Mozambique - Beira Transport Corridor Project

Document of The World Bank FOR OFFICIAL USE ONLY G6Z 12 O-oZ C Report No. 77094MOZ STAFF APPRAISAL REPORT MOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT JULY 18, 1989 Southern Africa Department Infrastructure Operations Division This document has a restricted disribution 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 Currency Unit - Metical (plural Meticais) USS1 * 715 Meticais (Mts) 05/01189 Mt 100 - US$0.14 MEASURES AND EQUIVALENTS 1 Meter (m) - 3.28 Feet 1 Square Meter (m.sq) 10.76 Square Feet 1 Hectare 2.47 Acres 1 Kilometer - 0.62 Miles ABBREVIATIONS AND ACRONYMS AfDB = African Development Bank BCA - Beira Corridor Authority BTC - Beira Transport Corridor CFM(C) - Caminhos de Ferro de Mocambique (Centro) CFM(N) ' Caminhos de Ferro de Mozambique (Norte) CFM(S) ' Caminhos de Ferro de Mocambique (Sul) CFM,EE Caminhos de Ferro de Mocambique, Empresa Estatal CPMZ - Companhia Pipeline Mozambique-Zimbabwe DANIDA = Danish International Development Authority DNPCF 8 Direccao Nacional de Portos e Caminhos de Ferro EDF - European Development Fund EEC European Economic Community GDR 3 German Democratic Republic MBA - Machipanda Beira Authority (Zimbabwe) MIS = Management Information System MT ' Metric Ton (1,000 kg) MTC = Ministry of Transport and Communications NORAD 3 Norwegian Directorate for Development Aid NRZ = National Railways of Zimbabwe RENFE = Red Nacional de los Ferrocarriles Espanoles RITES 2 Rail India Technical and Economic Services RSA Republic of South Africa SADCC = Southern Africa Development Coordination Conferenc; SATCC = Southern Africa Transport and Communications Commission SETEP 3 Secretariate of State for Technical and Vocational Training UEM = University Eduardo Mondlane U.K.ODA = United Kingdom Overseas Development Administration USAID = United States Agency for International Development WFP World Food Program  uaR rriLiAL USE ONLY MOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT TABLE OF CONTENTS Page No. CREDIT AND PROJECT SUMMARY ................. ... ............ . i - ii I. SECTORAL AND SOUTHERN AFRICA REGIONAL CONTEXT .......... 1 A. Introduction ......................... ... .. .... . 1 B. Geographic Setting ............... ............ 1 C. Role of Transport Corridors in Mozambique's Economy 2 D. Southern Africa Regional Coe t e xt 3 E. Government Strategyr........... *.......... 6 ... 5 F. IDA's Role...... 8 II. THE BEIRA TRANSPORT CORRIDOR............... 9 A. Physical and Institutional Framework . .. 9 B. Beira Portort.... ..... ..... . .... ... . 10 C. Beira-Machipanda Railwaya....... . ..... 13 D. Beira Machipanda R oad . ................ 15 E. Pipeline Beira-Mutare ..... . .......... 15 F. Tranna....................... 15 III. THE PROJECT R.. . ... ........... .... 16 A. Project Objectives ..... .... .16 B. The Beira Corridor Authority... 17 C. The Beira Corridor Program... 17 D. Overall Program Cost and Financing. 19 E. IDA Assisted Project Components.......... 21 F. Rationale for IDA Components... 27 G. Project Cost Estimates ..... 27 H. Financing ............. 28 I. Implementationn... .... .... 29 J. Procurement ....29 K. Disbursements ....30 L. Project Monitoring ... 31 M. Budget, Accounts, Audit and Evaluation .... 31 N. Enviroment ....32 This report is based on the findings of an appraisal mission to Mozambique in October November, 1988 consisting of World Bank staff Messrs D. Graham (Mission Leader), J. Rebelo, K. Kohriko and S. Nayak and consultants Messrs. Bonnin, Le Bussy, Gusmao, de Langen, Morra and Spit. Mrs. L. Iacono was responsible for report processing. This document has a restricted distribution and may be used by recipients only in the perfomrance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table of Contents (Cont'd) IV. ECONOMIC EVALUATION ...................... .. 32 As Introduction* ... . ........................*s..*......... 32 B. Traffic Forecast................................. 32 C. Economic Evaluation ..... ....... .. ....... . . .* 38 V. FINANCIAL EVALUATION ......................... so ...... 41 A. CFMC F....................... 41 B. DNPCF ....... #o , .... ,.. o ..... o 44 VI. AGREEMENTS, CONDITIONS AND RECOMMENDATIONS ............ 44 ANNEXES 1 Status of Overall Project Components 2 Training 3 Technical Assistance 4 Locomotive Requirements 5 Performance Targets 6 Economic Analysis 7 Financial Analysis 8 Organization Charts 9 Criteria for Investment in the Beira Corridor 10 Terms of Reference for the Project Coordinator and Procurement/Disbursement Officer MAPS IBRD No. 20101 Beira Corridor IBRD No. 20102 Beira Corridor IBRD No. 21356 Beira Port  MOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT CREDIT AND PROJECT SUMMARY Borrowers People's Republic of Mozambique Beneficiaries: Ministry of Transport and Communications (MTC), National Directorate of Ports and Railways (DNPCF), Beira Corridor Authority (BCA), Central Railway System (CFM(C)), Southern Railway System (CFM(S)), Northern Railway System (CFM(N) and Secretariate of State for Technical and Vocational Training (SETEP). Amounts SDR 31 million (US$ 40.0 million equivalent). Termst Standard IDA Terms with 40 years maturity. Onlendini Terms: The credit will be on-lent to the Mozambique Ports and Railways Administration (DNPCF) for a period of 30 years including five years of grace, at an interest rate of 7.65%. DNPCF will bear the foreign exchange risk. Project's Descriptions The main objective of the Beira Transport Corridor Program is the rehabilitation and upgrading of the Beira Corridor to restore its cost efficient transit functions. The IDA assisted project would contribute to this objective through the provision of technical assistance, manpower development and training and motive power. IDA components will provide for (i) technical assistance to run port and railway operations during the 1989-1995 period; (ii) manpower development and training for all levels of staff in the port and railways; (iii) rehabilitation and acquisition of motive power for main line and shunting operations; and (iv) implementation of a railway/port cost accounting system and a management information system. Benefits and Risks: The main benefits are the reduction of transport costs and increased efficiency which will allow Beira to compete with Durban (RSA), providing Zimbabwe and Malawi with cheaper alternatives to the sea and Mozambique with foreign exchange earnings. The main risks are (i) delays in project implementation and consequent inability to provide acceptable levels of service may affect financial and economic viability; and (ii) the security issue: in spite of the presence of armed forces guarding the corridor, possible disruptions may occur.  Estimated Costs: Local Foreign Total -US$ Million------------- General Projects 3.7 22.1 25.8 Port Projects 6.4 187.4 193.8 Rail Projects 3.4 94.6 98.0 Road Projects 5.0 22.3 27.3 Total 18.5 326.4 344.9 Financing Plan: Local Foreign Total - --------- US$ Million------------ Mozambique Government 4.3 - 4.3 Proposed IDA Credit 40.0 40.0 Parallel financing _ 29.5 29.5 Sub-total Project 4.3 69.5 73.8 Other Parallel financing of the Program 14.2 256.9 271.1 Total 18.5 326.4 344.9 Estimated Disbursements FY90 FY91 FY92 FY93 FY94 FY95 FY96 Annual 4.0 12.0 10.0 6.0 5.0 2.0 1.0 Cumulative 4.0 16.0 26.0 32.0 37.0 39.0 40.0 Economic Rate of Return: 14% for the whole Beira Corridor Program. 34% for the IDA assisted Project AF6IN July 1989  MOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT Documents Available on the Proiect File Report and Documents on Beira Transport Corridor 1. Beira Port Transport System, 10-Year Development Plan, SATCC (Southern Africa Transport and Communication Commission, Maputo, January 1986). 2. Comments and suggestions on: Beira Port Transport System, 10-Year Development Plan, World Bank, March 7, 1986. 3. Port of Beira, Pre-Feasibility Study of Channel Improvements, Volumes 1, 2 and 3, Bertlin and Partners, June 1978. 4. Beira Port Study, Phase 3, Master Plan Study, Volumes 2, 3, and 4, NEDECO, 1982. 5. Beira Port Project, Tender Documents, Reconstruction of Quays 2-5, Volume 1-6, NEDECO, March 1985. 6. Zimbabwe National Transport Study, SWECO, 1985. 7. Beira-Machipanda Road, Feasibility Study Report, Gavan McDonel and Company with Snowy Mountains Engineering Corporation and COTOP, Mozambique, April 1985. 8. Rehabilitation Study Beira-Machipanda Line and Maputo- Chicualalcuala Line, Mott, Hay & Anderson International Ltd., June 1982/April 1983. 9. Study of Railway Rolling Stock in the SADCC countries, Kampsax, Swedrail, June 1984. 10. Mozambique, An Introductory Economic Survey, Annex II, Transport Sector, World Bank, June 6, 1985. 11. Southern Africa Regional Transportation Strategy Paper, Louis Berger International Inc., for USAID, March 15, 1985. 12. Southern Africa Regional Transportation Strategy Evaluation, Data Base Update, Louis Berger International, Inc., for USAID, May 1986.  MOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT 13. Port Staff Training Program Study, SATCC, Norconsult, September 1986. 14. Reconstruction Quays 2-5, Beira, Economic and Financial Feasibility, Netherlands Economic Institute, October 1986. 15. SATCC Project #R-OP-1, Motive Power, Rolling Stock an; Operations Plan for CFM Center, Draft Final Report, 2 volumes, SLI Consultants, October 1986. 16. Study on Railway Training Programs, DE Consult, October 1986. 17. Study of Locomotives for DNPCF, Mozambique. Rites, September 1988. 18. Organization and Manpower Development Plan, SOFRERAIL, Norma and BCEOM, February 1989. 19. Needs Study for a Management Information System, SOFRERAIL, Norma and BCEOM, February 1989. 20. Terms of Reference for the Senior Training Instructors, Guiseppe Morra, December 1988. 21. Beira Transporc Corridor Program, World Bank, Report number 6698- MOZ, April 24, 1987. 22. Beira Port Transport System, Quarterly Progress Reports, 31 December, 1987 to 31 December, 1988. AF6IN April 1989  HOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT I. SECTORAL AND SOUTHERN AFRICA REGIONAL CONTEXT A. Introduction 1.01 The Economic Rehabilitation Program (ERP) Implemented by the Government of Mozambique (GON) since January, 1987 has initiated a major shift in economic policy and, in the two years since its introduction, has achieved considerable progress in the adjustment process and helped to revitalize the economy. Thus, after years of economic decline, GDP is estimated to have grown in real terms by 4.0% in 1987 and 4.1X in 1988 and a growth rate of 5.0% is forecast for 1989. In parallel with the ERP, there has been increased emphasis on rehabilitation of essential economic infrastructure, particularly in the transport and urban sectors, as a means of restoring economic growth and generating foreign exchange earnings. The Beira Transport Corridor Program, which is the subject of this report, is financed by 18 bilateral and multilateral donors and is the largest and most advanced of Mozambique's transport corridor rehabilitation projects. B. Geographic Setting 1.02. Mozambique has three international ports: Maputo in the south, Beira in the center and Nacala in the north. Maputo has direct rail connections with the Republic of South Africa (RSA), Swaziland and Zimbabwe; Beira with Zimbabwe &nd Malawi; ane Nacala with Malawi (see IBRD Map 20101 and paras 1.11-1.18). Indirectly, there are links with other landlocked countries: Zambia and Botawana. 1.03. As the map and Table 1.1 show, Mozambique's geography makes it eminently suitable as a transit country for the international trade of its neighbors. The rail and road distances to Mozambican po_-ts are considerably shorter than those to alternative ports in RSA.  Table 1.1 - Southern Africa: Rail and Road Distances to Malor Ports (Kilometers) Harare Lusaka Blantyre Francistown Maseru Lilonawe Gaboroz Capetown 2922 3135 3462 1980 1965 3806 1545 Port Elizabeth 2422 3234 2962 2000 830 3306 1565 Durban 2027 2839 2567 1605 666 2911 1170 East London 2362 2959 3174 1940 830 3246 1505 Richards Bay 1828 2640 2368 1855 844 2712 1420 Maputo 1193 2005 2545 1240 1175 2889 1675 Beira 605 2073 649 1308 - 993 1743 Nacala 1354 1888 814 2057 - 1158 - Dar-es-Salaam 3513 2045 1800 3200 - 1635 3635 Lobito 3758 2290 3374 3445 - 3030 3880 1) Rail distances from Blantyre & Lilongwe to RSA ports are via road between Blantyre and Harare 2) Distances from Zimbabwe to Richards Bay and Durban assume use of rail route through Swaziland. 1.04 These differences in distances, in turn, are reflected in differences in inland transport costs, as illustrated below: Inland Transport Costs to/from Harare, 1988 (Includes Railway and Port Charges) Via Beira Via Durban Difference (Z$) (ZM) (%) Per container - import 1413 2191 + 55 - export 1056 1989 + 88 General cargo, per ton 53 140 +164 1.05 As a result of the comparative advantage in transit transport costs, in the past, Malawi, Zimbabwe, Swaziland and to a lesser extent Zambia and RSA (Transvaal), relied on Mozambique's port and railway system to handle their overseas trade flows. C. Role of Transport Corridors in Mozambique's Economy 1.06 The transport sector is an important part of Mozambique's economy. It accounts for over 10% of the national social product and is potentially of vital importance as an earner of foreign exchange and a generator of employment. Prior to independence in 1975, the balance of payments had  - 3 - consistently shown a considerable deficit on the merchandise account which was mainly financed by a surplus on the services account. Thus in 1973, two years before independence, the deficit on the merchandise account of the balance of payments was US$115 million while the corresponding surplus on the services account was US$122 million. The transport sector accounted for US$109 million or 90% of this surplus. On the other hand, for a more recent year, 1983, the situation was entirely different: the merchandise deficit was US$355 million while the services account (including transport) of the balance of payments showed a deficit of US$16 million. This primarily reflects the steep decline of international transport services (rail and port) provided by Mozambique over the past decade; a decline of some 70% to 80%. The transport sector can and should play a much more important role as a foreign exchange earner than it is playing at present. D. Southern Africa Regional Context 1.07 Mozambique's neighbors too have a vital interest in Mozambique's transport system. They are very much interested in having the traditionally cheapest transport options to the sea restored. Moreover, this would simultaneously help to meet one of the primary objectives of the Southern Africa Development Coordination Conference (SADCC), the grouping of nine Southern African countries (Angola, Botswana, Lesotho, Malawi, Mozambique, Swaziland, Tanzania, Zambia and Zimbabwe): reduction of the current dependency on RSA for their international transport needs. 