Document of The World Bank FOR OFFICIAL USE ONLY Report No. 7069 PROJECT COMPLETION REPORT BRAZIL SECOND URBAN TRANSPORT PROJECT (LOAN 1839-BR) December 24, 1987 Latin America and the Caribbean Regional Office 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. T* WORt SANK macw nCIA Y Wa#hiIon. D.C. MM1 US.A. Oft of 0"CW4~6 - - December 24, 1987 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Brazil Second Urban Transport Project (Loan 1839-BR) Attached, for information, is a copy of a rcport ercitled "Project Completion Report on Brazil Second Urban Transport Project (Loan 1839-BR)" prepared by the Latin Anberica and the Caribbean Regional Office. Further evaluation of this project by the Operations Evaluation Dtpartment has not been made. Attachment This document has a _iritd disbutiond may be usod by nclants only in th petfbnuane of their official duties Its contents may not otherwise be diclued without Wold Bk authouatlon. F OR FICIL USE ONLY PROJECT COMPLETIOW REPORT BRAZIL SECOND URBAN TRANSPORT PROJECT (LOAN 1839-AR) TABMLE OF CONTENTS Page No. PREFACE ......................... I BASIC DATA SHEET ....................., ....................... ii HIGHLIGHTS ................................................. iv I. INTRODUCTION ............. .................................... I II. PROJECT PREPARATION AND APPRAISAL ............. ............... I III. PROJECT IMPLEMENTATION AND COST .............................. 3 IV. OPERATIONS AND TRAFFIC .... ........ ........................... X V. FINANCIAL PERFORMANCE ..... .......... ............... 1n VI. INSTITUTIONAL PERFORMANCE ................................ .... 12 VII. ECONOMIC REEVALUJATION . ....................................... 15 VIII. CONCLUSIONS .................. 1 lf TABLES 1. Disbursements Schedules ........................... 18 2. Comparison Between Appraisal Estimated and Actual Project Costs .............. 1q 3. Operating Statistics ....................... ................. 2n 4. TRENSURB - Financial Results (1986) ..... .................... 21 5. TRENSURP - Balance Sheet (1986) ............................. 22 6. Economic Reevaluation - Cost and Benefit Streams ............ 23 ANNEX 1. Comments from the Borrower.........e...................... . .* 24 MAP IBRD 14688 mns document h.s 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. i1- BRAZIL SECOND URUiN TRANSPORT PROJECT (LOAh 1839-BR) PROJECT CUIPTLETIOE REPORT PREKFACE The following is a project completion report on Brazil's Second Urban Transport Project, for which a loan in the amount of USS15q.O million was approved by the Executive Directors on May 6, 1980. The loan was closed on December 31, 1986, two years later than the appraisal envisaged. Loan funds amounting to US$26.0 million were cancelled. The completion report was prepared by the Rank's Latin America and Caribbean Regional Office and is based upon information obtained from the LAC Information Center, Staff Project Appraisal Report (Report No. 2759b-BR), Bank staff supervision reports, draft Project Completion Report prepared by TRENSURB, information collected during a mission to Porto Alegre in November 1985, and a consultant's report dated April 1986. In accordance with the revised procedures for project performance audit reporting, this Project Completion Report was read by the Operations Evaluation Department (OED), but the project will not be audited by OED staff. A copy of the draft report was sent to the Borrower for comments. Their comments were received and have been incorporated into this final report and reproduced as an Annex. - il - BRAZIL SECOND URBAN TRANSPORT PROJECT (LOAN 183%-BR) PROJECT C0"PIETION REPORT &ASIC DATA SHEET Key Project Data Appraisal Expectation Actual Total Project Cost (US$ million) _/ 312.8 249.4 Loan Amount (US$ million) 159.0 133.0 Disbursed 159.0 133.0 Cancelled - 26.0 Date Physical Component Completed 3/84 3/85 Proportion of time overrun (%) - 25.0 Economic Rate of Return ( t) 19 11.5 Institutional Performance Performance of the Borrower - Fair Performance of the Executing Agency - Fair Other Project Data Original Plan Actual Appraisal - 12/79 Board Approval 5/6/80 5/6/80 Loan Agreement Date 6/2/80 6/2/80 Date of Effectiveness - 12/9/80 Closing Date 12/31/84 12/31/86 Name of the Borrower Federative Republic of Brazil Executing Agencies State of Rio Grande do Sul Empresa de Trens Urbanos de Porto Alegre (TRENSURB) Fundacao Metropolitana de Planeiamento (METROPLAN) Rede Ferrovi aria Fecieral S.A. (RFFSA) Empresa Brasileira de Planejamento de Transportes (GEIPOT) Empresa Brasileira de Transportes Urbanos (EBTU) Fiscal Year of the Borrower 1/1-12/31 Follow-on Project: None / Costs are not readily comparable since the project scope increased during implementation. ----4-- __ a~~~~~~~~~~~~~~~~ - ii i - BRAZIL SECOND URBAN TRANSPORT PROJECT (LOM 18 3-BR) PROJECT COMPLETION REPORT BASIC DATA SHEET (CONT'D) Staff nate of Date Staff Weeks Weeks Report Identification 8/77 2 2 4 9/02/77 Preparation 2/79 4 2.5 8 3/12/79 Preparation 4/79 1 2 2 5/08/79 Pre-appraisal 7/79 6 3 18 8/09/79 Preparation 8/79 1 0.5 0.5 9/14/79 Appraisal 9/79 6 3 18 12/05/79 Supervision 6/80 5 2 10 7/15/80 Supervision 10/80 2 2 4 10/23/80 Supervision 3/81 2 2 4 3/04/81 Supervision 7/81 2 2 4 R/28/81 Supervision 2/82 3 2 6 2/18/82 Supervision 10/82 3 0.5 1.5 12/20/82 Supervision 6/83 3 0.5 1.5 6/20/83 Supervision 10/83 3 0.5 1.5 11/30/83 Supervision 3/84 2 1 2 4/13/84 Supervision 10/84 3 1 3 10/25/84 Supervision 3/85 2 1 2 4/12/85 Limited Supervision 9/85 1 0.5 0.5 9/26/85 Completion 10/85 1 1 1 t.cumulated Actual and Estimated Schedule of Disbursements (US$ million) Actual as X Appraisal of Appraisal IBRD !Y Estimate Actual Estimate 1981 0.1 - - 1982 26.7 3.4 13 1983 90.1 14.1 16 1984 158.6 56.8 36 1985 159.0 122.0 77 1986 - 133.0 84 - iv - BRAZIL SECOND URBAN TRANSPORT PROJECT (LOAN 1839-BR) PRO.JECT COKPIZTION REPORT RIGHLIGRTS The Second Urban Transport Project, subject of this report, was designed to improve the distribution cf land use within the Oorto Alegre corridor with the fastest growing population, reduce the petroleum fuel consumption in the same corridor, and avoid traffic congestion both in the corridor and in the metropolitan center of the city. The objectives were to be achieved by providing an efficient conventional suburban electric railway, with associated car parks, bus feeder services and interchange facilities, while at the same time restricting car parking availability in the downtown metropolitan area (para 1.02). The major physical components of the project were successfully completed within one year of the projected completion date and at a cost 20% below appraisal estimates, considerable achievements for a project of this size and complexity (para 8.01). Pr3ject implementation was mostly problem-free, except for some disagreements which arose in the context of the application of international competitive bidding rules (taras 3.10-3.12). The pioject included technical assistance in the form of managerial support to TRENSURB, to establish the company as a functioning railway operator and to recommend complementary transport policies to he carried out by Nucleo Metropolitano de Transportes Urbanos (NMTU), and complementary land use plans to be carried out by the Fundacao Metropolitana de Planelamento (METROPLAN) and/or the municipalities in the corridor (para 3.13). TRENSURB's management consultants carried out their tasks effectively and the company has been operating the railway services for two years (para 3.14). The other technical assistance, related to studies on the reorganization of bus services, the development of commercial areas close to TRENSURB stations and of station car parking facilities were also satisfactorily carried out under the project. However, implementation of the recommendations emanating from those studies has been limited mainly because of political difficulties (paras 3.15-3.17). The most serious shortcoming of the project was the failure to undertake and implement a public transport tariff study as originally envisaged (para 3.15). This study would have revealed the financial cost of keeping TRENS'URB's tariffs at their current low levels and provided a basis for establishing a short term revenue maximizing tariff structure. However, the