Colombia - Fifth Railway Project

CONFIDENTIAL Report No. 776 INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT PROJECT PERFORMANCE AUDIT REPORT on COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) June 11, 1975 Operations Evaluation Department  PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) PREFACE This report presents an audit of achievements under the Colombia Fifth Railroad Project, for which Loan 551-CO of July 25, 1968, in the amount of US$18.3 million, was finally closed on December 31, 1972. The audit is based mainly on correspondence and reports in Bank files (Loan and Guarantee Agreements, Appraisal Report, Progress Reports, Supervision Reports, and correspondence between the Bank and the Borrower), as well as discussions with Bank staff members. In February 1975, a one-week visit was made to the Republic of Colombia in connection with this report. The valuable assistance of the Colombian National Railroads is gratefully acknowledged.  PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) PROJECT DATA Amount of Loan US$18.3 million Amount Disbursed 1 US$12.4 million Date of Loan Agreement July 25, 1968 Date of Effectiveness September 20, 1968 Original Closing Date June 30, 1971 Final Closing Date December 31, 1972 First Supervision Mission October 1968 Final Supervision Mission December 1972 Exchange Rates: 1968 ------- USSl = Col$16.95 1969 ------- US*1 = Col$17.93 1970 ------- US$1 = Col$19.17 1971 ------- US$1 = Col$21.00 1972 ------- US$1 = Col$22.88 1973 ------- US$1 = Col$24.89 1/ About US$5.9 million was cancelled as a result of a successful barter arrangement with Poland for the procurement of rails.  PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) TABLE OF CONTENTS fage No. Summary Introduction 1 The Bank and the Loan 2 Project Implementation 6 Project Cost 8 Operating Performance of the Colombian National Railroads 8 Financial Performance of the Colombian National Railroads 14 Economic Justification of the Project 16 The Role of the Bank 18 Conclusions 20 Annexes 1. Investment Program of Colombian National Railroads, 1968-70 2. Consulting Services 3. Estimated and Actual Investment Program of Colombian National Railroads, 1968-72 4. Derailments on Colombian National Railroads, 1967-73 5. Selected Operating Statistics of Colombian National Railroads, 1968-73 6. Forecast and Actual Revenues and Expenses of Colombian National Railroads, 1968-73 7. Forecast and Actual Balance Sheets of Colombian National Railroads, 1968-72 Map: Colombian National Railroads  PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) SUMMARY In 1965, the Colombian National Railroads (CNR) applied to the Bank for a loan to support a Fifth Railroad Project intended to help finance CNR's revised investment program for the 1968-70 period. After about three years, in July 1968, Loan 551-CO for US$18.3 million was signed. The con- sideration of this application was first delayed by the Bank, for about a year, because of CNR's poor performance under the ongoing Fourth Railroad Project (Loan 343-CO for US$30 million in 1963). In 1967, following some improvements in CNR's operating efficiency, the Bank was willing to consider the application, but further delay occurred when the Government decided to procure the material for track rehabilitation through a barter agreement with Poland. The project supported by the loan represented a modification of CNR's original investment program, with many important changes introduced by the Bank. Some investment items were deleted because, with only minor improvements in efficiency, they would not be necessary, and extensive con- sulting services were added. The planned total investment in the program was US$81 million,and the main project components and their share of the total investment were: track renewal (46%), motive power and rolling stock (32%), construction and equipment of workshops (10%), increases in stations' capacity (5%), consulting services (4%), and miscellaneous (3%). Implementation of the project proceeded slowly, with delays of about two years in completion of some of the main components. The closing date of the loan was postponed by one and a half years and it was finally effected on December 31, 1972. Only 79% of the proposed track rehabilitation work was completed, and problems occurred with the quality of the materials used as the rail showed excessive wear after a few years. Although procurement of rolling stock was delayed by about one year, the locomotives arrived on schedule. But in both cases, there were problems with the equipment and further re- pairs were needed. The consultants for train operations and control were appointed only in late 1971, three years later than expected. Finally, not all workshop construction was completed. It is difficult to .attribute these delays to a single factor, but the unstable management situation between mid-1969 and 1970 and the replacement of most of the top management team played an important role. - 11 - Contributing factors in the case of track renewal were delays in procure- ment of the rails, severe winter damage in 1969-70, and low supply of ballast and sleepers. A factor in the case of rolling stock was that the local industry could not meet the delivery schedule. The cost overrun, measured on the basis of actual versus ex- pected investments, amounted to 24%, but the actual overrun is higher, as not all the items included in the investment plan were completed. The highest cost overruns were in motive power and rolling stock and in increased stations and line capacity. Operating efficiency achievements were minor. The number of derailments, already extremely large, increased substantially, equipment utilization rates continued to be low, and average speed remained stable. The availability of motive power and rolling stock increased, but this increase can be attributed primarily to the statistical effect of incor- poration of the new equipment. While trainload increased, until 1971 this improvement can be largely attributed to the growing importance of the Atlantic Railroad (the average trainload on this line is substantially higher than the average on the whole network). In 1971-72, an important improvement in trainload occurred, which cpn be closely associated with the work of the consultants appointed in 1971 in train operations and control. The financial targets were not met, although the financial situa- tion improved slightly as compared with 1963-67. The Bank was overopti- mistic in assuming that a sudden change in the financial situation was possible. The operating ratio was expected to be below 100 by 1968, but it was over 100 throughout the 1968-73 period (except in 1969), reaching its highest value of 123 in 1972. The insufficient cash generation led to a weak cash position. Internal cash generation was expected to account for 32% of total cash requirements in 1968-72, but it actually accounted for only 6%. The main factors that prevented CNR from showing a better finan- cial performance were the increasing road competition in the light of a lack of major improvements in the quality of railroad services, and the evolution of foreign trade in relation to the main products transported by the railroads. The economic rate of return at audit is estimated to fall between 7% and 10% (depending on the figures used for operating costs), as against 15% at the time of appraisal. The audit estimate is close to the 7% rate of return for the rehabilitation program in the 1960-69 period, which was estimated in a previous Operations Evaluation Department study of the - 111 - Colombian transport sector. The lower than expected rate of return is ex- plained by the lack of major improvements in operating efficiency, which has prevented the potential benefits of the investment from fully material- izing, and by the rather stagnant traffic level. The Bank made useful contributions in the project preparation process but more emphasis should have been given to manpower problems, im- provement of operating efficiency, and development of an aggressive Com- mercial Department in CNR. Also the Bank was right in stressing the need for consultants to improve CNR's technical level. The consultants finally en- gaged can be credited with important improvements in track and equipment maintenarce, although their final impact on CNR's operations has been reduced due to the difficulties in retaining in the organization, or even in the same assignment, the staff who had worked with the consultants as counter- parts. In retrospect, it seems that closer supervision was necessary by the Bank, as is now being done in the Sixth Railroad Project (Loan 926-CO for US$25 million in 1973). CNR's lack of experienced staff and changes in senior management in mid-1969 made close supervision necessary. Another negative factor in relation to supervision missions was the frequent changes in the staff making supervision visits and the lack of emphasis on operational matters. For example, the first three supervisions were carried out by different staff and not until the third one, in 1970, did the Bank express concern about the high number of derailments. Only in 1971, after derail- ments had increased by 145% over 1967, did the Bank become seriously inter- ested in the actions that CNR was taking to prevent them. In conclusion, the Fifth Railroad Project fell below expectations. This project points out once again the importance of improved operating efficiency as a means of ensuring that the benefits of an investment plan can materialize. Only through improvements in operating efficiency is it possible to obtain savings in operating costs and better service, which, in turn, lead to better financial performance.  