1.08 Table 1.2 summarizes trade flows for each SADCC country through SADCC and RSA ports for 1987. As could be expected, all of Angola's import and export traffic moved via Angolan ports, while Mozambique and Tanzania also relied primarily on their own ports. Botswana and Lesotho, on the other hand, used South African ports exclusively, while Swaziland shipped almost half of its overseas traffic via South Africa. The BLS countries (Botswana, Lesotho, Swaziland) all participate in the Southern Africa Customs Union, which facilitates their use of South African ports and surface transport. From a distance perspective Botswana could, if all lines were in operation, make use either of Maputo or Durban ports, which are 1,030 and 1,170 km from Gaborone, respectively. Lesotho bas no choice by virtue of its location but to use South African Transport System (SATS). However, for Swaziland, there is a clear distance advantage of using Maputo port over Durban, as the former is 275 km from Matsapa and the latter 540. Malawi shipped approximately three-fourths of its overseas goods via Durban but also made use of Beira, Dar es Salaam, and minimal use of Nacala port. For Malawi, the distance penalty of being forced to ship via Durban is significant, as most routes to Durban are in the order of 3,500 km in length versus 1,770 to Dar es Salaam, 640 to Beira, and 815 to Nacala. Zambia shipped most of its goods via Dar es Salaam. Zimbabwe made significant use of both Beira (mainly for petroleum products) and Durban ports, with a small amount of traffic via Maputo. It is clear that significant savings could be made, particularly by Malawi, Zimbabwe and Swaziland, through greater use of SADCC ports, particularly those in Mozambique.  DISTRIBUTMON OF OVERSEAS TRADE TOUIH SADCC PORTS. 1987 (1M6 Tonnes) PORT DAR ES MAPUTO/ TOTAL RSA COUNTRY SALAAM NACALA BEIRA M UATOLO LOB TO 2V SADCC PORTS y ANCOLA 273.6 p/ 273.8 BOTSWANA 133 LESOTHO 61 MALAWI 21.0 1.5 67.7 90.2 319 MOZAMBIQUE 184.5 W 882.1 a 841.4 31 1308.0 SWAZILAND 686.0 585.0 464 TANZANIA 1884.0 1/ 1884.0 ZAMBIA 1269.0 J 1269.0 16 ZIMBABWE 1203.5 396.6 1860.1 926 SOUTH AFRICA 479.7 479.7 OTHER 427.6 11 89.3 6.4 516.7 TOTAL: 8601.0 186.0 1692.6 2830.1 278.6 8666.3 = 5= == I/ Includes Petroleum Products V 1988 Figures y Excludes Cabotago i/ Preliminary Base Year Estimates for Revised SATCC Forecast. Source: SATCC 0  - 5- E. Government Strateey 1.09 As shown in Table 1.3, since independence in 1975, the transit function of Mozambique has steadily declined, until recently, from a peak of 18.26 million tons handled by the system in 1973, of which 95Z was transit traffic, to 4.25 million tons by 1986, of which 3.6 million were transit. The improvement observed in 1987 is the result of initial rehabilitation activities (see paras 2.03 - 2.16 below). 1.10 The principal reasons for the dramatic decline in Mozambique's transit function are: (i) the departure of most Portuguese personnel at the time of independence and the lack of an adequate training program, leaving the transport system with a severe shortage of competent personnel to manage and operate the system; (ii) the steady deterioration of port and rail physical facilities since the early 1970s because of lack of maintenance and rehabilitation; (iii) the loss of much of the RSA (Transvaal) and Swaziland traffic since independence; (iv) the closure by Mozambique of the border with Rhodesia 1 from 1976 to 1980, and; (v) the activities of armed bandits, since 1980, in Mozambique. 1.11 As a result of the bandit activities, the rail connections between Malawi and the ports of Nacala and Beira as well as the rail link between Zimbabwe and Maputo are closed at present. The rail and pipeline connections between Zimbabwe and Beira are guarded by Mozambican and Zimbabwean troops and are open to traffic, although rail and port capacities are limited. 1.12 Given the importance of the transport corridors to Mozambique's economy, their rehabilitation has become a key element in the Government's strategy of structural adjustment, which seeks to revitalise the economy and reverse the economic decline of previous years. Thus, in parallel with the ERP launched in 1987, is a major program of investment in the rehabilitation of urban and transport infrastructure facilities as a means of rapidly restoring economic growth and generating foreign exchange earnings. Implementation of the strategy, with the help of bilateral and multilateral donors, has been quite successful. The following paragraphs describe the current status of the three corridors. l/Zimbabwe since independence in 1980. Throughout this report, Rhodesia is used in reference to the period before independence.  - 6 - Table 1.3 MOZAMBIQUE Beira Transport Corridor Project Rail Transport and Port Throughput - rreight, 1973 - 1987 (mil lion) A. Rail Transport (million) South System Central System North System Total 3 1 Yesr (Maputo) (Beira) (Nacala) Systems # Ton Ton Km Ton Ton Km *Ton Ton Km Ton Ton Km 1973 14.13 1828 6.93 1634 .53 202 20.70 3644 (85%) (93%) 1975 9.94 1324 2.93 916 .47 1? 13.41 2414 (84%) (79%) 1979 6.86 542 1.52 459 .48 214 8.88 1220 (77%) (68) 1983 3.47 4S8 .48 131 .39 176 4.37 787 (77%) (77%) 1984 3.17 376 .30 70 .21 90 3.68 536 (67%) (71%) 1985 2.62 188 .31 90 .06 10 2.90 2.90 (84%) (83X) 1988 2.64 194.4 .368 98.7 .03 7.7 2.96 301 (78%) (87X) 1987 2.06 159.4 .482 136.3 .06 11.9 2.60 307 (72%) (81.1%) S. Port Throughput (million ton) Maputo B-lra Nacala Total ! 1973 14.17 2.98 .79 18.26 (95%) 1976 10.92 3.01 .76 14.88 (93%) 1979 8.22 1.72 .78 10.94 (89% 1983 4.08 1.61 .63 6.56 (80%) 1984 3.07 1.39 .60 6.18 (78%) 1986 2.76 1.39 .20 4.54 (87%) 1986 2.48 1.33 .22 4.25 (85S) 1987 2.67 1.96 .30 6.08 (83%) 1/ Total also includes the minor volumes of freight moved by the Zambezia Railways; figures in parenthesis indicate percentage of International traffic. 2/ Total includes also the throughput of some minor ports; figures in parenthesis indicate percentage of international transit traffic. Source: Ministry of Transport and Communications, 1987.  -7- (a) The Nacala Corridor 1.13 The Nacala system, operated by CFH(Norte), consists of the port of Nacala and a single line track between Nacala and Malawi, through northern Mozambique. The system used to serve Malawi's foreign trade and Mozambican exports of tea, cotton and cashews from the provinces of Nampula and Niassa. For a period, between 1973 and 1976 after the Zambian-Rhodesian border closure, the line also carried some Zambian cargo. The 814 km line (from Blantyre) was closed in 1984 because of bandit activities, and despite current multi-donor assistance to rebuild the track and the port, which commenced in 1984, the line remains closed to Malawi's traffic. The port is being improved with donor support from Finlands container facilities are being provided and staff trained, while the rail link with Malawi is being improved with Canadian, Italian, French, Portuguese and EEC assistance. Subject to security considerations, the rehabilitation program is expected to be completed by 1991/2. (b) Maputo Corridor 1.14 The Maputo port transport system, operated by CFM(Sul), is considered SADCC's premier alternative to South African routes for overseas trade. It serves southern Mozambique, Swaziland and southern and western parts of Zimbabwe, as well as the north-eastern Transvaal (RSA). There are three rail connections with Maputo port. The link between Zimbabwe and Maputo that follows the river border (known as the Limpopo line) is 534 kms long. Althou&h the line is longer than the Beira Corridor, it has a higher capacity both up and down since it traverses flat, open terrain. Closed to commercial traffic since 1984, this line is currently being rehabilitated with assistance from the United Kingdom (U.K.), the EFC, Portugal and Canada with completion expected in 1990/1. 1.15 The Ressano Garcia line, 88 kms long in Mozambique, links Maputo with RSA. This route has been completely rehabilitated and currently carries over 1.5 million tons of cargo per year. Principal commodities include coal, sugar, citrus and steel from South Africa and Zimbabwe. In 1986, approximately 500,000 tons of Zimbabwe's external trade passed through Maputo. Zimbabwean traffic destined for Maputo currently passes via South Africa, transversing the border at Beitbridge and connecting with the Mozambican system at Ressano Garcia. The third rail link, the Goba line, connects Maputo port with Swaziland (68 km in Mozambique). The line is in poor condition and is frequently closed because of security problems. Nonetheless, the line carried approximately 500,000 tons of cargo in 1986. The Government of Italy has made a commitment to reha&ilitate the entire length of the line inside the Mozambique border. 1.16 Maputo port is currently handling approximately 2.5 million tons of cargo from landlocked countries per year. The theoretical design capacity of the port is 14.0 million tons per year and it is capable of handling 7.0 million tons in its current condition. (c) Beira Corridor 1.17 This corridor, operated by CFM(Centro), is made up of railways, roads and a pipeline that connect Zimbabwe and other interior countries with the port of Beira. The pipeline, connecting Beira with Mutare (just  - 8 - inside the Zimbabwe border), currently pumps nearly a million tons of oil products annually to Zimbabwe, representing 100X of Zimbabwe's total oil import requirements. The National Railways of Zimbabwe (NRZ) and Mozambique Railways - Caminhos de Ferro de Mozambique (CFM) - have completed the upgrading of the rail line between Dondo (28 km from Beira) and Mutare, Zimbabwe. Theoretical capacity on the line is 2.0 million tons per year. 1.18 The Beira Corridor System saw a dramatic increase in the level of traffic in 1987 and 1988. About 40,000 tons per year of Malawi's sugar goes through Tete by road to be transhipped by rail from Harare to Beira. The use of the Beira port for Zambian copper increased in 1987 due to a major shift from the South African port of East London. Zambia now ships about 60% of its copper exports via Dar es Salaam and 40% (about 80,000 tons per annum) through Beira. In addition, as Beira corridor facilities improve, Zimbabwe is expected to divert much of its overseas trade to that port. (d) CFM,EE 1.19 An important part of the Government's stratSgy is the transfer of the responsibility for the three main corridors from the Direccao Nacional de Portos e Caminhos de Ferro (DNPCF), which is a directorate of the Ministry of Transport and Communications (MTC - see Annex 8), to a new autonomous parastatal, Caminhos de Ferro de Mozambique, Empresa Estatal (CFM,EE). The draft statutes and operational guidelines, which are acceptable to IDA, were approved by the Council of Ministers on April 26, 1989, while the financial statements of the new entity are being finalised with the help of technical assistance provided by the Government of France. The intention is that the new entity should become financially viable as soon as possible and in any event not later than 1995, eliminating the need for Government subsidies. Agreement to this effect was obtained during negotiations (para 6.01). F. IDA's Role 1.20 IDA assistance to Mozambique to date has focussed primarily on supporting the implementation of the country's ERP. In June 1985, a first IDA Credit of SDR 45.5 million (Rehabilitation Program - Credit 1610-MOZ) was approved to help meet Mozambique's priority needs in the industrial, t.ansport and agricultural sectors. Performance under the Credit has been satisfactory, with about US$41.5 million disbursed by end-February, 1989. The Credit has provided the Ministry of Trade (MOT) with an improved procurement capability, and has allowed enterprises benefitting from the financing to produce more efficiently through provision of spare parts, raw materials, and technical assistance at the managerial level. The Governments of the Kingdom of Norway and Italy provided an additional US$22 million of co-financing with the Credit. A second IDA credit (Credit 1806- MOZ), of SDR 15.6 million to support the Energy Technical Assistance and Rehabilitation Project, designed to bring about quickly a substantial improvement in the supply and distribution of electricity and petroleum products to the main trade and processing areas, was approved by the Executive Directors in May, 1987. US$1.9 million of this Credit had been disbursed as of February 28, 1989. This was followed by a third IDA credit  - 9 - (Credit 1841-MOZ) of SDR 54.5 million (Second Rehabilitation Project) approved in August, 1987, in conjunction with an Africa Facility Credit of SDR 14.5 million and a Swiss Special Joint Financing Grant of SWF 16.9 million. Disbursements totalled US$68.0 million as of February 28, 1989 out of a total of US$110.0 million (including cofinancing). An Education and Manpower Development Credit of SDR 11.7 million (Credit 1907-MOZ) was approved in May, 1988, an Urban Rehabilitation Credit of SDR 44 million (Credit 1949-MOZ) in August, 1988 and a Health and Nutrition Credit of SDR 21 million (Credit 1989-MOZ) in May 1989. As of February 28, 1989 US$1.51 million has been disbursed from Credit 1907 and US$1.85 million from Credit 1949. 1.21 IDA'S role in the transport sector has so far mainly been to help coordinate donors on the Beira Corridor Program, the largest of Mozambique's transport rehabilitation programs, through bringing together economically synergistic components and by doing the overall economic and financial analysis. Thus when, early in 1986, SADCC presented for consideration by donors a detailed, 10-year, US$660 million program, IDA suggested a phased program which would restore the transport capacity of the Corridor, largely through rehabilitation of facilities and technical assistance. At a donors meeting in Beira in April 1986, a revised program spanning a period of five years was presented for donors' review and was approved by Mozambique and SADCC with some adjustments. The approved program included rehabilitation of the infrastructure services (roads, water, sewerage, erosion control, etc) ano housing for the city of Beira. To make the implementation of the program more manageable and to ease the housing shortage and improve urban services during construction activities of the corridor itself, IDA included assistance for rehabilitation of the cities of Beira and Maputo in the SDR 44 million IDA Credit for Urban Rehabilitation approved in August 1988 (Report No. 7279-MOZ, July 1, 1988). IDA worked closely with the European Development Fund (EDF) to draw up the terms of reference for a manpower development and training study for CFM(C) which was financed by the EEC and the recommendations of which are incorporated in the proposed project. The IDA Project Preparation Facility (PPF) has been used to fund a locomotive study for Mozambique and a management information system needs assessment for CFM(C), both of which have served as the basis for components in the proposed project. PPF funds are also being used to finance a technical assistance and training needs study for the whole of Mozambique's transport sector. IDA is currently undertaking a Transport Sector Review for Mozambique with the assistance of the U.K. Overseas Development Administration (ODA) and is also undertaking a study of the financial viability of all of the transport corridors in the SADCC region with funding from the Netherlands, USAID, Denmark and Canada. II. THE BEIRA TRANSPORT CORRIDOR A. Physical and Institutional Framework 2.01 The Beira Transport Corridor, located in central Mozambique (see Maps Nos. 20101 and 20102), consists of (i) Beira Port; (ii) the single track 315 km rail link from Beira to Machipanda on the Zimbabwe border, midway between Beira and Harare; (iii) a 282 km two-lane highway from Beira to Machipanda; and (iv) a 300 km oil pipeline from Beira to Mutare, just across the border in Zimbabwe.  - 10 - 2.02 CFM(C) is responsible for the operation and maintenance of Beira port and the central rail system which, in addition to the Beira-Machipanda link, includes the Sena line linking Beira with Malawi. The Sena line has been closed for a number of years because of the security situation and only the Beira-Machipanda line is operating regularly. The pipeline is a joint state-private enterprise, operated by CPMZ (Companhia Pipeline Mozambique Zimbabwe). The road falls under the responsibility of the Ministry of Construction and Water. International freight forwarding companies are well represented in the port and, to encourage private enterprise, agreement was reached at negotiations that BCA will submit proposals to IDA by December 31, 1990 for plans to expand private sector participation in port and railway related activities such as warehousing and container freight stations (para 6.01). B. Beira Port (a) Basic Facilities 2.03 The port is situated on the left bank of the Pungue River at latitude 190.50 and longitude 340.51 east. May 3 (IBRD Map No. 21356) shows the layout of the port. The first section of the existing wharves was built immediately after World War I with major extensions undertaken after 1930. In 1969, a pipeline was constructed to link the port with Zimbabwe. Currently, facilities for the working of vessels and handling and storage area amount to 350,000 sq.m. and covered areas to approximately 59,000 sq.m., of which approximately 34,500 sq.m. are used for transit storage. Specialized facilities include a cold store with a capacity of 1,000 tons for fruits and 500 tons for frozen meat, a coal yard with a holding capacity of 150,000 sq.m. and a loading capacity at the wharf of 700 tons per hour, seven tallow tanks with a storage capacity of 2,630 tons and three molasses tanks with a total capacity of 26,000 tons. The handling rate for tallow and molasses is approximately 120 tons per hour. Depth alongside is 8 m at Berths 2-5 and 8-9 m at Berths 6-11. 2.04 During the past decade the condition of the port, particularly of the wharves and loading equipment, has deteriorated. This led to initiation of a rehabilitation of the port (financed by the Netherlands) in 1982 which included the paving of the operational area, upgrading of the coal loading facilities to a capacity of 1.2 million tons per annum (t.p.a.), new cargo handling equipment, improvement to communications, lighting and navigation systems and, since 1987, technical assistance in port operations and training. The rehabilitation program has also provided a paved storage of 21,000 sq.m. behind Berths 9 and 10 to serve as a temporary container terminal and enable the working of ro-ro vessels at the berths. Reconstruction of Berths 2-5 started in January 1988 under EEC financing. (b) Navigation 2.05 The port is situated some 20 km from the open sea. The channel for shipping currently has an aveiage depth of 6 m. below chart datum; however, there are some shallow spots due to siltation, with a depth of only 4.60m below chart datum. There are considerable variations in the frequency of high waters (over 6 m) and low waters (less than 1 m). Due to the tidal  - 11 - range, at the channel depth of 6 m., vessels with a draft of 9.7 m, corresponding to a fully laden 20,000 dead weight ton (DWT) vessel, can transit the channel during 602 of the high water. The number of ship days of delay due to waiting for tide and for daylight were recorded at 200 in 1980. (c) Organization and Management 2.06 The port is administered by CFM (C) which also has responsibility for the railway system from Beira to the borders with Zimbabwe and Malawi. An organization chart of CFM (C) is shown in Annex 8, Chart 3. CFM (C) is headed by a director-general. Reporting to him are the director in charge of port operations and maintenance and his counterpart for the railways. The port and the railways share planning, financial and marketing services with each having specialist operating and engineering services. Full time port employees total 2,519; with casuals (part time) numbering 3,317. A breakdown of the personnel establishment is given in Annex 2. 2.07 The port works three shifts, totalling 24 hours: 0700 to 1500 hours; 1500 to 2300 hours; 2300 to 0700 hours. The organization of labor for cargo handling operations generally follows conventionally accepted practice. A gang working at hatch consists of one foreman, one tally man, one winchman, one forklift driver and, depending on the nature of the cargo, nine (general cargo) to 1; (bagged cargo) stevedores of whom four are normally deployed at the wharf and the rest in the hold. At the shed/storage area receiving cargo is a complementary storage gang made up of one foreman and 9 to 17 men. Considering the type of cargo that is being handled at the port and the labor intensive nature of the operation, labor is not excessive. (d) Traffic 2.08 Port throughput, after reaching a peak of 3.0 million tons in 1975, had been steadily declining (see Table 1.3) until recently due mainly to the loss of transit traffic to South African ports. The trough was reached in 1986 when the port handled 1.33 million tons of which 0.84 million were petroleum products (over 902 of which was for Zimbabwe). 1987 witnessed an increase of 47X to a total of 1.95 million tons of which 1.01 million were petroleum products. Total traffic increased by a further 13% in 1988 to 2.20 million tons. The total number of containers (TEU's) handled in 1987 were 10,072, up from 4,907 in 1986, an increase of 105%. Of the 10,072 containers handled in 1987, 6,989 were transit traffic (4,997 export and 1,992 import). Container traffic rose by a further 50% in 1988 to 15,279 containers. The port is, therefore, experiencing a traffic increase as a result of improved management and improved efficiency. (e) Cargo Handling Equipment 2.09 Cargo handling equipment currently consists of 51 electric quay cranes, 74 fork lift trucks, 4 heavy duty (lifting capacity 22 tons) fork lift trucks, 9 shunting tractors, 10 mobile cranes, 10 tug masters and 30 trailers. Equipment availability in 1987 was:  - 12 - Electric quay cranes 70Z Forklift trucks 592 - 87% Heavy duty forklift trucks 61% Shunting tractors 33Z Mobile cranes 40% Tug masters 56% Trailers 642 Sources CFM(C) (f) Port Operations 2.10 Annex 5 presents a summary of past performance of port operations and envisaged targets with improvements under the project. After a long decline, port operations have significantly improved in 1987-1988, partly due to the very effective technical assistance programs initiated by Port of Amsterdam Consultants (financed by Netherlands) and Plantrans (financed by Finland and Sweden) and partly due to better equipment availability. Handling rates for all categories of cargo increased, with container handling in particular reaching a peak loadinglunloading rate of 250 tons equivalent per hour. This is the more important since container ships are amongst the most costly vessels and without these improvements in handling times shipowners might be reluctant to call at Beira. However, further improvements are needed in the handling of non-containerized cargo, where progress has been limited due to the systematic procedure of direct unloading: trains are loaded alongside shies, with ships spending much of their time in port waiting for wagons. The situation has been exacerbated by the shortage of shunters, the lack of coordination of shunting operations, the limited use of pallets for bagged commodities and the lack of bulk unloading/loading facilities for commodities such as corn, sugar fertilizers, etc. These constraints are being addressed under the project, together with provision of spares to improve availability of handling equipment and training for port employees, and are expected to result in improvements in handling rates as summarised in Annex 5. (g) Transit Facilitation 2.11 The transit of goods to Zimbabwe via the Beira Corridor is not affected by customs delays. Customs clearance is unusually fast at Beira for international transit goods. Facilitation documents were simplified in 1986, and a new unified document (Combined Port and Customs release document) has come into use. This document, filed in 7 copies, is used for Customs, the port and the railways. Shipping agencies must attach to it the bill of lading and manifest, and, in case of breakbulk cargo, invoices to assess the value of goods. Customs clearance time for import transit goods are generally several hours to one day, and rarely exceed two days. For export transit goods, the necessary documents include the bill of lading and shipping instructions, and in case of breakbulk, the bill of  - 13 - entry, stamped by the Government of Zimbabwe, to assess the value of the goods. Customs clearance is in no way a bottleneck at the present time, as it is on other links to landlocked countries in Africa. 2.12 A specific and important example of Mozambique/Zimbabwe coordination is the joint preparation by the Mozambican Beira Corridor Authority (BCA) and the Zimbabwean Machipanda Beira Authority (MBA) of a transit agreement to expedite operations, prevent problems, and in case of disagreements, provide for mutually acceptable arbitration. A draft agreement is currently under discussion between the two governments. It is a dated covenant that the agreement be finalized, by June 1990 (para 6.01). C. Beira-Machipanda Railway (a) Basic Facilities 2.13 The 315 km Beira-Machipanda line is single track with 38 passing loops. After the section Beira-Inhamitanda (see Map 2), the route rises till Gondola (Mozambique) with steep gradients and sharp curves. From Gondola to Almada the line goes through rather level terrain but from Almada to the Zimbabwe border the line once again has sharp curves and steep gradients. From the border to Dondo, the track has been reconditioned by the National Railways of Zimbabwe (NRZ) over the last two years, including the replacement of defective sleepers with steel sleepers, the replacement of old 30 kg/m rail by 18 m lengths of fished 40 kglm rail and rebalasting. (b) Motive Power 2.14 CFM(C) has 7 mainline diesel locomotives on the Beira-Machipanda line, of which 3 were operational in December 1988, 4 being out of service due to accidents and attacks by armed bandits. A further 13 mainline diesels have been inaccessable at Inhaminga on the Sena line since 1983 due to bandit activity. These locomotives are considered to be beyond economic repair. CFM(C) also has 12 garratt steam locomotives which are mainly used for domestic traffic, but occasionally haul inte alational freight trains. At present, much of the motive power on CFM(C) is leased from Zimbabwe Railways (1800 loco-days per year at a cost of Z$4,682 per day or about US$2,500 equivalent). Shunting operations in Beira port are handled by only one steam shunter, which is clearly insufficient. (c) Rolling Stock 2.15 The CFM(C) freight fleet totals 3,237 wagons of which 2,424 are currently operational (See Table 2.1 below). Most international traffic is carried in NRZ wagons for which CFM(C) pays a substantial rental. DANIDA will finance a component of the program for the repair and rehabilitation of wagons, which was appraised in September, 1988 and is expected to begin in 1989 and last for approximately three years.  - 14 - Table 2.1 - CFM(C): Locomotives and Rolling Stock Main Line Operational Locomotives Total in 1988 Availability Diesel 7 3 602 steam 12 12 502 Totals 19 15 Wagons/Type Tank 149 90 60Z Covered 662 542 822 High Sided 1486 976 652 Low Sided 888 776 872 Flat 52 40 772 Total: 3237 2424 752 Source: CFM(C) (d) Workshop 2.16 CFM(C) has a steam locomotive workshop in Beira but virtually no spares and materials. Part of this workshop was converted under a component financed by USAID and is to be equipped and stocked to undertake diesel repairs as part of a USAID Regional Rail Systems Support Project. This project, which is to start in 1989, will provide a total of US$34.5 million in grant aid to Mozambique, mainly to CFM(S). In Beira, US$1.7 million of tools and equipment would be provided for the workshop. (e) Staffing 2.17 CFM(C) currently employs 3,000 railway staff on the Machipanda- Beira line out of a total railway staff (including Sena line) of some 5,300. Given the low traffic volume currently being handled this reflects some overstaffing, but recent studies by CFM(C) indicate that, with the expected increases in traffic in the near future, there will be no overstaffing. (f) Traffic 2.18 In 1987 CFM(C) handled 481,500 tons of freight of which 397,000 tons were transit traffic on the Beira-Machipanda line. This is only 10% of what it handled a decade earlier, but compares with 368,100 tonnes in 1986 of which 286,100 were transit traffic, an increase in transit traffic of 39% from 1986 to 1987. Preliminary results for 1988, however, show no further growth over 1987, largely due to shortage of locomotive power.  - 15 - D. Beira-Machipanda Road 2.19 The 282 km road is to be rehabilitated with financing from the African Development Bank (AfDB), the EEC, Sweden and Japan. At present, traffic on the road is very light partly because of its poor condition caused by lack of maintenance over the last decade and inadequate original design of certain sections. When rehabilitated, it will have a carrying capacity of about one million tons annually. E. Pipeline Beira-Mutare 2.20 This pipeline carries finished products from the port to Zimbabwe and transports 100% of Zimbabwe's imported liquid fuel. In 1987 it carried some 930,000 tons, up from 772,700 in 1986 (preliminary estimate for 1988, 920,000 tons). The pipeline is in good condition and has a capacity of upward of 1 million tons, enough to cater to Zimbabwe's needs for the foreseeable future. F. Training 2.21 Since the mid-1970s, CFM(C) has suffered from an acute shortage of skilled labor and, most importantly, a lack of qualified and experienced technical supervisory and management staff. This problem is further compounded by the country's low levels of education and literacy. Currently, DNPCF is responsible at the national level for the planning, implementation, supervision, coordination, and evaluation of all CFM training activities. DNPCF carries out this responsibility with the assistance of consultants through the National Railway Training School in Inhambane and three Regional Training Centers (CFM(S) Maputo, CFM(C) Beira and CFM(N) Nacala/Nampula. Although considerable training activities have been carried out by the Inhambane School and the Regional Centers since the early 1980s, the overall results have been somewhat limited due to a variety of problems including (i) weak training capacity at national and regional levels; (ii) lack of a sound national training program which would clearly delineate DNPCF's and regional centers' responsibilities and resources (financial, human, and physical) required to carry out these responsibilities; (iii) low levels of education and literacy of most trainees; (iv) the odd location of the Inhambane School; 2 (v) shortages of (a) effective training courses; (b) relevant training equipment and materials; (c) adequate physical training facilities; (d) experienced instructors with good knowledge of the language(s) spoken by trainees; (vi) shortages of trainees because of lack of incentives and other organizational constraints; and most importantly, (vii) a weak personnel management system which does not provide the right kind of motivation through competitive salaries and career development opportunities. 