned by a sound understanding and acceptance of market pricing principles. This framework is largely lacking in Chinese cities. In China, the relatively sudden advent of motorization and consequent lack of experience in traffic management, the current dependence on bicycle travel, the lack of efficient public transport systems (either bus or rail), and the uneven application of market pricing principles all complicate the introduction of area pricing in the short term. It remains, however, the most enticing approach to gaining control of motor vehicle congestion. F. INTERNATIONAL EXPERIENCE OF MOTORIZATION This section looks at the actual experience of transport developments in several Southeast Asian cities, all of which are further along in the motorization process than China. As a background to this discussion, the graph in Figure 10 presents the growth of the total vehicle fleet over the period 1960-90 for the main Southeast Asian economies. The line for each country spans the period, starting bottom left at 1960 and finishing top right at 1990. Patterns are quite consistent for the country economies, with a strong correlation between vehicle ownership levels 92 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions and per capita GDP. Korea shows that high growth can be sustained over a 30-year period, and Japan indicates that vehicle fleet growth can continue even at very high GDP per capita levels. However, the two city-states of Hong Kong and Singapore show that the pattern can be modified with high vehicle-taxation policies to restrain vehicle ownership growth. FIGURE 10: EAST ASIA MOTORIZATION, 1960-90 1000.00 1 Japan Singapore 100.00 0 ~~~~~~~~~~~~HK 9 U~~~~~~~Nblaysia , Rl~~~~~~~Pilippines //1/Korea 10.00 a.Ian Indonesia / / ina 10.00 100 1000 10000 100000 GDP Per Capita The significance for China is clear. While motorization is at an early stage in China, Figure 10 indicates that China is starting to follow the East Asia pattern, supporting the projections of China motorization presented in Section B of this paper. This section, therefore, examines recent developments in transport in several major East Asian cities as they are going Stephen Stares and Liu Zhi 93 through the motorization process to see what lessons they might have for Chinese cities embarking on the same process.29 Hong Kong 2 Hong Kong is a small city region of 1,070 km , mostly comprising steep hills and uninhabited islands. The densely developed built-up area covers just 96 km2, much of it reclaimed from the sea. With few natural resources, Hong Kong has thrived on light industry, on financial services, and as a major regional seaport. Strong economic growth has been sustained for many years, averaging around 8 percent per year since the early 1960s. Population roughly doubled from about 3 million in 1960 to nearly 6 million today. Facing increasing road congestion in the mid- 1960s and a rapidly increasing population, Hong Kong embarked on a series of transport and land development studies, both major and minor, which continue to this day. Quantified transport and land-use planning is a hallmark of the Hong Kong planning process that has guided the massive transport infrastructure development program of the past 30 years. Despite an intensive construction program, it was recognized early on that new roads could not keep pace with rising car ownership and that chronic traffic congestion would threaten the viability of the road-based public transport system, which carried around 85 percent of all passenger journeys in the early 1970s. Therefore, a decision was made in 1973 to deter further vehicle ownership by imposing large increases in vehicle taxation; this was reinforced by further increases in vehicle taxes in 1982. These measures were successful (see the figure in Box 3). Supported by other measures, including parking controls and increased tolls on road tunnels, and with strong traffic management to extract maximum efficiency from the available road system, peak-hour urban road speeds in 1986 were maintained at 25 to 28 km/h. However, vehicle growth resumed in 1986, and traffic congestion is again an issue. Average peak-hour speeds have now dropped to at most 22 km/h (1993 measurement). Given high densities of urban development, much located in narrow corridors of reclamation, the early transport studies gave high priority to the construction of a rail-based mass transit system. Construction of the mostly underground Mass Transit Railway (MTR) started in the mid-1970s and three lines have been opened to date. With high ridership (2.2 million passengers/day), the MTR is considered one of the most successful metro systems in the world, and the only one with any chance of recovering all its construction and operating costs from passenger revenues. During the same period, the Kowloon-Canton Railway (KCR) expanded services to the rapidly growing towns of the New Territories located outside the main urban area by electrifying its mainline and constructing a new light rail system (for a total 1.0 million passengers/day). Both MTR and KCR are owned and operated by public corporations. Despite the construction or upgrading of five major rail lines (about 100 km) in the past 15 years, buses, minibuses and taxis (3.5, 1.7, and 1.0 million passengers/day, respectively) 29 The authors were greatly assisted in preparing the city write-ups by Sock-Yong Phang (Singapore), George D. Esguerra (Manila), Rodney J. Stickland (Bangkok), Jason Brooks (Kuala Lumpur), and Bob C.L. Lee (Taipei). 94 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions continue to carry well over half of all public transport passengers. Many buses are now air- conditioned and provide good levels of service, partly protected from road congestion by bus- priority traffic management measures. Preference is given to buses, with the number of minibuses kept fixed and the few additional taxi licenses granted each year (around 170) that are auctioned to the highest bidder. Trams and ferries complete the public transport scene. All these services are provided (profitably) by private companies under Government-regulated franchises. Hong Kong has excellent transport infrastructure and public transport services, a result of many years of intensive planning and investment. It is also a highly competitive system with services available over a range of quality and price. Road and rail construction continues, including works to serve the new airport and provide new links with China, but it is recognized that construction solutions are not sufficient on their own. The major transport challenge continues to be rising vehicle ownership and associated traffic congestion that again threaten the essential bus, minibus and taxi services. Various countermeasures are under consideration, including further increases in vehicle taxation, road pricing (piloted successfully in 1985 but not implemented), and increased tunnel tolls (although complicated by franchising agreements). Seoul, Korea Seoul's urban land-use/transport system has evolved into a model favored by many Chinese city planners and policymakers. A densely developed metropolis, Seoul is the home for 17 million people. The growth of its central city has been physically controlled by a strictly enforced greenbelt, with much new development taking place in high-rise satellite towns scattered outside. These satellite towns are linked to the central city by high-performance railways and highways. The central city itself is served by four metro lines (121 km) and an extensive bus network (422 routes and 8,700 buses). Yet, Seoul is now facing the biggest transport challenge in its history: rapidly growing traffic congestion. In the early years of economic development, all cars were imported and subject to high duties, which put vehicles out of reach for most people. This changed with the development of the domestic car industry, which the Korean Government saw as an engine of the economy and a key export earner. To help establish the industry, the Government initially set low taxes for domestic vehicle purchase and reduced the fuel tax and other vehicle usage fees. Combined with growing incomes, this has produced continued rapid increases in car ownership nationwide. In Seoul, the number of passenger cars has grown by 20 percent per year since 1985. Average auto speeds declined from 33 km/h in 1989 to 20 km/h today, and peak-hour traffic jams are now a daily phenomenon. Recognizing that the central city, with a population density of 18,000 persons per square kilometer, had little room for road capacity expansion, Seoul focused instead on expanding the metro system. However, the metro has not made any appreciable impact on road congestion. Moreover, although it carries 5 million passenger-trips daily, and is very crowded during rush hours, fare revenues do not cover operating expenses (including depreciation). With further subway extensions typically penetrating thinner and thinner markets, increasing subsidies are likely in the future. As in the past, further subway expansion is unlikely to reduce road congestion because continuing growth in road traffic will soon fill spare capacity created by any road users diverted to the metro. Stephen Stares and Liu Zhi 95 The bus system has been the biggest loser to growing traffic congestion. Carrying 10 million passenger-trips per day (40 percent of total person-trips), the system is operated primarily by 90 private businesses under Government regulation of fares and routes. To date, it has not required any operating subsidies and has been regarded as very efficient by international standards. But growing traffic congestion caused serious declines in bus speeds and deterioration of service reliability, resulting in loss of ridership and revenues as passengers turned to private cars and taxis. At the same time, declining bus speeds increased both operating and capital costs. Facing ridership losses (5 percent per year) and cost increases (15 percent per year), Seoul's bus operators demanded action. In response, the Seoul Government recently implemented extensive bus-lane priorities across the central city; these are reported to have improved bus speeds substantially and shifted some car users to buses. While Korea's successful auto industry strategy has brought substantial benefits to the country's economy and the people's well-being, road congestion caused by uncontrolled motorization in urban areas is in danger of eroding these benefits. By opting for subway development and avoiding any direct action to reduce road congestion, the Seoul Government exposed itself to increasing subsidies for metros and deteriorating quality of bus transport, which still remains the largest public transport carrier. The recent actions on bus priorities, which both helped buses and restricted car usage, have substantially improved the situation, but vigilance will be required to cement these gains. Bangkok, Thailand Bangkok, with a population approaching 8 million, is located on low land adjacent to the Chao Phraya River. Until the early twentieth century, its transport was based on the river, Khlongs (canals) and waterways. The development of the city has proceeded in a largely uncoordinated manner, due to the absence of any effective planning controls or strong administrative direction. The road network is characterized by very wide, multilane main roads-many constructed over or along the alignment of abandoned khlongs-and a large number of small local feeder roads (sois) that are narrow, poorly connected and indirect. Overall densities are relatively high in the inner areas, but are significantly lower in the outer areas where development is limited by the absence of an adequate road network. There has been a tremendous increase in the vehicle fleet in Thailand in recent years. New vehicle sales have shown a sustained growth of over 15 percent per year, with a staggering 33 percent per year increase in private car sales over the last 10 years. Half of all vehicle registrations (75 percent for cars) are in Bangkok. Despite these high rates of growth, the absolute levels of vehicle ownership are still relatively low and substantial growth may be expected before saturation is reached. Fuel sales in the region have been increasing by over 9 percent annually, representing a doubling in traffic volumes every eight years. The dominant characteristic of Bangkok's transport system is overwhelming traffic congestion. Within the inner areas, speeds average less than 10 km/h for much of the day, and conditions in the suburbs are rapidly becoming similar. The lack of alternative nonroad-based transport and the ever-sprawling suburban development results in congestion extending upward of 20 km from the center for up to 16 hours per day. 96 Thleme Paper 1: Motorization in Chinese Cities: Issues and Actions With thie exception of very limited local commuter rail services and some boat services 011 the river and khlongs, there are no segregated public transport services in Bangkok. Those without access to cars or motorcycles are dependent on buses, minibuses, taxis and a mix of srmialler public transport modes-including semilegal motorcycle taxis. Buses carry some 6 m-illion passengers per day, but are severely affected by the high levels of traffic congestion and face increasing competition from the ever-increasing number of private cars and motorcycles. 'I'he problems have long been recognized and studied, and many proposals for mass transit schemes and urban expressways have been promoted over the past 20 years. A series of privately funded transit schemes have been proposed, although none of the projects are close to being operational, and construction is well behind schedule on the two schemes on which work has started. Full futiding has yet to be secured for these projects, which are claimed to be self- Finaincin g. Progress on the implementation of urban expressways has been more impressive and i network of tolled routes is now in operation, with more sections planned or under construction. 'iost have beeni implemented using a mix of public and private finance: considerable difficulties have been experienced in securing the finance and the associated contractual arrangements between the operator and the Government. These expressways, however, do little to solve the ueileilying problems and serve to funnel yet more traffic onto the overloaded central street systeni. The maini hope for improving travel conditions lies in the completion of the rail mass transit schemes. Restrictions on private car usage would be desirable, but are considered both i'practical and unacceptable in the Thai context, with the commitment to personal liberty and ndividual freedom. .n"7apore Singapore is a small island city-state with a population of 3 million and a land area of J80 in'. It has no natural resources (except for its strategic location and excellent natural ir-.rbor) and has pursued a successful industrial policy of attracting foreign direct investment to 's ma nufacturing, financial and business services sectors. Economic growth has averaged 8 er cent per year since the early ]960s. 'T'he Government has undertaken long-term land-use and transport planning since the late 1 960s. The state has been responsible for industrial park developments, as well as the construction of large high-rise public housing estates that presently house 86 percent of Singapore's households. It also leases land to the private sector for commercial, residential and trcreationial developments. The island is well served by adequate public transport services and excellent transport infrastructure. Public transport operators are either owned or regulated by the Government. The two- linie 66-kin Mass Rapid Transit rail system, owned and managed by public corporations, was opened in stages between 1987 and 1990. The capital cost was entirely financed by the Government and passenger revenues are expected to cover operating costs only. Further expansion of the MRT is underway or planned, and the Government has also made a commitment to Iink additional areas to the MRT via Light Rail Transit systems. Bus services are provided by two major operators that are profitable, publicly listed companies. Buses enjoy Stephen Stares and Liu Zhi 97 exclusive bus lanes as well as priority right-of-way at designated traffic junctions. Minibuses operate high-quality public transport services between certain housing estates and the dowintovw area during peak hours. Road congestion became a major problem in the early 1970s. Tackling the motorization problem early, the Government implemented an Area Licensing Scheme in 1975 (see B3ox 6). The scheme that started as a morning-peak licensing scheme has been extended to become a whole-day scheme. Road pricing has also been introduced along an expressway and otlier tlIsatc charges include parking fees and fuel taxes. Usage charges are complemenited by strict X ehiclc ownership quota policies that aim to restrict vehicle growth to 3 percent per year. Since V)9t prospective new car owners must first acquire a Certificate of Entitlement (COE) obtained at a monthly tender. The October 1995 tender price of a COE for a mediuLm1-size car (1,001 lo 1,600 cc) was about US$30,000. Other duties, taxes and charges took the average price oft a medium-size car to over US$90,000 in 1995, perhaps five times the CIF value. ([hel-e is nL domestic automobile industry in Singapore so all cars are imported.) The combination of quotas and pricing policies has successfiully reduced roadl conigestio n in Singapore; peak-hour speeds in the restricted zone average 30 km/h. Motor vehicle charges have also proven to be highly revenue-productive, accounting for nearly one-quarter of total tax revenue in recent years, more than adequate to finance expenditure on transport in]rastruLtiIr- However, despite (or perhaps because of) Singapore's success in containinig the imotor car, the Government is now under great pressure to expand vehicle ownership and roacd facilities. A recent poll showed that young Singaporeans have two major aspirations: to live In a biungalowS (currently 86 percent of the population live in high-rise apartmenits), and to own a car (currendly just 1 in 10 own one). For the majority, these must be dreams, but Govermilenit is comilmitied to respond. The 1996 Government White Paper, "A World-Class Land Transport Systeim, pledges to greatly increase road construction (possibly including undergrounid expressways in the central area), to increase car ownership to I in 7, but at the same time to extend vehicle usage controls to maintain free traffic flows-electronic road pricing is scheduled for imiplemllenitationi in 199/. Plans also include continuous improvement of the public transport system and lfurtler decentralization of employment from the central city area to four regional centers. Metro Manila, Philippines Metro Manila, covering an area of 636 km2, is the couLntry's capital region coniprshy eight cities and nine municipalities. With 8.9 million residents, Metro Manila accounts for about 13 percent of the national population, one-third of all urban dwellers, and aboul one-tlir(i of the national economic output. With trade liberalization, the economic upturn in 1990, and Governmenlt's policy to make the price of locally-assembled cars affordable through the "People's Car Progran<," ve.aicle ownership and usage have increased rapidly. The private transport share has riseni hlorn 126 percent in 1980 to the present level of 35 percent. Out of the 80,000-unit anllual sales of the domestic car industry, two-thirds are registered and used in Metro Manila. The growtlh in pria ate vehicles, increasing at the rate of 12 percent per year, has outpaced growth in road capacit and 98 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions traffic congestion has spread to nearly all the arterial roads of Metro Manila. Peak-hour travel speeds are now down to 10 km/h in major traffic corridors. Public transport services are predominantly road-based, consisting largely of privately- owned jeepneys and buses for primary and secondary routes. Short-distance bus transport forms an integral part of Metro Manila's transit system. Under the liberalized regulatory regime, public transport operators, particularly bus companies, have recently increased and modernized their fleet. As early as the 1960s, the strategic transport development plan for Metro Manila called for the introduction of urban rail transit systems to provide trunkline services. Despite financial difficulties, a 15-km Light Rail Transit (LRT) line was constructed in 1984 to serve two of Metro Manila's most-traveled corridors. The Manila LRT system enjoys daily ridership of about 400,000 passengers. Despite the low flat fare of P 6.00 ($0.22) since 1991, the fare box ratio (gross revenue divided by direct operating expenses) has remained high at 1.5, comparable to other urban rail systems in the world. Thus, revenues help offset equipment depreciation, but make little or no contribution to the original construction costs. The LRT line has helped shape the urban development of Metro Manila, triggering redevelopment of the old business districts within the City of Manila and commercialization of areas around the LRT stations. The transport problems of Metro Manila are mainly attributed to the deficiencies in public transport services and the urban road network. The traffic situation is dominated by the sheer number of private cars and the operation of space-consuming jeepneys, with the latter now operating on traditional bus routes. Traffic management measures, despite the use of a computer-controlled signal system, have remained uncoordinated, and suffer from inconsistent traffic enforcement by a multitude of traffic authorities. The end result has been chronic congestion on most roads, which further degrades the level of service of public transport modes, particularly buses. In an attempt to combat traffic congestion, in December 1995 transport authorities adopted an odd-even number plate permit