Colombia - Sixth Railway Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 4670 THE WORLD BANK PROJECT PERFORMANCE AUDIT REPORT COLUMBIA: SIXTH RAILWAY PROJECT (LOAN 926-CO) August 22, 1983 Operations Evaluation Department 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.  FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT COLOMBIA SIXTH RAILWAY PROJECT LOAN 926-CO) TABLE OF CONTENTS Page No. Preface ................................................................. i Basic Data Sheet ................................................... ii Highlights ......................................................... iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. BACKGROUND .................................... ..1...... 1 II. THE PROJECT ......................... 3 III. PROJECT IMPLEMENTATION AND RESULTS.......... ............ 6 IV. CONCLUSIONS .... 11 Tables 1. Appraisal Estimates and Actual Costs...........14 2. Expected vs. Actual CNR Performance..................... 15 PROJECT COMPLETION REPORT I. Introduction.............................................. 17 II. Project Preparation and Appraisal......................... 18 III. Project Implementation and Cost............................. 19 IV. Traffic and Operations.............................. 26 V. Financial Performance ......................... .. ......... 28 VI. Economic Re-evaluation........................................ 32 VII. The Role of the Bank.... .................. ......... . ... o 34 VIII. Conclusions................. .... ........................ 35 Tables 1. Achievement of CNR's Program of Action 1973-1976.......... 37 2. Actual and Expected Physical Completion... ............ 38 3A. Actual and Appraisal Estimates of Investments............. 39 3B. Actual and Appraisal Estimates of Project Costs........... 40 4. Selected Operating Statistics 1970-1979.........oo ...... 41 5. Actual Revenues, Expenses and Net Income..... ......... ..42 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 au horization. Table of Contents (cont'd) Page No. 6. Summary Balance Sheets.................................... 43 7. Actual and Appraisal Expectation of Financing of Investment Plan............................ ........... 44 8. Uneconomic Lines and Passenger Train Services Closed...... 45 9. Traffic Diversion Effect...........46 10. Total Transport Cost "With" and "Without" the Project................................................ 47 Map PREFACE The following is a performance audit report on the Sixth Railway Project in Colombia for which Loan 926-CO (US$25.0 million) was made in August, 1973. The project was considerably delayed and had to be cut back significantly to accomodate cost overruns and shortages of funds. The Loan Account was closed in August 1980, but disbursements under the Loan continued until November 1980 when the last disbursement was processed and funds were exhausted. The attached Project Completion Report (PCR) was prepared in 1981 by the Latin America and Caribbean Regional Office with the assistance of the Borrower and is based on information furnished by the Borrower, as well as that collected during normal supervision of the project. The Operations Evaluation Department (OED) has reviewed the Minutes of the Executive Directors' Meeting at which the project was considered, the President's and Appraisal Reports, project files and the PCR, and has inter- viewed Bank staff familiar with the project. Based on this information and on data that has become available since the PCR was written, the audit dis- agrees with the conclusions of the PCR and reassesses the likely project outcome in the attached Audit Memorandum. Salient aspects of the project's experience are summarized in the Highlights. No mission was undertaken by OED staff to audit this project. The draft audit report was sent to the Borrower for comments; however, none were received.  - ii - BASIC DATA SHEET KEY PROJECT DATA Actual or Appraisal Current Item Estimate Estimate Total Cost of Investment Plan (US$m) 95.6a 61.6/a Total Project Cost (US$m) 45.9 34. /a Investment Plan Cost Underrun (%)- 3 /a Project Cost Underrun (%) - 2a Loan Amount (US$million) 25.0 25.0 Disbursed - 25.0 Repaid to January 1983 3.5 Outstanding to January 1983 - 20.4/b Date Physical Project Components Completed 12/75 08/80 Proportion Completed by Original Date (%) 100 55 Proportion of Time Overrun (%) - 350 Economic Rate of Return on Investment Plan (%) 12 Negative/c Financial Performance Poor Institutional Performance Poor Cumulative Estimated and Actual Disbursements (US$ million) FY74 FY75 FY76 FY77 FY78 FY79 FY80 FY81 Estimated 11.0 25.0 - - - - - - Actual 1.2 10.6 17.4 23.2 24.2 24.7 24.8 25.0 Actual/Estimated (%) 11 42 70 93 97 99 99 100 OTHER PROJECT DATA Item Original Plan Actual First Mention in Files 05/70 Appraisal 06/71 Negotiations - 06/73 Board Approval 07/73 08/73 Loan Agreement 08/73 08/73 Effectiveness 10/73 12/73 Closing Date 06/76 08/80 Borrower Ferrocarriles Nacionales de Colombia Executing Agency Ferrocarriles Nacionales de Colombia Fiscal Year of Borrower January 1-December 31 Follow-on Project Seventh Railway Project Loan Number 2090-CO Amount (US$ million) 77.0 Loan Agreement Date 12/28/82 /a Figures are meaningless for comparison purposes because many project items were cancelled or only partially implemented (see Table 1). /b Includes an exchange adjustment of -US$1.1 million. 7c The PCR gives an ERR of 11%; however, based on updated information vis-a-vis the PCR, the audit has concluded that the project yielded a negative rate of return (PPAM, para. 31).  - iii - MISSION DATA Month/ No. of No. of Staff- Date of Mission Year Weeks Persons weeks Report Preappraisal 10/70 0.5 2 1.0 11/02/70 Preappraisal 03/71 0.5 1 0.5 04/21/71 Appraisal 06/71 2.0 4 8.0 07/13/71 Appraisal 10/71 2.0 3 6.0 11/10/71 Reappraisal 12/71 2.0 6 12.0 02/23/73 Supervision I 08/73 1.0 2 2.0 09/21/73 Supervision II 02/74 2.0 3 6.0 03/22/74 Supervision III 05/74 2.0 3 6.0 06/05/74 Supervision IV 07/74 1.5 3 4.5 07/29/74 Supervision V 09/74 0.5 1 0.5 10/08/74 Supervision VI 11/74 2.0 3 6.0 11/25/74 Supervision VII 02/75 2.0 3 6.0 02/25/75 Supervision VIII 05/75 2.0 2 4.0 06/05/75 Supervision IX 08/75 3.0 2 6.0 09/17/75 Supervision X 11/75 1.5 1 1.5 11/20/75 Supervision XI 03/76 2.5 3 7.5 04/09/76 Supervision XII 05/76 1.5 3 4.5 05/26/76 Supervision XIII 09/76 1.5 2 3.0 10/15/76 Supervision XIV 02/77 1.5 4 6.0 03/17/77 Supervision XV 08/77 1.0 3 3.0 09/22/77 Supervision XVI 01/78 1.0 1 1.0 02/17/78 Supervision XVII 11/78 2.0 2 4.0 12/05/78 Supervision XVIII 04/79 2.0 5 10.0 05/22/79 Supervision XIX 08/79 1.0 3 3.0 09/12/79 Supervision XX 12/79 1.5 3 4.5 01/23/80 Supervision XXI 02/80 0.5 3 1.5 03/03/80 CURRENCY EXCHANGE RATES Name of Currency (Abbreviation) Colombian Peso (Col$) Appraisal Year US$1 = Col$ 22.50 Intervening years' average US$1 = Col$ 36.16 Completion year average US$1 = Col$ 48.24  - iv - HIGHLIGHTS The project was not successful. It did not achieve its main objec- tive of rehabilitating CNR-s main lines or improving operating conditions. Continued excessive derailments, which caused much damage to motive power and rolling stock, and poor locomotive availability were major factors in prevent- ing an improvement in operations and in the quality of service. Thus, the railway continued to lose traffic to highway transport and revenues to de- cline; as a consequence, increasing financial losses were incurred. CNR's failure to reach the expected financial targets during the project period imposed an undue burden on the -Government's budget and, while 'Government transfers were of a considerable magnitude they were neither timely nor sufficient. The resulting shortage of funds and cash flow problems caused CNR's working capital to deteriorate and led to the project being implemented over five years instead of the two forecast; financial problems also led to important project items being cancelled. The following points are of particular interest in connection with this project: - a key objective of the project was to reduce derailments by means of track rehabilitation and improvements in operating practices. To improve the track it was expected to relay 576 km on concrete sleepers (SAR, para. 4.05) and on 434 km to use good quality wooden sleepers and ballast (SAR, para. 4.10). Only 3 km of concrete sleepers were laid under the project on a trial basis; on the 417 km of track that were rehabilitated poor quality timber sleepers were used (PCR, para. 3.04). The record is not clear on the extent to which the loan provided financing for concrete sleepers or the reason why they were ultimately not purchased. The loan, however, made specific provision for purchasing locally produced timber sleepers. Delays occurred in the purchase of ballast crushing machines, compactors and cars, and rails and rail welding equipment, all of which delayed track rehabilitation (PCR, para. 3.04) and prevented the project from improving operations and reducing costs; - wagon turnaround time was expected to improve as a result of the project. While there was some improvement in 1974 and 1975, the turnaround actually worsened overall during the project period to 1981, mainly due to delays at the ports (PCR, para. 3.02) where the 'Government failed to provide rail facilities in a timely manner, or guarantee that wagons moved out of port areas quickly (PPAM, paras. 23 and 36); - the railway suffered from lack of competent management, partly due to frequent senior and middle management changes (PCR, para. 3.02), but also due to insufficient Government institutional support (PCR, para. 6.06 and PPAM, paras. 27, 35 and 36); - Government funds to finance project investments and other expenses while increasingly significant (PPAM, para. 30) were insufficient and delayed (PCR, para. 5.02 and PPAM, para. 36); - V - - the economic reevaluation presented in the PCR shows an economic rate of return (ERR) of 11%, about as high as the appraisal forecast of 12% (PCR, para. 7.09). Based on information that has become available since the PCR was prepared, the audit questions this estimate as the project did not contribute to reducing the rail- way's financial, or the country's economic losses generated by CNR (PPAM, para. 31). The audit estimates the project to have yielded an overall, system wide, rate of return which would be substantially lower, probably negative (PPAM, para. 31). However, system wide averages used both in the PCR and by the audit may mask any benefi- cial impact the project may have had on specific lines or services. No information has been presented in the PCR to identify these or the extent of any beneficial impact of the project on them (PPAM, para. 33). Further, to the extent that capacity constraints can be alleviated, there would be a beneficial effect on the overall rate of return, as visualized by the Region (PPAM, para. 31). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM COLOMBIA SIXTH RAILWAY PROJECT (LOAN 962-CO) I. BACKGROUND 1. The Andes Mountains present formidable barriers to communication among the main population centers in Colombia. For this reason, the country is constituted by separate and almost self-sufficient regions. It was not until the early 1960s that, under a drive toward integration and moderniza- tion the transport system began to evolve into a national network. 2. The Bank'Group has had a considerable involvement in the development of the Colombian transportation system. Eight loans and one credit have been made for highways since 1951, totalling US$375.5 million. In addition, US$19.9 million have been lent for rural roads in five agricultural projects. Seven loans have been made for railway development, including the project under review, totalling US$196.6 million. One loan was made for domestic aviation (US$61.0 million) and an IFC loan (US$15.0 million) was granted for an oil pipeline. The grand total is US$648.1 million. 3. Until the early 1950s, the railroad system consisted of two separate networks: the western system, linking Puerto Berrio on the Magdalena river, to the cities of Medellin, Cali, and Buenaventura on the Pacific Ocean; and the eastern system from Bogota to Puerto Salgar (also on the Magdalena river) and south to Barbosa, Belencito and Neiva. The first three Bank railway loans (68-CO for US$25 million in 1952, 119-CO for US$15.9 million in 1955, and 267-CO for US$5.4 million in 1960) helped finance construction of a line from Puerto Salgar north passing through Puerto Berrio to the port of Santa Marta on the Atlantic. This line, known as the Atlantic Railroad, not only connected the eastern and western systems, but also joined the Atlantic and Pacific coasts to each other via Medellin, and to Bogota. However, the railway system was not connected to the Atlantic ports of Barranquilla and Cartagena and this, together with the fact that trucks can operate more easily in mountainous terrain, limited the scope of railway operations. While in the mid 1960s the railway handled close to 17% of total freight traffic in the country, by the mid 1970s the railway share of traffic had decreased to less than 10% and since then it has further decreased to less than 5%. 4. In 1963, the network was consolidated and an integrated system was created by standardizing the track gauge to 914mm. The fourth and fifth loans for railways (343-CO for US$30 million in 1963, and 551-CO for US$18.3 million in 1968) helped finance the 1963-65 and 1968-70 segments, respec- tively, of a Ten-Year (1963-72) Investment Plan for the system consolidated under the Colombian National Railways (CNR). - 2 - 5. Experience under the first five projects was summarized in "Bank Operations in Colombia: An Evaluation" (OED Report Z-18 and its Annexes, May 1972), and in the Project Performance Audit Report on the Fifth Railway Project (OED Report No. 776 of June 1975). The reports noted that CNR's performance had been poor, that substantial delays and cost overruns had been experienced, and that traffic had grown significantly less than anticipated. In addition, financial support from the'Government had been inadequate, amounting to only 67% of the funds expected in the period. Because of these negative developments, the Bank delayed considering the Sixth Project now being audited, by over a year, to induce CNR management and the Colombian 'Government to take steps to improve performance. During this waiting period traffic increased somewhat, a new cost-based tariff was introduced, CNR's operating ratio improved to below 100, and the 'Government provided needed funds. However, other deficiencies, as noted below, persisted. The sixth project was approved in August 1973. 