2.22 In collaboration with technical assistance provided by donor countries and EEC, DNPCF has been taking remedial actions to strengthen and expand CFM training capacity at both national and regional levels. 2/Inhambane is hundreds of kilometers from any of the main railway systems, with which it has very poor communications. This makes it difficult to attract candidates to the school.  -16 - Consultants financed by EEC have completed on behalf of BCA a comprehensive organization and manpower development plan for CFM(C). The plan identifies training needs and the resources required to prepare and implement training programs for port and railways personnel. Based on the consultants' recommendations and data collected during appraisal, the project includes funds for the preparation and impleret4tion of two parallel four-year programs for the training, retraining, and upgrading of some 4,500 port and railways personnel. Technical Education and Vocational Training 2.23 The responsibility for developing policies and programs of technical vocational education and training rests with the Secretary of State for Technical and Vocational Training (SETEP). Up to the late seventies, over two-thirds of the population was illiterate. Considerable progress has been made since then in school enrollment and adult literacy schemes. However, the country's educational and literacy levels are still low. It is now estimated that only about 5% of the school age population attend secondary school. 2.24 The University Eduardo Mondlane (UEM) in Maputo offers only five major basic programs, including engineering. The UEM was negatively affected by post-independence events which caused a sharp decline in enrollment until 1978. Although enrollment increased in subsequent years, recent statistics indicate that only a few hundred students graduate per year, and less than forty from the faculty of engineering. Of the several institutions responsible for technical/professional training, four in Beira run by SETEP are expected to make a useful contribution to certain aspects of CFM(C)'s training progrems, with assistance to be provided under the project (see Annex 2). III. THE PROJECT A. Proaram Oblectives 3.01 The primary objective of the program is the rehabilitation and upgrading of the Beira Corridor in order to restore its cost efficient transit functions which are vital to Malawi and Zimbabwe and constitute an important source of foreign exchange and employment for Mozambique. The secondary objectives of the program are as follows: (i) to help the Government achieve its objective of providing cheap transport for the agricultural, productive areas along the Beira Corridor; and (ii) to facilitate capacity building by developing skilled, semi- skilled and managerial local staff to gradually replace the technical assistance program, and introduce effective financial management, particularly in the CFM(C).  - 17 - B. The Beira Corridor Authority 3.02 In December 1985, the Government of Mozambique (GOM) established the Beira Corridor Authority (BCA) and appointed the Deputy National Director of Ports and Railways to be the BCA Director. The objective of BCA is to 'direct, plan, coordinate, mobilize and supervise the entire process leading to implementation of the rehabilitation of the Corridor's system of transportation and communications, including projects of a complementary nature, specifically those for infrastructure for the city of Beira" (Bulletin of the Republic, Resolution Establishing Beira Corridor Authority, Series 1, Number 1, January 1, 1986). To strengthen BCA, Sweden, Denmark, Norway and Finland, in 1986 provided BCA with an advisory team, stationed in Europe and available on a part time basis. In 1987 this team, with some overlap, was replaced by a 9-man, full time team of which eight are presently stationed in Beira. This team, also financed by the Nordic counties, has considerably strengthened the capacity of the BCA to coordinate and speed up implementation and freed the BCA Executive Director to focus on the strategic aspects of the Beira Corridor System and the liaison with Mozambique's partner countries in SADCC as well as the donor community providing financing for the rehabilitation. Donor coordination meetings are held on an "ad hoc" basis as necessary. The performance of BCA has been fully satisfactory. C. The Beira Transport Corridor Program 3.03 BCA is undertaking the first phase of a 10-year development program, with an estimated cost of around US$345 million and with funding largely lined up from 18 bilateral and multilateral agencies (see Tables 3.1 and 3.2). The program consists of rehabilitation and upgrading of the port, rail, and road facilities and provision of technical assistance and training. The overall Beira Transport Corridor program includes four categories of sub-projects: railway, port, road and general sub-projects, the last being mainly for the upgrading of the city of Beira, where present infraotructure and services are seriously deteriorated (the bulk of these are being addressed under the IDA-assisted Urban Rehabilitation Project, Cr.19490-MOZ). The criteria on which investments are planned give priority to rehabilitation of existing infrastructure and manpower development (see Annex 9). A summary description of each category of sub-projects is given below (cost estimates shown include contingencies). 3.04 The main port sub-proiects are: (a) Dredging the entrance channel to deepen it from the current depth of 6 meters to 8 meters. This will enable larger vessels of PANAMAX size, third generation ro-ro vessels and up to 60,000 DWT product tankers to use the port and will result in considerable cost savings. The contract for this component has been awarded and work started in February, 1989 and will be completed by the end of 1990 (cost US$18.9 million, financed by The Netherlands); thereafter, mairntenance dredging will be necessary to maintain the depth of the channel. Equipment for maintenance dredging is proposed for EEC financing (cost US$12 million);  - 18 - (b) Multipurpose and container handling terminal, with a capacity of up to 100,000 containers p,- year including reconstruction of berths 2-5 and new buildings. The quay foundations of berths 2-5 (645 meters) had deteriorated to such an extent during their 60 year life that the structure might have collapsed at any time; reconstruction was therefore necessary and started in June, 1988 (cost US$67.5 million financed by EEC 48.0 million ECU, Netherlands 4.7 million Guilders and FINNIDA US$4.3 million); (c) New oil terminal to make it possible to handle oil product tankers of a size up to 60,000 DWT, as against the existinig 20,000 DWT limitation and thus reduce the cost of shipping; oil products. The final design is underway with call for tenders expected in March 1989 (cost US$13.9 million, financed by NORAD); (d) Reconstruction of the port railways. Beira port was designed as a railway port, with cargo being loaded and unloaded directly to and from rail wagons. With the development of multi-purpose and container terminals and the handling of ro-ro and container vessels, direct loading and unloading of rail wagons at berths becomes inefficient. Hence the port rail network needs to be redesigned. Detailed design has started, implementation expected to start in 1989 (cost US$13.1 million financed by Italy); (e) Equipment for multi-purpose and container handling terminal. The first phase of rehabilitating existing equipment (cranes) is completed. The second phase of providing 4 forklift trucks (two 42 tonnes and two 28 tonnes) plus four new tugmasters is also complete. This enables the port to handle 40 ft. containers, an important step in its ability to compete with RSA ports. A third phase includes new ship to shore cranes (out to tender) to be implemented in 1989/90 (cost US$18.2 million financed by AfDB) and other equipment and technical assistance to be financed by FINNIDA/SIDA (US$24 million) and the Netherlands (US$6 million); (f) Tug and pilot boats - procurement of two harbour tugs, two pilot boats and two work and line boats to provide the port with modern service vessels and so reduce the turnaround time of vessels in port. The pilot boats are in service and the other boats are to be delivered in 1989; (cost US$10.6 million financed by DANIDA); and (g) Technical Assistance to Manage the Port, currently being undertaken by Dutch and Finnish/Swedish experts, needs supplementing with a training component, proposed for IDA financing (cost US$5.9 million financed by Netherlands, Finland and Sweden, additional TA and training are included in the IDA component and described in paras 3.10 to 3.17).  - 19 - 3.05 The main rail sub-proiects are: (a) Rehabilitation of the Beira-Machipanda line. This work which consists of resleepering and rebalasting, has been completed, except for a 25 km stretch between Dondo and Beira (cost US$9 million, implemented by NRZ and financed by NORAD, SIDA, Austria, Canada and USAID); (b) Track maintenance equipment - tender documents being prepared (cost US$3.5 million financed by Austria); (c) New radio telecommunications and train control systems are intended to improve communications and train control to increase efficiency and line utilization (cost US$8.4 million, fin-aeed by Italy); (d) Acquisition of locomotives to replace leased Zimbabwean locomotives and provide additional capacity to cope with increased traffic and shunting (see paras 3.18-3.25 for details). IDA financing is proposed for the shunting component, while bilateral financing from Canada and untied financing from Japan are proposed for the main-line diesel locomotives. (e) Manpower development and training. Technical assistance in line positions to reinforce CFM(C) with respect to general management, supervisors and key personnel in technical departments and to establish training programs for Mozambican personnel. Programs are already in place for the port funded by the Netherlands, Finland and Sweden and a recently completed organization and manpower development study financed by the EEC has determined the remaining needs for the port and for the railway which are proposed for financing by Spain and IDA (see paras 3.10 - 3.17 for details). 3.06 Road sub-prolects include mainly the upgrading of the road between Beira and Machipanda (US$27.0 million, EEC, Sweden, AfDB). 3.07 General sub-Prolects, apart from the rehabilitation of the infrastructure and services and housing provision for the city of Beira, include the development of a cost-based tariff system (US1.3 million) and a management information system (US$1.0 million) both of which are proposed for IDA financing (see para 3.26 for details). D. Overall Program Cost and Financing 3.08 Overall program cost and financing are shown in Tables 3.1 and 3.2.  -20- MOZAMBIQUE Beira Transport Corridor Project TABLE 3.1 PROGRAM COST ESTIMATES ---- (US$ Million) ---- COMPONENT LOCAL FOREIGN TOTAL A. PORT SUB-PROJECTS (q Channel dredging 0.34 25.00 25.34 (a) Multipurpose and containerterminal 3.50 61.00 64.50 (ii) Oil terninal 0.22 11.60 11.82 Iv) Port raiways 0.13 11.00 11.13 (v) Handling equipment 0.00 20.02 20.02 (vi) Tug and pilot boats 0.00 10.60 10.60 (vi) Technical asstance and training 1.00 17.64 18.64 (i) Other 0.50 9.19 9.69 SUB- TOTAL 5.69 166.05 171.74 B. RAIL SUB-PROJECTS (1) Beira-Machipanda line rehabilitation 0.50 8.50 9.00 (i) Track maintenance equpment 0.00 2.80 2.80 (Iii) Radio telecomsain control systems 0.00 8.96 8.96 (i) Rehab/aquisition of new locos 0.50 33.60 3410 MMn Maementand trairng 1.00 1860 19.60 (v) Other 1.13 8.63 9.76 SUB- TOTAL 3.13 81.09 84.22 C. ROAD SUB-PROJECTS (I Beira-Machipanda road upgrading 4.43 18.93 23.36 D. GENERAL SUB-PROJECTS () M.LS. and costing systems 0.00 2.20 2.20 (iiQ Housing for TA and emergency power 3.02 12.08 15.10 f(iui Other 0.24 7.22 7.46 SUB- TOTAL 3.26 21.50 24.76 TOTAL BASE COST 16.51 287.57 304.08 Physicl contingencies 1.13 23.48 24.61 Price contingences 0.91 15.32 16.23 TOTAL COST 18.55 326.37 344.92  - 21 - Table 3.2 - Financing Plan USS Million Source of Finance General Port Rail Road Total EEC 1.0 67.1 - 5.1 73.2 NORAD 2.4 13.9 1.0 - 17.3 SIDA 12.4 8.2 1.1 3.5 25.2 FINNIDA 4.2 16.1 - - 20.3 DANIDA 2.6 16.5 4.1 - 23.2 FEDERAL REPUB GERMANY - - 1.8 - 1.8 THE NETH ERLAN DS 1.4 24.8 0.2 - 26.4 BELGIUM - 2.3 - - 2.3 ITALY - 13.1 8.5 1.2 22.8 U.K. - - 1.8 - 1.8 AUSTRIA - 4.8 - 4.8 SPAIN - - 5.0 - 5.0 AfDB - 19.8 - 17.5 37.3 JAPAN - - 8.0 - 8.0 USAID - 10.0 - 10.0 CANADA - - 21.7 - 21.7 IDA 1.8 12.0 25.2 - 39.0 GOM - - 4.3 - 4.3 UNDP - 0.5 - 0.5 TOTAL 25.8 193.8 98.0 27.3 344.9 E. IDA Assisted Program Componen'-s 3.09 Program components included in the IDA-assisted project are: (i) technical assistance to run port and railways operations in the Beira Transport Corridor during the 1989-95 period; (ii) manpcwer development and training at all levels of CFM(C); (iii) rehabilitation and acquisition of motive power for main line and shunting operations; (iv) technical assistance for the study and implementation of a railway/port cost accounting system and a management information system in CFM(C); and (v) studies and technical assistance for future investments. (i) Technicat Assistance to CFM(C) 3.10 Beira port and the crucial rail line between Beira and Machipanda are managed by CFM(C) under the general authority of Beira Corridor Authority (BCA). At present, there is considerable concern over the technical and managerial effectiveness of CFM(C). The major constraint on the efficiency of the organization has been identified as being the lack of  - 22 - managerial and technical, skilled and semi-skilled personnel due to the abrupt loss of skilled manpower at the time of Mozambique's independence. The efficient functioning of CFM(C) is crucial to the success of BCA's plan for the development and expansion of the Beira Corridor transport system. A study financed by the EEC, the terms of reference of which were agreed with IDA, has prepared a comprehensive Organization and Manpower Development Plan for CFM(C), which forms the basis of the TA and training included in the project. Manpower training targets for CFM(C) were established during the study and a 2lan of organizational development was agreed to enable CFM(C) to perform efficiently in the context of both today's restricted traffic flows and the predicted expansion of traffic over the next few years. The study also identifies the technical assistance required in the short to medium term while Mozambique staff is being trained. The TA component of the project provides the financing required to implement the manpower development and training proposed. Additional financing will be provided by bilateral donors for their ongoing technical assistance teams in Beira, including container terminal operations (Finland and Sweden), port operations (Holland) and railway operations (Spain). Details of the training and technical assistance programs are shown in Annexes 2 and 3. Agreement was reached during negotiations on the terms and duration of contracts for the key management posts for CFM(C) (para 6.01). 