scheme on key routes during peak periods. This scheme has so far proved ineffective, so plans are being made to extend it to cover public transport vehicles, including taxis. However, the hopes of the authorities rest on implementing major road construction projects and three new LRT lines, targeted for opening by 1998. Kuala Lumpur, Malaysia The Federal Territory of Kuala Lumpur provides a natural focus for business and 2 commercial activities undertaken in Malaysia. Covering an area of some 245 km , Kuala Lumpur has a population of some 1.3 million people and also provides a source of employment for residents of neighboring states. Population growth in Kuala Lumpur during the past decade has been rapid, largely as a result of urbanization, and economic growth has paralleled the national GDP growth rate of over 8 percent per year. A key component of the Government industrial strategy has been the development of a national car industry (Proton), and this has greatly reduced costs of car ownership. Since the introduction of the Proton Saga in 1985, growth in passenger car ownership has been rapid, with vehicle registrations increasing at more than 8 percent per year during the past decade. Import duties remain high on all foreign cars. Stephen Stares and Liu Zhi 99 The combined pressures of an increasing population and high car ownership have led to rapid growth in traffic demand within Kuala Lumpur. Traffic congestion is now pronounced within the central area of the city. Traffic police are required to man all major intersections during peak periods and the duration of traffic congestion is spreading throughout the day. Investment in road network construction has focused on the development of effective orbital road systems to relieve traffic demand in the central city areas. Much of this development has been achieved through private-sector involvement and many of the new road schemes will operate as toll roads. Buses carrying some I million passengers per day provide the main public transport service in Kuala Lumpur but have long been viewed as a mode of last resort, reflecting problems of overcrowding, poor travel conditions, and unreliable journey times. The Government has recently embarked on a major reorganization intended to improve bus service standards and promote private-sector investment. They plan to rationalize the role of buses and minibuses and construct new passenger interchange facilities away from the urban center. The Government is also evaluating the potential benefits of introducing bus-priority measures within the central area of the city. Rail facilities provided by Keretapi Tanah Melayu Bhd (KTM) have recently been upgraded, including construction of a second parallel track, track electrification, and construction of new stations. Work was completed in 1995 and has led to improved commuter rail service within the Kuala Lumpur conurbation. There has also been a long-standing commitment by the Government to implement a Light Rapid Transit (LRT) network within Kuala Lumpur. Two main routes (covering some 50 km) are currently being implemented by private-sector companies and will each provide a carrying capacity of some 16,000 passengers per hour. They are scheduled to be completed before the start of the Commonwealth Games in 1998. The implementation of an LRT system will greatly increase the overall capacity of the public transport system and reduce its sensitivity to road congestion. However, it is recognized that the full benefits of the LRT system will only be realized by integration with other rail and bus services. Since the early 1990s, the authorities in Kuala Lumpur have acknowledged that it is not possible to accommodate the demand for personal mobility through new road construction alone. Instead, the authorities have embarked on a series of major initiatives designed to achieve a modal shift of 40 percent from private to public transport by the year 2000. The initial emphasis has been on the development of a fully integrated public transport system, as described above. However, it is also recognized that measures will likely be needed to limit the growth in motor vehicle traffic. Measures currently being considered include revised parking policies, high- occupancy vehicle lanes and the implementation of road pricing. Taipei, Taiwan (China) Taipei has an area of 272 km2 and a population of 2.64 million. The city is surrounded by mountains on three sides and Tansui River in the west. Its metropolitan area has extended westwards over the river to many cities of Taipei County linked by more than 10 bridges. The Central Business District (CBD) is very dense in the middle-west area, and the residential area is located beyond. 100 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions Private vehicles play a major role in Taipei's urban transportation system, with approximately 0.9 million registered motorcycles and 0.42 million registered cars. They carry 30 percent and 25 percent, respectively, of daily passenger travel. The total number is still growing, with more than 10,000 new vehicles per month. There is currently no specific policy for limiting vehicle ownership, in part to help the domestic manufacture of cars and motorcycles. Growing incomes make it easy to own a vehicle and poor public transportation service encourages private vehicle usage. Traffic congestion is considered a serious problem in the CBD and on commuter routes. The average traffic speed is 15 km/h during the peak, 18 km/h during the off-peak. Vehicle parking is another issue in Taipei. Demand exceeds the supply of public off- street parking places, so that roadside parking and illegal parking are very common in the city. This reduces city street capacity and exacerbates congestion. Recently, the City Government decided to provide more incentives to private development of carparks. Buses and taxis are the two major public transportation modes in the city, accounting for about 35 percent of all passenger movements. Currently, 10 bus companies are operating with a total fleet of 3,345 buses and providing 247 service lines inside the city, which link to the neighboring cities in Taipei County. The number of taxis operating in the city is estimated to be more than 50,000, including those registered in Taipei City and neighboring cities. However, the transport capacity provided by public transportation is far below the demand, especially in peak hours, and the service quality is poor. To mitigate this problem, the City Government has drafted many schemes, for example, dedicated bus lanes, on-time bus information, comfort interior design of buses and automatic fare collection system, in an effort to attract more trips from private transportation modes. A major goal of the City Government is to improve the service level of public transportation in order to shift more trips from private transport modes to pubic transportation. A five-line Mass Rapid Transit (MRT) system has been under construction since 1988 and is expected to be completed by 1998. The first line was scheduled to commence service in 1994 but, due to a series of construction and equipment problems, the line is still under inspection and testing. It is now expected to begin operations in 1996, at the same time as the second line. Currently, limited rail commuter services carry about 5 percent of daily passengers. Conclusions from the Southeast Asian Experience Table 8 below attempts to summarize the main features of transport development in the Southeast Asian cities presented in this section, and of the outcome in terms of traffic congestion. The two extremes of city transport development in Southeast Asia are represented by Bangkok and Singapore. Bangkok suffers from chronic traffic congestion resulting from uncontrolled growth in vehicle ownership, an inadequate city street system, and little effective control of land use or public transport services. Some hoped-for improvement is offered by the rail transit schemes being planning or under construction, but these are plagued by financial, political and physical problems. Singapore, on the other hand, can be said to have solved the traffic congestion problem, using a combination of strict land-use controls, severe restraint of Stephen Stares and Liu Zhi 101 vehicle ownership through draconian pricing policies, controls on vehicle usage of the busier roads, again using pricing methods, and development of an exceptional public transit system exploiting both buses and metro, where appropriate. TABLE 8: TRANSPORT POLICY VS. OUTCOME IN SOUTHEAST ASIAN CITIES City and Country Motor Land Use Industry Urban Rail Bus Car Congestion Controls? Strategic? or Metros? Priorities? Restraints? Controlled?La HongKong V X ''4 V4 4 Seoul, Korea 4 4 ' 4 x x Bangkok, Thailand X X ('4) X X XX Singapore 4 X 4 4 4 '4'4 Manila, Philippines X X X X X XX Kuala Lumpur, Malaysia 4 '4 (X) (4) X X Taipei, Taiwan (China) X ('4) (') X X X = No, '4= Yes (4) = In implementation or planning La A double mark indicates a higher degree of success or failure. The other cities fall somewhere between these two extremes. The relative degree of success or failure in controlling congestion seems to depend on the extent of control and intervention exercised by the government in the areas of land-use control, vehicle ownership/ usage restraint, and provision of good public transport services. The two city-states of Singapore and Hong Kong have been most active in restraining vehicle ownership, but have two key advantages that neither the other cities nor China possess: a restricted land area with no rural hinterland to consider, and no domestic motor vehicle manufacturing industry to satisfy. However, Singapore has also been extremely successful in controlling motor vehicle usage in the city center, and this owes nothing to its city-state status. On the evidence of Singapore, road usage pricing can be effective when implemented with other integrated transport policies. All cities have identified fixed rail systems as a key element in their transport strategy. Several have successful systems in operation and all are constructing or planning new lines. There are three special features to note. First, almost all the cities are the capital city of their territory, so financing becomes a national matter rather than just a city one. Most large cities in China contemplating expensive urban rail construction would not be supported by national funds. Second, all the cities were at an advanced stage of economic development before starting on urban rail development and therefore had substantial resources to call on. Many large Chinese cities are not at this stage yet. Third, and most importantly, rail systems alone are not the answer. The successful rail-system cities are those vigorously pursuing other parallel policies to control land use and transport. It is difficult to summarize experience from other countries in any precise way, partly because objective comparative data are hard to come by, and partly because the influence of the 102 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions particular geography, history and customs of each city is difficult to capture and yet has a profound impact on the outcome. Hlence, international experience can provide some clues on likely transport policy outcomes, but specific policies for China must be developed in the local context. G. STRATEGY CONSIDERATIONS FOR CHINA Motorization is proceeding rapidly in China, and this process is essentially irreversible. It is mostly driven by the substantial economic benefits brought to industry, and in particular to the forms of widespread light industry which have been so crucial to China's recent economic growth. However, the problems of motorization are mostly felt in the urban areas, and traffic jams are now becoming a daily feature of major cities. T his paper has reviewed the issues raised by this, and the range of possible actions that can be taken. These are discussed below in the context of current policies in China. Road Management and New Construction Available road space is relatively low in many Chinese cities, although not as low as sometimes quoted. This leads to the following considerations: * Existing road space can be better managed. With the existing fragmented responsibilities for municipal transport, this could be assisted by the development of interagency teams to tackle traffic management in a comprehensive manner. * Bus services are particularly vulnerable to traffic congestion and need protection. Bicycles do not mix well with motorized traffic, but play an essential role in city- center transport. Therefore, it makes sense to allocate existing road space among road users, segregating to the extent possible buses, bicycles (already extensively segregated) and other road vehicles. This means protecting specific roads or sections of roads for exclusive use by buses or bicycles, at least for critical periods in the day. * In addition to the above measures, to make the existing street system perform more efficiently, special measures are needed to protect vulnerable areas where excessive vehicle use is detrimental, particularly those dominated by pedestrians, using traffic- calming methods. * Considerable road construction will be required in future years, but this will mostly be part and parcel of the process of urban expansion and city-center dedensification. * Some urban expressway construction will be useful, but many city masterplans currently in development are well in excess of world norms for urban expressway construction, will likely prove unaffordable, and are unlikely to resolve city traffic problems in the manner intended. Better evaluation and justification of high-class road networks is necessary. * In general, new road construction on its own will not solve the traffic congestion problem-traffic will tend to fill up all available road space. New construction and road management must be combined with management of transport demand. Stephen Stares and Liu Zhi 103 Manqagement of Transport Demand Cities need to take action to avoid chronic traffic congestion, which profoundly reduces the efficiency and quality of life in cities. * A major cause of traffic congestion is the systematic underpricing of transport services, which both burdens the road system with excess traffic and produces financial deficits to be made up from general revenues. O A more consistent transport pricing policy is highly desirable, as discussed in Theme Paper 10, "Shaping the Future: Getting Prices Right." O In particular, higher prices for vehicles and fuel, imposed by means of supplementary taxes or fees, would provide additional municipal revenues and usefully dampen the present rapid and unmanageable growth in transport demand. * Worldwide experience shows that the keys to controlling traffic congestion are strong policies to control the use of vehicles, particularly cars and motorcycles, in city centers at peak times: O Control of city-center parking by pricing and regulatory means are the most widely used policies, despite some inherent flaws in this approach. O More sophisticated area pricing and road pricing policies promise to be the more effective over the long term and deserve close study. 0 China currently has a window of opportunity to establish effective policies before motorization is too far advanced. • Such measures could be supplemented by restrictions by vehicle type or by area, but measures applied to date in China appear ad-hoc, implemented as problems arise, without analysis of needs and benefits of each vehicle class and with little restriction on car usage. All nonpricing methods of restricting vehicle usage deserve very careful evaluation to ensure that they do not bring substantial economic disbenefits. Motorcycles Motorcycles present a special problem: * Without controls, they are likely to grow extremely rapidly and swamp city street systems. * Higher-quality motorcycles, with strict noise and emissions controls and priced accordingly, could improve their environmental impacts and somewhat restrict demand, but further controls would likely be necessary. 104 Theme Paper 1: Motorization in Chinese Cities: Issues and Actions * Increasing motorcycle prices by supplementary taxes or fees would help control the growth in demand, but would unnecessarily penalize ownership and use in rural areas where they have a strong economic role. * Until urban road pricing becomes a reality, some areawide restrictions on urban motorcycle use appear to be necessary. Motor Vehicle Industry Projections of the rate of motorization in China made by domestic motor industry studies are probably conservative. This suggests: * that a viable domestic industry can be established on the basis of domestic demand; * that it will not be necessary to "force" the pace of motorization by subsidies or financial incentives or subsidies to new owners; * that cities can be left to set policies to control and guide the ownership and usage of motor vehicles without endangering the viability of the motor vehicle industry. Michael P. Walsh 105 THEME PAPER 2: MOTOR VEHICLE POLLUTION CONTROL IN CHINA: AN URBAN CHALLENGE MICHAEL P. WALSH' INTRODUCTION AND BACKGROUND Over the past one or two decades, the vehicle population has begun to grow rapidly in China. For example, in Figure 1, the total road network is growing rapidly, as is the total amount of passenger mileage and freight turnover. Further, as illustrated in Figure 2, there has been a steady shift from rail to road over this same time frame. While these vehicles have brought many advantages-increased mobility and flexibility for millions of people, more jobs, and enhanced many aspects of the quality of life-the benefits have been at least partially offset by excess pollution and the adverse effects which result. FIGURE 1: GROWTH OF ROAD MILEAGE AND TRANSPORTATION Road Mileage Passenger Mileage, Freight 1,200 500 1,000 400 S 4 s * ~~~~~~~~~~~~~400 EJ_ ~~~~~~~~~<4 800- 300 600-4 200 400 200 100 0. 10 1980 1982 1984 1986 1988 1990 1992 1981 1983 1985 1987 1989 1991 1993 Paved Blacktop Only 27% 3.3 km/1000 people Road Mileage Passenger Mileage Freight Turnover (1000 Km) (billion passenger km) (billion ton km) The purposes of this paper are to review the environmental consequences of an increase in road traffic and to summarize the strategies that could be used to ameliorate them. The next section will review the adverse effects of pollution. This will be followed by an assessment of the dominant role usually played by motor vehicles as a source of this pollution. Then, the ) Michael P. Walsh is an Environment Consultant, Washington, DC, I Inited States. 106 Theme Paper 2: Motor Vehicle Pollution Control in China: 4n Urban Challenge current and likely future situation in China will be evaluated, followed by a review of strategies that could reduce vehicle pollution, in general, and then specifically in China. VEHICLE EMISSIONS AND AIR FIGURE 2: TRENDS IN SURFACE TRANSPORTATION QUALITY CONCERNS Billion Passenger Kilometers 500 Cars, trucks, motorcycles, scooters and buses emit significant 400 quantities of carbon monoxide, hydrocarbons, niitrogeni oxides anld * Railway fine particles. Where leaded 200 El Road Passenger Miles gasoline is used, it is also a significant source of lead in urban 100 air. As a result of these emissions, 0 many major cities around the world 1980 Year are severely polluted. This section will review some of the consequences of these pollutants. Lead There has been an explosion of knowledge during the last two decades with regard to the adverse health impact of long-term exposures to low levels of ambient lead.2 In response to this growing body of data, most industrialized countries and several developing countries have introduced unleaded gasoline and several have or will soon prohibit the use of leaded gasoline entirely. The toxic properties of lead at high concentrations have been known since ancient times as lead has been mined and smelted for more than 40 centuries. Precautions in its use have been widespread for centuries, but it has only been recently that its adverse impacts at very low levels have been fully appreciated. The seminal work in this area is the 1979 report by Dr. Herbert Needleman and his colleagues, which showed that children with high levels of lead accumulated in their baby teeth experienced more behavioral problems, lower intelligence quotients (IQs) and decreased ability to concentrate.3 More recent evidence indicates that it is not only the length and severity of exposure to lead that results in the health damage but the age at which exposure occurs. This is especially important because "Of all the persons in the community, the newborn child is the most prone to injury from overexposure to lead for several reasons, and the damage that may be caused then will have the greatest long-term social and economic consequences."4 2 "Lead Exposure and Human Health: Recent Data on an Ancient Problem," Needleman, Technology Review, March/April 1980. "Air Quality Criteria for Lead," Office of Research and Development, US Environmental Protection Agency, Washington, DC, December 1977. 3 "Deficits in Psychological and Classroom Performance of Children with Elevated Dentine Lead Levels," Needleman, et al., The New England Journal Of Medicine, Vol. 300, Number 13, March 29, 1979. 