China - Railway strategy

Report No. 10592-CHA China's Railway Strategy February 25, 1993 Transport Operations Division China and Mongolia Department East Asia and Pacific Regional Office FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (As of January 1993) Currency name = Renminbi Currency unit = Yuan (Y) = 100 Fen $1.00 = Y 5.75 $0.174 = Y 1.00 FISCA'. YEAR January 1 - December 31 WEIGHTS AND MEASURES ctkm = converted t"m or traffic unit (1 pkm = 1 tkm) km - kilometer [- 0.621 mile (mi)] mp = million passengers mpkm = million passenger-kilometers mppy 5 million passengers per year mt million tons mtkm - million ton-kilometers mtpy = million tons per year ntkm = net ton-kilometer pkm = passenger-kilometer (= 0.621 passenger-mile) tkm = ton-kilometer (= 0.621 ton-mile) A1BREVIATIONS AND ACRONYMS 6FYP - Sixth Five-Year Plan (1981-85) 7FYP - Seventh Five-Year Plan (1986-90) 8FYP - Eighth Five-Year Plan (1991-95) 9FYP - Ninth Five-Year Plan (1996-2000) ECO - Expanded Cofinancing Operation GDP - Gross Domestic Product GNP - Gross National Product MIS - Management Information System MOF - Ministry of Finance MOR - Ministry of Railway MPT - Ministry of Post and Telecommunications OECF - Overseas Economic Cooperation Fund PSO - Public Service Obligation RIS - Railway Investment Study SPC - State Planning Commission TMIS - Transport Management Information System UNDP - United Nations Development Program  FOR OFFICIAL USE ONLY CHINA'S RAILWAY STRATEGY Table of Contents Page No. EXECUTIVE SUMMARY . . . . . . . . . . . . . . . v I. TRANSPORT DEMAND AND SUPPLY . . . . . . . . . . . . . . . . . . 1 A. Past Trends of Transport Demand . . . . . . . . . . . . . . 1 Traffic . . . . . . . . . . . . . . . . . . . . . . . 1 Unsatisfied Demand . . . . . . . . . . . . . . . . . . . 2 Total Demand . ...... . . . . . . . .. . . . 4 B. Future Demand for Transport . . . . . . . . . . . . . . . . 4 Total Transport Demand .... . . . . . . . . . . .. 4 Railway Transport Demand .way.ra..o . .t.D . .m.a.n.d. 5 C. Development of the Transport System . . . . . . . . . . . 5 Size of the Transport System . . . . . . . . . . . . . . 5 System Bottlenecks . . . . . . . . . . . . . . . . . . 6 D. Transport Demand and Supply Management . . . . . . . . . . 7 II. THE RAILWAY SYSTEM TODAY ......9 A. Traffic and Traffic Allocation . . . . . . . . . . . . . . 9 B. Management and Relationship with the Government . . . . . . 10 Organization . . . . . . . . . . . . . . . 10 Role of the Government . . . . . . ...... . . . 10 C. Finance, Tariffs, and Investment . . . . . . . . . . . 11 Profitability and Tariffs . . . . . . . . . . . . . . . 11 Planning and Investment . . . . . . . . . . . . . . . . 12 D. Maintenance . . . . . . . . . . . . . .. . . . 14 E. Comparison with Other Railway Systems . . . . . . . . . . . 15 System Size and Transport Output . . . . . . . . . . . . 15 Productivity of Track, Equipment, and Labor . . . . . . 16 Transport Management Structure . . . . . . . 16 This report is based on extensive discussions between the Ministry of Rail- ways' staff and the Bank's railway team during an October 1991 identification mission, a March 1992 preappraisal mission, and a June/July 1992 appraisal mission for the proposed Sixth Railway Project. The missions members included: Victor Alalouf, Philip Anderson, David Butcher, Peter Cook, Hennie Deboeck, William Harris, Nikola Holcer, Omar Jung, Emile Karman, Udo Marggraf, Robert McAfee, Mohammed Rasheed, Louis Thompson, W.H. Thompson, Lee Travers, Shunso Tsukada, Thawat Watanatada, and Huikang Xu. The report was written primarily by Thawat Watanatada (Task Manager) and Peter Cook (RIS Technical Leader), with contributions from Louis Thompson (Railway Advisor), and others (see Contributions). The Division Chief is Daud Abmad, the Lead Economist is Shahid Yusuf, and the Director is Shahid Javed Burki. This document has a restricted distribution and may be used by recipients only in the performance | of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. I  - it - Paxe No. III. PROGRESS SINCE THE 19809 A. Improving Railway Management . . . . . . . . . . . . . . . 18 Railway Law . . . . . . . . . . . . . . . . . . . . 18 Economic Contract ................ ......... 18 Decentralization in the Railway System . . . . . . . . . 18 Incentive Pay Systems . . . . . .......... 19 B. Diversifying into Nonrail Activities . . . . . . . . . . . 19 C. Coping with Inflation . . . . . . . . . . . . . . . 9 . v 20 D. Accommodating Past Traffic Growth . . . . . . . . . . . . . 20 E. Meeting Future Expansion Needs . . . . . . . . . . . . . . 22 IV. FUTURE ROLE OF THE RAILWAY .................. 25 A. Major Challenges . . . . . . . . . . * . . . . * * . . . . 25 MOR's Transport Businss e . .. . . . . .. 25 MOR's Nontransport Business . . . . . . . . . . . . . . 26 B. Strengths and Weaknesses of the Railway System . . . . . . 26 C. General Strategy for MOR's Transport Business . . . . . . . 27 D. Government's Supportive Actions . . . . . . . . . . . . . . 28 V. KEY TASKS FOR MOR'S TRANSPORT BUSINESS . . . . . . . . . . . . 29 A. Achieving More Effective Railway Regulation . . . . . . . . . . 29 Promoting Competition ....ti.on . . .................. . 29 Improving the Government's Control of MOR . . . . . . . 30 Creating an Independent Regulatory Agency . . . . . . . 31 Reforming the Railway Tariff System . . . . . . . . . . 31 B. Bolstering Transport Management . . . . . . . . . . . . . . 33 Enhancing Organizational Structure . . . . . . . . . . . 33 Rationalizing the Railway System Through Divestiture . . 34 Developing Railway Human Resources . . . . . . . . . . . 34 Implementing a Management Information System . . . . . . 34 C. Strengthening Railway Investment and Finance . . . . . . . 35 Buttressing Investment Planning . . . . . . . . . . . . 35 Choosing the Best Means to Expand Railway Capacity . . . 35 Improving Railway Financial Management . . . . . . . . . 36 Expanding the Scope of Financial Resource Mobilization . 37 D. Enhancing the Quality and Efficiency of Railway Services through Modern Technology and Techniques . . . . . . . . . 38 Improving Service to Shippers . . . . 38 Modernizing Container Transport Service . . . . . . . . 38 Upgrading Passenger Service . . . . . . . . . . . . . . 39 Raising Operating Efficiency through Better Maintenance 39 Boosting Environmental Protection . . . . . . . . . . . 40 VI. MEDIUM-TERM BANK ASSISTANCE STRATEGY . . . . . . . . . . . . . 42 A. The Bank's Contributions in the Past . . . . . . . . . . . 42 B. Objectives and Scope of Bank Assistance . . . . . . . . . . 42 CONSRBUTIONS . . . . . .... ...... ... .... ... .... .... .... ....... .... 44  - tii - Page No. TABLE-S 1. China: Freight Traffic by Mode . . . 45 2. China: Passenger Traffic by Mode . . . . . 46 3. Transport Investment vs. Economic Output in China . . 47 4. Number of frOR Staff by Activity Type . . . . .. 48 5. A.-tual Revenues and Expenses, 1986-91 . . . . . . 49 6. International Comparison of Selected Railways . . 50 1. Future Demand for Transpor t 51 2. Railway Organization and Staffing Policies . . . 58 3. Railway Law of the People's Republic of China . . . . . . . . . 62 4. Railway-Government Relationship . . . . . . . . . . . . . . . . 65 5. RIS Analysis System . .. . . . . . . . . . . . . . .. . . .- . 67 6. Result of RIS Aoflysi na lysis.... . . . . 73 7. Railway Track Maintenance n.c. . . .... .. .. 85 8. Railway Equipment Maintenance . . . . . . . . . . . . . . . . 86 9. Selected Information on Railway Policy Reforms in the 1980. . . 87 10. Su-nary of Ongoing and Proposed Programs of Technical Cooperation between MOR and the Bank ................... 91 11. Possible Contents of Future Railway Projects . . . . . . . . . 95 12. Suummary of Objectives of Components of MOR-Bank Railway Projects . . . . . . . . . . . . . . . . . . . . . . . 97 13. Timing of Possible Actions Under MOR-Bank Cooperation . . 98 CHARTS C.1 General Organization Chart of the Ministry of Railways . . . . 99 C.2 General Organization Chart of the Government of China . . . . . 100 TABLES IN TEXT TI International Comparison of Government-sponsored Transport Investment a & * 7 T2 Percentage Breakdowns of Alternative Railway Investment Programs for 8FYP (1991-95) . . ................... 22 FIGuS in TEXT Ti Trends for Satisfied Demand, Unsatisfied Demand, and Total Demand for Freight Transport (All Modes) . . . . . . . . . . . 2 T2 Trends for Traffic by Mode .................. 3 T3 International Comparison of Land Transport Networks . . . . . . 6 T4 Trends in Utilization of Chinese Railway Assets (1980-91) . . . 21  - iv - MAPS M.1 Total Freight Traneport Demand Between Regions in 1989 (IBRD 24531) 1.2 Total Freight Transport Demand Bet:ween Regions in 2000 (IBRD 24532) M.3 China Railway System (IBRD 24578) M.4 Railway Bottlenecks in '989 (IBRD 23817R) M.5 Railway Organization (IBRD 23792) M.6 RIS Preferred Railway Investment Package (Low Inventmernt) (IBRD 23816R) 4.7 RIS Preferred Railway Investment Package (Higi. Investment) (IBRD 24618) M.8 Railway Bottlenecks in 2000 Under RIS Preferred Package (IBRD 24624)  - v - CHINA'S RAILWAY STRATEGY Executive Summary i. The main objectives of this paper are: first, to contribute to the Ministry of Railways' strategies for the 1990ss and second, to provide a basis for Bank assistance to China's railway system in the next five years. The paper provides a background on the transport sector, with a primary focus on the railway system, describing the system as it stands today and the progress it has made since the 1980s in modernization and reforms. As China is rapidly evolving into a market economy, the paper discusses the main challenges the railway system now faces, as well as the major tasks that need to be done to meet these challenges, and how the Bank plans to assist the Ministry of Rail- ways in performing these tasks. ii. Intercitv Transiort Problems. The capacity of the intercity trans- port system in China has become increasingly insufficient slnce the mid-1980s to handle the volume of intercity freight and passenger trafiic, as evidenced by the growth of bottlenecks on the transport system, the rationing of trans- port capacity on railway lines, and the poor quality of service experienced by shippers and passengers. The inadequacy of transport capacity is mainly caused by two factors: (a) the rapid growth of traffic since the onset of economic reforms in 1979 (averaging 8 percent per year for freight and 12 percent per year for passengers during 1979-89)s aud (b) past underinvestment in transport infrastructure. iii. The Railway System. Given its vital role in the country's economic and social fabric, the railway system is often regarded by the Chinese as "the backbone of China." For both economic and historical reasons, the railway system is the dominant mode of intercity transport in the nation: it carries more traffic than all other modes combined. Economically, the Chinese rail- ways enjoy economies of scale that result in low costs of rail transport per passenger-km and ton-kl compared with road transport and some river transport. Historically, the competitiveness of the Chinese railways has been reinforced by decisiono of the national government before 1978 to allocate more invest- ment funds to railways than to highways and waterways. However, from 1978 to 1991, the government moved its emphasis away from the railways in favor of other modes. Since 1992, the government has shifted the emphasis back to the railways and given top priority to railway development. iv. Despite its size, the Chinese railway network provides little cover- age for China's population and land area. The network's coverage, be it mea- sured in terms of total route length per capita or total route length per unit of arable land area, is one of the lowest in the world--lower than the cover- age provided by railway networks in Russia, India and Brazil. The railway system is intensely used and ranks only second to Russia in the amount of combined freight and passenger traffic per route-km of track. v. The Ministry of Railways is charged with administering the national standard-gauge railway network, comprising about 53,000 km of main routes. MOR employs about 3.4 million persons. It is a profitable operation, respon-  - vi - sible for financing all of its operating and capital expenditures (through both self-go-nerated funds and borrowings on market terms). Although NOR employs relatively old technology, its equipment and plant utilization is among the world's highest. MOR is well managed and has a disciplined work force. vi. Since the early 1980s, MOR has undergone a number of changes to: (a) make its relationship with the government more effective, through enact- ment of a Railway Law to provide a framework for railway regulation and intro- duction of an economic contract compr.sing (i) a financial responsibility system that gives MOR a higher degrec of financial autonomy; and (ii) an incentive system tl.at ties staff salaries to their productivity; (b) boost its investment and financial planning capability; (c) strengthen its internal management through decentralization of executive powers; and (d) modernize its track, equipment, and information technology. These changes have helped MOR serve the economy more b_,ectivel,. vii. MOR's General Market Strategy. The next decade is likely to see highway, waterway, and civil aviation mount increasingly serious competition against the railway system in both price and service. Highway transport is expanding rapidly with more high-quality roads and more modern trucks and buses, which provide competitive services over short to medium distances. Waterway transport is modernizing to include container transport in the coastal and river delta areas where it is highly competitive in serving import/export traffic. Air transport is growing fast with more modern air- craft and the birth of regional airlines to encourage competition. As a result, the railway system will likely lose some of its traffic. Some long- distance railway passenger traffic is likely to divert to civil aviation, while short- and medium-distance traffic comprising both passenge-7s and light manufactured products are likely to shift to highways and waterways, espe- cially in coastal areas or where modern highways run parallel to railways. viii. Yet, the present dominance of the railway in the Chinese transport system, combined with the expected strong growth of the Chinese economy and recent technology gains in railways, ensures a continuing major role for rail- way transport over the next 20 years and beyond, particularly for bulk and semi-bulk commodities over medium to long distances and for long distance passenger traffic. To fulfill its role as a provider of public transport services, MOR should strive to (a) provide a range of safe and adequate trans- port services that meet customer needs on a commercial basis; (b) control costs and refrain from exercising its monopoly power at the expense of the public; and (c) maintain financial self-sufficiency, i.e., being able to finance all of its operating costs and capital investments, the latter mostly through self-generation of cash but also through capital market borrowing. ix. In so doing MOR should focus on the categories of traffic where the railway has a comparative advantage: (a) medium- to long-distance bulk traffic, especially in unit train loads or multiple carloads (and at the same time let truck transport pick up more of short- to medium-distance general merchandise traf- fic, with less-than-carloads or single-carload);  - vii - (b! container traffic, with door-to-door services in full cooperation with other modes5 (c) medium- to long-distance passenger traffic (potentially incluiding some with high-speed services); and (d) high-volume suburban rail commuter service. x. Key Issues. To effectively serve the broad tiansport markets out- lined above, MOR must successfully meet new challenges in a rapidly changing macroeconomic environment. Among the major issues facing MOR in the 1990s, the two most important ones relate to expansion and regulation/management of the ,ailway system. These issues are becoming more urgent as China's move toward a market economy is being accelerated by the current new surge of eco- nomic reforms. xi. A first issue is the concern that the planned expansion of railwe capacity will not satisfy the growth of railway transport demand through th, year 2000. Although the Chinese government hac recognized the vital impor- tance of expanding the railway to support future economic growth, the cur- rently planned railway network capaclty still may not close more than half of the anticipated gap between railway transport demand and supply in the year 2000, if China's economy grows at a rate similar to that in the 1980s (8 to 9 percent per year). This is expected to cause a loss in economic output of 0.5 to 1 percent of GNP. The supply problems have two sources: first, there is less than optimal use of the existing capacity of the railway system due to the lack of modern operating and planning tools; and, second, an inadequate level of capacity expansion inveetment (Y 40-60 billion below needs) is now in the plans supported by the national government. These problems have been com- pounded by an inadequate level of railway tariffs and surcharges, which even after a major increase in July 1992 is still barely enough to finance the modest railway investment program currently planned. This situation is improving as a result of recent actions taken by the government and MOR in: (a) developing and applying a decision support analysis system (with World Bank support) under a Railway Investment Study (RIS); (b) raising the economic growth rate for planning from 6 percent per year to 8 to 9 percent; and (c) boosting a surcharge on freight traffic to build up an earmarked Railway Construction Pund. However, the threat of growing economic 1088 still exists if more essential actions are not taken to meet the growing needs. xii. The RIS analysis syst6m has shown that there is significant acope for improvement in system throughput: about 10 percent through improved freight routing procedures and another 10 percent through optimization of the scale, location and timing of investments for rail capacity expansion. RIS resalts have also suggested that a different mix of investment from currently planned would yield better results; specifically, they show substantial bene- fits for a shift of resources from new line construction to electrification projects which yield twice as much capacity per yuan of investment. xiii. A second issue is the concern that the existing framework of railway regulaticQ and the existing system of railway transport management do not give NOR enough flexibility to price and market railway transport services in response to increasing competition from other modes and to changing customer  - viii - needs. Government controls currently restrict MOR's ability to price services and its ability to provide new services in response to demand. These controls will only conetrain MOR's scope of action in responding to market conditions. xiv. Rev Tasks for MOR's Trans2ort Business. To resolve the two key issues raised above, as well as other issues related to policy and management reforms and to systemwide technological modernization, MOR needs to undertake tasks in four major areas--railway regulation, internal management, investment and finance, and railway technology and techniques. xv. Achieving more effective railway regulation deals with measures fort (a) promoting competition--by rationalizing transport pricea across modes, relaxing controls of market entry, developing infrastructure networks equitably for all modes, and encouraging competiticn within the railway system; (b) improving the government's control of MOR--by enhancing the economic contract between MOR and the government with (i) more emphasis on quality of railway service, (ii) closer integration of the contract with the railway strategic plan tied to the country's macroeconomic policy and forecasts, and (iii) reinforcement of the contract with government actions such as improving the materials distribution system in order to enable MOR to fulfill its economic and social role in a more cost-effective way; (c) creating an independent regulatory agency--mainly to monitor and publish MOR's performance statistics in areas that are important to the government and the public, including profitability, safety, and service quality; and (d) reforming the railway tariff system--by (i) raising the railway tariffs and surcharges to a level needed for MOR's financial self- sufficiency, (ii) removing distortions in the tariff system through proper costing of rail services, and (iii) allowing MOR the needed flexibility to set rail tariffs (within constraints) ir. response to market conditions. xvi. Bolsterina transport management involves measures for: (a) enhancing the organizational structure of MOR's transport business-- by sharpening the financial accountability of railway managers, and focusing personnel incentive systems less on meeting physical output targets and more on meeting service quality standards (while ensur- ing that MOR's already high physica:: efficiency of operations will not be jeopardized); (b) rationalizing the railway system through divestiture, to improve management efficiency and foster competition--by spinning off selected nontransport functions, potentially including railway con- struction and manufacturing, "diversified economy" activities, and health, education, and housing functions;  - ix - (c) developing railway human resources--by providing (i) compensation policies, and (ii) further training of railway staff to meet the needs of a market-economy railway; and (d) implementing a railway management information system--Dy providing railway managers at different level3 with the right kinds of finan- cial, operating, and service performance information needed to moni- tor and control railway operations and to make well-informed deci- sions. xvii. Stretnathening railway investment and finance covers measures for: (a) buttressing railway investment planning--by employing more exten- sively modern analytical tools and adopting transport demand fore- ^asts that match more closely the economic growth prospects; (b) choosing the best means to expand tAilway capacity--by (i) better matching the scale, timing, and location of railway investment to future traffic patterns, (ii) adopting more cost-effective operating policies and technologies for track and rolling stock, and (iii) improving freight routing patterns to decrease the total ton-km volume of freight traffic to be carried by the railway system (para. xii); (c) improving vailway financial management--by (i) revamping MOR's accounting system to make it conform better with international prac- tices, (ii) estaslishing a costing system for proper tariff setting and financial controls, (iii) strengthening financial planning and reporting to foster accountability, and (iv) improving the use of scarce foreign exchange; and (d) expanding the scope of financial resource motilization to fund needed railway investment--by tapping funding sources other than tariff revenues and current creditors, for example, by making joint investments with other investors, and through borrowing from foreign capital markets (e.g., through the Bank's Expanded Cofinancing Oper- ations (ECOs)]. xviii. Enhancina the gualitv and efficiency of railway services through modern technologv and techniques entails measures for: (a) improving service to shippers--by providing more responsive and more reliable service to shippers, with the help of a computer-based Transport Management Information System (TMIS) currently under implementation; (b) modernizing container tran port service--by providing a more customer-oriented, door-to-door service with higher frequency, faster travel times, and greater reliability, in close coordination with other modes; (c) upgrading passenger services--by (i) installing a computer-based passenger reservations system on a networkwide basis, (ii) monitor-  - x - ing and enforcing customer service standards, (ii) providing cus- tomer information systems for customers, and (i:) improving on-ground and on-train services; (dj raising operating efficiency through better maintenance--by improv- ing the design and manufacturing technology for critical equipment parts, reducing the number of designs and models of locomotives, modernizing trar monitoring technology and mechanizing track main- tenance, etc.; (e) boosting railway environmental protection--by (i) btrengthening MOR's capability for environmental assessment, monitoring, and con- trol, and (ii) developing strategies and action plans for solving railway pollution problems (e.g., pollution related to passenger trains). xix. ?edium-Term Bank Assistance Stratezy. Since 1983, the Bank has assisted the government through five national railway operations and one local railway operation, totaling $1,365 million in loans and credits. The primary objectives of the first four national operations were to finance railway con- struction to expand traffic capacity in high-priority corridors and to intro- duce new technology for improving railway operating efficiency. The fifth national operation began a new approach of supporting railway policy reforms and technological modernization on a systemwide basis. Future railway opera- tions in the medium term (i.e., the next five years) are intended to implement the new "policy-technology" approach more extensively. XX. Under this approach, the Bank would support MOR in four major areas. First, the Bank would help MOR formul&te and implement a variety of policy initiatives, such as to strengthen railway investment planning, reform the railway tariff system, and revamp the railway accounting system. Second, it would help MOR undertake institutional development of the railway system, potentially in the areas of manpower development and rationalization, divesti- ture of nontransport functions, corporate strategies for railway construction and manufacturing, etc. Third, it would help MOR finance investment programs to expand railway capacity (determined on an economic basis). And fourth, the Bank would continue to help MOR modernize its technology on a systemwide basis related to management information systems, computer and telecommunications equipment, plant and equipment maintenance, and environmental protection.  CHINA'S RAILWAY STRATEGY I. TRANSPORT DEMAND AND SUPPLY 1.1 Over the past decade, transport demand has grown faster than the supply of transport services and infrastructure in all modes in China, thereby creating major bottlenecks in the transport system. The explosive growth of the Chinese economy has been reflected in the demand for transport while the expansion and modernization of the transport infrastructure and services have been proceeding at a slower pace. This chapter first discusses the evolution of transport demand, then the growth of supply, followed by the implications of system bottlenecks fv, the railways. A. Past Trends of TransDort Demand 1.2 Traffic represents the part of demand that is satisfied by the transport system. In a situation of transport system shortages, there is also unsatisfied demand in locations where potential shippers and passengers are unable to find a means of transport. This has been a problem particularly for medium- to long-distance transport in China. As discussed below, the growth of total demand for transport can be estimated by combining the data on growth of traffic (i.e., satisfied demand) with information on unsatisfied demand. 1.3 Traffic. Since 1979, despite supply constraints, China has experi- enced very rapid growth of traffic, consistent with the rapid expansion of the economy, which averaged about 8.8 percent par year in real GDP terms during 1979-91. In this same period, total freight traffic (in ton-km) on all modes increased by 7.3 percent per year and passenger traffic (in passenger-km) grew by 10 percent per year 1/ (Figure Ti; Tables I and 2). This growth trend tapered off in later years as the supply constraints became more severe, transport prices were increased, and the growth of the economy slowed tempo- rarily during the stabilization period of 1989-91. Increases in passenger and freight traffic flows between provinces represent a large portion of the growth in interprovincial trade and related transport demand in the past decade. The fastest growth in demand for transport has been between the rapidly growing provinces along the east coast (especially Guangdong province and those provinces that surround Shanghai) and along the routes connecting these areas with the other regions of the country and the outside world. The flows of coal from the provinces in the Coal Base (Shanxi, Shaanxi, and Inner Mongolia) represent the largest of these flows (Map M.1) and have been growing the fastest of all longer-distance freight traffic flows (about 50 percent faster than average rail freight growth). Linkages for input materials and industrial products from the northeast and southwest have also increased sig- nificantly9 along with manufactured goods shipped from the high growth areas to other regions. Because these freight flows travel relatively long dis- tances, the demands on the capacity of the interprovincial transport system I/ Both freight and passenger traffic includes transport by railway, high- way, inland and coastal waterways, and air; freight traffic also includes transport by pipeline.  - 2 - have been increasing significantly and have a significant impact on a large part of the national transport system. Figure TI: TRENDS FOR SATISFIED DEMAND, UNSATISFIED DEMAND, AND TOTAL DEMAND FOR FREIGHT TRANSPORT (ALL MODES) 2,400 - S2,200- nsatisfied . 1,00 - Demand C ~2,00O-OO .+H+PH E 0110 | 1,600- * S~~~~atsfied 1,2000- -| , - -- 1985 1986 1987 1988 1989 1990 1991 1992 Year 1.4 Railway freight and passenger traffic grew rapidly during 1979-88, averaging annual rates of 6.3 percent for freight and 11.6 percent for passen- gers, but experienced setbacks during 1989-91 that temporarily prevented the bottlenecks from worsening (Figure T2). However, the growth of railway trans- port demand is expected to increase again in the future. The growth of rail- way freight traffic dropped to 3.6 percent per year during 1989-91, as a result of two factors: first, a series of rate increases during 1990-91 totaling about 40 percent; and, second, bottlenecks on key railway lines blocking additional traffic flows between many distant pairs of crigins and destinations that did not have alternative routes with available capacity. Railway freight traffic is expected to grow faster as railway tariffs stabi- lize and the key bottlenecks are removed. In 1992, a year in which railway tariffs went up 35 percent in July, railway freight traffic grew an estimated 5.5 percent. Railway passenger traffic dropped by 7.3 percent in 1989 and a further 16.2 percent in 1990, following the introduction of administrative controls on personal travel in mid-1989, as well as sharp fare hikes in Sep- tember 1989 that averaged 112 percent. Intercity passenger travel by railway is expected to regain its higher growth in the future as suppressed demand is released from administrative controls and per capita income increases. In 1992, railway passenger traffic rose an estimated 10.7 percent. 1.5 Unsatisfied Demand. The less visible aspects of the gap between transport supply and demand are: (a) the unsatisfied rail transport demand over medium to long distances that was estimated conservatively at a total of  -3- Fi&ure T2: TRENDS FOR TRAFFIC BY MODE ailion ton-km 1,000i 400 600 | ,m Pipelines T Road N Inland waterway a] Railway eIl:bn Passenger-km 100 50: /1~~~~~~' Pasege Traffic Year | CPvi Aviation InWaterway teRoad LRailway] .lion tons of freight in 1989 (9.8 percent of the total potential rail trnsnport demand in tons) 2/ and up to 30 percent of the passenger ke Annexc 6 for furthetr discussion of unlsatisfied deand. The unstis- .ed demand as a percentage of ton-kin is slightly higher du. to larger Lan average haul distances for the unsatsfied dend.  - 4 - traffic served for some services; and (b) losses to the economy estimated at up to Y 22 billion per year, equivalent to about 1 percent of 1991 GDP.3/ This gap has grown from a very small proportion of total traffic in the early 1980G (less than 2 percent of total demand in 1985) to its present level (9.8 percent) at a rate roughly proportional to the growth in the number of bottlenecks on the national transport network (para. 1.12; Figure Ti). 1.6 Total Demand. The growth in total demand for transport (actual traffic plus unsatisfied demand) is rising significantly faster than traffic growth. The above estimate of unsatisfied demand implies that total demand for freight transport has grown at a rate approximately 15 percent faster than traffic (i.e., 8.3 percent per year over the period 1979-91 for total demand compared with 7.3 percent for traffic). Total demand faded from 1989 to 1991, but returned to its long-term growth trend in 1992 as economic growth increased (Figure T1). Based on trends in the past 10 years, the elasticity of total demand for freight with respect to GNP is estimated at 0.95 (i.e., it grows at 95 percent of the GNP growth rate) (Annex 1). The faster increase in passenger transport demand combined with the greater supply constraints on passenger trains resulted in a greater increase in unsatisfied demand for passenger transport than for freight, at least since 1985. The past growth rate of total passenger transport demand is estimated to be in the range of 12 to 15 percent per year, i.e., in the same range as growth for relatively unconstrained road transport (13.9 percent per year from 1979 to 1991). This implies that total passenger transport demand may be growing 30 to 40 percent faster than passenger traffic growth and that the main constraints are from rail transport shortages. The elasticity of total demand for intercity per- sonal travel is roughly estimated at 1.5 (i.e., total demand grows at 150 percent of the growth of GNP per capita) (Annex 1). B. Future Demand for TransDort 1.7 Total TransDort Demand. There are both positive and negative fac- tors that determine future total demand for intercity transport. For freight transport, the major positive factors are growth of the economic output, improvements in the quality of transport service in the aggregate, and growth in interprovincial trade (para. 1.3). The major negative factors for freight are expected increases in the prices of transport and commodities, and ratio- nalization of the commodity distribution system. For passenger transport, the major positive factors are growth of the per capita personal income, growth of the population, an increase in the share of services in the economic output, improvements in the quality of transport services, and relaxation of the travel restriction policy. The major negative factor for passenger traffic is an expected increase in railway passenger fares. 1.8 Because of the lack of knowledge on the precise relationship between the demand for transport and these factors, future transport demand can only be imprecisely predicted. Based on an analysis of these factors with a likely 8 percent per year rate of GNP growth, future growth of total transport demand 3/ China Railway Investment Study, Volume II, Yellow Cover version, Ministry of Railways and World Bank, February 14, 1992, pages 8 and 84, and Annex 6 of this report.  - 5 - in the 1990. is expected to be only slightly less rapid than in the 1980., i.e., about 6 to 8 percent per year for freight and 12 to 13 percent per year for passengers. Light industry and manufactured goods are likely to have the highest growth rates; bulk freight (coal, grain, fertilizers, etc.) has grown at sustained rates of about 6 percent per year, on average, since the start of economic reforms and is expected to continue to grow at slightly lower rates than in the past. These growth rates imply that the transport system as a whole will still require substantial expansion of capacity to meet the demand. A large portion of the projected growth in total transport demand is derived from the growth in transport demand between provinces and regions which is increasing faster than intraprovincial traffic (Maps M.1 and M.2). 1.9 Railway TransDort Demand. Future demand for railway transport depends on the above factors, as well as on the prices and quality of railway transport services relative to those of the other modes. The next decade is likely to see highway, waterway, and civil aviation mount increasingly serious competition against the railway system in both price and service. Highway transport is expanding rapidly with more high-quality roads and more modern trucks and buses, which provide competitive services over short to medium distances. Waterway transport is modernizing to include container transport in the coastal and river delta areas where it is highly competitive in serving import/export traffic. Air transport is growing fast with more modern air- craft and the birth of regional airlines to encourage competition. As a result, the railway system will likely lose some of its traffic. Some long- distance railway passenger traffic is likely to divert to civil aviation. Some passengers and some light manufactured products by rail over short to medium distances are likely to shift to highway and waterway, especially in coastal areas or where modern highways run parallel to railways. 1.10 On the other hand, the railway system is generally superior to the other modes with respect to safety, energy efficiency, use of land, and cost effectiveness. It is likely to retain its competitive edge with its low-cost services, particularly for bulk and semi-bulk commodities over medium to long distances, where in-transit time and service reliability are not a critical factor. These commodities account for about 75 percent of railway freight traffic at present and are expected to sustain a high share in the foreseeable future. Also the railway system can guard against much of its high-valued freight traffic switching to the competition, by improving its services, espe- cially through the provision of door-to-door container transport services that use the railway for line haul transport and the highway for pickup and deliv- ery. The railway system is also likely to be the preferred mode of transport for passenger traffic over a wide range of medium distances that are too short for the airplane to take advantage of its speed and too long for the bus to be competitive with respect to travel time. Taking these various factors into account, future railway transport demand is forecast to be in the range of 4 to 6 percent per year for freight and 10 to 12 percent per year for passen- gers (Annex 1). C. Development of the Transport System 1.11 Size of the Transport System. The transport system in China has grown considerably over the past four decades but it is still relatively sparse. Between 1952 and 1992, the length of the highway network increased  - 6 - ninefold to 1,014,000 km, and the length of the railway network more than dou- bled to 53,000 route-km, to cover a large part of China (Map M.3). Despite this progress, China has one of the sparsest transport networks in the world (Figure T3). Figure T3: INTERNATIONAL COMPARISON OF LAND TRANSPORT NETWORKS Railway Networks 1,200z . Track Length (km) per million 3,000 . . . ... ...... Population 800 . . l...r7 6...... Track Length ...... 20 (km) per 1 0 sq km arable land. 100 0 Thousands s tRoad Networks 40, . ~~~~~~Road length (km) . ~~~~~~~per million 30 . 7 Population . .. . . . . . . . . .. . . .. . . . . . . .. .. .. . . . . _ > f ~~~~Road Length . _ ~~~~~~~(km) per 10 sq 20 klin rbeln 1.12 System Bottlenecks. Past growing demand on Chinas's relatively sparse transport network has widened the Sap between the supply and demand for  transport infrastructure and services. The visible manifestations of this gap are: (a) the lack of sufficient, reliable transport service in some regions (especially for heavy goods and bulk commodities such as coal); (b) the exis- tence of transport bottlenecks; and (c) administrative rationing of limited transport capacity. Bottleneck links on the railway network, where traffic in at least one direction reaches 95 percent of the design capacity or more, leaped from 7 percent of the network in 1985 to 37 percent in 1989 (Map M.4). While improvements to management and organization in the provision of trans- port services can alleviate the bottlenecks to some extent, the major reason for these bottlenecks is the lack of stable, reliable sources of investment funds, resulting in underinvestment in transport, which averaged only 1.3 per- cent of GNP during 6FYP (1981-85) and 7FYP (1986-90) (Tables TI, 3) compared with 2.0 to 3.8 percent for several other countries with economic growth simi- lar to or slower than China's. An important consequence of this underinvest- ment was that the railway system did not acquire sufficient financing to meet the demand for its services. Table T1.1: INTERNATIONAL COMPARISON OF GOVERNMENT- SPONSORED TRANSPORT INVESTMENT (As a Percentage of GNP) Transport investment Country Period as percent of GNP Japan 1964-73 3.5-3.8 Korea 1979-81 2.0 Brazil 1979-81 2.4 India 1980-89 2.4 Soviet Union (former) 1980-89 2.8 China 1981-85 (6FYP) 1.3 1986-90 (7FYP) 1.3 1.13 Since the administrative rationing of medium- and long-distance transport capacity in China gives a higher priority to freight than to passen- gers, Chinese travelers have been limited to one of the lowest levels of per- sonal mobility for intercity travel in the world, with about 550 km traveled per capita per year compared with 800 km per capita for India in 1989. This situation has improved somewhat for the higher income and business travelers, with expanded air transport capacity since 1990, but it has not improved fast enough to alter the gap between transport supply and demand. D. Transport Demand and Supplv Management 1.14 There is little doubt that to bridge the gap between transport demand and supply by the year 2000, the GNP share of transport investment in China needs to be increased significantly, from the present 1.3 percent to 2 percent or more. Without this increase there will be losses to the economy due to congestion costs and lost production--to the tune of 1 percent of GNP or more. This explains why investments to expand the capacity of bottlenecked  - 8 - railway lines tend to have very high economic rates of return, often more than 30 percent. 1.15 If total transport investment as a share of GNP were to be increased from, say, 1.3 to 2.3 percent, how should the increment be allocated between the railway and the other modes? This question cannot be answered with any degree of precision without undertaking a large exercise of multimodal invest- ment analysis on a nationwide basis. However, a rough indication may be obtained from the observation that annual railway investments since 1981 have averaged about 50 percent of total investment in transport in China. The fact that many key bottlenecks have developed in the railway network suggests that railway investments designed to eliminate these bottlenecks may have higher economic rates of return than investments in other modes of transport. This implies that perhaps more than 50 percent of the investment increment should be made in the railway system. 