completion of the study and the implementation of its recommendations are unlikely to jome about unless all responsible authorities recognize its need (para 6.09)._/ An economic reevaluation of the project, carried out by TRENSITRR at the end or 1985, yielded an internal rate of return of 11.5%, as compared with the 24% estimated at appraisal. The main causes for the 1/ Please refer to footnote 7 on page 7 of this report. -v- lower-than-anticipated return are: (1) population and demand growth in the last five years significantly below appraisal estimate; (ii) proportion of passengers In the corridor using TRENSURB significantly below that anticipated; and (iii) operating costs, especially staff costs, above those anticipated (paras 7.03-7.05). Final real inveatment costs, substantialtv below original estimates, partially offset the decline in benefi:s. The economic benefits being generated by the prolect are not reflected in the financial results of TRENSURB. Although it was expected that TRENSURB would earn sufficient tariff revenue to cover its working costs, it was never expected to cover its full operational costs and interest charges. Some form of Government subvention was therefore inevitable. However, the present level of Government subvention, amounting to more thin six times the total passenger revenue, considerably exceeds the original expectations (para 5.06). The Project Completion Report offers the following conclusions (paras 8.01-8.04): (a) It is not sufficient that the railway service has been successfully implemented. The success of the project depends upon the service attracting sufficient passengers to make the investments economically viable and the railway profitable. This will he highly contingent upon the resumption of economic growth in Porto Alegre and in the evolution of the fares for the users of alternative transport modes. In addition, the attractiveness of the system would be enhanced by the implementation of the recommendations of the studies carried out under the project (bus integration, car parking restrictions, etc.); (b) One major lesson to be learned from this project is that, In cases in which there is a need for substantial policy decisions and the implementation of measures involving different Government agencies subject to conflicting pressures, their commitment should he realistically assessed during project preparation. The steps to ensure that these policies are impJemented on time should he carefully identified at the same time. In the particutlar case of the Second Urban Transport Project, a more detailed action program including specific policy implementation targets and appropriate measures for its enforcement should have been agreed upon during appraisal; and (c) Regarding the original objective of TRENSITR3 covering its working costs, the appraisal was unrealistic to expect that this would take place during the initial operation period. Only after the system has demonstrated Its reliability to the users, will the application of a revenue maximizing policy be feaslble. In the meantime, TRENSURB should review its staff needs in order to streamline costs and carr out a tariff study to determine the optimal user charges. / The Bank should closely follow-up the implementation of these taiskq in the context of its ongoing and future involvement in the sector and continue to impress upon the Government the benefits to be gained from a financially healthy public enterprise as well as the existence of tools much more effective than transport subsidies to achieve income distribution objectives. 2/ Please refer to footnote 7 on page 7 of this report. -~-1 PROJICT CO(PLITIOU REPORT BRAZIL SECOND URBAN TRAlSPORT PROJECT (LOAN 1839-BR) I. INTRODUCTION 1.01 The Bank's involvement with urban transport in Brazil began in mid-1978 under the First Urban Transport Project, implemented in response to an increased demand for efficiency in the provision of urban transport services throughout the country. The First Urbar- Transport Project was aimed at improving the efficiency of bus services in selected corridors and making public transport more widely available to low income sectors of the population. The Project was carried out in five major metropolitan areas, including Porto Alegre. The Porto Alegre component of the First Urban Transport Project consisted of improvements to the signalization of bus corridors, minor improvements to the operation of already completed projects, and assistance in the paving of routes used by bus services. 1.02 The Second Urban Transport Project, subject of this report, was designed to improve land use within the corridor with the fastest growing population, reduce che petroleum fuel consumption in the same corridor, and avoid traffic congestion both in the corridor and in the metropolitan center of Porto Alegre. The objectives were to be achieved by providing an efficient conventional suburban electric railway in the North-South corridor, with associated car parks, bus feeder services and interchange facilities, while at the same time restricting car parking availability in the downtown metropolitan area. 1.03 This PCR is based upon the following sources of information: (a) the Staff Project Appraisal Report (Report No. 2759b-BR); (b) Bank staff supervision reports; (c) LAC Information Center; (d) Draft Project Completion Report prepared by TRENSURB; (e) information collected during a mission to Porto Alegre in November 1985; and (f) a consultant's report dated April 1986. II. PROJECT PREPARATION AND APPRAISAL 2.01 The prospect of a Second Urban Transport Project in Brazil was first suggested by the First Urban Transport Project preparation mission in 1977. That mission recommended that rather than delay the preparation of the First Project by considering complex suburban rail components, it would be better to include them in a separate second project. A project -2- identification mission took place in February 1979. Petween 1977 and 197%, consultants were contracted to carry ouc an economic feasibility study of the proposed project. The project was conceived in two phases, the firsc being the construction of a new line from the business center of Porto Alegre to Sapucaia in the outskirts, a total of 26.7 km, and the second phase, an extension t.om Sapucaia to Novo Hamburgo (Map IBRD 146R8). The Project as described in the remainder of this renort refers only to the first phase unless otherwise specified. 2.02 The feasibility study confirmed the economic lustification of the project, on the basis of an assumption that 102 of far passengers in the corridor would be attracted to the new rail system. / In order to achieve this level of transfer, the project would include t¶e provision of car parking facilities at some of the stations. In addition, the agency for the coordination and regulation of transportation in the metroolitan region of Porto Alegre, Nucleo Metropolitano de Transportes Urhanos / (NMTU), would .plement policies which would restrict the use of car parWing space in the wntown area. NMTU would also be required t.o engage consultants to study a best way of modifying bus routes to make them complementary to, rather -.an competitive with, the rail service. 2.03 From the early stages of project design it was intended that a separate company (TRENSURB) be established to operate the railwaY. For practical reasons, the company would start up as a subsidiary of the Federal Railway Company (RFFSA). 2.04 An important objective of the project was to improve land uise and transport planning in the metropolitan region of Porto Alegre. / The support given by the Project to the newly created NMTU to foster its development into an effective transport regulatory and coordinating agency, was considered an important benefit. It was hoped that the project woul1 assist in generating an overall revision of the method of assigning bus route licenses, determining public transp-rt tariff policies, and implementing auto restraint policies. The Metropolitan Planning Agency (METROPLAN) was to be involved in identifying potential changes in land use. Consultants were to be appointed to examine possibilities for creating and developing commercial centers at specific stations on the new line. The 1/ The Borrower points out that the feasibility study also considered the diversion of 50% of bus passengers to the system (see page 1 of Annex j). 