Madagascar - Railway Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 5434 PROJECT PERFORMANCE AUDIT REPORT MADAGASCAR FIRST RAILWAY PROJECT (CREDIT 488-MAG) January 29, 1985 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the perforuance of their oflicial duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT MADAGASCAR FIRST RAILWAY PROJECT (CREDIT 488-MAG) TABLE OF CONTENTS Page No. Preface.............................................. .............. i Basic Data Sheet.............................. ...............0...... ii Highlights -....... .. ................... ... .... ... ............. .. iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. PROJECT BACKGROUND ....... ...... ......... .e ... .oo ...... .. 1 II. PROJECT IMPLEMENTATION .................. ............. 6 III. POINTS OF SPECIAL INTEREST .... ............. ............... 12 IV. CONCLUSIONS .............................. ............... . 15 Attachment: 1. Transportation Policy and Action Plan.................... 19 2. Borrower Comments.... o .................. ....... ...... 31 PROJECT COMPLETION REPORT Summary ....... .. ........... 0........0... ............0...... 33 I. Introduction................................ o ............... 35 II. Project Preparation and Appraisal......................... .36 III. Implementation and Costs........ .. ........ ...... .. - ..... 38 IV. Traffic and Operations.......... o ... ...................o 43 V. Financial Evaluation.............. .... ..... ...... 46 VI. Institutional Development.................... ............ 49 VII. Economic Reevaluation.................................... 53 VIII. Role of IDA................... . ............ ...... 57 IX. Conclusion.......... . ............................. .58 Tables 1. Project Cost and Financing................. o .............. 59 2. Project Implementation.................................... 60 3. Comparative Traffic Statistics................ ....... 61 4. Comparative Income Statements.......... ............. 62 5. Comparative Balance Sheets.................. ....... 63 6. Selected Operating Statistics....... ........ ........... 64 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. PROJECT PERFORMANCE AUDIT REPORT MADAGASCAR FIRST RAILWAY PROJECT (CREDIT 488-MAG) PREFACE This is the performance audit report of the Madagascar First Rail- way Project (Credit 488-MAG) of US$6 million approved by the Board in January 1974. The Credit was declared completed in July 1979, the last disbursement took place on November 19, 1979, leaving a balance of US$764.41 which was cancelled. This document consists of a Project Performance Audit Memorandum (PPAM) prepared by the Operations Evaluation Department (OED) and of a Project Completion Report (PCR) prepared by the East Africa Regional Office. OED studied the project documentation in the Records Center and reviewed the PCR against the Appraisal and President's Reports, the legal documents, and the transcripts of the Executive Directors' meeting which con- sidered the project. Discussions were held with Bank staff on project formu- lation, implementation and follow-up. OED has found that the PCR provides an adequate record of the experience under the project for all components except the one intended to assist the Government's transport sector planning capacity. The audit elaborates on this topic as well as on the problems associated with railway autonomy. The draft report was sent for comments to the Borrower, whese replies indicating that they had no comments, appear as Attachments 2 and 3 to the Project Performance Audit Memorandum. - 1i - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET MADAGASCAR FIRST RAILWAY PROJECT (CREDIT MAG-488) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$ million) 8.9 11.8 Overrun (%) - 34 Credit Amount (US$ million) 6.0 6.0 Date Physical Components Completed/a 06/76 07/79 Proportion Completed by above date (7) 100 52 Time Overrun 3 years Economic Return 35% Less than 10% Financial Performance Good Poor Institutional Performance Good Poor CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS FY74 FY75 FY76 FY77 FY78 FY79 FY80 (i) Appraisal Estimate 2.0 3.9 6.0 6.0 6.0 6.0 6.0 (ii) Actual - 2.2 3.0 4.5 5.2 5.8 6.0 %of (ii) to (1) - 57 52 75 87 97 100 OTHER PROJECT DATA Original Actual or Item Plan Revision Est. Actual First Mention in Files - - 12/03/70 Negotiations 11/12/73 11/20/73 11/20/73 Board Approval Date - - 01/15/74 Credit Agreement Date - - 06/27/74 Effectiveness Date - - 09/26/74 Closing Date - - 07/31/79/b Borrower RNCFM Executing Agency Ministry of Transport Fiscal Year of Borrower January 1 - December 31 Follow-on Project Name Madagascar Second Railway Project Credit Number 903-MAG Amount (US$ million) 13.0 Credit Agreement Date 06/25/79 Ja The project was scaled down. 7b- Date of final disbursement 11/19/79. - iii - MISSION DATA Month/ No. of No. of Date of Item Year Weeks Persons Manweeks Report Identification 04-05/71 2 2 4 09/21/71 Preparation 08/72 1 1 1 09/21/72 Preappraisal 10/72 1 1 1 11/15/72 Appraisal 01/73 3 3 9 03/14/73 Post Appraisal 10/73 1 2 2 11/09/73 Total 8 9 17 Supervision I 01/74 1 1 1 03/11/74 Supervision II 07/74 1 1 1 09/05/74 Supervision III 01-02/75 1 2 2 05/30/75 Supervision IV 09/75 1 2 2 11/24/75 Supervision V 01/76 2 1 2 04/14/76 Supervision VI 09/76 1 2 2 11/12/76 Supervision VII/a 03/77 1 3 3 04/29/77 Supervision VIII/a 10-11/77 1 1 2 02/21/78 Supervision IX/b 04/73 3 3 9 05/24/78 Supervision X/b 09/78 1 1 1 10/05/78 Supervision KI/b 12/78 2 1 2 01/11/79 Supervision XIT7b 04/79 1 1 1 06/07/79 Supervision XIiI/c 06/79 1 2 2 08/02/79 Supervision XIVic 02/80 2 3 6 04/21/80 Completion/c 09/80 2 1 2 10/20/80 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Malagasy Francs (FMG) Year: Appraisal Year Average 1973 Exchange Rate: US$1 = 215 1974 US$1 = 223 1975 US$1 = 224 1976 US$1 = 249 1977 US$1 = 236 1978 US$1 = 209 1979 US$1 = 201 /a Also for preparation of Second Railway Project. 7 Also appraisal Second Railway Project. /c Mainly Supervision Second Railway Project. - iv - PROJECT PERFORMANCE AUDIT REPORT MADAGASCAR FIRST RAILWAY PROJECT (CREDIT 488-MAG) HIGHLIGHTS The bulk of Madagascar's 10 million inhabitants live in the quarter of the island's surface comprising the Tananarive and Fianarantsoa provinces. In the early 1970s, plans for the construction of a paved road from Tananarive to Tamatave suggested that the 375 km main line of the national railway (RNCFM) would soon face serious competition. At the time, RNCFM enjoyed a strong advantage on the Tananarive-Tamatave corridor over competi- tive transport using the unpaved road but required extensive improvements of its infrastructure, rolling stock, and operAtional practices. RNCFM's major handicap was that, since 1965, it had been operating as an ordinary Govern- ment department. The Association had been assisting the transport sector since 1966 (PPAM paras. 2 and 3). Both within RNCFM and in the Ministere de 1'Amenagement du Territoire (MAT) which was responsible for the development of the transport sector as a whole, there were shortages of experienced staff and very little systematic planning (PPAM paras. 5 and 8). The Association felt that although transport investments were indispensable, they ought to be accompanied by training of Malagasy officers who, over time, would be able to undertake all planning and management functions. Similarly, commercial prac- tices and cost-based tariffs would permit economically justified competition between road and rail. Throughout the identification and preparation phases of the First Railway Project in 1971-73, the Association argued, and the Government did not disagree, that intermodal competition would be a viable policy only if RNCFM were granted financial autonomy. The First Railway Project was a holding operation to aid RNCFM in replacing outdated equipment, and in determining the long-term role it could play within the transport sector. The project also included technical assis- tance for MAT so that Government could map out development requirements for the entire transport sector and part of this assistance was for training of Malagasy officials in transport planning and coordination. During negotia- tions, the Government seemed to agree with the overall approach proposed by the Association and gave assurances it would support its implementation (PPAM para- 15). Project components are distinguishable into (a) Items for RNCFM, and (b) Technical assistance for the Ministere de l'Amenagement du Territoire (MAT). Items for RNCFM included works, procurement of rolling stock, and consulting services. The 62 km track renewal program was completed with 2- year delay and increased from an app.-aisal estimate of FMG526 million to FMG644 million, due largely to implementation delays (PCR, paras. 3.08 and 3.09 and PPAM paras. 19-22). The planned procurement of rolling stock involved the purchase of 50 freight wagons and 20 passenger coaches with an -v - estimated combined cost of USS2.14 million. Bid prices were much higher than anticipated and fund shortages led to the actual procurement of 35 wagons and 5 coaches only. Deliveries were a year late because the Ministry of Finance delayed the award of contracts (PCR, paras. 3.12-3.14 and PPAM paras. 23- 25). Consulting services involved the provision of technical assistance to RNCFM in two phases. Phase I, which was satisfactorily carried out, focussed on improvements of management, operations, and financial policies and prac- tices. Phase II was planned to identify investment needs, rationalize services, refine commercial policy and tariff structure, and train staff but was not implemented because the question of RNCFM financial autonomy remained unresolved (PCR, 3.16 and PPAM paras. 26-28). Technical assistance for MAT should have supported transport planning and coordination and train Malagasy nationals for this work (PPAM para. 29). The team did not achieve its objec- tives because of the Government's lack of commitment to systematic planning (PCR, 6.10-6.12 and PPAM para. 30). Total project costs incroased from an estimated FMG1900 million to FMG2552 million (PPAM para. 42). Because of cost overruns the project was reduced in scope and completed with a 3-year delay. The economic reevalua- tion in the PCR states (PCR para. 7.03) that the project as a whole was mar- ginal (witl: an economic rate of return reestimated at less than 10% (PPAM, paras. 43 and 56)), even though the economic rate of return estimated at appraisal had been 35% (SAR, para. 5.09). The covenant requiring RNCFM to obtain a 2.5% return on net fixed assets through 1976 and 3% thereafter was not observed but the Bank took no actLon (PPAM para. 57). The audit has concluded that where RNCFM enjoyed relative freedom of action (track strengthening, improvement of data collection) the results were reasonable. Where RNCFM had to operate within bureaucratic or political constraints (tariff setting, revenue recovery from other Government depart- ments) and in areas transcending railway operations (national transport plan- ning, training of Malagasy staff) achievements were limited. The failure of management, institution-huilding and policy objectives was due to Govern- ment's reluctance to share the Association's commitment to economic effici- ency and free competition. The Association was aware of the Borrower's reservations but chose to adopt, wrongly in the audit's view, a conciliatory attitude which led to the Second Railway Project of 1979 (PPAM paras. 56-58). The audit acknowledges that both the First and the Second Railway Projects benefited the people of Madagascar by keeping the railway operation- al. However, the distinction is clear between development projects and deferred maintenance operations and the latter should not be mistaken for the former (PPAM para. 59). The main lesson is that the Bank should not launch projects aimed at economic efficiency and the encouragement of competition when the Borrower has reservations about the desirability of such objectives. Perhaps the Association couli have done more to explain its developmental philosophy and to give careful attention as the Borrower set forth his own views on equity and growth. This may have revealed areas where the two sides might have col- laborated productively. The absence of such a dialogue brought about the unsatisfactory results of the First Railway Project (PPAM, para. 60). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM MADAGASCAR FIRST RAILWAY PROJECT (CREDIT 488-MAG) I. PROJECT BACKGROUND 1. Madagascar, the world's fourth largest island (590,000 sq. km.), lies 400 km. off the east coast of Africa. Most of the 10 million inhabi- tants, 90% of whom are engaged in agriculture, live in the Tananarive and Fianarantsoa provinces, located in the central plateau. The island was granted independence in June 1960, underwent a revolution in 1972, withdrew from the French franc zone in 1974, and underwent a second revolution in 1975. Since 1975 the Government has pursued policies intended to foster self-sufficiency and to increase State control of the economy. 2. The Bank Group's first lending operation was the Highway Recon- struction Project (Credit 90-MAG, US$10 million, 1966), to help build two sections of the Antananarivo-Mahajanga Road. An OED Project Performance Audit Report (No. 1409, dated January 3, 1977) found that the project was successful, that road construction costs were below appraisal estimates, and that the recalculated rate of economic return was greater than estimated at appraisal. The Second Highway Project (Credit 134-MAG/Loan 570-MAG, US$8 million, 1968) helped finance road and bridge construction. Contractors encountered difficulties and this delayed completion by about a year. There was a cost overrun of about 30% and the OED Project Performance Audit Report (No. 811, dated July 18, 1975) found that the economic Justification of one of the roads was doubtful in view of the low (8%) recalculated rate of econo- mic return. The Third Highway Projct (Credit 351-MAG/Loan 876-MAG, US$30 million, 1973) provided for construction of primary roads, for detailed engineering studies and for a review of the traffic counting system. Sub- stantial cost increases led to the modification of the project in September 1975, resulting in a supplementary credit of US$5.6 million and a reduction of the total road length to be built. The OED Project Performance Audit Report (No. 2143, dated July 27, 1978) found that while the overall recalcu- lated rate of return (14%) was acceptable, two roads showed rates below 10% because anticipated agricultural benefits did not materialize due to the deteriorating political and economic situation. The OED Report discussed project deficiencies, including the insufficient funding for road maintenance and the failure to train Malagasy nationals. Bank Group assistance to the highway sector continued under the Fourth Highway Project (Credit 641-MAG, US$22 million, 1976), the Fifth Highway Project (Credit 938-MAG, US$24 million, 1979; plus a US$10 million EEC Special Action Credit), while the improvement of feeder roads was assisted through a number of agricultural and rural development projects: the Lake Alaotra Irrigation Project (Credit 214- MAG, 1970), the Morondava Irrigation and Rural Development Project (Credit 322-MAG, 1972), the Mangoro Forestry Project (Credit 525-MAG, 1974), and the Village Livestock and Rural Development Project (Credit 506-MAG, 1974). - 2 - 3. Outside the road sector, Bonk Group assistance was extended to maritime transport through the Tamatave Port Project (Credit 200-MAG, US$9.6 million, 1970) which financed the extension of port facilities, the creation of the Toamasina Port Authority and the provision, through technical assis- tance, of management personnel and of staff training. The OED Project Performance Audit Report (No. 2229, dated December 22, 1978) concluded that: (a) physical targets had been satisfactorily met, (b) the recalculated rate of economic return was 7%, compared with 15% at appraisal, (c) traffic stag- nated due to socio-political conditions, (d) the Institution-building efforts had not been successful, and (e) the financial performance of the Port Authority had been adversely affected not only by the declining traffic but, also, by the increasing labor costs which political developments had forced the port to absorb. The Report concluded that financial autonomy had not been achieved and was not likely in the near future. 4. By 1971, the Bank Group had been assisting transport for almost six years and Bank staff were familiar with sectoral strengths and weaknesses. The condition of the national railway (RNCFM) suggested a new area deserving attention and an IDA mission in April 1971 identified a possible railroad project. The railroad provided Tananarive, the capital and economic center of the country, with its only reliable access to the main port at Tamatave. The mission concluded that the railway continued to survive because of the near-monopoly it enjoyed over the Tananarive-Tamatave transport corridor. However, an all-weather road between the two cities was planned for comple- tioi by 1977 and from then on road and rail would have to compete on an equal footing. In anticipation, measures had to be taken to increase the capacity of the rail system and to reduce its operating costs. 5. The estimated total cost of the project identified in April 1971 was US$16 million equivalent and represented the first three years (1972-74) of the RNCFM Five-Year Plan. In addition to infrastructural renovations, modernization of rolling stock, and operational improvements, the Bank mis- sion recommended a number of policy measures which would permit RNCFM to com- pete effectively with road transport once the Tananarive-Tamatave road had been completed. The proposed measures aimed at the strengthening of inter- modal competition and included (a) granting RNCFM autonomy to pursue rational commercial, financial, budgetary, employment and investment practices; (b) implementing a tariff reform with a view to establish a competitive and cost-oriented tariff structure; (c) rationalizing of railway operations; (d) ensuring that trucks using the Tananarive-Tamatave road would pay for its maintenance; and (e) establishing a time-table for the gradual elimination of entry restrictions in the trucking industry. 