1.16 Because of the scarcity of capital in China, the investment to expand transport capacity must be made at minimum possible cost. This should be done through both demand and supply management measures. As to demand management, the government has taken major strides in the last two years by raising the prices of transport services close to their economic long-r-in marginal costs, so that only relatively small additional adjustments are required ln the future. Regarding other demand management measures, the gov- erument is considering actions to: (a) rationalize the administrative commodity distribution system to reduce total transport requirements (e.g., by better matching supply and demand centers, mainly through an expanded role of the market) (para. 4.13); and (b) improve efficiency in the use of existing rail transport capacity, according to the economic priorities of different types of traffic (where capacity is rationed) (para. 4.14). Regarding supply management, the government could: (a) minimize the cost of capacity expansion (per ton-km per year of new capacity) on a given rail line or yard, by improving operating poli- cies or upgrading transport technology (e.g., by increasing the axle loads on rails) (para. 5.24(a)); (b) maximize the economic benefits of network capacity expansion, by optimizing the scale, location, and timing of capacity expansion projects under budget constraints (paras. 2.18-2.19, 5.24(b)]; and (c) improve freight routing patterns in order to utilize network capacity more efficiently, which could add some 10 percent to system throughput (paras. 2.18-2.19, 5.24(c)).  - 9 - II. THE RAILWAY SYSTEM TODAY 2.1 Given its crucial role in the country's economic and social fabric, the railway system is frequently regarded by the Chinese as "the backbone of China." The railway system today encompasses a nationwide, interconnected standard-gauge system comprising 4,400 route-km of local railways and 53,000 route-km of the national railway network, which is administered by the Minis- try of Railways (NOR) (Map M.3). The national railway system has a locomotive fleet of about 13,800 units, comprising about 5,900 steam, 6,100 diesel, and 1,800 electric units. With 1989 being their final year of production, MOR's steam locomotives are being phased out by attrition. The railway system also has a rolling stock fleet of 27,600 passenger coaches and 370,000 freight cars. Despite its use of relatively old technology, the Chinese railway sys- tem is a physically efficient operation compared to most other railways in the world. The Ministry of Railways is a profitable operation and receives no cash subsidies from the government. It is responsible for funding all of its operating costs and capital investments--through both self-generated funds and borrowings on market terms. With the help of the government, the Ministry has undergone a number of changes in the past decade, in a long-term process to transform itself from a government organization into a market-oriented entity. This transformation process has resulted in improvements in its regulatory and contractual relationship with the government; its capability for investment and financial planning and financial resource mobilization; and the technology for its information, track, and locomotive and rolling stock systems. How- ever, the railway system still faces formidable challenges in completing the transformation. A. Traffic and Traffic Allocation 2.2 The total volume of combined freight and passenger traffic carried by MOR ranks third in the world. Coal is the largest item (accounting for 40 percent of total tonnage handled by the railway) and a critical source of energy for the economy (accounting for 75 percent of all energy consumed in China). Major commodities that make up the bulk and semi-bulk traffic (coal, grain, petroleum, construction materials, iron and steel products, etc.) account for about three quarters of total freight traffic. The remaining quarter is classified as "miscellaneous." This miscellaneous category includes general merchandise and the types of traffic that can be moved in containers. Between 1980 and 1990, bulk and semi-bulk traffic on the railway grew at 5.9 percent per year, while miscellaneous traffic grew at the higher rate of 7.9 percent per year. 2.3 Despite its great potential, however, container traffic still is a minuscule part of railway traffic. At 10 million tons in 1990, the total freight tonnage carried in containers by rail is less than one percent of total rail freight tonnage. The Ministry has launched a program to modernize and expand container transport services in cooperation with other modes. 2.4 The majority of railway freight traffic is planned and contracted on an annual basis at the national level. The State Planning Commission (SPC) specifies the total volume of freight traffic to be hauled and the volume of coal traffic to be transported out of the energy base in northern China. SPC  - 10 - also issues guidance on the number of passengers and the amounts of certain freight traffic to be carried. The State Council (China's cabinet) on occa- sions directs MOR to provide certain emergency services. B. Management and Relationship with the Government 2.5 Organization. The Ministry of Railways employs 3.4 million individ- uals, of whom 51 percent work on rail transport. The remaining 49 percent work on nonrail transport activities: 17 percent on construction; 13 percent on manufacturing; 8 percent in a wide variety of diversified ventures suck as local truck pick-up and delivery companies and retail outlets; and the remain- ing 11 percent on other activities including attending to the welfare and higher education needs of railway staff and their dependents (Table 4; Annex 2; Chart C.1). Through control measures exercised by MOR, the 3.4 million work force is expected to remain roughly constant between 1991 and 1995. The rail transport function is carried out by 12 regional administrations that operate as separate profit centers with a considerable degree of autonomy (Map M.5). These administrations are empowered to (a) make day-to-day operating decisions; and (b) manage small construction projects, retain part of the operating profit, and staff their organizations (paras. 3.4-3.5). MOR's head- quarters retains authority over several key functions, such as selecting and managing large construction projects, determining the traffic targets of the regional administrations, and scheduling trains (except local trains). 2.6 The nonrail transport part of MOR is a conglomerate of companies engaged in a variety of activities that are mostly railway-related. MOR's own factories (about 75 of them) produce the bulk of MOR's requirements for loco- motives, passenger coaches, freight cars, signaling and communications, and parts. Most construction for the railways is carried out by some 20 MOR con- struction companies. Design, engineering, and architectural services also are provided in-house, by special institutes and bureaus. Furthermore, the Minis- try operates 11 universities, 20 staff colleges, and more than 60 vocational schools, as well as over 4,000 hospitals and other medical service facilities. 2.7 Finally, there is a group of more than 300,000 MOR nonrail transport staff who undertake a range of business activities, including trucking ven- tures that provide delivery and pickup services for shipments to and from railway stations. This so-called "diversified economy" group is becoming more important as a means to absorb surplus rail transport personnel that result from improvements in the efficiency of railway transport operations. During 1991-95, while the railway transport and construction staff are subject to a 20 percent cut in size, the diversified economy group is expected to increase the size of its personnel by about 150 percent. 2.8 Role of the Government. The national Railway Law lays down the responsibilities of the national and local governments, MOR, railway custom- ers, and the public at large with respect to the construction, management, and operation of railways in China (para. 3.1; Annex 3). For the national rail- ways, the 12 regional railway administrations (para. 2.5) are charged with the responsibility for providing railway transport services under the management of MOR's headquarters.  2.9 According to the Railway Law, the overall responsibility for railway regulation rests with the State Council. The State Council, through its coor- dinating agencies, especially SPC, the Ministry of Finance (MOF), and the State Price Bureau, and in cooperation with special-function agencies (e.g., the National Construction Bank), exercises its authority to (a) direct MOR's development plans, based on the long- and medium-term development plans at the national level; and (b) approve a number of decisions including the following (Annex 4; Chart C.2): (i) the total size of railway investment in a Five-Year Plan, (ii) major railway capital construction and upgrading projects, (iii) changes in railway tariffs, (iv) the total level of MOR's borrowings in each year and each Five- Year Plan, and (v) wage levels of railway employees. 2.10 Under an economic contract set up with the national government, the Ministry of Railways is responsible for financing all of its operating and capital costs (through both cash from its operations and borrowings on market terms). In return, the Ministry is allowed to keep all of its net income after paying a 5.3 percent sales tax on gross revenues and a lump sum tax of Y 2,000 million per year (4.3 percent of 1991 gross revenues). This financial responsibility, however, should be seen in the context of distorted prices in China (paras. 2.11-2.12). In addition, the contract provides the Ministry with incentives in the form of an annual pay per employee that varies with staff productivity. Staff productivity for a given year is measured in terms of the volumes of freight and passenger traffic per employee, the total lengths of different types of railway capacity expansion projects built per employee, and the numbers of locomotives and rolling stock produced per employee (para. 3.3). C. Finance. Tariffs, and Investment 2.11 Profitability and Tariffs. The Ministry of Railways runs a finan- cially profitable operation with its gross revenues well above its operating expenses (Table 5). However, its profitability is distorted in two opposing directions. In one direction, its profitability is overstated by two factors: (a) underreporting of depreciation expenses; and (b) credit subsidies due to lower-than-market interest rates on some of its borrowings and price subsidies on some of the materials, fuels, and electricity it purchases. MOR's account- ing procedures tend to underreport depreciation expenses because they value assets on a historical basis and because they depreciate assets at predeter- mined low rates (e.g., 4 percent for fixed assets). However, these factors are being tempered by the following measures: First, the Ministry is begin- ning to revamp its accounting system (with Bank assistance). Second, the implicit credit and price subsidies are being phased out: between 1990 and 1992 the level of these subsidies was reduced from 20-25 percent to 13-18 per- cent of railway operating costs.  ^ 12 - 2.12 In the opposite direction, the Ministry's profitability has been understated because of a chronic problem with its tariff systems the prices it charged for railway freight transport services (base tariffs plus surcharges) were until 1992 kept well below their economic long-run marginal costs ("economic costs" means costs calculated with the effects of deprecia- tion understatement and implicit subsidies removed). This underpricing situa- tion has improved recently: in July 1992 a freight surcharge for an earmarked Railway Construction Fund was raised from 0.2 to 1.2 per/tkm, thereby increas- ing railway freight revenues by about 35 percent. This increase has brought freight transport prices to 95 to 115 percent of their average economic long- run marginal costs on a systemwide basis,1/ and at the same time, enabled NOR to fund 80 to 85 percent of the planned investment program of Y 116 bil- lion for the Eighth Five-Year Plan (8FYP) (1991-95). The Ministry expects to obtain further increases in base tariffs and surcharges in the next few years to close the existing funding gap and to finance expected increases in invest- ment resulting from higher economic growth expectations (paras. 3.16, 5.21). MOR is also likely to raise additional funds from both domestic and interna- tional borrowings. 2.13 The underpricing problem aside, the railway tariff system has two major deficiencies. First, the tariff system does not sufficiently account for (a) the sensitivity of the traffic being served to the tariffs, (b) the underlying costs of the services, and (c) the existence of competition. A major reason for this is that MOR does not yet have a functional railway cost- ing system (it is implementing one). And second, control of MOR's tariffs by the national government (through the State Price Bureau) does not give MOR enough flexibility to adjust its tariffs to reflect varying cost and market conditions across its vast network.2/ With Bank assistance MOR plans to propose a revision of its tariff system, among other things, to remedy these deficiencies. 2.14 Planning and Investment. The Ministry carries out three types of plans: (a) the long-term strategic plan, which focuses on strategic issues over a horizon of more than five years, such as what kinds of passenger ser- vices are to be provided; (b) the five-year plan, which basically identifies specific railway investment projects proposed for implementation during the plan period (as part of the national government's five-year planning process); and (c) the annual plan, which basically specifies a yearly program for con- struction of the projects approved for implementation under the Five-Year Plan, as well as a yearly program of rehabilitation of track and equipment. Under the five-year plan, capital construction and upgrading projects costing Y 50 million or more must be approved by SPC and those costing Y 200 million or more must be approved by both SPC and the State Council. In addition, the total amount of MOR's capital expenditure each year must not exceed a ceiling opecified by MOF. Il/ The economic long-run marginal costs vary considerably according to the type of railway construction, from building new lines to double-tracking or electrifying existing lines. 2/ MOR is allowed, however, to set higher tariffs up to certain limits for newly constructed lines, as a way to recover their investment costs.  - 13 - 2.15 To sustain China's rapid economic growth, the government must ensure that its scarce capital resources are deployed in the most efficient way, especially for large investment programs such as for the railway system. However, until recently the railway investment planning process had two short- comings as recognized by MOR stafft (a) it underestimated future traffic volumes and therefore investment needs, and (b) it lacked analytical tools. 2.16 The past tendency of the government to adopt relatively low traffic forecasts as a basis for sizing its future investment programs is reflected in the current plan for railway investment in the 1990s, which is based on traf- fic growth rates that are significantly lower than actual rates during the 1980.. For freight, the annual growth rate adopted for the period 1991-2000 of 4.2 percent is much lower than the rate of 6.4 percent realized during 1981-90 under severe capacity constraints. For passengers, a growth rate slightly higher than that for freight (4.5 to 5.0 percent per year) has been adopted, which is also well below the actual rate of 6.6 percent realized during 1981-90, when railway passenger demand was heavily suppressed. This shortcoming in planning seems to have resulted from a combination of the fol- lowing factors: (a) Forecasts of economic growth (and related railway traffic targets for MOR) were based on the goal of quadrupling GNP during 1980-2000 --rather than on macroeconomic conditions that were expected to pre- vail in the 1980. or are likely to prevail in the 19909 (i.e., an economic forecast that takes into account the likely impacts of recent and prospective economic reforms).3/ (b) The size of future railway investment was estimated on the basis of traffic forecasts that were determined on the assumption that rail- way traffic in future years would be constrained by limited railway capacity. 2.17 An undesirable consequence of using low traffic forecasts may be characterized as a "vicious circle" in railway investment planning: a low traffic forecast would likely lead to a capacity expansion investment program that is too small for future traffic. This in turn tends to result in inade- quate railway tariffs and therefore inadequate railway revenues, which in turn would likely cause insufficient funding of railway investment and hence inade- quate expansion of future railway transport capacity and subsequent con- straints on economic growth. Recognizing this problem, the government and MOR have recently taken initial steps to address it (paras. 3.13-3.20 and 5.21- 5.23). 2.18 Compounding the effects of this vicious circle are the results caused by the lack of analytical tools for network analysis. Until recently, MOR did not have computer software to determine the optimal location, scale, 3/ The goal of quadrupling GNP between 1980 and 2000 implies an average annual growth rate of 7.2 percent over the 20-year period. To meet the quadrupling GNP goal, given that GNP already grew at a relatively high annual rate of 8.9 percent during 1981-90, it needs to grow at a lower rate of 5.5 percent per year during 1991-2000.  - 14 - and timing of railway capacity expansion investments. Therefore, MOR has had to select the location and timing of network capacity expansion projects in a suboptimal manner. Due to the interdependence of traffic on different links in the railway network, the implementation of a suboptimal set of projects has placed unnecessarily low limits on the annual throughput of the railway system as a whole. 2.19 Recognizing this deficiency, MOR decided in 1989 to develop a com- puterized decision-support system for investment planning under the Railway Investment Study (RIS) in cooperation with the Bank (Annex 5). Now developed and functional, the RIS decision-support system has been used, in a prelimi- nary analysis, to show that (a) potential gains of roughly 10 percent or pos- sibly more in total traffic throughput (about 120 billion ton-km or 110 mil- lion tons per year) could be achieved by optimizing the sequence and timing of capacity expansion improvements through 1995, and (b) an additional 10 percent potential gain in capacity could be obtained by better routing of freight traffic (Annex 6), all with approximately the same total investment bud- get.4/ D. Maintenance 2.20 Maintenance and rehabilitation of the track system are carefully done. But because manual methods and small tools are used, the quality of maintenance and rehabilitation tends to be compromised (Annex 7). The lack of mechanization in track maintenance, the relatively low weight of much of the rails used, and the poor design of the concrete sleepers and fasteners--these factors contribute to the relatively large magnitude of maintenance and reha- bilitation MOR must provide. To keep up with incraases in train speeds, traffic densities, and axle loads, with Bank assistance the Ministry is mecha- nizing maintenance and rehabilitation of the heavily trafficked lines and improving the standards and design of track components. These actions will result in not only lower train operating costs but also higher utilization of scarce track capacity. 2.21 MOR's fleets of locomotives, passenger coaches, and freight cars are generally well maintained and their availability has been kept high by world standards. For instance, MOR's DF4 diesel locomotives are available for ser- vice about 89 percent of the time, compared to the average availability rates for diesel locomotives of about 82 percent in India and 91 percent in the United States. However, MOR's approach to maintenance of its equipment is relatively inefficient for a variety of reasons (Annex 8). For all equipment, scheduled overhauls and inspections are time-consuming. The 18 different designs of diesel locomotives make it difficult to get parts. Passenger coaches and freight wagons are maintained at several hundred depots, which often perform activities similar to those of the factories. The factories, on the other hand, are not specialized by function. This situation prevents the realization of production economies from the use of modern machinery and pro- 4/ The optimization used by the RIS team was based on a heuristic method that yields a conservative estimate of the potential gain. Optimization using a more exact metixod based on integer programming would tend to yield a larger gain in traffic throughput.  - 15 - ceases. With Bank assistance, MOR is preparing and implementing action plans to solve these problems. E. ComRarison with Other Railway Systems 2.22 MOR is one of the world's most remarkable railways in many respects. A good appreciation of the Chinese railway may be obtained from Table 6, which compares MOR with railway systems in five other countries--France, India, Poland, the United States, and the former USSR--on the basis of (a) system size and transport output; (b) productivity of track, equipment, and labor; and (c) structure of the transport management system.5/ 61 7/ At this general level of comparison, a useful measure of transport output that com- bines both passenger and freiglt traffic into one index is the "converted ton- km," or ctkm, defined as the sum of passenger-km of personal travel and net ton-km of freight. 