2/ The agency was created in December 1979, during project preparation, to replace the several state agencies previously involved in the coordination and regulation of transport in the Metropolitan Region of Porto Alegre. 3/ Should read "the North-South Corridor of the Metropolitan Region of Porto Alegre (see page 1 of Annex 1). same consultants were to be used to help METROPLAN update and revise its land use master plan. 2.05 The establishment of TRENSURB, a necessary prerequisite for signing the loan agreement, took place in April 19RO. A Rank loan of 1TSS159 million was granted on June 2, 1980 to finance the foreign expenditure component of the project. USS89.37 million (56.2%) was to finance the purchase of unit trains and related spare parts, USS14.19 million (8.9%) for signalling and electrification equipment, USS9.34 million (5.9%) for track superstructure and a further US$3.93 million (2.5%) for maintenance equipment and consultants studies. A further US$42.17 million (26.5x) was unallocated and intended for price and physical contingencies. 2.06 The Federal Urban Transport Authority (EBTU) was designated as the agent for the Borrower and was required to onlend the loan proceeds to TRENSURB. The Municipality of Porto Alegre was required to undertake issue and implementation of the pollcies in respect of car parking restraint. Other parties to the Agreement were METROPLAN (in respect of commercial development associated with the railway) and the State of Rio Grande do Sul (in respect of assisting in implementing the car restraint poliev and, through NMTU, in identifying bus routes to be discontinued, bus routes to serve as feeder links to the railway, and making recommendations regarding bus companies adversely affected by the project. NMTUY was also to undertake a study of the level and structure of all public transport tariffs in the Metropolitan Region of Porto Alegre including charges for satellite and other automobile parking). III. PROJECT IMPLEMENTATION AND COST Timing and Costs 3.01 The Loan Agreement became effective on December 9, 19RO. Problems with the availability of local funds and delays catsed by differences between the Bank and the Borrower concerning the application of ICB procedures, resulted in implementation delays and postponement by two years of the loan closing date, which was originally envisaged for Decemner 31, 1984. 3.02 The loan agreement was amended four times (Table 1). 4/ The first amendment, which took place in January 1983 consisted of the reallocation of loan funds to compensate for fluctuations in the exchange rate which causad excess funding in some categories and deficiencies in others, and to aczount for the firming up of some cost estimates. In particular, procurement of unit trains resulted in a price 48% less than that originally estimated. 41 The original table was corrected in accordance with the Borrower's comments (see page 2 of Annex 1). -4- The reasons for the lower-than-expected prices are two-fold: first, bidder's quotations were below appraisal estimates because of depressed world market conditions and, second, the appreciation of the US doliar vis-A-vis the yen when the contract for the Japanese trains was entered nto. In the case of Category 3, allocation of loan funds was increased by dbout US$18.5 million because of the inclusion of a telecommuolications system (originally not Bank-financed) and of an additional electric substation. The maintenance equipment category was also increased to permit the puirchase of track and catenary maintenance equipment, not originally envisaged for Bank financing since maintenance services were tc be contracted out to the Federal railways. Furthermore, new exrenditure categories were added to the loan: one for statlon equipment (automatic ticket control units) and the other for civil, works comprising the San Pedro Station, infrastructure works and pedestrian under/over passes. 3.03 A second amendment to the Loan Agreement Was introduced in September 1983. Part of the categorv of consultancy services was redesignated as technical assistance with 100% of Bank financing of foreign expenditures; the remaining consultancy services continued to be financed by the Bank at 60% of total expenditures.J/ At the same time, a number of complementary civil works (35% Bank-fi7nanced) were added to the program. Most of the additional works were designed to provide linkages between sections of established communities which would be severed by the new rail line, by means of viaducts and access roads. It was necessary to incorporate in the project these essential civil work._, which were originally planned to be financed by the municipalities and State governments, because budget constraints imposed on thes.- authorities prevented their undertaking them within the agreed timetable. Other extra works included extensions to already proposed buildings for maintenance and operations. In addition, the new category of stgtion equipment was excended to include tic':et issuing and checking machines._/ 3.04 Another important change to the Loan Agreement was agreed to in December 1983. Reductions in the 1982 and 1983 budgets of Government agencies, together with escalating civil works prices and changes in the exchange rate, resulted in a potential gap in the 1983 and 1984 financing plans for the Project. To compensate for exchange rate differentials and assist in project implementation, it became necessary to increase the Bank's participation in the civil works program from 35% to 65%. This also assisted in making up for deficits in the financing plan. It was Possible to introduce the higher disbursement percentage under the Special Assistance Program for Brazil. 5/ The increased Bank financing related only to the technical assistance to TRENSURB for its operations, management and financial planning (Part F (1) of the project as described in the Loan Agreement, as amended on January 11, 1983]. 6/ The Borrower points out that no ticket issuing machines were purchased (see page 3 of Annex 1). -5- 3,05 In late 1983, the scope of the consultant's services was extended. The scope of work of the consultant was initially defineI on the assumption that a number of tasks, tn particular in the areas of setting iin operational procedures and staff training, would he carried out by TRENSURB's own staff with minimum assistance from the consultant. However, a Presidential Decree (No. 88004, of December 1982) forbidding federally-owned companies to increase the numher of authorized staff positions, precluded TRENSURB's staff recruitment as scheduled. Only in September 1983, through an exception decree, TRENSURB was granted the right to hire additional staff. However, since the new staff to be hired would not be capable of carrying out some of the tasks which were on the project's critical path, TRENSURB proposed, and the Bank agreed, to contract the execution of those tasks to the consultant. 3.06 The reformulated scope of work of the consultant included: (a) in the area of personnel selection, training, development and compensation: (i) a more intensive assistance to TRENSURB in recruiting staff; (ii) the preparation of manuals for the training center (operational procedures and norms for the center, and training manuals for the main positions in the area of operations); and (iii) assistance in the setting-up of the training center and in implementation of the training program; and (b) in the area of design and installation of systems and procedures: (i) redefined tasks for the design and installation of the financial management system, to allow for more intensive computerization than originally envisaged; (ii) expanded assistance for the setting up of the maintenance management system; (iii) updating and detailed design of the operations plans; and (iv) preparation of the operations manuals of procedures, norms, and regulations. The scope of the consultant's work in other areas hasically remained unchanged. 