4 "Exposure to Lead in Childhood: The Persisting Effects," Moore, Nature, Vol. 283, January 24, 1980. Michael P. Walsh 107 Another series of health studies in the United Kingdom confirmed these findings.5 They add further evidence that lead contributes to behavioral problems, lower IQs and decreased ability to concentrate. Even after taking up to 15 social factors into account, a 3 IQ number deficit was consistently found. While not necessarily statistically significant in any individual study (which is largely influenced by the size of the sample among other factors), the body of data consistently shows the effects. In addition, the studies of Dr. Winneke in Germany offer further evidence that "neuropsychological effects are causally related to very low blood lead levels."6 The effects are not necessarily the dominant ones in any particular instance but they are real, a matter of concern and preventable. Several comprehensive studies of the health issue have been conducted and their major conclusions are summarized below: * In 1980, the US National Academy of Sciences comoleted an extensive study of "Lead in The Human Environment."7 A major finding of this study is "The evidence is convincing that exposures to levels of lead commonly encountered in urban environments constitute a significant hazard of detrimental biological effects in children, especially those less than three years old. Some small fraction of this population experiences particularly intense exposures and is at severe risk." The Academy then recommended that "A serious effort should be made to reduce the baseline level of exposure to lead for the general population of the United States." * In August 1982, as part of its review of the existing lead program, the US Environmental Protection Agency (EPA) reanalyzed the issue and summarized the results in this way: "The majority of the comments emphatically rejected the proposition that lead was no longer a public health problem. Sixty-four comments were received from the professional health community and academia. Sixty of these opposed any loosening of the lead standard, and many suggested that tighter controls would be desirable. Thirty-two comments were received from local and state governments. All of these supported retention of the current standard to protect the citizens' health. Most of the commenters pointed to previous studies, as well as their own experiences, to demonstrate that lead has an adverse effect on people at very low dosages, and that the more the problem is studied the lower the 'acceptable level' of lead becomes. They concluded that protection of public health and welfare demands that all reasonable steps be taken to eliminate lead from the environment."' In October 1982, EPA decided as a result of this review to reduce the amount of lead in gasoline even further. 5 "The Relationship between Blood Lead Concentrations, Intelligence and Attainment in a School Population: A Pilot Study," Yule, Lansdown, Millar and Urbanowicz, Devel., Mfed Child Neurol. 1981, 23, 567-576. 6 Comments at Conference, Lead in Petrol, Winneke, May 1983. 7 "Lead in the Human Environment," National Academy of Sciences, Washington. DC, 1980. 8 Federal Register, Vol. 47, No. 167, Friday, August 27, 1982. 108 Theme Paper 2: Motor Vehicle Pollution Control in China: ,ln (Urban Challenge I In April 1983, the US Court of Appeals completed its review of the EPA decision to lower the gasoline lead levels.9 In its opinion the Court noted, "there is compelling evidence that gasoline lead is a major cause of lead poisoning in young children." In making this assessment, the Court found that "recent studies suggest that the recognized danger point of 30 micrograms per deciliter is too high and that lead reduces intelligence at blood lead levels as low as 10 to 15 micrograms per deciliter.... Other studies have correlated blood lead levels of 10 to 15 micrograms per deciliter with altered brain activity." The Court concluded that "the demonstrated connection between gasoline lead and blood lead, the demonstrated health effects of blood lead levels of 30 micrograms per deciliter or above, and the significant risk of adverse health effects froin blood lead levels as low as 10 to 15 micrograms per deciliter, would justify EPA in banning lead from gasoline entirely." * Finally, also in April 1983, in the United Kingdom, the Royal Commission on Environmental Pollution concluded that "the safety margin between the blood lead concentrations in the general population and those at which adverse effects have been proven is too small.... It would be prudent to take steps to increase the safety margin of the population as a whole." They continued, "that measures should be taken to reduce the anthropogenic dispersal of lead wherever possible...... Based on the growing body of data showing adverse effects from lead, in 1985 the US EPA reduced the maximum allowable lead content in leaded gasoline to 0.1 grams per gallon. As part of that rule-making, EPA uncovered evidence linking lead in the blood and high blood pressure." A subsequent study, in which 249 children were monitored from birth to two years of age, found that those with prenatal umbilical-cord blood lead levels at or above 10 micrograms per deciliter (.tg/dl) consistently scored lower on standard intelligence tests than those at lower levels.'2 Most recently, British researchers reviewed every epidemiologic study on lead and IQ published since 1979 that had over 100 children and measured IQ as a function of blood or tooth lead levels. Based on a meta-analysis of all the data, they concluded that a doubling of body lead burden from 10 to 20 ,ug/dl in blood levels was associated with a mean fall of about I to 2 IQ points." 9 United States Court of Appeals, No. 82-2282, Small Refiner Lead Phase-Down Task Force, et al. v. US EPA, April 22, 1983. 0 "Lead in the Environment," Ninth Report, Royal Commission on Environmental Pollution, April 1983. 1 Schwartz, J., H. Pitcher, R. Levin, B. Ostro, and A.L. Nichols. 1985. Costs and Benefits of Reducing Lead in Gasoline: Final Regulatory Impact Analysis, Report No. EPA-230-05-85-006, US EPA, Washington, DC. 12 Needleman, 1989. 3 Pocock, S.J., et al., "Environmental Lead and Children's Intelligence: A Svstematic Review of the Epidemiological Evidence," BMJ 1994, November 5; 309: 1189-97. Michael P. Walsh 109 In summary, the available evidence indicates that "there is no known physiological function served by lead in mammalian metabolism. As far as cells are concerned, each molecule of lead has the potential to disrupt the chemical basis of normal cellular function. For nerve cells, this interference is particularly destructive because communications between cells in the brain depends upon precisely controlled movements of such molecules such as calcium, sodium, potassium and chloride. Lead can interfere, on a molecule by molecule basis, with these essential elements."'4 Lead Scavengers When lead additives were first discovered to improve gasoline octane quality, they were also found to cause many problems with vehicles. Notable among these was a very significant buildup of deposits in the combustion chamber and on spark plugs, which caused durability problems. T o relieve these problems, lead scavengers were added to gasoline at the same time as the lead to encourage greater volatility in the lead combustion by-products so they would be exhausted from the vehicle. These scavengers continue to be used today with leaded gasoline. Ultimately, a significant portion of these additives are emitted from vehicles. This is important because, unfortunately, these lead scavengers, most notably ethylene dibromide, have been found to be carcinogenic in animals and have been identified as potential human carcinogens by the National Cancer Institute.'5 Therefore, their removal along with the removal of lead may result in significant beneFits to health. Carbon Monoxide (CO) Carbon monoxide-an odorless, invisible gas created when fuels containing carbon are burned incompletely-poses a serious threat to human health. Persons afflicted with heart disease and fetuses are especially at risk. Because the affinity of hemoglobin in the blood is 200 times greater for carbon monoxide than for oxygen, carbon monoxide hinders oxygen transport from blood into tissues. Therefore, more blood must be pumped to deliver the same amount of oxygen. Numerous studies in humans and animals have demonstrated that those individuals with weak hearts are placed under additional strain by the presence of excess CO in the blood. In particular, clinical health studies have shown a decrease in time to onset of angina pain in those individuals suffering from angina pectoris and exposed to elevated levels of ambient CO."6 Nitrogen Oxides (NOj) As a class of compounds, the oxides of nitrogen are involved in a host of environmental concerns impacting adversely on human health and welfare. Nitrogen dioxide (NO2) has been linked with increased susceptibility to respiratory infection, increased airway resistance in 4 "Lead Poisoning," Dr. Ellen Silbergeld, Toxic Substance Control ANewsletter, Autumn 1982. 15 "Automotive Emissions of Ethylene Dibromide," Sigsby, et al., Society ofAutomotive Engineers, #820786. 16 "Effect of Carbon Monoxide on Exercise Performance in Chronic Obstructive Pulmonary Disease," Aronow, et al., American Journal of Medicine, 1977, "'Health Effects of Exposure to Low Levels of Regulated Air Pollutants, A Critical Review," Ferris, Journal of The Air Pollution Control Association, May 1978. 110 Theme Paper 2 Alotor V"ehicle Pollution Control in China: An tIrban Challenge asthmatics, and decreased pulmonary function."' It has been shown that even short-term exposures to NO2 have resulted in a wide-ranging group of respiratory problems in school children--cough, runny nose and sore throat are among the most common.'" Further, in France, in an ingenious experiment, Dr. Orehek has shown that asthmatics are especially sensitive to even one-hour exposures.'9 A small group of asthmatics were initially exposed to carbachol, a bronchoconstrictor representative of urban pollen, anid then to NO2; adverse effects such as increased airway resistance were experienced by some of the individuals at levels as low as 0.1 parts per million for 1 hour. The oxides of nitrogen also participate in the formation of the family of compounds known as photochemical oxidants and in acid deposition. Finally, as a result of secondary transformations in the atmosphere, NO, emissions are converted to nitrates, thereby increasing the accumulationi of particulate in the air.20 Photochemical Oxidants (Ozone) The most widespread air pollution problem in areas with temperate climates is ozone, one of the photochemical oxidants that results from the reaction of nitrogen oxides and hydrocarbons in the presence of sunlight. Motor vehicles are a major source of both of these precursor pollutants. Ozone causes eye irritation, cough and chest discomfort, headache, upper respiratory illness, increased asthma attacks and reduced pulmonary function.2' It has also been demonstrated in numerous studies that photochemical pollutants seriously impair the growth of certain crops. For example, the Congressional Research Service of the US Library of Congress found that "the short-run or immediate impacts of ozone are evident in annual crop yield decreases estimated at $1 .9 to $4.3 billion."22 Particulate (PM) A series of studies released in the last few years indicate that particulate may be the most serious urban air pollution problem. By correlating daily weather, air pollutants and mortality in six US cities, scientists have discovered that nonaccidental death rates tend to rise and fall in 7 "Air Quality Criteria for Nitrogen Oxides," Draft, US Environmental Protection Agency, June 1980, "Health Effects of Exposure to Low Levels of Regulated Air Pollutants, A Critical Review," Ferris, Journal of The Air Pollution Control Association, May 1978. 18 "The University of Akron Study on Air Pollution and Human Health Effects," Mostardi et al., Archives of Environmental Health, September/October 1981. 9 "Effect of Short-Term, Low-Level Nitrogen Dioxide Exposure on Bronchial Sensitivity of Asthmatic Patients," Orehek, et al., The Journal of Clinical Investigations, Volume 57, February 1976. 20 Atmospheric nitrate is essentially secondary, formed from reactions involving oxides of nitrogen to form nitric acid. 21 "Air Quality Criteria for Ozone and Other Photochemical Oxidants," US Environmental Protection Agency, April 1978. 22 "Air Pollution Impacts on Agriculture and Forestry," Biniek, Congressional Research Service, Library of Congress, May 1982. Mlichael P. Walsh I11 near lockstep with daily levels of particulates-but not with other pollutants.23 Because the correlation held up even for very low levels-in one city to just 23 percent of the federal limit on particulates-these analyses suggested to the researchers that as many as 60,000 US residents per year may die from breathing particulates at or below legally allowed levels.24 More recently, another study has emerged showing a strong linkage between particulate air pollution and mortality.l5 The study is distinctive in that it used a prospective cohort design that allowed for direct control of other individual risk factors such as cigarette smoking, diet, etc. In addition, the study was larger and represented a larger geographic area than any other study to date. Air pollution data from 151 US metropolitan areas were linked with individual risk factors in 552,138 adults who resided in these areas when they were enrolled in this study in 1982. Deaths were ascertained through 1989. Sulfates and fine particulate air pollution were associated with a difference of approximately 15 to 17 percent between mortality risks in the most polluted cities and in the least-polluted cities. Even in cities that meet the US Federal clean air standards, the risk of death is 2 to 8 percent higher than in the cleanest cities. Certain particles appear to be especially hazardous. For example, diesel particles, because of their chemical composition and extremely small size, have raised special health and environmental concerns. Diesel particulate matter consists mostly of three components: soot formed during combustion, heavy hydrocarbons condensed or adsorbed on the soot, and sulfates. In older diesels, soot was typically 40 to 80 percent of the total particulate mass. Developments in in-cylinder emissions control have reduced the soot contribution to particulate emissions from modern emission-controlled engines considerably, however. Much of the remaining particulate mass consists of heavy hydrocarbons adsorbed or condensed on the soot. This is referred to as the soluble organic fraction of the particulate matter, or SOF. The SOF is derived partly from the lubricating oil, partly from unbumed fuel, and partly from compounds formed during combustion. The relative importance of each of these sources varies from engine to engine. A comprehensive assessment of the available health information on diesel particulate was carried out by the International Agency For Research on Cancer (IARC) in June 1988 and concluded that diesel particulate is probably carcinogenic to humans.26 Studies conducted at the Fraunhofer Institute have suggested that the diesel particle itself, stripped of the organic and other materials on the surface, may also be carcinogenic. Confirmatory studies under the auspices of the Health Effects Institute (HEI), a jointly funded Industry-Government program, recently verified this conclusion. These "results, and recent 23 "An Association between Air Pollution and Mortality in Six US Cities,' Dockery, et al., The New England Journal of Medicine, December 9, 1993. 24 "Air Pollution and Daily Mortality in Philadelphia," Dr. Joel Schwartz, presented at the 1991 meeting of the American Lung Association, Anaheim, CA. May 1991. 25 Pope at al, 1995. 26 The term "carcinogen" is used by IARC to denote an agent that is capable of increasing the incidence of ma!ignant tumors. 112 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge findings from other laboratories, suggest that (1) the small respirable soot particles in diesel exhaust are primarily responsible for lung cancer developing in rats exposed to high concentrations of diesel emissions, and (2) at high particle concentrations, the mutagenic compounds adsorbed onto the soot play a lesser role, if any, in tumor development in this species."" This is quite significant as it indicates that it is important to control the particles themselves and not just the organic material sitting on the surface of the carbon. In a subsequent analysis, HEI raised questions about this conclusion.28 The authors argue that because the rats were exposed to very high concentrations over their full lifetimes, the observed effects are more likely the result of the impairment of the rat's ability to clear particles from its lungs, leading to inflammation and rapid cell proliferation. They note that similar effects did not occur in hamsters and results were mixed with mice. While further studies are carried out to determine which element in diesel particles is most hazardous, the prudent course of action seems to be to reduce both the organics and the particulate mass. To put the concerns with diesel NO, and particulate into perspective, one recent study attempted to quantify the health benefits associated with reducing diesel particulate and nitrogen oxides.29 Based on a careful review of the available health information, the authors concluded that reducing one gram per mile of particulate or NO,, over a 100,000-mile vehicle lifetime would produce benefits of $11,432 and $1,175, respectively. Focusing specifically on the 1992 heavy-duty vehicle fleet in Los Angeles, the authors conclude that a 50 percent reduction in NO, and PM,( emissions, would be worth $9,200 and $13,500 per vehicle, respectively. A 90 percent reduction would have a value of $16,600 and $24,300 per vehicle, respectively. It is important to emphasize that these amounts reflect the value of the health benefits alone. Earlier studies have indicated that the economic benefits of reduced soiling and improved visibility are also quite significant. Physics and Chemistry of Particulate.0 Atmospheric particles originate from a variety of sources and possess a range of morphological, chemical, physical, and thermodynamic properties. Examples include combustion-generated particles such as diesel soot or fly ash, photochemically produced particles such as those found in urban haze, salt particles formed from sea spray, and soil-like particles from resuspended dust Some particles are liquid, some are solid; others contain a solid core surrounded by liquid. Atmospheric particles contain inorganic ions and elements, elemental carbon, organic compounds, and crustal compounds. Some atmospheric particles are hygroscopic and contain particle-bound water. The organic fraction is especially complex. Hundreds of organic compounds have been identified in atmospheric 27 ,Pulmonary Toxicity of Inhaled Diesel Exhaust and Carbon Black in Chronically Exposed Rats," Mauderly, et al., Health Effects Institute Research Report Number 68, October 1994. 28 HEI, 1995. 29 ,On the Costs of Air Pollution from Motor Vehicles," Small, K. and Kazimi, C., Department of Economics, University of California at Irvine, forthcoming in The Journal of Transport Economics, January 1995. 