2.23 System Size and Transport Output. MOR is among the largest rail- ways in the world. At 1.7 million employees, its rail transport work force ranks second, and at 53,000 km, its route length ranks fifth. MOR's combined freight and passenger transport output (at 1,321 billion ctkm in 1990) is third behind the former USSR (at 4,262 billion ctkm in 1989), whose land mass is more than twice as large as China's, and the United States (at 1,614 bil- lion ctkm in 1988), whose economy is more than twelve times as large. 2.24 Although it handles mixed passenger and freight traffic, MOR caters much more to freight than to passengers: its passenger share of combined freight and passenger traffic is only 20 percent, compared with 55 percent for India, 53 percent for France, and 34 percent for Poland. The countries that have a smaller passenger share of combined rail traffic are the United States (1 percent), which relies much more on other modes of transport for intercity personal travel than on the railways, and the former USSR (10 percent), which used to severely restrict the mobility of its citizens. 5/ The comparisons must be made with care, for the following reasons. First, the statistics for different countries are not entirely consistent and therefore are not entirely comparable. Second, different countries have inherent differences that must be taken into account in explaining the contrasts between their railways. 6/ The statistics for the United States include Class I railroads, which carry 96 percent of rail freight traffic, and Amtrak and commuter lines, which carry virtually all of intercity rail passenger traffic. Class I railroads represent about 80 percent of total rail route length in the United States. The remainder belongs to regional and local railroads. The statistics for China exclude local railways (4,400 km of total route length). 7/ The term "transport management system" is used to distinguish the part of MOR that deals with the provision of transport services (both rail and nonrail) from the part that deals with the other a-sivities, such as manufacturing and construction.  - 16 - 2.25 Productivity of Track. Equipment, and Labor. Despite serious administrative and technological limitations, MOR is able to maintain a high level of operating efficiency on the railway system. This efficiency is evi- dent in the high productivity rates of its track and equipment. The produc- tivity of railway track, at 24.8 million ctkm per route-km in 1990, is second only to that of the former Soviet Union (at 28.9 million ctkm/route-km in 1989) and is much higher than those of India (at 6.5 million ctkm/route-km in 1988), the United States (at 7.9 million ctkm/route-km in 1988), and the other countries. 2.26 Similarly, in the productivity of wagons, at 2.9 million tkm of freight hauled per wagon in 1990, China is about on par with the former USSR (at 2.8 millon tkm in 1989) and again is well above the other countries. Indeed, China and the former USSR are in a class by themselves as far as track and wagon productivity is concerned. China's locomotive productivity, at 97.1 million ctkm per locomotive in 1990, ranks second behind the former USSR (at 129.2 million ctkm/locomotive in 1989). MOR is recognized for having the best turnaround time for freight cars in the world (four days for a 990-km average turnaround distance). 2.27 The Chinese railway is able to achieve these feats of operating efficiency despite the relatively olc technology it employs. For example, 42 percent of the locomotives being used by NOR are still steam-powered, whereas no steam locomotives have been used in the United Staites or the former USSR for a number of years. 2.28 MOR's productivity of labor, at 777 ctkm per thousand employees in 1990, is moderately higher than those of the French, Indian, and Polish rail- ways, but is much lower than those of railways in the United States (at 6,232 ctkm per thousand employees in 1988) and the former Soviet Union (at 2,350 ctkm per thousand employees in 1989). The comparison indicates that there is significant room for MOR to improve its labor productivity. This problem is not the most urgent, however, because currently the wages received by the railway employees in China seem much lower than in other countries. At present, staff payroll and health and other welfare costs represent about 22 percent of total railway operating costs--a low proportion by world standards. 2.29 Transport Manazement Structure. The structure of MOR's transport management may be called a "monolith," which is similar to those of railways in former command economies such as in Poland and the ex-USSR, and also those of railways in market economies some 50 years ago, such as in France or the United States. In the monolith structure, each functional department exer- cises total and undifferentiated responsibility over all services.8/ For example, a department may control the maintenance of all coaches, wagons, and locomotives, or another department may control the running of all trains. Given that NOR has been accountable largely for maximizing the quantity and efficiency of transport output, this structure has served MOR well in the past--in an environment where NOR, as the dominant mode of transport, has been able to dictate the services it wishes to provide with little regard to pref- erences of customers. The monolith structure, however, would be less suitable 8/ See Moyer and Thompson, "Options for Reshaping the Railway," Infrastruc- ture and Urban Development Department, The World Bank, March 1992.  - 17 - when China becomes a market economy, because it does not give enough flexibil- ity for MOR to make changes in its services in response to emerging competi- tion and changing customer needs. 2.30 In contrast to the monolith structure, many railways in market econ- omies have undergone organizational transformation into a more modular, market-oriented structure in the past 20 years, to enable them to serve cus- tomers better and compete more effectively with other modes, especially trucks and waterways. Typical market-oriented structures may be classified as "lines-of-business," "wholesaler," "competitive access," "toll-rail enter- prise," or a combination thereof.9/ The French rail:iay has a dominant lines-of-business structure, while the railways in the United States--the freight railways, Amtrak, and the commuter railways--have a combination of all four market-oriented structures (Table 6). The railways under a market- oriented organizational structure tend to share the following traits: (a) They make extensive use of modern information technology and manage- ment tools--including computerization of transport operations, passenger reservation systems, freight and passenger information systems, financial controls, investment planning systems, and man- agement information systems. (b) They use outside suppliers and service providers to perform a vari- ety of tasks such as infrastructure maintenance, administrative, and marketing functions--for example, to maintain wagons and coaches under contract; provide payroll and employee welfare services; and provide special passenger transport over certain routes. The choice of options is dictated primarily by economic considerations. (c) They are market-focused in that: (i) for the markets they serve, they have well-defined accountability (with the government or share- holders) with respect to financial profitability and service qual- ity; and (ii) in designing and pricing their services, they have a relatively clear view of the markets being served, with rec -ct to the competition, the customer preferences, and the financial impact. 2.31 Although the above-mentioned characteristics are common among rail- ways with market-oriented structures, they can be developed at least to some extent within a monolithic structure. Injecting such traits into a monolithic railway should facilitate its transformation into a market-oriented railway. Indeed the Ministry of Railways has alraady started the process of injection: it has (a) begun (with Bank assistance) to modernize railway information tech- nology such as digitalizing its telecommunications network and installing a computerized Transport Management Information System (THIS) (para. 5.30); (b) parceled out railway-related activities from rail transport units to sepa- rate entities such as to trucking ventures in the diversified economy group (para. 3.8) and through "labor contracts" (para. 3.7); and (c) evolved minis- terial bureaus into railway manufacturing and construction companies with increased autonomy. 9/ See Moyer and Thompson, op. cit.  - 18 - III. PROGRESS SINCE THE 1980s A. Improving Railway Management 3.1 Railway Law. In the past ten years, China has taken a number of steps to improve railway management. One of the government's landmark steps has been to issue the Railway Law in 1991 (Annex 3; paras. 2.8-2.9). The Railway Law authorizes the State Council to set forth detailed regulations for railway operations; and the MOR headquarters to set forth "technical regula- tions" such as design standards for track and equipment, plan and construct national railway projects, and set railway tariffs subject to approval by the State Council. The Law entrusts the 12 regional administrations not only to provide railway transport services, but also to meet safety and on-time stan- dards, provide basic amenities and services to passengers, and safeguard against environmental pollution along the railway lines. The Law also empow- ers the MOR headquarters in Beijing to manage the regional administrations. 3.2 The Railway Law's regulatory aspect is important because, before enactment of the Law, there had been no legal framework for the operations of the railways, and therefore no formal allocation of authorities and account- ability between various parties existed. 3.3 Economic Contract. In 1986, the government took a landmark step in decentralizing the administration of the national railway system: it created the economic contract between MOR and the government (para. 2.10). The eco- nomic contract hat expanded MOR's financial responsibility and autonomy, as well as provided MOR's staff with a powerful set of incentives to improve their productivity by various means, including through control of the size of railway work force. Before 1981, MOR had little financial responsibility or autonomy: it paid a 15 percent sales tax on gross revenues and turned over all its net income to the government. The government then allocated budgets to the Ministry for capital investment. This arrangement gave MOR little con- trol of railway capital budgets. Neither did it provide MOR with incentives to control the costs of railway operation and construction, or to allocate railway investment in an efficient manner. The 1986 economic contract and its 1991 successor have removed most of these deficiencies. 3.4 Decentralization in the Railway System. The decentralization efforts in the past ten years, including a new set of policy changes announced in July 1992, have resulted in not only a shift of powers from the national government to the Ministry, but also from the Ministry headquarters to the regional administrations and the construction and factory companies. The second type of decentralization has occurred in the management of projects, the use of profits, and the administration of personnel, to improve productiv- ity and accountability of the regional administrations (Annex 9): (a) Proiects. Previously, the Ministry headquarters managed virtually all railway capital construction projec*s. Now, the headquarters gives the regional administrations budgets to manage rehabilitation, upgrading, and small-scale construction projects. The Ministry headquarters still retains the power LO manage large-scale capital construction projects and projects which have systemwide impacts.  - 19 - (b) Profits. Previously, the regional administrations were not allowed to retain any profit in excess of a specified level. Now, the regional administrations are allowed substantial discretion in using their profits, such as investing in diversified economy activities, paying out bonuses provided the over-ll salary caps are not exceeded, etc. (c) Personnel. Previously, detailed staffing decisions were made at the Ministry headquarters. Now, as part of the evolving enterprise structure within MOR, some of the design institutes and bureaus recently have been restructured into companies with some freedom (e.g., the Locomot4ve and Rolling Stock Company). These semi- autonomous units can decide not only on their internal organiza- tional structures but also on their total staff size, composition and pay incentives, to serve their own needs. Furthermore, these units can promote their professional staff up to the level of Deputy Director. 3.5 In addition, the decentralization has resulted in policy changes that permit the administrations and companies the powers to: (a) make and sell their products, provided that they meet their quotas in, say, traffic to be carried; and (b) export and import goods and services (up to the level of foreign exchange earned from their exports). 3.6 Incentive Pay Systems. In 1986 the Ministry introduced a monthly incentive pay system to improve the management of profits and personnel of the administrations. This system ties the total wage payment of the staff of an administration to the performance of the unit as a whole for the month in question according to a fixed formula (Annex 9). The concerned administra- tion, then, distributes the total wage payment it receives to its staff according to an incentive formula which is devised at its own discretion (but still must be consistent with the railway's wage policy). The total monthly wage payment for the staff of an administration is primarily geared to a pro- ductivitv index. In the case of an administration, this productivity index is the "converted ton-km" handled by the administration in the given month (defined earlier as the sum of passenger-km and net freight ton-km). This primary productivity-index payment is modified by rewards or penalties accord- ing to: (a) the number of times freight wagons are loaded, unloaded or dis- patched; (b) the number of times trains are disassembled or assembled at bot- tleneck yards; (c) the extent to which operating costs are reduced (d) the extent to which the total revenue and coal transport targets are met; and (e) the safety records (Annex 9). 3.7 In 1986 the Ministry introduced a labor contract system to boost staff incentives. Under the system railway workers are contracted to perform specific tasks for specific pays. Between 1980 and 1990 the number of railway workers under labor contracts went up from zero to 310,000, or about 9 percent of the total MOR work force. B. Diversifying into Nonrail Activities 3.8 In 1986, MOR began a major program to divert a portion of its trans- port staff to a wide variety of ventures. Since then these ventures have been  - 20 - growing rapidly, now producing about $200 million in operating profits in 1990, compared with $40 million in 1986. About 50 percent of the profits are attributable to trucking ventures that move freight traffic between customers' warehouses and rail stations. The size of MOR personnel engaged in this diversified economy grew from zero in 1985 to about 300,000 in 1990. The size of NOR personnel engaged in manufacturing and several other nonrail transport activities also grew during 1985-90. As a result, MOR now projects that the share of nonrail transport personnel will increase from 50 percent in 1990 to 60 percent in 1995. This means that the share of rail transport staff will decrease to about 40 percent by 1995 (Annex 2), thus greatly improving the staff productivity in railway operations as the total MOR staff size is expected to remain constant during 1991-95. C. Covina with Inflation 3.9 During the 1980., MOR's profitability and its ability to finance future capital investments were threatened by inflation, as well as by increases in subsidized prices of fuels and materials for the railway. These price increases raised the costs of operating the railway, laying new track, and acquiring new locomotives and rolling stock. To remedy the situation, the government took two actions during 1989-91. First, it permitted MOR to raise passenger fares by 112 percent in 1989, and base tariffs for freight by 24 percent in 1990 and 7 percent in 1991. Second, in 1991 the government authorized the railway to set up an earmarked Railway Construction Fund, to be raised by levying a surcharge on freight traffic. That year, the government also approved an initial surcharge of 0.2 fen/ton-km, representing an average increase of about 7 percent. Despite these increases, NOR stayed just ahead of inflation by end-1991: During 1981-91 the consumer price index in China went up 93 percent, while combined freight and passenger revenues per con- verted ton-km (passenger-km plus net ton-km of freight traffic) went up 110 percent. During that period MOR had difficulties in raising self-gener- ated funds to finance its investment. It was only in July 1992 when the Rail- way Construction Fund surcharge was increased from 0.2 to 1.2 fen/ton-km that MOR became able to stay significantly ahead of inflation and to generate more funds to meet its investment needs. Still, the current level of base tariffs and surcharges is expected to enable MOR to fund only 80 to 85 percent of its Y 116 billion 8FYP investment plan, which at present is considered to be inadequate for the planning economic growth rate of 8 to 9 percent per year (paras. 3.16-3.17). D. Accommodating Past Traffic Growth 3.10 During 1981-91 railway freight traffic vent up 92 percent and rail- way passenger traffic 90 percent. MOR was able to accommodate a large part of these enormous traffic increases by substantially raising the utilization of its track, wagons and locomotives. During this period, MOR raised: (a) the productivity of railway track (measured as ctkm per route-km) 79 percent; (b) the productivity of wagons (measured as ton-km of freight traffic handled per wagon) 43 percent; and (c) the productivity of locomotives (measured as ctkm per locomotive) 43 percent (Figure T4). During the same period, the total route length of MOR's network went up only 7 percent.  - 21 - Fiaure T4: TRENDS IN UTILIZATION OF CHINESE RAILWAY ASSETS (1980-91) 100- - 4 CTKMIROUTEUKM + NTKMIWAGON OWNED * CTKMILOCO. OWNED 3E8 0-. . .. ...... ..... . .. . . . .. ........ .. . .. .. . E 0 o 6 0 - . ....... ........ ... ........ ....../ ''y 40 - .. 200 . ... . .......... ................. . . C 0 _ 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Year 3.11 The more intense utilization has its dark side, however, in the form of line and yard congestion,, which translates into more delays and loes reli- able services to customers, and also in the form of more frequent breakdowns in railway track and equipment. Indeed, the railvay has increased its trans- port capacity by raising its operating efficiency at the expense of service to its customers: it exercises close control of wagon allocation to shippers and imposes limits on passeuger service availability. More than 40 percent of monthly wagon requests vere denied in 1990. Furthermore, many of- MOR's heav- ily trafficked lines need rehabilitation sand upgradingo, the backlog of track that requires rehabilitation is accelerating, and the capacities of two thirds of major railway yards are expected to be reached by 2000 (Annex 6). 3.12 MOR's achievements in railway asset utilization have been helped by three factors that are expected to wane in the future: (a) aome initial slack of transport capacity in the system; (b) relatively simple technological upgrading (e.g., double tracking) sand (c) stringent requirements on shippers (e.g., to order a wagon the shipper must give between 18 to 48 days advance notice, and vhen a vagon arrives the shipper is permitted only some four hours to load or unload it). As the volume of traffic increases faster than invest- ment to expand railway capacity, the scope for MOR to achieve additional capacity by raising the utll4zation of its existing assets is getting smaller, for the following reasons. First, there will be less slack in the transport capacity of railway system. In Just four years, from 1985 to 1989, the number of bottlenecked railway links vent up from only 7 percent of the total number of railway links in the network to 37 percent (par&. 