3.07 During the final reallocation of funds between cateeories (Novemoer 1984), an additional US$1.0 million was allocated to technical assistance from the unallocated category. This was necessary in part because of the increase in the scope of the consultant's work described in the preceding paragraphs, and also because of the need to extend the services of the financial experts in order to ensure the completion of the implementation of a financial accounting system as wel.l as the development of subsystems for stores control and fixed assets accounting. 3.08 Total project cost was US$249.4 million, about 20% below appraisal estimate. Several reasons contributed to wide differences bevween the appraisal estimated and actual costs for individual project items (Table 2). In every case, because of changes in the exchange rate, US dollar costs were well below original estimates. However, increases in quantities of work for some project items more than offset the price decline (i.e. pedestrian and road vehicles under'over passes, stations and administration and maintenance depot buildings). In addition, some items - 6 - which were not Bank financed were not included in the estimated costs at appraisal. Such was the case of the consulting services for general engineering design, supervision of works construction and training; vehicles and office equipment; and complementary civil works. In any event, even though the size and scope of the prolect were expanded, loan funds in the amount of USS26.0 million were not utilized and were cancelled. Procurement 3.09 As envisaged during appraisal, track infrastructure works, pedestrian and road vehicle under/over passes and stations were to he financed by the Borrower and contracts were awarded under local competitive bidding procedures. These works were satisfactorily completed in early 1985 at a total cost of about US$80.0 million (Table 2). As indicated in para 2.05, other major components of the project such as unit trains, signalling, telecommunication and electrification systems, track superstructure, maintenance equipment, complementary civil works (para 3.03), and consulting services were to be financed by the Bank loan. ICB procedures were followed for these components. 3.10 Even before the project was appraised, the Issues Paper identified a potential conflict between Bank and Brazilian national policy with respect to the application of international competitive bidding rules. In the evaluation of bids for the unit trains, the Brazilian Government sought to add two items to the quoted CIF price of the foreign bids: the Merchant Fleet Renewal Fee and the Port Improvement Tax. The Bank considered both items as taxes, the Brazilians considered them as costs. There was also a difference of opinion in respect of the ocean freight rates quoted by the lowest bidder as evaluated by the Bank. The bid evaluation made hy the Brazilian Government increased the freight charges quoted by the lowest bidder, arguing that they were below the Conference Rate. The bid invitation, according to the Bank interpretation, only required the quotation of a rate by a shipping line belonging to a conference of which Brazil was a member, and not necessarily a conference rate. The lowest bid satisfied this condition. 3.11 The addition of the disputed Fleet Renewal and Port Improvement charges together with the Bank's permitted margin of preference for national suppliers, resulted in a bid by a Brazilian manufacturer heing the lowest. In terms of basic price, this bid was 21% higher than that of the lowest bidder. The bid evaluation results were communicated to the Bank at the end of August 1981. They were not accepted by the Bank; negotiations ensued, and by July 1982, the basic difference of opinion was resolved with the award to the lowest evaluated bidder, without the addition of the disputed charges. However, difficulties arose at the time of signing the contract in respect of the shipping line to be used, and the placing of the insurance contract for freight in transit, with the Borrower arguing that they should be placed with Brazilian companies. These difficulties were finally resolved in early 1983 with the Borrower's acceptance of the freight and insurance companies quoted by the manufacturer. 3.12 Problems also were encountered in the procurement process for signalling equipment. One of the bids was rejected on the grounds that -7- the bidder had not supplied all the company documentation as specified. The Bank at first suggested that there were insufficient grounds for rejection since the omission was only a technicality. A lengthy explanation of the Brazilian position was included in a letter from the President of TRENSURR dated March 16, 1983. The Bank finally agreed to the rejection, and the procurement for signalling equipment experienced onlv a six-week delay. Technical Assistance 3.13 The project included technical assistance for a value of U5S3.9 million (including contingenclts), of which 60% was to be financed hy the Bank. The principal component of the technical assistance was to be in the form of managerial support to TRENS1RB, to establish the company as a functioning railway operator. Other technical assistance was designed to recommend complementary transport policies to be carried out by NMTIT, and complementary land use plans to be carried out by Metroplan and/or the municipalities in the corridor. 3.14 The management consultant was appointed in July 1982 and provided valuable assistance at a vital time in the establishment of TRF.NSURB. The recruitment constraints placed on TRENSURB by the federal government resulted in insufficient counterpart staff being available. This caused delays to the consultant's work because tasks expected to be undertaken by TRENSURB staff, on the consultant's critical path were not carried out on schedule. The consultant's contract was extended and revised to include some of the delayed tasks, particularly those relating to staff training. The additional work represented an increase of 40% in the consultant's workload. The consultant carried out the tasks effectively and, in general, on time. The few delays which occurred could he attrihuted to factors outside the consultant's control. 3.15 Three of the consultancy contracts, related to studies on the reorganization of bus services, the feasibility of TRENS1TRR Phase 2, and on land-use for satellite automobile parks, were awarded to a local consulting firm, which completed its assignment. Subsequentlv, the same local firm was selected to carry out the study of public transport tariffs, including those to be used by TRENSURB, however, the contract was not signed because this firm ran into financial difficulties. After consultation wlth CBTU, it was agreed that GEIPOT would undertake the study, however, the study gad not been completed at the time of writing of this report (February 1987)./ 3.16 The other three studies, related to the development of commercial areas and services close to TRENSURB stations, the satellite automobile parks, auto restraint policy and monitoring (undertaken by a consortium), and to the urban development along the area of influence of TRENSURB (prepared by Metroplan) were completed in 1982. 7/ The tariff study was finally received by the Bank in November 1987 together with the Borrower's comments (see Borrower's letter included in Annex 1) and is now under review. 8- 3.17 Although the performance of the consultant for the studv and recommendation of complementary policies was acceptable, there was a lack of urgency in awarding the contracts and there has since been a lack of interest in implementing the recommendations. Supervision missions frequently brought the problem of implementation to the attention of the responsible authorities but with little success, mainlv because of the political difficulties expected to arise from such implXmentation. IV. OPERATIONS AND TRAFFIC Operations 4.01 Table 3 pre3ents a comparison hetween the operating statistics estimated at appraisal and those recorded in early 1986. Passenger services started on March 3, 1985 with a pilot service operating during limited 9 hours. The service has built up and is now operating 17 hours per day,9/ with a maximum peak frequency of ten trains per hour, five trains per hour in the off-peak and three trains per hour on the late night service (after 20:30 hr). The peak period frequency is lower than that envisaged at appraisal (10 per hour vs 11 per hour) but higher than that envisaged for the off-peak (five per hour vs four per hour). The total number of daily services (192) represents 87% of that envisaged in the appraisal evaluation (220 per dayV Since the trains are at present operating with one unit all of the day, v_ instead of with two in the peak and intermediate peak periods, the daily unit-kilometers (5,165) are only 74% of those expected (6,992). It follows that the total capacity offered is lower than envisaged. 