30 Excerpted from US EPA Draft PM Criteria Document, April 1995. Michael P. Walsh 113 aerosols, including alkanes, alkanoic and carboxylic acids, polycyclic aromatic hydrocarbons, and nitrated organic compounds. Particle diameters span more than four orders of magnitude, from a few nanometers to 100 micrometers (km). Combustion-generated particles, such as those from power generation, from automobiles, and in tobacco smoke, can be as small as 0.01 gm and as large as I gm. Particles produced in the atmosphere by photochemical processes range in diameter from 0.05 to 2 gm. Fly ash produced by coal combustion ranges from 0.1 to 50 ltm or more. Wind-blown dust, pollens, plant fragments, and cement dusts are generally above 2 gm in diameter. Recent measurements of the FIGURE 3: SIZE DISTRIBUTION OF TYPICAL size distributions of primary PARTICLES particles confirm US EPA conclusions that most fugitive dust r0 above 10 microns ernissions are in particles larger 120% t < 10micron than 2.5 gm and that the majority of 100% U <1 micron emissions from combustion sources 80% are in sizes smaller than 2.5 im.n Figure 3 was derived from a major 60% characterization of different source 40% emissions in California conducted 20% during 1986. Hot-exhaust samples were diluted to ambient a/ Diesel Fugitive Dust temperatures prior to sampling onto filter media through impactor inlets with 50 percent cut-points of 1, 2.5, 10, and -30 gm. The figure shows that combustion products are nearly always less than 2.5 ltm in size. In particular, they show that diesel truck emissions are almost all less than 1.0 tm in size; particles in this size range are especially hazardous because when breathed in, they are able to penetrate to the deepest part of the lung where the critical gas exchange takes place. Sources of Suspended Particles. The ambient atmosphere contains both primary and secondary particles; the former are emitted directly by sources, and the latter are formed from gases (SO2, NO,, NH3, VOCs). Fugitive dust is a primary pollutant. Major sources of particle emissions are classified as major point sources, mobile sources, and area sources; these are anthropogenic. Natural sources also contribute to ambient concentrations. Fugitive dust is a major contribution to PMIo at nearly all sampling sites, although the average fugitive dust source contribution is highly variable among sampling sites within the same areas, and is highly variable between seasons. Primary motor vehicle exhaust in the United States makes up as much as 40 percent of average PM1o at many sampling sites. Vegetative burning outdoors and residential wood burning are significant sources in residential areas. Fugitive dust from paved and unpaved roads, agricultural operations, construction, and soil erosion constitute around 90 percent of nationwide primary emissions in most countries. Fugitive dust consists of geological material that is suspended into the atmosphere by natural wind and by a.nthropogenic activities from sources 114 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge such as paved and unpaved roads, construction and demolition of buildings and roads, storage piles, wind erosion, and agricultural tilling. Mobile sources are major emitters of primary particles, oxides of nitrogen, and volatile organic compounds. They are also minor emitters of sulfur dioxide (SO2) and ammonia (NH3). On-road motor vehicles using gasoline-and diesel-fueled engines are by far the largest component of mobile source emissions in most countries, and the emissions estimation methods are most highly developed for these vehicles. Motor vehicle exhaust contains high concentrations of organic and elemental carbon, but their ratios are much different from those found in wood combustion with the abundance of elemental carbon being nearly equal to the organic carbon abundance. Exposure to Particulate Emissions. For any air pollutant, the total exposure of an individual consists of a variety of sequential exposures to a variety of microenvironments. They are typically, outdoors, indoors at-home, at-work, in-traffic, and many other indoor microenvironments. A typical particulate exposure scenario can be seen in Figure 4. Other Toxics The 1990 Clean Air Act (CAA) directed the US EPA to complete a study of emissions of toxic air pollutants associated with motor vehicles and motor vehicle fuels. The study found that in 1990, the aggregate risk is 720 cancer cases in the United States. For all years, 1,3-butadiene is responsible for the majority of the cancer incidence, ranging from 58 to 72 percent of the total motor vehicle toxics risk. This is due to the high unit risk of 1 ,3-butadiene. Gasoline and diesel particulate matter, which are considered to represent motor vehicle polycyclic organic matter (POM), are roughly equal contributors to the risk. The combined risk from gasoline and diesel particulate matter ranges from 16 to 28 percent of the total, depending on the year examined. Benzene is responsible for roughly 10 percent of the total for all years. The aldehydes, predominately formaldehyde, are responsible for roughly 4 percent of the total for all years. A variety of studies have found that in individual metropolitan areas, mobile sources are one of the most important and possibly the most important source category in terms of contributions to health risks associated with air toxics. For example, according to the US EPA, mobile sources are responsible for almost 60 percent of the air pollution-related cancer cases in the United States per year. Air Quality Standards Because of all the above concerns, the World Health Organization (WHO) and many individual countries have adopted air quality standards to protect public health and the environment. A cross-section of such standards is summarized in Table 1. FIGURE 4: A TYPICAL PARTICULATE EXPOSURE SCENARIO 280 260 * Indoors 240 0~~ In Transit 240 0 Outdoors Well-Ventilated Kitchen, 220 CO) 200 :L 180 Cafetera, Smoking Section o 160 Outside Cigar 2 140 _ Behind Smoky Diesel Truck Smoker's Office Dining Cafeteria, oom ,r 120 Nonsmoking C100 1080 t Vi Buexhaust edro Office IC Livig 60 ) Commutinglil A ommutlng Room 40 - - -Bedr2oomn - 8 Suburbs Street Suburbs, Outdoor City. Outdoor Jogging 20 -LbayU0CFdCafeteria I2I0I I I I I LlbzarylU oIcaItb I In__e I I Li Ing Ro r I I 34 5 6 7 8 9 10 1112 1 2 3 4 5 6 7 8 9 10 1112 Midnight A.M. Noon P.M. Time of Day 116 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge Table 1: AMBIENT AIR QUALITY STANDARDS (,g/m3, except for CO which is in mg/rn3) CO_ _ NO2 = = SO2 SPM 03 Pb 8 hr. I hr. Annual 24 hr. 1 hr. Annual 24 hr. I hr. Annual 24 hr. 8 hr. I hr. Annual 24 hr. WHO 10 30 150 400 50 125 350 120 100 150 0.5 United States 10 40 100 80 365 50/a 150* 235 15/b Japan 22.8 80 107 267 10( 118 0.1 Germnany 30 80 200 140 400 100 2 Italy 50 200 40 100 40 100 Netherlands 6 40 175 500 830 150 120 0.5 2 Canada 6 15 60 200 400 30 150 450 60 120 30 100 Taiwan (China) 22.9 100 133 267 240 Thailand 20 50 320 100 300 1 100 330 _____ 10 /a USA SPM standards are for PM 10, others are for TSP. /b Quarterly Average. MOTOR VEHICLES ARE A DOMINANT POLLUTION SOURCE Worldwide, cars, trucks, buses, and other motor vehicles continue to play a dominant role in local, regional and global air pollution. They are major sources of carbon dioxide; volatile organic compounds (VOCs) and nitrogen oxides, the precursors to both tropospheric ozone and acid rain; carbon monoxide (CO); toxic air pollutants such as diesel particulate; and chlorofluorocarbons (CFCs). Throughout the European Union FIGURE 5: SOURCES OF EMISSIONS IN THE as a whole, for example, both on- and off- EUROPEAN UNION, 1990 road vehicles are the largest source of CO, 140% NO, and nonmethane hydrocarbons, as E3 OfRoadVehicles illustrated in Figure 5.31 1 . 0 100% In densely populated urban areas, 80% vehicles can be a major source of 60% particulate as well. For example, as 40% recently noted in the United Kingdom, "currently, road vehicles account for 74 20% percent of nitrogen oxides and 94 percent 0% NOx NMHC CO of black smoke emissions in London. On their own, diesels account for 32 percent and 87 percent of total emissions (43 percent and 92 percent of vehicle emissions) for these two pollutants respectively."32 As illustrated in Figures 6 and 7, motor vehicles are also major emissions sources in the United States and Japan. In the northeastern United States where the air pollution problem is 31 "The Estimation of the Emissions of Other Mobile Sources and Machinery Subparts Off-Road Road Vehicles and Machines, Railways and Inland Waterways in the European Union," Andrias, Samaras and Zierock, September 1994. 32 "Diesel Vehicle Emissions and Urban Air Quality," Second Report of the Quality of Urban Air Review Group, prepared at the request of the Department of the Environment, December 1993. Michael P. Walsh 117 especially severe, EPA projected that highway vehicles will account for approximately 38 percent of the total NO, inventory and 22 percent of the total VOC inventory in 2005, in spite of the introduction of tighter motor vehicle standards in the 1990 Clean Air Act.33 FIGURE 6: SOURCES OF EMISSIONS IN THE FIGURE 7: SOURCES OF EMISSIONS IN UNITED STATES, 1993 JAPAN, 1989 140% 140% U Road Traffic D Off Road Vehicles * Other 1 2 ad Traffic Othe 120% 100% 100 80% 80%/ 20% 20% 0% M0% ~ NOx NMHC CO PM-10 NOX Pariculae While not as well-documented, it is increasingly clear that motor vehicles are also the major source of many of the pollution problems that are plaguing the developing world. By way of examples, the motor vehicle contribution to air pollution problems of Manila and Korea are summarized in Figures 8 and 9. FIGURE 8: SOURCES OF EMISSIONS IN THE FIGURE 9: SOURCES OF EMISSIONS IN REPUBLIC OF KOREA, 1991 MANILA, 1990 140% 140% M Road Traffic e] Other * Road Trafc D Other 120% 120% 100% l o% e0% 80% 40% 40% 0% 0% NOx PM-10 TOG Co Nox TSP HC CO Virtually the entire global motor vehicle fleet runs on fossil fuels, primarily oil. For every gallon of oil consumed by a motor vehicle, about 19 pounds of carbon dioxide (containing about 5.3 pounds of carbon) go directly into the atmosphere. In other words, for every 15-gallon fill-up at the service station, about 300 pounds of carbon dioxide are eventually released into the atmosphere. Globally, motor vehicles account for about a third of world oil consumption and about 14 percent of the world's carbon dioxide emissions from fossil fuel-burning. For the United States, motor vehicles consume about 50 percent of the oil used each year and emit about 25 percent of the carbon dioxide. 33 These emissions estimates are based on the most accurate data currently available. The Agency continues to analyze emissions data and modeling assumptions. Consequently. these estimates could be subject to change. 118 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge A major source of CFCs in the atmosphere is motor vehicle air conditioning, and in 1987 approximately 48 percent of all new cars, trucks, and buses manufactured worldwide were equipped with air conditioners. (CFCs also are used as a blowing agent in the production of seating and other foam products, but this is a considerably smaller vehicular use.) Annually, about 120,000 metric tons of CFCs are used in new vehicles and in servicing air-conditioners in older vehicles. In all, these uses account for about 28 percent of global demand for CFC- 12. As agreed under the Montreal Protocol, CFCs are to be completely phased out of new vehicles by the turn of the century, a welcome but long overdue step. CFCs emitted over the next decade will cause damage to the planet for the next two centuries. THE SITUATION IN CHINA TODAY Background The number of motor vehicles in China is very small compared with the developed countries-the total vehicle population is only approximately 6 million; conversely, the population of China is almost 1.2 billion, the largest in the world. Moreover, 71 percent of the vehicles are trucks, less than 23 percent are passenger cars, and only 15 percent are diesel-fueled vehicles, a much lower ratio than in virtually any other country in the world. Nevertheless, over the FIGURE 10: PASSENGER CAR FLEET AND PRODUCTION past several years the annual ('000) vehicle production and FLEET PRODUCTION registrations have begun to 10,000 1,400 increase rapidly, averaging 10 Motor Vehicle Fleet Car Fleet 11200 and 12 to 14 percent per year, 8,000 + A -W_ and12to14perenpra, 8Motor Vehicle Production Car Productio 1,000 respectively, since the late 6,000 - -- 800 1970s and one can expect that _---_ '_X this tendency will continue for 4,000 _ >- 600 the foreseeable future. As a - 400 result, as shown in Figure 10, 2,000 200 the vehicle population in China o 0 _"_ = '0 has doubled from less than 4 1986 1987 1988 1989 1990 1991 1992 1993 Car Production Expected I mm by 2000, 3.5 mm 201 0 million in 1986 to almost 8 Car Fleet Expected 22 mm by 2010 million in 1993. At the same time, vehicle production has approximately tripled from about 400,000 to about 1.3 million vehicles per year. Not surprisingly, as has been the case in most other countries, most of the vehicles tend to be concentrated in or near urban areas. As a result, as illustrated in Figure 11 for just three cities, Guangzhou, Beijing and Shanghai, the urban vehicle population has grown even more rapidly than the national vehicle population since the early to mid-1980s. If these trends continue at the current rate, as shown in Figure 12, the urban vehicle populations will be two to four times greater by 2010 than they are today; in fact, their rates of increase are likely to be more than linear. MichaelP. Walsh 119 FIGURE 11: VEHICLE POPULATION FIGURE 12: VEHICLE POPULATION TRENDS IN CHINESE CITIES TRENDS IN CHINESE CITIES Thousands Thousand 700 700 2,OGO 2,000 600 I 600 1,500 1,500 500 GUANGZHOU 500 1 -U--- ~~~~~~~~~~~~~~1.000 -,0 400 BEIJING 400 300 300 550So 200 SHANGHAI 200 5 D0 100 100 1,960 1,970 1,980 1,990 2,000 2,010 O 0 197 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 GUANGZHOU BEIJING SHANGHAI 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 -U- -- It is also important to note that the types of vehicles can differ substantially between cities. For example, as illustrated in Figure 13, almost two out of every five vehicles in Beijing is a light truck or utility vehicle; in contrast, almost two thirds of the motor vehicles in Guangzhou are motorcycles. Further, the growth rate of motorcycles in Guangzhou far exceeds that of other motor vehicle types (Figure 14). This is significant because of the high-pollution characteristics of these vehicles, especially those with two-stroke engines. FIGURE 13: 1992 VEHICLE POPULATION IN FIGURE 14: VEHICLE REGISTRATION IN DIFFERENT CHINESE CITIES GUANGZHOU (NUMBER OF VEHICLES) 38.0% 9.9% 12.0% 400,000 17.4% 13Z%| 350.000 GMryces 13.2% 200.000 MSmaIT0Carks 250.000 fiLarge T,aks 17.1% 27.5% 649%, 50,000 0LreCr 100,000 .,. i BEIJING GUANGZHOU s0 - * CARS B LIGHT TRUCKS x 12 HEAVY TRUCKS & BUSES 0 MOTORCYCLES Environmental Problems In spite of the relatively low FIGURE 15: LIKELY MOTOR VEHICLE-RELATED vehicle population, air pollution AIR QUALITY TRENDS IN CHINESE CITIES problems caused by motor vehicles have NORMALIZED AIR QUALITY LEVELX POLLUTANT started to emerge in the major cities of , S China (Figure 15). One reason, as noted / SNPM-1D0 above, is that the vast majority of the 12 / CO vehicles in use in China are driven in 10 _NOX major cities. In addition, while some of 8 the vehicles in China are manufactured 6- by joint-venture enterprises and by 4/ enterprises under the license of 2 developed countries, another portion are o 1 2000 Y designed, developed, and manufactured R990 2000 2010 120 Theme Paper 2: Motor Vehicle Pollution Control in China: An UJrban Challenge by domestic companies using designs that are more than 20 years old. One result is that the CO and hydrocarbon (HC) emission levels in these engines are estimated to be about 10 to 20 times the levels emitted from the controlled vehicles in the United States or Japan. Furthermore, the operating speed of motor vehicles in the major cities is quite low due to the crowded streets and mixed traffic (motor vehicle, motorcycle, bicycle, tractor, even carts); the average speed inside the third ring road of Beijing is lower than 20 km/h, which results in a high level of CO and HC emissions. Therefore, pollution levels in the major cities of China are already unacceptably high, especially for CO and HC. CO and HC levels frequently exceed healthy levels, and their ambient concentrations parallel vehicle traffic patterns, that is, they tend to peak during the morning and evening peak traffic time. Within the city proper in Beijing, the average concentration of CO exceeds the limit prescribed by the National Ambient Air Quality Standard. Furthennore, the concentration levels in the street and in the residential areas near the street are much higher than the average values. Finally, the proportion of days exceeding the standards is increasing in parallel with the increase of motor vehicles in use. According to an air quality survey of Beijing in the late 1980s as summarized in Table 2, motor vehicles contribute about half of the total CO, HC, and NO, emissions coming from all pollutant sources. TABLE 2: MOTOR VEHICLE EMISSION POLLUTANTS' CONTRIBUTION IN BEIJING URBAN AREA (WITHIN THE THIRD RING ROAD) (Percent) Pollutant Winter Time Summer Time Annual Average CO 26.1 60.0 39.1 HC 62.7 86.8 74.8 NO, 38.0 54.7 46.2 Lead is another pollutant of concern in the major cities of China. Lead levels in the urban area of major cities such as Beijing are usually I to 1.5 ptg/m3, and even reach 14 to 25 Ag/M3 in some areas. Overall, particulate problems caused by motor vehicles in Beijing are minimal. One reason is that driving motorcycles is restrained in many major cities of China including Beijing. In addition, the proportion of diesel-fueled vehicles is relatively small compared to other countries. Most of the diesel vehicles are heavy-duty trucks, which are not allowed to drive in urban areas. In Beijing, therefore, the only significant diesel vehicle population is buses. Serious smoke, suspended particulate matter, and SO2 pollutants in China mainly result from the coal burned by power stations, plants, and residents. Based on current vehicle emissions rates and the likely future growth in urban road traffic, one can foresee a tremendous growth in vehicle-related air pollution problems in the near Michael P. Walsh 121 future along the lines illustrated below unless government policies are adopted to reverse current trends. Motor Vehicle Emissions Controls Rather than waiting for that to happen, and then trying to rapidly develop solutions in a crisis mode, China has already begun to address motor vehicle pollution. All the motor vehicles driving in China are required to be registered, and equipped with a pair of registration number plates. The local vehicle management offices, subordinated to the local bureau of public security, are in charge of this registration. A domestic motor vehicle will not be registered, unless the model of this motor vehicle has been listed in the "Index of Enterprise-Produced Motor Vehicles and Their Products" issued by the China National Automotive Industry Corporation (CNAIC) and the Ministry of Public Security. Before being listed in the index, the vehicles must pass an approval test carried out by "The Type Approval Organization for New Motor Vehicle Products" authorized by CNAIC. Included in the approval test are idling emissions tests for gasoline-fueled vehicles and free-acceleration smoke tests for diesel-fueled vehicles. In addition, a full-load smoke test is required for diesel engines. During 1994, new light-duty vehicles offered for sale in Beijing were required for the first time to pass the ECE regulation 15-04 emissions standards. In addition, each motor vehicle should be checked for its idling emissions or free-acceleration smoke at one of the local inspection lanes during registration. In the case of imported motor vehicles, the approval test, including safety and emission tests, of a sample vehicle is conducted in one of the labs authorized by the State Administration of Import and Export Commodity Inspection (SACI), which is the responsible body for motor vehicle imports and exports. Further, during registration, each imported vehicle must be checked at the inspection lane, the same as the domestic vehicles. A periodic inspection is required once per year for every category of vehicles. It is conducted by the local vehicle management office in an inspection lane using the idling emissions or free-acceleration smoke standards. Furthermore, idling emissions or free-accelerator smoke inspections are carried out at the end of assembly line for each gasoline or diesel-powered vehicles by the major auto manufacturers. The National Environmental Protection Agency of China is in charge of formulating and revising the standards for controlling major vehicle emission pollutants, while the local Environmental Protection Bureaus are responsible for supervising and putting emission standards into effect. A number of additional steps are under development and will be phased in over the next few years. These include the following: * The gasoline octane number available for trucks will be raised from 80 to 90 Research Octane Number (RON) after 1995, to increase the thermal efficiency of gasoline engines, and thus decrease their fuel consumption. 122 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge * Unleaded gasoline will begin to be distributed to major cities, provincial capitals, and economic developing zones in 1995.34 * In order to reduce CO and HC by 30 percent on medium- and heavy-duty gasoline- fueled vehicles, exhaust emission standards for these vehicle are being promulgated, and will be put into effect from the beginning of 1995. The test procedure will be the 9-mode used by the US EPA in the period from 1970-83. * For the purpose of further reducing HC emissions, it is planned to require fuel evaporative emission controls. The appropriate standards are being evaluated and it is expected that they will be phased in during 1995 and 1996 for light- and heavy- duty vehicles. The limit values will be 2 g/test and 4 g/test, respectively. * The measurement method at idle speed will include a raised idle speed, as prescribed in International Standards Organization (ISO) 3929. It is intended that emissions of in-use vehicles will be checked at two engine speeds that are used frequently in urban areas, and on the basis of these tests determine whether the carburetor should be repaired or replaced. v For diesel smoke, the standard at free acceleration for newly produced vehicles is expected to be tightened to 4.5 units in 1995, and 4.0 units in 2000, while the values for vehicles in use will be 5.0 and 4.5, respectively. * As for standards of smoke at full load, the test procedure and limit value to be changed are under discussion. In addition, standards for controlling exhaust emissions of medium- and heavy-duty diesel engines are to be formulated, and the test procedure will be equivalent to that of ISO. Conclusions Regarding Current Situation Although China has a relatively modest vehicle population compared to developed countries, it has been increasing rapidly over the past 15 years, and will continue to have a high growth rate for the foreseeable future. Some air pollution problems caused by motor vehicles have started to emerge in the major cities of China since the middle of the 1980s, especially with regard to CO and HC, and the NO, problem has been getting worse in the past few years. Lead pollution from vehicles is also a concern, but smoke, particulate, and SO2 are mainly emitted from other pollutant sources. Looking at Beijing as an example, and based on the official growth rates, vehicle emissions are expected to grow substantially in the coming years in spite of the adoption of several new requirements as summarized in Figure 16. Using more likely (and somewhat higher) growth rates, Figure 17 shows that the emissions trends could be even worse. 34 China already produces some unleaded gasoline and actually exports some to Singapore. The remainder is mixed with leaded fuel at terminals since no segregated distribution system exists at present. Michael P. Walsh 123 FIGURE 16: BEIJING MOTOR VEHICLE FIGURE 17. BEIJING MOTOR VEHICLE EMISSIONS TRENDS (City Plan Growth EMISSIONS TRENDS (More Likely Growth Projections, Summer Conditions) Projections, Summer Conditions) EMISSIONS (NORMALIZED TO 1992) EMISSIONS (NORMALIZED TO 1992) 3 3 6 - 2.5 - _ 0 2.5 C . . c 2 - _. 2 HC 4 4 HC 1.5 - i *-- . X 1.5 NOx NOx -~~~~~~~ ~~2 2 0.5 _ - 0.5 C02 i 1 C02 0 0 0 O 0 VEHICLES 1992 2000 2005 2010 2015 VEHICLES 1992 2000 2005 2010 2015 YEAR YEAR STRATEGIES TO REDUCE MOTOR VEHICLE POLLUTION Since a great deal has been learned FIGURE 18: ELEMENTS OF A about reducing emissions from vehicles, one COMPREHENSIVE VEHICLE POLLUTION should not conclude that higher emissions and CONTROL STRATEGY more motor vehicle-related air pollution are inevitable. Strategies exist to both lower C emissions per kilometer driven and reduce VEHICLE actual driving; application of both approaches TECHNOLOGY can be used to ameliorate the otherwise likely future pollution increases. Many of these strategies-eliminating lead from gasoline, APPROPRIATE D vehicle inspection and maintenance, vehicle MAINTENANC MAE demand management, etc.