1.12). Secont, future technological modernization vill become more sophisticated and require a  - 22 - longer lead time to implement. And, third, railway customers will require significantly better quality of service. E. Meeting Future Expansion Needs 3.13 To handle railway transport demand in the 1990s, MOR prepared in the late 19809 a capital investment program of Y 116 billion for 8FYP. This pro- gram comprises an array of capital expenditures, with a percentage breakdown shown in Table T2. About 30 percent of the Y 116 billion program represents projects that have been extended from 7FYP. The new line construction in the 8FYP can be divided into three categories (Map M.3): (a) recently initiated new line construction on the periphery of the network that both creates access to new areas and increases rail network connections (Nanning-Kunming, Zhongwei-Baoji, Jining-Tongliao and Guangzhou-Shantou); (b) new lines in the center of the network, comprising sections of a new north-south corridor (Beijing/Tianjin-Wuhan/Jiangjiu-Hong Kong) between the heavily congested Beijing-Guangzhou and Beijing-Shanghai corridors; and (c) some relatively short east-west corridors intended to ease bottlenecks on coal routes running from the Coal Base (Shanxi, Shaanxi, and Inner Mongolia) to major industrial centers along the east coast (including Yuangping-Shijuazhung and Yangguan). The electrification projects in the program are planned for sections of the intensely trafficked, double-tracked Beijing-Shanghai and Beijing-Guangzhou lines, and for rail lines in mountainous regions where diesel power is rela- tively inefficient and double-tracking is expensive as a means to expand line capacity. Parallel track construction projects (double or triple tracking) are concentrated in bottleneck areas, connecting Beijing to the Northeast (e.g., Qir'tit-agdao-Shenyang); Guangzhou to Shanghai; and the resource-rich Southwest t, tha center of the country. Table T2: PERCENTAGE BREAKDOWN OF ALTERNATIVE RAILWAY INVESTMENT PROGRAMS FOR 8FYP (1991-95) Original RIS Preferred Packase program Low High New line construction 36 26 26 Double tracking of existing lines 13 14 15 Electrification of existing lines 7 12 12 Terminal upgrading 4 4 4 New 1-.^zomotives and rolling stock 26 32 33 Track rehabilitation and maintenance 7 6 5 Other categories 7 6 5 Total 100 100 100 3.14 The RIS system was employed to assemble two alternative railway investment packages, by means of: (a) a heuristic procedure for optimizing the scale, location, and timing of railway projects; and (b) an efficient traffic-routing procedure (Annex 6). Of these two "RIS Preferred Packages," one is of similar size to the original Y 116 billion program (Low Investment)  - 23 - (Map M.6) and the other is 30 percent larger in the 8FYP and 50 percent larger over the period 1991-2000 (High Investment) (Map M.7). The percentage break- downs of the RIS Preferred Packages are shown in Table T2 in comparison with the breakdown of the original program (see Table A6.5 for details). While neither of the RIS Preferred Packages suggests a significant change in the share of double-tracking from the original program, both packages suggest that: (a) the share of new line construction be decreased (from 36 percent in the original program to 26 percent); and (b) the share of electrification be increased (from 7 percent to 12 percent) (Table T2). The large changes in tha shares of new line construction and electrification suggested by the RIS anal- ysis is not surprising. This is because the unit cost of capacity expansion by electrification is only 50 to 60 percent of that by the construction of a new single-track line. 3.15 Both RIS Preferred Packages suggest that the share of new locomo- tives and rolling stock be increased, from 26 percent in the original program to 32 or 33 percent. Because the expenditures on locomotives and rolling stock are roughly proportional to the projected traffic throughput of the railway system, the 32 and 33 percent shares of expenditures on locomotives and rolling stock under the RIS preferred investment packages imply that, for the same level of budget, the RIS methodology may help MOR design a railway network that offers 23 or 27 percent more traffic carrying capacity than the original program. The capacity augmentation reflects two major kinds of sys- tem improvement: (a) about 10 percent more throughput from more efficient systemwide traffic routing; and (b) about 10 percent more throughput if the scale, location and timing of investment are optimized for the currently planned investment (Annex 6). This suggests that there is scope for more action by MOR to increase railway network capacity in a more cost-effective Way. 3.16 Although the Y 116 billion investment program is about 70 percent larger than the 7FYP in real terms, it is still based on a relatively low economic growth rate of 6 percent per year, compared to the 8.9 percent achieved during 1981-90. The resumption of economic growth, from 7.0 percent in 1991 to 12.0 percent in 1992, together with the new drive for further eco- nomic liberalization that began in early 1992, caused Chinese planners to revise their future economic growth expectations steeply upwards. In mid-1992 the government decided to adopt for planning purposes GNP growth rates in the range of 8 to 9 percent through the year 2000. The new economic growth rate has raised the expected annual growth rate of railway transport demand from 3.4 percent to 4.4 percent for freight and from 7.8 percent to 10.0 percent for passengers. This has enormous implications on suture railway investment needs: the current level of planned investment probably will meet only two thirds of the higher expected growth in railway transport demand (Annex 6). 3.17 To accommodate future growth of railway transport demand under the 8 percent economic growth scenario through the year 2000, an order-of-magni- tude analysis has estimated that investment in railway capacity would have to be increased 40 to 60 percent over the current level of planned railway investment (Annex 6). If only the currently planned investment is made, the railway will experience many bottlenecks, amounting to an estimated 31 percent of the railway network in 2000 (Map M.8), compared to 37 percent in 1989 (Map M.4). If rail capacity investment is increased by 30 percent (i.e., Y 25-  - 24 - 35 million) in conjunction with improved traffic routing procedures as in the High Investment Package, MOR could expect to reduce the extent of bottlenecks to about 25 percent of the railway network. Investment of a further Y 20- 30 billion in constant 1992 prices by the year 2000 would be needed to increase expected throughput to about 98 percent of demand and eliminate all critical bottlenecks (Annex 6). 3.18 The above estimate is conservative for two reasons. First, it assumes that a set of accelerated reforms will be undertaken that will decrease the growth of rail transport demand or increase railway network capacity. The reforms include two major demand management measures: (a) rationalization of bulk commodity distribution; and (b) increases in rail freight tariffs (both base tariffs and surcharges), which would significantly check demand for rail transport (Annex 1). The reforms also include two sup- ply management measures: (a) to improve freight traffic routing; and (b) to optimize the scale, location, and timing of railway investments (para. 3.14). These two measures together are estimated to yield a 20 percent increase in network throughput capacity. A second reason is that the estimate is based on the simplifying assumption that passenger transport demand grows at the rate of growth of freight transport demand. In reality, passenger transport demand has been growing and is expected to grow at a much faster rate than freight transport demand (paras. 1.6, 1.9). 3.19 Recent increases in the tariff surcharge and planned increases in base tariffs have eliminated most of the potential gap in financing for 8FYP, but they do not cover the gaps in 8PYP and 9FYP that will result from the large railway investment level needed to support the currently expected 8 to 9 percent per year economic growth rate (para. 3.17). An analysis of finan- cial self-sufficiency of MOR's rail transport business concludes that if the railway investment requirements increase 50 percent in real terms from the currently planned level through 2000, the base tariffs, freight surcharge., and borrowings must be raised substantially in real terms in order to bridge the potential funding gaps. 3.20 Besides the question of MOR's ability to fund the investment pro- grams, the question also has been raised as to MOR's absorptive capacity to carry out the programs. This is because the investment program for 8FYP is about 70 percent larger in real terms than the program for 7FYP (Y 51 billion) (para. 3.16). NOR's response to this question is that it should have more than ample capacity to execute the investment programs, by mobilizing a work- force of some 30,000 for engineering design, 600,000 for the construction of railway infrastructure, and 430,000 for the manufacture of locomotives, freight wagons and passenger coaches.  - 25 - IV. FUTURE ROLE OF THE RAILWAY 4.1 The main focus of this chapter is the transport function of MOR,1/ but it will cover briefly MOR's nontransport functions, including railway manufacturing, construction, and other activities. 4.2 In the ongoing process of transforming the Ministry of Railways into a market-oriented entity (para. 2.1), the major goals for its transport func- tion should be for the Ministry to: (a) provide a range of safe and adequate transport services that meet customer needs and compete with other modes on a commercial basis; (b) control costs and refrain from exercising monopoly power at the expense of the public; and (c) maintain financial self-sufficiency, i.e., being able to finance all of its operating costs and capital investments, the latter mostly through self-generation of cash but also through capital market borrowing. A. Maior Challenges 4.3 MOR'g Transport Business. To meet the goals mentioned above, the Ministry munt successfully meet new challenges in a rapidly changing macroeco- nomic environment as China moves toward a market economy. MOR's two most important challenges are: - first, to accommodate future traffic growth in the most cost- effective way, by breaking the "vicious circle" in railway invest- ment planning (para. 2.17) by choosing the best means of capacity expansion; and - second, to provide a wider variety and better quality of railway services in the face of growing customer needs and increasing competition from other modes, mainly through more effective regula- tion of MOR's transport function and improvement in MOR's tranaport management system. There are also other major challenges which include the tasks of: (a) coping with the financial burden resulting from the removal of implicit price and credit subsidies; (b) implementing more efficient operating policies and tech- nologies; (c) developing manpower resources to meet future needs of the rail- way; and (d) protecting the environment resulting from railway-related pollu- tion. J/ The transport function of MOR comprises both rail transport activities (involving 1.7 million MOR staff) and nonrail transport activities (involving local pick-up and delivery services, which are treated as part of the "diversified economy" ventures.  - 26 - 4.4 MOR's NontransRort Business. The nonrail transport business may he divided into four major groups: (a) construction companies; (b) manufacturing companies; (c) diversifi'%d economy companies; and (d) welfare-related units that comprise hospitalh, zlinics, universities, schools, railway housing agen- cies (para. 2.6). These groups, together with the 12 regional administrations which perform the transport function, make the Ministry of Railways a highly diversified organization. With continued expansion of the diversified economy group, MOR will be even more diversified (pare. 2.7). 4.5 Two major challenges confronting MOR in the context of its nontrans- port business relate to the questions of possible divestiture of some of these nonrail transport companies and units, as well as the question of labor ratio- nalization, i.e., how best to deal with MOR staff, especially the rail trans- port staff who are being made redundant by improvements in staff productivity. The question of possible divestiture should be addressed in the context of readiness of candidate companies and units to be divested, e.g., transferabil- ity of the railway welfare systems outside of the railway system, corporate strategies for these divestiture candidates, possible forms of ownership and financial controls for them, possible consolidation of some divestiture candi- dates, actions required to make some divestiture candidates more profitable and more ready for divestiture, etc. (para. 5.16). The question of labor rationalization should be addressed in the context of the prospects for their retraining and reassignment and the absorptive capacity of the diversified economy group which has served as the primary means of absorbing redundant MOR staff. MOR plans to address these issues with Bank assistance (paras. 5.16, 5.18). B. Strengths and Weaknesses of the Railway System 4.6 From the standpoint of meeting its challenges in the 1990., the Chinese railway has main strengths and weaknesses which may be summarized as follows: 4.7 Main strengths: (a) The railway is an efficient, well-functioning system, run by a capa- ble and disciplined staff. (b) Its utilization of locomotives and rolling stock rivals most other railways in the world. (c) It is financially viable, with strong potential to increase revenues further as needed to pay for the cost of network expansion. (d) It is a dynamic organization, which is adapting to the needs of the economy (albeit slowly). (e) It serves industrial and export development in the interior of China with less harm to the environment than a large highway and trucking infrastructure. (f) It is more energy-efficient than truck transportation for medium to long distances.  - 27 - (g) Through expanded use of technology and organizational restructuring, the railway has excellent long-term growth and market retention capabilities. 4.8 Main weaknesses: (a) Because of the fast-growing demand and capacity constraints, manage- ment attention has focused on maximizing operational efficiencies as opposed to improving customer service. (b) Since there is no competition within the railway subsector and only limited competition from other modes, there is little external pres- sure for the railway to innovate and modern.ze. (c) Because it does not have a marketing group within the railway, it does not adequately monitor its competing modes of transport or the long-term needs of its customers. (d) Despite recent improvements, it employs some outdated technology, which not only makes the costs of operation and capacity expansion higher than necessary, but it also makes improvement in the quality of customer service a more difficult task. (e) There are few, if any, policies in place to coordinate the railway's transport services with those of other transportation modes, partic- ularly trucks. C. General Strategv for MOR's Transport Business 4.9 To fulfill its role as a provider of transport services, MOR should adopt a general strategy that will respond to the challenges and opportunities in ways that capitalize on its strengths and minimize its weaknesses. This strategy must respond to market forces as well as government demands, on the basis of a clearly defined relationship between the railway and the govern- ment. 4.10 The evolving macroeconomic environment with more emphasis on eco- nomic reforms will progressively lead to increased market competition in the next decade. Consequently, the role of the railway will inevitably change, and the railway will need to be more responsive to the needs of freight and passenger customers. Quality and cost of service will become more icnportant in deciding whether the railway or its competitor satisfies transport demand. The railway will need to provide more economic door-to-door transport and become more specialized in order to respond effectively to the needs of each transport market that it faces. 4.11 Although the railway must ultimately decide on a specific strategy to face each transport market, there are broad areas of specialization that respond to the general market trends. In particular, the railway should focus more on the following major categories of traffic: (a) medium- to long-distance bulk traffic in unit train loads or multi- ple-car loads (and at the same time let the truck pick up more of  - 28 - short- to medium-distance general merchandise traffic, with less- than-car loads or single-car loads); (b) container traffic, with door-to-door services in full cooperation with other modes; and (c) medium- to long-distance passenger traffic (potentially including some with high-speed services). Conversely, the railway should gradually containerize less-than-car-load traf- fic and phase out single-car, short-haul traffic by substantially increasing tariffs for these services or making policy decisions to discourage the ser- vices or both. 4.12 For certain services that are being performed at a financial loss (e.g., some lightly trafficked lines and possibly suburban commuter services), unless tariffs can be adequately raised the government should consider changing MOR's role from being a subsidizer back to its core role as being a transport provider. In this context, the railway should carefully examine its core functions in order to identify those functions that are provided at a financial loss, but that are gublic service obligations (PSOs) and therefore require proper compensation from the government. By specifying contractual terms for these services with the government, the railways role can then &hift from that of a subsidizer to that of a transporter receiving direct subsidies for each PSO on a contractual basis. D. Government's Supportive Actions 4.13 Once the desired roles for MOR are agreed, it is critical for the government to take actions in areas that are beyond MOR's control to help MOR play its role in the most effective way. These actions include the shift to contractual, direct subsidies, the rationalization of railway transport demand by improving the commodity distribution system and restructuring of railway tariffs and prices of major commodities (especially coal) to reduce excessive use of transport (para. 5.8(c)]. 4.14 Until the demand for railway transport is entirely met, a possibil- ity that may not be realized before the year 2000, the government has no alternative but to ration scarce railway transport capacity. In rationing railway capacity, a key question is: How should the limited capacity be allo- cated between passengers and freight? When railway capacity is severely lim- ited, as is the case of the Chinese railway system at present, the choice between passengers and freight is a difficult one: giving too little capacity to passengers means continuation of the policy of restricting intercity mobil- ity of Chinese citizens, as well as interprovincial trade; giving too little capacity to freight, on the other hand, means restriction on China's economic growth. The choice should therefore be made carefully. One possible approach is to base the choice on the relative economic priorities of different types of freight traffic and different purposes of passenger travel. This type of analysis requires knowledge of the willingness-to-pay for travel of different kinds of passengers. Such knowledge can be obtained from some research and analysis.  - 29 - V. KEY TASKS FOR MOR'S TRANSPORT BUSINESS 5.1 Within the broad corporate strategy for the transport business out- lined above, the railway must carry out tasks, with support of the government, in a variety of areas related to regulation, internal management, investment and finance, and systemwide technological modernization. A. Achievina More Effective Railway Regulation 5.2 Regulation of MOR should aim at helping MOR as China's largest pro- vider of transport services to achieve the following objectives: (a) to structure tariffs and other charges in order to improve the economic efficiency of railway operations; (b) to provide safe transport services of adequate quality to the public at the lowest possible cost; and (c) to maintain financial self-sufficiency without making unnecessarily high profits. 5.3 The issues of appropriate regulation of railways are complex, and obviously cannot be solved by simple formulas. However, one key lesson has emerged from experience with successful deregulations of transport enterprises in market economies since the mid-1970s: no matter how sincerely motivated or adequately empowered they may be, regulatory agencies inherently restrict the railway's ability to compete effectively with the other modes. In other words, the least possible regulation is the best. 5.4 With the above lesson in mind, an effective approach to regulating the railway system should be to minimize its control where applicable and to foster competition where possible. Within this broad approach, the existing railway regulatory system should be enhanced by four key measures: first, by promoting competition; second, by improving the government's control of MOR; third, by establishing an independent regulatory agency, not to constrain MOR but mainly to monitor MOR's performance; and fourth, by reforming the railway tariff system. 5.5 Promotina Competition. Competition in the provision of transport services will progressively become keener in the 19909 as the other modes develop. This process will be accelerated as more economic reforms are imple- mented in parallel. By the late 1990. or perhaps the early 2000s, the Chinese railway may well have to face market situations similar to those faced by other railways: a variety of traffic MOR has been handling, including light manufactured goods, may progressively shift to truck transport. The new macroeconomic environment will encourage not only intermodal competition and but also more demand for high quality and specialized transport services. 