PROJECT PERFORMANCE AUDIT REPORT COLONBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) INTRODUCTION The loan for this project (551-CO) represented the Bank's fifth involvement with the Colombian National Railroads (CNR) and was to help finance CNR's revised investment program between 1968 and 1970. The railroad network consists of 3,436 km of single track having very narrow gauge (914 mm as against 1,435 mm in Europe and North America). At the time this loan was considered, the railroad faced serious problems in operations and track maintenance because of gradients of 3%-4% in mountainous areas and sharp curves with radii of 80 m required by the difficult Colombian topography. These problems had contributed to an alarming number of accidents (an aver- age of nine derailments/day in 1967). The Bank's association with CNR started in the early 1950s. At that time, the railroad system was composed of two separate major networks: the western system linking Puerto Berrio and Buenaventura (an important port on the Pacific Ocean) and the eastern system radiating from Bogota to Puerto Salgar, Barbosa, Belencito and Neiva. The first three Bank 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 and equipment of a con- nection between Puerto Salgar and the Atlantic port of Santa Marta, passing through Puerto Berrio. This line, known as the Atlantic Railroad, not only connected the eastern and western systems (by means of the Puerto Salgar- Puerto Berrio link), but also joined the Atlantic and Pacific coasts to each other and to Bogota. The Atlantic Railroad supports the highest traffic den- sity of the network. The average daily tonnage is roughly three times that of the whole network and consists mainly of bulk commodities being moved to and from the ports. A Ten-Year Investment Plan (1963-72) for the railroads was developed as a result of a 1960 transport survey, prepared at the Bank's initiative. The fourth Bank loan (343-CO for US$30 million in 1963) helped finance the 1963-65 program of the Ten-Year Investment Plan. CNR's performance under these four Bank loans has been poor. Construction of the Atlantic Railroad was completed under the first three loans, but with substantial delays and large cost overruns that seriously affected the profitability of the investments. Achievements under the fourth loan also did not meet expectations. The investment program was not completely carried out because the Government did not provide adequate financial support - 2 - and CNR was not able to generate internally the necessary funds. Over the period of this loan, lack of internal funds' generation was the consequence of lower than anticipated growth in freight traffic and CNR's inability to raise tariffs because of increasing road competition, while, at the same time, operating costs kept pace with inflation. THE BANK AND THE LOAN CNR first approached the Bank in relation to this Fifth Railroad Project in late 1965. The project was a logical extension of the Fourth Railroad Project and would help finance the 1968-70 program of the ongoing 1963-72 Investment Plan. Consideration of the application for the Fifth Railroad Project was delayed, and the loan documents were not signed until July 1968. The Bank first delayed consideration in 1966 because of CNR's poor performance under the ongoing Fourth Project. The latest information for 1965 indicated that freight traffic had remained fairly stable at the 1962 level and the financial situation had not improved since 1962. Although the appraisal of the Fourth Project had projected the operating ratio in 1965 as 92, it actually was 129, or higher than the operating ratio of 121 in 1962. Track rehabilitation was well behind schedule, uneconomic lines were not being closed, workshop rehabilitation was being delayed, and avail- ability of motive power and rolling stock was low. Also, the Government was behind schedule in providing funds for the investment plan -- only 67% of the expected funds for the 1963-65 period had been provided. In early 1967, the Bank was willing to reconsider its position with regard to the application for the Fifth Project because of two impor- tant changes in 1966 in the external conditions under which CNR was operat- ing. First, the new Government that took office in mid-1966 had a more favorable attitude toward the railroads. It was seriously interested in transforming the railroads into a sound commercial enterprise and therefore was willing to provide the necessary funds for the investment program. Second, foreign trade restrictions had been eased and, as a result, overall traffic from the ports had increased substantially. Railroad traffic in- creased by 25%, and this increase, together with application of a tariff system developed by the consultants, SOFRERAIL, led to an operating ratio below 100 for the first time in CNR history. Although this second external change could not be considered to reflect a permanent situation, it did cause CNR's financial situation to improve, and this improvement created a more favorable atmosphere for CNR. But before the Bank committed itself to consideration of the loan application, it wanted some assurance that changes would be made within CNR - 3 - to achieve a more efficient and viable enterprise. By early 1967, the Bank informed the Government that it would consider a new loan if there were substantial improvements in the following areas: (a) Track: Detailed plans for track rehabilitation and main- tenance over the next five years should be completed before the end of 1967. (b) Locomotives: The availability of diesel locomotives should be improved to 85% from the 70%-80% availability in 1966,and experienced expatriate staff should be engaged at an early stage to ensure that this is accomplished. (c) Rolling Stock: Items to be scrapped should be identified and segregated before April 1967. Availability of rolling stock should be improved to 90% in the next 12 to 18 months. (d) Accidents: A commission should be appointed to pinpoint the major causes of accidents and to suggest remedies. (e) Workshops: Plans for complete reorganization of the work- shops should be prepared with the help of experienced consultants. (f) Mechanical Department: The Bank should be informed about the plans for strengthening this department, which was responsible for main- tenance of locomotives and rolling stock, because the continuous use of consultants was not in itself the answer to achieving efficiency in the long run. (g) Relations with the Government: Assurances should be given by the Government that CNR will receive timely financial support to carry on its investment program and will be permitted to adjust its tariffs. In addition, an assurance should be obtained from the Government that it will not unreasonably withhold agreement to the closing of uneconomical lines. CNR's reaction to this request was favorable. It indicated willingness to appoint experienced consultants to assist in the plans for track renewal, to improve the availability of rolling stock, and to reor- ganize workshops. A commission was appointed to study the causes of accidents, and the Government made available all funds promised for 1966. At the same time, CNR asked the Bank whether it would consider the loan application. The new situation was examined at the Bank. The Bank felt that CNR and Government were taking steps to resolve the problems it had identified, and the reasons for not considering the loan application had been largely remedied. The Bank also took into consideration the likelihood that it might lose influence to press for future improvements unless it acted on the application. CNR was seeking other sources of finance for its invest- ment plan, and suppliers credits had been obtained from Mexico and Spain for purchases of rolling stock and motive power. These factors led the Bank to inform the Government in April 1967 that it would consider appraising the loan after: (a) completion of detailed traffic and financial forecasts, which would have to show continuous and significant improvement in the railroad's financial position starting from 1966; (b) appointment of SOFRERAIL as consultants on track problems; (c) agreement by the Government to provide adequate and timely financial support throughout 1967; and (d) assurances from the Government that foreign exchange would be provided for CNR's essential needs, track renewal, and equipment. The preceding conditions were met by late September 1967, and an appraisal mission was mounted in December 1967. But