6. On completion of the Fifth Project progress on track rehabili- tation was observed to have been slow, amounting to only 79% of the planned figure even after a two year extension of the project. Delays were mainly due to slow delivery of rails, low deliveries of ballast and poor procurement practices which led to reduced deliveries and poor quality of sleepers. A one year delay had occurred in the procurement of freight cars mainly due to problems with their specifications. Delivery of 60 diesel locomotives was on schedule but overheating problems were experienced due to poor engine specifi- cations on a 60 km main line section with high gradients. Workshop construc- tion was delayed - in fact, the workshop for rolling stock repair had not been built at all. CNR had been given access to yard facilities at the port of Buenaventura two years behind schedule and closure of uneconomic lines had been postponed. Appointment of consultants to assist in improving operations had been delayed three years. Derailments stood at an average of 340 per million ton-km mostly because of lack of control of train operations, failure to strictly enforce train operating regulations, lack of incentives for engineers to avoid derailments, and poor track conditions. Workshop capacity was hampered by a shortage of experienced staff and their high turnover. However, availability of motive power and rolling stock had, in general, improved due to the acquisition under the project of 60 new diesel locomotives and 1,200 new freight cars. Equipment utilization was still low and uneven and while car and train loads increased (due to the higher capacity of the new cars and the longer trains that could be operated on the Atlantic line with the new locomotives), car turnaround times had also increased. CNR's finan- cial performance had not improved as anticipated - in fact, it had deterio- rated because traffic did not rise as expected, tariff increases did not cover inflation, and working expenses had increased by over 80%. Cash generation was consequently much lower than expected and CNR had to borrow heavily from local banks to stay financially solvent. - 3 - 7. A major comment of the 1975 PPAR related to the adequacy of Bank supervision. In particular, it was emphasized that the Bank appeared to rely completely on the use of consultants to achieve operational improvements. It was noted that a more active role by the Bank could have accelerated adoption of the necessary steps to improve operating efficiency. It was also remarked that the Bank had overestimated CNR's ability to retain staff trained by the consultants. In this respect, a more comprehensive approach to man- power and management problems was seen as necessary. The audit suggested that the Bank put more emphasis on improving operating efficiency. Bank super- vision missions were noted to have concentrated mostly on finances and procurement (admittedly problem areas) and less on operating performance. It was thought that a "Plan of Action" with specific time-phased targets would have been useful. II. THE PROJECT 8. Processing of the project involved two preappraisal missions, two appraisal missions, and a re-appraisal mission, over the period from October 1970 to December 1972, and loan negotiations were held in June 1973. The long gestation period of this project reflected not only the doubts and serious questions raised within the Bank about CNR and its role within the transport system of the country, but also indecision on the part of the 'Government (in particular, the Planning Department) on whether or not to finance the continuing rehabilitation of the railway and the service of the debt contracted by CNR (IBRD, and other foreign and local debt). The'Govern- ment was particularly concerned at the prospect of having to provide large sums of money to continue to support CNR's operations in the very likely event that revenues would fail to cover (as had happened before) administrative and operating costs. 9. CNR's investment plan for the period 1973-76 was reviewed by the Bank and, after considerable modifications during project preparation, apprai- sal and reappraisal, a plan was agreed upon estimated to cost US$94.0 million, with a foreign exchange component of US$60.4 million. Of total costs, over 50% was for track rehabilitation and civil works, 23% for motive power, rolling stock and workshops, and the remainder for telecommunications, freight handling equipment, dwellings and consulting services, plus provision for the repair of damages caused by heavy rains. The project was tailored to assist in implementing the first two years of the plan (1973-1974) and included a Program of Action to improve railway operating efficiency and its financial performance. Special covenants in the Loan Agreement required the'Government to provide, in a timely manner and in sufficient amounts, the funds necessary to implement the project. Covenants also required CNR to carry out a tariff, traffic and investments study in the Magdalena River valley (where about 80% of CNR's traffic was carried), to carry out a management study, and to imple- ment actions leading to the closure of uneconomic lines and services. 10. Investments under the project were estimated to cost about US$44.2 million equivalent with a foreign exchange component of US$32.2 million. These investments included materials (rails, sleepers, fastenings, switches, bridge parts, etc.), rail welding equipment, ballast cars, 28 diesel locomo- tives, spares for locomotives and rolling stock, trucks and trailers for a road/rail operation, freight handling equipment, telecommunications equipment, equipment for the repair of rain damages, as well as consultant and training services. However, the Bank loan (US$25.0 million) was to finance only part of the cost of materials and equipment for track rehabilitation and mainte- nance (62%); locomotive and freight car spares (15%); and freight handling equipment, trucks and trailers, telecommunications equipment, equipment for rain damage repair, and consulting services (23%). Due to CNR's poor finan- cial situation, the loan was also to finance commitment and interest charges during the project period (about US$1.6 million). 11. The project focused on specific solutions to CNR's principal problem areas. Most importantly, track condition was to be improved by the use of concrete sleepers and welded rails which, together with adequate crushed stone ballast, would provide a more stable roadbed and, in turn, reduce derailments and track maintenance costs. To support track improve- ments the loan provided financing for procuring hard rails (for use on sharp curves), some wooden sleepers procured locally, rail fastenings designed for use with the concrete sleepers, ballast crushers and 50 new ballast cars. Also included in the program was the replacement of 530 old switches of which the loan would finance 230. 12. Equipment availability was to be improved with the purchase of subassemblies for diesel locomotives, and bogies, roller bearings and equip- ment to overhaul freight cars. Technical assistance to improve workshop practices and train mechanics was to be continued and management consultants were to study CNR's salary structure and suggest changes to reduce the high turnover of qualified workshop personnel. Also, completion of a new workshop at Facatativa, together with new preventive maintenance practices for rolling stock, were expected to help cut repair times and increase availability. 13. Operations were to be improved with the acquisition of new tele- communications equipment for train control and new freight handling equipment for working at terminals. The Program of Action in the project included a time-phased plan for the introduction of improved scheduling, composition and switching of trains, higher car loads, lower car turnaround times, and the reduction of derailments. 14. Improved train control and terminal working was to be achieved with the introduction of the new equipment and higher net loads would result largely from their greater carrying capacity. However, reducing and maintain- ing low car turnaround times requires continuous in-depth analysis of train operations and wagon movements, an effort that CNR had failed to undertake. Much of the car turnaround problem, in fact, related to delays at the ports, so that improvements could only result from studying the interface between the railway and port operations, improving railway facilities at ports, and improving the cooperation between CNR and port authorities. These issues were - 5 - not addressed under the project in relation to the port of Buenaventura and only addressed peripherally in relation to the port of Santa Marta in a study of intermodal choices along the Magdalena River corridor. 15. In the past, track rehabilitation had moved at a slow pace and had not been of a high standard. Track maintenance was now expected to improve as the track itself was rehabilitated to a better standard. Provision was made in the project to continue the assistance of consultants retained under the previous project, in the areas of track maintenance and renewal, and workshops. 16. CNR's finances were also expected to improve rapidly. The opera- ting ratio was expected to improve from 122 in 1973 to 99 by 1977. This was to be achieved through increased revenues from (i) an increasing volume of freight traffic, (ii) selective tariff increases, (iii) a concentrated mar- keting effort and (iv) cost savings from reductions in staff and the elimina- tion of uneconomic lines and services (SAR, paras. 2.09 and 3.13). Freight was projected to grow at an average annual rate of 4.4%, account having been taken of the fact that in the past projections had been optimistic and traffic had not materialized (SAR, para. 4.01). Also, because in 1971 a high standard paved highway paralleling the Cali-Buenaventura line had been completed and a paved road parallel to the Bogota-Santa Marta line was almost finished (both with Bank assistance), it was expected that the two lines, which were the financial backbone of the railway, would face increased competition from trucks (SAR, paras. 2.06, 2.07 and 2.11). The traffic forecasts counted, also, on attracting more long-haul traffic to the Atlantic line based on encouraging the use of a new truck/rail transshipment link to the ports of Cartagena and Barranquilla; this expectation was, however, predicated on CNR's management being able to cope with an entirely different type of trans- port operation to that with which it was familiar. 17. As opposed to the rather optimistic expectations of the Fifth Project, the Sixth did not anticipate much of its costs to be derived from CNR's own cash generation. Total cash needs for the project, debt service and working capital were estimated at Col$1,627 million and CNR was only expected to provide Col$62 million. Col$970 million would come from the Bank's loan, supplier credits, and a foreign commercial bank loan, Col$85 million from local bank loans, and Col$510 million from the'Government. Under arrangements incorporated in Section 2.03 of the'Guarantee Agreement, the'Government undertook to promptly provide CNR with Col$85 million and US$20 million to help pay all outstanding short term liabilities and foreign debt service obligations for the coming three years. A Project Fund was established in the Banco de la Republica to be used exclusively by CNR, into which the'Government would pay these amounts and subsequently make monthly payments based on CNR's cash flow projections for each succeeding three month period, including funds to meet debt service obligations and any estimated shortfall in operating revenues. These arrangements were considered satisfactory so as to avoid repetition of the serious cash shortage experienced during implementation of earlier projects. They were contingent, however, on the Project Fund being supplied with cash. 18. Training was an important element of the project but it was concen- trated on improving rolling stock maintenance, and on operating and commercial practices. Also, improving top and intermediate level management was con- sidered crucial. To this end, consultants who had been assisting CNR for several years were to be continued and the loan provided US$1.5 million for these services and a study of CNR's personnel planning and management functions. The consultant was not only to make recommendations but also to assist CNR during a six month period in implementing the recommendations. The management consultant was also to study the salary structure and suggest changes as necessary to reduce the high turnover of qualified personnel. III. PROJECT IMPLEMENTATION AND RESULTS 19. Physical implementation of the project was to start early in 1973, be completed by the end of 1974 and the loan account was to be closed in June 1976. However, because of delays in fulfilling legal conditions, the loan did not become effective until the end of 1973. The last loan-financed physical component of the project was not completed until August 1980 and the loan account was closed in November 1980. 20. Project implementation was correspondingly protracted and important elements of the Plan were not carried out. The track rehabilitation program was cut short and delayed. Of the 804 km of track expected to be rehabili- tated 235 km were done between 1974 and 1975, and 182 km between 1976 and 1980 for a total of just 417 km in the whole period (PCR, para. 3.04). The main reason for the delay was the shortage of sleepers. In fact, CNR management cancelled purchase of the high quality, treated timber sleepers foreseen in the project and purchased those of lower quality and cost. Still, quantities were insufficient, mainly because of the low price paid by CNR.!/ The programmed relaying of 576 km of line with concrete sleepers, a key element of the track rehabilitation program, was not implemented at all; and because the concrete sleepers were not procured, the elastic fittings, which were pur- chased under the Bank loan to be used with them, remained unused. The 230 switches needed to replace worn out ones in yards and stations, a major source of derailments, were not purchased either, and loan funds were reallocated to other uses. Procurement of 197 km of hard rail necessary to replace curved sections on which wear had been severe - another source of derailments - was considerably delayed due to procurement disagreements between CNR and the Bank, and a world wide shortage of steel. Similarly, procurement of welding equip- ment, track maintenance equipment, ballast crushing machines, and 50 ballast cars was also delayed for a long time. Finally, on the mechanical side, sorely needed spares for locomotives and freight cars did not become available until 1980. Thus, the physical part of the project stretched into 1980, well beyond the whole investment plan, with some key items not carried out. 1/ Annual purchases of wooden sleepers declined from 858,000 in 1974 to only 175,000 in 1979. - 7 - 21. In contrast, items that would have improved operations, such as freight handling equipment and the trucks and trailers, were not excessively delayed. 22. Because of the undue stretching out of the project, the impact of inflation was significant and it is now difficult to compare actual costs with appraisal estimates. The PCR (Subsection B, Project Cost, paras. 3.05 and 3.06), compares investments actually made over the 1974-1980 period with the appraised cost for the full 1973-1976 plan, and a brief footnote gives a com- parison between the project as appraised and actual investments made under it. To give a better picture of what was planned and what was achieved, the audit has combined in Table 1 financial and physical data where available. The many blanks in the Table reflect a lack of data that makes it now difficult to evaluate clearly the project's physical achievements as against those planned. 23. Operationally, CNR performed well during the two years 1973 and 1974. Freight and passenger traffic rose 10% and 9% respectively compared to 1972. This surge in traffic was largely the result of the better performance of the economy in those years and of large imports of grain resulting from a food shortage. However, traffic declined abruptly thereafter to a level in 1980 no better than that of 1970. A major cause for the drop in traffic was the declining availability of locomotives which fell from 87% in 1974 to 28% in 1981/82. To a lesser extent, a decline in the availability of cars and the failure of the truck/ rail operation serving Cartagena and Barranquilla (with which CNR experienced serious management problems) contributed to the lower traffic. 24. The problem with the locomotives was largely related to poor design. CNR, in fact, acquired a locomotive with insufficient power to operate over steep gradients at high altitudes and high temperatures with a full load. Even after some improvements were made by the manufacturer, further problems persisted mainly because of CNR's poor operating practices as, for example, the mismatching of locomotives in tandem units. As a result, the more power- ful locomotives became overloaded and rapidly deteriorated. In 1978, acknowl- edging that the locomotives were less powerful than originally expected, CNR decided to operate them in groups of three matched units. By then, however, it was difficult to produce many such groups. 