3.11 Additional technical assistance needs in the ,ort are quite limited due to the fact that the very effective TA programs being financed by the Netherlands, Finland and Sweden cover most areas of port operations. The effectiveness of these programs, which involve twenty experts over a four year period up to 1991, is not only due to the experience of the individuals involved but also to the fact that they work in line management positions with operational responsibilities. The existing positions cover port traffic, the container terminal, workshops, general cargo operations, maintenance, marketing and manpower development. The donors concerned have indicated their readiness to extend these contracts up to 1995. The additional positions needed and which IDA would finance are in the general service areas: (i) cost accounting specialist (48 months); and (ii) procurement expert (48 months) 3.12 Unlike the port, the railway has received only limited technical assistance in the last few years, from India and the German Democratic Republic (GDR), and has only now received the first 11 members of a team from Red Nacional de los Ferrocarriles Espanoles (RENFE) financed by Spain. The positions to be filled by Spanish TA (400 man-months) are described in Annex 3 together with the additional positions to be financed by IDA (a further 400 man-months). To help provide a unified and motivated' team, RENFE staff would focus on operational and technical positions, with IDA funding general manpower development, marketing and accounting positions for CFM(C) as a whole and railway maintenace positions. Building on the success of the technical assistance in the port, the positions in the railway would also be for line management with operational responsibilities. To ensure the effectiveness of the technical assistance effort, specific achievement goals will be set and performance monitored through the Project coordinator (see para 3.13 below).  - 23 - 3.13 A Project Coordinator has been selected who will be responsible for coordinating and monitoring the effectiveness of the technival assistance in the railway and training throughout CFM(C). His position will be funded by UNDP with the Bank as executing agency. The terms of reference for this position have been agreed with G0M and CFH(C) and are included in Annex 10. (ii) Capacity Building for CFM(C) 3.14 Two parallel programs for the training, retraining and upgrading of port and railways personnel at all levels will be administered by CFM(C) with technical assistance. Capacity building is a long-term process which requires continuing monitoring and sustained investment. Accordingly, the project would finance the first five years of a long-term manpower development program (15 year horizon) which would provide training courses for some 4,500 port (2,500) and railways (2,000) personnel from managers and middle level technicians to equipment operators, mechanics and administrative staff. Because of the low level of education and literacy of most personnel, it is expected that during the first five years substantial training technical assistance will be required to initiate the programs and to train Hozambican counterparts who would be gradually assigned to expatriate personnel. It is also expected that in addition to relevant technical subjects, many trainees will have to undertake more general subjects such as English, Portuguese, mathematics and chemistry in order to strengthen their basic education. 3.15 Financing would be provided under the proposed credit for (a) strengthening and expanding CFM(C) training capacity and its training center; (b) providing housing for instructors; (c) purchasing textbooks, training equipment and materials, audio-visual training aids anl developing relevant courses; (d) recruiting suitable instructors; (e) training abroad (fellowships) for selected middle an high level port and railways management staff as well as training officers and senior Mozambican instructors; (f) strengthening four technicallvocational schools which will assist with the training effort; (g) implementation of an incentive scheme for CFM(C)'s trainees; (h) a transportation sector survey of manpower requirements and related training needs; and (i) a mid- and end-project training evaluation, including the preparation of the next phase follow-up of the proposed training program. Annex 2 and attachments provide details of the training programs, including their objectives, elements and cost estimates. In order to carry out additional courses for CFM(C) staff, four technical/vocational schools in Beira would receive funds under the proposed credit for additional instructors and training equipment and materials. 3.16 A training section, with technical assistance, has been established and its chief appointed to be responsible for the planning, budgeting, implementation, supervision, and evaluation of all CFM(C) training activities. The section includes a port and a railways unit, and adequate staff to carry out its responsibilities including the supervision of the CFM(C) training center. After an adequate period of training abroad and in-service, as counterparts to the Chief of the Training Section and the training managers, CFM(C) training staff would become gradually more responsible for supervising the implementation of all training activities. CFM(C) has also established a Training Steering Committee within CFM(C)  - 24 - whose main responsibility is to provide policy guidelines, determine priorities, and periodically evaluate the training programs. The project will provide training technical assistance for a total of 552 man-months to help CFM(C) manage the execution of all training activities. Because of the substantial amount of technical assistance and training inputs expected to be provided by other agencies/donors, a close coordination is essential among all those involved with training to prevent costly duplication of efforts. Draft terms of reference for training technical assistance financed under the project are available in the Project File. Financing has also been provided under the project for an independent mid and end- project training evaluation, including the preparation of follow up training program. A transportation sector survey of manpower and training needs to guide further actions in the sector on a country-wide basis, is currently underway, financed out of PPF funds. Because of the nature of training, the Association would finance 100Z of all local and foreign costs. 3.17 Remuneration of CFM(C) Workers. To increase the productivity of CFM(C) workers, it is proposed that a study be undertaken of means to improve worker incentives. During negotiations, assurat.ces were obtained that BCA will undertake such a study using PPF funds and, by December 31, 1990, will prepare a detailed program for the implementation of the study recommendations, exchange views with the Association and circulate it to donors for comment. (iii) Acquisition of Locomotives and Shunters 3.18 In view of a continuing shortage of motive power for mainline and shunting services, DNPCF commissioned a study in 1988 by RITES of India financed out of PPF, to determine the needs of locomotives for the whole of its network and analyse the economics of rehabilitation versus new purchases. From this study it became apparent that much of the locomotive fleet was old and in need of replacement or rehabilitation. The current locomotive position of DNPCF is shown in Table 3.3 below; Table 8.3 - DNPCF Locomotive Fleet Number In Fleet Under/ 10 Yrs Over Awaiting or 10-25 26-30 30 No.to be Major Effective TYPE OF LOCO Less Years Years Years Total Scrapped Repairs Fleet GE - U-20C Diesel-electric } 49 34 - - 83 19 18 46 mainline ltcos } AEI-Suizer Diesel-electric } - - 10 - 10 8 __ 2(in Mat mainline locos } Romanian Diese'-hydraulic } 31 - - - 21 1 22 8 Shunting locos } Carratt Steam Locos (CFM(C)) - - - 14 14 __ e 8 Shunting Steam Locos (CFM(C)) - - - 10 10 2 6 3 Simple Steam Locos (CFM(C)) - - - 8 8 2 4 Simple Steam Locos (CFM(N)) - - - 6 5 2 2 1  - 25 - 3.19 The requirements of locomotives on the Mozambican Railway Systems were studied on the basis that all three railway corridors (excluding the Sena Line) will be open for international and domestic traffic from 1991 onwards. Based on 24 hour working of all the three lines and excluding the future possible requirements of the Sena line, the number of additional main-line locomotives to be acquired has been estimated up to 1995 as follows: 1991 - 9 1992 - 4 1993 - 2 1994 - 4 1995 - 6 25 3.20 If however, security conditions permit only 12 hour working in 1991, the number of additional locos to move the traffic demand in 1991 would be 36, as opposed to 9 with 24-hour operations. In so far as the Beira Corridor is concerned, the number of additional locos required to compensate for 12 hour working in 1991 is 9. Since the Beira port is expected to develop the full capacity to meet the full traffic demand in 1991, it is crucial that locomotive capacity will be available to ensure that materialisation of returns on the large capital investments in the port is not delayed. It is, therefore, proposed to acquire 15 locomotives in 1991 itsclf, instead of 9 in 1991, 4 in 1992, and 2 in 1993. When discounted to 1991, the cost of reasonably insuring against the risk of the security problem not being fully solved ia 1991, by purchase of the 6 locomotives in advance, works out to 0.711 times the cost of a new locomotive or US$1.21 million. As against this, the possible loss of benefit caused by insufficient motive power capacity to move the traffic forecast would be considerably greater (about US$9.4 million annually). The acquisition of the 15 locomotives is also a safeguard against any possible slippages in the ongoing rehabilitation program. 3.21 As regards shunters, the existing steam locos will continue to be used at outstations, with the acquisition of diesel shunters for the Beira complex, where 6 are required. 3.22 The proposed USAID-funded Regional Rail Systems Support Project (para 2.16) covers the needs of rehabilitation and major overhaul arrears of diesel mainline locomotives, and parts for maintenance for a three year period. It also provides funds for tools, equipment and facilities, as well as technical assistance to maintenance workshops. Funds for major overhauls and rehabilitation works during the project period, (estimated at US$1.5 million), however, are not covered under the USAID project and are included in the IDA assisted project. 3.23 A rehabilitation program for the comparatively new diesel shunters (Romanian) is currently under way, and is expected to be completed in 1990. Funding of an additional $1.5 million for spares and technical assistance to complete the rehabilitation program is included in the IDA financing.  - 26 - 3.24 The proposed locomotive allocation policy on the CFM(C) is as follows: (i) Passenger services, service trains, and shunting services at outstations (other than Beira) are to be met by steam locomotives, which would not be economical for mainline freight services. (ii) Mainline diesel locomotives will be used to operate freight services. (iii) Shunting services in the Beira complex are to be provided with diesel shunters. (iv) The steam locomotives, which are of vintages ranging from 1946 to 1955, will have to be phased out, in due course, on an age-cum-condition basis. 3.25 The railways in Nozambique, in contrast to most other Sub-Saharan African railways have the advantage of a standardized main-line diesel locomotive fleet. In view of the imperative need to limit the variety of locomotives to one or two not very dissimilar types, in order to maintain the benefits of standardization, the acquisition of the new locomotives is proposed to be done with appropriate bilateral assistance (5 locomotives with untied Japanese assistance and 10 locomotives with Canadian assistance). As regards the six shunters, it is proposed to acquire them out of IDA funds under ICB. These locos would be confined to working on the Beira Corridor, where, the necessary expertise and inventory would be developed. The allocation of these locomotives will be as stated in para 3.24 (ii) and (iii) above, which was confirmed at negotiations (para. 6.01). (iv) Cost Accounting and Management Information Systems 3.26 CFM(C)'s accounting system, although better than in other DNPCF branches, is incapable of providing management with the elements required for day-to-day and long-term management. Not only is there lack of qualified personnel, but also the system is not geared towards cost accounting which would provide adequate information on a cost/profit center basis. Furthermore, CFM(C) does not have a management information system to facilitate and expedite the accounting functions. The purpose of this component is, therefore, to contribute to the strengthening of financial management of the CFM(C) railways aad Beira Port. The cost estimates included in Table 3.3 are based on the recommendation of the M.I.S. needs study financed by the PPF. (v) Studies and Technical Assistance for Future Investments 3.27 To assist in the ongoing rehabilitation of the Limpopo line (para 1.14), 100 man-months of IDA funded assistance is proposed for the management team of the Brigada de Melhoramentos do Sul (BMS), the  - 27 - development and construction arm of CFM(S). The U.K.ODA-funded study of the rehabilitation of the Limpopo line3 identified a critical shortage of management in BMS and recommended expatriate technical assistance to fill the gap during the rehabilitation period 1989-91. IDA comented on the study, generally accepting its findings but urging restraint on the size of the investment program. The IDA-funded technical assistance, terms of reference for which are available in the Project File, would identify future technical assistance and training needs after the line becomes operational in 1991, which might form part of a future IDA-assisted project. The IDA credit would also fund a review of the technical assistance needs oi DNPCF as well as the preparation of future projects in the transport sector. F. Rationale for IDA Components 3.28 IDA's role in the project has been to assist in ensuring the viability of the whole investment pr ;ram through (a) analysis of the costs and benefits of each component and of the overall program; (b) acting as a catalyst for other investors in the project, as in the case of locomotives; and (c) as a lender of last resort in the capacity building program to complement investments made by others. IDA's investments in the project are thus ones which IDA is uniquely equipped to finance and are directed towards the self-sufficiency and financial viability of the Beira Corridor and other elements of the DNPCF within the framework of the ERP. G. Prolect Cost Estimates 3.29 Total estimated cost of the IDA assisted project is US$73.75 million (see Table 3.4). The costs have been estimated net of taxes. I/Study for the Total Rehabilitation, Operation and Maintenance of the Limpopo Railway Line; Mott Hay and Anderson International Ltd., June, 1988.  - 28 - Beira Transport Corridor Proiect Table 3.4 - Pro1ect Cost Estimates FY89-95 (US$ Million) Local Foreign Total A. Technical Assistance to Manage: (i) the Railway 1.17 8.36 9.53 (ii) the Port 0.15 1.00 1.15 B. Training for: (i) the Railway 0.58 3.88 4.46 (ii) the Port 0.55 3.66 4.21 C. Aquisition of Locomotives: (i) Main-line 25.00 25.00 (ii) Shunters - 6.00 6.00 D. Rehabilitation of Locomotives .45 2.55 3.00 E. M.I.S. and Cost Accounting Systems 0.20 4.00 4.20 F. Locomotive and MIS Studies (PPF) - 0.20 0.20 G. Studies and TA for Future Investments .35 3.15 3.50 Total Base Cost 3.45 57.80 61.25 Contingencies: Physical 0.36 5.35 5.71 Price 0.44 6.35 6.79 Total Cost 4.25 69.50 73.75 Financing: GOM 4.25 Local Costs Japan 8.00 Item C. (i) Spain 3.50 Item A. (i Portugal 1.00 Item G. Canada 17.00 Item C. (i) IDA 40.00 The Rest H. Financing 3.30 In view of the continuing severe domestic resource constraints, the Government's financial contribution will not exceed local costs of US$4.25 million or 5.8% of total project cost. Hence, full financing of foreign costs (US$69.50 million, net of taxes) will be necessary. IDA funds of US$40 million have been allocated to the project in FY90 and untied financing from Japan (US$8.0 million) and parallel financing from  - 29 - Spain (US$3.5 million) and Portugal (US$1.0 million) have been assured. Parallel cofinancing of US$17.0 million for locomotives is being secured from CIDA. It is not envisaged that any of the donors will request the World Bank to administer their funds, either through Trust Fund arrangements or through other cofinancing arrangements involving review of withdrawal applications. I. Implementation 3.31 The IDA-assisted components will be implemented over a period of six years and be completed by December 31, 1995, with a closing date of June 30, 1996. DNPCF and its successor CFM,EE through its subsidiary CFM(C), will be responsible for the implementation of the project. Conditions of effectiveness would be (a) a signed legal agreement (Beira Corridor Implementation Agreement) specifying the mutual obligations of all agencies involved in implemenitation of the project and, (b) a signed subsidiary loan agreement between the borrower and DNPCF (para 6.01). A Project Coordinator has been appointed to supervise the implementation of the Project (see para 3.13). In addition, expatriate staff will be employed to assist CFM(C) in implementation. The implementation schedule is shown in Annex 1. To measure progress in meeting anticipated improvements in CFM(C) efficiency and productivity, CFM(C) has agreed to a comprehensive set of operational targets for the project period (Annex 5). These targets will be reviewed by IDA at annual implementation reviews each September. J. Procurement 3.32 Procurement arrangements for the IDA-assisted project are summarized in Table 3.5 below: Table 8.6 - Procurement Arrancements (USS Million) IDA FINANCED FINANCED BY DIRECT oTHER OTHER PROJECT ELEMENT ICB IS/LS CONTRACTING PROCEDURES DONORS/GOM TOTAL A. Goods: 1. Shunting locomotives G.0 6.0 2. Main line diesel locomotives 25.60 26.0 3. Spare parts, tools, vehicles, computers and ottice equipment 2.0 2.0 4.0 B. Technical Assistanc, and Training: 1. Railway/Port TA A Training 2S.8 8.6 34.4 2. Cost accounting and MIS 4.0 .2 4.2 3. PPF studies 0.2 0.2 Total: 6.0 2.0 2.0 86.6 33.8 73.8 3=S = _= ====  - 30 - 3.33 Goods: Equipment and goods are expected to be financed both under the proposed credit and by co-financiers. Such equipment and goods, financed under the proposed credit, would be procured by ICB in accordance with the Association's Procurement Guidelines. Eligible domestic manufacturtrs would be afforded a preference of 15 percent or applicable duty, whichever is lower, under ICB procurement. A margin of 7.52 preference to domestic contractors would also apply. In cases where the borrower can show to the Association's satisfaction that the items required are clearly established manufacturers' parts, in which case they may be obtained directly from the original suppliers through negotiated direct contracts up to a maximum of US$2.0 million in total. Where small orders are under US$100,000 each, up to a maximum of US$2.0 million in total for which international or local shopping through obtaining at least three price quotations will be applied. Goods estimated at US$25 million and consultancy services estimated at US$6 million parallel-financed by co- donors will be procured in accordance with the procedures of the respective agencies. All documentation for procurement of goods financed by IDA in excess of the US$100,000 will be subject to IDA :eview prior to issuance of tenders and award of contracts. 3.34 Technical Assistance: Technical Assistance and training funded by IDA will be provided by speciAlist consultant firm(s) who will be appointed in accordance with the World Bank Guidelines for the use of consultants. K. Disbursements 3.35 Disbursements from the IDA Credit will be made on the following basis: Mi) 100% of foreign expenditures and 70% of local expenditures for (a) acquisition of shunting locomotives and (:) materials, equipment, furniture and supplies for provision of housing for expatriate specialists and instructors; (ii) 100% of foreign expenditures for vehicles, spare parts, computers, office equipment, materials and supplies; and (iii) 100% of the total cost in the case of consultants' services, training, studies and external audit services. assistance for transport planning; and (iii) training; Disbursements would be made against full documentation, except for payments against contracts of less than US$50,000 for goods and US$20,000 for consultants services which would be disbursed against certified statements of expenditure. Supporting documents would be retained by CFM(C), and made available for review by visiting Bank missions and by auditors. 3.36 To facilitate project implementation, a U.S. Dollar Special Account would be opened in a commercial bank, and operated and maintained on terms and conditions acceptable to IDA. The authorized allocation for  - 31 - the Special Account would be US$4.0 million, representing anticipated eligible expenditures for a period of about four months (excluding anticipated large payments which would be handled through letters of credit or direct payment). The Special Account would be replenished on receipt of properly documented withdrawal applications prepared by the BCA on a regular monthly basis promptly after receipt and reconciliation of bank statements from the commercial bank. 3.37 The estimated schedule of disbursements from the IDA Credit based on the assumption of effectiveness by January 31, 1990, is as follows: Disbursement Disbursement at the During the Year End of the Year (USSmillion) (USSmillion) Percenta#e FY90 4.0 4.0 1 FY91 12.0 16.0 40 FY92 10.0 26.0 65 FY93 6.0 32.0 80 FY94 5.0 37.0 93 FY95 2.0 39.0 98 FY96 1.0 40.0 100 L. ProJect Monitoring 3.38 DNPCF (and its successor CFM,EE) and CFM(C) will, with the assistance of the Project Coordinator, submit quarterly progress reports as per an agreed format covering, inter alia, the following aspects: (i) Progress of the project work with cost data; (ii) Training and technical assistance; (iii) DNPCF/CFM(C) financial performance; and (iv) CFM(C) performance in meeting the agreed operational targets. M. Budget, Accounts, Audit and Evaluation 3.39 DNPCF prepares annual operating and capital budgets before the beginning of eacn fiscal year which are submitted to the Ministry of Transport and Communications (MTC) and subsequently to the Ministry of Finance. DNPCF was assisted by consultants (RITES) in the preparation of their consolidated financial statements for 1986 and 1987. The accounts are audited by the Ministry of Finance rather than by external auditors and it was a pre-condition of the appraisal of this Credit that the consolidated financial statements audited by the Ministry of Finance be available. During project implementation, CFM(C) will provide the Association with audits of the annual financial statements prepared by independent auditors acceptable to the Association within 6 months of the completion of each financial year. At che same time, CFM,EE will also  - 32 - provide audits of its annual consolidated financial statements prepared by the Government Audit Department The auditors would also review records of amounts withdrawn on the basis of SOEs and their opinion would include a separate paragraph covering these withdrawals. No later than six months after Closing Date (June 30, 1996), DNPCF will provide the Association with a Completion Report which evaluates the operations, execution, costs and benefits of the project; the performance of CFM(C) and the Bank Group, and lessons learned. N. Environment 3.40 Since most of the IDA assisted components relate to rehabilitation of existing assets and provision of technical assistance, it is not considered that they will have an adverse effect on the environment. In the city of Beira, IDA is already contributing to improvement of the environment through the Urban Rehabilitation Project, which includes rehabilitation of the sewerage system, solid waste management and erosion control measures to return the coastline to its natural habitat. For the Beira Transport Corridor Program as a whole, BCA has provided assurances at negotiations that environmental aspects have been taken into account. In the port, the Program will lessen the risk of accidents and subsequent pollution through improved navigational aids and deepening of the access channel. In the railway, the diesel workshop modernization will improve safety and the rehabilitation and acquisition of locomotives will have a beneficial effect through the consequential phasing out of atmosphere- polluting steam locomotives. IV. ECONOMIC EVALUATION A. Introduction 4.01 The economic evaluation of the Beira Transport Corridor Program is based on an updated version of the traffic forecasting and allocation models developed by the Netherlands Economic Institute (NEI) for IDA's earlier study of the Beira Corridor (para 1.20), and the latest (1988) Southern Africa Transport and Communications Commission (SATCC) commodity- wise traffic forecasts by country. This combination provides commodity- wise traffic flows for the Beira Corridor and competing routes and gives the best estimates of the project's rate of return. The NEI methodology also forms the basis of a wider study being undertaken to review all of the Southern Africa Transport Corridors. This study, which is being undertaken by the Bank under the sponsorship of SATCC with funding from the Netherlands, USAID, Denmark ana Canada, reviews the financial and economic viability of all of the transport corridors in the SADCC region with a view to increasing their effectiveness through enhanced cooperation within the region. It is expected to be completed by the end of 1989. B. Traffic Forecast 4.02 The traffic forecast for the Beira corridor for the period 1989 to 2005 is summarised in Table 4.1. It considers future traffic for three scenarios: a) the entire Beira Corridor program, including the IDA  - 33 - hOZAMBIQUE BEIRA TRANSPORT CORRIDOR PROJECT Table 4.1:Beira Corridor Traffic Forecast (OOO's Metric Tons) A B C project without xIthout case IDA assisted all Components projects 1989 Mozambique 484 484 468 Zimbabwe (excl. POL) 274 274 0 Zimbabwe (POL) 770 770 770 Malawi 50 50 0 Zambia 100 100 0 Total 1678 1678 1238 1992 Mozambique 579 579 497 Zimbabwe (excl. POL) 829 437 0 Zimbabwe (POL) 836 836 836 Malawi 0 0 0 Zambia 120 0 0 Total 2364 1852 1;33 1995 Mozambique 727 727 572 Zimbabwe (excl. POL) 937 490 0 Zimbabwe (POL) 908 908 908 Malawi 377 0 0 Zambia 170 0 0 Total 3119 2125 1480 2000 Mozambique 892 892 724 Zimbabwe (excl. POL) 1048 454 0 Zimbabwe (POL) 1041 1041 1041 Malawi 411 0 0 Zambia 170 0 0 Total 3562 2387 1765 2005 MozambLque 1159 1159 902 Zimbabwe (excl. POL) 1162 521 0 Zimbabwe (POL) 1196 1196 1196 Malawi 468 0 0 Zambia 170 0 0 Total 4155 2876 2098 Source: SATCC, September 1988, Cotmod4ty-wise forecasts per country, WB 1988 mission, analysis of traffic allocation per corridor  -34- assisted projects (the project case); b) the Beira corridor program but without the IDA assisted projects (the without IDA projects case); and c) without any program (the without all projects case). This last scenario is hypothetical, since many Beira corridor projects are already under execution and many others are already financed by bilateral agencies. However, this scenario represents the base case against which (a) and (b) will be evaluated and allows, therefore, an assessment of the overall economic viability of the corridor project with and without the rail project components which are included in the proposed IDA assisted project. The traffic forecasts include traffic generated by Mozambique, Zimbabwe, Malawi and Zambia, by all modes in the corridor, i.e., rail, road and pipeline. 4.03 In the project case the total volume of traffic through Beira, in 1992, is estimated at 2.36 million MT (metric tonnes), including 0.96 million MT of petroleum oil lubricants (POL). This includes all Zimbabwe's POL imports plus approximately 50% of the remaining overseas trade from Zimbabwe (0.8 million MT, the other 50% going through Maputo), and approximately 120,000 MT of Zambia's forecast copper exports. Overseas trade of Malawi only starts using Beira again at a significant level after the Sena line reopens (in this analysis it is assumed that the Sena line will reopen in 1995). To compare the forecasts in Table 4.1 with the Beira port statistics it is necessary to convert the latter into MT. The 1987 throughput in Beira was 1.5 million MT (including POL), which are roughly 1.9 million port tons. 4.04 Without the railway project components included in the IDA assisted project, the total capacity of the railway system linking Beira with its hinterland is assumed to remain at the present (1938) level, i.e., around 400 to 500,000 MT per year. This assumes locomotive power availability through the present lease arrangements with NRZ or variations which will guarantee the minimum power required to transport those tonnages. Without both port and railway projects (i.e. the without all projects case), no international rail traffic could be expected, since the quality of service would be so poor as to divert any international traffic to alternative routes. Also, the traffic originated in Mozambique's central region would grow less rapidly, as reasonable inland railway connections are necessary if the growth rate forecast for Mozambique's export traffic in the project case is to be attained. 