4.02 The availability of unit trains for passenger service is lower than that originally expected. A problem with corrosion of the floors in the region of the door frawes is being rectified by the manufacturer under the terms of the guarantee but the rectification work is resulting in a low availability of units. The utilization of the available units is better than anticipated because of the high off-peak frequency. 4.03 The level of maintenance of unit trains is being adversely affected by the low number of maintenance staff available. TRENSURR has 8/ Paragraphs 3.15, 3.16 and 3.17 were changed to reflect the Rorrower's comments (see page 3 of Annex 1). 9/ The Borrower points otut that the service is now operating 17.5 hours per day. °/ Each unit consists of four cars. -9- been partially successful in avoiding the full impact of public sector staff recruitment constraints, but maintenance staff levels are still below those considered necessary (para 5.10). The problem will become more serious a few years from now when the units begin to require general overhauls after 500,000 km. 4.04 The average operating speed is at present 41 km/hr. The expected average operating speed was 42 km/hr. The lower-than-expected speed results from lack of detailed analysis of possible speed profiles related to track geometry and some temporary restrictions imposed by the unit train manufacturer. The lourney time from Mercado Station to Sapacuia is 39 min, compared to an appraisal e3timated time of 38 min. This additional time requires that the number of peak hour trains in service he increased hy two to i,aintain the ten-per-hour frequency. 11/ 4.05 The number of passengers recorded as travelling per day has increased from 95,000 in the first week of service to 103,000 in July 1985 and reached an average of 113,000 in 1986. 12/ The appraisal evaluation was based upon an estimate of 260,000 passengers per day for 1985. There are several reasons for the substantial difference between the projected and actual passenger level. 4.06 First, the projected population growth in the transit corridor has been much lower than expected. The depresssed level of the Brazilian economy in the years since project appraisal has resulted in a slowdown in the rate of population migration from rural areas to the metropolis. Planning policies which were expected to concentrate anticipated urban growth along the corridor have not been implemented (slowdown in migration has made them unnecessary), and so the limited actual population growth has been more dispersed through the metropolitan area. The estimated population of the metropolitan area in 1985 was 2.73 million, 17% below the appraisal forecast of 3.20 million. 4.07 The lower-than-appraisal-expected population growth is reflected in the traffic counts on the principal road in the same traffic corridor, whicTh indicate traffic levels between 9% and 30% lower than those projected in the appraisal. With lower traffic levels and less traffic conaestion, the incentive for car passengers to transfer to rail is less than anticipated. 11 Subsequent to the writing of this report the average operating speed has increased to 46 km/h, thus reducing the journey time from Mercado station to Sapacuia to 35 minutes. See page 4 of Annex 1 for further details of operational changes reported by the Borrower. 12/ See page 4 and Table 3 of Annex 1 for the updated traffic statistics reported by the Borrower. - ic - 4.08 Second, there was an assumption that commercial development at selected points along the corridor would take place, increasing the number of trips from the central area. The appraisal made little or no allowance for the time necessary for this development to take place. As the service has been in full operation for less than two years, there has been little opportunity for an impact on lprd use patterns. The major impact of this factor is unlikely to be evident for at least five years from the start of services. 4.09 Third, the appraisal assumed that city centre car par,ing would he restricted, feeder bus services would he implemented in the suburban municipalities, and cat parking facilities would he made availahle at suburban stations. In practice, there has been no restraint on city centre car parking, only two feeder bus services are operating, and car parking available at suburban stations is minimal. 4.10 Finally, there is probably an underestimation of present passenger flows since the automatic ticket barriers only recently became operational and the temporary manual inspection of tickets allowed for a significant level of fare evrasion. This is estimated at between 10% and 20%. This underestimation, however, may have a limited influence on longer-term passenger demand expectations since many of the non-paying passengers may he deterred from travelling when the system becomes less susceptihle to evasion. 4.11 To offset the factors leading to lower-than-projected user levels described in the preceding paragraphs and attract passengers, the present tariff has been set lower than that envisaged at appraisal. The appraisal assumed a rail tariff of 120% of the minimum hus tariff. The present rail tariff is about 80% of the minimum bus tariff. Nevertheless, passenger demand is less than half of that anticipated, a fact that negatively affects the financial results and economic reevaluation of the project. V. FINANCIAL PERFORMANCE Passenger Revenues 5.01 The appraisal report indicated that TRENSURB operations should earn sufficient traffic revenue to cover working costs and to make a significant contribution to operating costs. No specific financial projections were included in the appraisal report. Instead, annual operating costs were calculated for six different levels of operation, ranging from 100,000 to 500,000 passengers per dayv Passenger revenue was not specified but it was estiniated at appraisal that it would be set above working costs per passenger (Cr$ 4.82; 18 IJS cents in mid-1197 values) and below the then current average bus tariff (CrS 7.00 or 27 US cents in - 11 - mid-1979 values)J13/ 'inancial sensitivity analyses were made with tariff's of Cr$ 2, CrS 3, ZiS 4, CrS 5 and CrS 6 per passenger. The analyses with base case operating costs indicated a working ratio of 100 at a tariff between CrS ' and CrS 3. With personnel costs incrersed by 100%, this working ratio was entimated to be achieved at a tariff of between CrS 4 and CrS 5. 5.02 Technical assistance was included in the loan to finaace a ;,tudv of public transport tariffs in Porto Alegre with the intention of leading to a recommended tariff for TRENSURB, hut the tariff study has not so Tar been completed despite continual pressure from the Bank on the Borrower. / When passenger operations started in March 1985, the tariff was setiat CrS 450 (equivalent to US 8.7 cents). The tariff has twice heen increased and was,in March 1987, CzS 2.00 (equivalent to US 12.0 cents). To encourage interchange passengers, a combination TRENSURB/feeder hIs ticket is available on the route in Canoas. The tariff for the combination ticket is CzS 3.50 (US 22 cents). These fares are still below levels which would at least cover working costs (para 5.07). 