another factor further delayed preparation of the appraisal report. The Government was negotiating with Poland for track rehabilita- tion material in exchange for surplus coffee. Bank missions were sent in January and March 1968 to assess the effects of the outcome of these nego- tiations on the project to be financed. By March, negotiations for pro- curement of track materials were proceeding slowly. To prevent further delays in preparation of the appraisal report, the Bank proposed to include US$6.5 million in the loan for purchases of track rehabilitation material, on the understanding that this amount would be cancelled when procurement was completed. Negotiations were held in June 1968. The main modifications in- troduced into CNR's investment program by the Bank were: (a) deletion of purchases of 1,000 additional freight cars on the ground that these cars should not be needed as current usage of that equipment was low (about 38 km/day/car) and with only minor improvements in operating performance they would not be needed; (b) deletion of purchases of new passenger cars because rehabili- tation of the existing cars was more economical; -5- (c) -reduction in track maintenance purchases from US$4 million to US$3 million to allow for gradual transition to mechanization; and (d) increase in the provision for consulting services. These modifications led to a revised investment program for the 1968-72 period. Total expenditure was estimated at US$81 million (Col$1.3 billion) for the following major components: (a) Ways and Structure Renewals and Improvements (46% of total expenditure). Extensive track rehabilitation and welding of rails; purchase of 110 ballast hopper cars and some mechanical maintenance equipment; ex- tensive ballasting on lines being rehabilitated and on the Atlantic Railroad; and strengthening of a few bridges. (b) Construction and Equipment of Workshops (10% of total expen- diture). New workshops at Facatativa and Cali; machinery and tools; and new facilities to complete dieselization. (c) Motive Power and Rolling Stock (32% of total expenditure). Purchases of: 60 new diesel locomotives, 1,200 new freight cars, and 14 additional motors for railcars; rehabilitation of 1,050 existing freight cars; and modernization-of about 250 passenger cars. (d) Increased Station and Line Capacity (5% of total expenditure). Extension of marshalling yards and crossing stations; purchases of freight handling equipment; remodelling of warehouses; and improvement of inadequate station buildings. (e) Miscellaneous (7% of total expenditure). Completion of communications network; construction of staff housing; and provision of local and overseas staff training as well as extensive consulting services. In July 1968, the Bank made a US$18.3 million loan to help finance the US$38.7 million (Col$630.8 million) foreign exchange cost of the US$61.9 million (Col$l billion) expenditure for the 1968-70 program of the 1968-72 railway investment plan. The balance of the foreign exchange cost was provided through long-term suppliers credits for purchases of 60 diesel locomotives and 1,200 freight cars, which had already been obtained. The main investment items envisaged for the 1968-70 period are shown in Annex 1. Special covenants were included in the lending documents. Some related to employment of consultants, and details concerning their services are described in Annex 2. Others related to financial targets and to pro- vision of foreign exchange for spare parts, Government financing, and suit- able yard and access facilities for CNR at the port of Buenaventura. -6- PROJECT IMPLEMENTATION Implementation proceeded slowly, with delays of one to two years in completion of some of the main components: rehabilitation of track, purchases of rolling stock, and provision of consulting services. The original closing date was postponed by one and a half years, and the loan was finally closed on December 31, 1972. The track rehabilitation program provided for work on 811 km during the 1968-72 period. A bartellagreement with Poland for purchases of rails was completed in mid-1968.- However, by the end of 1972, only 641 km (79% of the planned figure) was completed and the expected number of kilometers to be rehabilitated in each year was achieved only in 1971. In all other years, the actual number of kilometers rehabilitated was far below expectations, as shown in the following table: Actual as Proportion of Year Expected Actual Expected -------------km------ 1968 54 22 41 1969 163 91 56 1970 211 155 73 1971 209 269 126 1972 174 104 60 Total. 811 641 79 It is difficult to attribute the delays in track rehabilitation to any singLe reason, except in 1968-69 when delivery of the rails was delayed until mid-1969, about one year later than expected. But even thereafter the program was not able to compensate for previous delays or to match ex- pectations with regard to the number of kilometers to be rehabilitated each year. Further delays occurred as a result of severe winter damage in 1969-70. Tn addition, the low supply of ballast and sleepers, the produc- tion of which did not change much from the 1968 level and fell short of esti- mates over the 1968-73 period, was another limiting factor. The low supply of ballast can bc attributed both to CNR's organizational -roklems and to difficulties with the operation of crusher plant:s purchased under the Fourth Rail-road Project. However, the low supply of sleepers can be full- attributed to CNR because the price that CNR had agreed to pay for the sleepers was well below the market price. Delays in payments and lack of aggressive procure- ment practices by CNR's Purchasing Department contributed to the problem. It seems that the problem of low supply of ballast and sleepers would have delayed implementation of the program even if the rails had 1/ Successful completion of this agreement led to cancellation of US$5.9 million from Loan 551-CO in 1970. -7- arrived on time. The Bank was right in insisting as early as the first supervision mission in late 1968 that the necessary steps should be taken to ensure an adequate supply of these items. But it seems that CNR's management underestimated the magnitude of the shortage. Another problem related to track rehabilitation was the quality of material used. The rails showed excessive wear after a few years, and had to be removed from curves and installed on straight areas. The reasons for such excessive wear are not yet clear because an extensive investigation by SOFRERAIL proved that the technical characteristics of this material were within the range set at the time of purchase. Procurement of rolling stock was delayed by about one year, and problems also occurred with regard to the quality of some materials used. Rolling stock was manufactured by a joint venture of a Mexican firm and a Colombian consortium -- frames and accessories were produced in Mexico and then assembled in Colombia. The first delay arose when the shipment of frames and accessories from Mexico was considerably damaged due to insuffi- cient packing and poor handling. Another delay happened when the Colombian consortium failed to maintain delivery schedules. The final delay occurred as a result of frequently broken wheels after the cars were put into service. When that occurred, it was decided to replace the locally-produced wheels with imported wheels. Locomotives were delivered on schedule, but when they were put into service they overheated on some steep grades. After an investigation, the injection system was modified. Locomotives now are reported to be operating normally. In regard to the consulting services, teams were working on track renewal and workshop organization by late 1968, and they had a positive impact on CNR. The teams introduced the preventive maintenance schedule for motive power and rolling stock as well as the program for periodic track maintenance, which are the basis of CNR's current maintenance operations. In addition, the consultants prepared a valuable detailed instruction manual for track maintenance, which is currently being used by CNR. But while the consultants had a positive effect on track and workshops, they had a much smaller impact on train operations and control because of the substantial delay in their appointment. The teams for train operations and control were rtot engaged until 1971, about three years later than expected, despite continuous reminders by the Bank. The delay in appointing the latter team seems to have been caused by some internal resistance within CNR and by CNR's unstable management situation from mid-1969 to 1970. This situation had resulted because of the change in CNR's top management after a 1969 amendment of the National Constitution, providing that CNR's General Management should be appointed and removed by the President of Colombia instead of by -8- CNR's Board of Directors. The consultants for train operations and control were finally appointed only because the Bank had included their appointment as a prerequisite for consideration of CNR's application for the Sixth Railroad Project (Loan 926-CO for US$25 million