4.05 The traffic forecast has been prepared in four steps starting with total traffic generation followed by traffic distribution, modal split and traffic assignment to the several routes and modes. To follow this method it was necessary to: a) estimate the total external trade of Zimbabwe, Malawi and Zambia by commodity, which was done on the basis of SATCC forecasts; b) estimate the total overseas trade of these same countries by commodity; and c) estimate the total generalized transport costs per commodity by different routes (Beira, Maputo, Durban, Nacala). The generalized cost applied in this study includes railway cost, road cost, handling in ports, time cost of cargo, insurance, pilferage, and differences in ocean freight costs between the ports of exit. The allocation of traffic to routes took into account points of origin of the  - 35 - cargo within the hinterland, total generalized cost and availability of specific handling equipment in ports, specific agreed transport contracts and port capacity constraints (Beira, up to 1991). All Zambian traffic, except part of Zambian copper exports, have been allocated to the Dar-es-Salasm route. Insofar as national traffic is concerned the forecasts include a commodity-wise forecast of overseas import and export traffic from the central region of Mozambique and of the national coastal shipping traffic (cabotage). Details are given in Annex 6. 4.06 The most important factor in the allocation of traffic between the competing routes is the generalized cost of transport incurred in using them. This includes not only inland costs but also the differences in ocean freight rates between the ports of Beira, Maputo and Durban. Table 4.2 shows the difference in generalized costs between the Beira corridor and the other routes for containerized cargo. Detailed transport costs are given in Annex 6. Table 4.2 - Generalized Transport Cost: Beira vs. Other Routes (US$/MT) for Containerized Cargo at 12 MT/TEU From/To Beira Durban Maputo Maputo present 1992 Present & 1992 (via RSA) (via Limpopo) Present & 1992 1992 -------------------------------------------------------------------__----- To Harare 121 81 115 179 89 From Harare 100 61 105 138 73 ----------------------------------------------------------------------__--- Cost level in 1988 includes railway/road, handling in port, time cost, insurance, pilferage, difference in ocean freight. 4.07 The generalized cost approach produces a traffic allocation between the Beira and Durban routes which is quite close to the tonnages transported in 1987 (Table 4.3). Indeed, while direct out-of-pocket costs through Beira are cheaper than via Durban, the present generalized costs by the latter are lower and this explains why only 17% of Zimbabwean traffic (other than POL) goes via Beira while 55% goes via Durban.  - 36 - Table 4.3 - Actual and Estimated Cargo Flows 1987. Zimbabwean Overseas Traffic (exc. Rol) ________-_______________________________________________________________________ Total Estimate Based on the Generalizes (in 1000's Actual Distribution Cost Allocation of port tons) Beira Maputo* Durban Beira Maputo* Durban --------------------------------------------------------------__-------------__- EXPORTS} + } 1595 273 397 925 IMPORTS} 100% 17% 25Z 58% 171 281 55% .._____..________________________________________________________________________. *via RSA 4.08 Traffic allocation to the Beira corridor becomes more complicated when the interdependence between the Beira line and the Limpopo line (Zimbabwe to Maputo) is taken into account. Given the recent donor support to the rehabilitation of this line it has been assumed that it will be fully operational .n 1991, which is a conservative assumption from the point of view of the Beira corridor. If the Limpopo line is available, more Zimbabwean traffic will go through Maputo than in the present situation, since in 1988 the only route from Zimbabwe to Maputo is via the RSA. Table 4.4 shows the influence of the opening of the Limpopo corridor on the traffic forecast for the Beira corridor. The Limpopo line, if fully operational, would attract in 1992 some 230,000 MT (142 of total Zimbabwean traffic), which otherwise would take the Beira route. Most of the containerized cargo to/from Zimbabwe will go through Beira, most of the bulk (especially exports of minerals, steel and sugar) goes through Maputo. Table 4.4 - Traffic Allocation of International Traffic from Zimbabwe between the Beira Corridor and the Limpopo Line ----------------------------------------------------------------__--------- Both corridors open Beira corridor open, Limpopo closed 1992 Beira Maputo Beira Maputo Durban via RSA EXPORTS 401 60% 51% 20% 21% IMPORTS 76% 24% 95% 5Z 0% TOTAL 49% 51Z 63% 22% 15% ---------------------------------------------------------------------__----  - 37 - 4.09 The impact of the opening of the Limpopo line on the economic rate of return of the Beira corridor is significant. An important part of the benefits generated by the development of the Beira corridor consists in the savings in inland transport costs as compared to the least cost alternative "without the project". If the least cost alternative is the Limpopo line instead of the route to Maputo via RSA or to Durban, the inland transport savings are much lower (see Table 4.2). By developing both routes (Beira and Limpopo) at the same time, part of the potential benefits of the Beira corridor are attributed to the Limpopo route. Paras 4.13 - 4.19 below give the distribution of benefits by line and the economic rate of return of the program, for each of the scenarios. 4.10 The assumption that the Nacala line opens in 1991, and the Sena line only in 1995 is also disadvantageous to the Beira corridor, implying that no Malawi cargo will transit through the port of Beira between 1991 and 1995. From 1995 onwards, however, it can be assumed that the traffic on the Sena line will be consider3bly higher than on the Nacala line. If only the inland transport costs were taken into account, the Malawi traffic would be split approximately 50/50 between Beira and Nacala. However, the expected level of shipping services in Nacala is substantially lower than in Beira (no major Conference seriices) causing considerably higher ocean freight cost and time costs in Nacala for moving on liner vessels to North West Europe. Allocation of traffic between Beira and Nacala is presented in Annex 6. 4.11 As mentioned above the present and future level of shipping services in the ports involved is a decisive factor for the development of the corridors. At present, although the official tariff for the North Western Europe conference line vessels is equal for Durban, Maputo and Beira, the level of service in these ports is quite different; special tariffs can be negotiated in case of large shipments, such as the tariff negotiated by the Zimbabwe Tobacco Ass. for containerized tobacco exports through Durban. Also, "outsiders', i.e. vessels which do not belong to conferences, charge different tariffs for Beira, Maputo and Durban. Moreover, eastbound connections (Far East, Australia) and South/North America connections would, at present, require transhipment (at considerable extra cost) at Durban, if Beira is used. Discussions with shipping line managers suggest that when Beira port becomes fully operational with the expected handling rates and a minimum flow of about 6000 MT or 500 TEUs/week, the major differences in level of service between Beira and Durban will disappear. The same holds true for Maputo, if the minimum cargo flows are met. Nacala cannot be expected to reach a comparable level of shipping services for liner trade, irrespective of the port facilities available, because the overall level of traffic will remain rather low and the port lies outside the main shipping routes. 4.12 The distances between Zimbabwe and the port of Beira (Harare- Beira 605 km; Mutare-Beira 280 km) are such that a significant amount of road traffic might be expected when rehabilitation of the road between Machipanda and Beira is completed in the early 1990s. This prompted an analysis of the modal split in the Beira corridor, to forecast the traffic which mi&ht be diverted from rail to road. On the basis of generalized transport costs which include a two-week advantage for road transport, an  - 38 - empty return for trucks (502 load factor) and a US$10/MT road toll to reflect infrastructure costs, road traffic appears to be interesting for only a small proportion of the high value international cargo. This is partly due to the high cost of fuel in Zimbabwe, which weighs heavily in the tariff charged for heavy trucks. Another factor in favor of rail is that existing transport infrastructure in Zimbabwe is strongly based on transport by rail. Also, the Mozambique government has indicated that, in the short run (up to 1995), it will continue to regulate international traffic in the corridor until the railway manpower is developed and the present operational constraints are removed, allowing only selected cargoes to use the road. In the long term, when traffic is deregulated, road transport may become more attractive for cargoes originating in Zimbabwe and even in the Lusaka region (Zambia), because the Beira road route will be cheaper than the present Dar-es-Salaam road route. Without the project (i.e. with very poor rail and port service), it is forecast that the road would carry roughly 70,000 MTlyear of international traffic, in 1995. In the situation with the project (i.e. with greatly improved rail and port servi.ce), the volume handled by road will be 40,000 MT/year of international traffic plus 120,000 MT/year of local traffic. The latter has average haul distances of 200 km, over which road is cheaper than rail. To assess the impact of increased efficiency of road transport and eventual deregulation of traffic on future modal split, a drop of 30% in road tariffs was tested. The results show that road traffic would, in this case, increase to 225,000 MT in 1995 and 280,000 MT in 2005. The rail/road modal split on the Beira Machipanda route (i.e. without the traffic from the Sena line), changes from 96%:4% to 791s21% in 1995, if the traffic is deregulated and truck tariffs drop by 30%. These modal shifts do not affect the overall rate of return of the corridor. However, they affect the financial rate of return of the railway which, when setting its new tariff schedule, will need to take account of the effects of eventual international traffic deregulation on the modal split. C. Economic Evaluation 4.13 The economic evaluation takes into account all the costs of the Beira corridor program (investment, technical assistance, training, operating and maintenance costs). Details are given in Annex 6. Four types of benefits have been considered: a) inland transport benefits for diverted traffic, i.e. using the Beira corridor instead of other corridors; these benefits are calculated as the difrerence between total generalized transport costs for Malawi and Zimbabwe overseas trade in the 'without the programu and in the uwith the program" situations; b) waiting and service time benefits for normal traffic at the port: waiting time benefits resulting from reduced waiting time and costs in the Port of Beira for normal cargo, i.e. the total volume of cargo forecast in the "without the program' situation. The difference in waiting costs per ton between the "with" and "without" project situation is multiplied by the amount of "normal' traffic to derive the waiting time benefits. Service time benefits have been calculated using the NEI port simulation model, including the queuing model, for the port of Beira; c) waiting and service benefits for generated traffic; d) freight cost benefits for normal traffic due to the use of larger vessels or a better level of service (containerized cargo). No waiting time benefits or service time benefits  - 39 - are taken into account for diverted traffic, because it is reasonable to assume that port capacity in Maputo and Nacala is large enough to handle the additional traffic without longer waiting times than in Beira and handling rates at the three ports are the same. Costs and benefits are in November 1988 prices, including physical contingency and using the exchange rates at that time. 4.14 The Beira corridor program cannot be evaluated in isolation from other ongoing efforts to open the Limpopo and Nacala lines and, although somewhat later, the project to reopen the Sena/Moatize line, which links Malawi to Beira. In the "without project" situation such as it exists today (and assuming the Beira corridor program had not started), all Malawi and Zimbabwe traffic would have to use RSA ports (Durban) or Maputo via RSA railways, at high inland transport cost. Such a base case definition for the Beira corridor project, would imply large inland transport benefits to the traffic diverted from Durban and Maputo (via RSA) to Beira. If, however, the Limpopo line to Maputo opens in 1991, the least cost alternative to the Beira corridor will then be the Limpopo corridor. The remaining inland transport benefits of using the Beira corridor instead of the Limpopo corridor (diversion of traffic from Beira to Maputo) are much lower than the inland transport benefits of a Durban - Beira traffic diversion. The same is true for the Nacala-Beira diversion versus the Durban-Beira diversion for Malawi traffic. In calculating the ERR of the overall Beira corridor project, only benefits of the Maputo - Beira diversion and Nacala - Beira diversion have been attributed to the Beira corridor project. By doing so, in the present project base case, the majority of the inland transport benefits derived from Zimbabwe and Malawi overseas traffic are attributed to the Limpopo and Nacala corridors much earlier (1991 rather than 1995), which makes this evaluation quite conservative. Table 4.5 quantifies the impact of the base project case (case 4 in Table 4.5). Table 4.5 - Total Inland Transport Cost (US$ million)* for Zimbabwe and Malawi Overseas Trade 1995 traffic forecast, excl. POL Availability of routes Zimbabwe Malawi O Only Durban (D) and Maputo via RSA 148.7 72.0 1 D, Maputo (via Limpopo) only 90.0 72.0 (Durban) 2 D, Beira (via Machipanda+Sena) only 79.6** 18.4 3 D, Maputo (L) + Nacala, no Beira corridor 90.0 24.5 4 D, Maputo (L) + Beira (M + S) + Nacala 82.4 18.4 *Cost includes inland transport cost + time cost + insurance & pilferage + difference in ocean freight **Assuming steel, sugar and minerals could be handled in Beira  - 40 - 4.15 In the present evaluation, comparison is made of case 4 (all corridors, situation with the project) with case 3 (no Beira corridor, i.e. situation without project). In this case US$ 13.7 million are the cost savings attributable to the Beira corridor (11%) and US$106.2 million are the savings attributable to the Limpopo and Nacala corridors (89%). Table 4.5 also shows that, if case 2 is compared to case 4 no significant additional benefits would be produced, assuming that the port of Beira could handle all the traffic without capacity constraints. This, however, is not the case because Beira would not be able to handle all cargo without major new investments, whereas Maputo and Nacala can. 4.16 The proposed project has an economic rate of return of 14.1% and a net present value of US$ 55.3 million at a 10% discount rate. Annex 6 presents the detailed cashflow calculation. Given that the project was evaluated in a very conservative fashion by attributing the majority of inland transport benefits to the Limpopo and Nacala lines, it can be safely concluded that the project is economically sound. If the opening of the Limpopo and Nacala lines is assumed for 1993 instead of 1991, the ERR of the project is 16.7%. In this case, in 1992, with the alternative lines closed, the inland transport benefits of the corridor are as much as US$47 million (with the Sena line still closed). The conclusion is, therefore: 1) if the Limpopo ,nd Nacala lines are open to traffic as scheduled (1991), which is questionable given the security situation, the Beira corridor project is economically sound with a reasonable and conservatively estimated ERR for a project of this magnitude. It carries an important share of Zimbabwe traffic and (after 1995) Malawi overseas traffic; 2) if, for any reason the Limpopo and/or Nacala lines would remain closed, an efficiently managed Beira corridor project shows a very good rate of return, which increases dramatically for each year of postponement of the opening of the two other corridors. There is also the vital strategic role in the transportation of goods of Zimbabwe and Malawi which will be able to decrease their dependency on the RSA and this was not quantified because of the subjectivities involved. 