-are win-win strategies in that they not only lower pollution / but also bring substantial economic or other CLEAN benefits (Figure 18). FUELS Generally, the goal of a motor vehicle pollution control program is to reduce emissions from motor vehicles in-use to the degree reasonably necessary to achieve healthy air quality as rapidly as possible or, failing that for reasons of impracticality, to the practical limits of effective technological, economic, and social feasibility. Achievement of this goal generally requires a comprehensive strategy encompassing emission standards for new vehicles, clean fuels, strategies designed to assure that vehicles are maintained in a manner that minimizes their emissions and traffic and demand management and constraints. These emission-reduction goals should be achieved in the least costly manner. Standards for permissible levels of exhaust and evaporative emissions from motor vehicles should be based on a realistic assessment of costs and benefits, keeping in view the technical and administrative feasibility of proposed countermeasures. Technological approaches to achieve the desired emission standards may include fitting new vehicles with emission control devices such as catalytic converters or particulate traps or requiring such devices to be retrofitted to existing vehicles, modifying fuels or requiring the use of alternative fuels in certain vehicles, 124 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge and traffic and demand management and policy instruments. However, many of the potential benefits of these countermeasures will be squandered if they are not buttressed by regulatory and economic instruments that assure vehicle owners, manufacturers and fuel suppliers have sufficient incentives to achieve the desired goals. A key element of the overall strategy, therefore, must be effective enforcement to ensure maximum compliance with standards. Vehicle Inspection and Maintenance Opportunities Today's vehicles are absolutely dependent on properly functioning emission controls to keep pollution levels low. Minor malfunctions in the emission control system can increase emissions significantly. Major malfunctions in the emission control system can cause emissions to skyrocket. A relatively small number of vehicles with serious malfunctions frequently cause the majority of the vehicle-related pollution problems. Unfortunately, it is rarely obvious which vehicles fall into this category, as the emissions themselves may not be noticeable and emission control malfunctions do not necessarily affect vehicle driveability. Effective inspection and maintenance (I/M) programs, however, can identify these problem cars and assure their repair. For countries with only minimal if any controls on vehicles, a simple T/M program can be a good pollution-control starting point as even vehicles with no pollution controls can benefit from improved maintenance. A simple idle check on CO and HC emissions from gasoline vehicles or visible-smoke check on diesel vehicles can be used to identify the highest polluters and those vehicles that would most benefit from remedial maintenance. Hong Kong, whose air quality problem is primarily excess particulates, trained a small group of smoke inspectors who then patrolled the streets, identifying vehicles with excess smoke and requiring them to be repaired or pay a fine. Such a program requires minimal capital investment and resources. As vehicle technology advances, more sophisticated test procedures may be necessary including loaded mode tests, which use a dynamometer to simulate the work that an engine must perform in actual driving. Substantial advances are occurring in I/M programs. For the most advanced vehicles, those equipped with electronic controls of air-fuel and spark management systems and equipped with catalytic converters to reduce CO, HC and NO,, a transient test that includes accelerations and decelerations typical of actual driving can provide additional emissions reduction benefits. Fuel Modifications and Alternative Fuels Conventional vehicle fuels have undergone substantial modification in recent decades and will likely be improved even more in the future; in parallel, alternative fuels such as ethanol, methanol, natural gas and liquefied petroleum gas (LPG) continue to receive attention. The major trend underway worldwide is the gradual replacement of lead in gasoline, both to reduce lead emissions and to facilitate the use of pollution control technologies such as the catalytic converter. Additional gasoline improvements include reduced volatility, increased oxygen content, reduced aromatics and more widespread use of detergent additives. Such fuel modifications can, if carefully introduced, substantially improve the environmental impacts of gasoline. Michael P. Walsh 125 Conventional diesel fuel can also be improved by the reduction of sulfur and aromatic content and the use of detergent additives. For countries that are initiating a vehicle pollution-control program, first priority with regard to fuels should be placed on improving those fuel qualities that are important for good combustion-in the case of gasoline, raising the octane number to at least 90 RON, and in the case of diesel, raising the Cetane number to at least the 40s and if possible the 50s. This will allow the use of modern engines. Next, the focus in most countries should be on reducing the amount of lead in leaded gasoline and introducing a grade of unleaded gasoline. This involves not only producing the unleaded fuel but providing an infrastructure for its distribution. For example, China at present, produces almost 20 percent of its gasoline as unleaded and actually exports unleaded gasoline to Singapore but, because of the lack of an adequate infrastructure, distributes very little unleaded gasoline domestically. The possibility of substituting cleaner-burning alternative fuels for conventional fuels has drawn increasing attention during the last decade. Motivations advanced for this substitution include conservation of oil products and energy security, as well as the reduction or elimination of vehicle emissions. Alternative fuels include methanol (made from natural gas, coal or biomass); ethanol (made from grain); vegetable oils; compressed natural gas (CNG) mainly composed of methane; LPG composed of propane, butane, electricity, hydrogen; synthetic liquid fuels derived from hydrogenation of coal; and reformulated gasoline and diesel, including oxygenated blends. Additives (like lead compounds in gasoline) are introduced in small quantities to improve storage, distribution, or performance characteristics of fuel. The principal alternative fuels presently under consideration are natural gas and methanol made from natural gas, and in limited applications, LPG. Environmental assessment of alternative fuels should not be based solely on vehicle end-use emission characteristics but should account for pollutant emissions associated with the production, storage, and distribution of these fuels. Partly in response to environmental pressures to eliminate lead in gasoline and partly in response to energy needs, increasing amounts of alcohols and ethers are being used either as high-octane blending components or as substitutes for gasoline. Alternative fuels can make significant contribution to improved air quality and are increasingly playing a role in urban areas. Most often, these fuels are used with special groups of vehicles that can have a large impact on the environment (such as transit buses or taxicabs) and can be fueled at central location, thus minimizing the need for a widespread fueling infrastructure. However, care must be taken when considering the use of alternative fuels to assure that the desired effect is achieved. For example, while particulate emissions and visible smoke can be reduced for a diesel vehicle by the use of alternative fuels such as CNG, emissions of other gaseous pollutants such as NO, and aldehydes may be much higher than from diesels if they are not addressed. When CNG vehicles are equipped with the appropriate pollution controls, they can be very clean and make a significant contribution to improving air quality. 126 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge Vehicle Emission Control Technology Measures Technology has been developed and introduced on millions of gasoline-fueled vehicles worldwide that has demonstrated the ability to lower CO, HC and NO,, emissions by approximately an order of magnitude compared to vehicles without controls. The backbone of these systems is the catalytic converter and similar approaches are now being gradually phased into the two-wheeled vehicle market. As the world community increasingly embraces today's state-of-the-art controls on these vehicles, advanced controls are emerging, which will likely become increasingly widespread throughout the next decade. Development of diesel control technologies beyond crude smoke controls started later but is now advancing rapidly. By the mid-1990s it is expected that exhaust aftertreatment systems will be increasingly available; in the interim, engine modifications are readily available. Vehicle Demand Management Cleaner vehicles and fuels alone will not solve the air pollution problem in many large cities unless accompanied by strategies designed to reduce the growth in the number of vehicle miles traveled (VMT). VMT has steadily increased over time in most areas of the world. Such increases offset a significant portion of the emission reductions potential associated with the introduction of cleaner vehicles and fuels into the motor vehicle fleet. To avoid a similar scenario in the future, air quality and transportation planners must develop more effective strategies for reducing the amount of discretionary automobile travel, especially in single-occupant vehicles. Strategies to reduce VMT may include a host of diverse measures including the segregation of high-occupancy vehicle lanes designed to encourage multiple-occupant vehicle commuting; promoting mass transit development and expansion projects; the imposition of parking freezes and other programs to restrict vehicle use in congested urban areas; parking fees or other types of vehicle use taxes; vehicle-free zones; fringe and transportation corridor parking facilities; ride-sharing/van-pooling programs; bicycle lanes and storage facilities; flextime programs to reduce office commuting during peak-hour periods; and land use controls to promote more efficient development. The Role of Enforcement Advances in automotive technologies have made it possible to dramatically lower emissions from motor vehicles. Initial crankcase HC controls were first introduced in the early 1960s followed by exhaust CO and HC standards later that decade. By the mid-1970s, most major industrial countries had initiated programs to control motor vehicle emissions. Today, it is well established that use of the existing state-of-the-art emission control technologies based on catalytic converters can substantially lower emissions from gasoline-fueled vehicles. Diesel particulate standards and control technologies have tended to lag but are starting to advance rapidly. Taking full advantage of these advances requires a carefully thought-out compliance program. For countries with no vehicle manufacturing industry, it can be mandated that vehicles Michael P. Walsh 127 being imported into the country have received a type approval certificate from their country of origin certifying that they achieve whatever limits are being imposed. In this case, no testing burden is placed on the country receiving the vehicles. At the other extreme, in countries with a developed vehicle manufacturing sector (such as Malaysia) that exports to highly industrialized countries, it would be possible to introduce a comprehensive compliance program. It should assure that attention to emission standards is paid at the vehicle design stage before mass production begins. It should also ensure quality assurance on the assembly line. And through an enforceable warranty and recall system, it should deter manufacture of nonconforming vehicles. Furthermore, vehicle owners should be encouraged to carry out maintenance on emission control devices as required by the manufacturer, and the service industry regulated to perform this maintenance properly. The cornerstone of an effective compliance program in most countries will be its vehicle I/M. I/M programs can improve emissions from vehicles equipped with virtually no pollution controls as well as from the most advanced systems. To put I/M in perspective, it is important to understand that today's cars are absolutely dependent on properly functioning emission controls to keep pollution levels low. Minor malfunctions in the emission control system can increase emissions significantly, and, as noted earlier, major malfunctions in the emission control system can cause emissions to skyrocket. By assuring good maintenance practices and discouraging tampering and misfueling, I/M remains the best demonstrated means for protecting a national investment in emission control technology and achieving the air quality gains that are needed. As a general matter, maximum I/M effectiveness occurs with centralized l/M systems using loaded mode tests. These programs are also much lower cost overall and more convenient to the public. However, as noted earlier, even simple idle tests can be helpful for older vehicles with limited or no vehicle pollution controls. Smoke checks of diesel vehicles on the road can also be very beneficial. VEHICLE COMPLIANCE AND ENFORCEMENT STRATEGIES Various enforcement tools are available to address each stage of a vehicle's life cycle. The certification or type approval process that requires testing of prototype cars prior to production can impact on vehicle design at low mileage and, to a limited degree, on the durability of emission controls. It can also impact on vehicle maintenance to the degree that review of manufacturer-proposed maintenance schedules can constrain the manufacturer from requiring excessive maintenance and thereby reduce the potential effectiveness of recall and warranty and from requiring less maintenance than was performed on his certification prototype. Some prototype maintenance, nevertheless, is not required to be recommended to the consumer. Its major advantage, that it impacts on vehicle design early in the design process before actual production begins, is also its major weakness, that it inherently deals with somewhat artificial, prototype cars in an artificial environment. By its very nature, therefore, it cannot address production problems nor deterioration due to age or real world driving and ambient extremes nor the amount and quality of maintenance that will actually be performed in-use. Assembly Line Testing requires testing of new production vehicles and is the only technique that can be used to assure before sale that vehicles when built are in fact meeting 128 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge emission standards; however, its impact on durability of design depends on requiring allowances for deterioration that are derived from other programs, such as certification or in-use testing. Further, like certification, it cannot impact on the amount or quality of maintenance performed in-use. The recall and warranty programs can provide some incentive to individuals to properly maintain their vehicles and are the only programs that can directly affect the actual in- use durability of vehicles. These programs are subject to the limitations of dealing only with properly maintained vehicles and a generally less-than-perfect response on the part of individual vehicle owners, and they would appear only to impact on vehicles, which have been polluting to an excessive degree, already in use and to be, therefore, remedies after the fact. Further, much of their potential effectiveness is lost after vehicles are in use more than one or two years. However, the major impact of these manufacturer-directed in-use programs is a significant deterrent to the design and/or manufacture of vehicles that will fail to comply initially or as a result of deterioration from actual use. Inspection and Maintenance (I/M) is the only compliance technique that assures that in-use vehicles are properly maintained. By requiring that vehicles pass a retest, it impacts directly on the quantity and quality of maintenance and also impacts on design through the warranty and recall programs that use I/M as a surveillance tool. I/M is probably the most effective "antitampering" program because of the intensive surveillance built into the periodic inspection. Such surveillance is particularly helpful in addressing vehicle maladjustments that cause vehicles to exceed standards. However, where I/M is not in effect, gross tampering by dealers may be substantially deterred by an aggressive program. The benefits of I/M, however, are limited by the adequacy of the short test used, the ability of the service industry to make proper repairs and the potential tampering that could occur following the test to allow the vehicle to emit high emissions throughout the year. It seems clear, therefore, that the ideal program must include all of the above elements. INSTITUTIONAL REQUIREMENTS While technical and policy solutions are available to dramatically lower emissions from vehicles, the greatest challenge is frequently to develop a workable institutional arrangement that blends a variety of government agencies and ministries at the local, regional and national levels while providing adequate opportunity for input from fuel suppliers and vehicle manufacturers, on the one hand, and nongovernmental organizations (NGOs) on the other. There can be no set formula as a practical arrangement must utilize the existing legal and governmental structure and shape it to address the problems. Key ingredients in any arrangement, however, would seem to be clear lines of authority and responsibility and a public decision-making process that provides opportunity for public input. Frequently, national authorities will assume the responsibility for setting appropriate air pollution targets or goals based on local studies, as well as information available from WHO among others. Either the Environment or Public Health Ministries or their equivalent are appropriate organizations to Michael P. Walsh 129 perform this role. In addition, national authorities will usually adopt minimum emissions standards for new vehicles and fuel quality; again environment ministries could perform this role although, in many countries, the vehicle standard setting is left to Transportation Ministries (with policy input from the Environment Ministry) and the fuel quality is defined by the Energy Ministry. Local governments usually have the authority to go beyond the national requirements depending on the air pollution levels in a given city. Strategies such as l/M and clean fuels are especially well-suited to a local focus. Transportation planning and land use control usually best lend themselves to regional approaches. A STRATEGY FOR PROGRESS IN CHINA The potential for reducing emissions and improving air quality must start with an assessment of existing vehicle emissions. Many factors affect the total inventory of motor vehicle emissions. Understanding these factors helps one to better structure a total inventory of these emissions and to determine optimal programs for their control. Having inventories that accurately reflect different control measures allows one to track the effectiveness of a given regulatory program. Some of the more important factors follow: * Emission Factors For New Vehicles. These emission factors can ideally be determined from the emission standards for new vehicles subject to regulation or from emission data available for similar vehicles (such as similar vehicle design, control technologies) used in other countries. * Deterioration of Vehicle Emissions With Vehicle Age and Mileage. Estimating how vehicle emissions deteriorate with time and mileage is critical in assessing in-use emissions. Different types of vehicles with different technologies (noncatalyst gasoline-fueled vehicles, catalyst-equipped, diesel-fueled, etc.) deteriorate differently with increased mileage and time. This type of information has been determined from testing of in-use vehicles carried out by several different countries. * Tampering Effects. This adjustment accounts for vehicle owners or drivers intentionally altering or disabling an emission control system. Examples are disconnecting air pumps, catalysts, evaporative emission control systems, exhaust- gas recirculating systems, and ignition timing. These vehicles are sometimes tampered with in the mistaken belief that vehicle performance, fuel economy or other factors (such as maintenance time or costs) will be improved. The effects of tampering on vehicle emissions can be obtained from data obtained from various countries. However, the incidence of vehicle tampering should be estimated for the particular country where the inventory is being developed. * Vehicle Maintenance. If a vehicle (with or without emission controls) is not maintained according to the manufacturer's recommendations (namely, tune-ups, replacement of spark plugs or emission control components, carburetor adjustments), the vehicle will have significantly higher emissions (that is, greater deterioration) than one properly maintained. These higher emissions must be accounted for. Some 130 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge initial estimates can be made based on similar information available from in-use vehicle emission levels measured in other countries. * Inspection and Maintenance and Antitampering Checks. The presence of an effective I/M program that identifies high-emitting vehicles and assures