25. Another factor contributing to the decline in rolling stock avail- ability was the continued incidence of derailments and accidents. The damaged stock overloaded the workshops, slowed down repair times and interfered with the execution of preventive maintenance. The actual causes of derailments and accidents are still not precisely known or remedied. The major cause is assumed to be, after years of rehabilitation, the bad track. Undoubtedly, track conditions continued to be bad in places because of the continued use of poor quality, inadequately treated sleepers, failure to install the concrete sleepers as planned, undue wear on the rails, and the continued poor bal- lasting. The number of derailments which had been over 7,000 in 1970 declined to 3,100 in 1979; while this is a considerable improvement, 3,000 derailments per year is still far too high. - 8 - 26. The Program of Action provided for a very modest reduction in staff, from 11,400 in 1973 to 11,200 in 1976. By 1976 total staff had risen to 12,100 mainly because retirees were retained on the payroll as CNR had no funds to pay their retirement benefits; however, in 1979 the'Government financed paying off most of the retirees and total employment fell to about 10,350. The numbers declined at the wrong end, however; skilled technicians in the workshops went down while employment in management and administration increased. 27. The consultants continued to train many technical employees but much of the effort was lost because the trained staff quickly left for better paying jobs elsewhere. The consultant's advice on improving operations and maintenance was overtaken by the many day-to-day problems experienced by CNR, such as landslides, floodings, derailments and locomotive failures. The management consultant's recommendations were on the whole not accepted. The appointment of a deputy general manager as a career railway officer, for instance, was particularly contested and not resolved for a considerable time. Subsequently, appointees have also become subject to the political changes introduced in management from time to time. 28. The decline in operating quality and traffic was reflected in CNR's deteriorating financial situation. The working and operating ratios, which had improved to 91 and 99, respectively, in 1974, rose to 139 and 147 in 1976 (against appraisal targets of 91 and 104), and were 122 and 126 in 1979. The causes were of a familiar pattern. CNR faced increasing competi- tion from truckers and the low level of traffic prevented CNR from increasing tariffs sufficiently to cover cost increases (while consumer prices rose by 587% from 1969 to 1979, passenger fares and freight rates rose only by 390% and 459%, respectively). In turn, despite the project, costs were not reduced or productivity increased enough for the railway to effectively compete with trucking. Thus, as overall traffic volumes in the country increased the railway was not able to take advantage of them and the pattern of deteriora- tion and unreliable service continued. 29. CNR's costs rose considerably, particularly staff (70% of total working expenses), due to additions in the administrative area and to higher benefits demanded by the unions without a corresponding increase in produc- tivity. As salaries and also fuel prices increased sharply, CNR's working expenses became almost double the appraisal estimate by 1977 (Col$1,823 million versus Col$938 million) and continued to rise. The resulting working losses rose from 9ol$65 million in 1973, before the project started, to Col$1,456 million1Y in 1980 when the Investment Plan was concluded. Since these were all cash losses, CNR could provide nothing towards debt service or investment financing and constant 'Government financial support was necessary. While the'Government did provide an increasing amount of funds to CNR, it was unable to assume the whole burden of financing operating losses, social 1/ Without including social security expenses paid on behalf of CNR by the Government. - 9 - security payments, debt service and investment. Thus, CNR resorted to local borrowing by way of bank overdrafts, and to delaying payments to suppliers. The result was a continuing erosion of its working capital, and the current ratio, which had been 2.0 or more until 1975 declined rapidly to only 1.0 by 1979. Current liabilities (which includes payables and short-term financing) rose from Col$266 million at the end of 1974 to Col$1,657 million by 1979. 30. The mechanism set up in the'Guarantee Agreement to channel financial support for the railway failed to operate properly. While the initial deposit to the Project Fund was made, subsequent monthly payments became irregular and soon ceased. However, in total, the 'Government did, in fact, provide con- siderable funds to CNR: 1973 1974 1975 1976 1977 --------Col$ million (current)------ Funds for Investment 185.3 174.4 90.0 99.2 120.0 Debt Service payments (including IBRD, local and foreign Loans) 75.3 83.9 149.7 352.6 464.8 Social Security payments 125.9 131.0 160.8 293.4 308.2 Contribution to offset Operating Deficits - - - 82.5 350.0 Total 386.5 389.3 400.5 827.7 1,243.0 In 1979, the national budget included Col$2,855 million for CNR of which Col$478 million was for rehabilitation of track and equipment, Col$1,011 million for debt service, Col$696 million for pensions, Col$550 million for social security payments, and Col$120 million to cover part of CNR's expected cash deficit for the year. The extent of the'Government's financial support for CNR can be gauged by examining the summary balance sheets in Table 6 of the PCR. Total cumulative government investment in CNR rose from Col$2,684 million in 1974 to Col$7,534 million in 1979. 31. The Appraisal Report estimated an ERR for various project components separately: the concrete sleepers, hard rails, ballast cars, locomotives, the trucks and trailers needed for the road/rail operation; also an ERR for investments in the Bogota-Santa Marta line separate from that for the project as a whole. By contrast, the PCR presents only an economic reevaluation limited to the project as a whole. It concludes that the project yielded an economic return of 11%, about the same as expected at appraisal (12%) but lower than the'Government's cut-off rate for projects in the transport sector (PCR, para. 7.10). In the light of the generally poor performance of the railway, its high cost, particularly due but not restricted to the high - 10 - incidence of derailments and accidents, its poor service and the much lower volume of traffic compared to appraisal forecasts, the audit questions the validity of the estimate. A number of factors combine to produce this result, among them, as explained in the PCR, (i) the lower than expected investments, (ii) the higher than expected benefits in the first two years of the project (when many of the investments and actions associated with the project were not yet in place), and (iii) the somewhat faster increase in the period to 1979 (17% vs. 12%, Table 2) of highway transport costs relative to rail costs. The avoided traffic diversion to road transport, the main benefit of the project, was, however, about 50% below expectations and, contrary to appraisal expectations, the project did not prevent rail costs from increasing (rail costs were expected to first decrease and then remain constant in real terms). The PCR's estimate is in large measure the conse- quence of assuming that benefits measured for 1979 would continue to occur for the next 20 years. However, more recent information indicates that since 1979 CNR's traffic has continued to decline, indeed that it has fallen to levels below the appraisal's "without" project estimate (Table 2), invalidating the PCR assumptions and sharply reducing project benefits. In fact, because traffic levels are now so low, system wide average costs per unit of traffic moved have become higher than highway costs, signifying that not only is the railway as a whole losing money but the country could be wasting real resources. If the ERR calculation is repeated with the PCR's methodology, and the traffic trend of the last 10 years is assumed to continue, the project is likely to show a negative rate of return. It is, however, the judgement of the Regional staff that past traffic trends have resulted from track and motive power capacity constraints. As these constraints are removed under the ongoing Seventh Railway Project, traffic levels are expected to improve and the past trends to be reversed. 32. Evidently, appraisal expectations regarding the evolution of CNR's traffic did not materialize and expectations regarding the project's effect on costs did not come about. While at appraisal system wide railway costs were Col$2.38 per ton-km (measured in 1979 Col$), and had slightly decreased to Col$2.13 by 1974 when project implementation began (Table 2), by 1980, when the project was completed, they had risen to Col$3.79 per ton-km (Table 2) and have since continued to rise. 33. However, figures in Table 2, as those in the PCR, are system wide averages; they may hide the fact that for some commodities on some lines railway costs and the relationship between rail and trucking costs may be more favorable to the railway. For instance, while in 1971 the system wide average cost was Col$0.47 per ton-km, for the Bogota-Santa Marta line it was just Col$0.28. If the difference between system wide costs and those of the Atlantic line were to still exist today, costs on that line would be close to, if not lower than trucking costs, indicating that at least that line is viable. However, information is not available to permit a judgement on whether this is the case. Moreover, while the Appraisal Report was specific on investments that were to be made on the Bogota-Santa Marta line, and the economic assessment of the project as a whole was predicated on the expected benefits from that line, no information is offered in the PCR on the distri- bution of project expenditures by line and their specific effect on each of them. In fact, it appears that project investments were distributed through- out the network and scattered on both high and low return works rather than - 11 - being concentrated on activities with high expected yields. This departure from the more focused investment objectives specified at appraisal, coupled with CNR's failure to generate system wide improvements and shut down uneco- nomic lines and services, helps explain the project's overall poor impact. Of course it could be argued, as does the PCR (para. 9.02), that without the project the railway's situation would have been even worse, that services would have had to be suspended and that it is to the credit of the project that this did not happen. However, this was not the "without project" condi- tion against which the project was appraised; it has therefore, not been evaluated against that alternative. It is evident to the audit that, despite the project, the railway has deteriorated to a critical level and that a continuation of this trend would most likely lead to closing down of parts of the system. However, a study of the railway's future, and its continued viability in whole or in part, requires a different type of analysis than that presented in the PCR, one that treats all existing assets as sunk investments without alternative use. 34. Because of the poor accounting and insufficient information (see Table 1), it is now impossible to evaluate which portions of the system and what services have benefitted from the project. The system wide averages tend to underrate the advantages of the viable portion of the network and overrate the disadvantages of the non-viable, uneconomic portions of it. As at apprai- sal there should have been on conclusion of the project a more concentrated effort to identify by line and service the effects of the project. IV. CONCLUSIONS 35. Although at appraisal the Bank admitted that after five projects and twenty years of effort the railway still provided poor quality service, failed to achieve financial targets and was less significant in the Colombian trans- port sector than before, it forecast a continued economic role for it, partic- ularly for the Bogota-Santa Marta line. In support of this expectation an investment plan was endorsed that focussed on track rehabilitation and on measures, including technical assistance, to improve services and, thus, defeat the cycle of deteriorating infrastructure, derailments, poor service and loss of traffic. CNR's performance in the years immediately prior to project implementation improved, but as the project was executed performance deteriorated again and the railway fell even more into a vicious circle of declining traffic and revenues, rising financial deficits, insufficient financial support from the Government and poor service. However, part of the project implementation period coincided with an economic slow-down in Colombia and with a period of balance of payment difficulties which prevented the Government from supporting the railways adequately. Public sector invest- ments in the railway were, in fact, reduced to 0.6% of the total transport budget in 1979 compared with 9.1% in 1972 (PCR, para. 5.02). Natural disas- ters compounded CNR's difficulties (the flooding of tracks near rivers, - 12 - collapse of a major bridge on the Atlantic line and rock slides which severed the line between Medellin and Cali)..Y However, frequent changes in CNR management were disruptive of efforts at improvement, and other factors related to CNR's management, or well within its control, greatly contributed to the lack of success. 36. Thus, the 'Government had to provide increasing financial support to CNR; however, funds were never sufficient or timely enough to meet the railway's multiplying needs. While the 'Government is hardly to blame for attempting to contain the financial drain imposed by the railway, it is a fact that financial difficulties were a contributory cause of slow project imple- mentation and of decisions contrary to the intent of the project such as the continued purchase of low quality wooden sleepers, inadequate ballast, the cancellation of the concrete sleepers and new switches, and delays in the purchase of spare parts for locomotives, rolling stock, and new rails. Also lacking was adequate cooperation between *Government agencies. It was well known that the railway depended largely on providing for the traffic at the ports; the Government should have ensured, therefore, that both at Santa Marta and Buenaventura not only adequate physical facilities were provided for rail operations (yards, warehouses, cargo handling facilities) but that port and railway managements cooperated fully. Such was not the case. 37. The Bank rightly recognized the need to better control project implementation as a means of increasing the probability of achieving the proposed targets. It increased project supervision beyond regular efforts. However, it might have been useful to bring more pressure to bear on CNR and the Government to ensure that the advice provided by successive supervision missions in detailed "aide memoires" was heeded, and that the help of the resident staff in Bogota was sought to achieve this. Instead, the'Government, CNR and the Bank got themselves involved in controversies over procurement matters which detracted from focussing on the issues and major problems at hand. 38. The PCR concludes and the audit agrees that for the project to have succeeded it would have been indispensable to have competent management and the full support of the 'Government. This applies to any project. The PCR also concludes that had the project not been carried out CNR operations and equipment would have deteriorated more rapidly and rail service would have had to be suspended, a situation avoided by the project. This may be true but was certainly not the justification for the project as presented to the Board. 