in 1973). In addition to the previously described delays in implementation, workshop construction was behind expectations. A new workshop at Facatativa for repair of locomotives was completed in 1970, but a workshop for repair of rolling stock was not built. Important delays also occurred with regard to other loan covenants. CNR was not given yard and access facilities to the port of Buenaventura until late 1971, two years behind schedule. In addition, political consid- erations delayed closure of uneconomic lines although the consultants' recommendations had been completed by late 1969. PROJECT COST CNR's total investment during the 1968-72 period was Col$1,634 million as against the appraisal estimate of Col$1,320 million (including contingencies) (Annex 3). Comparison of tHose figures indicates a cost overrun of 24%, but the actual overrun is higher because in some cases, such as the track renewal program and workshop construction, not all works in- cluded in the project at appraisal were carried out. Actual investments in motive power and rolling stock as well as in increased station capacity are substantially higher than estimated at appraisal. In the case of motive power and rolling stock, a large difference occurred between actual and expected investments in the 60 locomotives and 1,200 freight cars that were purchased with suppliers credits. The only explanations obtained for this difference were some technical changes in the locomotives and variations in the exchange rate. In the case of the increased station capacity, the cost overrun was largely due to extensive work in the marshalling yard and cargo facility at km 5, which was not in- cluded in the original investment program. OPERATING PERFORMANCE OF THE COLOMBIAN NATIONAL RAILROADS Derailments The high number of derailments by 1967 (about 3,149 or 340/million train-km) was regarded in the appraisal of the Fifth Railroad Project as an important factor preventing CNR from improving its operations. -9- The Bank expected that the causes of such accidents would be care- fully studied. Implementation of the recommendations of such studies, together with the track rehabilitation program and appointment of consult- ants in train operations and control, was expected to lead to a substantial reduction in the number of derailments. But this objective was not achieved, and the situation became even worse, with derailments per million train-km increasing from 340 in 1967 to 783 in 1973, with a peak of 844 in 1970 (Annexes 4 and 5), as can be observed in the following figures: 1967 = 100 1968 117 1969 145 1970 248 1971 230 1972 205 1973 231 The frequency of derailments in Colombia is much higher than in other Latin American countries for which information is available. In 1970, the number of derailments per million train-km was 844 in Colombia, 379 inl/ Argentina, 219 in Paraguay, 133 in Bolivia, 69 in Mexico, and 18 in Chile.- The unfavorable evolution of the number of derailments is mainly explained by; a) lack of adequate planning and control of train operations; b) failure of management to strictly enforce train operation regulations; and c) delays in the progress of track rehabilitation while, at the same time, a substantial number of diesel locomotives were put into service. Lack of in-depth studies of the causes of the accidents 1 have prevented the most important causes from being identified. But it seems that even with the prevailing poor condition of the track, the number of accidents could have been significantly reduced through the development of train operation regulations specifically adapted to track conditions and thr5ugh strict enforcement of the regulations. 1/ It should he noted, however, that the gauge in Colombia is in general narrower than in other Latin American countries and the rolling stock has.not been fully adapted to the narrow gauge. Therefore, Colombia is bound to have somewhat more derailments than the other countries mentioned. 2/ SOFRERAIL made a special study only in 1973. -10 - The slowness in appointment of consultants for track operations and control seriously delayed the development of operating regulations. There was a general lack of studies to determine the train composition and speed adapted to track conditions. Even when there were safety regulations (for example, speed limits), enforcement was not strict. As late as 1971, speed recorders were installed in some locomotives, but this action was not completely effective because the recorders were often found to be damaged. Another factor that contributed to the high number of derailments was the lack of incentives for the personnel to avoid derailments. First, the responsibility of the personnel involved in accidents was seldom deter- mined because of the lack of in-depth studies of the causes of accidents. Second, derailments constituted "good business" for operating personnel by making it possible for them to work overtime. The high accident rate has had a negative effect both on operating efficiency and on traffic levels because it has resulted in unreliable travel time. A reliable estimate of travel time often is more important than speed in choosing a particular transport mode. And a major obstacle that faces the railroads is that, in general, travel time for rail transport cannot be estimated as accurately as for road transport. The importance of this factor has been recognized by some railroad administrations, such as that in Spain, where the railroads guarantee a certain travel time under specific contracts with its main customers. In the pase of CNR, a precise estimate of the unrealiability of travel time was not possible within the limits of this audit. But an idea of the magnitude of this problem can be obtained from the following table which sh s the most likely range of travel time for some routes of the network.-' Route Most Likely Range of Travel Time ------------------------- days-------------- Santa Marta-Bucaramanga 11.1 to 4.5 Santa Marta-Medellin 9.6 to 3.0 Santa Marta-Bogota 13.2 to 2.6 Santa Marta-Neiva 17.6 to 6.2 Santa Marta-Belencito 16.0 to 4.7 Capulco-Bucaramanga 13.2 to 3.8 Capulco-Medellin 9.2 to-1.8 Capulco-Bogota 12.6 to 4.4 1/ This information has been elaborated on the basis of data on the mean and variance of travel time in a sample of shipments included in: The Netherland Institute of Economics, Analysis of Transport Investment Alternatives in the Magdalena Corridor. - 11 - Route Most Likely Range of Travel Time -------------------------days -------------- Capulco-Neiva 12.8 to 5.4 Capulco-Belencito 13.1 to 5.3 Medellin-Bucaramanga 10.5 to 3.1 Medellin-Bogota 10.4 to 3.9 Other important effects of the derailments are the higher operating costs that they imply. Despite the extent of the derailments, CNR still has not made a detailed estimate of the costs associated with these accidents. The only information available was derived by the Bank during the appraisal of the Sixth Railroad Project in 1971. A rough estimate of only the direct average cost per derailment was estimated at US$150, considering the costs of track repairs, hours lost to train crews, and damage to locomotives and rolling stock. Based on that estimate, derailments in 1970 should have accounted for 6% of the total working expenses for that year. But this is an under- estimation of their incidence as there are other important indirect effects to be taken into account, such as: a) disorganization of train operations; b) postponement of regular track maintenance because workmen had to repair track damaged by derailments; c) lower availability of rolling stock and locomotives, which in turn may have led to overinvestment in equipment to maintain existing traffic levels; and d) damage to cargo. Availability of Motive Power and Rolling Stock The availability of motive power and rolling stock in 1968 was relatively low -- 79% and 75%, respectively. Preventive maintenance was almost nonexistent and experienced workshop staff was in short supply. At appraisal, availability had been expected to increase to 85%-90% for loco- motives and 90%-95% for freight cars. But specific target dates had not been established. The availability of motive power and rolling stock has increased since 1968. In 1973, the availability of diesel locomotives was 87% (Annex 5), within the range expected at appraisal. But the availability of rolling stock - 12 - was only 80%, below the range expected. An important consideration had to be made, however, in analyzing the actual improvements in these indicators: a substantial number of new diesel locomotives (60) and freight cars (1,200) were brought into service in the 1968-73 period. In the case of locomotives, incorporation of these new units resulted in a 71% increase in the diesel motive power fleet of 1968, and this increase explains most of the improve- ment in the availability index. In the case of freight cars, the analysis is more difficult, but, in any case, the actual improvement is also below the observed one. The main reason for this rather low performance is the abnormally high number of derailments, which imposed an excessive burden on maintenance operations. In 1968, derailments resulted in damage to 7,000 cars and in 1973, to 10,000 cars, with the peak damage during the 1968-73 period being 11,569 cars in 1970 (Annex 5). The.total rolling stock fleet was about 6,400 cars, so each car was, on the average, damaged twice in one year. At the same time, the capacity of workshops to absorb this sub- stantial amount of repairs was seriously hampered by a shortage of experienced staff. Personnel shortages continued to be a problem over the 1968-73 period, despite training efforts by the consultants, because of the high turnover of personnel. This high turnover can be explained by CNR's salary levels and structure. Experienced senior staff continue to leave the railroads for better paid positions in the private sector (most of the mechanical engineers trained in France have left the railroads). 