that they are repaired (or emission control systems replaced/repaired if tampered with) can help eliminate excess emissions resulting from the previous two factors. Estimates of the benefits of such programs are available from tests conducted in different countries. * Technology Mix. The fraction of vehicles using different technologies (such as diesel, noncatalyst, oxidation catalyst, three-way catalyst, etc.) is critical in estimating total vehicle emissions. The fraction of technologies used in a given country (and their appropriate emission factors) must be known or estimated. * Vehicle Age. The number of vehicles of a given age is important to know since older and/or higher-mileage vehicles usually have higher deterioration. This type of data is generally available from vehicle registration data obtained by most governments. - Number of Vehicles. The total number of vehicles of a given model year in a given area is generally obtained from vehicle registration data and must be known to calculate an inventory. A critical element is to make accurate projections for future years. - Vehicle Miles Traveled Per Vehicle Per Year. The number of miles a given type of vehicle travels per year must be known. This number usually varies with vehicle age, with older vehicles traveling fewer miles annually. The number of miles traveled will be different from one country to another. * Vehicle Misfueling. If both leaded and unleaded gasolines are sold in an area where some vehicles require the use of unleaded fuel to protect the catalytic converter, the fraction of catalyst vehicles that misfuel with leaded gasoline must be determined. * Fuel Characteristics. Fuel volatility can be an important determinant of vehicle evaporative emissions (which can account for as much as half of the total hydrocarbon emissions). Other fuel characteristics such as sulfur content, distillation characteristics, and oxygen content may also be important. * Ambient Temperature. The average daily temperature (generally maximum and minimum) must be known to predict vehicle emissions. Generally, separate inventories are calculated for warm-weather conditions (when ozone levels are at their peak) and cold-weather (when carbon monoxide levels are high). A typical breakdown of emissions for major urban areas in the Asia Pacific region would usually indicate that motorcycles are a major contributor to both hydrocarbons and organic particulate and a significant source of carbon monoxide and lead. Diesel vehicles are the major source of sulfate and a significant source of carbonaceous particulate. Passenger cars dominate the carbon monoxide and lead problems and contribute significantly to nitrogen oxides and Afichael P. Walsh 131 hydrocarbons. Therefore, different vehicle categories must be the focus of attention to address different aspects of the overall vehicle pollution problem. The major elements of an overall vehicle pollution control strategy were summarized earlier. In short, one must both reduce emissions per kilometer driven while simultaneously reducing the amount of driving. Emissions per kilometer driven can also be lowered by altering some aspects of the driving itself-average speed, degree of acceleration, etc. A natural and consistent tension exists between altering driving characteristics and reducing driving since frequently strategies designed to increase average speed by improving traffic flow actually enable a given roadway network to carry more vehicles per hour and effectively increase overall vehicle emissions. First priority, of course, should be directed at restraining future vehicle growth rates. Economic measures, physical restrictions and selective policies will, of course, each play a role. However, even if overall vehicle growth could be constrained to only 5 percent per year, well beyond the current average in many countries in the Region, vehicle emissions would explode over the next 15 years. In addition to growth restraint, therefore, a series of additional strategies are necessary. Inspection and Maintenance l/M programs have been demonstrated to lower emissions from existing vehicles in two ways: * By lowering emissions from vehicles that fail the test and are required to be repaired. * By encouraging owners of vehicles to take proper care of them and to avoid the potential costs of repairing vehicles that have been tampered with or misfueled. Based on all the data available, it is estimated that a well-run l/M program is capable of very significant emissions reductions, on the order of 25 percent for HC and CO and about 10 percent for NO,. The less significant NO, reductions reflect solely the lower tampering rates from I/M and antitampering programs since, at present, there has been no focused effort to specifically design I/M programs to identify and correct NO, problems. It is also important to note that the reductions start out slowly and gradually increase over time because l/M programs tend to lower the overall rate of fleet emissions deterioration. Maximum l/M benefits are thereby achieved by adopting the program as early as possible. Stringent Motorcycle Standards Two-stroke motorcycles are a major source of white smoke, hydrocarbons and particulate emissions. For example, as illustrated in Figure 19, in Bangkok two-stroke-powered motorcycles are estimated to emit almost as much particulate per kilometer driven as heavy duty diesel trucks and buses. 132 Theme Paper 2: Motor Vehicle Pollution Control in China: In Urban Challenge FIGURE 19: ORGANIC PARTICULATE Technologies available to control Emissions (Grams Per Kilometer) emissions from two- and three-wheeled vehicles are similar to those available for 2.71 other Otto cycle-powered engines. Reducing the content of lubricating oil in the fuel is one 1.7 - possible approach. Refining the fairly simple type of carburetors used would help 0.85 significantly reduce HC, CO, and smoke 0.1 0.12 0.26 emissions. Even catalytic converters are technologically feasible for these engines.35 cp"g5 GP. -es P 13 %W51 g.0C'\ (15o5 Many modern engines use a separated Vehicle Types lubrication system, which brings about overall leaner fuel/oil ratios and is therefore favorable for smoke reduction. Since 1986 mopeds with catalysts have been available in Switzerland and in Austria, and since 1992 motorcycles in Taiwan (China) have been similarly equipped. It is a fair conclusion to state today that the historical problems of high smoke and unburned hydrocarbons from two-stroke technology are no longer technologically necessary. New technology promises to resolve these concerns. As examples, direct cylinder electronic fuel injection, electronic computer control, and catalytic exhaust conversion are now commonplace solutions. In addition, modern, advanced two-stroke engines such as those under development from the Orbital company indicate that these engines can even be cleaner and more fuel-efficient than four strokes. Improved Fuel Quality Gasoline Throughout much of the industrialized world, unleaded fuel has been the norm for more than a decade. Japan has actually been the world leader in this regard, with more than 90 percent of the gasoline in that country being unleaded for almost two decades. In 1994, it is estimated that approximately two-thirds of all gasoline sold in the world was unleaded. Even when leaded fuel is used, the lead content should be reduced to no more than 0.15 grams per liter. Beyond the reduction or elimination of lead, it is possible to make additional modifications to gasoline, to "reformulate" it to reduce both regulated and unregulated emissions of concern. As part of a comprehensive policy to reduce vehicle emissions, fuel reformulation has the potential not only to offset any increased risks associated with the introduction of unleaded gasoline but to complement the elimination of lead health risks with an overall reduction of the toxic and ozone-forming potential of gasoline and gasoline vehicle emissions. 35 OECD [1988a]. Transport and Environment, Paris. Michael P. Walsh 133 The potential for "reformulating" gasoline to reduce pollutant emissions attracted considerable attention in the United States as pressure to shift to alternative fuels increased during the mid- to late-1980s. One result was a major cooperative research program between the oil and auto industries. During the early 1990s, this was followed by a similar effort in Europe. The result is that a great deal has been learned about the potential for modifying gasolines in a manner that can significantly improve air quality. An additional advantage of fuel reformulation is that it can reduce emissions from all vehicles on the road in much the same way that reducing lead in gasoline can reduce lead emissions from all vehicles. The most significant potential emission reductions that have been identified for gasoline "reformulation" have been through reducing volatility (to reduce evaporative emissions), reducing sulfur (to improve catalyst efficiency), and adding oxygenated blend stocks (with a corresponding reduction in the high-octane aromatic hydrocarbons that might otherwise be required). The potential benefits of improving various fuel parameters are summarized below. Lowering Volatility. Fuel volatility, as measured by Reid vapor pressure (RVP) has a marked effect on evaporative emissions from gasoline vehicles both with and without evaporative emission controls. Tests on vehicles without evaporative emission controls showed that increasing the fuel RVP from 9 pounds per square inch (psi) (62 kilipascals-kPa) to approximately 12 psi (82 kPa) roughly doubled evaporative emissions.)6 The percentage effect is even greater in controlled vehicles. In going from 9 psi (62 kPa) to 12 (81 kPa) RVP fuel, the US EPA found that average diurnal emissions in vehicles with evaporative controls increased by more than five times, and average hot-soak emissions by 25 to 100 percent.37 The large increase in diurnal emissions from controlled vehicles is due to saturation of the charcoal canister, which allows subsequent vapors to escape to the air. Vehicle refueling emissions are also strongly affected by fuel volatility. In a comparative test on the same vehicle, fuel with 11.5 psi (79 kPa) RVP produced 30 percent greater refueling emissions than gasoline with 10 psi (64 kPa) RVP (1.45 vs. 1.89 g/liter dispensed).38 In response to data such as these, the US EPA has established nationwide summertime RVP limits for gasoline. An important advantage of gasoline volatility controls is that they can affect emissions from vehicles already produced and in-use and from the gasoline distribution system. Unlike new-vehicle emissions standards, it is not necessary to wait for the fleet to turn over before they take effect. The emissions benefits and cost-effectiveness of lower volatility are greatest where few of the vehicles in use are equipped with evaporative controls. Even where evaporative 36 McArragher, J.S. et a]. 1988. Evaporative Emissionsfrom Modern European Vehicles and their Control. SAE Paper No. 880315. SAE International, Warrendale, Pennsylvania. 37 US EPA. 1987. Draft Regulatory Impact Analysis: Control ofGasoline l'olatility and Evaporative Hydrocarbon Emissions From New Motor Vehicles, Office of Mobile Sources, United Sates Environmental Protection Agency, Washington, DC. 38 Braddock, J.N. 1988. "Factors Influencing the Composition and Quantity of Passenger Car Refueling Emissions-Part 11." SAE Paper No. 880712. SAE International, Warrendale, Pennsylvania. 134 Theme Paper 2: Motor Vehicle Pollution Control in China. 4n Urban Challenge controls are in common use, as in the United States, control of volatility may still be beneficial to prevent in-use volatility levels from exceeding those for whichi the controls were designed. In its analysis of the RVP regulation, the US EPA (1987) estimated that the long-term refining costs of meeting a 9 psi (62 kPa) RVP limit throughout the United States would be approximately $0.0038 per liter, assuming crude oil at $20 per barrel. These costs were largely offset by credits for improved fuel economy and reduced fuel loss through evaporation, so that the net cost to the consumer was estimated at only $0.0012 per liter. Oxygenates. As noted earlier, blending small percentages of oxygenated compounds such as ethanol, methanol, tertiary butyl alcohol (TBA) and methyl tertiary butyl ether (MTBE) with gasoline has the effect of reducing volumetric energy content of the fuel, while improving the antiknock performance and thus making possible a potential reduction in lead and/or harmful aromatic compounds. Assuming no change in the settings of the fuel metering system, lowering the volumetric energy content will result in a leaner air-fuel mixture, thus helping to reduce exhaust CO and HC emissions. Impact of Oxygenate Used: MTBE. It appears that MTBE can be added to gasoline up to 2.7 percent without any increase in NO,,. There are two opposing effects taking place with the addition of oxygenates: enleanment, which tends to raise NO,, and lower flame temperatures, which tend to reduce NO,. With MTBE levels above 2.7 percent, the lower flame temperature effect seems to prevail. Impact of Oxygenate Used: Ethanol. Available data indicate that ethanol can be added to gasoline at levels as high as 2.1 percent oxygen without significantly increasing NO, levels but above that point levels could increase significantly. For example, EPA test data on over 100 cars indicates that oxygen levels of 2.7 percent or more could increase NO,, emissions by 3 to 4 percent.39 The auto/oil study concluded that there was a statistically significant increase in NOx of about 5 percent with the addition of 10 percent ethanol (3.5 percent 02). Impact of Oxygenate Used: ETBE. Ethyl rertiary Butyl Ether appears to be an attractive source of oxygenates but, unfortunately, to date, too little data exist regarding its NOx impact to make a reasonable judgment as to its impact. The auto/oil study found about a 6 percent increase in NO, but the results were not statistically significant. Other Fuel Variables: Sulfur. Lowering sulfur in gasoline lowers emissions of CO, HC and NO,, from catalyst-equipped cars. As noted by the auto-oil study, "The regression analysis showed that the sulfur effect (lowered emissions) was significant for HC on all ten cars, for CO on five cars, and for NO, on eight cars. There were no instances of a statistically significant increase in emissions."40 To the extent that oxygenates are sulfur-free, their addition would tend to traditionally lower gasoline sulfur levels. Based on the auto/oil study, it appears that NO, would go down about 3 percent per 100 parts per million sulfur reduction. 39 Personal Communication. 40 Auto/Oil Air Quality Improvement Research Program, Technical lBulletin No. 2, "Effects of Fuel Sulfur Levels on Mass Exhaust Emissions," February 1991. Michael P. Walsh 135 Other Fuel Variables: Other. According to the auto/oil study, "NO, emissions were lowered by reducing olefins, raised when Tgo was reduced, and only marginally increased when aromatics were lowered."4' In general, reducing aromatics and T90 caused statistically significant reductions in exhaust mass NMHC and CO emissions. Reducing olefins increases exhaust mass NMHC emissions; however, "the ozone forming potential" of the total vehicle emissions was reduced.42 With regard to toxics, the reduction of aromatics from 45 percent to 20 percent caused a 42 percent reduction in benzene but a 23 percent increase in formaldehyde, a 20 percent increase in acetaldehyde and about a 10 percent increase in 1,3-Butadiene. Reducing olefins from 20 percent to 5 percent brought about a 31 percent reduction in 1,3-Butadiene but had insignificant impacts on other toxics. Lowering the T90 from 360 to 280F resulted in statistically significant reductions in benzene, 1,3-Butadiene (37 percent), formaldehyde (27 percent) and acetaldehyde (23 percent). Cost Effectiveness. It is difficult to estimate the costs and the cost effectiveness of fuel modifications because refiners differ widely in terms of the characteristics of the fuels they produce. Individual fuel component control costs and the effects of changes in one fuel component on the other fuel components are integral parts in the determination of the cost effectiveness. In the US EPA's analysis, these two integral parts were estimated from the results of refinery modeling performed by Turner, Mason and Company (for the Auto-Oil Economics group) and Bonner & Moore Management Science (for EPA) and on survey results presented by the California Air Resources Board (CARB). The total cost (or manufacturing cost) of producing a reformulated gasoline is the sum of the capital recovery cost and the operating cost. An example of the individual fuel component costs and the associated incremental percent reduction in VOC emissions are shown in Table 3. EPA proposed a range of VOC standards and NO, standards based on particular combinations of fuel component controls that reduce VOC (and VOC plus NOJ) emissions at a cost of less than $5,000 and less than $10,000 per ton, respectively. EPA believes that these ranges represent the upper limit of costs that will be incurred by many ozone nonattainment areas in achieving attainment. Estimates of the costs and cost effectiveness of California RFG continue to come down. At the time it developed its regulations, CARB estimated the costs to be $0.12 to $0.17 per gallon. Recently, an EPA analysis placed the costs at $0.08 to $0.11 per gallon. This analysis estimated the cost effectiveness of the California RFG to be $4,100 to $5,100 per ton of VOC and NOx control; Federal phase 1 RFG was estimated to cost $3,100 per ton of VOC control.43 41 Auto/Oil Air Quality Improvement Research Program, Technical Bulletin No. 1, "Initial Mass Exhaust Emissions Results from Reformulated Gasolines," December 1990. 42 ,Auto/Oil Air Quality Improvement Research Program-What Is It and What Has It Learned?", Colucci and Wise, June 7, 1992, Presented at XXIV Fisita Congress, London, England. 43 "The Case for California Reformulated Gasoline-Adoption by the Northeast," Dr. R. Dwight Atkinson, May 1993. 136 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge Table 3: COMPONENT CONTROL COSTS AND VOC EMISSION REDUCTIONS Incremental Cost Cumulative VOC Component Control Level (c/gal) Reduction (%) Oxygen 2.0 Wt% 1.67-3.36/a 9.0 Benzene 1.0 vol% 0.69 9.0 RVP 8.1 psi 0.57 17.6 RVP 7.4 psi 1.67 25.3 Sulfur 160 ppm 0.35-0.57 26.4 Oxygen 2.7 Wt% 0.59-1.18/a 28.5 Olefins 5.0 vol% 1.81-2.44 30.2 Sulfur 50 ppm 1.45- 1.86 31.2 Aromatics 20 vol% 0.61-0.98 31.4 /a Based on MTBE. Diesel Fuel Modifications to diesel fuel composition have now also drawn considerable attention as a quick and cost-effective means of reducing emissions from existing vehicles. The two modifications that show the most promise are a reduction in sulfur content, and in the fraction of aromatic hydrocarbons in the fuel. Sulfur Content. In addition to a direct reduction in emissions of S02 and sulfate particles, reducing the sulfur content of diesel fuel reduces the indirect formation of sulfate particles from SO2 in the atmosphere. In Los Angeles, it is estimated that each pound of SO2 emitted results in roughly one pound of fine particulate matter in the atmosphere. In this case, therefore, the indirect particulate emissions due to SO2 from diesel vehicles are roughly as great as their direct particulate emissions. SO2 conversion to particulate matter is highly dependent on local meteorological conditions, however, so the effects could be greater or less in other cities. Aromatic Hydrocarbons. A reduction in the aromatic hydrocarbon content of diesel fuel may also help to reduce emissions, especially where fuel aromatic levels are high. For existing diesel engines, a reduction in aromatics from 35 percent to 20 percent by volume would be expected to reduce transient particulate emissions by 10 to 15 percent and NOx emissions by 5 to 10 percent. HC emissions, and possibly the mutagenic activity of the particulate soluble organic fraction (SOF), would also be reduced. Modeling studies of the refining industry have shown that aromatic reductions of this magnitude can often be obtained through alterations in diesel fuel production and blending strategy, without a need for major new investments in additional processing capacity. Reduced diesel fuel aromatic content would have other environmental and economic benefits. The reduced aromatic content would improve the fuel's ignition quality, improving cold starting and idling performance and reducing engine noise. The reduction in the use of catalytically cracked blending stocks should also have a beneficial effect on deposit-forming tendencies in the fuel injectors, reducing maintenance costs. On the negative side, however, the Michael P. Walsh 137 reduced aromatics might result in some impairment of cold flow properties, due to the increased paraffin content of the fuel. Fuel Additives. A number of well-controlled studies have demonstrated the ability of detergent additives in diesel fuel to prevent and remove injector-tip deposits, thus reducing smoke levels. The reduced smoke probably results in reduced particulate emissions as well, but this has not been demonstrated as clearly, due to the great expense of particulate emissions tests on in-use vehicles. Cetane-improving additives are also likely to result in some reduction in HC and particulate emissions in marginal fuels. Alternative Fuels For Buses The possibility of substituting cleaner-burning alternative fuels for diesel fuel has drawn increasing attention during the last decade. Motivations advanced for this substitution include conservation of oil products and energy security, as well as the reduction or elimination of particulate emissions and visible smoke. The principal alternative fuels presently under consideration are natural gas and methanol made from natural gas, and in limited applications, LPG. Whether to use alternative fuels and if so which fuel requires a detailed study of the costs of implementing and sustaining the fuel supply system in a given location. More Stringent Car and Truck Standards Advances in automotive technologies have made it possible to dramatically lower emissions from new motor vehicles. Increasingly, countries around the world have been taking advantage of them. Once good I/M programs and unleaded gasoline are introduced, state-of-the- art pollution controls should be pursued. WORLDWIDE OVERVIEW Historic Patterns of Vehicle Production and Use In 1950, there were about 53 FIGURE 20: GLOBAL TRENDS IN VEHICLE million cars on the world's roads; only four REGISTRATIONS decades later, the global automobile fleet is over 456 million, almost a tenfold increase. 