5.6 The government should not only welcome emerging competition in the transport sector but also promote it. Competition should be promoted from outside as well as within the railway system. External competition is already occurring mainly from the truck and waterway modes for freight and from the  - 30 - bus and airline modes for passengers.l/ It should be further encouraged-- mainly through rational pricing, further relaxation of market entry and con- trols, more extensive development of highway and waterway infrastructure net- works and greater modernization of trucks, vessels, and barges. Internal competition is beginning, as seen in an example in which railway staff working in the diversified economy group is permitted to offer local pick-up and delivery services to the public (Annex 2). This appears to be a good example of internal competition provided that the private sector is given an equal opportunity to compete. Internal competition through participation of the private sector should be strongly encouraged wherever possible--for example, by permitting private individuals or companies to bid for the provision of railway transport services on certain routes. 5.7 Competition, however, does not apply everywhere in China. Because of the present lack of development in the competing modes, the national rail- way system still has a monopoly or a near-monopoly in many if not most of the railway transport markets. For these markets, the railway system should be regulated, but a less strict form of regulation than the existing one could be more effective, as discussed below. 5.8 ImRroving the Government's Control of MOR. This could be achieved by augmenting the economic contract system for MOR through three basic means:2/ (a) Generation of a Strategic Plan. This task would be carried out jointly by MOR, SPC, and the State Council. Its main purpose would be to prepare a long-range plan for the railway that relates China's macroeconomic policy and forecasts to NOR's marketing plans. The market plans involve identification of the major market segments to be served by the railway. This task, which could be carried out as part of MOR's 10-year strategic plan, would identify all policy issues that affect the railway, such as issues related to an overall plan for financial self-sufficiency including cost recovery from users as well as public services; freedom for MOR to set railway tariffs (paras. 5.11-5.13); public service obligations (PSOs) of the railway; and policy for treating railway staff (reassignment, retraining, etc). It then would analyze options for dealing with the issues, and recommend policy decisions. (b) Enhancement of the Economic Contract. This task, which should be carried out jointly by MOR, SPC, and the State Council, would build up the existing economic contract (para. 2.10). It would recommend a formal statement of expectations, authorities, and obligations of MOR and the government over a period of time (say, 5 years), related to the fulfillment of the railway's economic and social role. This 1/ The provision of truck and waterway transport by collectives and individ- uals is lightly regulated with respected to pricing and entry and exit (Guangdong Transport Study Report, 1990). 2/ Adapted from Techniques for Railway Restructuring by L.W. Huff and L.S. Thompson, Railway International, October 1990.  - 31 - task would be consistent with the Railway Law with respect to the rights and obligations of the railway. It would address the ques- tion of how best to provide MOR with incentives to improve quality of transport services. (c) Develonment of an Enabling Action Plan. This task would recommend a list of actions that SPC and the State Council should agree to take to enable MOR to fulfill its social and economic role in the most cost-effective way. Such actions may include the rationalization of railway transport demand (including the restructuring of prices of major commodities), rationalization of the administrative system of materials distribution, and rationing of scarce railway capacity between passengers and freight (paras. 4.13-4.14). 5.9 MOR is interested in conducting an exercise along the line sketched above through a cooperative program with the Bank under a future railway operation. Moreover, MOR is interested in reinforcing this exercise with a program to strengthen its transport management system (paras. 5.14-5.20). 5.10 Creating an Independent Regulatorv Agencv. Under the present regu- latory system, MOR carries out a significant amount of self-regulation: MOR headquarters enforces the newly issued Railway Law as well as monitors on a daily basis the performance of the regional administrations in terms of: (a) the level of transport output; (b) the efficiency of the equipment and plant; and (c) the on-time and safety records of train operations.3/ Although MOR seems to carry out these functions well, an independent regula- tory agency should be more effective in monitoring and publishing MOR's per- formance statistics in areas that are important to the government and the public, such as profitability, efficiency, safety, and service quality (including the degree of customer satisfaction). This arrangement would also help avoid any appearance of conflict of interest. To avoid overburdening MOR with unnecessary constraints, the regulatory powers for this kind of agency should be kept to a minimum. 5.11 Reforming the Railway Tariff System. Deserving a special emphasis is reform of the railway tariff system, because of its critical role in three areas: first, raising needed funds to finance railway capacity expansion; second, enabling MOR to market and price its transport services to compete with the other transport modes and to meet the needs of railway customers; and third, boosting economic efficiency by reducing excess transport demand on the railway system and by encouraging MOR and shippers to increase mutual gains through contract pricing. The railway tariff reform should consider incorpo- rating the following principles, which have been adapted from successful but hard-won experience of many railways in market economies: (a) Unless the government is prepared to pay the difference directly, railway tariffs should not be less than the increase in cost associ- ated with the addition of the traffic in question and the increase 3/ These performance statistics are wired to MOR Headquarters at 18:00 hours every day from the regional administrations. Hence the name "the 1800 report.'  - 32 - due to inflation and/or the removal of subsidies. In other words, railway tariffs (including surcharges) at replacement cost should not be below full variable costs, including deprecietion at current prices. (b) If the government does not intend to pay for fixed costs A/ (i.e., the difference between variable costs and total costs) or to inject funds for future capacity expansion, then railway tariffs must exceed variable costs by an amount sufficient to cover the total costs and the amount of financing needed (in the form of debt ser- vice coverage and/or in the form of direct funding of the capital investment program). This is accomplished efficiently by a railway rate structure which deviates the most from variable cost where demand is the least sensitive to price (the "inverse elasticity" rule).5/ (c) Except for the three rules discussed below, the railway should be allowed almost total flexibility in setting tariffs--to meet compe- tition, serve the needs of its customers, and meet the financial targets set for it. This flexibility includes quality of service guarantees backed by rate rebate provisions, nonpublic contract rates, and rates based on the provision of capital investment (wagons, unloading facilities, specialized equipment, etc.) by the shipper or customer. The flexibility is subject to the following boundaries: (i) The regulatory agency should see that total revenues of the railway do not exceed its total needs, including investment; (ii) The regulatory agency should ensure that individual rates do not rise so high as to reflect an abuse of market power by the railway; (iii) The regulatory agency should set rules that will promote the efficiency of the railway and counteract the influence of its strong market position. 5.12 Implementation of these rules is not easy, and is subject to abuse on the part of both the railway and the regulatory authority. Success depends on the exercise of restraint by both. Increasing competition, possibly in the next 5 to 10 years, should make rules (i) and (ii) less important. Rule (iii) has been hard to implement. The best approach to emerge has been to set an A/ Virtually all governments in market economies, just as the Chinese gov- ernment, have decided not to do so. S/ In the Western economies, this has led in some cases to highly complex tariff structures distinguishing among users with respect to time of day, season, direction, commodity or class of service, size of shipment or group, quality of service required, and many other factors. In China, the tariff structures should be kept simple, initially by applying this rule to broad groups of commodities by service type.  - 33 - arbitrary but reasonable margin (roughly 200 percent in the United States and Canada) by which the proposed tariff may exceed variable cost without suggest- ing abuse of market power, and thereby requiring specific government approval. Proposed rates beyond this threshold may be reviewed--but not necessarily overturned if the regulatory agency decides that the rate does not unduly restrict the flow of the traffic involved. 5.13 The government and NOR have recently agreed to some increase in tariff flexibly for new lines and new services. However, more flexibility is needed to allow MOR to better serve the economy. With Bank assistance, MOR has launched a tariff study with three main objectives: (a) to ensure MOR's financial self-sufficiency; (b) to reduce structural distortions in the tariff system; and (c) to give MOR enough flexibility in pricing and marketing its services. The study is expected to propose tariff reform recommendations later this year. B. Bolstering TransDort Management 5.14 Although there is commitment by top MOR management and by higher levels of the government to adapt MOR to China's changing needs, because of its size MOR is slow to change. As China evolves further into a market econ- omy, the railway management and staff need to gain a deeper appreciation of the ways in which freight customers' production and distribution costs are influenced positively as well as negatively by the railway's rates and service characteristics, particularly its reliability. They also need to gear their organization, financial controls, performance goals, and personnel incentive systems to ensure that the future market-oriented role of the railway will be fulfilled. The performance standards and personnel incentive systems should focus less on meeting transport output targets and more on meeting service quality standards. It is important to ensure that the high physical effi- ciency of railway operations MOR now enjoys will not be sacrificed in the quest to boost service quality. Both physical efficiency and service quality could be achieved through an enhanced economic contract between MOR and the government (para. 5.8(b)], as well as an adequately structured transport man- agement system that is supported by properly trained staff and modern informa- tion technology. 5.15 Enhancing Organizational Structure. MOR should examine its own internal organizational structure and effect the necessary changes to make the railway more effective in fulfilling its role. A particular issue that should be addressed is the appropriate structure of the transport management system to be adopted by MOR as it enters a new phase of market-oriented management. The current transport management system of MOR may be characterized as a "monolith" (para. 2.29). This structure allows MOR to produce transport out- put at high physical efficiency, but does not provide enough flexibility foi MOR to respond sufficiently to growing competition and changing customer needs. Even within the monolithic structure, however, it is possible for MOR to take concrete steps to sharpen its competitive edge and market orientation to some extent (para. 2.31). To advance further would require the national railway system to evolve a different organization structure. Depending on the objectives set for MOR, and on the trends expected for the development in the Chinese economy, various structural options, including "lines-of-business" and "wholesaler" options in particular, deserve consideration.  - 34 - 5.16 Rationalizing the Railway System through Divestiture. A potential mear.s to improve efficiency of the railway system is to reduce its size and make it less vertically integrated--by spinning off into separate enterprises some selected railway units that do not perform the core railway transport fur.ction, namely, the railway construction, and manufacturing companies, as well as the diversified economy ventures (para. 2.7) and those units that perform welfare functions. Such a divestiture has the potential to generate cash for financing much needed railway capacity expansion investment (para. 5.27(e)). It also should improve economic efficiency by (a) relieving the onus of management on the railway system; (b) encouraging stronger competition among divested units (e.g., allowing divested railway construction companies to compete more freely for construction projects); and (c) permitting the divested companies and the remaining railway transport units to separately develop recruitment and compensation policies that suit their own needs, unen- cumbered by the weight of a larger bureaucracy. MOR plans to launch a cooper- ative program with the Bank to address the Dossibility of divesting nonrail transport functions. 5.17 Developing Railway Human Resources. Manpower development is an integral part of the possible modernization of MOR's transport management system. The Chinese railway system needs to upgrade the skill levels of its technical staff in order to fully exploit the advanced technologies that are being introduced. The railway also needs to train its management staff in order to meet stronger competition from other modes. Fresh recruits should be carefully screened and their numbers weighed against the needs to contain or reduce staff levels. 5.18 As part of the modernization exercise, MOR would forecast the opti- mal size and composition of its staff targeted to be achieved in the next decade. It then would examine the existing staffing and all major personnel policies, and prepare a manpower development plan to achieve the staff tar- gets. The manpower development plan would address the compensation policies, including the pay scale, performance incentives, employee benefits, reassign- ment and retraining of railway staff, and recruitment of new employees. The plan would also address the question of how to deal with redundant railway staff (para. 4.5). 5.19 Fostering MOR's human resources would enable the railway to achieve stronger management, with flexibility in hiring, firing, and rewarding of per- formance. MOR plans to cooperate with the Bank in this effort. 5.20 Implementing a Management Information System. Efficient and effec- tive management of a vast transport organization such as MOR requires care- fully designed management information systems. Such systems would provide railway managers at different levels with the right kinds of financial, oper- ating and service performance information needed for them to monitor and con- trol railway operations, and to make well-informed decisions. The Bank has supported MOR's quest to develop an MIS capability. MOR has completed testing its pilot MIS project at the Harbin regional administration and is developing a plan to implement a networkwide MIS in conjunction with the computer-based Transport Management Information System (TMIS) being installed (para. 5.30). To benefit from this MIS on a large-scale basis, MOR is taking actions to:  - 35 - (a) complete implementation of the pilot MIS in the Harbin administra- tion; (b) undertake a MIS study at the MOR headquarters level that would com- plement the system being installed in the Harbin administration; and (c) develop plans for revising the MIS in conjunction with the creation of an interface between it and the TMIS, and implemerting the revised MIS in the regional administrations. C. Strengthening Railway Investment and Finance 5.21 Buttressing Investment Planning. To accommodate the traffic growth in the 1990s, the vicious circle in investment planning must be broken (para. 2.17). To this end, two major steps have been taken recently. First, the government has since mid-1992 officially adopted for planning purposes a more realistic economic growth rate of 8 to 9 percent per year, compared to 6 per- cent employed previously (para. 3.16). As a result, infrastructure investment plans, including railway investment plans, are expected to be revised upwards. Second, with Bank assistance the Ministry of Railways has carried out fore- casting of railway transport demand under a scenario in which GNP is assumed to continue to grow at 8 percent per year through the year 2000. 5.22 This scenario has been adopted by the RIS decision support system to: first, prepare an investment program that would be used as a basis for setting railway tariffs and forecasting MOR's borrowing needs; and, second, predict the consequences of not having enough budget to finance the needed investment program. The results of this RIS analysis exercise indicate that: (a) the level of railway investment program for capacity expansion needed to accommodate the forecast traffic level will be some 40 to 60 percent larger than the currently planned level; (b) if the currently planned level of rail- way investment is implemented, then the level of unsatisfied demand for rail- way transport will be 80 percent higher than currently projected; and (c) the total economic loss associated with the unsatisfied demand will be 0.5 to 1 percent of GNP (Annex 6).6/ Thus, the investment to expand capacity would reduce unsatisfied demand and thereby lessen economic losses due to bottle- necks. The benefits from this reduction are on the same magnitude as the expansion investment itself. RIS analysis has also been carried out to assess traffic flows and determine priority investments under budget constraints. 5.23 At present, MOR is discussing with SPC the possibility'of enlarging the railway investment plans in response to the new planning economic growth rate. 5.24 Choosing the Best Means to Expand Railway Capacity. The Ministry is faced with the task of expanding railway network capacity at a much faster rate than in the 1980s. This is a major challenge because MOR is already having difficulty in financing the capital program from its borrowings and 6/ This loss will be twice as large if MOR does not achieve the increase in throughput due to improved routing efficiency assumed in the RIS analysis.  - 36 - internally generated funds (paras. 2.12, 3.9). Therefore, MOR should strive to stretch its financial resources as far as possible by more aggressively choosing the most cost-effective means to expand rail transport capacity. There are three main approaches: (a) Adopt a rigorous means to identify high-priority investments and develop an economically optimal investment plan under budget con- straints. An order-of-magnitude RIS analysis has indicated that optimization of the location, scale, and timing of railway invest- ments could yield gains of 10 percent or perhaps more in systemwide rail traffic throughput. The 10 percent gain represents some $4-5 billion of savings in railway construction cost through the year 2000 (Annex 6; para. 2.19). (b) Implement technological measures and operating methods aimed at increasing railway capacity at lower capital and operating costs (e.g., by running more unit trains, introducing bypass-blocking concepts, allowing heavier axle loads, reducing the dead weight of wagons, and the like). The results of the capacity simulation study conducted by NOR in cooperation with the Bank suggest possible key measures that can be taken to increase railway capacity by as much as 30 percent at relatively small incremental costs (including mea- sures such as increasing the train length further than being imple- mented now, boosting the power-to-weight ratio of the train, decreasing minimum headways and increasing the train stopping dis- tance).71 In addition to this capacity simulation study, the Bank is supporting a new study by MOR aimed at determining and imple- menting cost-effective technological measures and operating methods on a comprehensive basis. (c) Improve the routing patterns of freight traffic in order to accommo- date more long distance traffic on the railway system. This action would thereby increase the total ton-km of freight traffic to be carried by the railway system. A preliminary analysis conducted with the RIS system has produced results that seem to indicate that improvements in freight routing patterns can potentially increase the total ton-km of railway freight traffic by about 10 parcent within current line capacities (paras. 2.19, 3.15). With some enhancements the RIS system can be used to carry out a more rigorous analysis of freight routing patterns. 5.25 Improving Railwav Financial Manaxement. Besides reforming the tar- iff system, MOR needs to implement other financial reforms, in order to pro- vide proper management of its operating performance and financial condition and to lay down a foundation for longer term capital and operational planning for the railway as a transport enterprise. Specifically, the railway needs to: 7J1 As summarized in World Bank Infrastructure Note No. RW-3, "How to Improve Capacity: The Case of China Railway," by E.P. Anderson, L.C. Davis, and C.D. Van Dyke (April 1992).  - 37 - (a) prepare all financial statements according to guidelines established by MOF (including income statements, balance sheets, and fund flow statements); (b) revamp the railway accounting system, to make it uniform throughout and conform better with international practices; (c) establish a costing system for determining the variable and fixed costs of its transport services to guide investment planning, cost control, selection of alternative railway services, and tariff set- ting; (d) institute more effective financing plans and internal control sys- tems, to enhance the financial accountability of various railway and nonrailway units; and (e) improve the use of scarce foreign exchange, by focusing more on the acquisition of foreign know-how, equipment, machinery, and the like, for which there are no domestic substitutes, as distinguished from items such as rails and other steel products that are available in China. 5.26 The Bank will continue to support MOR's efforts to improve its financial management. Under ongoing railway operations the Bank has supported a railway costing study. Under a future railway operation the Bank would be supporting MOR's efforts to prepare and implement action plans to improve (a) the railway accounting system, (b) the railway tariff system, and (c) MOR's capability for financing its investments. Additionally, a future Bank operation could support MOR's efforts to augment its cost control. 5.27 ExRanding the Scope of Financial Resource Mobilization. MOR should consider using as many means as possible to obtain funds to finance its vast investment requirements. Besides raising railway tariffs and surcharges, MOR should consider: (a) sharing the cost of installing fiber-optic cables with other users (para. 5.30); (b) entering into joint ventures with other investors (domestic and foreign) to provide specialized transport services (e.g., import- export container transport services); (c) letting major coal mines, coal users, or other railway customers own freight wagons; (d) taking advantage of the new economic liberalization policy by jointly investing with foreign investors on new railway construction projects; (e) divesting nonrail transport functions and using proceeds from the divestiture to finance capacity expansion (para. 5.16); and  - 38 - (f) expanding the scope of borrowing, both domestically and internation- ally, for example, by raising more domestic bonds, and borrowing from international capital markets [e.g., through an Expanded Cofi- nancing Operation (ECO) supported by the Bank]. 5.28 Regarding international funding, the major existing sources, namely, the Overseas Economic Corporation Fund (OECF) of Japan and the World Bank, are already approaching their lending limits for the Chinese railway system. Since these two sources combined contribute no more than 10 percent of MOR's capital investment requirements, MOR should consider expanding the scope of international borrowing into private capital markets. A potential first step would be for MOR to undertake an ECO with a future Bank loan. This would enable MOR to borrow hard-currency funds from a private capital market under a guarantee of government-related risks provided by the Bank. Such an ECO would help attract new funding sources to MOR, reduce costs of borrowings, and extend maturities (up to 15 years) by tempering political risks to lenders. D. Enhancing the Qualitv and Efficiency of Railway Services through Modern Technology and Techniques 5.29 Improving Service to ShiDpers. With its existing technology, MOR has made significant gains in (a) expanding network capacity, (b) increasing the utilization of locomotives and rolling stock, and (c) reducing the costs of operation. Some of these improvements in freight operations, however, have been at the expense of the shipper that has to place its order for wagons between 18 to 48 days in advance of the loading date, to accept a wagon that is not ideal for the commodity, and to load the wagon within four hours after it arrives. Shippers in a modern market economy would not be able to compete successfully within these rigid constraints. As China becomes a market- oriented economy, Chinese shippers will require a more flexible transport system to react efficiently to dynamic market forces. This will require the introduction of modern computer-based, transport management information systems (THIS) that will enable the railway to make substantial, rapid changes to its transport plans without hurting MOR's excellent utilization of locomo- tives and rolling stock. A major upgrade of the railway's telecommunications system is of course a prerequisite to the implementation of a TMIS. The for- mation of marketing groups at the administration level might also be helpful. The objectives would be to increase revenue and operating profits in the near term and to develop closer relationships with shippers in the longer term. 5.30 The Bank has supported the Ministry's evaluation of alternative approaches to developing and implementing a TMIS with an associated telecommu- nications system. A future Bank operation proposes to finance implementation of a THIS and associated telecommunications system. To capitalize on their large economies of scale, and make more capital available to railway telecom- munications, the Bank is encouraging MOR and the Ministry of Post and Telecom- munications (MPT) to jointly install fiber-optic cables along railway corri- dors. 5.31 Modernizing Container Transport Service. Given the importance for the railway to adapt itself to changes in transport demand in the future, par- ticularly the needs of higher-value cargo for better quality service, and the international trends toward containerization of imported and exported cargo,  - 39 - the railway is giving higher priority to container service. Container service requires a different approach from the traditional approach to rail transport service in China. This type of service must be more customer-oriented, with higher frequency, faster travel times, and greater rel4a1illity than regular freight service, and it should be closely coordinated X 31 other modes, as well as with MOR's own local pick-up and delivery servic-s. These require- ments necessitate changes in train operations, organization, and staff atti- tudes. The Bank is providing technical assistance to MOR's container trans- port study that addresses all these issues. Also, the Bank is helping MOR initiate a pilot project to test and demonstrate the effectiveness of a modern container transport service. Based on the recommendations of the action plan developed under the container transport study, the Bank could help MOR estab- lish a nationwide network of container service. 5.32 Upgrading Passenger Service. As the economy is becoming more market oriented, railway passengers will require higher standards of service. To meet these increasing requirements, MOR must address several areas of quality improvement, including: (a) the provision of more passenger capacity at full cost, where there is substantial unmet demand; (b) the installation of passenger reservations systems on a networkwide basis; (c) the setting up, monitoring, and control of standards of customer service; (d) the provision of customer information systems for passengers; (e) the improvement in both on-ground and on-train services, including the provision of more amenities at passenger stations, the contract- ing out of catering services on trains, the answering of customer queries, etc.; and (f) the improvement in passenger and pedestrian safety. 5.33 The Bank is encouraging MOR to undertake a feasibility study for a passenger reservations system in the near future. The cost of implementing such a reservation system will be relatively small compared with the cost of the ongoing upgrade to the telecommunications system. 5.34 Raising Operating Efficiency through Better Maintenance. Despite the care given to their maintenance, MOR's plant and equipment tend to wear out faster or fail more often than usual, because of: (a) the heavy utiliza- tion of plant and equipment; (b) the inadequate designs of some critical com- ponents; and (c) the heavy reliance on manual methods, hand tools, and anti- quated machine tools that lowers the quality of maintenance. These three factors tend to cause disruptions in railway operations, which in turn reduce both the operating efficiency and transport capacity of the railway. 5.35 As an example, the maintenance of the track system, while generally good, is vulnerable to the problems mentioned above. First, although the  - 40 - design of NOR's rails is basically sound, the rails are often too light for the traffic, causing them to deteriorate relatively fast. Second, inadequate designs of concrete sleepers and fasteners are causing added rail wear. And, third, the poor drainage of much of the rock ballast is shortening the life of the rails, sleepers, and fasteners. To solve these problems MOR has taken steps with Bank assistance to: (a) install mechanized on-track machinery to clean ballast and restore track alignment and surface; (b) replace existing track with heavier and more durable rails; tc) improve the designs of concrete sleepers and fasteners; and (d) develop better methods of track maintenance that rely on the use of advanced on-track machinery, track recording cars, and other inspection equipment. Also, with Bank assistance MOR plans to improve track maintenance management and to acquire modern machinery for mechanized track maintenance. 5.36 As another example, the maintenance of the diesel locomotive fleet, while generally good, also needs to be more efficient. The key measures that need to be taken by MOR include: (a) improving the design and manufacture of equipment parts; (b) consolidating the maintenance functions and work loads of repair shops; (c) reducing the number of designs of locomotives and rolling stock; and (d) achieving better standardization of equipment parts. The Bank has helped MOR improve the maintenance of locomotives, freight cars and passenger coaches, through a variety of measures including the provision of critical components, and the establishment of a unit exchange maintenance system for foreign locomotives. The Bank will also help MOR modernize the design of several components as well as the machine tools for making them. 5.37 Boosting Environmental Protection. Since the setting up of a department within MOR for public health and environmental protection in 1972, MOR's environmental activities have been strengthened through the establish- ment of environmental training schools, rules and regulations, and a series of investments to correct environmental problems. By not using modern technology and techniques of environmental protection, however, NOR's environmental man- agement and control activities have not been as effective as they should be. MOR needs to: (a) familiarize its environmental staff with modern technology for monitoring and control of air and noise pollution and of solid and liquid waste; (b) upgrade the capability of its environmental staff for assessing environmental impact and for developing strategies and priorities for tackling environmental problems; and (c) conduct studies of its current problems with passenger health, waste from trains, litter along the railway track, air and noise pollution in railway factories and workshops, and industrial waste, to develop and implement action plans for remedying the problems. 5.38 A first cooperative program for railway environmental protection between MOR and the Bank has been undertaken, with funding available from UNDP. The main objective of this program is to prepare an action plan to alleviate some of MOR's most pressing environmental problems, namely, pollu- tion related to passenger trains and noise pollution caused by passing trains.  - 41 - This progr,a is bolng followed by a second program designed to strengthen ORX's capability in performing environmental protection activitie, a. well as In devoloping medium- and long-term strategies for alleviating railway envi- ronmental pollution. These action plans are intended to be implemented under a future bank railway project.  - 42 - VI. MEDIUM-TERM BANK ASSISTANCE STRATEGY A. The Bank's Contributions in the Past 6.1 Since 1983, the Bank has assisted the government through five national railway operations and one local railway operation, totaling $1,365 million in loans and credits. The primary objectives of the first four national operations were to finance railway construction to expand traffic capacity in high-priority corridors (Map M.3) and to introduce new technology for improving railway operating efficiency. The fifth national operation begar a new approach of supporting railway policy initiatives and tech- nological modernization on a systemwide basis. Future railway operations in the next five years are intended to implement the new "policy-technology" approach more extensively. 6.2 In the course of assisting MOR the Bank has tried to expose MOR to new ideas on systemwide modernization, but, to the extent possible, to assist NOR only with initiatives that the railway believes to have strong potential for a high payoff. At the beginning of the Bank's railway operations in China, MOR was unfamiliar with most modern management techniques and technology, and therefore was reluctant to embark upon programs of technical cooperation whose benefits it did not see clearly. Since then, the dialogue between MOR and the Bank has bren enriched through a range of such programs. A summary of these programs and newly proposed programs is given in Annex 10. Now, MOR has become more convinced about the potential benefits of such tech- nical cooperation with the Bank, and it is much more receptive to technical assistance from the Bank and others in a wide variety of subjects ans disci- plines. 6.3 A recent audit by the World Bank's Operations Evaluations Depart- ment, which reviewed the First Project in the context of the subsequent four operations, supports the Bank's "policy-technology" approach. It makes a major observation: Significant gains in economic efficiency can still be made by increasing operating efficiency, but larger gains are likely to be made on a broader basis, by alleviating transport bottlenecks through capacity expan- sion, or by reducing freight traffic intensity, or by providing higher quality transport services. B. Objectives and Scope of Bank Assistance 6.4 Future Bank railway operations in China in the medium term (i.e., next five years or so) would help MOR deal with the issues discussed in Chapter V on the basis of the policy-technology approach mentioned above. Specifically, these operations would support four of MOR's objectives: (a) Policv initiatives, in railway investment planning, tariff setting and regulation, economic contract, staff compensation and incen- tives, financial management, financial resource mobilization, etc,; (b) institutional develoDment of the railway system, potentially with respect to manpower development and rationalization; development of corporate strategies for rail transport units, railway construction  - 43 - and manufacturing companies, and diversified economy ventures; divestiture of nontransport functions; and modernization of railway welfare functions (including health, education and housing); (c) capacitv expansion investment on a rational economic basis; and (d) svstemwide technological modernization, covering management informa- tion systems, computer and telecommunications technology, service- quality improvement, plant and equipment maintenance, and environ- mental protection. 6.5 The Bank's support would be provided through (a) technical assis- tance financed by the Bank's own resources or under bilateral or multilateral grants (e.g., from the Japan Grant Facility or UNDP); and (b) Bank operations. Possible contents of these operations are listed in Annex 11. Annex 12 summa- rizes the key issues or objectives that ongoing and future railway project components are intended to address. Annex 13 shows an approximate time frame for implementation of possible actions under these components.  - 44 - CONTRIBUTIONS Apart from those mentioned on page (i), the following people have contributed to this report: * Robert Burns (Reviewer: White Cover) * Pedro Geraldes (Reviewer: White CoverJ * Tariq Hassan (Railway Law] * Frida Johansen (Reviewer: Green Cover) * Hernan Levy (Reviewer: White, Green and Gray Caversl * Kang-Ning Xu (Graphics, Statistics)  - 45 - Table 1 Table 1 CHIMNA: FreiLht Traffic bv Mode Domestic Pipe- civil Ocean Rail Road watety Une aviation Total Shiogint Traffic (bilion ton-km) 1952 60.2 L4 11.8 - 0.032 73.4 2.8 1977 456.9 25.1 102.1 387 0.076 622.9 174.1 1979 559.9 74.5 139.3 47.6 0.123 8214 317.1 1980 571.7 76.4 m1l 49.1 0.141 849.4 353.2 1981 57L.2 78.0 L50.7 49.9 0.170 850.0 364.3 1982 612.0 94.9 170.8 50.1 0.198 9280 376.9 1983 664.7 1084 181.1 53.4 0.229 L007.7 397.7 1984 724.8 153.6 V6.1 57.2 0.311 1,132.0 437.4 1985 812.6 169.3 237.1 60.3 0.415 1,279.7 532.9 1986 876.5 21L8 270.0 61.2 0.481 1,419.9 594.8 1987 947.1 26M0 288.9 62.5 Q650 L565.2 657.6 1988 987.8 322.0 310.4 65.0 0730 1,686.0 696.6 1989 1L039.4 337.5 349.8 62.9 0690 L790.2 768.9 1990 1,062.2 335.8 345.1 62.7 0.820 L806.6 814.1 1991 1,097.2 342.8 396.5 62.1 L010 L899.6 899.0 Mode Split (%) 1952 82.0 2.0 16.0 - - 100.0 1977 73.4 4.0 16.4 6.2 - 100.0 1979 68.2 9.1 17.0 5.8 0.0 100.0 1980 67.3 9.0 17.9 5.8 0.0 100.0 1981 67.2 9.2 17.7 5.9 0.0 100.0 1982 65.9 10.2 184 5.4 0.0 100.0 1983 66.0 10.8 18O 53 0.0 100.0 1984 64.0 13.6 17.3 5.1 0.0 100.0 1985 63.5 13.2 18.5 4.7 0.0 100.0 1986 61.7 14.9 19.0 4.3 0.0 100.0 1987 60.5 17.0 18.5 4.0 0.0 100.0 1988 58.6 19.1 18.4 3.9 0.0 100.0 1989 5&1 18.9 19.5 3.5 0.0 100.0 1990 588 1&6 19.1 3.5 0.0 100.0 1991 57.8 18O 20.9 3.3 0.1 100.0 G;rowth rate % 0.. 1952 -7 8.4 12.1 9.0 - 15.7 89 18.0 183 - 84 41.7 8.3 7.1 35.8 12.3 10.0 1984 -85 12.1 10.2 20.9 5.4 33.4 13.0 2L8 1985 -86 7.9 25.1 13.9 L5 15.9 1L0 11.6 986 - 87 8. 25.6 7.0 2.1 35.1 10.2 10.6 1987 - 88 4.3 2L0 7.5 4.0 123 7.7 5.9 1988 - 89 5.2 4.8 127 (3.2) (5.5) 6.2 10.4 1989 - 90 2.2 (5) (1.3) (0.3) 18.8 0.9 5.9 1990 - 91 3.3 2.1 14.9 (L0) 23.2 5.1 10.4 1982 - 91 6.7 153 9.8 2.4 19.8 8.3 10.1 * tauiU.omlYs- aahm,t. 9r Sowm SnihUmd Yesbaoct-IW.TmW. Lt-LpSl)  - 46- Table 2 Table 2 CHINA: Passenger Traffic by Mode avil Railway Road Waterway aviation Total Trafric (biion gassenter-km) 1952 20.0 2.3 2.5 0.024 24.8 1977 102.3 44.8 9.8 1.834 158.7 1979 121.6 60.3 11.4 3.499 196.8 1980 138.3 73.0 12.9 3.956 228.2 1981 147.3 83.9 13.8 5.016 250.0 1982 157.5 96.4 14.5 5.950 274.4 1983 177.6 110.5 15.4 5.896 309.4 1984 204.6 133.7 15.4 8.350 362.1 1985 241.6 172.5 17.9 lL7UO 443.7 1986 258.7 198.2 18.2 14.600 489.7 1987 284.3 219.0 19.6 18.205 541.1 1988 326.0 , 252.8 20.4 21.700 620.9 1989 303.7 266.2 18.8 18.679 607.4 1990 261.3 262.0 16.5 23.408 563.2 1991 282.8 287.1 17.7 30.132 617.7 Modal split (%) 1952 80.6 9.3 10.1 0.1 100.0 1977 64.4 28.2 6.2 12 100.0 1979 61.8 30.6 5.8 1.8 100.0 1980 60.6 32.0 5.7 L7 100.0 1981 58.9 33.6 5.5 2.0 100.0 1982 57.4 35.1 5.3 2.2 100.0 1983 57.4 35.7 5.0 L9 100.0 1984 56.5 36.9 4.3 2.3 100.0 1985 54.5 38.9 4.0 2.6 100.0 1986 52.8 40.5 3.7 3.0 100.0 1987 52.5 40.5 3.6 3A 100.0 1988 52.5 40.7 3.3 3.5 100.0 1989 50.0 43.8 3.1 3.1 100.0 1990 46.4 46.5 2.9 4.2 100.0 1991 45.8 46.5 2.9 4.9 100.0 Growth rate % p.a. 1952 - 77 6.7 12.6 5.6 18.9 7.7 1983 - 84 15.2 21.0 0.0 41.6 17.0 1984 - 85 18.1 29.0 16.2 40.1 22.6 1985 - 86 7.1 14.9 L7 24.8 10.4 1986 - 87 9.9 10.5 7.7 24.7 10.5 1987 - 88 14.7 15.4 4.1 19.2 14.7 1988 - 89 (6.8) 5.3 (7.8) (13.9) (2.2 1989 - 90 (14.0) (1.6) (122) 25.3 (7.3 1990 - 91 8.2 9.6 7.3 28.7 9.7 1982 - 91 6.7 12.9 2.2 19.8 9.4 * Excudeac-gangguupai. S StatilYesYmbookctfbns-9Tbl.e2-a.p513 I _SZU?u(~,  - 47- Table 3 Table 3 Transport Investment vs. Economic Output in China (Y billion) Annual GNP Transport transport (current investment Year investment prices) as % of GNP 1966 -70 3FYP 15.0 /_1 935.0 1.6 1971 -75 4FYP 31.8 /_1 1,323.0 2.4 1976 -80 5FYP 30.2 /1 1,758.0 1.7 1981 -85 6FYP 41.9 3,130.5 1.3 1986 -90 7FYP 89.1 6,875.3 13 1980 5.9 447.0 1.3 1981 3.6 4773 0.8 1982 5.2 519.3 1.0 1983 7.2 580.9 12 1984 10.0 696.2 1.4 1985 15.9 856.8 1.9 1986 16.7 969.6 1.7 1987 17.4 1,130.1 1.5 1988 19.0 1,406.8 1.4 1989 18.0 1,599.3 1.1 1990 18.0 1,7695 1.0 1991 30.0 1,9855 1.5 1980 - 1991 average 13 /_1: Investment of FYPs includes Post and Telecommunications (about 4% of total). Source: Statistical Yearbook of China- 1992, Table 5-22, p. 162, table2-12, p. 31 Statistical Yearbook of China- 1991, Table 5-22, p. 160 Other data based on SAR of China Sixth Railway Project, table 1.05, p. 88 Vilu'hI3kwlh1&ftlV6  Table 4 - 48 - Table 4: NUMBER OF MOR STAFF BY ACTIVITY TYPE (thousands at year end) 1980 1985 1990 No. 2 No. 2 No. S Rail transport 1,556 60 1,861 60 1,709 50 Diversified economy /a Truck 0 0 0 0 140 4 Other 0 0 0 0 160 5 Subtotal O 0 O 0 300 9 Industry 356 14 405 13 430 13 Construction 501 19 628 20 571 17 Others 191 7 229 7 371 11 Total 2.615 100 3.124 100 3.381 100 La Includes all "diversified services" such as trucking services (see Annex 2).  