1/ Service between Medellin and Cali has not been reestablished and the railway is now effectively two systems, one on the Cauca river valley with a branch to Buenaventura, and another along the Magdalena river valley with connections to Medellin and Bogota. - 13 - 39. The project has been succeeded by another, the seventh with the Colombian National Railways. The loan to support that project (Loan 2090, US$77.0 million) was approved by the Board in February 1982, but has not yet become effective for much the same reasons the Bank delayed considering the Fifth Project in 1968 (PPAR No. 776, June 1975), and this project in 1972 (para. 7), namely, the 'Government's delayed and insufficient financial support. The PCR makes a case for the continued existence of the railway (para. 9.03) provided certain actions are taken by CNR and by 'Government. Such actions cannot be long delayed. PROJECT PERFORMA CE AUDIT REPORT COLOMBIA SIXTH RAILWAY TROJECT (LOAN 926-Cgl Appraisal Estimates and Actual Costs (In Constant 1973 Col$m and us$m equivalent) APPRAISAL ESTIMATES ACTUAL EXECUTION Quantities Costs Quantities Expenditures 1973-76 1973-74 Loan 1973-76 1973-74 Loan Loan Loan Plan Project Financed Plan Prnject Fianced Plans Pro3ect/b Fnanced Plan Financed Cu Um Col$m E$n EL C - is$- us$- Col$ US$m Col$m US$. US$n 1. Permanent Way Department Track Rehabilitation (km) 434 364 7 193.8 8 6 143 5 6.4 3 1 417 235 7 360 7 16.0 174.0 7 7 1 S.itches (No ) 754 530 230 35 0 1.6 24.2 1 1 0.5 Cancelled - - - - - - Concrete Sleepers (km) ) 576 152 oil ) 289.9 12 9 76 3.4 nil nil nil ) 29 2 1.3 25 4 1.1 1.2 Fastenings (No.) ) 959,00 253,0 253,000 180,000 180,000 [80,000 Hard Rails (tons) 13,020 13,020 13,020 81 4 3.6 81.4 3 6 3 6 13,020 13,020 13,020 30 3 1.3 - - 5.3 infrastructure and Bridges I I 7 110 4 4 9 46.7 2.1 1 3 7 ? 77.9 3 5 52.3 2 3 2.5 Ballasting and Resleepering I7 49.8 2.2 28 4 1 3 - 7 - 66 7 4 0 19 6 0.9 - Aluminothermic Welding I ? ? 46.5 2 1 23.5 1 0 0 9 7 7 29 7 1 3 22 1 1 0 1.8 Maintenance Equipment and Ballast Crushers a ? ? 66.0 2 9 43.3 1 9 1 9 ? 7 45 4 2.0 27.8 1 2 2.9 Ballast Cars (No ) 100 50 50 30.2 1 3 15.1 0.7 0.7 1 50 38 1 1.7 6.4 0 3 2.1 Sub-total 903 0 40.1 482 7 21.5 13.5 678 0 31 1 327.6 14 5 15.8 2. Mechanical Department Diesel Locomotives (No.) 28 28 - 160.5 7 1 160 5 7 1 - z8 28 - 133.8 5 9 133 8 159 - Locomotive Spares - - - 86 5 3.8 43.7 1 9 1 9 - - - 178.3 7.9 87.9 3.9) Freight Car Spares - - - 98 7 4 4 43.5 1.9 1.3 - - - 109.9 4.9 12.6 0.6) 4.3 Machinery and Equipment for Workshops nil - 62.6 2.8 - - - 17.4 0 8 7 7 0.3 - Sub-total 408 3 18 1 247.7 11.0 3 2 439 4 19 5 242.0 10.7 4.3 3 Operations Department Stations and Yards 7 - 56.5 2.5 16.1 0 7 - 49 3 2.2 26 6 1 2 - Warehouses - 10.1 0.4 0 7 - - - - - - - - Freight Handling Equipment ? ? 19 9 0 9 3.5 0 2 0.2 ? 1 10 6 0.5 10 6 0 5 0.3 Road Trucks (No.) ) 28 28 28 1 17 17 17 ) Road Trailers (No.) ) 64 64 64 426.1 1 2 26 1 1.2 1 2 17 734) 0.5 94 0.4 09 Passenger Coaches I - - - - I - - 4 7 0 2 2.6 0.1 - Rail Cars ? - - - - - - 0.5 - - - - Containers (No.) 136 o11 - 4.6 0.2 - - - - - - Sub-total 117.2 5 2 46.4 2 1 1 4 76.2 3.4 49.2 2.2 1.2 4 Miscellaneous Dwellings - 15 1 0.7 7 4 0.3 - ? 7 - 9.7 0 4 1 8 0.1 - Telecommunications 182.6 8 1 31 8 1 4 1 7 7 33.4 1 5 18 1 0.8 1.0 Consulting Services 7 81 3 3 6 48 5 2 2 1 5 1 7 47.4 2 1 39 6 1 8 1.3 Winter Damages Repair - 43.2 1 9 43 2 1.9 1 0 - 80 3 3 6 74 2 3 3 - Sub-total 322 2 14 3 130 9 5 8 3 5 170 8 7.6 133.7 6 0 2.3 TOTAL 1,750.7 77 8 907.7 LO 4 21 6 1,364 4 61 6 752 5 33 4 23 6 Centingencie- 363 3 16 2 87 9 3 9 1.8 - - - - GRAND TOTAL 2,114.0 94.0 995 6 '4 3 23.4 1,364 4 61 6 752 5 33 4 23 6 Interest During Construction 35.9 1 6 35.9 1.6 1 6 1 ? 19 1 0 8 1.4 GRAND TOTAL INCLUDING INTEREST DURING CONSTRUCTION 2,149.9 95.6 4,0315 L5.9 75 0 771 6 34 2 25 0 /a Actual Plan period was 1974-1980 lb Actual Project period was 1974-75 c Fastenings for concrete sleepers February, 1983 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA SIXTH RAILWAY PROJECT LOAN 926-CO) Expected vs. Actual CNR Performance 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 Forecast Traffic with Project (billion ton-km) 1.24 1.30 1.37 1.44 1.48 1.48 1.48 1.48 1.48 1.48 Forecast Traffic w/out Project (billion ton-km) 1.17 1.06 .94 .82 .70 .70 .70 .70 .70 .70 Forecast Avoided Diversion (billion ton-km) .07 .24 .43 .62 .78 .78 .78 .78 .78 .78 Forecast Rail Costs with project/a (79 Col$ million) 2536 2453 2470 2491 2527 2527 2527 2527 2527 2527 Forecast Rail Costs w/out project (79 Col$ million) 2413 2201 2000 1777 1571 1370 1168 1168 1168 1168 Forecast Rail Unit Cost with (79 Col$/ton-km) 2.05 1.89 1.80 1.73 1.71 1.71 1.71 1.71 1.71 1.71 Forecast Rail Unit Cost w/out (79 Col$/ton-km) 2.06 2.08 2.13 2.17 2.24 1.96 1.67 1.67 1.67 1.67 Forecast Road Costs/b (79 Col$/ton-km) 2.72 2.72 2.72 2.72 2.72 2.72 2.72 2.72 2.72 2.72 Actual Traffic /c (billion torr-km) 1.33 1.33 1.14 1.16 1.22 1.23 1.11 0.86 0.62 0.55 1 Actual Avoided Diversion (billion ton-km) .16 .27 .20 .34 .52 .53 .41 .16 .0 .0 H Actual Rail Costs/d (79 Col$ million) 3162 2829 2699 3400 2975 3129 2964 3257 3141 5189 1 Actual Rail Unit Costs (79 Col$/ton-km) 2.38 2.13 2.37 2.93 2.44 2.54 2.67 3.79 5.07 9.43 Actual Road Costs/e (79 Col$/ton-km) 3.07 3.15 3.24 3.32 3.41 3.51 3.60 3.68 3.77 3.85 /a Freight traffic costs exclusive of overhead. /b Costs including overhead and road infrastructure costs. /c The difference with PCR Table 9 is caused by the use of "net paying traffic", which is comparable to appraisal figures, as opposed to "total net traffic", which includes the railway's own freight. /d Working expenses plus depreciation and social security payments less a proxy for costs attributable to passenger services equal to passenger revenues plus 33%. /e Operating costs for an average 10-ton truck with a 55% load factor operating on an average Colombian road, including taxes and duties as a handsome proxy for the cost of providing and operating the infrastructure. April 1983  - 17 - PROJECT COMPLETION REPORT I. INTRODUCTION 1.01 The Andes Mountains in Colombia present formidable barriers to communication among the main population centers in the country, which, constitute separate and almost self-sufficient regions. It was not until the early 1950s that, under the drive toward integration and modern- ization, the transport system began to evolve into a national network. 1.02 The Magdalena River had been the only transport route between the central region and the Atlantic coast until, in the early 1960s, construction and upgrading of the Western Road (Cartagena-Medellin-Cali-Pasto), the Eastern Road (Santa Marta-Bucaramanga-Bogota-Neiva), and the two main transverse roads that connect them in the central region were substantially completed. Since 1961, the Atlantic Railroad has provided a connection between the port of Santa Marta and the two major urban centers of Bogota and Medellin. During the last 25 years, the country's port capacity and air transport services have also expanded considerably. 1.03 During the period 1950-1978, the transport sector's contribution to GDP grew at an annual rate of 6.1%, which is larger than the annual growth rate of 5.2% of total GDP for the same period. At the same time, the share of the transport sector in GDP rose from 5.1% in 1950 to 6.6% in 1978. 1.04 The Colombian inter-urban transport system moved about 19.5 billion ton-km in 1978, of which about 75% was moved by truck over the national highway network. Coastal shipping accounted for about 11%, river transport about 7%, railways some 6% and aviation about 1%. Although agricultural output dominates, manufactured goods are becoming an increasing share of Colombian freight. Roads dominate passenger traffic, accounting for almost 71% of the pass-km, while aviation's share is close to 27%; railways serve only about 2% of the passenger traffic. 1.05 Colombia has a road network totaling about 67,000 km, of which almost 23,000 km constitute the national highway system, 31,000 km are departmental roads and 11,200 km are feeder roads. The balance of about 2,000 km of roads are private, municipal or in the national territories. About one-third (7,900 km) of the national system is paved, while only about 3% of the departmental roads are paved. All other roads are gravel or unsurfaced. The road network is not well maintained. At least half of the paved roads need asphalt overlays or strengthening in the next ten years to preserve the pavement structure. The rail network consists of 3,403 km (Map IBRD 3667R1), of which 2,912 km are in operation. The remaining 491 km, not in operation, are mostly part of the Pacific network (Antioquia and Pacific Divisions), including the connecting link to the Atlantic network on which the Itagui (near Medellin) - La Felisa line which was washed out in 1975 and on which traffic has been suspended ever since. 1.06 This project completion report (PCR), covering the sixth railway project, has been prepared with the assistance of the Borrower, who provided detailed reports for the Bank's review. It covers the component of the project financed by Loan 926-CO for US$25.0 million, the sixth loan to the Colombian National Railways (CNR). The objective of the project was the physical, economic and financial rehabilitation of CNR. - 18 - II. PROJECT PREPARATION AND APPRAISAL A. Background 2.01 The Bank's association with the railways in Colombia began in 1951, and the first three railway loans (68-CO, 119-CO and 267-CO) helped to finance the construction and equiping of the Atlantic line, which has become the backbone of the railway system. In 1963, the network was consolidated through the creation of CNR and was integrated by standardizing the track gauge to 914 mm. Two subse- quent loans in 1963 and 1968 (343-CO and 551-CO) contributed to the rehabilita- tion of track and rolling stock and to the improvement of management practices. In addition, these loans helped to finance the services of consultants to provide technical assistance to the railways. CNR received, in total, US$88.7 million under these five loans from the Bank. Performance under the last two loans, however, was below expectation: implementation of the projects were frequently slow and financial results were disappointing. Derailments increased significantly up to 1970 as a result of delayed rehabilitation, inadequate maintenance, unexpected track washouts due to severe rainy seasons and poor operation. B. Project Appraisal 2.02 CNR first approached the Bank on the matter of the Sixth Railway Project in 1970. The investment plan was prepared by CNR with the assistance of their consultants. The investment plan submitted by CNR requesting financial support was analyzed, and the Bank concluded that CNR had an important role to play in the economy of Colombia, particularly with regard to the Atlantic line and its connection to Medellin. A minimum investment plan was found justified; it was designed to break the,cycle of deteriorated infrastructure, derailments, poor service and loss of traffic. The project also included numerous measures, in a Program of Action,to improve the services and to provide technical assistance to strengthen the management, operations and finances of CNR. C. Project Description and Goals 2.03 The total investment plan for 1973-1976 was estimated to cost US$94.0 million equivalent with a foreign exchange component of US$60.4 million equivalent. Of this total, some 52% was for track and civil works; 23% for motive power, rolling stock and workshops; 10% for telecommunications; 5% for consulting services and the rest for freight-handling equipment, dwellings, and winter damages. 2.04 The project consisted of the first two years, 1973-1974, of the above-mentioned investment plan and included also a Program of Action to improve the operational efficiency of the Railway. The project was esti- mated to cost about US$44.2 million equivalent, with a foreign exchange component of US$32.2 million. The Bank loan for the amount of US$25.0 million covered mainly essential track materials, i.e., rails, fastenings, switches, infrastructure and bridges, rail-welding equipment, track maintenance equipment including ballast crushers and ballast cars. These items combined represented 62% of the loan and reflected the emphasis given to the much needed improvement - 19 - of CNR's permanent way. Other items included in the loan were spare parts for rehabilitation of locomotives and freight cars (15%) as well as freight handling equipment, trucks and trailers, telecommunications equipment, consult- ing services and repair of winter damages. 2.05 Using the above-financed equipment and works, CNR was expected to make substantial improvements in operations. The Program of Action specified operational targets for the 1973-1976 period, including availability of diesel locomotives and freight cars, average net load per car/train, car turnaround time, staff reductions, improvements in management and operational procedures and the overall financial position and profitability of the railways (Table 1). Consultants were to be employed in the fields of manage- ment, track maintenance and rehabilitation, workshops and operations, with the latter category including operational plans for the improvement of services on specific lines and the reduction of services and closure of uneconomic lines. A number of improvements had already been made in some of these areas by the time the project was presented to the Board; for instance, the program for closure of uneconomic lines and reductions in services and closure of stations had already started, improved train operating plans had been introduced and tariffs had been completely revised. 2.06 Because of difficulties of the Colombian Government in defining its position regarding the railway's future role in the economy of the country and its attitude toward the solution of CNR's financial crisis, the project was appraised in 1971 and re-appraised in 1972. III. PROJECT IMPLEMENTATION AND COST A. Project Implementation 3.01 Implementation of the project was to start early in 1973, and the closing date was estimated to be in June 1976; however, because of delays in loan effectiveness, the implementation of the project started at the end of 1973, with the exception of the 28 locomotives procured with supplier credit, which arrived in 1973; the last physical components of the loan were completed in August 1980, and the loan was closed on November 7, 1980. 3.02 CNR's performance was good in the early period of this loan (1973-1974). Locomotive and wagon availability was either on, or better than, the target. The financial performance was substantially better than the plan of action targets. The number of derailments declined, and procurement was progressing well. Freight traffic in 1973 and 1974 was about 11% higher than in 1972. The management consultants, hired under the technical assistance program of the project, completed their studies on time, and CNR started implementing their recommendations. In short, CNR was well on the way to becoming a strong and viable entity. The improving trend, however, was effected by unforeseen circumstances in 1975. The network was damaged by line washouts, the economy slowed, and balance-of-payment difficulties adversely affected imports, re- ducing the railway's long-haul traffic of bulk commodities such as wheat and - 20 - fertilizer. While total tonnage dropped and revenues shrank, costs increased substantially as a result of higher social benefits demanded by the Unions (para 5.04); actual salaries were Col$ 1,070.0 million in 1976 as compared with Col$ 435.9 million in 1973, an increase of 145% (Table 5). The unfavorable balance of payments situation, inflation and a fund shortage forced the Govern- ment to curtail payments to the railway. Thus, maintenance of the plant, neces- sary for efficient and economic operations, was not carried out. In addition, frequent changes of CNR's top management affected the quality of performance and contributed to the flight of good middle-managers. During 1976-1979, the Commercial Manager was changed six times; the Technical Manager, four times; the Financial Manager, six times; and the Administrative Manager, four times. The General Manager was changed four times in the last six years. The Deputy General Manager, who should be a career railway expert, was changed twice. This high turnover of management and supervisors severely constrained the institution's performance. Poor coordination between the railway and the ports of Santa Marta and Buenaventura caused long wagon waiting time (11-12 days) at the ports, reducing the railway's traffic carrying capacity. At a result, users switched to other more reliable modes of transportation, causing a decline in railway traffic. 