5.03 Tables 4 and 5 Present, respectively, TRENSURB's preliminary Financial Results and Balance Sheet for 1986. As seen in the tables, operating revenue amounted to USS2.37 million equivalent. This compares with an estimate in the appraisal report of USS1R.14 million for the first year of operation. / Total income for 1986 was TUSS18.40 million, including US$16.2 iiTllion in government subsidy. Although it was expected that TRENSURB would earn sufficient tariff revenue to cover its working costs, it was never expected to cover its full operating costs (working costs plus depreciation) and interest charges. Some form of government subvention was therefore inevitable. However, the present level of Government subvention, amounting to more than six times the total passenger revenue, considerably exceeds the original expectations. Working Costs 5.04 The present pattern of traffic operations was nlanned for A passenger level of 260,000 passengers per dav. In particuilar, the number of unit trains acquired was based upon this level. The average working cost per passenger, as estimated bv TRENSURB, for 1085/86, is US 11 cents. The total 1986 working cost is USS12.63 million equivalent for an average 3/ The minimum bus fare for the shorter routes was Crq 5.n. 14/ See footnote 7. 5/ This figure was corrected to reflect the Borrower's comments, see page 4 of Annex 1. - 12 - traffic level of 113,000 passengers. These figures compare with appraisal report figures of US 18 cents and USS14.04 million respectively at 260,000 passengers per day and IJS 21 cents and USS6.20 million equivalent for I00,000 vassengers per day. 5.05 The discrepancy between the total costs projected in the appraisal Leporc and those achieved is largelv attributahle to the categorv of lahor costs. A comparison between the appraisal projected staff nurhers hy category and present TRFNSURB staff numbers in the same categories, is shown in Table 3. The present staff costs are USS8.81 million equivalent per year, compared with USS5.50 million (260,000 passenger level) in the appraisal assessment. 5.06 The average cost per staff memher is considerably less than that estimated in the appraisal report. The appraisal costs were TJSSR,647 and US$9,280 per year for the 260,000 and 100,000 per day passenger levels respectively. The actual cost is onlv US95,705 per person. 5.07 However, the present TRENSURB staffing level of 1,294 (eno-19QR) is almost double that envisaged in the hasic proposal, to carry 26n,n00 passengers per day. The staffing levels were substantially underestimated at appraisal; the consultant subsequently recommended a staffing level of 1,790 for the full system. This underestimation was especially evident in the administration, station and workshop categories. Nevertheless, the organization chart that forms the basis of the consultant's estimates indicates a rather top heavy administrative structure more typical of a public sector railway than of a commercially-oriented enterprise. Increases of this order in working costs, coupled with low fares have adversely affected tF-e financial performance of the enterprise. VI. INSTITUsTIONAL PERFORMANCE 6.01 Institutional performance needs to he considered in two distinct parts. First, the performance of the new company, TRENSURR itself, and second, the performance f the institutions responsible for the complementary transport and land use policies envisaged at appraisal. Both institutional groups have interrelated roles, crucial to the eventual success of the project; a satisfactory performance by one of them can easilv be negated by the poor performance of the other. 6.02 Considering the project objectives at appraisal, the creation of a new company with shareholding by various existing public sector authorities and the construction and putting into operation by that companv of a completely new service within a five-year period is an ambitious target in any context. Likewise, the investments in the new services were to serve as a driving force in restructuring urban transport and land use patterns in a complex metropolitan area, also within a five-year span. In retrosoect, it can therefore be said that project objectives at appraisal were overly ambitious. - 13 - 6.03 The degree of success achieved varied. TRENSURR itself succeeded in completing all the physical works and equipment procurement, and has teen operating a passenger service for almost two years. In contrast, few of the complementary policies have been implemented. That is not to say that no beneficial results at all have been obtained, since the various studies carried out under the project are providing useful inputs to the modest town planning efforts being carried out. However, implementation of the studies' recommendations, in the context of the rapid transit system, is taking place at a pace much slower than originally envisaged. TRENSURB 6.04 TRENSURB was already established as a subsidiary of RFFSA when the Loan Agreement was signed. The early development of an organizational structure and the filling of senior posts with competent and experienced professionals was handled efficiently. The first serious problems arose when the federal government prohibited the recruitment of further staff by all federal agencies. This prohibition came at a time when TRENSTTRR needed to expand its staff rapidly in the process of changing from a construction to an operating company. Nevertheless, subsequently, TRENSURB was able to overcome this problem by obtaining an exemption from this prohihition and by temporarily placing some in-line responsibilities on the management consultant. TRENSURB's performance is still hampered by interference from local, state, and federal governments and pressure groups, while striving to get established as a transportation company. For its performance to improve significantly, it should be afforded a greater degree of autonomy, while at the same time fully implementing the complementary town planning policies. Other Institutions 6.05 At the time of the appraisal, much reliance was placed upon the newly created NMTU. It was agreed that NMTU would: (i) grant and review franchises for intra- and intermunicipal bus services and taxi operators; (ii) specify bus routes; (iii) coordinate levels of service at intermodal connections; (iv) specify, regulate and review safety standards for all transport modes; (v) provide transport enterprises with guidelines for the setting of tariffs, monitor such tariffs, and require such enterprises to propose tariff changes; (vi) regulate the use of private cars through the ,coordination of municipal parking policies, user charges, and taxes, and encourage the use of public transport; (vii) review and advise on all proposed transport infrastructure investment decisions in the metropolitan area; and - 14 - (viii) prepare and update annually an indicative five-year rolling transport plan to serve as a guide for public and private investment decisions. 6.06 There is no doubt of the need for such an institution as NMTU but in practice it was too optimistic to expect that an under-resourced and not fully accepted institution could achieve so many objectives in the context and time-span of one project. Support for NMTU was weak, and, as a whole, it was unable to overcome the reaction which arose against it because many of its functions interfered with established interests. To sum up, NMTU was subject to considerable opposition and it lacked the political support which was essential to pursue its objectives successfully. 6.07 For a long time, it was not certain whether NMTU had the legislative authority to undertake the tasks it was responsible for under the Loan Agreement. The first problems become apparent in 1980 when it was realized that NMTU could not contract directly with a consultant because it had not been formally established as a legal entity. The status of NMRTU was still unclear early in 1984. At that time, the State could not decide whether NMTU should function under the State Governor or the State Secretary for Transportation. Finally, in mid-1984, NMTII was dishanded, and its responsibilities assumed by METROPLAN. Unfortunately, the same indeterminations regarding status are now associated with METROPLAN's legislative authority to carry out the same tasks, a fact that continues hindering the achievement of METROPLAN's multiple objectives. 