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. Excessive work resulting from the derailments and shortage of experienced staff seriously diminished the positive effect of the major reorganization of the work proposed by the consultants. Other Efficiency Indicators The appointment of consultants in train operations and control was included in the project for the purpose of obtaining the best opera- tional use of motive power and rolling stock. But no specific targets were set at appraisal in relation to performance indicators such as: speed, car turnaround time, carload, trainload and train length. Reference was made, however, to equipment usage, which was regarded as rather low (about 100,000 km/year/locomotive and 15,000 kmreight car/year). But equipment usage continued to be low over the 1968-73 period, as illustrated in the following table: - 13 - Performance Indicator 1968 1969 1970 1971 1972 1973 Locomotive km/year (thousand) 123.8 79.3 84.4 81.0 70.5 78.7 Freight car km/year (thousand) 16.9 17.1 15.5 13.6 16.8 14.9 Net ton-km/train 144.0 154.0 181.0 211.0 293.0 301.0 Freight cars/train Loaded 6.4 6.8 7.2 8.4 10.8 10.8 Empty 2.8 2.9 3.6 4.3 5.0 5.1 Freight car turnaround time (days) 10.3 12.0 14.2 15.8 16.7 16.2 Average speed (km/h) 22.9 22.2 20.7 20.3 19.7 19.4 Average load/freight car (ton) 22.5 22.9 24.9 26.3 28.3 28.8 The evolution of these performance indicators shows an uneven development. Although carload and trainload improved, car turnaround time increased substantially and train sp.eed declined gradually although dieseli- zation was completed in 1970. In the explanation of this uneven evolution, it is necessary to distinguish between two periods -- 1968-71 and 1971-72. In the 1968-71 period, the increase in car turnaround time and trainload is largely a statistical effect because of the growing importance of the Atlantic Railroad. The rela- tive importance of the Atlantic Railroad increased by 34% over this period and has remained fairly stable thereafter. Average trainload and average length of haul on the Atlantic Railroad are substantially larger than on other sections of the network. Therefore, the increase in the relative importance of the Atlantic Railroad has produced an increase in the average car turnaround time and trainload for the whole network between 1968 and 1971. In the 1971-72 period, a much larger increase occurred in trainload and train length -- about 50% and 40%, respectively -- that can be closely associated with the work of the consultants in train operations and control, appointed in 1971, The decline in usage of motive power is the consequence of a 71% increase in the diesel locomotive fleet in 1969-70, while traffic remained fairly stable. Similar factors explain the decline since 1968 in the average number of km/year/freight car. Only in early 1974 were some actions taken toward better usage of rolling stock. By 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. Preliminary indications show a favorable evolution in car turnaround time, but considerable effort still has to be devoted to faster loading and un- loading in the stations if this indicator is to improve significantly. Finally, train speed did not increase, despite the substantial track rehabilitation effort, and this suggeststhe need for detailed studies in train scheduling and control. - 14 - FINANCIAL PERFORMANCE OF THE COLOMBIAN NATIONAL RAILROADS When the Fifth Railroad Project was appraised in 1968, the Bank expected a substantial improvement in CNR's financial situation as a con- sequence of economies to be achieved through the impact of the investment plan as well as the increases in traffic and tariffs that were to follow improvement in the service. A small operating surplus, after depreciation, of about Col$8 million was expected in 1968 (the net loss in 1967 had been Col$47 million), increasing to about Col$124 million in 1973. The operating ratio was expected to remain below 100 over the 1968-73 period, improving from 98 in 1968 (it had been 115 in 1967) to 77 in 1973. By the end of 1973, these objectives had not been met (Annex 6). CNR had a net operating loss throughout the 1968-73 period (except in 1969) and, consequently, the operating ratio was above 100, reaching its highest value of 123 in 1972.1/ The following table shows the expected and actual evolution of the main financial indicators: Financial Indicator 1968 1969 1970 1971 1972 1973 ----------------- Col$ million -------------- Net Operating Surplus (Loss) (After depreciation and before interest) Expected 8 22 55 80 97 124 Actual (22) 2.7 (66.3) (104.4) (116.9) (65.3) Operating Ratio Expected 98 95 88 83 81 77 Actual 106 99 114 122 123 110 The financial performance of CNR cannot be regarded as completely unsatisfactory, however, if the 1968-73 period is compared with the 1963-67 period. But, in retrospect, the Bank was overoptimistic in the financial forecast for the Fifth Railroad Project, when it assumed that a sudden improvement in CNR's financial situation was possible in such a short time. The Bank recognized that these objectives were ambitious and subject to the vicissitudes of the Colombian economy, but it considered that they could be realized if CNR maintained the momentum for change and improvement evident at that time. The estimated and actual financial evolution can be summarized as follows: 1/ Despite these problems, CNR is still one of the best railroads in Latin America from the viewpoint of its financial performance. - 15 - Actual Forecast Actual Financial Indicator 1963-67 1968-73 1968-73 Average Operating Ratio 116 87 112 Average Working Ratio 104 72 98 Average Net Operating Surplus (Loss) (1968 Col$ million) (67) 61 (54) Change in Average Revenue/ /a /a ton-km (1968 Col$) -22% +27%- -21% Ja Refers to the 1967-73 period. The Bank projected working expenses to remain roughly constant during the 1968-73 period, while it projected average revenue/ton-km to increase by 27% over the 1967 level, assuming also that any increase in working expenses (for example, wages) would be offset by further tariff increases. In terms of CNR's past performance, this was clearly unrealistic because in the 1963-67 period average revenue/ton-km in constant prices declined by 22%. A much more realistic approach was followed in the appraisal of the Sixth Railroad Project in 1973, when a gradual improvement in the working ratio was forecast. The assumed increases in working expenses and revenue/ ton-km are expected to lead to the first operating ratio below 100 in 1977. This overestimation of CNR's capability to generate a substantial operating surplus was also reflected in the Financial Plan. During appraisal, the Bank expected that CNR's internal cash generation over the 1968-72 period would be Col$822 million (in constant 1968 prices), accounting for about 32% of the cash requirements during that period. The Government was expected to contribute 42% of the total requirement and the balance was expected to be borrowed. But the actual internal cash generation in the 1968-72 period was only Col$178 million (in current Colombian pesos of each year), repre- senting only 6% of the actual requirements. Consequently, CNR had to rely on extensive borrowing from local banks, which finally accounted for 47% of the total required as against 26% expected at appraisal. 