6. Between July and November 1971 the Government and the Association corresponded on the proposals of the Identification Mission and the record shows the Government's reluctance to consider any drastic measures related to competition, tariff reform and the development of the trucking industry. Granting autonomy to the railway presented particular difficulties which were known to the Association. RNCFM had been an autonomous entity in 1965 but Parliament transformed it into a Government department and when, at the end of 1971, a bill was introduced to restore autonomy to RNCFM, Parliament voted - 3 - it down. After the May 1972 government reorganization, RNCFM was made a department of the Ministere de l'Amenagement du Territoire (MAT). 7. Despite the Covernment's unreceptive attitude, the Association con- tinued with project preparation. However, the report of a pre-appraiRal mission which visited Madagascar in August 1972 confirmed earlier negative findings and concluded that: (a) RNCFM was operating in a legal vacuum since all decrees relevant to its status and mode of operation had been abolished; (b) RNCFM discipline and productivity were suffering because of the prevailing uncertainty over the future of the railway; (c) Financial management and accounting were particularly weak, and; (d) Continuing operation of the railway as a Government department would imperil its survival as a well-functioning transport enter- prise. The mission, feeling that autonomy was crucial, discussed this issue with the Government and established unequivocally that this was a political rather than an operational matter. Even so, and without any indication that autono- my could be made politically acceptable, the mission recommended that appraisal be conditioned on the Government's agreeing to the railway's finan- cial independence. It is significant that the mission was not able to discuss road/rail competition because no one on the Government side was will- ing to explore this question. 8. At the time of the pre-appraisal mission in August 1972, RNCFM had extended its investment program to cover the 1972-76 period, had increaged the cost estimate of the program to USS32.2 million equivalent and hoped that the Association would contribute about USS15 million. Two months later, In September 1972, a Highway Projects Supervision Mission reported that the RNCFM Investment Plan had been further increased to about US$39 million and that it was hoped that the Bank would provide half of that amount. These successive changes suggested that investment programs were, in substance, little more than shopping lists. 9. Even though RNCFM was not granted autonomy, the Association in due course decided to proceed with Project Appraisal. This took place in January 1973 and covered a much smaller scope than originally envisaged: total esti- mated cost was about US$9 million equivalent and the proposed Bank credit was US$6 million. Scaling down was due partly to the lack of preparation and partly to the still unsettled conditions prevailing in the country and dem- onstrated by a drop of almost 20Z in freight traffic between 1971 and 1972. The proposed project consisted of relaying of 60 km of track, procurement of freight wagons and passenger coaches, and technical assistance to RNCFM and to MAT. The technical assistance components are noteworthy because they formed the centerpiece of Bank efforts for institutional development. Agree- ment was eventually reached with the Government that there would be two distinct groups of expatriate staff. The first would provide -4- cechnical assistance to RNCFM and would concentrate on setting up railway accounts on a commercial basis and on developing an investment program. The second group would assibt MAT in overall transport planning. Although MAT itself agreed it would he desirable to have such assistance, the Planning Secretariat (Secretariat d'Etat au Plan) voiced reservations about relying to such an extent on expatriate personnel and expressed a preference for direct assistance by Rank staff through frequent missions. 10. The March 14, 1473 Issues Paper summarized as follows the project rationale and objectives as well as the obstacles likely to be encountered during Implementation. Rationale 11. (a) RNCFM plays a vital role in Madagascar's economy. Its main line (375 km) is the only reliable means of surface transportation between Tananarive and Tanatave. If a paved road between the two cities is built, the railway will lose traffic but in the meantime it will have to continue transporting virtually all imports and exports. (b) Tnstufficient rolling stock, sharp grades, narrow curves and termi- nal problemn have been making it difficult for RNCFM to meet traffic demand. A more enlightened tariff policy could lead to better wagon loading factors and to other improvements in railway operations. Preliminary figures for 1973 show a sharp decline in tntal traffic but demand is expected to pick up and grow by about 57 a year. Notwithstanding any possible operational ia- provements, RNCFM would still need additional roll:ng stock. (c) The process for Pranting operational autonomy to the railway is underway: a draft decree has been prepared and enactment is expected short- ly. However, RNCFM requires consultancy services to improve its accounting, management and planning practices. (d) For the transport sector as a whole, Investment priorities and transport polIcy are still unclear. Long-term investment planning is not being practiced. FAT has almost no qualified staff in transport planntng and coordination and requires expert assitance. Project Objectives 12. Given the uncertainties of the future role of the railway and the lack of technical and economic project preparation, RNCFM agreed with the Appraisal Mission that the 1973-75 Investment Plan be reduced to cover strictly necessary and Immediate needs. Accordingly, the Issues Paper pro- posed that the project would provide RNCFM with the minimum amount of infra- structural improvements and additions to ro:ling stock to enable it to carry projected traffic through 177: it would improve RNCFM managerial and opera- tional efficiency through the provision of consulcancy services; it would finance a Ludy of the railway's long-term prospects; and it would finance expatriate experts to assist MAT in transport planning and coordination. 13. Potential problems were foreseen as follows: (a) Some opposition could be expected regarding the technical assis- tance elements of the project. (b) The railway might not be able to meet the proposed financial re- quirements. During 1973-75, RNCFM was to contribute 40% of Its Investment Plan costj and to meet its debt servicing obligations. The Government ought to be asked for guarantees that it would cover any shortfalls in the event that the railway could not generate sufficient funds internally. (c) The autonomy question might not be resolved. Credit negotiations should take place only after the Association was satisfied that the autonomy issue was being satisfactorily resolved. 14. The April 18, 1973 Decision Memorandum noted that Government had prepared a draft decree which would transform RNCFM into a public enterprise with operational and financial autonomy. This draft would be reviewed by the Association before deciding whether autonomy was still an issue. The Memorandum also recorded that loan negotiatiols would take place only after the Association received assurance that acceptable consultants would be appointed for the improvement of RNCFM managerial and operational efficiency. The project was processed on that basis and negotiations took place in November 1973. 15. The November 27, 1973 note to files, written by Association staff upon the conclusion of Negotiations of the Railway Project records agreement reached between the Government and the Association on traffic forecasts, modifications of the 1973-75 RNCFM Investment Plan, and items to be included in the project. Important points specifically addressed during negotiations were: (a) Increase in revenues Although earnings after 1972 were poor, RNCFM expected them to improve from 1974 on and Covernment agreed to reimburse the railway for revenue losses incurred because of concessionary fares. How- ever, this would not provide RNCFM with sufficient net operating income to ensure an annual rate of return on net fixed assets of 2.5% through 1976 and 3% in 1977 and RNCFM and the Government agreed to increase tariffs by an average of at least 52 as of April 1, 1974. (b) Management and accounting consultants Because of the delay in inviting proposals, the appointment of con- sultants would be a condition of effectiveness. (c) Decree setting up RNCFM as an autonomous body The enactment of the decree would be a condition for Board presen- tation. - 6 - (d) Government attitude regarding RNCFM The Government indicated it would support the railway so as to keep it going as a financially viable concern. To this end, it would take measures to create the required commercial climate and to develop the necessary traffic patterns. 16. The Bank's Board of Directors approved the project on January 15, 1974. Starting in April 1974, successive letters from the Arsociation re- minded the Government that as long as the decree on RNCFM autonomy was not in force the Credit could not be signed. On May 17, 1974 the Government inform- ed the Association that the decree had been approved and the Cr!dit was signed on June 27, 1974. The Association's Legal Department judged that all conditions had been met and the Credit became effective on September 26, 1974. II. PROJECT IMPLEMENTATION Objectives 17. The project was conceived as a holding operation to aid RNCFM in renewing come track and replacing outdated equipment, to determine its long- term prospects, and to outline the development requirements for the entire transport sector (SAR, Summary, para. ii). While some track replacement and rolling stock procurement were accomplished, the project did not meet its other objectives, particularly relating to restructuring of the railway and planning of the transport sector. The main reasons were the unstable politi- cal climate, unsupportive Government policies and the stagnation of the economy (PCR, Summary, para. iii). Components 18. The project comprised the portion of the RNCFM 1973-75 Investment Plan for which financing had not been committed at the time of the November 1973 negotiations, plus technical assistance to MAT for transport planning and coordination (SAR, paras. 4.03-4.07). Project components are distin- guishable into (a) Items for RNCFM, and (b) Technical Assistance for MAT. (a) Items for RNCFM Civil Works 19. Renewal of rails and sleepers was planned on the Lohariandave- Andavibe section (62 km) of the main line, plus a new crossing station, extension of yards, sidings and workshops, and construction of a few build- ings, primarily for staff housing. 20. The invitation to bid for track materials was published in November 1973 but due to delays in transmitting copies to foreign embassies the sub- -7- mission of bids was extended to January 1974. All track materials were received by March 1975 (PCR, para. 3.08). 21. The track renewal program was scheduled for completion in 1976 but suffered a two-year delay. According to the railway, this was due to supply problems with small materials and to a shortage of wagons to carry ballast from the quarries. A reallocation of the credit was requested to purchase the needed wagons and this was approved by the Association in November 1977. According to the Association, works execution was hampered by poor organiza- tion and planning on the part of RNCFM and by major realignment works being done concurrently on the Ambilla-Brickaville section. The Association ques- tioned the priority of these realignoent works but RNCFM decided to complete them by force account (PCR, para. 3.09). 22. The total cost of the track renewal program amounted to FMG644 million, compared to an appraisal estimate of FMG526 million. The difference is mostly due to local cost increases resulting from implementation delays. The extent to which other items (new crossing station, extension of yards, etc.), and amounting to about 8% of the total project cost, have been carried out could not be established by the audit (PCR, Table 1). Rolling Stock 23. The project envisaged the purchase of 50 freight wagons and 20 pas- senger coaches with an estimated combined cost of US$2.14 million (SAR, para. 4.08). The lowest bid for wagons was US$33,000 against an appraisal estimate of US$20,000. Fund shortages led RNCFM to propose, and the Associa- tion to agree with a reduction in the number of wagons from 50 to 35. Delivery was about a year late mainly because the Ministry of Finance delayed award of the contract (PCR, para. 3.12). 24. The unit price in the lowest bid for passenger coaches was about US$200,000, as against the appraisal estimate of US$65,000. Under the cir- cumstances, the Association could finance only five coacheq out of the twenty originally envisaged and to make up the difference, the Caisse Centrale de Cooperation Economique (CCCE) agreed to finance ten additional coaches. The final contract was awarded with a 20-month delay in June 1976 and the coaches were delivered in November 1977 (PCR, para. 3.13). 25. Differences between estimated and actual prices of rolling stock are partly explained by steel price increases from 1973 to 1975, partly by an underestimate at appraisal, and partly by a difference in standards which the audit believes ought to have been firmly agreed upon during negotiations: the Association advocated simple coaches while RNCFM wanted more elaborate units (PCR, para. 3.14). Consulting Services for RNCFM 26. Consultants were to be engaged under the project to render tech- nical assistance to RNCFM. Phase I of these services would focus on improve- ments of RNCFM management, operations, financial and accounting policies and practices, the recommendation of tariff adjustments and the assessment of - 8 - manpower requirements. Phase II would identify RNCFM investment needs, rationalize its services, refine its commercial policy and tariff structure, and train staff (SAR, para. 4.06). 27. Consultants were appointed in April 1974 to carry out Phase I. Their final report was presented in August 1975, their recommendations were considered sound and were in general well accepted by the railway. However, the launching of Phase II was postponed for about a year because it had by then become evident that far from being resolved, the question of RNCFM autonomy was still pending. 28. At the end of 1976 the autonomy question was still unresolved and the Association agreed that, at least, studies included in the Phase II work program ought to proceed. The same consultants who had worked on Phase I were retained in February 1977 to work on these studies and to review Phase I recommendations related to personnel and accounting matters. Implementation of Phase I recommendations regarding far more crucial issues (operations, finance, management, cost accounting and data processing) were postponed to the Second Railway Project (PCR, para. 3.16). (b) Technical Assistance for MAT 29. The project provided for technical assistance to MAT in transport planning and coordination and in training Malagasy nationals for this work. During negotiations, the Government agreed that the expatriate staff would assist the Ministry to plan, review and coordinate transport studies; to for- mulate transport pricing and coordination policies; and to prepare sectoral investment plans (SAR, para. 4.07). 30. The technical assistance team was headed by a transport economist seconded by the Association and included three more expatriate experts whose services were made available under a contract awarded to a consortium of two consulting firms in December 1975 (PCR, para. 3.17). Although the team pro- duced some useful studies and trained some national staff, transport planning did not become integrated into governmental activity partly because of the Government's lack of commitment to the principles espoused by the Association (PCR, paras. 6.10-6.12) and partly because the Association was unwilling to insist that transport planning and coordination deserved the importance assigned to them during project preparation and negotiations. Forecasts and Actual Performance (a) Traffic and Operations 31. Due to the depressed economic conditions, general freight traffic remained at an annual volume of about 650,000 tons and in 1977 was approxi- mately 30% below appraisal estimates. In 1978-79, chromite traffic, an important element of the railway's business, dropped from 180,000 to 80,000 tons because of reduced ore sales (PCR, Table 7). Despite the decreased demand, and despite the additional freight wagons, RNCFM suffered from capa- city constraints. Tananarive began to experience shortages of essential - 9 - commodities and Government instructions concerning movement of priority goods disturbed the flow of railroad traffic. In 1979/80 the Government borrowed heavily abroad to finance imports and the railway found itself unable to meet the suddenly increased traffic demand (PCR, para. 4.01). Passenger traffic grew faster than expected and in 1977 was 22% above the appraisal estimate. Thereafter, the rate of passenger traffic growth slowed down to 4% primarily because of capacity constraints (PCR, para. 4.02). 32. Locomotive availability was sufficient during most of the project period but decreased in 1978/79 because the inferior fuel used generated serious maintenance and spare parts problems. Freight wagon availability was also sufficient but wagon productivity dropped in 1978/79 because problems with locomotives led to the cancellation of train departures. Passenger coach availability was low but traffic was accommodated by overcrowding pas- sengers and delaying the scrapping program. In 1978/79 passenger trains-kms decreased due to the shortage of locomotives. Staff productivity increased because railway personnel was reduced by attrition and because growth of passenger/kms led to an overall increase of traffic units (PCR, para. 4.03). 