4.17 A sensitivity test was performed in order to analyse the vulnerability of the project to substantially lower traffic volumes than presently expected (-25% international traffic than in the base case). This led to an ERR for the overall Beira corridor project of 12.5% instead of the original 14.1%. Therefore, even under those circumstances the project remains acceptable. 4.18 In order to evaluate the proposed IDA-assisted components it was assumed that railway traffic in the corridor would remain at present levels (400 to 500,000 MTlyear) without these components. This is a somewhat optimistic assumption because traffic would probably decrease from present levels if the additional motive power proposed is not available. However, the assumption is again justified on the grounds of obtaining a conservative estimate. Comparing the situation "with the IDA-assisted railway projects" with the situation "without railway projects" produces an ERR of 34.7% which is quite attractive for a project of this type. This result also shows how vital for the Beira corridor is the provision of motive power and technical assistance. A sensitivity analysis was performed to evaluate the impact of a 25% reduction in international treffic, to the base case. The ERR decreases from 34.7% to 28.3%.  - 41 - 4.19 The overall effect of a more aggressive road competition may cause a shift of traffic from rail to road and although the overall ERR for the corridor will be the same, the ERR of the railway projects will be lower. To test this scenario the effect of a further decrease of 25% in rail traffic was evaluated. The ERR obtained was 25Z which is still very healthy for such a project. Table 4.6 summarizes the results of the economic evaluation and sensitivity analyses. Table 4.6 - Beira Corridor - Economic Evaluation NPV @ 10% CASE TESTED ERR (US$ MILLION) 1. Overall program with Limpopo and Nacala open in 1991 14.1% 55.36 2. Same as (1) with 25% drop in international traffic 12.5% 33.29 3. Same as (1) but Limpopo and Nacala open in 1993 16.7Z 85.27 4. IDA assisted project 34.71 84.72 5. Same as (4) with 25% drop in international traffic 28.3% 59.98 6. Same as (5) with 252 further reduction of rail traffic 25.9% 50.02 V. FINANCIAL EVALUATION A. CFM(C) (a) Accounting System 5.01 CFM(C) maintains an accounting system on an accrual basis in sufficient detail to adequately present the financial operations of the railway and the port. Revenues earned by each operation are clearly identified. Working expenses which apply directly to each operation are so charged with joint or common costs being accumulated separately. The accounting system is, however, primarily oriented towards the preparation of the annual accounts rather than the accumulation of financial data for management purposes on a timely basis. The system is compater-based but the data processing center is currently not operational. CFM(C) has retained a team from Rail India Technical and Economic Services (RITES) to assist in the preparation of financial accounts.  - 42 - 5.02 CFM(C) does not maintain a traffic costing system. As a result, there are no clear data on which to Sase a tariff policy. During project preparation, bCA obtained financing from the EEC for a RailwaylPort Cost Based Tariff Study, the Terms of Reference of which were agreed with the Bank. The proposed project will provide funding for the technical assistance team needed to install the traffic costing system and train staff. 5.03 At present, CFM(C) has no Management Information System (MIS). During project preparation a needs study for a management information system was undertaken financed by PPF. The proposed project would provide funding for the technical assistance to install the MIS and train staff and for the necessary hardware/software. The action plan and the timetable to implement the recommendations of the PPF-financed study were reviewed and agreed at negotiations (para 6.01). (b) Past Performance 5.04 The combined Profit and Loss statements for port and railway operations are summarized in Table 5.1 below and shown in detail in Annex 7. For FY87 port operations resulted in a net profit while railway operations just about broke even with a combined profit for the two operations of US$7.2 millions. Table 5.1 - CFM(C) Profit and Loss 1987 Railway Port Total (US$ Million) Revenues 7.43 10.69 18.12 Working Expenses 7.70 2.33 10.03 Operating Surplus (Deficit) (0.27) 8.36 8.09 Depreciation 0.54 0.11 0.65 Other 0.10 0.11 0.21 Profit/(Loss) (0-91) 8.14 7.23 5.05 Prior to 1987, (CFM(C) had made a loss each year, partly due to the articially low official exchange rate of US$1 = Mt. 40 which prevailed prior to 1987 and which converted CFM(C)'s revenues, mainly foreign exchange earnings, into a relatively small amount of Meticais (or conversely, if the accounts are expressed in U.S. Dollars, inflated the main item of expenses, wages and salaries, out of proportion); but also due to the dramatic decline in traffic which occured in the late 70s and early 80s and is only now beginning to recover.  - 43 - 5.06 Port throughput declined from a high of 3.5 million tons in 1964 to 1.33 million tons in 1986 and then increased to 1.95 in 1987. However, port costs are generally quite variable as 60% of port personnel are stevedores who are mostly employed only when needed. Tonnage carried by rail also declined steadily from 5.9 million tons in 1973 to 0.3 million tons in 1984 but has gradually increased since then to 0.48 million tons in 1987. Unlike the port, the railway's costs remain relatively fixed in the short run with labor accounting for approximately 662 of the working expenses. 5.07 CFM(C) does not maintain a traffic costing system so it is not possible to determine which commodities are carried profitably, or which commodities contribute the most to the loss. However, using average figures, and assuming labor costs remain constant, the operating breakeven point for the Beira-Machipanda line is approximately 500,000 tons, which is considerably less than the forecast traffic of some 1.15 million tons for 1995. (c) Financial Forecasts 5.08 The financial projections assume that the IDA component of the investment plan will be on-lent by the Government to DNPCF/ CFM(C) at a rate of 7.65% over a period of 30 years including a grace period of 5 years. The projected operating statements for the years 1990-2005 are set out in Annex 7 . The financial assumptions used in the projections are shown in the annex. (d) Tariffs 5.09 On completion of the first phase of the program in 1991, it is expected that the capacity of the port and railway, handling speeds, transit times and reliability will improve to such an extent to permit selective increases in tariffs on the Beira corridor while remaining competitive with alternative routes. The current cost-based tariff study will help to determine the precise amount of tariff increases that are reasonable and it was agreed at negotiations that the first phase of adjustments would take place not later than December 1991 and would enable CFM(C) to achieve a 75% working ratio in FY1992, which would be sufficient to meet debt servicing requirements. In view of the established transport pattern for Zimbabwean traffic via RSA and the unsure timing within which this traffic can be attracted back to the Beira Transport Corridor, improvements in the working ratio in subsequent years will be agreed between BCA and IDA in the annual project implementation reviews. For the purposes of financial projections, an average increase of 20% has been assumed, which is well within the cost differential for the alternative Durban route (see para 1.04). On this basis the incremental financial rate of return is 12%. (e) Foreign Exchange Retention 5.10 Foreign Exchange for CFM-C. Lack of foreign exchange is a major handicap for the Beira Corridor operation despite the fact that CFM(C) is an important net earner of foreign exchange. This is due to the fact that all foreign exchange is kept by the Bank of Mozambique. CBM(C) needs  - 44 - foreign exchange to (i) purchase required fuel, spare parts, materials, replacement tools/small equipment, and (ii) consumption goods for workers so that for part of their wages they would have access to imported goods. In fact CFM(C) lacks personnel with the specialized experience to obtain foreign exchange rapidly from the Bank of Mozambique. To resolve this problem, a qualified member of CFM(C)'s staff has been appointed to the position of Procurement/Disbursement Officer to expedite foreign exchange withdrawals, for which specialised training will be provided during the period preceding Credit effectiveness (see Annex 10 for terms of reference). B. DNPCF 5.11 The financial statements for the three main elements of DNPCF, consisting of the transport corridors of CFM Norte, Centro and Sul, have been consolidated with the help of French Government-funded technical assistance provided by SOFRERAIL and are shown in Annex 7. It is estimated that DNPCF generated a surplus of some US$7.2 million in 1987, which is expected to rise substantially in future, as also are debt service payments. However, there are problems of non-compatibility of accounting methods between the three railway groups and differences in the basis of calculating interest expenses. The ongoing French technical assistance program will help improve the accounting base and advise on the establishment of CFM,EE (see para 1.19). During negotiations assurances were obtained that CFM,EE will adopt policies that will lead to its financial viability, .ncluding achieving a working ratio of 75% in FY 1995, the financial target in subsequent years to be determined in agreement with IDA in the light of progress on the Nacala and Maputo Transport Corridors (para 6.01). VI. AGREEMENTS, CONDITIONS AND RECOMMENDATIONS A. Agreements and Conditions 6.01 During negotiations, agreement was reached with the Government of Mozambique (GOM) and BCA, on the following items: (i) The terms and duration of contracts for the key management posts, including a Project Coordinator, which are crucial to the successful operation of CFM(C) (paras 3.10 - 3.12, Annexes 3 and 10); (ii) A manpower development program for CFM(C) and an action plan to implement it (para 3.10); (iii) An action plan and timetable to design and implement an incentive scheme for CFM(C) workers not later than December 31, 1990, this scheme to be developed with technical assistance financed by IDA (para 3.17); (iv) An action plan and timetable to implement the recommendations of the MIS study (para 5.03); (v) Operational targets for port and railway (Annex 5);  - 45 - (vi) Utilization of locomotives by CFM(C) (para 3.25)1 (vii) Criteria for future investment proposals and the current composition and phasing of the investment program in the Beira Corridor (para 3.03). (viii) Selective adjustment of raillport tariffs in CFM(C) as improved services become operational, the first phase of adjustments to take place not later than December 1991 to enable CFM(C) to achieve a working ratio of 75% iA FY1992, (para 5.09); (ix) Promulgation of the necessary administrative and legislative measures establishing a legally and financially autonomous entity (CFM,EE) to succeed DNPCF and transforming the CFM South, Central and North systems as separate sub-entities with independent costing, financial, accounting and management units by December 31, 1989. (para 1.19); (x) DNPCF (and its proposed successor CFM,EE) to adopt policies that will lead to financial viability, including a working ratio of 75% in FY1995, the financial target in each of the subsequent years to be determined in agreement with IDA (para 5.11); (xi) Finalization of a transit agreement with Zimbabwe, by June 1990, for the Beira Corridor (para 2.12); (xii) PropJsals to IDA, by December 31, 1990, to formulate feasible ways to increase private sector participation in port and railway related activities such as warehousing and container freight stations (para 2.02); and (xiii) Annual Project implementation reviews to be jointly conducted with IDA by Sep;ember 30 of each year during the execution of the Project (para 3.31). 6.02 Conditions of Effectiveness are: (i) . signed legal agreement (The Beira Corridor Implementation Agreement) specifying the mutual obligations of all agencies involved in implementation of the IDA assisted project components (para 3.33); and (ii) A signed subsidiary loan agreement between the borrower and DNPCF (para 3.33) B. Recommendation 6.03 Subject to the above assurances, the proposed project forms a suitable basis for granting an IDA Credit of SDR 31 million (US$40 million) to the Peoples republic of Mozambique. AF6IN July, 1989  - 46 - BtlFA IPRNANSPOF'P CORRIDOR P1OIJELT .i-EX I REsRA CCRRt - - - 1987 [-- 1989 1990 1991 1992 1993 PRA-M-A ORGANIZATION PLAN 4 FI - - PR-,-18 FINANCIAL. PLAN PR-t-.IC I4NAGENNT SYSTEM PR-N-2S WOEM ANPRTFN -T PR-M-2C WOWERS HOUSING 0- : . .... ....... PR-1-3 BEIRA TOWN SruoY I PR-N-4 COPE STAFF OCA PR--5 (Hi HOUSING COMPLEX ! PR-N-S (Ell E)ERGENCY POWER 4 PR-N-5 (TI) PROCURENENT FUND P-TA-I TECH. ASSISTANCE I * P-A-1 CHANNEL DREDGING I I ; P-A-2 NAVIGATIONAL AIOS I-1 P-CE-1 (A) CONTAINER BERTHS | ; P-CE-1 (B) WCKSHOP i4 P-CE-I (C) TRANSIT SHEOD P-CE-I (D) OFFICE - C P-CE-2 OIL TER14 NAL ..-4 I . P-CE-3 POR' RAILWAYS I j 4 P-CE-5 COLD STORAGE P-CE-SO PORT FACILITIES P-CE-13 CONTROL CENTER 4_ F c. P-CE-14 POWER SUPPLY '- . -. - P-NE-IA EQUIPNWNT REPAIR ; P-E-SB NEw EQUIPMENT I P-E-IC CONTAINER CRANES - - * :, _ _ ._ P-lE-2 TUGSi PILOT BOATS P-NE-3 HANDLING EOUIPNENT P-ST-s TELECON NETWORK -- - R-TA-1 TECH. ASSISTANCE _ R-CE-1 MACHIPANDA LINE ._ - 0-4 R-CE-4 SLEEPfER . .S.. *- .... R-CE-4 QUARRY . .i . . .c..... ....i.... R-CE-4 SENA LINE i I I ."'-..A l O 11 1 P-CE-7 MAINTENANCE EQUlfmENT I.. A R-NE-I WAGON RENAB. C f O R-NE-2A LOCOMOTIVE REHAB. I *R-E-20 NEW LOCONOTIVES I I * I * R-NE-4 LOCO WORKSHOP EQUIPWET z - | R-NE-7 RESCUE CRANES | | I R-ST-I TELECOM REPAIRS R-ST-2 RADIO TELECOM SYSTEM R-ST-3 TRAIN CONTROL SYSTEN .....I.... R-SI- WAGON CONTROL SYSTEM |( R-ST-5 TELE4ONE EXCHANGE I RD-aE-IAI ROAD REPAIR-SOfALA RO-CE-lA2 ROAO REPAIR-MANICA RD-CE-sE MAINTENANCE UNIT I I r*-CE-lC ROAD RECONSTRUCTION j T-CE-2 NICIWAVE LIN W | RE-I ANO 3 BEERTH REPAIR - I . ~ ~ ~ ~ __ __ .1 __ _. _ , ct~c