6.08 The main institutional achievement under the project was the completion of the studies for complementary policies (paras 3.14 and 3.15), with the exception of the tariff study. As already discussed, the implementation of the recommendations emanating from those studies is proceeding slowly. More specifically, some bus feeder services have been established (i.e. to the Canoas, Esteio and Sapucaia stations), recommendations on commercial development are being included in the medium-term municipal development master plans, and some development investments are already taking place in the Esteio station area of influence. With regard to the car parking rgstriction, measures were taken only in the area of the Rodoviaria Station. / 6.09 The most serious shortcoming is the failure to undertake and implement the public transport tariff study. As already discussed, TRENSURR is charging tariffs which cover only one third of its working costs, in part as a result of implementing a tariff policy with different objectives than those included in the Loan Agreement. The tariff study, now under the responsibility of METROPLAN and CBTU, is urgently needed if the financial problem of TRENSURB is to be alleviated in the short term. This study should reveal the financial cost of keeping tariffs at their current levels, and provide a basis for establishing a short term revenue maximizing tariff 6/ The Borrower points out that car park restriction measures were also - implemented in the Mercado Station. - 15 - structure. However, the completion of the study and the implementation of its recommendations are unlikely to come about unless all responsihle authorities recognize its need. Politically, actions to increase revenues and/or decrease costs may be lifficult, especially if TREN9URB is perceived as a social service, rather than a commercial enterprise. / VII. ECONOMIC REEVALUATION 7.01 The appraisal report included an economic evaluation of five alternative public transport projects for the North-South corridor of Porto dlegre. The selected alternative yielded an economic internal rate of return of 24%. Sensitivity tests were carried out with investment costs 15% nigher, passenger time values at zero, onlv 10 of car passengers transferring to rail instead of 10% in the first year increasing to 25% by the 30th year, and an overall reduction of 15% in benefits. The rates of return for the sensitivity tests were 22%, 17%, 19%, and 22% respectively; i.e., they showed a low sensitivity to the range of tests made. 7.02 In practice, the costs and henefits have hoth fallen outside the range of the sensitivity tests. The investment costs are 80% of those in the base case, TRENSURB passenger levels at present are 40% of those in the base case, operating costs are 10% higher, and road traffic is only about 80% of that projected for 1985. The reevaluation is hasically to test whether the reduction in investment costs sufficiently offsets the fall in benefits to maintain an acceptable rate of return. 7.03 A reevaluation was carried out at the end of 1985 by TRENSTJRB, using basically the same methodology as the appraisal evaluation but comparing only the implemented project and a 'without project' case. This reevaluation indicated a rate of return of 11.5% (Table 6). 7.04 The main causes of the fall in the rate of return from that of the appraisal report are: (i) population and demand growth in the last five years significantly below that anticipated; (ii) proportion of passengers in the corridor using TRENSURB significantly below that anticipated; and (iii) operating costs, especially staff costs, above those anticipated. 7/ As reported by the Borrower, the preparation of the study was subsequently placed under the responsibility of Geipot. With the assistance of CBTU and TRENSURB, the study was finally completed in September 1987 (see footnote 7). - 16 - 7.05 The reduction in economic growth In Brazil in the early 1P8Os, i.e., the underlying factor in (i) above, was not foreseen and cotild not reasonably have been so. However, the sensitivity test for demand did not cover a wide enough range, given the high growth rates projected. The projected modal split was optimistic, at least for the early vears of the project. A lower proportion of commuters using the new service in the early years of the project has a disproportionate effect on the benefits and hence on the rate of return. The original cost estimates underestimated the staff requirements for an independent railway operation; nevertheless, nresent staff levels are higher than technically necessarv / VIII. CONCLIJSIONS 8.01 The major physical components of the project were successfully completed within one year of the projected completion date and at a cost 20% below appraisal estimates, considerable achievements for a project of this size and complexity. However, the failure to implement complementary measures is having a negative effect on the operational and financial performance of TRENSURB. It is not sufficient that the railway service has been successfully implemented. The success of the project depends on the service attracting sufficient passengers to make the investments economically viable and the railway profitable. This will he highly contingent upon the resumption of economic growtit in the metropolitan region of Porto Alegre and in the evolution of the fares for the users of alternative transport modes. In addition, the attractiveness of the svstem would be enhanced by implementation of the complementary actions recommended by the studies carried out under the project (bus integration, car parking restrictions, etc.) 8.02 One major lesson to be learned from this project is that, in cases in which there is a need for substantial policy decisions and the implementation of measures involving different levels of Government agencies subject to conflicting political and social pressures, their commitment should be realistically assessed during project preparation. The steps to ensure that these policies are implemented on time should be carefully identified at the same time. In the particular case of the Second Urban Transport Project, a more detailed action program including specific policy implementation targets and appropriate measures for its enforcement should have been agreed upon during appraisal. 8.03 Although the economic reevaluation yielded a marginally acceptahle return, this was mostly due to fortuitous developments that resulted in a final real investment cost substantially below original estimates. However, the economic benefits being generated by the project are not reflected in 18/ The Borrower does not concur with the report's opinion regarding present - staff levels of TRENSURB (see page 5 of Annex 1). - 17 - the financial results of TRENSURB, since tariffs charged are not in line with the quality of service provided. 8.04 Regarding the original objective of TRENSURR covering its working costs, the appraisal was unrealistic to expect that this would take place during the initial operation period. Only after the system has demonstrated its reliability to the users, will the application of a revenue-maximizing policy be feasible. In the meantime, TRENSURB should review its staff needs in order to streamline costs and carry out a tariff study to determine the optimal user charges. The Bank shoald closely follow-up the implementation of these tasks in the context of its ongoing and future involvement in the sector and continue to impress upon the Government the benefits to he gained from a financially healthy public enterprise as well as the existence of tools much more effective than transport subsidies to achieve income discribution objectives. - 18 - TARLF. 1 PROJECT CGOMPMO IEPORT MEMl IMME TRMIPT PROECT (LOAM i8lA-M) Olsburaents Setidulas (US$ I I Ion) Origin l Revislon Revision Rowleslon Revision Final Allocation Jan. 1983 Sept. 1983 Dec. 1983 NOV. 1984 Allocation (I) Traick Superstructure 9,340,000 10,530,00 9,400,000 9,400,000 9,400,000 8,410,57.75 (2) Unit Tralns ad related spares 89,370,00 77,780,000 77,800,000 73,000,000 73,000,000 66,999,373.84 (3) Slgi fling, electrification, etc. 14,190,000 32,700,000 29,000,000 29,000,000 29,000,000 27,834,677.19 (4) intebamnos equilpmnt 2,220,000 9,200,000 9,200,000 9,200,000 9,200,000 7,907,650.10 ;S) Pasenger s'stlan equlprnt 1,50o0o,0 4,300.000 4,300,000 4,300,000 3,363,465.99 (6) Conultantsb Servics 2,500,000 (a) Technical Asslstance 1,8Q,O000 1,800,o0 2,800,000 2,919,028.19 (b) Consultanb' Services 1,710,000 700,000 700,000 700,000 175,093.36 (7) C lvi I Work 10,000,000 (a) Parts A and 0 (t) thru 8 (4) 10,000,000 13,600,000 13,600,000 8,182,936.3 (b) Parts 9 (5) thru a (71 7,900,000 11,700,000 11,700,000 5,770,189.43 (8) Uncliocabd 42,170,000 14,790,000 8,900,000 6,300,000 5,300,000 - 0 - 70TO.: 159,000,000 1590000 59,000,000 159,000,W0 139,00,000 132,950,137.53 Sources (Date) - ~~~~~~~~~~~~1 - 19 - TABLE 2 PROJRCT COMPLETION REPORT BRAZIL SECOND URBAN TRANSPORr PROJECT (LOAN 1839-BR) Cwaparison Between Appraisal Estimated and Actual Project (but. (US$ million) Estsmr.