33. Thus, Operational performance was in general satisfactory under "normal" conditions. However, RNCFM could not react adequately when con- fronted with developments which required independence in action and a flexible, non-bureaucratic response: drop in fuel quality, traffic pressure, and increased political interference (PCR, para. 4.05). (h) Finance 34. RNCFM financial results did not come up to the levels forecast at appraisal. On the contrary, the financial situation deteriorated progres- sively and the railway failed to achieve the annual return on net fixed assets required under the Project Agreement (PCR, para. 5.01). The main reasons were freight traffic stagnation, higher operating costs, inadequate tariff increases (para. 15), and non-materialization of income from two fore- casted sources. 35. Traffic stagnation was the direct result of the depressed economic conditions (PCR, para. 4.01). Higher operating costs were primarily due to the steep rise of staff expenditures after 1976 (PCR, para. 5.02) and this in spite the concurrent reduction of personnel numbers. Tariff increases were implemented too late and on too small a scale (PCR, para. 5.03). Income from other sources (profits from CIBA, a RNCFM subsidiary, and reimbursements from the Government of concessionary fares) did not materialize (PCR, para. 5.04). 36. Although deteriorating, both the current and the debt-equity ratios remained strong during the project period. Nevertheless, the apparent strength concealed two expanding problems. In the first place, current assets and liabilities more than doubled during the 1973-78 period because Government accounts receivable continued to grow and this, in turn, forced RNCFM to delay payment of its own current accounts. In the second place, the debt-equity ratio deteriorated because of the increasing operating losses and - 10 - the additional borrowings such as the amounts secured through the Second Railway Project of 1979 (PCR, para. 5.05). (c) Institution-building 37. At the time of appraisal, RNCFM was run as an ordinary government department, subject to all civil service rules. At the urging of the Associ- ation, and as a condition of credit effectiveness, the railway was given the status of a state-owned Public Industrial and Commercial Establishment in January 1974. Other institutional shortcomings noted at appraisal were lack of coordination among RNCFM departments, lack of correspondence between accounting methods and commercial principles, and the defective personnel policies which made it difficult to attract and retain qualified staff. The project provided for consulting services to study these problems and make re- commendations. From the moment the project became effective the Association assumed that the decree setting up RNCFM as a "Public Industrial and Commercial Establishment" would give the railway operational and financial autonomy. Shortly after the consultants submitted their Phase I Report, in August 1975, it became obvious to the Association that the basic assumption was wrong (PCR, paras. 6.01-6.04). 38. The consultants made two principal recommendations: (a) RNCFM statutes should be modified to give the railway modern management methods, and (b) given the railway's special needs, a new personnel statute was required. RNCFM agreed with these recommendations and collaborated with the consultants tc draft the necessary statutes. However, the Ministry of Finance disagreed that any changes should be made and rejected the draft statutes. This led to the deferral of the substantive elements of Phase II and the consultants' work was limited to areas not related to the question of RNCFM autonomy (PCR, para. 6.04). 39. Further 4ncertainties surrounding the railway's future status arose subsequently when the Government proposed to apply a "Charte des Entreprises Socialistes" to all public bodies, including the railway. Then, in September 1976, the Government issued a document entitled "Politique et Plan d'Action dans le Domaine des Transports" (Attachment 1) which while explicitly favor- ing nationalization of transport enterprises and Government intervention in the operation of the entire transport sector, did not make any reference to RNCFM and consequently did not provide any useful guidance for the resolution of the railway's managerial, opcrational, personnel and financial problems. Nevertheless, the tone and implicit intent of these developments went against all agreements previously reached between the Government and the Association insofar as the status of RNCFM was concerned but the Association neither objected to nor questioned the rationale of the Government's positions. How- ever, the contradiction between what the Government and the Association wanted for RNCFM became only more stark after these developments; as para- graphs 6 and 7 show, the Government had from the beginning a different attitude from the Association about the proper role of RNCFM. The Associa- tion failed to explore the basis of this attitude and did not try to find an accommodation for it. - 11 - 40. The project also provided technical assistance to MAT (para. 29). While some achievements were recorded in making studies and training counter- part staff, critical decisions by the Government in the transport sector remained largely untouched by improved planning procedures (PCR, para. 6.10). The major cause of failure was the diver:ence between what the Government felt and what it had agreed to during negotiations. The Association and, by extension, the expatriate staff working at MAT, had to assume that agreements reached at negotiations represented the genuine posi- tion of the Government. However, as events proved, the Government was committed neither to granting autonomy to RNCFM, nor to promoting economic efficiency by implementing cost-based transport tariffs, nor to encouraging competition among the different transport modes. The audit cannot therefore agree with the assertion made in the PCR (para. 6.12) that "it is likely that most of the difficulties could have been avoided if at the outset of the study a senior Government official had been selected to supervise the team's work". (d) Economic Reevaluation 41. For the project as a whole, the rate of economic return was estimated at 35% at appraisal (SAR, para. 5.09). Individual rates of return for the major components (accounting for 79% of totRl costs) were estimated as follows: Track renewal program 20%+; Freight wagons 58%; Passenger coaches 40%. 42. Details of estimated and actual costs for the project as a whole (FMG1900 million and FMG2552 million respectively), are shown in Table 1 of the PCR. The audit finds it disturbing that actual costs for "other works" are unavailable and that local costs for consulting services to MAT increased from FMG42 to FMG183 million. The PCR text contains no satisfactory explanation either for the absence of data in the former case or for the quadrupling of costs in the latter. This suggests that appraisal estimates were off and that project reporting (PCR, para. 3.18) was not as thorough as it could have been. 43. The audit agrees with the statement made in the PCR (para. 7.02) that economic reevaluation cannot be done in precise terms because there are major discrepancies between the appraisal/actual benefits and because the railway did not operate under "normal" conditions during 1978-80. The audit also agrees with the PCR (para. 7.03) that the project was at best marginal and that in the case of certain components, such as passenger coaches, the recalculated rate of return is negative (PCR, para. 7.10). - 12 - III. POINTS OF SPECIAL INTEREST The Development Dialogue 44. The Association's developmental rationale is anchored on economic efficiency and its objectives were straightforward when, in April 1971, it proposed to extend assistance to the railways (para. 4). Since the Govern- ment lacked trained staff to formulate an overall transport policy, the Association proposed to include in the Railway Project resources for strengthening transport planning at the national level. Reasonable as this approach may have been, no account seems to have been taken of the Govern- ment's attitude which did not assign the same importance to economic effi- ciency as did the Association, an attitude which apparently antednted the 1972 change of Government but may have been sharpened thereafter. 45. Project negotiations did not reveal this divergence of views possibly because one side felt that legal documents recording agreement reached (SAR, para. 7.01) were sufficient guarantee of commitment while the other side regarded assurances given as subject to modification. Neverthe- less, there were signs that communication was not always on the same wave- length. For example, the April 18, 1973 Decision Memorandum assumed that the transformation of RNCFM into a financially and operationally autonomous entity was imminent and recorded the Association's resolve to negotiate the project only when this had been accomplished and when consultants for RNCFM were about to be appointed (para. 14). Neither was RNCFM granted autonomy, nor were consultants appointed but, even so, negotiations did take place (para. 15). Project processing was advanced by making the enactment of the autonomy decree a condition for Board presentation and the appointment of consultants a condition of loan effectiveness. The credit was presented before enactment of the autonomy decree and this enactment also become a con- dition of effectiveness. Postponing the resolution of important issues continued throughout the implementation of the project and was, indeed, extended well into the implementation of the Second Railway Project of 1979. The different priorities attached by the Association and the Government to the fundamental question of economic rationalism were not, so far as the audit could determine, ever touched upon possibly from fear that such a dis- cussion would do irreparable harm to Government-Association relations. The audit believes that avoidance of an open discussion has benefitted neither side. 46. The record reveals that right from the start the Association was aware of Government attitudes which were at variance with the Government's expressed acquiescence in the Association's proposals. An internal memoran- dum dated January 30, 1974 summarized the status of technical assistance re- cruitment for MAT and reported that delays were due partly to bureaucratic procedures and partly to Government unwillingness to entrust planning func- tions to expatriates. Even more compelling is the evidence on the lack of Government commitment to financial covenants. In late March 1974, RNCFM wrote the Bank asking for a postponement of tariff increases from April 1 to - 13 - July 1 (para. 15a). The Bank agreed and then, in successive communications dated August 23, 1974, January 9, 1975, March IR, 1975, and May 5, 1975 attempted to find out why the increases had not yet been implemented. On June 11, 1975 the Ministry of Finance wrote that the matter was being studied but the November 24, 1975 Supervision Mission report states that, five months later, the tariff increases had not yet been introduced. 47. The November 24, 1975 Supervision Mission Report also noted that procurement was delayed by a year; that cost overruns obliged IDA to finance five instead of 20 originally planned passenger coaches; that the RNCFM con- sultants had submitted their Phase I report which, although favorably re- ceived by the railway, could not be followed up by Phase II since the ques- tion of autonomy was still unresolved; that traffic had not recovered from the national disturbanres of 1972; and that RNCFM continued to make losses partly because the tariff adjustments had not yet been introduced. On January 21, 1976, the Bank's Railways Advisor wrote a note to the Eastern Africa Regional Office questioning whether, on the basis of the Supervision mission findings, any further lending to RNCFM could be justified. From that point on, the record shows that whereas Back-to-Office Reports (BOR), written by Supervision Missions immediately upon their return to Washington, are terse and gloomy (for instance, the BOR dated March 3, 1976), Full Supervi- sion Reports (FSR) on the same mission are positive and optimistic (FSR, dated April 14, 1976). Similar differences are noticeable in the BOR dated October 18, 1976 and the FSR dated November 12, 1976 as well as in the BOR dated November 18, 1977 and the letter, dated December 29, 1977, sent to the Government as a summary of the October 1977 supervision mission. 48. The audit cannot say whether or not there is a correlation between the diplomatic language and the positive tone of official communications from the Association during 1977 and the December 21, 1977 request from the Ministry of Planning for a second railway loan. It cannot, however, be argued that the second project was approved because the experience from the first had been encouraging: the May 24, 1978 report summarizes supervision of the First and appraisal of the Second Railway Project. Breaches of cov- enants are acknowledged and attributed to unsettled conditions; the RNCFM is reported to lose money and to have no autonomy; the institution-building objectives of the First Project are hardly mentioned; the proposed Second Project would also be a "holding operation" but would not go into any aspect even remotely connected with sectoral planning. The Second Project is in effect a railway maintenance operation pure and simple and this suggests that the Association gave up trying to assist Madagascar in building up its national transport planning capacity possibly because the Government had demonstrated how indifferent it was to such an exercise. The question arises whether, under the circumstances, the Association could justify its continu- ing involvement in the transport sector but the atiswer escapes the ambit of this audit. 49. Still, when the Second Railway Project was being appraised, the First Project was far from being completed. On August 18, 1978 the Associa- tion informed the Government that in order to get the agreed 3% return on net fixed assets an immediate 30% tariff increase was indispensable and asked - 14 - what the Government's position was on the matter. Although no answer was provided, the Association wrote to the Government, on the occasion of the annual Project Implementation Review on May 9, 1979, that the First Railway Project was satisfactorily executed. One week later, the BOR, dated June 7, 1979, on the April 1979 supervision mission, contradicts the May 9 letter by showing that the project continued to be plagued by a varIety of financial, administrative and operational problems. Transport Planning 50. Of all the project components, the provision of technical assis- tance to MAT (paras. 29-30) had the largest scope for long-term institution- building benefits. Partly to accommodate the Government's desire for fewer consultants (para. 9), the Association arranged for the head of the technical assistance team to be a Bank staff member who took up his duties in March 1975. The team was completed with the arrival of three more experts who began work between October 1975 and January 1976. 51. The Terms of Reference (SAR, Annex 7) and the November 23, 1976 BOR show that the team had to produce studies touching on all major and many minor aspects of transport planning. Nevertheless, the Government considered a number of these documents (for example, "Transport Routier des Marchandises", Rapport Provisoire, May 1976) to be incompatible with its socio-economic philosophy and restricted their circulation even within its own departments. The Association became aware of these restrictions and in a December 28, 1976 letter, requested copies of all reports but the record shows that Government never responded and the matter was not followed up. 52. The technical assistance team worked under difficult circumstances and its Progress Reports (dated February, August and November 1976) note the absence of Government instructions regarding priorities of the differcnt items on the work program, the lack of counterparts, the official silence to the team's recommendations, and the jurisdictional problems arising from the antagonism between MAT and the Ministry of Public Works. The audit has found no evidence that the Association encouraged the Government to view the recom- mendations of the expatriate experts as inputs to an institution-building effort which, over time, would allow Madagascar to carry out transport plan- ning with its own staff. On the contrary. the Association seems to have accepted the Government's lack of interest in systematic planning and this explains the absence of any planning-related component in the Second Railway Project. RNCFM Finances 53. The deterioration of railway finances was the main concern of all supervision missions (PCR, para. 5.08). The Association took it for granted that since the railway performed an important economic function (para. 1la), it deserved assistance to become more efficient, to provide a better service, and to earn enough revenue so as to cover its operating expenses and invest- ment requirements. These assumptions were fortified by assurances given during the November 1973 negotiations that Government would support the rail- - 15 - way so as to keep it going as a financially viable concern and that, to this end, Government would take measures to create the required commercial climate and to develop the necessary traffic patterns (para. 15d). 