0 This insufficient cash generation also led to a weak cash position over the 1968-72 period. The liquid ratio has always been below expected values and, moreover,has been below 1 since 1970 (Annex 7). Several factors prevented CNR from achieving better financial performance: slow traffic growth, inability to raise tariffs sufficiently, and increased working expenditures. Freight traffic was fairly stable over the 1968-71 period, with an increase of 11% in 1972-73 (Annex 5). This outcome reflects the evolution - 16 - of the main export products that are transported by the railroads. While traditional railroad traffic remained stable, new traffic could not be attracted because of the low-quality service offered and the lack of aggres- siveness of CNR's Commercial Department. Moreover, CNR had difficulty in retaining the transport of products traditionally carried by rail because of increasing road competition and extensive track damage caused by the severe winter in 1969-70. For example, after completion of the Cali- Buenaventura road, the railroads' share of traffic from the port of Buena- ventura declined sharply and the Cali-Medellin line, which was put out of operation in the winter of 1969-70, still is out of operation. Another contributing factor was CNR's inability to raise tariffs sufficiently. Although tariff changes during the 1967-73 period led to increases of 62% in average revenue/ton-km, they were not enough to offset inflationary pressures, and average revenue/ton-km declined in real terms by 21% in the 1967-73 period. Given the competitive situation prevailing in Colombia's transport sector, further tariff increases would have required substantive improvements in the quality of service, which in turn was seriously hampered by delays in operating improvements. Working expenses increased by 80% in the 1968-73 period (Annex 6), reflecting not only the effects of inflation and higher wage rates, but also the expenditures necessary to repair track damaged during the winter of 1969-70, increased allocations for track maintenance after implementation of the track maintenance program devised by SOFRERAIL, and, finally, higher pension payments that have to be paid to CNR staff retroactively to 1965. The combined effect of these factors more than offset any economies from the investment plan in 1970-71, when, even with a rather constant traffic level, working expenditures increased in real terms by 10% in relation to 1968-69. Another factor contributing to the increase was the delay in appointment of consultants in train operations and control, which prevented CNR from obtaining important improvements in operating efficiency until 1972-73. ECONOMIC JUSTIFICATION OF THE PROJECT The economic rate of return over the 1968-72 investment program was estimated to be 15% at the time of appraisal, while it was estimated to be between 7% and 10% at the time of audit. This range results from alternative hypotheses with regard to tne evolution of operating costs as traffic grows. The audit estimate of the rate of return is close to the results of the previous Operations Evaluation Department study of the Colombia transport sector. In that study, a rate of return of 7% was obtained for the rehabilitation program in the 1960-69 period. - 17 - The audit rate of return is based on the evolution of CNR's operating expenses and actual investments during the 1968-72 period. The evolution of CNR's operating expenses is shown in the following table: Year Operating Expenses Traffic Units (millions of 1968 Col$) (millions) 1968 344.8 1,301.5 1969 336.6 1,295.0 1970 368.8 1,297.5 1971 370.9 1,290.5 1972 355.1 1,397.0 1973 332.0 1,544.0 The "without" situation is defined on the basis of the operating costs for 1968, assuming that they would grow in real terms as indicated in the appraisal report if investments were not made. The "with" situation is defined on the basis of the actual operating costs in each year. The bene- fits are then the difference between total operating costs in the "without" and "with" situation. Benefits after 1973 were projected using the same assumptions as in the appraisal. The 1970-71 period was excluded from the analysis because the increase in operating costs can be attributed somewhat to exogenous factors: repair of the extensive 1969-70 winter damage, retroactive payments of higher pensions to CNR's retired staff, and an accounting change by which some expenses relating to the Atlantic Railroad, previously charged to in- vestment, were charged to working expenses. As the exact influence of these factors cannot be isolated, 1970-71 was excluded from the analysis, the implicit assumption being that operating costs would have remained roughly at the 1968-69 level.1/ The main reasons for the lower rate of return at audit than at appraisal are the lower than expected reduction in operating costs and the rather stagnant traffic level. The traffic level issue has already been analyzed, and the operating cost issue is analyzed in the following paragraph. The lower than expected reduction in operating costs is the con- sequence of less than full usage of the new infrastructure and equipment, which in turn had a lot to do with delays in the appointment of the con- sultants in train operations and control as well as the generally unstable management situation which has already been described. The slowness of the improvement in CNR's operating efficiency has led to underusage of the 1/ This assumption does not affect the rate of return in a significant way. Interpolating the benefits for 1970-71 from the ones in 1969 and 1971, the rate of return is increased by less than .5 percentage points. - 18 - investment made in the 1968-73 period. Average train speed has remained fairly stable, the number of accidents has not been reduced despite the extensive track rehabilitation program, and improvements in trainload have not been noticeable until after 1971. Thus not all of the potential benefits have been obtained, and this explains the low actual rate of return for this investment program as compared with the appraisal estimate. THE ROLE OF THE BANK The Bank played an important role in preparation of the Colombia Fifth Railway Project, first through delaying consideration of the loan application and then through introducing changes in CNR's original proposals by suggesting more intensive use of consulting services and by deletion of some of the investments. Consideration of the loan application was delayed for about one year, and that delay seems justified, in retrospect, because it led CNR's management to try seriously to improve efficiency standards at a time when it was more concerned with increasing CNR's carrying capacity through new investments in motive power and rolling stock than through more efficient usage of existing equipment. Establishment of conditions to be fulfilled by CNR before consideration of the loan application was, therefore, appro- priate and had a positive effect. In retrospect, the Bank seems to have been right in stressing the importance of overcoming the lack of experience of CNR's staff and including consulting services in the project. Experienced consultants were to set up new organization schemes (workshops, track renewal, track maintenance, and train operations), and initially were to operate them. The consultants can be credited with important improvements in CNR's operations, particularly in the areas of equipment and track maintenance. However, it was not possible in this audit to assess whether the quality and quantity of the consultant services included in the project were adequate. It seems though that CNR could have profited more from the consultants' work. For example, the consultants were expected to provide "on-the-job" training for CNR staff. But for this method to be successful, the trained personnel had to stay in their jobs long enough for CNR to be able to benefit from the experience gained by them. The Bank overestimated CNR's ability to retain the trained personnel, and some left while others, such as mechanics, changed jobs within CNR. The result was that CNR could not fully use its trained personnel. The main reasons for the departures or transfers were CNR's salary levels and structure. It seems that this problem could have been detected by the -19 - Bank at appraisal and therefore a more comprehensive approach could have been pursued to remedy the lack of experienced CNR staff. Deletion of purchases of additional rolling stock was another worthwhile Bank initiative. Usage of existing rolling stock was low at the time of appraisal, and only moderate improvements in operating efficiency would make investments in freight cars unnecessary. Therefore, purchases of these items were deleted from the project. This action also had the positive effect of creating a need for improved efficiency in CNR. The Bank could have been more influential in two other areas of project preparation. First, the Bank could have put more emphasis on improved operating performance. As shown in the economic justification of the project, CNR's rather low operating efficiency has prevented most of the benefits from the investments from materializing. A plan of action, similar to that of other railroad projects at the time (for example, Spain) could have been useful. This plan could have included specific targets in relation to different operating performance indicators like speedcar turn- around time, carload, trainload and train length. It also could have had positive effects on the corresponding departments of CNR by pressing them to achieve certain targets at fixed dates. Second, it is surprising that the Bank did not consider development of a more aggressive Commercial Department within CNR to be of high priority. Such an aggressive department might have been an important factor in pressing for improvements in operating efficiency. In regard to project implementation, the Bank has to be credited with the appointment of the consultants for train operations and control. But closer supervision by the