800X000X000 800,000,000 On average, the fleet has grown by about 9.5 million automobiles per year over this 600,000,000 - 600,000,000 period. Simultaneously, as illustrated in Figure 20, the truck and bus fleet has been 400,000,000 - 400,000,000 growing by about 3.6 million vehicles per year. While the growth rate has slowed in 200,000,000 -- - 200,000,000 the highly industrialized countries, 0 0 population growth and increased 1930 1940 1960 1960 1970 1980 1990 urbanization and industrialization are 1935 1945 1965 1965 1975 1986 138 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge accelerating the use of motor vehicles elsewhere. If the approximately 100 million two-wheeled vehicles around the world are included (growing at about 4 million vehicles per year over the last decade), the global motor vehicle fleet is now approaching 700 million. Europe (including Eastem Europe and the former USSR) and North America each have about 30 percent of the world's motor vehicle population. The remainder is divided among Asia, South America, Africa, and Oceania (Australia, New Zealand and Guam), in that order. North America has about 40 percent of the world's trucks and buses, followed closely by Asia and then Europe. In terms of per capita motor vehicle registration for various regions, the United States, Japan, and Europe also account for the FIGURE 21: DIVISION OF VEHICLES AND PEOPLE lion's share of the ownership and use of AROUND THE WORLD motor vehicles. Indeed, as shown in Figure 21, the non-OECD countries of Africa, Asia 3% 32% (excluding Japan) and Latin America are * OECD NORTHAMERICA home to more than four-fifths of the O OECD ASIAPACIIC world's population, yet account for only REST OF THE WORLD 13% one-fourth of world motor vehicle registrations! PEOPLE VEHICLES Government Responses To The Problem Gasoline-Fueled Vehicles. Vehicle pollution control efforts reflect an approximately 30-year effort to date. Initial crankcase HC controls were first introduced in the early 1960s followed by exhaust CO and HC standards later that decade. By the early to mid-1970s, most major industrial countries had initiated some level of vehicle pollution control program (Figures 22-25). FIGURE 22: LIGHT-DUTY VEHICLE STANDARDS (HC Plus NOX Combined) Grams per Kilometer 1.2 GERMAN PROPOSAL 0.8 For The European Community 0.6 ". :*_.^. JAIPAN 0.4 EUROPEAN COMMUNITY 0.2~~~~~~~~~~~-- 0.2 ,d, t _ CALiFORNIA 0 1990 1992 1994 1996 1998 2000 2002 2004 1991 1993 1995 1997 1999 2001 2003 2005 The US, Europe and Japan Use Different Test Procedures Which InfluencesThe Actual Stringency of the Requirements Michael P. Walsh 139 FIGURE 23: PASSENGER CAR EMISSION STANDARDS (Diesel Particulate) Grams Per Kilometer 0.4 EUROPEAN COMMUNITY 0.3 US FEDERAL CALIFORNIA 0.2 . .- *~ S GERMAN PROPOSAL 0.1 . For The European Community & -- o--& - -& 4 -*- - 4VZZV4Z S Z 4 I#*: 0 1990 1992 1994 1996 1998 2000 2002 2004 1991 1993 1995 1997 1999 2001 2003 2005 The US and Europe Use Different Test Procedures Whichi Influences The Actual Stringency of the Requirements FIGURE 24: HEAVY TRUCK EMISSION STANDARDS (Diesel Particulate) Grams Per Kilowatt-Hour 0.4 F -~ nJAPAN (Possible) EUROPEAN COMMUNITY 0.3 -t L] ' N C '3 C L3I n Li Cl Li E9 USTRUCKS 0.2 - *. - . -4....V4.. _ _ _........... . ..-.4 US BUSES GERMAN PROPOSAL For The European Community 0 1993 1995 1997 1999 2001 2003 2005 - 1994 1996 1998 2000 2002 2004 The US, Europe and Japan Use Different Test Procedures Which Influences The Actual Stringency of the Requirements FIGURE 25: HEAVY TRUCK AND BUS EMISSION STANDARDS (Nitrogen Oxides) G ram s Per Kilowatt-Hour 2 0 UNITED STATES is E U R O P E A N C O M M U N ITY GERMAN PROPOSAL For The European community 5 _ a s | ~~~~~~~~~~~~~~~~~~~~~~~J A P A N o I I I I I I I l l l l | JAPAN (Possible) 1 990 1992 1994 199h 1998 2000 2002 2004 1 991 1 993 1 995 1 997 1 999 200 2003 2005 0 TheIUS, Europe and Japan Use DifferentTest Procedures W hich7 Can Influence The Actual Stringency of the Requirements _ _ .~~~~~~~~~~~~~~~~~~~~~~~~~~__ _- 140 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge During the mid- to late-1970s, advanced catalytic aftertreatment technologies were introduced on most new cars in the United States and Japan. During the mid-1980s, Austria, the Netherlands and the Federal Republic of Germany adopted innovative economic incentive approaches to encourage purchase of low-pollution vehicles. Australia, Canada, Finland, Austria, Norway, Sweden, Denmark and Switzerland all decided to adopt mandatory requirements. Even rapidly industrializing, developing countries or regions such as Brazil, Chile, Taiwan (China), Hong Kong, Mexico, Thailand, Singapore and Korea have adopted stringent emissions regulations. In 1990, the European Union decided to require all new light-duty vehicles sold within the European Community by 1992/93 to meet emission standards roughly equivalent to US 1987 levels. Further, in 1994, they voted to require a second step to go into effect by 1996. Just as Europe was moving toward parity with US standards, the United States and, to a much greater extent, California embarked on a course that could prove just as momentous to the 1990s as the 1970 Clean Air Act was to the 1970s and 1980s. With passage of the Clean Air Act amendments of 1990, the US Congress adopted requirements that will double the durability of passenger car emission control systems, tighten emissions standards, require cleaner conventional and alternative fuels, add cold-temperature standards and reduce toxic emissions. Simultaneously, to further reduce motor vehicle emissions, the California Air Resources Board (CARB) has established stringent new vehicle exhaust emission standards, the so-called Low-Emission Vehicle or LEV program. Compliance with these standards will be achieved through a combination of advanced vehicle emission control technology and clean-burning fuels. Diesel Vehicles and Engines. Not surprisingly, in view of the already serious health concerns and the continued growth in the use of diesel vehicles, many countries are pushing equally hard to reduce diesel emissions as gasoline vehicle emissions. The United States, Europe and Japan have all now adopted NO, and particulate standards that will require significant technological progress during the 1990s. Just as the 1970s represents the period when advanced gasoline vehicle technologies passed from the laboratory to the marketplace, the 1990s promises to be a similar era for diesels. Motorcycles. Stringent motorcycle standards were adopted in the United States several years ago but as the motorcycle population declined, especially those using two-stroke engines, concern with these vehicles has waned. Motorcycles were largely ignored in the European Union as well. However, in 1988 the Swiss government broke the logjam by deciding to set new emission standards for motorcycles to go into effect by October 1990. Shortly thereafter, on July 1, 1991, the motorcycle standards in Taiwan were tightened to 4.5 grams per kilometer for CO (from 8.8), and 3.0 for HC and NO,, combined (from 6.5), based on the ECE R40 test procedure. With introduction of these requirements, Taiwan had the most stringent motorcycle control in the world, requiring use of catalytic-converter controls on all new two-stroke motorcycles. As these .tight standards begin to spread during the 1990s, Taiwan will very likely institute even more stringent requirements in a few years. Zero-emissions motorcycle standards have been proposed in Taiwan for a limited portion of 1998 motorcycles. Michael P. Walsh 141 SUBSTANTIAL PROGRESS HAS OCCURRED IN MANY COUNTRIES Since 1970, the United States has had an aggressive effort underway to reduce emissions from cars and improve air quality. This program has combined many elements including the introduction of leaded gasoline, tight standards for new vehicles, in-use vehicle inspection and maintenance efforts, and most recently the use of reformulated and low-volatility gasoline. As a result, over the course of the past 25 years, the emission rate for on-highway cars in the United States has declined dramatically. As newer vehicles equipped with advanced emissions controls have replaced older, higher-polluting ones, there has been a clear downward trend in emissions of all three pollutants. This is especially encouraging in light of the continued rapid growth in vehicles and vehicle miles traveled by cars during this same period; in 1990 there were 50 million more cars on US highways than there were in 1970. Had emissions per mile not been reduced, passenger cars in 1990 would have emitted 65 percent more CO, HC and NO, than they did in 1970. In other words, as illustrated in Table 4, instead of passenger car CO having been reduced from 68 million metric tons to 27, these emissions would have climbed to 112 tons. Table 4: EMISSIONS TRENDS IN THE UNITED STATES (1970-90), PASSENGER CARS (Tons per year) Carbon monoxide Hydrocarbons Nitrogen oxides 1970 Actual 67.9 8.87 4.36 1990 Actual 26.9 2.65 2.34 1990 Potential /a 112.0 14.6 7.2 /a What would have occurred had pollution controls not been put on cars over this period. Figure 26 illustrates the auto emissions FIGURE 26: TRENDS IN EMISSIONS FROM US reductions to date, 60 percent for CO, 70 CARS (Normalized to 1970 Levels) percent for HC and 46 percent for NO,. Lead emissions from all highway vehicles have also been reduced dramatically; between 1970 and g - Ml TRA0EL0D 1993, highway vehicle lead emissions declined 200% from 171,960 short tons to 1,380. The point of this example is to show that adoption of a strong motor vehicle 10 i pollution control program can be very effective 5E% in reducing vehicle emissions. 0% Another example is the experience in .,,, I," Southern California's Los Angeles Basin, which has had the most aggressive motor vehicle pollution control program in the world over the past 40 years." From 1955 to 1993, peak ozone 44 "The Automobile, Air Pollution Regulation and the Economy of Southern California, 1965-1990," Jane Hall et al., Institute for Economic and Environmental Studies, California State University, April 1995. 142 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge concentrations were cut in half. The number of days on which federal ozone standards are exceeded fell by 50 percent from the 1976-78 timeframe to the 1991-93 interval. Further, the average annual number of days above the federal carbon monoxide standard fell from 30 to 4.3 during this same period and lead levels are now 98 percent lower than in the early 1970s. Most remarkably, this achievement occurred while the regional economy outpaced the national economy in total job growth, manufacturing job growth, wage levels and average household income. In short, a strong focus on environmental protection is not only not incompatible with strong economic development, they seem to be mutually reinforcing. VEHICLE POLLUTION CONTROL EFFORTS UNDERWAY IN ASIA In recent years, the most rapidly growing area of the world has been Asia. Not surprising, therefore, it already has a serious air pollution problem in many of its major cities. This review will summarize the air pollution situation, as well as the steps being adopted by several countries in the region. Several developing countries of Asia have made progress with some or all elements of these strategies; specific examples illustrating these efforts will be summarized below. Bangkok, Thailand Based on a review of available air quality data, it is estimated that roadside emissions of particulate, carbon monoxide and lead must be reduced by 85, 47 and 13 percent, respectively, if acceptable air quality is to be achieved in Bangkok. Recent data indicate that ozone levels downwind of the city may also be approaching unhealthy levels; therefore, it seems prudent to adopt measures which will reduce HC and NO, emissions, the ozone precursors, as well. In response to the serious air pollution threat, Thailand's current Seventh Plan has placed a high priority on improving air quality and definite targets have been set to control the amount of suspended particulate matter, carbon monoxide and lead on Bangkok's major streets. Current Program. A number of measures have been adopted to mitigate air pollution problems, particularly those caused by the transport sector. They are aimed not only at exhaust- gas emission controls but also at the improvement of fuel and engine specifications, implementation of in-use vehicle inspection and maintenance program, public transport improvement through mass transit systems, and the improvement of traffic conditions through better traffic management. Measures directed toward reducing vehicle emissions include: * Introduction of unleaded gasoline at prices below that of leaded gasoline (introduced in May 1991). * Reduction of the maximum allowable lead in gasoline from 0.4 to 0.15 grams per liter (effective as of January 1, 1992). * A plan to phase out leaded gasoline by 1996 (some companies have already eliminated sales of leaded gasoline). * Reduction of the sulfur content of diesel fuel from 1.0 to 0.5 percent as of April 1992 in the Bangkok Metropolitan Area and after September 1992 throughout the whole Michael P. Walsh 143 country; the use of low-sulfur diesel fuel has been mandatory in Bangkok since September 1993. * Reduction of the 90 percent distillation temperature of diesel fuel from 370°C to 357°C as of April 1992 in the Bangkok Metropolitan Area and after September 1992 througliout the whole country. * Required all new cars with engines larger than 1600 cc to meet the ECE R83 standards after January 1993; all cars were required to comply after September 1, 1993. * Taxis and Tuk-Tuks have already been largely converted to operate on LPG. * ECE R40 requirements for motorcycles were introduced in August 1993 and followed soon afterward by ECE R40.01; the Government has decided on a third step of control, which will be phased in this year. * ECE R49.01 standards for heavy-duty diesel-engine vehicles are now in effect. • The Government has decided to reduce the sulfur level in diesel fuel from the current 0.5 Wt. percent to 0.25 by 1996 and 0.05 by the year 2000. Currently, noise and emission testing are required and are conducted under the Land Transport Department's general vehicle inspection program. All new vehicles are subject to such inspection. For in-use vehicles, only those registered under the Land Transport Act (buses and heavy-duty trucks) and commercial vehicles registered under the Motor Vehicles Act (taxis, Tuk-Tuks and rental vehicles) are subject to inspection during annual registration renewals. It is expected that Land Transport Department (LTD) will require all in-use vehicles to be inspected soon. Vehicles in use for 10 or more years are subjected to an annual inspection while the newer vehicles will be subjected to inspection at different time periods. This will be determined by LTD. Private inspection centers are being licensed. Future Plans. Further investigations are underway to introduce more stringent standards for motorcycles as well as light and heavy trucks, and to purchase 200 CNG buses to reduce the smoke problem. A comprehensive motor vehicle pollution control strategy is being designed for Bangkok. The most critical data needs appear to be those related to motorcycle and diesel- vehicle particulate emissions factors. Unfortunately, it appears that locally generated data in this area is at least a year away. Further, better characterization of the particulate would be very helpful. In addition, as the new air quality monitoring network gets deployed, it will be critical to periodically update the air quality targets. Conclusions. Bangkok, like many other megacities in the world, has serious problems associated with the use of energy in transport sector. Several factors, including population growth and rapid economic expansion, etc., are fundamental factors needed to be considered for long-term planning. Rapid industrialization and urbanization, coupled with the lack of land use planing in the past, has contributed to the atmospheric pollution associated with the transport 144 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge sector. This problem has been intensified by the inadequate road infrastructures to absorb the rapidly growing vehicle population, which in turn causes congestion, and by the lack of mass transport system to offer good substitutes for private vehicles. These two factors encourage people to rely more on their private vehicles and hence have further contributed to the congestion problem. It is recognized that this problem can be alleviated through several means including the following measures: source reduction through improvement of fuel quality, I/M program, vehicle standards, and traffic and demand management (such as having a good mass rapid transit system). A great deal of work remains to be done, especially in the policy arena to control travel demand (demand-side management). Singapore In Singapore, motor vehicle emissions are a significant source of air pollution. The vehicle population has been steadily increasing over the past decade as a consequence of rapid urbanization and economic growth. At the beginning of 1993, the vehicle population stood at approximately 550,000. Land Transport Policy. Singapore's land transport policy strives to provide free- flowing traffic within the constraint of limited land. A four-pronged approach has been adopted to achieve this. First, the need to travel is minimized through systematic town planning. Second, an extensive and comprehensive network of roads and expressways, augmented by traffic management measures, has been built to provide quick accessibility to all parts of Singapore. Third, a viable and efficient public transport system that integrates both the Mass Rapid Transit (MRT) and bus services, is promoted. Finally, the growth and usage of vehicles are managed to prevent congestion on the road. Mobile Source Controls. Singapore's strategy for reducing pollution from motor vehicles is two-pronged: improving the engines and fuel quality to reduce emissions and using traffic management measures to control the growth of vehicle population and fuel consumption. The Pollution Control Department works closely with the Registry of Vehicles to implement the two-pronged strategy. * Between 1981 and 1987, the lead content in leaded gasoline was gradually reduced from 0.8 to 0.15 grams per liter. The use of unleaded gasoline was promoted in February 1990 through a differential tax system that made unleaded gasoline 10 cents per liter cheaper than leaded gasoline at the pump. All gasoline-driven vehicles registered for use in Singapore after July 1, 1991 must be able to use unleaded gasoline. These measures have resulted in the greater use of unleaded gasoline. About 57 percent of all gasoline sold in Singapore at the end of 1993 was unleaded. The sulfur content in diesel is currently limited to 0.5 percent by weight and will be reduced to 0.3 percent by weight from July 1, 1996 onwards. * The emission standards for gasoline vehicles have been progressively tightened since 1984 and the standards currently in force are the European Union Consolidated Emissions Directive 91/441 and the Japanese emission standards (Article 31 of Safety Regulations for Road Vehicles). Michael P. Walsh 145 * Since October 1992, motorcycles and scooters have been required to comply with the emission standards stipulated in the US Code of Federal Regulation 86.410-80 before they can be registered for use in Singapore. * Since January 1991, all diesel vehicles have been required to comply with smoke standards stipulated in the UN/ECE Regulation No. 24.03 before they can be registered for use in Singapore. * All in-use vehicles are required to undergo periodic inspections to check their roadworthiness and exhaust emissions while idling. Vehicles that fail the inspections are not allowed to renew their road tax. Traffic Management Measures. The situation in Singapore is a unique one. Singapore is essentially a city-state with a large population living on a small land mass. Urbanization, industrialization and infrastructural development are still progressing in earnest, fueled by a growing economy. With such a combination of factors, it is easy to see that there is a potential for serious environmental problems from both stationary and mobile sources if the sources are not managed or controlled properly. In the case of motor vehicles, the need to control their impact on traffic flow and the environment has given rise to a unique set of traffic management measures. (a) Vehicle Registration and Licensing, The expense of owning and operating a vehicle in Singapore has served as a dampener to the growth in the vehicle population. Car owners wishing to register their cars must pay a 45 percent import duty on the car's open-market value (OMV), a registration fee of $1,000 for a private car ($5,000 for a company-registered car) and an Additional Registration Fee (ARF) of 150 percent of the OMV. In addition, car owners pay annual road taxes based on the engine capacity of their vehicle. The road tax of company-registered cars is twice as high as for individuals. For diesel vehicles, a diesel tax that is six times the road tax of an equivalent gasoline vehicle is payable. To encourage people to