54. These promises were not kept and, almost six years later, during negotiations of the 1979 Second Railways Project, the Government once again agreed to do what it had already committed itself to do under the First Proj- ect. Specifically, that the railway would be operated on commercial prin- ciples; that tariffs would be increased by 25% effective January 1, 1979; that Government departments would reimburse promptly the railway for services performed; and that Government would reimburse RNCFM for the amounts foregone because of concessionary passenger fares. The December 21, 1981 PCR states that under the Second Project, tariff increases were again delayed by almost a year, that Government departments did not reimburse the railway in cash, and that the reimbursement for concessionary fares was not yet implemented (PCR, para. 5.08). In sum, the deterioration of RNCFM finances continued well into the Second Project. IV. CONCLUSIONS 55. Project objectives were clearly defined at appraisal: railway track and equipment needed modernization; railway operations had to be improved, especially in view of the forthcoming road competition; national transport planning needed strengthening; and Malagasy staff had to be trained. The audit has concluded that RNCFM was able to record some accomplishments in activities such as track renewal, which were within its special domain and required limited collaboration of other Government bodies. Where project activities depended upon wider political or bureaucratic support outside the railway (such as in finance, which required tariff increases and collection of amounts due from other Government agencies), little seems to have happened. In areas transcending railway operations (particularly in national transport planning, which required a commitment from the Government at its highest Levels), achievements were likewise limited. 56. The project was reduced in scope because of cost overruns and was completed with a three-year delay. Lower traffic, higher operating costs and a longer implementation period contributed to the drop of the rate of return to less than 10%, compared to 35% estimated at appraisal. The SAR was sanguine regarding prospects for economic recovery (SAR, paras. 3.28 and 6.07); the Government's commitment to grant autonomy to the railway (SAR, para. 3.03); and the Government's willingness to allow expatriates to engage in planning and policy-making functions (SAR, para. 4.07 and Annex 7). Reservations previously expressed by Bank staff (paras. 7, 9, 13a, and 14) and the difficulties experienced with the implementation of the Tamatave Port Project (para. 3), should, in the audit's opinion, have tempered the con- fidence of the SAR concerning the feasibility of the project's institutional and financial objectives. 57. The Association believed that RNCFM would become more efficient if it were granted autonomy. This implies full control over revenue and expen- - 16 - ditures and is an elusive target in public administrations patterned after the French model where the Ministry of Finance does not easily relinquish its role of final arbiter. Besides, it was not difficult to see that railway autonomy would have ultimately led to higher tariffs and, on these grounds alone, was bound to create opposition regardless of any analysis showing that the prevailing rates were too low. Furthermore, even if RNCFM had been granted autonomy, there is no way that this would have persuaded other Government bodies to settle their accounts (para. 36). Hence, in the audit's view, the autonomy issue was still born and the Association should have con- fronted the possibility that, in the foreseeable future, RNCFV would be run on subsidies. The audit believes that if this had been squarely faced, the Association could have focussed on measures intended not to set up the rail- way as a privileged enclave but, quite simply, to strengthen its financial position. For example, the need to increase tariffs and to take the other steps required so that RNCFM would obtain a 2.5% rate of return through 1976 and 3% thereafter deserved the most vigorous support and it is hard to under- stand the Association's inaction when important covenants were breached (Section 4.03, Project Agreement). 58. The audit found that throughout project implementation Association staff remained unconvinced that nationalization and disregard for market signals would bring about equity and growth but the Association never officially discussed the economic and financial implications of Government positions such as "Tariff setting and control by the State avoids a competition which is destructive for the transporters and excessive, therefore insupportable, for the transport users" (page 1 of the attached Transportation Policy and Action Plan"). 59. The PCR of the First Railway Project states (para. 6.08) that the Second Railway Project (Credit 903-MAG, 1979) is complemencary to the First. The audit reviewed the SAR and selected supervision reports of the Second Project but found no evidence of complementarity in such crucial areas as national transport planning and road-rail coordination. In fact, it found that the Second Railway Project is a deferred maintenance operation and that no attempt was made there ov in the subsequent Fifth Highway Project (Credit 938 of May 1979) to tackle decisively the broader organizational and finan- clal problems left unresolved by the First Railway Project. Still, the audit is not insensitive to what would have happened if, in the face of breach of covenants and of indifference to institution-building, the First Project had been suspended and the Second never approved. There is little doubt that transportation services on the country's principal corridor would have suf- fered, probably to the point of complete paralysis, especially since In mid- 1984, thirteen years after the identification of the First Railway Project, the Tamatave--Tananarive road, which had been planned for completion by 1977, is still unfinished. In this sense, both the First and the Second Railway Projects provided the people of Madagascar with benefits which no rate of return calculations can hope to quantify. - 17 - 60. The audit has concluded it is not productive for the Association to launch projects aimed at tconomic efficiency and the encouragement of compe- tition when the Borrower has reservations about the priority ol these objec- tives. Under such circumstances, the Associativn ought to explain patiently the underpinnings of its own developmental philosophy and to encourage the Borrower to elaborate on his own views regarding equity and growth. Doubt- lessly, this will be a time-consuming process requiring honesty and tact on both sideu. Nevertheless, only a frank dialague may perhaps Identify areas for genuine collaboration. Once mutual understanding has been achieved, the Borrower would he in a better position on the one hand to Indicate those areas in which he would welcome Association assistance and, on the other, to suggest policy measures he would he genuinely prepared to take. Assuming that no fundamental conflict emeryes between tho two positions and that the proposed measures are substantive rather than cosmetic, then a suitable proj- ect could be designed. The absence of such a dialogue brought about, in the audit's opinion, the unsatisfactory results of the First Railway Project.1/ 1/ The Region has the following general comments: 1. The audit report accurately reflects the sitruation In which this project Jound itself but does not attempt to fully explain the conditions in the country. Rather it raises the question of what shculd be the Bank's role when political or economic conditions are less than desirable. Should the Bank's function be primarily rescue or development? 2. At the time of appraisal it was obvious the country needed assistance for its railway. It still does. The project was not conceived as a rescue operation nor was the second raflway project. Insistence by the Bank that the statute. of the railway be changed finally were enacted uring the second project and it- position although still not completely effective has improved measurably. 3. Outstanding account. which did not improve during the first project have during the second. The harsh criticism of the audit, although probably justified on the recorded facts, does not attempt to take in to consideration the political climate and monetary situation which existed during the life of the project. 4. What is evident is that Madagascar would have beevi considerably worse off without the project and the economic health of the country is today much improved by the Bank's continuing involvement in the Rail- way's affairs." --/K- - 19 - ATTACHMENT 1 E-1287/84 July 11, I84 French Madagascar OED/Transport DBB:cc Rev.:ENMcM Ministry of Transportation, Supply and Tourism Transportation Policy and Action Plan September 1976 - 20 - A. Fundamentals of General Transportation Policy The main guidelines of the State's general transportation policy have already been defined in the Charter of the Malagasy Socialist Revolution, namely: - to link the various regions by means of a system of main roads; - to provide for the proper maintenance of existing main and second- ary roads and make optimum use of maintenance equipment being used below capacity; - to review the organization of transportation as a whole and plan the sector in such a way as to bring it more closely into line with the development needs of the production sectors. These are qualitative objectives that W.ill need to be quantified, planned and monitored, and the resources required to achieve them establish- ed. B. The Role of Transportation in the Economy Transportation plays a vital role in the general economic life of the nation and by virtue of its universality it plays a part at all stages of production, processing, distribution and consumption. Transportation is also a catalyst of social life, promoting the movement of ideas and people, through the propagation of culture and the intermingling of people from different regions. It is therefore a key sector which the State and the community must bring under their control. To achieve such control, in a socialist socio- economic setting, some degree of dirigisme Is required. Indeed, in the transportation sector, where improvisation was the rule on the eve of independence, there is still a shortage of qualified staff and, as a result of the paralyzing impact of so-called technical "assist- ance," personnel training in Madagascar has progressed at a rate dictated by those who wanted to hold on to their positions (gradual Malagasization as it was described at the time). Furthermore, the transportation industry is a heavy industry in the sense that it requires substantial capital investment, and only the State is in a position to meet this need under existing condi- tions. In view of the geophysical, economic and social situation in Madagascar, it is essential that the planning and coordination of transporta- tion -- both in time and space -- take account of national priorities. The State alone can play this part in a rational manner within the framework of the country's general development policy. - 21 - C. Transportation Objectives As regards transportation, the Ministry of Transportation, Supply and Tourism proposes two specific priority objectives: I. Meet transportation needs in terms of quantity, quality and frequency at minimum cost to the community; II. Contribute to the general development of the country. These objectives call for an action strategy, in other words the action policy adopted in the field of transportation must be such as to pro- vide immediate solutions to current problems while adhering to the long-term final objectives established under the Charter of the Malagasy Socialist Revolution. To meet these objectives, various measures will therefore be propused for each transportation mode based on the efficient and rational utilization of transportation facilities, in combination with other branches of the economy. I. Meeting Transportation Needs in Terms of Quantity, Quality and Fre- quency at Minimum Cost to the Community This clearly is the role to be played by transportation. The means needed to achieve the objec:Ives set are of three types. 1.1 Transportation coordination and planning Coordination and planning relate to service areas, tariffs and com- plementarity. 1.11 Service areas: For a rationtl distribution of transportation modes in space in order to achieve optimum utilization of resources. This implies coordination among the departments responsible for drafting infrastructure policy, as well as investment planning. 1.12 Tariffs: The State must be involved in setting tariffs, for example by establishing tariff brackets within which each decentralized administrative authority can set tariffs applicable within its jurisdiction, in the light of the special features and constraints of the region in question. For the State to fix tariffs and exercise control means avoiding competition that would not only be ruinous for the transport operators themselves but also place an excessive and hence unbearable burden on users. In effect this avoids throttling the economy. 1.13 Complementarity: The various modes of transportation must be complementary and not in competition with one another. Consider- able savings in investment can be achieved as a result. Technical coordination of the various modes of transportation -- namely rail/road, sea/road, air/road, sea/rail, etc., -- is therefore needed. - 22 - 1.2 Existing structures The existing structures of the enterprises that control transpotta- tion are those inherited from the colonial power. They grew up within the context of an economy completely dominated by outside interests. They %re therefore out of date and unsuited to the Government's goal of creating a Oev socialist society. These structures will need to be redesigned with a view to: 1.21 Takeover by the State or the community In views of the capital expenditure required, and the problem of tying up vast sums needed for competing priority commitments, a gradual approach seems to be the answer. The successive stages in this process would be as follows: - acquisition of interests in all transportation enterprises of economic, social or strategic importance; - obtaining a majority position on the boards of existing major enterprises; - total nationalization. 1.22 Consolidation of existing firms within each mode With the exception of railways and airlines, the basic characterifs tics of transportation companies are: - operational anarchy: total absence of any policy, large number of firms, absence of true cooperatives, development of lines without prior social and economic study; - inadequate resources; - poor distribution: concentration of resources on trunk routes and inadequate service on feeder routes; - obsolete equipment: intensive use, purely superficial maintenancep no depreciation; - lack of standardization of equipment, creating serious problems for spare parts and maintenance. In order to remedy these problems and give the State control over transportation, consolidation within each mode is now the rule, and any newly created transportation enterprise has to be authorized by the Ministry in charge of transportation. - 23 - This consolidation also makes it possible to increase and strength- en resource capacity, reduce operating costs, facilitate management control and ultimately avoid ruinous competition. The increased scale of financial operations will make it easier to deal with various management problems. 1.221 Air transportation: A single company is preferable for the many reasons listed above. There is, however, the problem of subidiaries, since neither the State nor the community is represented on their boards (for example TAM -- Travaux Aariens de Madagascar), the parent company being the sole stockholder. The regulations governing commercial civil aviation will therefore need to be revised, again within the framework of a single company, and applied both nationally and internationally with a view to: - democratizing air transportation through the setting up of an effective public service that is properly adapted to needs; and - diversifying our foreign relations through the development of an international airline network within the framework of our "omnidirectional" policy, which among other things will have the effect of developing the tourist industry. The revised regulations should provide the authorities with the means required to intervene effectively in all aspects of air transporta- tion. This will mean: - revising the agreement with Air Madagascar; - revising the organization of air taxi services and the statutes of TAM. Domestically, this program will be geared to the need: - to improve interregional service through decentralization of the system and customer services; - to restructure tariffs mainly to reduce charges on the secondary network with its public service role; - to reduce operating costs through a reduction in personnel expenses following the Malagasization of jobs now held by expatriates; - to improve productivity through the rational and optimum utiliza- tion of the fleet; - to base strategic long-term planning on established needs, market research, traffic projections, and the optimization of supply to meet potential demand. - 24 - Internationally, the program will be as follows: Acquire rights to routes, not only those that can be operated with the resources now available or that will be available to our domestic airline (Air Madagascar) in the near future, but also rights acquired in exchange for similar rights. It is understood that constructive solutions will be needed as part of a pattern of very close bilateral cooperation that will enable each country concerned to protect its interests and correct any factors that make for inequality (geographical situation, type of traffic, disparate re- sources). This might mean: - intercompany agreements (joint operation pools, charters, two-way operation arrangements, capacity limitation or sharing, revenue sharing, sharing of commissions etc.), as is currently the case. with East African Airways and Air Mauritius. - technical agreements (covering aircraft and crews). 