3.03 Another shortcoming was the fact that the railway was unable to collect demurrage charges from shippers who claimed that they were not responsible for delays in returning wagons to CNR. Furthermore, locomotive availability declined to 30% in 1980 because locomotives, with less horse- power than required were purchased, locomotives were used inappropriately at high altitudes (para 4.05) and repairs and maintenance were inadequate due to shortages of funds. 3.04 Actual and expected completion dates for the main project components are shown in Table 2, and the reasons for the delays in each case, are summarized below: (a) Some 235 km of track were rehabilitated between 1974 and 1975 in comparison with appraisal estimates of 364 km. The main reasons for this poor performance were the shortage of ties and the lack of funds (from the Government budget) for track materials and contractors' payments. (b) Early in 1975, and with the agreement of the Bank, CNR drew up a more realistic program covering 440 km of track to be rehabilitated in the period 1976-1978, of which only 182.4 km of track were rehabilitated. In total CNR rehabilitated 417.4 km between 1974 and 1980 as compared to 434 km originally forecast in the investment plan for the period 1973-1976. (c) The project included the procurement of 253,000 concrete sleepers. To this effect, 180,000 sets of rail elastic fittings at a cost of about US$1.2 million were purchased with funds from the Bank loan, and were available to be used by CNR in 1975. In spite of the fact that the concrete sleeper program was strongly recommended by the consultants and the Bank, it was never fully implemented by CNR because of CNR's management decision to continue the procurement of low quality cheaper - 21 - wooden sleepers and largely due to the lack of funds. On a trial basis, concrete sleepers were laid in 3 km of track, which at the end have proved to be a success. The elastic fittings are now to be used in CNR's investment program for 1982-1986. (d) During the first reallocation of funds in 1975, in order to give priority to other more urgently needed items for the track, and with the approval of the Bank, the program for the renewal of 530 switches was cancelled. (e) Because of the scarcity of funds for contracts, the start of the program of aluminothermic rail welding was delayed for a year and could only be completed at the end of 1979. (f) Several other items, such as rail fittings, rail spikes, permanent way equipment, ballast compactors, ballast crushers, compressors, hydraulic shovels, bulldozers, front-end-loaders, and ballast cars, locomotive spare parts, spares for freight cars, tele- communication parts and components were purchased during the course of the project and implemented late, as indicated in Table 2. (g) In 1976, CNR formed an intermodal transport company, together with a government-owned company called "Empresa Publica de Barranquilla" (EPB), to maintain and operate the fleet of 17 trucks and 34 trailers purchased through this loan at a cost of about US$1.0 million. At appraisal, the procurement of 28 trucks and 64 trailers was planned, but due to price increases, the number of units had to be reduced. The new company was called "Transmodal" and operated between Cartagena, Barranquilla and Cienaga (See Map IBRD 3667R1) with 10 trucks and 20 semi-trailers, and with the rest of the fleet operating between Ibague and Armenia. Financial results were good at the beginning, and Transmodal's services were in great demand. Later on, it was found that the terms and conditions of Transmodal operations were more in favor of EPB than of CNR, the owner of the fleet, and that the company was badly managed and the fleet not maintained properly. As a consequence, operations were stopped in July 1980 and the company was liquidated in March 1981. The trucks and trailers, apparently seriously damaged, had to be sent to Bogota, where they are being slowly repaired with the intention of putting them to work on the Facatativa-Puerto Salgar section as soon as funds are available. As a result, these invest- ments of about US$1.0 million were not well utilized. B. Project Cost 3.05 In current prices the actual total investment in the 1974-1980 period was Col$ 2680.8 million or US$75.7 million, as against the appraisal estimate of Col$ 1750.7 million or US$77.8 million (excluding contingencies) and - 22 - Col$ 2113.9 million or US$93.9 million (including contingencies). The apparent cost underrun (in US$) occurred because in most cases not all of the works included in the project at appraisal were carried out. The main explanations for the cost overrun (measured in current Col$) are higher prices because of what appears to be some underestimation of costs and local inflation at appraisal and the long implementation period. I/ Once bids were received, it was clear that there would be cost overruns for many items. As a result, adjustments in the quantities and/or cancelling of items were introduced, considering the least impact on the project. Procurement was reduced in the case of rails, rail fittings, equipment, trucks and trailers. Not all works and items included in the project at appraisal were carried out, such as the procurement of rail switches and concrete sleepers. 3.06 Because of price increases, and in spite of the fact that only 417.4 km of track was rehabilitated from 1974 to 19.80 as against 434 km of track programmed in the investment plan at appraisal, there was an overrun of 101%. The actual cost of rolling stock (50 ballast cars) were substantially higher than estimated at appraisal, US$1.7 million actual expenditure vis-a-vis US$1.34 million ex- pected for the procurement of 100 ballast cars. In the case of spare parts for locomotives and rolling stock there was an overrun of about US$4.0 million and US$0.5 million respectively because of the actual time-span which took the project to 1980, and because of the increasing need for spares caused by the high number of accidents and derailments in that period. Procurement 3.07 The railway equipment to be procured under the loan was divided into 35 groups, three of which were direct purchases (part of the spare parts for locomotives, rolling stock and railcars). The procurement process commenced towards the end of 1973. 3.08 There were inordinate delays in the procurement cycle, some being attributed to internal procedures and others to the frequent changes of CNR's managers, time was lost in needless inquiries and changes of procedures. The implementation schedule had several setbacks in non-Bank financed items because of the introduction, by the new railway management, of contractual procedures different from those previously in force in the railway and the suspension of all contractual work until these were complied with. These factors and the lack of adequate financial support from the Government inter- fered with the execution of the project; therefore, its completion was further postponed. I/ In constant prices, the actual investment in the 1974-1980 period was Col$ 1,364.4 million, 36% lower than the appraisal forecast including contingencies (Table 3-A). If the investment plan would have been carried out as originally envisaged during appraisal, at a cost of US$93.9 million, the actual cost in 1973 prices would have been approximately US$104.2 million or 11% overrun (because only aggregate data was available, it has not been possible to compare costs by category and some assumptions had to be made). Perhaps a more meaning- ful comparison is shown in Table 3-B; the cost of actual investments in constant 1973 prices during 1974-1975 was Col$ 752.5 million, 27% less than forecast in the appraisal report for the project implementation period 1973-1974. - 23 - 3.09 Procurement practices were particularly unsatisfactory in the case of ties. In an effort to avoid drastically reducing the number of ties procured, the the price was fixed by CNR at a low level, which resulted in an inferior quality. Annual purchases declined from 858,000 in 1974 to only about 175,000 soft wood ties in 1979. As a result, about 95% of the derailments due to the track were caused by bad ties. The Bank recommended that CNR should discontinue the practice of fixing the price and should purchase quality ties according to adequate specifications; unfortunately, CNR did not take into consideration these recommendations. 3.10 The major item in the Bank loan was the procurement of rails. 1/ The first tender for 13,300 tons of 18 meter rails was issued in November 1973 but no offers were received. The tender was reissued in 1974; again, no tenders were received. The tender was reissued for the third time in March 1975. This time five offers were received. Of these, two were from U.S. manufacturers for 12 meter rails. With the approval of the Bank, these were rejected, as were the other three, for not complying with the approved specifi- cations. 3.11 In June 1975, the Bank suggested in a cable that in the specifications, 18 meters be indicated as the first preference and 12 meters as an alternative but a Bank mission that visited Colombia in August 1975 agreed that only 18 meter rails were acceptable, accordingly, a tender notice was issued for the fourth time on June 9, 1975 for 18 meter hard rails. Two offers were received and both rejected. 3.12 In a letter dated November 13, 1975, the Bank urged CNR to issue a new tender for hard rails with the same specifications as used earlier, i.e. 18 meter rails. Meanwhile, a CNR technical committee recommended the inclusion of 18 and 12 meter rails, pointing out that, in comparing offers of the two types of rails, it would be necessary to take into account the cost of additional welding for 12 meter rails. A higher committee of CNR, however, rejected this recommendation and with the approval of the Bank (given by a Bank consultant, apparently overlooking the fact that the Bank has suggested the inclusion of 12 meter rails), the tender was issued for the fifth time on January 26, 1976 for 18 meter rails only. 3.13 On March 8, 1976, the Bank sent a cable advising CNR to include 12 meter rails in the tender that had been issued on January 26 and to postpone tender opening if necessary. CNR, however, wrote to the Bank on March 17 stating that it could not agree to a postponement of the closing date at this stage to incorporate the change proposed by the Bank because the Bank had repeatedly approved the specifications as issued. On March 25, the Bank, confirming a telephone conversation, cabled CNR regretting their decision not to comply with the proposed amendment. 1/ Case study No. 2 of O.E.D.'s Report No. 3557 "An Interim Report on Procurement Issues in Bank-Financed Projects". - 24 - 3.14 On March 26, the date on which the bids were due to be opened, CNR received another cable from the Bank reiterating the suggestion for amending the tender, post-poning the closing date and adding that failure by CNR to include 12 meter rails would prevent the Bank from financing the purchase. 3.15 When the tender was opened on March 26, six bids were received; five for 18 meter rails and one for 12 meter rails from a U.S. manufacturer at US$475 per ton, against the lowest offer made by a British firm for 18 meter rails at US$400 per ton. 3.16 On March 31, CNR sent a letter to the Bank requesting reconsideration of the position. In May 1976, the Bank informed CNR again that the Bank could not change its position and that a new tender should be issued. 3.17 The Government than issued a communique on June 2, 1976 denouncing "improper pressure exerted personally and by means of telephone communication by the staff of IBRD". The communique stated that the Government had suspended the utilization of a Bank loan and had sent a message to the President of the Bank to bring him up to date on the delicate situation created by "improper pressure". The Manager of the National Railway explained to the Colombian press that decision to suspend applications under the credit was reached because of "unacceptable and unjustified pressure" with a view to "annul a tender and open a new one with a view to the award of an order of several million dollars to a supplier from the United States". The Government simul- taneously announced that 18 meter rails would be purchased from the British firm which had submitted the lowest tender. 3.18 Next day, the Executive Director for Colombia circulated an aide memoire to the other Bank Executive Directors in which, after recounting the several occasions in which Bank staff approved the issue of tenders, and laid the blame with the Bank. 3.19 As a result of further discussions, the Bank agreed in June 1976 to finance the procurement of rails in accordance with the specifications issued to tenderers on January 26, 1976, to the lowest bidder. CNR agreed to ensure that the specifications for any future loan will be drawn up in such a way as to allow the widest possible ICB. The order for the procure- ment of rails was finally placed in August 1976, and the rails were received by CNR during the first part of 1977. D. Consultants' Performance 3.20 Two contracts were signed in 1974, during the initial stage of project implementation, with a firm which had earlier dealings with CNR; the contracts represented a continuation of previous contracts for workshops, track improvements and transport operations. The terms of reference dealt primarily with overall work to be performed and general personnel duties rather than setting forth a specific step-by-step schedule with detailed time limits allotted for every task. In retrospect, it is felt that the period of technical assistance required might have been shortened and the consultants' efforts applied in a more concentrated manner had the work program been spelled out more specifically and in greater detail at the beginning of the assignment. - 25 - 3.21 Less attention was given to the technical assistance aspect of the project than to other items. Further, CNR's counterpart staff were expected to continue fulfilling their previous responsibilities in addition to their assignments in the technical assistance program, which meant that they had to share their time among different activities and coult not concentrate on the training being offered. Consultant personnel also performed some routine tasks not related to the program. A greater degree of supervision in the Bank might have been helpful in alleviating some of these difficulties, which are frequently encountered in institution-building projects. Another difficulty, which influenced the impact of the consultants' recommendations, was that they did not take into account budget constraints which severely limited CNR's implementation of the recommendations. 3.22 As part of this project, another study was undertaken by CNR in 1974, with the help of another group of foreign consultants, aimed at a comprehensive review of CNR's organization and staff responsibilities at all management levels and the preparation of recommendations for its improvement. Although findings and recommendations of this study were satisfactory, they were not implemented adequately because of frequent changes in the management. 3.23 When the project was appraised, it was recognized that there was "a need for a long-term strategy for the development of the whole transport sector." In particular, a multi-modal and comprehensive study of the Magdalena River Basin, where the Atlantic Railway is operated, was to be carried out to generate improved coordination of investments and pricing decisions. Pursuant to Section 3.03 of the Guarantee Agreement of the Loan, the Colombian Government secured technical assistance through bilateral aid to carry out the study. The study was completed towards the end of 1974. 