Table 5 - 49 - Table 5: ACTUAL REVENUES AND EXPENSES, 1986-91 (in Yuan million) Average annual 1986 1987 1988 1989 1990 1991 growth rate (2) OperatinR Revenue 23.911 26.212 28.062 31.607 41.105 46.704 14 Oneratnax Expenses 15.154 17.290 20.272 24.982 27.807 32.509 16 Of which: Working expenses 12,161 13,976 16,858 21,058 23,451 27,645 18 Depreciation 2,993 3,314 3,414 3,924 4,356 4,864 10 Net Oneratina Income 8.757 8.922 7.790 6.625 13.298 14.195 10 Nonoperating income la 849 999 865 479 200 49 -43 Interest expense * 992 754 1,871 1,940 2,115 21 Business tax - 1,389 1,487 1,675 2,188 2,499 14 Lump-sum tax - - - - 2,000 2,000 Net Income 8.341 7.540 6.414 3.55a 7.370 7.630 -1 Working ratio Lk 542 562 612 682 582 602 Operating ratio /c 672 662 72Z 792 682 702 Unit operating revenue 2.12 2.13 2.14 2.36 3.14 3.43 622 increase (fenlctkm) since 1986 Unit operating expenses 1.34 1.41 1.55 1.86 2.13 2.39 782 increase (fen/ctkm) since 1986 a Revenues from railway manufacturing and construction companies. 1 Operating expenses, less depreciation, expressed as a percentage of operating revenue. ! Operating expenses, expressed as a percentage of operating revenue.  Table 6 - 50 _ Table 6s INTERNATIONAL COMPARISON OF SELECTED RAILWAYS China France India Poland US USSR 1990 1987 1988 1988 1988 1989 Comparison of System Size and Transport Output System Size: Route length (km) 52,700 43,600 76,200 26,600 204,088 147,500 Staff size (thousands) 1,700 222 1,624 360 259 1,814 Locomotive fleet size 13,600 4,266 8,810 4,795 19,755 33,000 Wagon fleet size 364,900 169,400 345,800 137,000 1,239,000 1,371,000 Coach fleet size 27,300 10,310 27,700 6,324 2,232 56,200 Transport Outputs: vreight traffic (billion ton-km) 1,060 50 230 110 1,594 3,852 ssenger traffic (billion pass-km) 261 60 264 56 20 411 -ibined traffic (bil converted ton-Ian) 1,321 110 494 166 1,614 4,262 )senger share of traffic (2) 20 55 53 34 1 10 Comparison of Productivity of Track. Equipment and Labor Track: Freight ton-km/route-km (million) 20.1 1.4 3.0 4.1 7.8 26.1 Passenger-km/route-km (million) 5.0 1.4 3.5 2.1 0.1 2.8 Converted ton-km/route-km (million) 25.1 2.5 6.5 6.2 7.9 28.9 Equipment: Freight ton-km/wagon (million) 2.9 0.3 0.7 0.8 1.3 2.8 Passenger-km/coach (million) 9.6 5.8 9.5 8.8 9.2 7.3 Converted ton-km/locomotive (million) 97.1 25.7 56.1 34.6 81.7 129.2 Steam share of locomotives (Z) 46 0 43 5 0 0 Wagon turnaround time (days) 4.0 5.8 11.6 n.a. 19.0 6.6 Wagon turnaround distance (km) 993 n.a. 1,270 n.a. 2,100 1,610 Labor: Converted ton-km/thousand employees 777 492 304 461 6,232 2,350 Labor cost/revenues (2) 10 90 51 n.a. 41 43 Labor share of operating costs (2) 15 n.a. n.a. n.a. 44 n.a. Comparison of Transport Management Systems Monolith Yes Yes Yes Yes Lines of business la Yes Yes Wholesaler /b Yes Competitive access /c Yes Toll-rail enterprise /d Yes /a Different lines of business are set up among freight railways, Amtrak and commuter railway authorities. lb US freight railways act as "wholesalers" providing train haulage services for "retailers" such as the American President Lines Corporation, which provides door-to-door container transport services for a variety of end-users. /c A US freight railway allows other freight railways to run trains on its system through trackage lights and joint terminal agreements. /d Amtrak allows US freight railways to run their tralns on its track system along the Northeast Corridor for a fee. Sourcest Ministry of Railways; Infrastructure Division, Infrastructure and Urban Development Department, World Bank.  - 51 - ANNEX 1 Page 1 CHINA'S RAILWAY STRATEGY Future Demand for Transport Factors Affecting Future Total Demand for Transport 1. Several factors are acting on transport demand in China, with both positive and negative effects. For freight transport, the positive factors are (i) economic growth and (ii) greater proportion of interprovincial flows and the negative factors are (i) increasing prices, (ii) change in economic structure, and (iii) increasing energy efficiency and rationalization of transport flows. The increasing market orientation of the economy could have both positive and negative effects on transport flows. For passenger trans- port, the positive factors are (i) population growth, (ii) increasing per capita incomes, (iii) less restrictive government travel controls, and (iv) greater travel by entrepreneurs in a market-oriented economy, and the negative factor is increasing prices. 2. Relationship to Economic Growth. The growth in freight transport demand has been roughly proportional to the growth of GDP, and this relation- ship can be expressed in the form of an overall elasticity of demand with respect to GDP. Based on performance over the last 10 years, the net elasti- city of demand with respect to GDP can be estimated at 0.95. Although this should not be perceived as a constant due to the influence of the other fac- tors mentioned above,l/ and the expected changes in regional transport pat- terns described below, it is a key predictor of future growth. Forecast growth in transport demand is related to the forecast growth rates in GDP (or GNP) through the use of the above elasticities, with modifications to reflect changes in other significant factors (Table A1.1). 3. Greater Proportion of Interprovincial Flows. The growth in national transport demand over the past ten years is closely linked to traffic flows between provinces linked to the areas of rapid economic growth along the east coast (especially in the provinces surrounding Shanghai) and in the south of China (e.g., Guangdong Province). The fastest growth in demand for transport has been either within these rapidly growing regions or along the routes con- 1/ An appropriate demand model would relate demand to a combination of fac- tors including GDP (or per capita income), population, relative price in real terms, and administrative or structural factors. Estimates of pas- senger trip demand elasticities using a model with two independent vari- ables (price and per capita income) in the recent Yangtze Economic Zone Transport Study (PPK Consultants, Travers Morgan and the YEZTS Comprehen- sive Transport Study Group, April 1992) gave elasticity values of 1.19 to 1.47 with respect to per capita income, equivalent to an elasticity of about 1.3 to 1.6 for total passenger trips, if per capita income grows at the same rate as GDP.  ANNEX 1 _ 52 - Page 2 necting these areas with the other regions of the country and the outside world.2/ The flows of coal from the provinces in the Coal Base (Shanxi, Shaanxi and Inner Mongolia) represent the largest of these flows (Map M.1) and have been growing the fastest of all longer-distance freight traffic flows (6.5 percent per year). Linkages for input materials and industrial products from the northeast and southwest have also increased significantly, along with manufactured goods shipped from the high growth areas to other regions. Because these freight flows travel relatively long distances, the demands on the capacity of the interprovincial transport system have been increasing significantly. In the future, this increasing proportion of medium- to long- distance transport demand, which is based on the location of resources and existing industrial plants relative to growth areas, will act to increase the transport intensity of economic growth in China and counteract the measures for reducing transport intensity discussed above (Map M.2). 4. Transport Pricos. The effects of increased prices on traffic levels have been much less than would be expected under conditions without transport system supply shortages and are expected to be even lower in the future. Railway freight traffic decreased by 6 percent compared with forecasts, fol- lowing rail tariff increases totaling 32 percent between 1989 and 1991. This implies a price elasticity of -0.2 for traffic, which is lower than the price elasticity for total demand, due to the effect of the large amount of unsatis- fied freight demand at current prices. Because the prices of all transport modes have been increased to levels roughly similar to their long-run marginal costS, future increases in transport prices are expected to be virtually zero in real terms, reflecting only increases in costs due to inflation and there- fore no decrease in demand from this factor (with the exception of the rail price scenario for funding higher investments, which is described below). 5. Change in Economic Structure. Chiua is noted for its high intensity of freight traffic--measured as total freight traffic handled (in both tons and ton-km) per US dollar of economic output. To produce one dollar of GNP,31 China transports about five times as many tons of freight and handles twice as many ton-km of traffic than India or Brazil (Figure A1.1), which have similar distributions of resources and activity centers. China's high freight intensity can be attributed to several factors--the dominance of coal as a source of energy (about 75 percent of all of China's energy consumption), the relatively small service sector and large heavy industry sector, many low- value products, low energy efficiency, and underpricing of energy, raw materi- als, and transport services. High freight intensity also is encouraged by decisions not to preprocess raw materials before transportation and the lack of attention to logistics costs in the central planning of industry location and distribution systems, all resulting in uneconomic transport of materials. 2/ The growth in total freight demand has been fastest in intraprovincial flows in the rapidly growing regions. 3/ GDP figures were adjusted by purchasing power to determine the most com- parable figures among countries with different foreign exchange rate policies. The adjustment is based on ratios of ICP adjusted GDP to offi- cial CDP calculated by World Bank staff for 1985 and applied to 1988 CUP.  - 53 - Page 3 On the other hand, when measured in relation to population size instead of economic output, China's freight traffic intensity per capita is one of the lowest in the world. This will change, however, with increasing per capita GDP. Figura Al.1t INTERNATIONAL CONPARISON OF FREIGHT TRAFFIC INTENSITY (per US$ of GNP--1987) ( *Tons/1,000 ICP G3DP [JTon-km/ICP GDP ) 6. Enerav Efficiency and Rationalization of TransDort Flows. The transport intensity of the Ch4nese economy is already declining due to shifts toward the market economy, in which shippers give more attention to minimizing logistics costs and to the Use of modern energy-efficient technologies in power generation and industrial production. This trend will undoubtedly con- tinue and will be reinforced by the rationalization of coal and 'other bulk counodity dlstribution (Annex 6). If they are implemented, these rationaliza- tion policies, combined with incresaed prices for coal and grain, are expected to reduce total ton-ke hauled for these coemodities by about S percent over forecasts by the year 2000, based on the present policies Market forces willl also lead to more rational use of the transport system in the future through more cost-effective industry location decisions, but these effects will t-ke a long time to be realized because current industrial and resource locations will determine the majority of flows of goods into the twenty-first century. The effect of these forces will be to further lower the transport intensity of the power sector and some industrial sectors to a small extent. Most of the lowering effect will be more than compensated by increases in interregional flows described above. ..............................................................~~~ ~~~ ~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  ANNEX 1 - 54 - Page 4 7. Passenger Demand. The factors influencing passenger demand are not as well known as those affecting freight in China and need more research by NOR. Different factors have different effects on business, commuter, and per- sonal travel and on tourism. However, the majority of trips for business and personal travel are related to population growth (forecast at 1 percent per year), per capita income elasticity (estimated at 1.5 and increasing due to greater mobility in the future), and price elasticity (estimated at -0.2 for traffic in shortage conditions and -1.2 for total demand).4/ The values of these factors indicate a growth rate for passenger demand that is signifi- cantly faster than for freight (Table A1.1). Alternative Scenarios of TransRort Demand 8. There are four basic scenarios that are useful for transport demand forecasts: low or high GNP growth without rationalization of transport dis- tribution or rail transport price increases (called Base Case) and low or high GNP growth with rationalization of commodity distribution and rail transport price increases. The implications of these different assumptions are given in Table AllI and discussed individually below. 9. GNP Growth Rates. The GNP growth rate targets for the Chinese econ- omy used in past planning for the Eighth and Ninth Five-Year Plans have been low compared with past performance, averaging less than 6 percent per year. Although this is one possible scenario, it appears increasingly less likely, because the current economic growth rate has accelerated past this level. Given the present thrust of economic reforms in China, it appears realistic to expect the economy to grow at least as fast as 8 percent per year in the 1990s, and possibly as fast as 10 percent per year. A growth rate of 8 to 9 percent per year has now been adopted by SPC as a revised basis for fore- casting to the year 2000 and this is considered more realistic for China's future by World Bank economic analysts. Therefore, an 8 percent GNP growth rate was adopted by the RIS team for investment analysis, as reported in Annex 6. Both the 6 percent growth and 8 percent growth scenarios are considered in Table A1.1. 10. Rationalization of Bulk Commodity Distributions. The Base Case freight demand forecast used by the RIS team includes an assumption of contin- 41 There was an increase in railway passenger fares of 112 percent in 1989 that resulted in only a 23 percent decrease in total rail passenger-km from levels forecast at constant price. These national-level figures reflect a combination of two separate situations: saturated markets where there is substantial unsatisfied demand with a shortage of ser- vices, and unsaturated markets where there are adequate services to satisfy demand. In the first case, traffic will change very little in response to price, while in the second case traffic decreases signifi- cantly in response to price increases. Saturated markets with low price elasticity due to large amounts of unsatisfied passenger trip demand are found for services between the larger cities in China. Total demand for trips was reduced at higher price levels, but it still exceeded the sup- ply at the new higher prices for many intercity rail services.  ANNEX 1 - 55 - Page 5 ued reform and incremental rationalization of transport flows. However, a more drastic rationalization of bulk commodity distribution was assumed for the alternative transport demand scenario, which reduces demand for coal, grain, cement and fertilizer transport by 5 percent compared with the base case (Table A1.1). This level of potential reduction has been suggested by recent studies of coal and grain distribution (Coal Transport Study and Grain Distribution Study).5/ 11. Rail Transport Price Increases. The Base Case assumes that baseu and tariffs increase with inflation (i.e., no real increase in economic terms). However, the alternative scenario assumes an increase in railway prices in real terms (38 percent for freight tariffs and 10 percent for pas- senger fares) that would be enough to finance the accelerated railway investment program needed to meet demand under the high growth scenario (Table A1.1). A lesser tariff increase for the low growth scenario was analyzed. 12. Forecasts using the above assumptions show that demand for rail transport can be expected to grow by 4 to 5.7 percent per year for freight and 8.4 to 11.3 percent per year for passenger trips in the Base Case, and that demand growth would decline to 3.4 to 4.4 percent for freight and 8.3 to 10.6 percent for passenger trips with the alternative scenario (Table A1.1). The consequences for railway investments are discussed in Annex 6. Modal Choice Issues 13. The expansion and modernization of the different transport modes has changed the relationships among them, especially in the sense of encouraging more specialization of each mode in its area of relative advantage: (i) Road transport has expanded with more modern truck and bus vehicles that provide fast local and medium-distance transport (up to 800 km or more where there are good highways). (ii) Waterway transport has expanded to include container transport and to utilize modern equipment in the coastal and river delta areas where export/import traffic make it is highly competitive, but it has remained more traditional in other inland areas. (iii) Air transport has grown rapidly with the acquisition of new modern aircraft and the introduction of competi- tion through regional airlines, taking a larger share of the medium- to long- distance passenger traffic. Finally, (iv) new oil and coal slurry pipelines are planned to be built in a few locations with a relative cost advantage, but they have a very limited capacity compared with rail transport.6/ 14. The more gradual expansion and modernization of tl-e railway mode have kept its costs low in comparison with the other modes, and have main- tained its competitive advantage despite the shortage of rail transport capa- 5/ It is expected that market-oriented activities will further reduce the transport requirements in some cases but also increase the transport requirement in cases where buyers diversify their sources to get the best prices. 6/ See the White Cover Report on the China Coal Transport Study, EA2TP, June 5, 1992.  ANNEX 1 - 56 - Page 6 city. The rail advantage is particularly evident for medium- to long-distance freight and passenger transport and for commuter trains. But the shortage of freight cars and line capacity has placed it at some disadvantage in compari- son with other modes where they are in a position to take up the extra demand left over from the railway and can provide more reliability of service to shippers who need guaranteed delivery times. Nevertheless, the railway system in China will continue to be the dominant mode for medium- to long-distance transport, due to its continuing price advantage, even with full cost pricing, including provision for investment in new capacity.7/ If the railway can keep up with demand and provide a market-oriented service with its relative price advantage, it is assured a growing demand for its services, according to the detailed analysis performed by the RIS team (Annex 6). 15. The potential rail demand shown in Table Al.1 has taken into account the increasing role of road transport in the future, which will attract cer- tain types of cargo and passengers away from rail, despite relatively low rail prices. The proportion of potential rail traffic that will be carried by the railway in the future will be determined by the future availability of rail services, which is determined by the proposed expansion of railway capacity (Annex 6). 7/ This conclusion is supported by all four recent national and regional studies undertaken in China: the Guangdong Province Comprehensive Trans- Dort Study, the Yanatze Economic Zone Trans2ort Study, the Chiga Rajll1 Investment Study and the China Coal Transgprt Studv.  - 57 - ANNXI Page 7 Table A1.1: DEMAND FOR TRANSPORT BY SCENARIO (billion pkm or tik) $62 N rwh 8 GNP aMrowthl pkm tl pln tlhmir 1989 Base Case ^g Total transport demand All modes 710 2,040 710 2,040 Rail demand lb 400 1,510 400 1,510 Total medium-long distance lc All modes 380 1,420 380 1,420 Rail demand /b 340 1,300 340 1,300 2000 Base Case Id total transport dand All modes 2,030 3,750 2,720 4,570 Growth rate (S) 10.0 5.7 13.0 7.6 Rail demand Lb 970 2,320 1,300 2,790 Growth rate (2) 8.4 4.0 11.3 5.7 Total medium-long distance /e All modes 1,080 2,610 1,460 3,180 Growth rate (2) 10.0 5.7 13.0 7.6 Rail demand lb 870 2,090 1,170 2,510 Growth rate (2) 8.9 4.4 11.9 6.2 Base Case with Rationalization of Bulk M Mt Distorlb..ution / eotal tranSpOrt emand All modes 2,030 3,660 2,720 4,460 Growth rate (Z) 10.0 5.5 13.0 7.4 Rail demand lb 970 2,320 1,300 2,790 Growth rate (2) 8.4 4.0 11.3 5.7 Total medium-long distance /c All modes 1,080 2,550 1,460 3,100 Growth rate (Z) 10.0 5.5 13.0 7.4 Rail demand /b 870 2,030 1,170 2,430 Growth rate (Z) 8.9 4.1 11.9 5.9 2000 Base Case with Rationalization of Bulk Im dl DistribUtion and Rai iransoort e ncreanes l1f Total transport demand All modes 1,940 3,470 2,460 4,000 Growth rate (Z) 9.6 4.9 12.0 6.3 Rail demand lb 910 2,170 1,140 2,420 Growth rate (2) 7.8 3.4 10.0 4.4 Total medium-long distance le All modes 1,030 2,420 1,320 2,780 Growth rate (2) 9.5 5.0 12.0 6.3 Rail demand lb 820 1,900 1,030 2,110 Growth rate (2) 8.3 3.5 10.6 4.5 J& Assuming 16 percent unsatisfied demand for all passenger trips and 20 percent for long- distance passenger trips, 14 percent for all freight, and 16 percent for long-distance freight shipments compared with present traffic levels. lk idedium- to long-range transport demand for rail, waterway, and air, for which the rail is the least-cost carrier (average haul distance was 900 km in 1989 and 1,000 km in 2000 for freight when unsatisfied demand is included). Total rail demand is assumed to be 116 percent of medium- to long-distance demand In tkic and pkm. See Annex 6 for more details. le Demand modeled in greater detail by the RIS team, equivalent to interzonal freight demand in RIS analysis. l4 Based on Base Case RIS analysis, which assumed no major additional price increases over pres- ent plans and no significant change in present distribution patterns. See Annex 6 e Assuming rationalization of coal, cement, fertilizer, and grain distribution, resulting in S percent less tlk in each case. /