1.222 Maritime transportation: Unlike air transportation, maritime transportation is provided by enterprises of various sizes and structures. The shipping industry is still heavily dominated by foreign irtersts and is characterized by: - Too many shipowners in relation to the number of ships, resulting in cut-throat competition with unfortunate financial results; - An absence of coordination of operating policies as regards both the service provided and transportation capacity, thus increasing costs; - An aging fleet of ships due to poor investment practices resulting from a virtually permanent cash shortage: - A lack of coordination of the demand for trAnsportation as regards both basic necessities and high-cost freight feeder on extension lines; - The existence of a totally heterogeneous port infrastructure with adverse effects on the composition of the fleet of the various shipowners as well as on their operating costs. This also makes it necessary to keep a large fleet of sailing ships in operation. In order to give the shipping industry the importance that will enable it to play its proper role in the country's general economy, and to facilitate State control and improve management and yields, consolidation of all shipping activities under one entity is essential. To this end, a national corporation is to be established which will cover all activities directly or indirectly relating to shipping. The - 25 - statutes of this new company will be in accordance with the terms of the Charter of the Revolution. Its function will be:. - To implement the State's general policy for the shipping sector; - To coordinate and plan all shipping activities; - To carry out or cause to be carried out or to monitor all trans- portation operations, lighterage, transshipment, consignments or charters; - To take up equity interests and establish subsidiaries; - To absorb enterprises in the same line of activity, or enter into management contracts; - Generally to cover all activities affecting shipping both upstream and downstream from transportation operations. In view of the commercial requirements that have to be faced and also the impact of freight costs on costs of production, it will be neces- sary: To reconcile sound management and productivity policies with public service needs; - To stabilize and reduce freight rates; - To bring operating policies into line with those of other trans- portation modes (road, air and rail); - To give priority to Madagascar flag vessels for routes from or to Madagascar routes; - To implement the provisions of the code of conduct of Maritime Conferences when they take effect. In order to promote and develop the long-distance operations of this company and to use to the full the resources it places at the disposal of customers, sales and purchases will be expressed as far as possible CIF and FOB, respectively. In addition, to reduce any monopoly effect on maritime operations, a shippers council will be established as the spokesman and partner of the company for maritime activities. The shippers council will manage the interests of the shippers. This would be the Malagasy Government's response to the UNCTAD recommendation that a shippers council be created. 1.223 Road and river transportation: It has been possible up to now to regulate only passenger traffic. - 26 - Even this system is characterized by: - Lack of coordination - Unhealthy competition - Inadequate resources - Spirit of insubordination among cooperatives and their members. Freight transportation is currently under study and an IBRD mission has been studying this subject for nearly 6 months. The salient characteris- tics of freight transportation remain dominant control by foreign interests (Devilleneuve, Extramad, Melia, Brai, etc.), the multiplicity of transport operators, and the diversity in the size of the enterprises. The aim is to centralize and coordinate the demand for transporta- tion in order to provide an adequate response to needs, to provide regular public service, to eliminate de facto monopoly by certain operators, to monitor transportation support operations, and to establish a licensing sys- tem for professionals in the field. A regional grouping of transporter operations is planned but the first move must be to clear up the situation and establish a sound basis of regulation. 1.224 River transportation: There is very little river transportation in view of the geography of Madagascar, but also due to the rainfall. In view of the advantages and very low cost per kilometer offered by this form of transportation, the objectives of the department responsible for this area will be as follows: - Preparation of an inventory of navigable waterways; - Study of equipment used; - Coordination with other modes; - Improvement of existing waterways (Pangalanes canal). 1.3 Coordination of transportation and infrastructure policies All transportation modes, even those that require no special route structures, need some form of infrastructure (airports, ports, etc.). In crder to avoid undue expenditure of effort and for the proper coordination and planning in time and space of the investment required, infrastructure policies will need to be coordinated with transportation expansion programs. - 27- CONTRIBUTING TO THE GENERAL DEVELOPMENT OF THE COUNTRY The ultimate objective of transportation policy is to contribute to the general development of the country. 2.1. This contribution to development will take the form of: 2.1. Freight transportation at lower cost - Through the study and application of a balanced tariff; - Through the supply of rational and appropriate transportation ser- vices. 2.2 Opening up isolated regions - Making the movement of products to market easier by consolidating the demand for transportation. The freight offices to be set up in each region will be responsible for coordinating the supply and demand for transportation, and for the rapid movement of basic necessities from where they are produced to distribution or consu- mer centers. This will help break up the de facto monopoly established by a few privileged transporters. - Expansion of the domestic market; - Promotion of external trade; - Speeding up of the rural development process; - Intermingling of people from differenr regions. 2.3 Consolidation of resources The increased availability of physical and financial resources will make it possible to replace foreign firms with Malagasy firms. Thus profits and accumulated capital will remain within the country, providing opportuni- ties for reinvestment, expansion and even diversification. 2.4 Promotion of tourism A properly adapted infrastructure combined with rational trans- portation facilities is indispensable for promoting tourism, the industry that brings in most foreign exchange earnings. The minister responsible for transportation will undertake to study tariffs. These will have to be attractive. To this end, charter flights and package tours, through both scheduled and unscheduled services, could play a - 28 - very important role. Flexible regulations will therefore be enacted for charter flights, provided this type of service does not undercut regular services. 2.5 Development of our external relations within the framework of an "omnidirectional" policy Madagascar's geographical situation as an island and its relative isolation from main traffic flows, with the exception of oil traffic, make it necessary for Madagascar to expand its external relations as part of its "omnidirectional" policy. This continues to be the main objective of its trade negotiations or agreements with foreign countries. These various lines of action are presented diagrammatically in the annex, showing the main thrust of the policy of the Ministry of Transporta- tion, Supply and Tourism in the area of transportation. The Minister of Transportation, Supply and Tourism /s/ Jean Bemananjara TRANSPOPT S TUR RES RVE COORDINATION WITH OTHER DEPARTHENTS STATE INVOLVEMENT oo*ina.an. .MAIN MISE .CON14TR ggy IONS Avet les Autres __< *EAIdet 06prtedet .EA deeso Instrument o e intervent .TraInspor 1%erlent 3 tous Its stades'- Transportation plays role Instrument d'Intervention de 1. Etavit' de rconomie at all levels of economy .SPport *conarrique ** sneustriet - Just I JIRAMA/SECREN type company for each sector Support for econo KW@SA il0ltT -1 Saule SocIff6 type JRAMAISECREN pour chaque setteur and industry DES ENTRzPisE5 EaI !o60ct# pour coiffer touites tes acjivqts marltirn.es .Substilution Entreprises Nationa!es J Enfreprises 4trtg#res at 1 Co. to cover all maritime activities fm Replac foreign firms with Malagasy firms Augmentation Inc eased scale of Facilitatior l:aier to 5 vings inEconor-e des-. do tosutface ulantl6e financial operationu du contrtrE exerci-e ans - myens control vipment- maltrais orination rlanninf- 01 o0 C'fl*nt enewal of Cocrrani2n TLriff Consolidat n by Fre;re -t ew ril;!rtel. an-pslisset.~n Plnfc1:n "A,licalpa wa 0-ionl E rII lI eII n1 quipment/investmLent Priuen of activi pr brach.s o Standardizatio of equipment DecV11tranl'.aton Decentrilization R duced operating cosEs* soupesto peaioa f exbi itv :A%%iujinne lata!uIII .D'ai.nltyexibilicvre.Am>hution de laroductiita Bett r productivity Speci liz- .unotn,raa,on uaI -Fods e o rso rcs.RtwmmoiundolnVh';doQQs Sati n -Darinvution ecs crurgps .renorli,se rieit Sur place ao" I.4eg Rationalization of .*A*, g.s I age * orking capital -0rganimationCes disettes - * Reinvestrent. within --Orgmation of investme t .Sptc.atston the country service areas u_NContribution to Contim gnrabN Oinnrt t d6veloppeti the country 10 Promotion of \tourism Pr leltranstertdes blens vement of freight a Par to destiavement OpeninC,-up of Par la promo-ior d .A moindre coot Ofcost dt2 r6glons 11t9.1ilatdfe%ou touris-ne V t c : .. . Evacu.:tin derroamt, Ioverent, of prodi-t Accens to !;itun .Transport 011oPt0 ;I r .!tc. tre. - U" 1 .119a16rsteentu march E:x&ai*k1utoI outuarket.u ti.b. 4t' infrastrii. Iiterinitarliml of pcopie . Induced ectfectu -2 o r.>- - 31 - ATTACHMENT 2 TRANSLATION November 23, 1984 Mr. Yukinori Watanabe Director Operations Evaluation Department With regard to the draft evaluation report on the First Railway Project (Credit 483-MAG) we have the honor to inform you that RNCFM has no comments on this draft. Director General National Railroad of Madagascar - 32 - ATTACHKENT 3 L-640/85 Dccembcr 19, 1984 French (Madagascar) OED DRJB:ak DEMOCRATIC REPUBLIC OF MADAGASCAR MINISTRY OF TRANSPORTATION, SUPPLY AND TOURISM Malagasy National Railways (RNCFM) Office of the Director-General Antananarivo, November 28, 1984 Mr. Yukinori Watanabe Director OED World Bank Washington, D.C. Subject: PPAR, First Railway Project (Credit 488-MAG) Dear Mr. Watanabe: In response to your letter of October 5, 1984 regarding the Project Performance Audit Report on the First Railway Project in Madagascar (Credit 488-MAC), I should like to confirm that we have no observations to make on the document other than to note the following particulars in connection with project works carried out by RNCFM: - the crossing stations at Andranokobaka (km 11 on the TCE line) and Ambalahoraka (km 203 on the TCE line) came into service in 1977 and 1979 respectively; - the works begun since 1976 to extend workshops and yards and construct miscellaneous buildings and accommodation facilities have gone ahead on schedule to date; - the new diversion between the Brickaville and Ambila-Lemaitso stations was opened to traffic in 1980. Yours, etc. Is/ Samuel Razanamapisa Director-General - 33 - MADAGASCAR FIRST RAILWAY PROJECT - CREDIT 488-MAG Project Completion Report Summary (i) The project was intended to replace outdated equipment and provide a modest increase in transport capacity, improve the railway's managerial and operational efficiency and set up a framework for transport planning. The total cost of the project was estimated at US$ 8.9 million of which US$ 6.0 million in foreign exchange, financed by IDA. The project was approved by the Board in January 1974 and was completed in July 1979. (ii) The project was reduced in scope due to cost overruns, and completed with a delay of three years mainly because of (a) a deliberate postponement of part of the consulting studies pending resolution of the question of the railway's autonomy, and (b) delays in track works. (iii) The project did not meet its objectives due to the unstable political climate in the country, changes in basic Government policies and the stagnation of the economy. The reevaluated economic rate of return on the phy:ical components of the project amounts to less than 10%; this is however probably conservative in view of the difficulty to evaluate what would have happened without the project. (tv) RNCFM's financial situation deteriorated continuously mainly because of the lack of competent financial management and the Government comLtment to implement the necessary actions. (v) The institutional objectives of the project were not realized. The consultants' work to improve the railway's management had only limited results in specific areas, but most of the recommendations cotild not be implemented because of the inaction of the Government in granuing the railway the necessary autonomy in financial matters. The consultants' team to the Ministry of Transport (MAT) did not succeed in setting up a general framework for transport policies mainly because of the Government's lack of identification with the objectives intended and the lack of itegration of this team in the decision process. (vi) Despite all the shortcomings of the project, IDA played a constructive role by maintaining an open dialogue on major policy issues and by providing practical advice to the railway. 37-- s 24 ce ' 4 - 35 - MADACASCAR First Railway Project- Credit 488-MAG Project Completion Report I. INTRODUCTION 1.01 The state owned railway - Reseau National des Chemins de Fer Malagasy (RNCFM) plays a central role in Madagascar's economy. It is the only reliable means of surface transportation between many of the island's economic centers and, most importantly, between its main port of Toamasina, and the capital city, Antananarivo, some 370 km inland, on which corridor it enjoys a monopoly situation. This monopoly, however, will cease as soon as a highway parallel to the rail-line is completed, which is currently scheduled for 1983. 1.02 This strategic role has been the key factor in formulating the Bank Group's support to the sector. Support began with a First Railway Project, approved in 1974, which is the subject of this report, and continued under a Second Railway Project approved in 1979. Both projects were mainly "holding operations" aiming at strengthening the railway's management and at addressing deferred maintenance rather than expanding the railway's transport capacity. 1.03 Because of its central role in the Malagasy economy, the railway's performance has been closely linked to the general economic and political environment which has prevailed in Madagascar since 1972. The period was marked by political unrest between 1972 and 1974 and by economic stagnation and fundamental changes in overall policies afterwards. This resulted in a lack of a coherent transport policy during the whole project period, in continuous uncertainties regarding basic Government orientations on major questions such as on the degree of autonomy of public bodies, and in delays in reaching agreements and implementing decisions on current matters such as on tariff increases or procurement approvals. In addition, because of the railway's strategic role, the Government often interfered directly in RNC?M's management, in particular, by abruptly changing the railway's top -aanagement twice and by imposing priorities on the movement of goods; these further hampered normal operations. 1.04 All these factors contributed to delays in project implementation, postponement of some parts of the project to the Second Railway project, and other operational problems. Thus, the real benefits of this project have not yet completely materialized. However, an analysis of the past experience may be useful in analyzing retrospectively the approach taken by the Bank in this difficult environment, and in providing guidance for successful implementation of the ongoing Second Railway Project. - 36 - II. PROJECT PREPARATION AND APPRAISAL 2.01 Discussions between the Government and the Bank on a major modernization program for the railway began in the early 1970's. In 1971, a Bank mission identified a possible project comprising the 1972-74 portion of the railway's 1972-76 Investment Plan, amounting to US$16 million. The plan was subsequently changed several times and the version presented during appraisal in 1973 covered the period 1973-77 and amounted to about US$43 million. 2.02 Extensive discussions on two major issues explain the long delay between identification of the project in 1970/71 and credit signature in August 1974. These issues were: (i) the need for the railway's management to have sufficient flexibility and authority in operational, financial and personnel matters; although operating in a legal vacuum, the railway was, in fact, run as an ordinary government department, subject to all civil service rules and financial and budgetary procedures; and (ii) the need for a coordinated transport policy, especially since the Government was planning to build a new road parallel to the railway between Toamasina and Antananarivo; this raised uncertainties over the railway's future role. 2.03 Despite the delay, these issues were never satisfactorily resolved. Although a post-appraisal mission was sent to review and agree on a proposed decree giving RNCFM greater autonomy, the decree enacted differed from the one agreed upon at appraisal by omitting any reference to the creation of an autonomous financial service. A compromise was finally reached on this issue by including such a service in the organizational chart of the railway and by including in the project technical assistance to conduct a major review of RNCFM's organization and legal set up. 2.04 Taking into consideration the railway's inherent problems and since this was the Bank's first project in the railway subsector, the project focussed mainly on studies and technical assistance to improve general management and transport coordination. Other investments were confined to providing equipment to meet the railway's immediate needs, because of the uncertainties regarding RNCFM's future. The project objectives were: (a) to improve the railway's managerial and operational efficiency, particularly in organization, finance, accounting, operations and personnel; (b) to investigate the Railway's long term prospects and requirements as well as the needs of the transport sector as a whole; and 37 - (c) to aid the railway in replacing outdated equipment and providing a modest increase in transport capacity. 