3.24 Unfortunately the study failed to throw a clear light on the problems it was supposed to clarify, particularly the long-term role of each mode of transport in the Magdalena corridor, and the relationship between the road and the railroad. The Bank's comments on this matter pointed out that the consul- tants were faced with two alternatives: (i) to try to optimize the total transport cost and, therefore, accept the eventuality of phasing out one mode of transport; or (ii) to decide a priori that all modes of transport should remain in operation and optimize traffic allocation among modes. The consultants used neither of these alternatives. For these reasons the Bank indicated that "the report cannot be efficiently used, in its present form, in the decision-making process." Actually, the study pre-assigned flows to modes without regard to cost/tariff relationships, and later tried to optimize traffic on each mode, but the railways, in particular, by means of total rather than long-run marginal costs. 3.25 In particular, on future prospects, the study concluded that the railway was doomed in the long-run unless it was permanently subsidized by the Government. Nevertheless, the consultants recommended that the railway should be maintained and improved for political reasons, rather than for economic ones. - 26 - 3.26 If the consultants' conclusions had been correct, in would have meant that any new project with CNR and the Colombian Government, basically a continuation of the rehabilitation of the railway started in Loan 926-CO, would have been a misallocation of resources which should have been avoided. However, a limited review restricted to the basic data in the study was used in the elaboration of the National Transport Plan, which concluded that the railway has an important economic role to play in Colombia, in particular in transporting low-value bulk commodities over medium and long distances. IV. TRAFFIC AND OPERATIONS 4.01 Traffic projections at appraisal assumed an average growth in freight rail traffic about 4.4% p.a., reaching 1,485 million-km (3,292 million tons) by 1977. Major traffic increases were foreseen in agricultural products, fertilizers, steel and, in general, long distance traffic between Santa Marta, Medellin and Bogota. On other sections, mainly Cali-Buenaventura and branch lines, traffic was assumed constant or slightly decreasing in spite of service, improvements, in view of growing road paving programs and truck competition. 4.02 On average, passenger service had barely covered variable costs, and there was a cross-subsidization from first class and motor railcars to other classes of passenger traffic. Shortly after the introduction of new petroleum prices in mid-1971, total rail pass-km increased substantially over that of 1970. This trend continued in 1972,and total traffic for this year reached 398 million passenger-km, as compared to 248 million pass-km in 1970. This large increase was due partly to large tariff increases on buses and partly to better service and introduction of faster trains and better train schedules. However, at appraisal, it was decided to observe the passenger traffic trend in the coming years before committing new investments, and the assumed traffic forecast for 1977 (343 million pass-km) was slightly lower than that of 1972. 4.03 As shown in Table 4, the freight traffic, which grew to 1,391 million ton-km (3.38 million tons) in 1974 declined during the remaining years of the project to a low of 1,128 million ton-km (2.6 million tons) in 1979, mainly because of the poor availability of locomotives (para 3.03 ). Wagon turnaround time improved in 1974 and 1975 to 13 days but was also affected by the lack of locomotives, and deteriorating to 16.6 days by 1979. It is important to mention that only in early 1974 some actions were taken toward better usage of rolling stock; at that time, and following the advice of the consultants, improvements in the system for allocating freight cars and a transportation plan for the Bogota-Santa Marta line were introduced. The operational plan for the Buenaventura-Medellin section was deleted because the line was repeatedly interrupted by land slides and finally closed between Itagui and La Felisa just south of Medellin. 4.04 At appraisal, availability of motive power and rolling stock had been expected to increase from 86% to 87% and from 84% to 86% respectively in the period 1973-1976. With the arrival of 28 new locomotives in 1973, - 27 - locomotive availability increased to 86.2% in 1973 and 86.9% in 1974, but then started to deteriorate to an all-time low of 30% in 1980. The availability of rolling stock (as shown in Table 1) was only 80% in 1973 and 75% in 1979. One of the main reasons for this rather low performance was and remains the high number of derailments, which impose an excessive burden on maintenance operations. In 1973, derailments resulted in damage to 10,248 cars and, in 1979, to 4,188 cars,a decrease of about 60%. The total rolling stock fleet was about 5,700 cars; thus, each car was, on the average, damaged about once a year. At the same time, the capacity of the workshops to absorb this substantial amount of repairs was seriously hampered by a shortage of spare parts and of experienced staff. Personnel shortages continued to be a problem over the 1973-1980 period because of the high turnover of personnel. This high turnover can only partly be explained by CNR's low salary levels and structure and by the staff demoralization due to frequent top management changes. Experienced senior staff continue to leave the railway for better paid positions in the private sector. Mechanics also tend to change jobs to become locomotive drivers after a few years because of the higher salary available for that job, and recruits have been difficult to attract. In addition, CNR management from 1975 onward favored the filling of vacancies in administrative areas, and the critical labor situation in the workshops became worse. All these factors seriously diminished the positive effect of the workshop reorganiza- tion program already started with the assistance of the consultants. 4.05 There has been a marked deterioration of the locomotive fleet since 1975, leading to a progressively declining rate of utilization of locomotives. Severe accidents and lack of spare parts and well trained personnel were adverse factors further aggravated by the serious mechanical problems experienced by the 88 diesel locomotives U-10B(55% of CNR's fleet) purchased from Spain. Lack of locomotives with sufficient horsepower, particularly at the difficult and high gradient section of the Atlantic Region's main line between Bogota and Puerto Salgar adversely affected the operations of the railway. On sections with steep gradients and difficult terrain between Bagazal and and Tribuna (65 km) the diesel engine fails to give a sufficient margin of power for acceptable reliability of service. Further aggravation of this problem was CNR's mismatching of locomotives in multiple units on the high gradient sections. As a consequence of this mismatching, the more powerful locomotive of the two was overloaded and gradually failed to perform. After several claims to the locomotive manufacturers, CNR made various unsuccessful attempts to negotiate a permanent solution, under these circumstances, it was only in 1978 and at a request of CNR that the Bank suggested as a temporary solution to the locomotive problem, the derating of some diesel engines and the use of a group of three matched units in this section. 1/ 1/ For more technical details on locomotive performance and related problems see Mr. C. Buratti's report of May 2, 1979 available in the Project files. - 28 - V. FINANCIAL PERFORMANCE A. General 5.01 During the first two years (1973-1974) of the Sixth Railway Project, CNR's financial performance was good (Table 5). The actual working and operating ratios were better than the appraisal targets; in 1974, the operating ratio was even less than 100. The working and operating ratios are summarized below: 1973 1974 1975 1976 1977 1978 1979 Working Ratio Appraisal Target 106.2 102.3 97.0 91.0 87.6 - - Actual 97.8 90.7 109.9 139.2 119.7 112.0 121.6 Operating Ratio Appraisal Target 122.3 117.5 111.3 103.6 99.0 - - Actual 110.2 98.8 118.7 146.9 125.5 117.0 125.5 In 1973-74, the Government's financial assistance to CNR was adequate and timely. As a result of traffic and tariff increases during 1972-1974, revenues increased by 72% (or Col$ 365 million), while operating costs increased by 38% (or Col$ 239 million). In 1974, total revenues of Col$ 874.2 million were greater than total operating expenses of Col$ 864.1 million, thus producing a net operating profit of Col$ 10.1 million. These improving trends indicated that CNR was on its way to financial recovery. 5.02 CNR, however, could not sustain its improving trends from 1975 onward because of unforeseen circumstances. With the economic slowdown and balance-of-payment difficulties, the Government could not continue its financial support to CNR adequately. The public sector investments in railways were reduced to a meager 0.6% of the total transport budget in 1979 as compared to 9.1% in 1972. The fund shortages forced CNR to curtail invest- ments and to neglect necessary repairs and maintenance of the railway plant. Even though CNR has been increasing tariffs, the increases were not sufficient to cover cost increases. A comparison of consumer price increases and the passenger and freight tariff increases for a ten-year period indicates that, while consumer prices in 1979 were 587% higher than those in 1969, the increases in passenger tariffs and freight rates were only 390% and 459% respectively. In recent years, CNR, however, has been increasing tariffs slightly above the rate of inflation, e.g., in 1980, the freight rates increased, on the average, by about 35% as against the inflation rate of 27%. 5.03 Another setback to railway progress was the traffic decline from 1,391 million ton-km in 1974 to 1,128 million ton-km in 1979 as a consequence of the rapid deterioration of locomotive availability. Between 1975 and 1979, - 29 - revenues increased by Col$ 1,200 million while operating costs increased by Col$ 1,568 million. The working and operating ratios deteriorated from 109.9 and 118.7 in 1975 to 121.6 and 125.5 in 1979, in spite of much curtailed maintenance costs because of a critical shortage of funds. By 1979, the net operating loss had reached Col$ 534.7 million. 5.04 Staff costs, which form 70% of the working costs, also increased considerably as a result of higher social benefits demanded by the strong railway union and the hiring of additional staff in administrative areas. Pensions and indemnities to retired employees as a percentage of basic salaries increased from 8% in 1969 to about 13% in 1979 because of the strong bargaining power of the Railway Union. The fuel cost has also increased since 1975, and at a much faster rate since 1978, because of Government efforts to equalize the local petroleum cost with the international price level. 5.05 The Appraisal Report estimated a return on net fixed assets of 0.2% and a debt coverage of 0.4 in 1977, but because of the factors discussed above, CNR did not generate any net operating revenue after 1974; therefore, neither of the two targets were met. 5.06 The Balance Sheets for 1972-1979 are shown in Table 6. During this period, long-term debt had increased continously from Col$ 1,961 million in 1972 to Col$ 3,214 million in 1979 in current pesos. The net capital first increased from Col$ 1,668 million in 1972 to Col$ 2,307 million in 1973, but started decreasing thereafter, reaching a low of Col$ 1,182 million in 1978. However, as a result of Government increased financial support, net capital showed an increasing trend from 1979. In general, CNR's liquidity position was extremely poor during 1975-1979 because of low revenues, high costs and high debts. The debt/equity ratio deteriorated from 54/46 in 1972 to 67/33 in 1979. The current assets show a gradual increase from Col$ 739 million in 1972 to Col$ 1,669 million in 1979; inflation, and the accumulation of obsolete and slow-moving inventories were the reasons for such increase. The fixed assets are shown at book values adjusted for foreign exchange rate fluctuations on those assets which were imported. The depreciation charges are therefore understated. B. Sources of Financing 5.07 The financing of the investment plan is summarized in Table 7, which indicates that the differences between anticipated and actual financing were relatively minor but because the time of the project execution, 1974-1980, was much longer than the forecast period and because many of the investments were not carried out, care should be taken in interpreting the results. C. Financial Covenants 5.08 The performance with regard to the financial covenants and targets was good in the first years of the project (1973-1974), but, thereafter, the progress was far from satisfactory. The borrower's and guarantor's compliance with main management and financial covenants is discussed below. - 30 - (i) Loan Agreement (a) Improve railway service and reduce or eliminate operating losses (Section 4.08). During the first years (1973-1974) of the execution of the project CNR closed a number of railroad stations, unremunerative passenger train services and branch lines and met the target working and operating ratios (a summary of the services closed is given in Table 8). How- ever, these improving trends stopped in 1975, the financial performance began to deteriorate and, as explained in Chapter IV, the targets were not met thereafter. (b) Independent Auditors (Section 5.02). During appraisal of this project, the Bank agreed to.the Government Auditors (Contraloria) serving as the railway's auditors. These auditors, however, were only certifying the monthly balance sheets and the financial statements, and were not issuing independent comments on the state of affairs of CNR's accounts and finances. At the Bank's suggestion, the Railway Board approved, in 1978, the hiring of private independent auditors. (c) Tariffs (Section 5.04). The overall tariff increases were higher than the ones specified in the Program of Action, but they were still not quite enough to absorb inflationary cost increases. The gap between cost and tariff increases is demonstrated by the fact that while operating costs increased by 587% between 1969 and 1979, average freight and passenger revenues increased by only 519% to 448% respectively. In spite of several Bank mission's recommendations the mechanism for triggering tariff increases was apparently inadequate. (d) Debt Limitation (Section 5.05). This covenant was not complied with by CNR, because of reduced railway revenues, increased costs, and lack of Government support. CNR had to borrow from local banks either to refinance existing loans or to pay off local contractors and meet other similar obligations. (e) Revaluation of Fixed Assets (Section 5.07). The borrower is in default of this covenant. The revaluation of fixed assets has not been carried out, except for a partial revaluation in 1972. CNR intended to carry out this revaluation, but the more serious economic and technical problems encountered kept it from doing so. CNR has now agreed with the Bank that the fixed assets would be inventoried and revalued by December 31, 1982 and at least every two years thereafter. (ii) Guarantee Agreement Provision of Funds to CNR (Section 2.03). The Government complied with parts a(i) and a(ii) of Section 2.03 by providing funds in local currency to aggregate eighty-five million pesos, and by providing funds in foreign currency to aggregate the equivalent - 31 - of twenty million dollars. The Government did not comply with the remainder of the same section, which required providing additional funds needed by CNR to meet its debt service obligations and to carry out the project. As explained earlier, the Government assistance was adequate only in the first two years. With regard to part (b) of Section 2.03, which refers to the opening of the Project Fund in the Banco de la Republica, the Government never deposited the required monthly contributions except for the initial deposit of Col$ 45 million when the Project Fund Account was opened. (iii) Other Covenants of the Loan Agreement As explained elsewhere in this report, Sections 2.04, 3.02, 4.01, 4.07 and Schedule 5, regarding closing date, employment of consultants, experienced and qualified management, carrying out the Investment Plan and the Program of Action respectively, were inadequately implemented. (iv) Other Covenants of the Guarantee Agreement (a) Compliance with Section 3.03 regarding studies in the Magdalena Valley is discussed in paragraphs 3.26-3.29. (b) Section 3.04 regarding improving the planning operations of MOPT was complied with, and the planning section is functioning well. VI. CNR's Institutional Performance and Development 6.01 The improvement of CNR's mana'gement was one of the principle objec- tives of the project. Both CNR and the Government recognized the need for strengthening CNR's management. Some steps, such as the creation of a Planning Office, had already been taken. As a first task, this office provided the General Manager with a Program of Action for 1973 on a divisional basis. The program set targets for revenues, expenditures and reduction of derailments for each Division,and the General Manager was in a position to control the execution of the Program in monthly meetings with the Divisional Managers. 