replace their old cars with newer, more efficient models, a Preferential Additional Registration Fee (PARF) system was iltroduced in 1975. Private car owners who replace their cars within 10 years are given PARF benefits that they can use to offset the registration fees they have to pay for their new cars. For cars registered on or after November 1, 1990, the PARF benefits would vary according to the age of the vehicle at deregistration. For cars registered before November 1, 1990, a fixed PARF benefit would be given upon deregistration based on the engine capacity of the car. To provide a higher PARF benefit to car owners who deregister their cars before 10 years, all PARF-eligible cars registered on or after November 1, 1990 receive higher fees if the vehicle is newer. (b) Vehicle Quota System. As high taxes alone would not ensure that the vehicle population grow at an acceptable rate, a vehicle quota system was introduced to achieve that objective. Since May 1, 1990, any person who wishes to register a vehicle must first obtain a vehicle entitlement in the appropriate vehicle class, through bidding. Tender for specified number of vehicle entitlements is conducted monthly. Successful bidders pay the lowest successful bid price of the respective category in which they bid. A vehicle entitlement is valid for 10 years from the date of registration of the vehicle. On expiration of the vehicle entitlement, 146 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge if the owner wishes to continue using the vehicle, he needs to revalidate the entitlement for another five or ten years by paying a revalidation fee (pegged at the 50 percent or 100 percent of the prevailing quota premium, respectively). (c) Weekend Car Scheme. The weekend car scheme was introduced on May 1, 1991 to allow more people to own private cars without adding to traffic congestion during peak hours. Cars registered under the scheme enjoy substantial tax concessions that include a 70 percent reduction in road tax and a tax rebate of up to a maximum of $15,000 on registration. Weekend cars are identifiable by their red license plates, fixed in place with a tamper-evident seal. They can only be driven between 7 pm and 7 am during the week, after 3 pm on Saturdays and all day on Sundays and public holidays. Weekend cars can be driven outside those hours but owners must display a special day license. Each weekend car owner is given five free day licenses per year and can buy additional ones at $20 each. (d) Area Licensing Scheme. The Area Licensing Scheme (ALS) was introduced in June 1975 to reduce traffic congestion in the city area during the peak hours. Only passenger cars were affected then. The scheme has gradually been modified to include all vehicles except ambulances, fire engines, policy vehicles and public buses. (e) Public Transportation. Public transport in Singapore is widely available and includes a mass rapid transit (MRT) system, a comprehensive bus network and over 13,000 taxis. Conclusions. Besides technical control measures (controls on engines and fuel quality), the use of traffic control measures has significantly contributed to the protection of the air quality in Singapore. Although the present measures appear to be adequate, Singapore will continue to look ahead for ways to improve them further. Pilot studies of three electronic road pricing systems are being carried out in Singapore and the most suitable system will be selected for implementation in 1997. Hong Kong Hong Kong's vehicle pollution control effort continues to focus on diesel particulate control because particulate is the most serious pollution problem at present in Hong Kong, and motor vehicles are estimated to be responsible for approximately 50 percent of the PM-10 emissions. Current Program. * With regard to diesel fuel, as of April 1, 1995, the sulfur level was reduced to 0.2 percent and it is planned to lower it to 0.05 percent by 1997 or 1998. * Diesel vehicle emissions standards were also tightened on April 1, 1995. All new passenger cars and taxis after that date must comply with either the US 1990 standards (PM=0.12 grams per kilometer, NO,=0.63) or the European Union Step I standards (93/59/EEC PM=0.14, HC+NO,=0.97) or the Japanese standards (PM=0.34, NO,=0.72 for vehicles weighing less than 1.265 ton or 0.84 for those above). Similar requirements will apply to all light and medium goods vehicles and light buses. For goods vehicles and buses with a design weight of 3.5 tons or more, Michael P. Walsh 147 either the 1990 US (PM=0.80 g/kWh, NO, 8.04) or the EURO I standards (PM=0.6 1 for engines producing less than 85 kW or 0.36 for engines producing more; NO,=8.0 for all engines) will apply. * In use smoke limits based on the EEC free acceleration test (72/306/EEC) will be lowered to 50 Hartridge Smoke Units (HSU); in certification, the limits will be 40 HSU. * Encouraged by a price differential of HK$1 per liter price reduction for unleaded gasoline compared to leaded, unleaded gasoline is now responsible for 71 percent of total gasoline sales. Notably, the benzene content of the unleaded gasoline is only 3.44 percent, virtually the same as leaded gasoline. Future Plans. An analysis of the motor vehicle related urban particulate problem indicates that 17 percent comes from buses, 63 percent from goods vehicles and the remainder from all vehicles under 5.5 tons. * As a matter of policy, Hong Kong is still trying to convert all light duty diesel vehicles including taxis to gasoline. Analyses are also being carried out regarding the possibility of converting some or all taxicabs to either CNG or electric. * With regard to I/M, the Government still has plans to introduce a mandatory program by May 1996. * Hong Kong also remains interested in the possibility of retrofitting buses with either catalysts or diesel particulate filters. They have submitted a proposal to the Asia-US partnership to fund such an effort and have also initiated discussions with potential suppliers in Europe. Korea A series of recent amendments in the Air Quality Control Law will gradually tighten Korea's vehicle emissions standards as summarized in Tables 5 and 6. The sulfur level in diesel fuel was reduced to a maximum of 0.4 Wt. percent during the period from February 2, 1991 through December 31, 1992; to 0.2 during the period from January 1, 1993 through December 31, 1995; and 0. 1 thereafter. Korea is also investigating possible improvements to their l/M program including the possible addition of I/M240. Research remains active in the use of diesel particulate filters. Three types of approaches are under investigation-burner systems that are seen as prime candidates for large vehicles; electrically heated systems that are seen as prime candidates for medium-size vehicles, and Cerium fuel-additive systems that are seen as the prime candidates for smaller vehicles. Research is also underway in Korea on electrically heated catalysts, CNG engines, two- stroke engines and lean NO, catalysts. 148 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge TABLE 5: EMISSION STANDARDS FOR NEW GASOLINE AND LPG VEHICLES Date of Imple- Exhaust Evap HC Vehicle Type mentation Test CO NO, HC (g/test) Small Size CarL a 1987 7/1 CVS-75 g/km 8.0 1.5 2.1 4.0 "1 2000 7/1 CVS-75 2.11 0.62 0.25 2.0 Passenger Car 1980 1/1 10-Mode 26.0 3.0 3.8 - 1984 7/1 10-Mode 18.0 2.5 2.8 - 1987 7/1 CVS-75 2.11 0.62 0.25 2.0 2000 1/1 CVS-75 2.11 0.25 0.16 2.0 Light Duty Truck L 1987 7/1 CVS-75 6.21 1.43 0.50 2.0 2000 1/1 CVS-75/c 2.11 0.62 0.25 2.0 2000 1/1 CVS-75 /d 6.21 1.43 0.50 2.0 Heavy Duty Vehicle 1980 1/1 6-Mode 1.6% 2,200 ppm 520 ppm 1987 7/1 US Transient 15.5 10.7 1.3 4.0 19912/1 13 Mode 33.5 11.4 1.3 2000 2/1 13 Mode 33.5 5.5 1.4 La Less than 800 cc of engine displacement. /b Gross vehicle weight of <3 tons. /c Gross vehicle weight of <2 tons. Ld Gross vehicle weight of between 2 and 3 tons. TABLE 6: EMISSIONS STANDARDS FOR NEW DIESEL VEHICLES Date of Imple- Vehicle Type mentation Test CO NO, HC PM Smoke Passenger Car 1980 1/1 Full Load - - - - 50% 1984 7/1 6-Mode 980 ppm 1,000/590/a 670 - 50% 1988 1/1 6-Mode 980 850/450 670 - 50% 1993 1/1 CVS-75 2.11 0.62 0.25 0.12 - 1996 1/1 CVS-75 2.11 0.62 0.25 0.08 2000 1/1 CVS-75 2.11 0.62 0.25 0.05 - Light Duty Truck 1980 1/1 Full Load - - - - 50% 1984 7/1 6-Mode 980 1000/590 670 - 50% 1988 1/1 6-Mode 980 850/460 670 - 50% 1993 1/1 6-Mode 980 750/350 670 - 40% 1996 1/1 CVS-75 6.21 1.43 0.5 0.31 - (0. 16)Lb Light Duty Truck <2 Tons 2000 1/1 CVS-75 2.11 0.75 0.25 0.12 All Other Light Duty Trucks 2000 1/1 CVS-75 6.21 1.00 0.5 0.16 - Heavy Duty Vehicle 1980 1/1 Full Load - - - - 50% 1984 7/1 6-Mode 980 1000/590 670 - 50% 1988 1/1 6-Mode 980 850/450 670 - 50% 1993 1/1 6-Mode 980 750/350 670 - 40% 1996 1/1 13-Mode 4.9 11.0 1.2 0.9 35% 2000 1/1 13 Mode 4.9 6.0 1.2 0.25 25% 0.1/c La Direct injection/indirect injection. lb Gross vehicle weight <2 tons. L/ City bus only. Michael P. Walsh 149 Taiwan (China) The Taiwan EPA has developed a comprehensive approach to motor vehicle pollution control. Building on its early adoption of US 1983 standards for light-duty vehicles (starting July 1, 1990) it recently moved to US 1987 requirements, which include the 0.2 gram per mile particulate standard, as of July 1, 1995. Heavy-duty diesel particulate standards almost as stringent as US 1990, 6.0 grams per brake horsepower hour NO, and 0.7 particulate, using the US transient test procedure, went into effect on July 1, 1993. It is intended that US 1994 standards, 5.0 NO, and 0.25 particulate, will be adopted soon, probably for introduction by July 1, 1997. Diesel fuel currently contains 0.3 weight percent sulfur. A proposal to reduce levels to 0.05 percent by 1997 is currently under consideration. The Executive Yuan on December 10, 1992 approved increases of up to 1,700 percent for the amount of fines to be levied against motorists who violate the Air Pollution Control Act. The new fine schedule raises the former maximum fine for motor vehicle pollution from $138 to $2,357. All forms of motorized transportation are included in the new fine schedule, including airplanes, boats, and power water skis. The new fines took effect in early 1993 after official public notice. Clearly, the most distinctive feature of the Taiwan program, however, is its motorcycle control effort, reflecting the fact that motorcycles dominate the vehicle fleet and are a substantial source of emissions. * The first standards for new motorcycles were imposed in 1984; 8.8 grams per kilometer for CO and 6.5 grams per kilometer for HC plus NO,, combined, using the ECE R40 test procedure. * In 1991, the limits were reduced to 4.5 grams per kilometer for CO, and 3.0 for HC and NO, combined. These requirements were phased in over two years and by July 1, 1993 were applied to all new motorcycles sold in Taiwan. As a result of these requirements, the engines of four-stroke motorcycles have been redesigned to use secondary air injection. All new two-stroke motorcycles are fitted with catalytic converters. * Since 1992, electric motorcycles have been available in the market but sales have been modest. * Motorcycle durability requirements have been imposed since 1991. All new motorcycles tested since that time are required to demonstrate that they can meet emissions standards for a minimum of 6,000 kilometers. * Since 1991, all new motorcycles must be equipped with evaporative controls. * In order to reduce the pollution from in-use motorcycles, EPA is actively promoting a motorcycle I/M system. In the first phase, from February through May 1993, EPA tested approximately 113,000 motorcycles in Taipei City. Of these, 49 percent were 150 Theme Paper 2: Motor Vehicle Pollution Control in China: An Urban Challenge given a blue card indicating that they were clean, 21 percent a yellow card indicating that their emissions were marginal, and 30 percent were failed. * Between December 1993 and May 1994, approximately 142,000 motorcycles were inspected, with 55 percent receiving blue cards, up 6 percent from the earlier program, and 27 percent failed, a drop of 3 percent. The major repair for failing motorcycles was replacement of the air filter at an average cost of $20. * In continuing regulations for the control of motorcycle emissions, EPA has adopted the Third-Stage Emission Regulation to be implemented from 1998. The new standards will lower CO to 3.5 grams per kilometer, and HC plus NO, to 2. In addition, the durability requirement will be increased to 20,000 kilometers. Finally, the market share for electric-powered motorcycles will be mandated at 5 percent. In addition, EPA will extend the periodic motorcycle I/M program. Conclusions As the above examples illustrate, substantial efforts have been and continue to be underway throughout many Asian countries to address their motor vehicle pollution problems. Several conclusions can be drawn from these efforts: * Several comprehensive motor vehicle pollution control programs have been developed in the region. * A wide variety of strategies are being implemented, tailored to the particular problems and capabilities in a particular country or city-one size does not fit all. * In virtually every serious effort to reduce motor vehicle pollution, cleaner fuels- especially unleaded gasoline and lower-sulfur diesel fuel-play a critical role. CONCLUSIONS The development of petroleum-powered motor vehicles has truly revolutionized society over the past century. The benefits of increased personal mobility and access to goods and services previously beyond the grasp of individuals cannot be denied. And, yet, the relentless growth in motor vehicle use has a dark downside that many have been slow to acknowledge, including a broad array of adverse public health and environmental impacts. The environmental damages caused by motor vehicle emissions are no longer debatable, and on a global basis they are increasing. The cars, trucks, and buses that make life better in so many ways emit more than 800 million tons of carbon per year. From their tailpipes come virtually all of the carbon monoxide in the air of our cities. Less directly, they cause much of the ozone and smog. And motor vehicles play a significant role in stratospheric ozone depletion. All of these pollutants contribute directly or indirectly to global warming. Over the last 40 years, the global vehicle fleet has grown from under 50 million to more than 500 million, and there is every indication that this growth will continue. Over the next 20 years, the global fleet could double to I billion. Unless transportation technology and planning are fundamentally transformed, emissions of greenhouse and other polluting gases from these Michael P. Walsh 151 vehicles will continue to increase, many relatively clean environments will deteriorate, and the few areas that have made progress will see some of their gains eroded. The worldwide challenges that these problems pose for motor vehicle manufacturers and policymakers are unprecedented. Nothing less than a revolution in technology and thinking at least as profound as the initial mechanization of transportation is needed. Manufacturers will come under increasing pressures to produce petroleum-powered vehicles that are ever cleaner, safer, more reliable, and more fuel-efficient. At the same time, they will need to develop new kinds of vehicles that will emit no pollution whatever. The amount of capital needed to accomplish these goals will be large and, making matters even more difficult, the pressures for these changes will arise not so much from traditional market forces but from public policies adopted in response to climate and other threats. While development of appropriate policies and technologies develops, countries can benefit from adoption of currently available policies and technologies. Various steps can be taken to reduce air pollution emissions from motor vehicles. These include incentives to remove older, higher-polluting vehicles from the road; tightening new vehicle emission standards for nitrogen oxides, volatile organic compounds, and carbon monoxide; developing and using cleaner fuels with lower volatility and fewer toxic components; enhancing inspection and maintenance (I&M) programs, including inspections of antitampering emission-control equipment; and extending the useful life for pollution-control equipment to 10 years or 100,000 miles rather than the current 5 years or 50,000 miles. The potential overall impacts of tighter standards, enhanced I&M, and extended useful life are especially significant because they help ensure that the benefits of clean-air technology will persist for the vehicle's full life. Additional reductions in vehicular emissions can be achieved by reducing dependence on individual cars and trucks and by making greater use of van and car pools, buses, trolleys, and trains. Improving urban traffic management by installing synchronized traffic lights, reducing on-street parking, switching to "smart" roads, banning truck unloading during the day, and so forth can also improve transportation system efficiency. Providing efficient, convenient, and affordable public transportation alternatives worldwide would produce multiple benefits. For every 40 persons who get out of their cars and onto a bus for a 10-mile trip to work, some 50 to 75 pounds of carbon are not emitted to the air. Greater use of public transportation would reduce congestion, cut fatalities and injuries from traffic accidents, and greatly improve air quality. Fortunately, such transportation improvements can be phased in over time. For example, roadways initially dedicated to bus traffic can later be upgraded to light rail or heavy rail if circumstances warrant. Wu Yong, Wang Jianqing, and Yao Zukang 153 THEME PAPER 3: MUNICIPAL TRANSPORT MANAGEMENT: A DOMESTIC VIEW WU YONG, WANG JIANQING, AND YAO ZUKANG I Transport is one of the basic demands of our daily life. With the growth of urbanization, all social activities, such as industry, commerce, technology, culture, education, public health and so forth, are increasingly concentrated in the urban area. Urban transport thus becomes one of the basic demands of modern society. It has two distinctive characteristics. First, demand for transport is desired by every member of a society. Second, it is only a derivative of other demands. Transport in its own right is not a final product for consumption but primarily a means to link the consumers and producers of goods and services together. Because of these characteristics, urban transportation is one of the areas of heavy government intervention around the world. Governments intervene to solve urban transportation problems and to meet the increasing demand. Their efforts aim at achieving efficiency in resources allocation, increasing productivity of the society, improving quality of living in urban residential areas and enhancing social equality. The form, degree and instruments of government intervention significantly influence the scale and level of urban public transportation development and the quality of urban transport services provided to different social sectors and residents; in particular, the low-income residents. Through many years practice and efforts, Chinese government has accumulated some experience in urban transport planning, construction, management, and regulation and policy- making. As China is now in transition from a centrally-planned-economy to a market economy, many changes have been taken place in the economic mechanism, sector structure and people's daily life. Accordingly, the structure and practice of urban transport management, formalized in the long period of the planned economy, are also in the process of transformation. They have to be reestablished based upon market principles. In the meantime, China's GNP has been growing at a 10 percent annual average rate since the reform and opening to the outside world. The living standard of city residents has been rising rapidly. In addition, the urbanization in China reached 22 percent in 1994 from a low 8.3 percent in 1978. Next only to India, China has now the second largest urban population in the world. Those factors contribute to a sharp rise in demand for urban transport. The volume of urban public passenger-transit increased from 13.2 billion passengers in 1978 to 29.2 billion passengers in 1994. The number of motor vehicles in urban areas increased from 9.5 million in 1978 to 5.6 billion in 1994. The rapid growth of economy and society, together with the reform and the open-door policy, pose a new challenge to the urban transport sector and provide an excellent opportunity for speeding up the urban Wu Yong is Division Chief, Comprehensive Planning Division, Urban Construction Department, Ministry of Construction; Wang Jianqing is Economist, Urban Construction Department, Ministry of Construction; and Yao Zukang is Professor, Department of Road and Traffic Engineering, Tongji University. 154 Theme Paper 3: Municipal Transport Management: A Domestic View transportation development as well. It is therefore very important to establish an urban transport management system fitting into China's development stage and country specifics, and to learn from other countries. Only by doing so will China be able to meet its increasing demand for urban transport to a reasonable degree and to ensure a healthy development of urban economy and society. This paper first gives a brief introduction to the existing urban transport management system in China; then it accesses the role of government in and its policy impact on urban transportation. Finally, it discusses the reform of urban transport management system. THE EXISTING URBAN TRANSPORT MANAGEMENT SYSTEM Urban transport is an integrated part of our society and economy. It can be further divided into two subsectors: intracity transport and intercity transport. This paper focuses on intracity transport. Intracity transport consists of three parts: - Infrastructure, including roads, rails, bridges and so on. * Urban transport operation systems. They include: (a) passenger transport by