2.05 The main project's physical components were: (a) renewal of rails and replacement of sleepers on 60 km of the main line, extension of yards and sidings, and construction of a few buildings; (b) provision of track maintenance equipment; (c) provision of 50 freight wagons and 20 passenger coaches; (d) technical assistance to RNCFM in the fields of management, finance, personnel and operations; and (e) technical assistance to the Ministere de L'Amenagement du Territoire on transport planning and coordination. 2.06 The project comprised that portion of the Railway's redefined 1973-75 investment plan for which financing had not been found. The final plan had been arrived at by postponing all major investments such as major acquisition of rolling stock and construction of new railway lines until the completion of a railway modernization and rationalization study included in the project. The elements of the plan not financed by the Bank Group were: (i) realignment of the Ambila Brickaville section financed by USAID; (ii) provision c,f seven locomotives under CCCE financing; and (iii) final engineering study of the Antsirabe-Fianarantsoa line, under Italian Technical Assistance. - 38 - Ill. IMPLEMENTATION AND COST General Implementation 3.01 A table comparing actual and expected project implementation is given in Table 6. Implementation was generally slow and the project completion date had to be postponed many times. The project closing date was delayed by three years, from June 30, 1976, to July 31, 1979. The last disbursement was made in December 1979 and the credit was officially closed on February 26, 1980. 3.02 The project was slow to start because of the delay in credit signature and effectiveness and the general inflationary context which made bidding by suppliers difficult. Further delays occurred in RNCFM getting approvals from the ministries concerned before it could award contracts. However, the major delay was a deliberate postponement of an important part of the project -- Phase II of consulting studies -- pending resolution of the question of the railway's autonomy. Finally, delays also occurred in carrying out track works because of the lack of adequate ballast wagons, the limited planning capacity of the railway, and major additional works unforseen at appraisal. 3.03 It is doubtful that all these delays could have been foreseen at appraisal. Recommendation to the railway to proceed immediately with procurement matters was made at appraisal and all tender documents had been reviewed before the credit was signed. The delays in track works show in retrospect either an over-optimism in the Railway's capacity to implement the track renewal program, an inadequacy of the project content, or both. It appears also -- in retrospect and leaving aside all "country" issues -- that the Bank's action on the question of the railway's autonomy was inadequate. If this question had been resolved decisively at the outset, when the Bank had project effectiveness as leverage, many delays and problems could have been avoided. This problem is dealt with in more details in the chapter dealing with "Institutional Development". Project Costs 3.04 A comparative cost table showing actual and expected costs is given in Table 1. The total cost overrun of the project amounts to 34%. This is, however, an underestimate since the project scope was reduced; if all wagons and coaches had been purchased, the cost overrun would have been about 55%. 3.05 The number of wagons to be financed by the project was reduced from 50 to 35, and the number of passenger coaches from 20 to 5 (plus an additional 10 financed by CCCE). The second phase of the consultants' studies was also amended and reduced in scope; this made possible a reallocation of US$380,000 to purchase 12 ballast wagons which were not included in the original plan. - 39 - 3.06 The cost overruns resulted from world-wide inflation following the doubling of oil prices and wide fluctuations of steel prices. Local project costs for track renewal exceeded appraisal estimates by 45%, mainly because of the delays in project implementation and salary increases to keep up with local inflation. A detailed analysis of the major components of the project is done below. Detailed Analysis of Major Components (i) Track Renewal 3.07 Invitation to bid for track materials was published in November 1973, but due to delays in transmitting copies of the invitation to foreign embassies, the submission of bids was extended to January 1974. Eight firms obtained tender documents; only two bids were received, one for rails and sleepers, and the other for fastings and fishbolts. Since no bid was received for welding materials, an invitation was sent to a limited number of suppliers. Retendering was considered but ruled out because (i) the advertising procedure was found acceptable despite the delays and (ii) retendering would probably have resulted in higher prices in the existing inflationary context. All track materials were received by March 1975. 3.08 The table below summarizes actual and expected completion dates of the 60km track renewal program. Actual Number Expected Actual as of kilometers Cumulative Percentage Years Renewed Cumulative Program of estimate 1975 17 km 17km 40km 42% 1976 20 km 37km 60km 62% 1977 2 km 39km 65% 1978 21 km 60km 100% 3.09 According to the railway, the track renewal program was delayed because of supply problems of small materials and a shortage of ballast wagons to carry the ballast from the quarry to the sections needing reballasting. A reallocation in the credit was requested, and approved by the Bank, in November 1977 to purchase the needed wagons. The Bank's view was that in addition to the shortage of ballast wagons, the railway's capacity was hampered by poor organization and planning and by major realignment works on Ambilla-Brickaville. Although the realignment works on the Ambilla-Brickaville section were originally financed by USAID, the overall program to Madagascar was stopped over general political questions and the railway decided to complete the works by force account. The Bank Group questioned the priority attached to these works but with no avail. 3.10 The total cost of the track renewal program amounted to FMG 644 million compared to an appraisal estimate of FMG 526 million without contingencies. This cost increase is mainly due to the increase in local costs (45%) resulting from implementation delays. - 40 - (ii) Tamping Machine 3.11 The tamping machine was ordered after international competitive bidding. The delay between bid evaluation and contract award was due to a major divergence of views between the Bank and the railway. The Bank insisted the contract was to be awarded to the lowest evaluated bidder since it met the required specifications, while the railway wanted to award the contract to another supplier in view of the higher performance of the machine offered (not specified in the bidding documents). The railway finally agreed to the Bank's recommendation. (iii) Rolling stock 3.12 Invitations to bid for rolling stock were made in May 1974 for the wagons and in July 1974 for the passenger coaches. The bids for wagons were opened in July 1974. Fourteen firms requested tender documents; three bids were received. The lowest bid price was about US$33,000 against appraisal estimate of US$20,000 including price contingencies, or a price increase of 65%. It was therefore proposed by the railway and agreed to by the Bank to reduce the number of wagons from 50 to 35 and to exclude from the project one set of automatic barriers. Wagon delivery was about a year late mainly because of the delay in awarding the contract by the Ministry of Finance over a question of the currency in which the bid was expressed. 3.13. Since only one bid was received for passenger coaches, it was not opened; instead it was proposed by the railway, and agreed by the Bank, to send a limited offer to selected firms. Tender documents were sent to 18 firms from eleven different countries. Because of the price increase expected, bids were asked alternatively for 20, 15 and 10 coaches. The unit price quoted in the lowest bid, of about US$200,000 per coach, exceeded by far the appraisal estimate of about US$65,000. As a result, the Bank could only finance five coaches out of the twenty originally planned. To help make up for the difference, CCCE agreed to finance ten additional coaches. The final contract was awarded June 1976, instead of September 1974 as originally planned, and the wagons were delivered in November 1977. 3.14 The large price differences in rolling stock can be partly explained by the wide fluctuations of steel prices between 1973 and 1975. Steel prices almost doubled between 1973 and 1974, before dropping at the end of 19751/. However, this alone does not explain completely the tripling of prices oT passenger coaches. It is likely that the price difference is explained by two additional factors: (i) an underevaluation at appraisal; and (ii) possibly a difference in standards: the Bank considered simple coaches with minimum standards while the railway wanted more luxurious coaches. (The Bank proposed reducing the standard of the coaches, but the 1/ Steel price per ton was $175 in 1973; $400 in 1974; $200 in 1975/77; T275 in 1978. - 41 - proposal made by the supplier was unacceptable to the railway because it would have implied basic changes in the specification of the coaches and a reduction in capacity). (iv) Consulting Services to RNCFM. 3.15 The project provided for consulting services to the railway to help it become an efficient autonomous public enterprise. Consulting services were divided into two phases. The first phase consisted of reviewing the existing situation and making recommendations in the following areas: (i) organizational structure, including legal set-up; (ii) financial management; (iii) operations; (iv) traffic costing and marketing; (v) statistics; and (vi) training and personnel policies. The consultants also had to prepare more precise terms of reference for the second phase, which was to consist of (i) preparation of a detailed investment program for the railway; (ii) study on rationalizing the railway's services; (iii) detailed implementation of Phase I reccomendations; and (iv) training of key personnel. Separate contracts covered the two phases. 3.16 French consultants were selected in April 1974 to carry out Phase I. The consultants' final report was presented in August 1975 and was in general satisfactory. Their basic recommendations were considered sound and were in general well accepted by the railway. Phase II was deliberately postponed for about a year, pending decision on the question of the Railway's autonomy since successful implementation of most of the Phase I recommendations depended on this crucial issue. At the end of 1976, it was decided to proceed with Phase II studies in phases to at least carry out those parts of the studies not directly dependent upon resolution of the autonomy question. In view of their performance during Phase I, French consultants were again selected to carry out the transport and investment studies in February 1977, and to review implementation of Phase I recommendations in the areas of personnel, inventory and accounting in February 1979. Implementation of the recommendations regarding operational and procedural improvements in the financial, planning, management and cost accounting, data processing and information systems was postponed to the Second Railway Project. (v) Consulting Services to MAT 3.17 The project also provided consulting Services to the Ministere de 1'Amenagement du Territoire (MAT) to improve the level and quality of transport planning in Madagascar. The team was coordinated by a general transport economist seconded by the Bank, who took up his duties in March 1975. The contract for consulting services was awarded in December 1975 to an American and French firm. The transport study team had to face continuous difficulties, which are explained in the chapter covering the institutional developments. The final cost of these studies was about 64% higher than appraised; this is mainly due to a quadrupling of local costs. - 42 - Reporting Requirements 3.18 Quarterly and annual reports were usually late. The first audit report was delayed due to complications arising for approvals required for appointing an outside auditor. Subsequent reports were sometimes delayed more than a year due to inadequate financial procedures and internal control, and changes in procedures introduced by the consultants. - 43 - IV. TRAFFIC AND OPERATIONS 4.01 Table 7 shows actual and expected traffic figures from 1973 to 1977 and actual figures for 1978 and 1979. Total freight traffic fluctuated between 750,000 and 830,000 tons per year, depending mainly on the amount of chromite carried. Excluding chromite, general freight traffic remained at about 650,000 tons p.a. during the whole period (or 200-210 million ton-km) and was about 30% below appraisal forecast in 1977. The traffic stagnation was the direct result of the depressed economic conditions. In 1978/79, chromite traffic dropped from 180,000 to 80,000 tons because of marketing difficulties. During the last few years, the railway also faced severe capacity constraints attributable to (i) a decrease in locomotive availability resulting from the poor quality of fuel after Madagascar had to change its source of supply; this increased maintenance needs; (ii) the loss of several locomotives and wagons due to derailments caused by sabotage and the deterioration of the track; and (iii) a more rigid scrapping program of locomotives than forecast. These capacity problems had repercussions on the supply of essential commodities to the capital city; in an attempt to resolve this problem, the Government imposed priorities in the movement of goods, which worsened the traffic movement by hampering normal operations. As a result, in 1979/1980, the railway could not meet the traffic demand because of heavy Government imports caused by a sharp increase in external borrowings. 4.02 Passenger traffic, on the other hand, grew faster than expected- at an annual rate of 7% - and was 22% above the appraisal estimate in 1977. This growth slowed down thereafter to about 4% per year due to the railway's capacity constraints. 4.03 Selected operating statistics are given in annex 10. Specific comments are as follows: (a) Locomotive availability remained at a very high level throughout the project period; the decrease in the last two years is due to a drop in fuel quality, and consequent lack of spare parts and maintenance problems. (b) Line locomotive productivity, measured by the number of km per available locomotives per year, declined from 80,000 km in 1973 to about 65,000 km in 1974775 when the new locomotives financed by CCCE were put in service; it increased to 92,000 km per year in 1977/78 when old, less powerful locomotives were downgraded to shunting service. It decreased thereafter due to the fuel problem. (c) Freight wagon availability also remained high during the whole project period. However, wagon productivity, measured by the number of net ton-km per year per available wagon, decreased in 1978/79 due to cancelled trains caused by lack of locomotives. This was partly - 44- compensated by an increase in load per train. Implementation of the scrapping program and major derailments in 1978/79 reduced the number of freight wagons to their 1972 level. (d) Despite the delay in delivering the new coaches, passenger traffic was allowed to increase by overcrowding passenger trains and delaying the scrapping program. (The number of passengers per passenger train in 1976, i.e., before new passenger coaches entered in service, was more than twice the 1972 level.) Passenger train-km decreased in 1978-79 due to the shortage of locomotives. (e) Staff productivity increased because of two factors: (i) reduction in personnel by attrition, and (ii) increase in traffic units since the number of passenger-km increased more rapidly than the decline in freight tonnage. 4.04 There was no Action Plan nor any specific covenant relating to operational improvements in view of the already high performance of the railway in operational matters. However, the consultant's terms of references included a review of the overall operations of the railway. During the second phase of the studies, this company proposed, and began implementing, a new transport plan to improve railway operations and rationalize train operations. This plan however, was never completely implemented because of the capacity constraints mentioned before, especially the decline in available traction. 4.05 Operational problems were closely followed up during supervision. Key statistics were prepared regularly and checked, and the consultants' recommendations were discussed. The sudden deterioration in the railway's operations during the last few years, however, (see Supervision reports on Second Railway Project) may show that although the operational performance of the railway was relatively good during implementation of the first railway project, there were latent weaknesses in maintenance procedures and transport operations, which went undetected. The railway was relatively efficient under "normal" conditions; but it could not react adequately when it was faced with "external" events (drop in fuel quality, traffic pressures and increased political interferences). fAsD LcA R . 