6.02 During negotiations, agreement was reached with CNR to contract management consultants to help improve the railway organization and the timely implementation of the expected recommendations. As explained in paragraph 3.25, CNR with the help of consultants prepared a study on management but it was never implemented. 6.03 Over the period of the project, however, some favorable changes have taken place in the organization. The Operations Department was split into the Technical Department and the Operations Department. In 1974, three new positions were created - the Deputy General Manager, the Commercial Manager (to control the marketing aspects of CNR), and the Coordinator of the Planning Unit. The Deputy General Manager directly controls the Division Managers and maintains a functional relationship with the Departmental Managers at headquarters. - 32 - 6.04 Beginning 1975, a number of qualified and experienced railway engineers and technicians left CNR for better paying jobs and many inexperi- enced professionals were hired from outside the railway. The continuity of railway supervisors therefore deteriorated (para 3.02). 6.05 Although total staff decreased from 11,408 in 1973 to 10,365 in 1980, there is still an excess of employees in the administrative areas. At the same time, there is a shortage of technicians and skilled workers in track, maintenance and workshop areas. There is an urgency to rectify this unbalance of staff and CNR has agreed to take necessary actions under the seventh railway project. 6.06 CNR's present inefficiency is partly due to a lack of good manage- ment. While part of the blame lies with the Government's ne lect of the rail- way in the past few years and the failure to provide funds, CNR's management is also to be blamed for some of the decisions taken to the disadvantage of the railways. For example,for more than three years, there have been vacancies for engineers in the Mechanical and Way and Works Department. Another example is that, in an attempt to reduce costs, the management stopped the traveling of instructors to train railway employees on the job. During 1978, only 13 out of 84 training courses were carried out. VI. ECONOMIC RE-EVALUATION 7.01 At the time of appraisal, the investment plan was broken up into two parts: (a) investments pertaining to the Bogota-Santa Marta Line, and (b) investments in the rest of the system. Economic rates of return (ERR) were calculated in each case and also for the investment program as a whole. The main quantified benefit was the avoidance of freight traffic diversion to roads. The ERRs obtained were 13% for the investment in the Bogota - Santa Marta line, 11% for the investment in the rest of the system, and 12% for both together (the Investment Plan). 7.02 The economic re-evaluation of the project has been based on the investment plan as a whole because of the lack of the necessary information for a more disaggregate analysis. The costs used in the re-evaluation are CNR's actual investments in the 1973-1980 period, and the economic benefits have been estimated by comparing the total transport cost for the country in the "without" and "with" situations, based on the actual railway traffic in the 1973-79 period . 7.03 The "without the project" case has been forecasted using the same assumptions as in the appraisal report. This is a conservative approach since it could be argued that because of CNR's difficulties, the "without the project" performance would also have been lower. Since there is no reasonable method to estimate such lower performance, the appraisal estimate has been used. It was then assumed that railway operating costs would increase at an annual rate of 2% from the 1973 level and that only the Bogota-Santa Marta line would remain open to traffic. Railway freight traffic was assumed to decline to a level of only 700 million ton-km by 1977 with the diverted traffic being transported by road. - 33 - 7.04 The total transport cost in the "without the project" case includes the road transport cost for the diverted traffic and the railway operating cost for the traffic that remained with the railways. While in the "with the project" case, no traffic diversion to road transport is assumed and, therefore, the total transport cost is just the railway operating cost. 7.05 The traffic that would have been diverted to road transport without the project was estimated by comparing the actual railway traffic in each year of the 1973/79 period with the corresponding figures for the "without" situation. This was substantially higher than the appraisal forecast in the first two years of the implementation of the project (1973/1974) and lower than expected for the remaining years (Table 9). 7.06 Road transport costs for the diverted traffic were estimated from the Colombia Ministry of Public Works and Transport's annual calculations of vehicle operating costs. These resulted in costs/ton-km in 1979 about 40% higher than the appraisal estimates. We have assumed a steady increase in this cost from the appraisal estimate in 1973 to this new value corresponding to 1979. The total road transport cost for the traffic that would have been diverted in each year is also shown in Table 9; while for 1973-1974 these figures are substantially higher than the appraisal forecast, they are close to those forecast thereafter because the reduced volume of traffic diversion had been offset by the higher road unit cost. 7.07 The other component of the total transport cost in the "without" situation is the railway operating cost for the traffic that has remained on the railway. These costs were estimated using the same assumptions as in the appraisal report. The total transport cost in each year for the "with" and "without" the project situations as well as the investment cost stream, are shown in Table 10. 7.08 The resulting ERR is 11%, which is surprisingly high considering the difficulties facing this project. The appraisal forecast was 12%. The main reasons for the relatively high ERR are the higher road transport costs, substantially lower than forecast investment levels in the 1973-1980 period (in constant prices) and much higher net benefits in the first two years of project implementation (1973-1974) 1/, which, in turn, are due to the higher traffic levels in these two years. The reduced actual investment, as compared with the appraisal forecast, is mainly explained by the much lower than anticipated procurement of switches and concrete sleepers (actual expenditure in these items was (in constant prices) only 9% of the forecast) and by the use, at appraisal, of a shadow exchange rate 10% higher than the offical rate. 1/ Only half of the benefits for 1973 have been considered since the 28 locomotives (para 3.01) were delivered early 1973 and were fully operational for only half the year. - 34 - 7.09 The main differences between the benefit and cost streams in this economic re-evaluation and those at the time of appraisal are the following: (a) Higher than forecast avoided traffic diversion in the first two years (1973-1974) and lower thereafter. (b) Higher than expected road transport costs, due mainly to the increase in fuel prices. (c) The combined effect of (a) and (b) is to generate much hi her than forecast economic benefits from avoided traffic diversion in 1973-1974 (1.5 times larger) and lower than the forecast thereafter. (d) The project cost (actual investments) used in the economic- re-evaluation is lower than the appraisal forecast (in constant prices) and the implementation period is longer (para 7.08). (e) The combined effect of items (a)-(d) preceding is to generate substantially higher than forecast net economic benefits for the first two years of project implementation and lower thereafter. 7.10 The economic return of 11% is slightly below the cut-off rate of 12%, which MOPT uses for evaluating transport projects. This indicates that, in retrospect, the project as carried out had only a marginal economic contribution to the sector, which is not surprising in light of the serious problems in project implementation. In fact, had road transport cost in real terms been as forecast during appraisal, the ERR would have been substantially lower. VII. THE ROLE OF THE BANK 8.01 The Bank has played an important role in the development of Colombia's transport sector. Its involvement dates back to 1949, when a transport sector mission reported the transport system to be in exceptionally bad condition. Since 1950, the Bank has lent about US$441 million in 21 loans to the sector. These investments accelerated the construction of an integrated highway netwoik and, more recently, have contributed to the rehabilitation and maintenanze of the network and to the development of an improved highway organ.zation. They have also contributed to the construction of over 670 km of main line on the Atlantic railroad from La Dorada to Fundacion, as well as the rehabilitation of others. The domestic aviation project is helping to improve basic aviation in infrastructure and to improve sector efficiency and planning. Bank participation in rural road construction has been through rural develop- ment projects, which also included other activities in the agricultural sector. - 35 - 8.02 The Bank was also deeply involved in preparing the Colombia Sixth Railway Project and in its implementation. In retrospect, the Bank seems to have been right in stressing the importance of setting up operational targets, but these were useful only as long as CNR had a management team committed to them. The Bank's efforts probably contributed to the success achieved during the first two years when with competent management and support from the Government, CNR achieved good results. In spite of the fact that the Bank through frequent supervision missions raised to the Government and CNR's management the issues referred to in the preceding paragraphs and proposed corrective measures, no full positive results were obtained because of the lack of interest from the Government. It is doubtful that from 1976 through 1979 the Bank's efforts materially influenced the performance of the railway. At one point towards 1977, the Bank seriously considered cancelling the loan, but about 90% of the loan was already committed and there was no leverage left, therefore, the idea of cancelling the loan was abandoned. 8.03 The process of the procurement of rails, (paras 3.10-3.19) has revealed that even without going into the technical merits of the two lengths proposed for the rails, it seems that most of the embarrassment could have been avoided had the question been fully considered at the time when clearance of bid documents was originally given, both in relation to what was required and what the market may have to offer. VIII. CONCLUSIONS 9.01 Considering the complexity of the Project, the Bank seems to have been right in supervising the project more closely than others. It is also clear, that for the success of a project, it is indispensable to have a competent management to implement it and the full support from the Government which did not happen in this project except in its earlier stage. 9.02 After the first two years of the project the lack of revenues and financial support frequent changes of CNR's top management and the continued policy changes severely constrained the institution's performance and delayed the project's implementation, showing results below appraisal expectation. With all its shortcomings in execution and in meeting operating and financial objectives, the limited investments made and related actions, arrested further deterioration of the railway system and allowed CNR to continue providing key transport services to Colombia. If the project had not been carried out, CNR operations and equipment would have deteriorated more rapidly and rail service would have been suspended. 9.03 Nonetheless, CNR has reached a stage at which continuation of existing operations would deteriorate the service to a point that, eventually it would be necessary to close down the entire railway network. An alternative course is to take appropriate technical, operational and financial measures, provided they are economically justified, to remedy the immediate problems, followed by a well conceived investment and action plan designed to improve railway operations. In view of recent changes in the world energy situation, including rising petroleum prices which have enhanced the railway's competitive position vis-a-vis road, and the potential for exploiting Colombia's coal - 36 - deposits which are among the largest in the world, as well as changes in traffic composition and patterns, the future role of the railway has changed. The Government has acknowledged this fact in the National Transportation Plan and, in view of this situation, is taking remedial actions and efforts to rehabilitate the railway. 9.04 With this in mind, the Colombian Government requested Bank support, which has resulted in the preparation of the proposed Seventh Railway Project, in which the most critical areas are addressed. Special emphasis has been given to measures to avoid the shortcomings of the Sixth Project. 