Bank and more attention to the problems of operating efficiency and derailments seems to have been necessary. The consultants for train operations and control were not appointed until late 1971, three years behind schedule. This delay is completely attrib- utable to CNR. From 1968, the Bank reminded CNR's management about the importance of this appointment, and, after the reminders failed, the Bank made the appointment a prerequisite for future lending to CNR. Only in that way were the consultants finally appointed. Bank supervision missions averaged one every 10 months between signature of the loan and the closing date. In retrospect, it seems that closer supervision was necessary because of the lack of experience of CNR's staff regarding the important changes in operations that the project attempted to induce. This situation was aggravated in mid-1969 in light of CNR's unstable management situation that followed the change in the method of appointing its General Management. The need for close supervision that would provide the necessary guidance and advice was recognized by the Bank as early as the first supervision mission in late 1968. It was suggested not only that - 20 - supervision missions should be sent frequently, but also that engineers en route to or returning from missions south of Colombia should make brief inspections. As a result of this experience, much closer supervision, averaging one mission every three months, is being carried out under the ongoing Sixth Railroad Project. The need for closer supervision under the Fifth Railroad Project also was pointed out by CNR's management during the visit to Colombia at the time of this performance audit. Frequent changes in supervision staff (for example, the first three supervision missions were carried out by different staff members) also was mentioned as a negative factor in the role of the Bank. Supervision missions were focussed mainly on procurement and financial matters. In retrospect, it seems that more attention should have been given to improvements in operating performance. Improved operating efficiency was necessary to enable most of the potential benefits of the investments to materialize. This improvement would lead to better service, which in turn would lead to savings in operating costs, increases in traffic, and, consequently, better financial results. The other aspect that was not given enough attention during super- vision missions was CNR's inadequate actions to reduce the number of accidents. Only in the third supervision mission (March 1970) did the Bank express some concern about the high number of derailments. And only after derailments had reached their peak in 1970, with an increase of 145% over 1967, did the Bank become seriously interested in CNR's actions to prevent derailments. The Bank knew that no major analysis of the causes of derailments had been made and that some equipment for the analysis, such as speed recorders, had been fitted only recently to part of the locomotive fleet. In retrospect, it seems that the Bank could have pressed these matters earlier. CONCLUSIONS The Fifth Railroad Project fell below expectations. Implemen- tation was delayed, the financial targets were not met,./ and the expected rate of return was not achieved -- the rate of return at audit is estimated between 7% and 10% as against 15% at appraisal. These shortcomings are mainly explained by the lack of improve- ments in operating efficiency. This project is a clear example that invest- ments alone are not enough to induce substantial improvements in operating 1/ It should be noted, however, that the financial results were not as bad as those of some other Latin American railroads. Although actual working ratios missed the appraisal targets, they were still below 100 for all the years 1968-73,except 1971 and 1972. - 21 - efficiency. Moreover, the institution must take the necessary steps to ensure that the potential benefits of such investments can materialize. In the case of CNR, several factors prevented these benefits from materializing: a shortage of adequately trained personnel, changes in top management positions, and some internal resistance to changes in operations. The Bank attempted to induce the necessary improvements in effi- ciency through the use of consultants. But this approach was only partially successful. CNR was not able to retain its trained personnel, considerable delays occurred in the appointment of the consultants for train operations and control, and frequent changes in top managerial positions slowed imple- mentation of the consultant's recommendations. In retrospect, it seems that a more comprehensive approach to CNR manpower and management problems was necessary. Also, the Bank relied completely on the use of consultants to achieve the necessary improvements in operating efficiency, and it seems that a more active role by the Bank could have accelerated adoption of the necessary steps to improve operating efficiency.  ANNEX 1 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RA AJ) PROJECT (LOAN 551-CO) Investment Program of Colombian National Railroads, 1968-70 Local Cost Foreign Cost Total Cost ----- Col$ million------ US$ million Col$ million I. Items to be Financed by Loan 551-CO 410 km of Rails and Acessories 117.6 121.3 7.5 Ll 238.9 110 Ballast Cars and Track Maintenance Equipment 2.4 48.9 3.0 51.3 Workshop Equipment and Tools 7.2 25.3 1.6 32.5 Parts for Passenger Cars and Railcars 8.0 21.7 1.3 29.7 Parts for Freight Cars 7.5 8.8 0.5 16.3 Freight Handling Equipment 3.8 8.6 0.5 12.4 Telecommunications and Signalling Equipment 11.4 6.2 0.4 17.6 Other Railway Material 2.8 8.5 0.5 11.3 Consulting Services and Training Abroad (Technical Cooperation) 22.4 27.5 1.7 49.9 183.1 276.8 17.0 459.9 Contingencies (7.5% for Foreign Cost only) 0 21.5 1.3 21.5 Subtotal I 183.1 298.3 18.3 481.4 II. Items to be Financed by Other Sources Workshop Buildings 60.0 0 0 60.0 60 Diesel Locomotives 0 156.5 9.6 156.5 1,200 Freight Cars 0 159.7 9.8 159.7 120 Passenger Cars (rehabilitation) 18.0 0 0 18.0 Way and Structure Renewals and Improvements 80.7 0 0 80.7 Increase in Station and Line Capacity 33.5 0 0 33.5 Miscellaneous 3.4 0 0 3.4 195.6 316.2 19.4 511.8 Contingencies 0 16.3 1.0 16.3 Subtotal II 195.6 332.5 20.4 528.1 Total I and II 378.7 630.8 38.7 1,009.5 /1 Includes $6.6 million subject to barter agreements. Source: IBRD/IDA Report No. TO-652aAppraisal of a Fifth Railroad Project, Colombia, July 11, 1968. ANNEX 2 Page 1 PROJECT PERFORKANCE AUDIT REPORT COLOMBIA FIFTH RALRAD PROJECT (LOAN 551-CO) Consulting Services A description follows of the areas where consulting services were required during the 1968-72 Investment Program: Track maintenance and rehabilitation. The general condition of the track is very poor. Maintenance has been neglected. In many sections the for- mation and drainage is inadequate and ballast and ties are insufficient in quantity and quality. Much of the rail and many of the switches are badly worn. There is need for large scale rehabilitation and the building up of a comprehensive maintenance organization. The absence of staff experienced in sound and up-to-date permanent way practice, including mechanical track work, makes it essential to appoint consultants (a) to draw up plans and programs, (b) to see to their implementation, and (c) to train CNR staff on the job. Action in this regard has already been initi- ated by CNR but will be required to continue throughout the project and much of the program period. Organization of workshops and the maintenance and repair of locomotives and rolling stock. Reequipping the workshops and reorganizing the work to be done are matters partially carried out already by CNR but further work in this direction will be required during the project period. Reor- ganization of the work to be done in the shops is a matter of urgent necessity. The present condition of almost all rolling stock, excluding the diesel locomotives, is very poor and, except at Medellin, there is little attempt at systematic preventive and periodic maintenance. Complete reorganization of the use of machines and tools, manpower and supplies is required; also systematic training on the job of CNR staff will be very important. As an index of the need for such action, the current availa- bility of passenger cars is 57%I and of freight cars 75%, both of which should be around 90% to 95%. Diesel locomotive availability, currently 80%, is also susceptible of improvement by 5% to 10%. The proposed loan included US$3 million for track maintenance equipment; this also will require sound and s'Ystematic maintenance in the workshops if full and effective use is to be made of it. Train operation and control. Train operation involves scheduling and control. At present, CNR operates with only rudimentary schedules and very limited control. With the introduction of a new telecommunications network covering the whole system, and also the complete dieselization of locomotive power, better schedules and control should be possible. ANNEX 2 Page 2 Consultants should be appointed to ensure that CNR obtains the