14Att.JftV »AoneCT (Mr4 TNtr) ¥o. 70 so. . to jRALWAY PR©2cT- Kid aa6a esae --- AC,TVke 'f AAFPI, ...... A %1. Pm o3 GTi©NuS - 46) - V. FINANCIAL EVALUATION Operating Statements 5.01 Comparative income statoments and halance sheet summaries for years 1973-79 are given in Tables 8 and 9. In generil, RNCFM's operating results have not comei up to the levels forecast at appraisal; rather its financial situation deteriorated, especially during the last project years. After almost breaking even in 1974 and 1975, RNCFM hegan sustaining increasing losses; as a result, It failed to achieve the 2.5% annual return on net fixed assets through 1976 iind 3% thereafter as required tinder the Project Agreement. 5.02 Thu main reasons were: (M) traffic stagnation, (ii) higher operating costs, (Lit) inadeqttate tariff increases, and (Iv) unavailability in income from two sources previously forecasted. First, as explained in paragraph 4.01 traffic stagnated during the 1973-1978 period. Secondly, personnel costs, which constitute 50% of total operating costs, rose steeply after 1976: 22% in 1976, 15% in 1977, and 21% In 1978. This followed a period of leveling in salary costs due to the reduction in railway personnel. The railway claims that these increases were needed to keep up with general pay increases in the country and that the 1978 increase was necessitated by salary adjustments resulting from the railway's ne. personnel statutes and the increased track maintenance efforts. 5.03 Thirdly, introductions of tariff increases were implemented too late and on too small a scale. Tariff increases, which should have been Introduced in 1974 were not effective until October 1976; and the 10% for freight and 25% for passenger fares were insufficient. The Government was reluctant to approve higher tariffs because they were seen as contrary to its new social policies and inflationary. The project Agreement provided that the railway make appropriate changes in tariffs and tariff structures to reflect the cost of services and meet the rate of return covenants by a date not later than March 31, 1975. The Government and the railway also agreed to consult with the Association on the implementation of recommendations of the traffic costing study to be undertaken by the consultants. These covenants were not implemented because the Second Phase of the consultant's study was postponed; and they became a part of the Second Railway Project. 5.04 Fourthly, revenues from two sources forecast in the appraisal report did not materialize; (i) CIBA, a RNCFM subsidiary producing sleepers and utility poles, did not operate successfully and was unable to produce any net gains; and (ii) reimbursement for concessionary passenger fares granted by the railway under Government directives were not received on a current basis. - 47 - Balance sheets 5.05 Although deteriorating, the current and debt-equity ratios remained strong during the whole project period. However, these two ratios hid increasing problems in two areas. First, current assets and liabilities more than doubled during the 1973-1978 period. This was mainly due to the large and increasing amounts of long overdue accounts receivable from the Government departments and agencies which, in turn, forced the railway to delay its own payments resulting in an increase in accounts payable. Secondly, the debt-equity ratio deteriorated because of the increasing operating losses and the additional borrowings related to the Second Railway Projects and additional investments. 5.06 The Railway complied with the Covenant setting limits on investments outside the project and on additional borrowings. 5.07 Although the deteriorating financial condition did not seriously affect implementation of the first railway project, the cumulative effect of continuous losses and of accumulation of arrears affected RNCFM's cash flow to the extent that it has increasingly delayed its own payments to suppliers, social security fund and towards debt service. This started affecting the railways' operations in the last two years. (The difficult cash position compounded the problem of increased maintenance needs due to the deterioration in fuel quality.) Availability of local funds may also become a problem on the implementation of the Second Railway Project. General Comments 5.08 The alarming deterioration of RNCFM's finances, was the main concern of all supervision missions and was discussed at length during the oappraisal of the Second Railway Project. During negotiations on the Second Railway Project, the Government agreed (1) that the railway needed to be operated on commercial principles; and (ii) that it would take the following actions: (a) increase tariffs by 25% effective January 1, 1979; th) reimburse all amounts outstanding for more than 90 days due from Government departments and agencies, and pay thereafter all current accounts within 90 days, and (c) reimburse RNCFM for the amounts foregone on the account of concessionary passenger fares. However, tariff increases were again delayed by almost a year; Instead of (b) the Government made by treasury advance (which is unsatisfactory since the problem of receivables remains and continues to deteriorate); and the reimbursement of concessionary fares has not yet been implemented. These actions were again requested by CCCE as a condition for an emergency program to the railway in 1980; only partial reimbursement of outstanding receivables was made. 5.09 In fact, it is doubtful that any specific measures would have remedied RNCFM's general financial situation, since most of the problems stem from a lack of competent financial management. The railway continues to be managed as a Government agency and has no autonomy or accountability in financial matters. Under these circumrstances, the railway's management - 48 - does not hold itself responsible for the railway's finances and concentrates on operational matters. This problem is not unique to the railway, but is spread throughout all public agencies. It is doubtful, therefore, that in the uncertain political climate and in the context of continuous changes in basic policies during the project period, the successive supervision missions could have succeeded in overcoming these problems. - 49 - VI. INSTITUTIONAL DEVELOPMENT RNCFM Intended Developments 6.01 At the time of appraisal, RNCFM operated in a legal vaccuum, without any statutes or by-laws regarding its structure and organization. It was, in fact, run as an ordinary government department, subject to all civil service rules and financial and budgetary procedures. In January 1974, it was given the status of state-owned Public Industrial and Commercial Establishment at the urging of the Bank Group (a condition of credit effectiveness). 6.02 The other weaknesses seen at appraisal were: (a) lack of coordination among RNCFM's different departments, especially on planning matters; (b) lack of a management information system; (c) need to review the entire accounting system and to base it on commercial principles; (d) lack of trained personnel due to inadequate personnel policies; and (e) poor coordination in operational maz::rs. 6.03 Consulting services to RNCFM were provided in the project to review all these problems and make appropriate recommendations. Implementation of the recommendations was to take place during a second phase. At the end of this process, the railway would have had the management tools needed to function properly as an autonomous entity, particularly in financial, budgetary, personnel and general planning matters. This however presupposed a basic change in the institutional framework which did not take place. Actual Developments 6.04 It was soon evident that the decree setting up RNCFM as a Public Industrial and Commercial Establishment failed to achieve RNCFM's autonomy. The first supervision mission after the consultants submitted their report on Phase I of the studies, reported as follows in the report dated November 24, 1975: "In decree 74-154 of May 10, 1974, RNCFM was given autonomy as an Establissement Public a Caractere Industriel et Commercial. The railway is in technical matters supervised by the recently created Ministry of Transport and Supplies and in financial matters by the Ministry of Finance. The latter control is provided inter alia through an Agent Comptable, who is appointed by and is responsible to the Ministry. While interferance from - 50 - the Ministry of Transport in matters of railway operations is reportedly minimal, the Ministry of Finance is exercising a detailed control, which in some cases has substantially delayed measures proposed by the railway, e.g., in regard to tariffs and procurement. French Consultants to Phase I of technical assistance program presented the following principal recommendations: (a) modifying RNCFM's statutes and regulations so as to give the railway modern management methods; and (b) enforcing a new personnel statute taking into account the railway's special needs. The consultants emphasize that these recommendations form the basis for the modernization program formulated by them. If they are rejected or substantially modified, the implementation of the program and the expected results will be jeopardized. The consultants have, together with RNCFM and in accordance with the principal recommendations, drafted new statutes which have been submitted to the ministries involved for comments. The Ministry of Finance has firmly objected to the draft statutes, recommending that the existing system be preserved. This goes against any understandings previously reached with the Government, and, in the mission's view, creates one of tht most serious problems for the future of the Railway." Consequently, the second phase of the study was deferred except for assistance in specific areas not directly related to the question of RNCFM's autonomy. 6.05 New uncertainties surrounding the railway's autonomy arose afterward, when the Government decided to apply a new "Chvirte des Entreprises Socialistes" to all public bodies, including the railway. The charter, however, has not yet been implemented. Although its effects will not be known before it is actually implemented, the general principles of the charter recognize the principle of autonomous and responsible management and provide simpler control procedures by the Government. 6.06 Confronted with these continuous uncertainties, which face all public bodies in Madagascar, the Bank Group took the following approach during the negotiations of the Second Railway Project: (a) discussions concentrated on the question of tesponsibility over the Railway's financial management. It was agreed that, pending modification in the Railway's statutes, the Agent Comptable would still be responsible for RNCFM's finances, but would confine his role to post-audit work. At the same time, the railway would develop accounting procedures based on commercial principles and appoint a competent financial manager; (b) the second phase of the consultant studies would receive the highest priority; and - 51 - (c) the Government agreed not to make any changes in RNCFM's statutes without prior consultations with the Bank. 6.07 Although not completely satisfactory, this was probably the only possible approach in view of the prevailing political context and the urgency of the investments needed (which basically addressed deferred maintenance). However, it is doubtful that the railway's financial situation will really improve before a solution is found to the crucial problem of the railway's financial autonomy. The railway is still considered basically as a Government Agency as can be seen from the following factors: (a) the railway's finances are still controlled by an Agent Comptable who reports to the Ministry of Finance, and not to the railway's management; (b) cash is borrowed indiscriminately from the Ministry of Finance whenever required; (c) no action is taken if another Government Agency does not settle its accounts; and (d) tariff setting is still a very sensitive issue, which depends more on political and social considerations than on the costs of services and other commercial principles. In addition, the new "Charte" has still not been applied to the railway; a competent financial manager has not been appointed yet; and the traffic costing studies postponed to the Second Railway Project have again been delayed. 6.08 The project clearly did not meet its main institutional objective (financial responsibility), because it was premature and probably too ambitious. This failure raises a basic question over the adaptability of the proposed measures to the local environment; the Bank tried, in fact, to impose a concept of a commercially oriented enterprise in a sector usually regarded as public service in African and European countries; this difference in philosophy was even more accentuated in Madagascar which was in the process of implementing a new overall social approach to the economy. Partial success was, however, obtained in personnel management (a new personnel statute was implemented in 1977), investment planning, (the investment program prepared by the consultants served as a basis for the Second Railway Project), and to some extent in accounting nethods, (new accounting manual implemented, financial procedures being implemented in the Second Railway project). The objective of achieving efficient management and financial responsibility in the railway should be seen as a long term objective carried out through the perspective of successive efforts; in this context, the Second Railway Project is complementary to the First. - 52 - 6.09 The problem in a long-term perspective, however, is how to set realistic financial covenants, how to enforce them and how to adapt basic institutional reforms to local conditions without sacrificing ultimate management objectives. The Second Railway Project was approved despite the fact that most of the covenants in the first had not been complied with, leaving little leverage to enforce the new ones. In addition, in the case of Madagascar, what is lacking is an agreement or. basic management principles, or more precisely a firm commitment to implement them; without this, all financial covenants become to some extent useless. Assistance to Ministere Amenagement du Territoire 6.10 The project also provided technical assistance to MAT for transport planning. In general, the technical assistance team did not achieve its major objectives, which were to (i) improve the level and quality of transport planning in Madagascar; (ii) advise the Government on transportation policies; and (iii) train local counterparts. Instead, it only produced scattered studies of limited utility. 6.11 The reasons for this failure are well explained in the Back-to-Office report dated November 23, 1976: (a) MAT was split into three ministries, which lacked communication among themselves; (b) At the time of the consultants' arrival, the Government's policies were in a transition period; this led to a lack of clear directives and to frequent changes in the consultants' work program. This was followed by conflicts with the consultants' recommendations when the Government policies became clearer. (c) The Government never showed a keen interest in the Transport Coordination Study, as reflected by the poor quality of counterparts. (d) Disputes on administrative matters interfered with the working relationship between consultants and Government officials. 6.12 In fact, the major cause of this failure was the Government's lack of identification with the purpose of the study. This was due partly to the fact t.iat the need for intermodal transport coordination was not well perceived, since Madagascar was still trying to develop a basic transport network; partly to the transition period Madagascar was in during which time basic economic policies were uncertain; and partly to the fact that no senior Government Official was associated with the team from the beginning. It is likely that most of the difficulties could have been avoided if at the outset of the study a senior Government official had been selected to supervise the team's work. - 53 - VII. ECONOMIC REEVALUATION 3/ 7.01 The overall economic return on the projects was estimated at 35% at appraisal, based on a limited economic life of 10 years to account for possible traffic diversion to a new road which was to parallel the railway along its most important corridor. This return had been calculated only on the investments for which benefits could be quantified, i.e., track renewal, freight wagons and passenger coaches, and for which separate returns had also been calculated as follows: Economic Rate of Category of Equipment Expected Benefits of Return (a) Track Renewal (60km) Reduction in maintenance cost (27%) 20% Reduction in derailments (79%) (b) Freight Wagons (50) To transport 21.5 million ton-km of freight which 58% would otherwise have to take the road, from 1977 onwards, at a higher cost (c) Passenger Coaches (20) To transport 51.8 million passenger-km who would 40% otherwise have to be transported by bus, from 1976 onwards, at a higher cost. 7.02 A recalculation of the overall benefits has proved difficult for the following reasons: (a) there are major discrepancies between the revaluated and appraised benefits, especially relating to track maintenance costs, derailment costs, road vehicle costs and productivity figures (see discussion below); and (b) the situation of the railway was "abnormal" in the last three years (1978-80) because of external factors (see para. 4.01); this hampers calculation of a normal fleet capacity, which is crucial to determining actual benefits. 