1/ 1/ The Bank would be financing locomotives for the difficult section Bagazal-Tribuna, gondolas for the coal traffic, spare parts for an intensive locomotive rehabilitation program, rails and fittings for the track, telecommunication equipment and technical assistance. The Bank has also received the commitment from CNR that appropriate ties will be procured, as well as to tackle difficult institutional problems including enforcing disciplinary measures. The Seventh Railway Project will closely follow CNR's institutional, operational and financial performance; these will be measured against time-phased targets which have been set at six- month intervals. It is proposed that the release of loan funds be related to the achievement of these targets. In addition, the success of the execution of the project and the achievement of the Action Program target would be especially examined by the Bank, CNR and the Government in a mid- term review. The continuation of disbursement of the Bank loan beyond this review would depend upon the progress achieved by then. - 3TABLE 3-A COLOMBIA SIXTH RAILWAY PROJECT (LOAN 926-CO) PROJECT COMPLETION REPORT Actual and Appraisal Estimates of Investments (In 1973 Col$ and US$ Millions) Appraisal Estimate of Cost 1/ Actual Cost (constant prices) 2/ Col$ US$ Col$ US$ Permanent Way Department 434 km track rehabilitation 193,808 8,614 360.7 16,031 754 switches 34,999 1,556 ------ Cancelled ------- 576 km concrete sleepers 289,893 12,884 29.2 1.298 175 km hard rails 81,422 3,619 30.3 1.347 Infrastructure & bridges 110,424 4,907 77.9 3.462 Ballasting & resleepering 49,804 2,213 66.7 3.964 Aluminothermic welding 46,502 2,067 29.7 1.320 Maintenance equipment & ballast crushers 65,990 2,933 45.4 2.018 100 ballast cars 30,193 1,342 38.1 1.693 Subtotal 903,035 40,135 678.0 30.133 Mechanical Department 28 diesel locomotives 160,492 7,133 133.8 5.946 Spares for locomotives 86,446 3,842 178.3 7.924 Spares for freight cars 98,743 4,388 109.9 4.884 Machinery & equipment for workshops 62,616 2,783 17.4 0.773 Subtotal 408,297 18,146 439.4 19.527 Operations Department Stations & yards 56,448 2,509 49.3 2.191 Warehouses 10,080 448 - - Freight handling equipment 19,900 884 10.6 0.471 Trucks & trailers 26,122 1,161 11.1 0.493 Passenger coaches - - 4.7 0.209 Rail cars - - 0.5 0.022 136 containers 4.610 205 - - Subtotal 117,160 5,207 76.2 3.386 Dwellings 15,120 672 9.7 0.431 Telecommunications 182,599 8,115 33.4 1.484 Consulting Services 81,250 3,611 47.4 2.107 Winter Damages 43 218 1 921 80.3 3.569 Subtotal 622873 170.8 7.591 Total 1,750,679 77,807 1,364.4 3/ 60.640 3/ Contingencies 363,316 16,148 _....__ --- GRAND TOTAL 22113.995 93955 1 364.4 60.640 1/ Estimated period at appraisal 1973-1976. 2/ Actual period 1974-1980. 3/ In current prices Col$ 2,680.8 million or US$75.7 million. Note: 1973 US$1 - Col$ 22.5 October 1981 - 40 - TABLE 3-B COLOMBIA SIXTH RAILWAY PROJECT (LOAN 926-CO) PROJECT COMPLETION REPORT Actual and Appraisal Estimates of Project Costs (In Constant 1973 Col$ Millions and US$ Millions). In Local Currency In Foreign Currency Appraisal Appraisal Permanent Way Department Actual 1/ Estimate 2/ Actual 1/ Estimate 2/ 364 km track rehabilitation 174,000 3/ 143,511 7,733 3/ 6,378.2 530 switches - T/ 24,177 - 4/ 1,074.5 152 km concrete sleepers 25,400 5/ 76,499 1,129 5/ 3,399.9 175 hard rails - 81,422 - 3,618.8 Infrastructure & bridges 52,300 46,741 2,324 2,077.3 Ballast & resleepering 19,600 28,397 871 1,262.1 Aluminothermic welding 22,100 23,489 982 1,044.0 Maintenance equipment & ballast crushers 27,800 43,341 1,236 1,926.5 50 ballast cars 6,400 6/ 15,097 284 6/ 671.0 Subtotal 327,600 482,678 14,559 21,452.3 Mechanical Department 28 diesel locomotives 133,800 160,492 5,946 7,133.0 Spares for locomotives 87,900 42,672 3,907 1,941.0 Spares for freight cars 12,600 43,513 560 1,933.9 Machinery & equipment for workshop 7-700 342 - Subtotal 242,000 247,677 10,755 11,007.9 Operations Department Stations & yards 26,600 16,128 1,182 716.8 Warehouses - 672 - 29.9 Freight handling equipment 10,600 3,544 471 157.5 Trucks & trailers 9,400 26,122 418 1,161.0 Passenger coaches 2,600 - 116 - Rail cars -_-_-_- Subtotal 49,200 46,460 2,187 2,065.2 Miscellaneous Dwellings 1,800 7,392 80 328.6 Telecommunications 18,100 31,830 804 1,414.7 Consulting services 6 training 39,600 48,471 1,760 2,154.3 Winter damages 74,200 43,218 3.298 1,920.8 Subtotal 133,700 130,911 5,942 5,818.4 Total 752,500 907,732 33,444 40,343.8 Contingencies - 87,935 - Total 752,500 995,667 33,444 44,252.0 Interest during construction - 36000 GRAND TOTAL 752.500 1,031,667 33.444 1/ Actual, refers to the period 1974-1975. 2/ Estimated period at appraisal, refers to 1973-1974. 3/ Only 235 km of track rehabilitated. 4/ The procurement of switches was cancelled. 5/ Only rail elastic fittings for concrete sleepers were procured, and concrete sleepers were laid in only 3 km of track on a trial b 6/ Part of the actual costs of 50 ballast cars. Note: 1973 US$1 - Col$ 22.5 August 1981 - 41 - COLOMBIA TABLE 4 STU RAILWAY PROJECT (LOAR 926-CO) PROJECT COM1PLETION REPORT Selected Operating Statistics 1970-1979 2u 122M 121 171 171 121l U9M I U7 21 11. 1111 1. Sy.ge Total Rout* km 3,431 3,431 3.431 3,403 3.403 3,403 3.403 3,403 3,403 3,403 In operation km 3,353 3.353 3,353 3.232 3.232 3.138 2,767 2,912 2.912 2,912 Out of popration km 78 78 78 171 171 265 636 491 491 491 Total Staff 8 11.526 11.654 11.516 11,408 11.662 12.002 12.107 11.345 10.980 10,345 11. Traffic Passeogera Caried million 2.95 3.16 4.27 4.20 4.56 5.18 4.05 2.97 2.56 2.46 7.&aenger km million 248 281 398 427 482 522 510 391 342 322 Averes* Journøy km 112 117 123 131 131 120 126 131 133 131 Total Net Tone million 3.06 3.04 3.07 3.06 3.38 2.92 2.99 2.88 2.96 2.60 N.t Pying Tone million 2.78 2.65 2.73 2.76 2.90 2.44 2.41 2.52 2.68 2.39 Net Not Pøying Tona million 0.29 0.39 0.34 0.30 0.48 0.48 0.58 0.36 0.28 0.21 Total Not Ton-km million 1.239 1.215 1,248 1.368 1.391 1.206 1.221 1.254 1.261 1.128. Not Paying Ton-km million 1.173 1,150 1,198 1,331 1.329 1.138 1,156 1.215 1,232 1,105 N.t Not Paytg Ton-km million 66 65 50 37 62 68 65 39 29 23 Total Cro.. Ton-km million 2,416 2,409 2.419 2.640 2.234 2.366 2.379 2,358 2.313 2,055 Freight Cross T.n-km million 2,190 2.159 2,124 2.313 2.413 2,028 2.047 2,088 2.088 1.854 P8gsenger Cross Ton-km million 226 250 285 325 321 338 332 270 224 201 Av.rag. Naut km 442 434 439 482 458 466 479 482 459 462 Total Freight Car-km million 70.41 69.89 64.81 70.13 75.74 63.23 63.49 64.20 63.59 55.83 l.ad*d Fr.ight Car-km million 47.14 46.28 44.18 47.43 50.53 42.99 43.51 42.33 42.66 38.24 Empty Frlght Car-km million 23.27 23.61 20.63 22.70 25.20 20.24 19.98 21.87 20.73 17.59 Ill. Traffic Deneity P...ngør-ka pr Røut, km (000) 70.4 79.6 113.8 126.8 144.8 161.8 184.3 134.2 117.1 109.2 Freight Net Toa-km par Routa km (000) 349.8 342.9 357.3 411.8 411.2 362.6 417.2 417.2 423.1 403.5 TV. Operatione Total Tr.in-km million 8.39 7.61 6.49 6.98 7.38 6.99 6.92 6.27 5.76 5.31 Tran-ka. Peøongare milliom 1.92 2.16 2.40 2.57 2.69 2.70 2.69 2.23 1.82 1.56 Train-km. Fr.ight million 6.47 5.45 4.09 4.41 4.69 4.29 4.23 4.04 3.94 3.75 v. Operating Efficincy (Fright) rog. Toa-km/Traiø-hss t 338.5 396.1 519.1 534.9 514.5 472.4 464.0 316.6 329.3 494.0 møt TØo-km/Trai-km t 181.3 211.0 293.0 301.8 283.0 281.0 R88.8 310.2 319.8 300.8 Not Too-kalCar-ka t 26.9 26.2 28.3 28.8 27.3 28.0 28.1 29.6 29.4 29.3 Car Turnrond Time daye 14.2 15.8 16.7 16.2 13.1 13.0 14.3 15.9 15.3 16.6 AveSg speed km/h 20.8 20.3 19.7 19.4 17.1 18.1 18.1 17.6 17.2 17.4 Loco. kn/day 1/ 231.9 222.5 193.7 216.4 223.9 215.7 219.7 225.4 243.7 251.6 Trtffi. Unit,/EMploy (000) 123.3 122.8 138.5 154.1 155.3 138.3 137.6 141.6 143.3 138.0 Diel0 Locomotive, % 85.2 85.4 86.1 86.2 86.9 83.2 80.3 73.8 68.1 57.9 Fre.,ht cara % 83.9 85.7 81.7 80.2 77.9 67.7 73.4 75.4 77.8 74.8 VI1. Derail.ent. Total MN-b.r~r 7.116 5.969 4.368 5.471 3.614 3.122 3.306 3.228 3.310 3.117 Total aours/lo.t 25.759 18.263 14.584 21.589 13,526 12.172 11.114 9,587 10.003 a... Hour. Lst/aDrailment 3.62 3.06 3.34 3.95 3.74 3.90 3.36 2.97 2.84 n.a. V&hiCle Dmaad 11.569 9,462 7.701 10.248 6,732 6,303 6,136 5.652 6.318 4,188 1/ Average per lotootive to ervece. Source: CNR Juoly 1980 CLOMBIA SIXTH RAILWAY PROJECT (LOAN 926-CO) PROJECT 00MPLETION REPORT Actual Revenues, Expenses and Net Income (Million Pesos) 1973 1974 1975 1976 1977 1978 J.. Appraisal Appraisal Appraisal Appraisal Appraisal Forecast Actual Forecast Actual Forecast Actual Forecast Actual Forecast 1/ Actual Actual Actual Operating Revenues: Freight - Cosmercial (including LCL) 466.0 519.9 495.9 689.9 520.6 659.9 547.2 807.1 564.3 1,168.1 1,545.0 1,721.1 - Non-Revenue (Government paid) 26.0 25.6 26.0 57.2 25.0 73.5 25.0 98.3 25.0 104.2 75.4 71.2 Passenger and Express 65.3 66.9 70.9 88.8 70.7 117.5 76.0 150.3 76.0 165.6 166.6 178.9 Miscellaneous - (storage, rent, demurrage, water & light) 22.5 25.9 23.0 38.3 26.0 46.0 28.5 44.9 32.0 83.9 154.0 125.4 Tariff Increase - - 65.2 - 145.0 - 242.6 - 354.7 - - Total (a) 579.8 638.3 681.0 874.2 787.5 896.5 919.3 1,100.6 1,052.0 1,521.8 1,941.0 2,096.6 Operating Expenses: Road Maintenance 166.5 179.6 186.0 225.0 203.8 304.5 223.9 458.9 245.4 540.6 634.4 730.4 Equipment Maintenance 126.3 100.3 149.6 120.8 168.3 156.5 185.8 239.0 207.0 291.0 349.0 388.7 Sales and Transportation 247.8 258.9 283.4 307.9 312.2 378.2 342.3 624.2 376.3 757.1 888.0 1,102.3 General 80.4 85.5 86.4 139.8 93.3 145.9 100.8 210.1 109.0 234.2 312.2 348.5 Total Working Expenses 621.0 624.3 705.4 793.3 777.6 985.1 852.8 1,532.2 937.7 1,822.9 2,183.6 2,549.9 I Depreciation 93.0 79.3 104.0 70.8 113.0 78.7 118.0 84.4 120.0 87.8 83.7 81.4 Reduction of looses in uneconomic lines 5.0 - 9.0 - 14.0 - 16.0 - 16.0 - - Total (b) 709.0 703.6 800.4 864.1 876.6 1,063.8 954.8 1,616.6 1.041.7 1,910.7 2,267.3 2,631.3 Net Operating Revenue/(Loss) c-(a-b) (129.2) (65.3) (119.4) 10.1 (89.1) (167.3) (35.5) (516.0) 10.3 (388.9) (326.3) (534.7) Not-operating Revenue (net) (sale of scrap and land-profit only) 19.0 18.5 19.0 17.1 15.0 18.3 10.0 129.7 10.0 38.9 19.7 9.7 Net Revenue/(Loss) (d) (110.2) (46.8) (100.4) 27.2 (74.1) (149.0) (25.5) (386.3) 20.3 (350.0) (306.6) (525.0) Interest 87.5 131.0 97.5 387.0 129.9 510.6 143.6 416.7 143.8 454.2 556.6 515.1 Net Income/(Deficit) (197.8) (177.8) (197.9) (359.8) (204.0) (659.6) (169.1) 803.0 (123.5) (804.2) (863.2) (1,040.1) Breakdown of Working Expenses Personnel 423.5 435.9 487.6 553.9 538.8 687.9 591.0 1,070.0 650.1 1,314.4 1,524.8 1,780.6 Fuel 28.4 28.2 32.2 35.9 36.5 44.5 40.6 69.3 47.0 86.2 98.7 115.3 Material 97.2 82.8 108.7 105.2 120.3 130.6 133.2 203.1 147.5 215.6 289.5 338.1 Other 71.9 77.5 76.9 98.3 82.0 122.1 88.0 189.8 93.1 206.7 270.6 315.9 Significant Ratios Working Ratio 106.2 97.8 102.3 90.7 97.0 109.9 91.0 139.2 87.6 119.7 112.0 121.6 Operating Ratio 122.3 110.2 117.5 98.8 111.3 118.7 103.9 146.9 99.0 125.5 117.0 125.5 Times Interest Earned - - - - - 0.3 - - 0.1 - - - Debt Coverage - - - - 0.1 - 0.2 - 0.4 - - Return on Net Fixed Assets - - - - - - - - 0.2 - - 1/ This appraisal forecast was shown in the Appraisal Report but the operating ratio was not shown in the Loan Document. Source: CNR November 1980 PROJECT COMPLETION REPORT Colombia: Sixth Railway Project (Loan 926-CO) SumDary Balance Sheets (Million Pesos) 1972 1973 1974 1975 1976 1977 1978 1979 ASSETS Current Assets Cash 40.0 27.4 118.8 134.0 91.3 91.0 148.5 241.9 Inventories 458.0 456.6 458.7 679.4 726.8 887.0 801.7 739.5 Other 241.0 324.2 203.7 217.8 288.4 397.2 549.9 687.6 Total 739.0 908.2 781.2 1,031.2 1,106.5 1,375.5 1,500.1 1,669.0 Fixed Assets Gross Fixed Assets 3,745.0 4,444.7 4,221.7 4,522.6 4,872.7 5,083.5 5,177.2 5,250.9 Less Accumulated Depreciation (534.0) (612.4) (681.6) (749.9) (829.9) (917.1) 993.4 1,072.8 Net Fixed Assets 3,211.0 3,832.3 3,540.1 3,773.6 4,042.8 4,166.4 4,183.8 4,178.1 Miscellaneous Assets 226.0 563.7 748.9 726.2 687.7 740.6 890.2 900.5 TOTAL ASSETS 5,068.2 5,531.0 LIABILITIES AND CAPITAL Current Liabilities 364.0 404.6 265.8 494.4 869.8 1,235.6 1,424.4 1,657.1 Long-Term Debts 1,961.0 2,300.3 2,675.7 3,394.0 3,449.9 3,727.9 3,811.5 3,213.6 Capital and Surplus 2,684.0 3,621.1 2,681.3 3,918.0 4,546.1 5,218.7 6,094.3 7,534.4 Less Accumulated Deficits (1,016.0) (1,013.9) (1.740.9) (2,440.3) (3.243.3) (4 047.5) (4)912.7 (5,944.5) Net Capital 1,668.0 2,307.2 1,940.4 1,477.7 1,302.8 1,172.2 1,181.6 1,589.9 Reserve for Social Benefits - (cesantias, pensions, vacations, etc.) 183.0 192.1 186.3 185.1 214.5 147.8 156.6 287.0 TOTAL LIABILITIES AND CAPITAL 4,76. 5,204.2 5_068.2 5310 5_837.0 6,282.5 6,574-1 RATIOS Current Assets/Current Liabilities 2.0 2.0 2.9 2.1 1.3 1.1 1.1 1.0 Current Assets less Inventories/Current Liabilities 0.8 0.8 1.2 0.7 0.4 0.4 0.5 0.6 Debt/Equity 54/46 50/50 58/42 70/30 73/27 76/24 77/23 67/33 Source CNR April 1981 COLOMBIA SIXTH RAILWAY PROJECT (LOAN 926-CO) PROJECT COMPLETION REPORT Actual and Appraisal Expectation of Financing of Investment Plan (In 1973 Col$ Millions and US$ Millions) 1/ In Local Currency In Foreign Currency Appraisal Appraisal Actual Expectation Actual Expectation Sources Col$ % Col$ % US$ % US$ % Government 477.74 35.0 474.0 22 21.23 35.0 21.07 22 IBRD (including interest during construction) 474.87 34.8 562.5 27 21.10 34.8 25.00 27 Credits from private banks & loans from local banks 112.02 8.2 535.0 25 4.98 8.2 23.78 25 Bilateral sources for locomotives & spares (Spanish suppliers & CAFETEROS) 174.22 12.8 152.6 7 7.74 12.8 6.78 7 CNR (Polish rails & other railway materials) 125.51 9.2 36.0 2 5.59 9.2 1.60 2 Further foreign loans to be arranged for 1975-1976 - - 353.9 17 - - 15.73 17 Total 4 100.0 2,114.0 100 60.64 100.0 93.96 100 1/ Total gross domestic product index: 1973 - 100.0; 1974 - 127.6; 1975 - 154.2; 1976 - 190.6; > 1977 - 244.6; 1978 - 280.2; 1979 - 347.7; 1980 - 386.0 August 1981 TABLE 8 COLOMBIA SIXTH RAILWAY PROJECT (LOAN 926-CO) PROJECT COMPLETION REPORT Uneconomic Lines and Passenger Train Services Closed Branch Lines 1. Cartago - Pereira (33 km) Pacific Network 2. Pereira - Manizales (78 km) Pacific Network 3. Pereira - Armenia (60 km) Pacific Network 4. Suarez - Popayan (94 km) Pacific Network 5. La Felisa - Itagui (Medellin) (130 km) Pacific Network 6. Chiquinquira - Barbosa (67 km) Central Division Passenger Services 1/ 1. Cali - Armenia (suspended) Pacific Network 2. Cali - Buenaventura Pacific Network 3. Cali - Tulua Pacific Network 4. Bogota - Chiquinquira Central Division 5. Bucaramanga - Barrancabermeja Central Division 6. Bogota - Girardot Central Division 1/ In addition, the frequency of a number of existing services had been reduced. Source: CNR October 1981 COLOMBIA SIXTH RAILWAY PROJECT (LOAN 926-CO) PROJECT COMPLETION REPORT Traffic Diversion Effect Railway Traffic Avoided Traffic Diversion Avoided Traffic Diversion Without Appraisal Appraisal Actual Forecast Project Forecast Actual Forecast 1/ Actual 2/ ----- (million ton-km)------ (71(million ton-km) 27-- --(Col$ millions 1979)---- 1973 1,368 1,240 1,178 62 190 159 494 1974 1,391 1,305 1,063 242 328 620 905 1975 1,206 1,370 946 424 261 830 762 1976 1,221 1,440 823 617 399 1,580 1,228 0 1977 1,254 1,485 700 785 554 2,010 1,794 1978 1,261 1,485 700 785 562 2,010 1,911 1979 1,128 1,485 700 785 428 2,010 1,526 1/ Road transport cost at appraisal Col$ 2.56 (1979 prices) times column (1). 2/ New estimate of road transport unit cost for each year was calculated assuming a smooth increase from the value for 1973 (appraisal estimate Col$ 2.56 in 1979 prices) to the new estimate for 1979 (Col$ 3.60 in 1979 prices). The total road transport cost in the actual situation is then calculated applying these unit costs to the volume of traffic indicated in column (2). April 1983 COLOMBTA SIXTH RATLWAY PROJECT (LOAN 926-CO) PROJECT COMPLETION REPORT Total Transport Cost "With" and "Without" the Project (Col$ Millions in 1979 Prices) Total Transport Cost "Without" the Project Road Transport Railway Total Transport Cost Cost Cost TOTAL "With" the Project Investment 1973 494 2,276 2,770 2,497 590 1974 905 2,112 3,017 2,326 1,008 1975 762 1,917 2,679 2,307 1,266 1976 1,228 1,707 2,935 2,918 866 1977 1,794 1,475 3,269 2,745 575 1978 1,911 1,505 3,416 2,798 112 1979 1,526 1,535 3,061 2,549 198 1980 1,526 1,566 3,092 2,549 412 1981 1,526 1,566 3,092 2,549 0 0 Revised April 1983  IBRD 3667 R] - - - - DECEMid *as ^v ENEZUELA .s . -Capulco - SANTANDER " WJ,lche'h d.. _- ,-h.- ýWlhs Gascia \ ucaramang aCadeno A N I Q U l A arrancabermeja e. a~~rareARA cA SA N T A N D E Rf Grecia uerqo SnBerro Cooboc Ntgu Medelden lmoo tg Borbosa R4 1 i ." , Ip - B leic- 11 Beleoconc L H~ ~ ~ C A DA ri a qudnqur on, a. ism RISA RA LDA .'ð CA aiua cpaquiró . La Fehispn anazal 6 CA ANA R i, 11 " Manizales CUNDI1' IV RiaCASANA R E Car'tao nF n co a aro A PereN a rab B GOT ** N) ArmniaBOGOTA dalueia U Zarae• ba,o El N SaNto -. · E pinal \fBuenaveitura O L l M A palmiro, Yumbo OPradera A L L E Calh /'m E T A 0 Su c0ri · antander LSnanao Newva rCOLOMBIA RISARALOA"NATIONAL RAILROADS C CAOUETA - MAIN LINES - - BRANCH LINES El D v iso '.)SECONDARY LINES ---.......... LINES CLOSED - -- -......- PROJECTE D L IN E S.,. 100 150 200 250agaza PUTUMAYO KlMERS- - DEPARTMENT BOUNDARIES ECUADOR -INTERNATIONAL BOUNDARIES