advantages of sound operating experiences in drawing up schedules and devising prac- tical means to secure the control and best operational use of locomotives and rolling stock. The current performance indices for diesel locomotives in use (285 km/day) and for freight cars available (38 km/day) are short, particularly the latter, of what should be achieved. A practical matter of immediate importance will be the devising of train schedules to allow the optimum track occupation for rehabilitation work at the same time as meeting public traffic requirements. The consultants will also make an assessment of future rolling stock needs. Uneconomic lines, In addition to the Manizales-Mariquita aerial ropeway, the following lines appear to be uneconomic: Palmira-Pradera (14 km); Armenia-Pereira (60 km); Pereira-Manizales (69 km); and Lenguazaque-Barbosa (116 km); and the Government and CNR agree that there should be consultant studies to determine possible future appropriate action. Management. In addition to these requirements, the services of consultants (Madigan Hyland de la Cruz and Co.) have been helpful in giving effective support to management for many years in the absence of sufficient experi- enced staff. CNR wish to continue making use of the services of these consultants during the Project period. Source: IBRD/IDA Report No. TO-652a, Appraisal of a Fifth Railroad Project, Colombia, July 11, 1968. ANNEX 3 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) Estimated and Actual Investment Program of Colombian National Railroads, 1968-72 (Col$ thousand) Actual as % Estimated Actual of Estimated Ways and Structure Renewals and Improvements 600,418 521,581 87 Workshop Construction and Equipment 122,492 59,192 48 Motive Power and Rolling Stock 408,167 826,154 202 Increase in Stations and Line Capacity 58,135 144,369 248 Miscellaneous 91,572 92,017 101 Contingencies 39,526 - - Total 1,320,310 1,634,313 124 Source: IBRD/IDA Report No. TO-652a, Appraisal of a Fifth Railroad Project, Colombia, July 11, 1968 and Colombian National Railroads. ANNEX 4 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) Derailments on Colombian National Railroads, 1967-73 flerailments Derailments per million km-trains 8,000 800 7,000- -700 i * 6,000- -600 5,000- 500 ** * * **C 4,000- -400 .. 3,000 - 300 0 2,000- 200 IIII 1967 1968 1969 1970 1971 1972 1973 Derailment per million train-km ********...* No. of Derailments World Bank-9778 ANNEX 5 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILRDAD PROJECT (LOAN 551-O) Selected Operating Statistics of Colombian National Railroads, 1968-73 1968 1969 1970 1971 1972 1973 System Total route lan 3,436 3,436 3,436 3,431 3,431 3,431 Total staff 11,549 11,387 11,526 11,654 11,516 11,408 Traffic Passengers - total (million) 3.69 2.65 2.95 3.16 4.27 4.20 - mainline - - 2.10 2.27 3.08 3.14 - season tickets - - .85 .89 1.19 1.06 Pass-km - total (million) 351 273 249 281 398 427 - mainline - - 236 267 380 410 - season tickets - - 13 14 18 17 Average journey (km) 95 103 112 118 123 131 Net paying tons (million) 3.24 3.05 2.78 2.65 * 2.73 2.76 Net paying ton-km (million) 1,126 1,159 1,173 1,150 1,198 1,331 Total gross ton-km (million) 2,160 2,197 2,190 2,159 2,119 2,315 Loaded freight car km (million) 50.01 50.58 47.14 46.28 44.18 47.43 Average haul (km) 347 380 422 434 439 482 Elmpty freight car-km (million) 21.71 22.03 23.28 23.61 20.6 22.7 Traffic Density Pass-km/route km (thousand) 102.2 79.5 68.4 77.8 110.8 119.5 Freight -- net ton-km/route km (thousand) 327.7 337.3 341.4 335.2 349.2 387.9 Operations Train-km - passengers (million) 2.09 1.99 1.92 2.16 2.40 2.57 TTain-km - freight (million) 7.80 7.48 6.47 5.45 4.09 4.41 Train-km - total (million) 9.89 9.47 8.39 7.61 6.49 6.98 Loco-km - steam (million) 3.38 1.83 0.45 0.40 0.29 0.15 Loco-lan - diesel (million) 8.42 9.52 10.47 9.72 8.32 9.06 Operating Efficiency (Freight) Gross ton-km/train-km 276.92 293.72 338.49 396.0 517.9 524.5 Net ton-kn/train-an 144.36 154.95 181.30 211.0 293.0 301.5 Net ton-km/loaded car-km 22.52 22.91 24.88 26.28 28.26 28.84 Car turnaround time (days) 10.3 12.0 14.2 15.8 16.7 T62_' Averdge speed (km/h) 22.92 22.25 20.77 20.26 19.72 19.43 Loco-km/day -. steam 63.0 36.4 5.0 1.1 1.0 0.09 Loco-km/day - diesel 265.2 181.2 196.4 221.9 156.2 142.7 Traffic units/employee (thousand) 127.9 125.8 122.2 121.5 137.9 152.6 Availability Steam locomotives (%) 45.3 25.4 7.0 7.0 18.6 17.3 Diesel locomotives (%) 79.5 83.3 85.2 85.4 86.1 86.7 Freight cars (%) 75.3 80.9 83.9 85.7 81.7 80.2 Derailments Total number/year 3,947 4,694 7,116 5,969 4,368 5,471 Total hours lost 15,083 16,024 25,759 18,263 14,372 21,589 Derailment/million train-km 399 495 844 784 698 783 Vehicles damaged 7,170 7,585 11,569 9,462 7,701 10,248 Sotrce: Colombian National Railroads. PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILOD PROJECT (LOAN 551-CO) Forecast and AQtual Revenues and Expenses of Colombian National Railroads, 1968-73 (Col$ million) 1968 1969 1970. 1971 1972 1973 Forecast Actual Forecast Actual Forecast Actual Forecast Actual Forecast Actual Forecast Actual Operating Revenues Freight 332.0 310.8 359.0 369.4 378.0 392.6 398.0 412.6 420.0 431.8 438.0 545.5 Passenger 38.0 35.1 41.0 34.0 41.0 32.4 41.0 37.4 41.0 56.8 41.0 66.9 Miscellaneous 17.0 18.7 17.0 15.9 17.0 15.8 17.0 17.6 17.0 20.1 54.0 25.9 Total Operating.Revenues 387.0 364.6 417.0 419.3 447.0 440.8 479.0 467.6 502.0 508.7 533.0 638.3 Operating Expenses Maintenance of Way 90.2 90.2 124.3 142.9 152.2 179.6 Maintenance of Equipment 56.8 62.1 80.8 95.6 98.1 100.3 Transportation 160.9 166.3 181.0 192.4 217.2 258.9 General Expenses 36.9 45.0 52.8 62.4 72.3 85.5 Total Working Expemes 333.0 344.8 332.0 363.6 326.0 438.9 328.0 493.3 330.0 539.8 332.0 624.3 Depreciation 46.0 41.8 63.0 53.0 66.0 65.2 71.0 78.7 75.0 85.8 77.0 79.3 Total Operating Expenses 379.0 386.6 395.0 416.6 392.0 504.1 399.0 572.0 405.0 625.6 409.0 703.6 Net Operating Revenue (Loss) 8.0 (22.0) 22.0 2.7 55.0 (63.3) 80.0 (104.4) 97.0 (116.9) 124.0 (65.3) Non-Operating Revenue 17.0 11.9 17.0 31.3 18.0 20.4 20.0 10.2 17.0 25.4 13.0 18.5 Net Revenue (Loss) 25.0 (10.1) 39.0 34.0 73.0 (42.9) 100.0 (94.2) 114.0 (91.5) 13.0 18.5 Interest 66.0 44.6 70.0 53.3 73.0 58.1 91.0 61.2 75.0 98.2 70.0 131.0 Net income (Deficit) (35.0) (54r.-7) (1. -) (B73) 0.0 (101.0) -9. (T374) - 9 . (189.7) 67. 0 (1-77.8) Ratios Working 86.0 94.6 80.0 86.7 73.0 99.6 68.0 105.5 66.0 106.1 62.0 97.8 Operating 98.0 106.0 95.0 99.4 88.0 114.6 83.0 122.3 81.0 123.0 77.0 110.2 Breakdown of Working Expenses Personnel 222.5 240.1 294.1 332.0 368.0 423.5 Fuel 23.2 19.9 16.1 18.9 24.8 28.4 Other Materials 61.5r 58.3 71.1 80.9 82.0 97.2 Others 37.6 45.3 57.6 64.7 65.0 71.9 Source: IBRD/IDA Report No. TO-652a, Appraisal of a Fifth Railroad Project, Colombia, July 11, 1968 and Colombia National Railroads. PROJECT PERFORMANCE AUDIT REPORT COLOMBIA FIFTH RAILROAD PROJECT (LOAN 551-CO) Forecast and Actual Balance Sheets of Colombian National Railroads, 1968-72 (Col$ million) 1968 1969 1970 1971 1972 Forecast Actual Forecast Actual Forecast Actual Forecast Actual Forecast Actual Current Assets Cash 32 20 36 26 43 23 46 32 50 40 Inventories 165 207 175 187 180 485 200 575 210 458 Other Current Assets 117 120 127 153 141 159 151 205 166 241 Total Current Assets 314 347 338 366 364 667 397 812 426 739 Fixed Assets Gross Fixed Assets 2,283 2,337 2,800 2,870 3,025 3,287 3,210 3,578 3,330 3,745 Less Accumulated Depreciation 250 269 270 315 320 378 390 453 460 534 Net Fixed Assets 2,033 2,068 2,530 2,555 2,705 2,909 2,820 3,125 2,870 3,211 Miscellaneous Assets _ __5 46 92 63 131 __ 2 170 90 215 226 Total Assets 2 409 2 461 2 960 2 984 3 200 3 628 3 387 4 027 3,511 4 176 Current Liabilities 70 95 75 151 80 190 85 304 90 364 Long Term Debt 821 890 1,171 1,165 1,194 1,500 1,162 1,673 1,063 1,961 Capital Government Contribution 1,776 1,776 1,973 1,950 2,155 2,143 2,330 2,387 2,479 2,684 Less Accumulated Deficit 408 430 439 445 439 556 430 774 391 1,016 Net Capital 1,368 1,346 1,534 1,505 1,716 1,586 1,900 1,613 2,088 1,668 Reserves 13 180 163 361 437 270 183 Total Liabilities 2,409 2,461 2,960 2,984 3,200 3,638 3,387 4,027 3,511 4,176 Ratios Current 4.5 3.7 4.6 2.4 4.6 3.5 4.6 2.7 4.7 2.0 Liquid 2.1 1.5 2.2 1.2 2.2 0.96 2.2 0.78 2.4 0.77 Debt/Equity 37/63 40/60 43/57 44/56 41/59 49/51 37/63 51/49 33/67 54/46 Source: IBRD/IDA Report No. TO-652a, Appraisal of a Fifth Railroad Project. Colombia, July 11, 1968 and Colombian National Railroads. C A R I B B E A N S E A SANTA MARTA BARRANQUILL CIENAGA lFundación 0 50 100 150 200 Km CARTAGENA0 Champan• 4 V EN E ZU EL A o ,~amarra. Capulco Puerto Wilches . BARRANCABERMEJA BCRMN -Carare PUERTO BERRIO MEDELLIN Paz del Rio Barbosa Bclencito Puerto ýn-za La Dorada Salgar TUNJA Alejandro Mariquita . 0 Lopez zipaquira fNýa.zales Betania CARTAGO PEREIRA BOGIOTA Zarzai El Salto Andalucia Buenos Aires Girardot Q Espinal BUENAVENTURA PALMIRA Yumbo Pradera CALL Timba Santander San Ignacio NE IVA NATIONAL RAIL-ROADS POPAYAN LEGEND TUMACO .i ,, .. Atlantic Line . El DiviSO ..... Other Lines Aeriail Caleway (out of service) E C U A D 0 R JANUARY 1970IB-24R