7.03 Based on appraisal figures and expectations, mainly traffic forecasts and the expected productivity increases, it is clear that the project was at best marginal and did not meet its objectives even though the planned road was delayed. Yet three important factors mitigate this somewhat negative conclusion: (i) The track renewal was necessary even at a 3/ See Economic Annex for detailed hypothesis. - 54 - lower than expected traffic level. The track was (and is) in very poor condition and deteriorating relatively quickly, imposing speed and load restrictions; the condition of the track proved to be a major bottleneck which has compounded the shortage of traction in the last three years. (ii) The expected increases in productivity levels were overly optimistic; productivity levels were based on their 1971 levels and were assumed to increase at 2% a year. However, after a major political crisis, all French technical assistance left the country in 1972, imposing a major reorganization in the railway. In addition, 1971 represented an abnormal peak when considering the long-term trend since 1960. (iii) The renewal needs of the railway, as calculated by the consultants in their investment program, exceeded what was expected at appraisal. However, even when considering these factors, the benefits of the project are still marginal as discussed below; a recalculated return would amount to less than 10%. Track renewal 7.04 The major benefits expected from the track renewal came mainly from (i) reduction in derailments (79% of benefits) and (ii) reduction in maintenance costs (21%). The average cost of a derailment was estimated at US$55,813, and maintenance costs on a non-renewed track at US$2,800 per km. Data given in the Second Railway Project show average derailment costs at US$5,250 and maintenance costs at US$700 per km, at 1978 prices versus 1973 prices in the First Railway Project. Such large differences cannot be accounted for as project benefits since only 60 km over a 350 km rail line were renewed. It is more likely that they result from discrepancies in the generally poor costing base available, and that they were overly influenced by extraordinary events; for instance, major derailments. 7.05 Despite the delay in the implementation of the track renewal program and the stagnation in traffic, it is likely that the benefits of this component were still substantial. It was reported that no derailments occurred on the new track and that speed restrictions had been lifted on this portion, while increased speed restrictions and derailments occurred on the rest of the track. A 16% return was calculated for track renewal in the Second Railway Project. A sensitivity analysis based on traffic stagnation still showed a return of at least 10%. This is probably conservative when considering the reduced capacity the railway faced during the last three years, its effect on the general economic activity in Madagascar, and the delay experienced in the construction of the road. Freight Wagons 7.06 The return on the 50 freight wagons amounted to 58% over a ten- year period in the Appraisal Report. The new wagons were to replace older wagons which needed to be scrapped. The Railway's fleet capacity was also expected to increase as a result of better productivity. Three factors, however, affected these particularly good results: (i) the general traffic did not increase as expected -- in 1980 it was at 64% of its 1971 level; (ii) the cost of the wagons was 65% higher than estimated at appraisal; and (iii) wagon productivity dropped between 1971 and 1972 from about 255,000 - 55 - net ton-km to about 200,000 net ton-km, declining with traffic from that level thereafter. 7.07 If the wagon produccivity had stayed at its 1971 level, the existing fleet, without the 35 wagons financed under the project, would have been sufficient to handle the actual traffic even with the planned scrapping program. The capacity of the railway would have still been sufficient even with the increased renewal program of about 160 sixty-year-old wagons as recommended by the consultants. 7.08 Based on a "normal" productivity of about 200,000 net ton-km per wagon, the railway would also have had sufficient capacity to meet actual traffic, because of the decline in traffic during the last three years. Since this decline was caused by a lack of traction, the benefits from the wagons will only start when the traction problem is resolved, i.e. towards 1982. Based on the latest information on road cost, the economic return on the 35 wagons would amount to 10%, mainly because no benefits are accounted for during the first five years. (Had the wagons been purchased in 1982, the return would have amounted to 22%, showing in fact that the investment is retrospectively five years premature.) This is probably a conservative figure since the new wagons were actually put in service immediately, allowing the railway to drop older and less efficient wagons from service; the benefits in this case, however, are difficult to quantify since they would come from differences in operating costs, on which no reliable data are available. Passenger Coaches 7.09 The return on passenger coaches was estimated at 40% at appraisal. In 1975, the economic return was recalculated at 16% in view of the higher cost of the coaches. In the Second Railway Project, the return on 10 additional passenger coaches amounted to 10%; they were not proposed for Bank financing. 7.10 The re-evaluated rate of return on the coaches procured under this project is negative despite a higher demand than expected at appraisal. This return results from the difference in road operating costs, since the latest available figures indicate that the cost per passenger-km will be higher on the train than on the road as soon as the new paved road is complete, which is expected to be in 1982/83. Summary: 7.11 In summary, the low recalculated rate of return comes from (i) lower than anticipated traffic levels and (ii) lower productivity. They were only partly compensated by a higher scrapping program than forecast. While the track renewal program was needed, the provision of wagons was premature and the provision of passenger coaches economically unjustified. Although this recalculation is probably on the conservative side, the project clearly did not meet its objectives; this could not have been foreseen at appraisal in view of past trends, although the rapidity expected in economic recovery was too optimistic, and the productivity - 56 - levels used are questionable. The return on passenger coaches depends heavily on unreliable operating costs data. One clear lesson of this project is not to expect rapid economic recovery after major political unrest. - 57 - VIII. ROLE OF IDA 8.01 Despite the shortcomings of the project, IDA played a useful and constructive role, especially during the supervision process. Although there were problems in achieving the institution-building objectives, institution-building is a long process, which should be viewed in the context of a continuous dialogue. In this perspective, the working relationship between IDA, the Government and the railway has been excellent. The supervision missions played particularly a useful role in keeping the dialogue alive and in taking an objective, intermediary role between the railway and the Government. 8.02 IDA was perhaps too lenient on the ,uestions of the financial measures needed and financial responsibility of the railway. This is, however, a matter of judgement which needs to be addressed in the context of all Bank Group operations in Madagascar in the existing political context, and in relation to the role of the railway in Madagascar's economy. As already pointed out in the financial and institutional chapters, what was lacking was a commitment of the Government to implement agreed management principles; in such a context, it is difficult to judge whether a change in the railway's statutes imposed from outside would have produced the hoped-for results. Alternative approaches, however, (such as the French "Contrat Programme") could perhaps have been tried , although it is not sure that these would have met with success, considering the unstable political climate in the country and the vital role of the railway in Madagascar's economy. 8.03 IDA was able to accomplish little regarding the technical assistance to Ministry of Transport, because of the lack of Government identification with the objectives of the study team. A real dialogue on transport issues never took place in Madagascar. The major conclusion of this unfortunate experience is perhaps the necessity of having a senior Government official directly responsible for such a team in future projects. 8.04 The design of the Second Railway Project benefitted from lessons learned during the first. The Second Project is still a holding operation aiming at improving the railway's management and operations. It focuses on the implementation of recommendations outlined in the first project and emphasizes simpler procedures to be implemented by local consultants. The physical elements of the project were based on the consultants' studies and took into account the actual implementation capacity of the railway. - 58- IX. CONCLUSION 9.01 The project was a valid starting point for improving the operations of the railway, considering that it was the introduction of IDA in this subsector. The scope and the content of the project were, in general, appropriate at the time it was appraised, but the objectives were probably premature and too ambitious. 9.02 The objectives of the project were not realized. This is mainly due to the unstable political climate, the changes in basic Government policies and the economic stagnation which occurred in Madagascar during the project period. This general environment makes it particularly difficult to evaluate what would have happened without the project; but it is clear that (i) the physical portion of the project was only partly justified -- while traffic did not increase as expected, the investments served partly to offset a deteriorating situation in the railway; (ii) the consultant's work to improve the railway's management had only limited results in specific areas, such as in personnel policies, but failed to improve markedly the railway's legal structure, management procedures, finances and operations; and (iii) the objective to set up a general framework for transport policies in MAT was not realized. 9.03 Despite these shortcomings, the project was useful and IDA played a constructive role. First, the above judgement should be mitigated since most of the benefits have been delayed; the exact benefits should therefore be evaluated after the Second Railway Project is fully implemented. Secondly, it is our belief that IDA played a major role, though unquantifiable and with no immediate results, by maintaining an open dialogue on major policy issues, which eventually will help the Government reshape its policies, and by providing practical advice to the railway. 9.04 One major lesson to be learned from this project is probably the need to carefully analyze the impact of the overall political and economic environment on a proposed project and the need to tackle basic issues in a broader perspective than the project itself. This involves particularly a need to reach agreement on the basic problems, as acknowledged by both parties, and on the means to tackle them. The impact of the environment however, does not explain all of the project shortcomings. It was clear that the railway needed strengthening; this could have probably best been achieved by giving more attention to practical management and operational procedures than by providing an overall study of the whole organization and trying to impose a completely new structure from outside. While the railway was relatively well run under normal circumstances -- by comparison with other railways in Africa, and considering that it is operated without outside technical assistance -- the experience of the last three years has shown that it was unable to react to external events which led to a crisis in 1979/1980. This could have perhaps been avoided or alleviated by a well-designed training program and some short-term operational assistance. MADAGASCAR CREDIT 488-HAG PROJEG? CWMPLETION REPdRT PROJECT COST AND FINANCING (FMG Million) Appraisal Estimate Actual Local Foreign Total Local Foreign Total I. RNCFH Investments A. Works 1. Track reneval (60 ka) 130 0 130 159 0 189 2. Other 199 0 199 200 0 2001/ B. Materials 1. Freight wagons 2/ (50) 8 187 195 13 276 289 2. Passenger Coaches/ (20) 4 260 264 72 720 792 3. Track Material 40 356 396 0 455 455 4. Miscellaneous quipment3/ 72 64 136 1 49 50 5. Ballast wagons21 - - - 2 88 90 Consulting servicesV 45 140 185 13 130 143 D. Contingencies Physical (102 A2) 20 - 20 - - - Price (7Z) 55 111 166 - - - Total 573 1,118 1,691 490 1,718 2,208 II. Consultini Services to HAT 42 167 209 183 161 344 Total Project Cost 615 1,285 1900 673 1,879 2,552 1/ Unavailable; appraisal apprairal estimates taken for comparison purposes. 2/ Freight wagons were reduced to 31 and passenger coaches to 15; of the latter, t!'e Bank Group only financed 5, and CCCE 10. (included) 3/ Only ballast tamper; set of automatic barriers deleted from project. 41 Part of consulting services was reallocated to purchase 12 ballast wagons, not included at appraisal. HADACASCAR CREDIT 468-MA; PROJECT COMPLETION REPORT PROJECT IMPLEETATION Tender Award of Docuuman Contract Start of Completion of Approved Invitation Approved by delivery delivery Credit CategorY By IDA to bid IDA 1 Contract date (or study) (or study) Actual Actual Expectedl Atual Expc ted- Actual Expected Actual Exvected- Actual Expected- I Covered freight vagone 4/74 5/74 4/74 9/74 8/74 3/75 6/74 7.'76 4/75 9/76 4/75 1 Passenger coaches 5/74 1/765-/ /74 6/76 9/74 11/76 9/74 11/77 9/75 12/77 12/75 I1 Track material 10/73 11/73 11/73 3/74 3/74 4/74 3/74 7/74 9/74 .3175 11/74 11 Welding material 10/73 11/73 11/73 9/74 3/74 3/75 - - - 9/75 - III Auto ballast tamper 4/74 6/74 5/74 1/7 7/74 3/75 8/74 - 2/75 10/75 2/75 III Automatic barri%a=2/ - 12/73 12/73 - - - - - - - Ballast vagons -' 3/78 - 6/78 - 9/78 - - - 4/79 - IV RNCFM Consultants Study c ehase I - 12/73 12/73 7/74 4/74 7/74 6174 8/74 5/74 8/75 5/75 Phase Id/ - - 3/75 - 7/75 - 7/75 2/77 5/77 6/76 V Techn. Aest. Transport Ministry - - - - - 12/74-8/75 - 2/75-10/75 - 8/78 - / Estimates from first supervision report dated March 11, 1974. / Cancelled / Reallocation - not included in:,project design. 4/ Not complete. 51 Second invitation to limited number of suppliers AEAGASCAR CREDIT 488-'AC PROJECT CCMLFTION REPOIT COMPARATIE TRAFFIC STATISTICS 1972 19?3 1974 1975 1976 1977 1975 1979 Actual Actual EOxected Actual Expected Actual Eetected Actual Ected ActcaExeted Actual Actual Freight traffic V('000T) Iwats* General 129 - 171 172 192 132 216 128 227 155 239 152 133 Petroleus products 138 - 120 138 132 130 145 133 159 139 174 133 166 Total 267 - 291 306 324 262 361 261 366 294 305 301 Rx"rts General 22 - 13 15 14 11 16 25 17 24 19 20 18 Chrowite 110 118 123 158 170 184 170 186 170 135 170 75 85 Total 132 - 136 173 184 195 186 211 187 159 189 95 103 Local 357 - 343 333 384 374 430 347 458 343 487 348 i11 Total 759 750 770 811 892 831 977 816 14029 795 1069 749 715 Total ton-ka (HN) 247 252 258 272 300 280 329 273 348 264 369 241 232 of vbich chromite Wr 17 3 -w 57 *? Others 204 205 210 210 233 207 262 200 281 211 302 192 198 Passengers Nuaber ('000) 2,580 2,951 2,780 3,607 2,8640 3,710 2,t,49 4,265 3,135 3,970 3,129 4,115 4,042 fassenger XI 192 208 203 254 205 248 211 293 217 276 225 296 303 1/ Excluding service traffic MADAGASCAR CREDIT 488-MAG PROJECT COMPLETION REPORT INCOME STATE2ENTS (FMG Million) 1973 1974 1975 1976 1977 1978 1979 Actual Expected Actual Expected Actual Expected Actual Expected Actual Expected A Actual Operating revenues Freight 2,038 1,915 2,129 2,195 2,190 2,440 2,332 2,589 2,548 2,764 2,450 2,653 Passengers 534 546 646 562 716 578 799 596 913 613 975 1,155 Miscellaneous 213 222 178 229 189 236 162 243 133 250 397 435 Concessionary fares - - - 100 - 103 - 106 - 109 - - CIBA revenues - - - 60 - 90 - 95 - 100 - - Tariff Increase - - - 107 - 156 - 165 - 174 - - Total 2,785 2,683 2,953 3,253 3,095 3,603 3,293 3,793 3,594 4,010 3,822 7 Operating Expenses Personnel 1,338 1,300 1,418 1,431 1,418 1,502 1,741 1,577 2,000 1,656 2,424 2,396 Material 600 570 685 628 674 690 768 760 875 836 1,181 887 Other 260 200 198 210 196 221 175 232 192 244 234 735 Depreciation 635 605 627 636 736 757 713 787 726 817 746 705 E Total 2,33 2,675 2,928 2,905 3,024 3,170 3,397 3,56 3,798 3,553 4,585 777 Net Operating Income (48) 8 25 148 71 433 (1041 438 (2041 457 731 20 Non Operating Items Interest Changes 45 81 57 116 82 158 139 172 143 162 180 197 Provision - 25 - 82 - 50 - 50 - 50 - 45 Other (revenues) - - 8 - 43 - Total 45 106 65 198 125 208 152 222 95 212 (155) 242 Net Income (93) (9) (40) (150) (54) 225 (256) 216 (299) 245 (608) 222) Working Ratio 0.79 0.77 0.78 0.70 0.74 0.67 0.82 0.68 0.85 0.68 1.00 0.83 Operating Ratio 1.02 1.00 0.99 0.89 0.98 0.88 1.03 0.88 1.06 0.88 1.20 1.00 Return of fixed assets (Z) - 0.5 0.2 2.5 0.4 2.9 - 2.8 - 3.0 - September 1980 0e MADAGASCAR CREDIT 488-MAG COMPARATIVE BALANCE SHEETS SUMMARIES1 (FMG Million) 1973 1974 1975 1976 1977 1978 1979 Actual E xected G ZeiaaL eEA d ExiAncu Expected Ata E.cted .dB. A&tual Assets C.arret Asset- - 1,284 1,704 1,471 2,076 1,543 2,303 1,626 2,244 1,746 3,226 3,968 Net Fixed Assets 13,749 16,336 15,377 16,311 15,605 16,486 15,458 16,422 15,281 16,196 17,164 Deferred Charges 85 - 181 75 218 72 221 86 181 67 62 Total 15,118 18,040 17,029 18,462 17,366 18,861 17,305 19,252 17,208 19,488 21,194 Liabilities Current Liabilities 757 952 1,094 817 865 1,132 887 1,533 858 2,224 3,041 Long term debt 1,981 1,357 3,405 1,999 3,746 2,351 3,447 2,670 3,134 2,921 4,165 Equity 12,280 15,731 12,530 15,646 12,755 15p378 12,973 15,049 13,216 1,4 13988 Total 15,118 18,040 17,029 18,462 17,366 18,861 17,305 19,252 17,208 19,488 21,294 Debt/Equity ratio 14,186 8/92 21/79 11/89 23/77 13/87 21/79 15/85 19/81 17/83 23/77 Current ratio 1.70 1.79 1.34 2.54 1.78 2.03 1.83 1.79 2.03 1.45 1.30 1/ Audited accounts since 1974 August 1980 劉。1,&&&‘弓。；&&&&&&&”藝藝籐籐，&&,&&,&&&;,；。，:，審：;::;;;;&,& 蔔”，惡”&&&&,&&,&&&&&&“投，,&&&&&&&&&”一“,&&&:;,&&&&:&&&:: 實―．丰啊編齋寫馨弓“,&&&,&,&,&“界套界界：召邊：森義：&,‘召”召”&&‘日：，甭，籐籐讓寫”&& 莖―‘籐‘&&”弓莽寫召丰臼霄投日萬償發斤寫露邑藝界：：讓‘二：：翁‘&:：雙提視︰”丰諜”召朵梁梁姿雜”&, 邑―遺弓病””諱邑界姿震造驚誠很發日誠栽表露套邑界暴讓：二屆讓：：為，&：華：養織：‘藝嬝籐蘇梁讓梁文森鳥，& &&&:&-！〕〕！．〕！,,！〕！．〕〔〕〕！〕！〕〔！。〕〕痲〔