*^ed Actual z Cobue Costs Difference Track Infrastructure 15.0 20.8 +39 Track Superstructure 34.4 16.8 -51 Pedestrian and road vehic'le under/over passes 35.6 39.5 +11 Stations 23.0 19.1 -17 Administration and maintenance depot buildings 10.7 15.0 +40 Maintenance depot equipment 2.8 6.6 +100 Permanent way maintenance equipment 0.5 Telecommunications 5.4 4.1 -24 Signalling 12.6 14.6 +16 Electrification 17.1 19.0 +11 Satellite car parking 4.4 0.1 -98 Expropriations 23.3 2.0 -91 Unit trains 114.5 64.8 -48 Unit train spare parts 9.6 Technical assistance 3.9 5.4 +38 Consulting Services for Projects, Studies and Supervision - 8.2 - Training, vehicles and office equipment - 1.4 Complementary Civil Works _ 12.0 TOTAL: 312.8 249.4 -20 = 1/ Including contingencies. Source: TRENSURB - 20 - TARLE I PROJECT CIrPTION RIUPRT AZIL SECOND UMll TRANISPOR POJECT (LOAN 1839-BR) Opetatlt Statel lea / Appraisal 1. Train Operations Estimate Actual A. General Information Origin Mercado Station Mercado Station Destination Sapucaia SaDucaia Route length 26.7 km 26.7 km Average train speed 42 km/hr 41 km/hr Operating day 18 hours 17 hours Average operating days per year 300 days Ano davs Unit train 4 cars 4 cars Unit train passenger capacity 1,200 1,n78 Unit train energy consumption 62.4 watt/hr 64.7 watt/hr per xross ton-km B. Passenger Traffic (per day) 260,000 1n3,nnn C. Number of Trains Peak hour 11 10 Off-peak hour 4 5 D. Headway between trains (minutes) Peak hour 5.5 h Mid-peak hour 6.6 12 Off-peak hour 15.0 12 E. Fleet Requirements (unit trains) Average peak hour 22 17 In repair/maintenance 3 A Total 25 25 Availability 85% 70% F. Personnel requirements (staff members) DirectorsF 9 12 Managers 10 22 Supervisors 12 28 Drivers and inspectors 33 1A2 Administration 60i 228 Station Staff 366 469 Workshops 58 293 Track Maintenance R6 An Total: 634 1,2q4 2 G. Operating Statistics Number of daily train trips 220 192 Number of daily unit train trips 344 192 Daily train-km 4,264 5,165 Daily unit train-km 6,992 5,165 Daily gross ton-km (1,000) 2,166 1,176 See Table 3 of Annex I for updated (1987) statistics. 2 This figure compares with a total of 1,790 staff recommended by the project consultant. The appraisal numbers were considerahly underestimated. Source: TRENSURB March 1987 - 21 - TABLE 4 PROJECT CONFLITION REPORT SECOND URBAN TRANSPORT PROJICT (LOMN 1839-SR) TRINSuU - Financial Results (1986) Appraisal Estimate Actual (US$ million) (US$ million) Reference Demand (passengers per day) 260,000 113,070 Income Passenger Revenue 18.14 2.37 Other Revenue - - Government Subsidy - 16.12 Subtotal: 18.14 18.49 Expenses Salaries and Wages 5.50 8.81 Energy and Lubricants 1.23 0.62 Materials and Other 1.30 3.20 Subtotal Working Costs: 8.03 12.63 Depreciation 6.54 12.96 I/ Total Operating Costs: 14.57 25.59 Operating Results 3.57 (7.09) Net Financial Expenses 0.23 (52.02) Monetary Adjustment - 56.73 Annual Profit/Loss 3.80 (2.38) Ratios Working Ratio 0.44 0.68 Operating Ratio 0.80 1.38 1I Including amortization of deferred assets. Exchange Rates 1979: Cr$ 26 1986: Cz$ 13.65 (In February 1986, Cr$ 1,000 became Cz 1.00) Source: TRENSURB - 22 - TABLE 5 PROWJCT COKPLTION REPOIT SIEL S800ND UMBAN TRANSPOIT PROJECT (LOAN 18319-BR) TRKiURi - Uslance Sheet (1986) Appraisal Estimate Actual (US$ million) (US$ million) Assets Current Assets 1.67 1.31 Investments 9.50 0.02 Stock 6.40 1.15 Fixed Assets 1/ 238.55 187.29 Deferred Assets - 78.19 Total Assets: 256.12 267.96 Liabilities and Equity Current Liabilities 0.82 2.70 Long-Term Debt EBTU Loan 159.00 146.95 4/ Other Debt _ 4.70 Equity Capital 92.49 90.48 2/ 5/ Reserves 3.81 23.13 3/ Subtotal: 96.30 113.61 Total Liabilities and Equity: 256.12 267.96 1, Preoperational expenses capitalized. 2/ Including monetary adjustments. 3/ Includes operating results. 4/ Including about US$10.0 miAlion of foreign exchange loss. 5/ Including US$9.69 of capital contribution. Source: TRENSURB - 23 - TABLE 6 PROJECT COMPLETION REPORT BRUAIL SECOND URBAN TRANSPORT PROJECT (LOAN 1839-BR) Economic Reevaluation - Cost and Benefit Streams (CZ 1Or) Investment Savings in Savings in Time Total Year Economic Costs Maintenance Costs Operating Costs Savings Benefits 1980 -3.97 0.00 0.00 0.00 0.00 1981 -14.80 0.00 0.00 0.00 0.00 1982 -22.64 0.00 0.00 0.00 0.00 1983 -31.41 0.00 0.00 0.00 0.00 1984 -97.36 0.00 0.00 0.00 0.00 1985 -20.15 -.42 -2.85 3.82 .55 1986 -6.36 -.42 -2.73 5.70 2.55 1987 0.00 -.42 -2.62 7.54 4.50 1988 0.00 -.42 -2.51 9.31 6.38 1989 0.00 -.42 -2.42 10.98 8.14 1990 0.00 -.42 -2.33 12.51 9.75 1991 0.00 -.42 1.42 19.00 20.00 1992 0.00 -.42 5.62 25.80 31.00 1993 0.00 -.42 10.31 32.93 42.82 1994 0.00 -.42 15.55 40.44 55.57 1995 0.00 -.42 21.41 48.37 69.36 1996 0.00 -.42 20.35 51.10 71.02 1997 0.00 -.42 18.96 54.00 72.53 1998 -24.48 -.42 17.19 57.09 73.86 1999 0.00 -.42 15.00 60.39 74.97 2000 0.00 -.42 12.32 63.93 75.82 2001 0.00 -.42 12.23 69.61 81.42 2002 0.00 -.42 12.04 75.39 87.01 2003 0.00 -.42 11.73 81.31 92.62 2004 0.00 -.42 11.32 87.41 98.31 IRR - 11.5% Source: TRENSURB -24 - ANNEX I Page 1 of 9 ORIGINAL COOENTS FROM THE BORROWER rQ TRENSURB EMPRESA DE TRENS URSANOS DE PORTO ALEGRE S. A. coc. 076653 /0001-6 OF-PRES-148/87 Porto Alegre, 16 de outubro ie 1987. Ref.: BIRD II- Emprestimo 1839-BR Relat6rio de Conclusao de Projeto Prezado Senhor Ribi: Recebemos desse BIRD, c6pia da minuta dc Relatorio de Conclusao de Projeto preparado pelo Banco, e gentilmente remetido para nossa apreciagao. Nossa equipe analisou 0 documento e emitiu comentarios, que anexamos, juntamente ccm o Estudo sobre Tarifas no Corredor Norte da Regiao Metropo- litana de Porto Alegre. Esses comentarios tem porobjeti- vo complementar algumas observag6es do PCR do Banco,e atua lizar os dados operacionais ccm informag6es do ano de 1987 ate o mes de agosto. Com este dccumento, aproxima-se do encerram.ento a se rie de informagoes, e ccmentarios de andamentos intercam- biados em correspondencias e em sucessivas reunioes de mis sao do Banco ao longo do periodo de vigencia do emprestimo 1839-BR. No entanto, continuamos a disposigao para quais- quer informag3es que o Banco desejar. Estaremos no aguardo do recebimento de um exemplar da versao definitiva do Relat6rio de Conclusao de Projeto. Atenciosamente, UETO DE CASTRO SERAFINI FILHO Diretor Presidente Ilmo. Sr. Rene A. Ribi Acting Chief Poliay-besed Lending, Energy, Industry and Urban Sectors Division Operations Evolution Departament Banco Mundial de Reconstrugao e Desenvolvimento WASHINGTON, DC, USA RUA ERNEST0 NEEUGEAUE!., tIN. POO ALEGRE - tS. - FON: PABX (0512) 43-4111 -25 - ANNEX 1 Page 2 of 9 TRENSURS EMPRESA DE TRENS URBANOS DE PORTO ALEGRE S. A. CGC. 90976853/001.56 OBSERVAQOES SOBRE 0 "PROJECT COMPLETION REPORT (PCR)" DATADO DE 11 DE SETEMBRO DE 1987 E ENCAMINHADO PELO BIRD EM 11.09.87 Pag. 2 - Paragrafo 2.02 A hipotese sugerida pelos Consultores refere-se a transferencia modal de 10% dos passageiros provenientes de autom6veis e 50% dos Snibus. Pag. 2 - Paragrafo 2.04 Os objetivos do projeto referem-se ao desenvolvi- mento do uso do solo no corredor norte-sul da regiao metro politana. Pag. 3 - Paragrafo 3.02 No que se refere a tabela 1: . Os valores langados estao representados em milh8es de d6lares; estao em desacordo com o cabegalho; . 0 valor correto do item (2) - Trens-Unidade,e de US$ 89,370,000; . 0 valor correto do item (1) - Superestrutura, e de US$ 10,530,000; . 0 valor resultante da revisao de Janeiro de1983, continua sendo US$ 159,0CO,OOO. Pag. 4 - Paragrafo 3.03 _ Nao foram adquiridas maquinas de emissao de bi- lhetes. RUA ERNESTO NEUGEBAUER, 1985 FONE PAUX (0512) 434111 PORTO ALEORE -F S - 26 - ANNEX 1 Page 3 of 9 TRENSURB E MPRESA DTRENS URSANOS OF PORTO ALEGRE S. A. Pag. 7 - Paragrafo 3.15 Os estudos contratados com a consultora (SAPSA) fo ram o estudo de viabilidade da 2' etapa; o estudo de rema nejo das linhas de onibus e o estudo das areas satelites de estacionamento para automoveis. Destes, os dois primeiros foram concluldos ate a minuta do relatorio final, e o ter- ceiro, chegou a ser apresentado em forme. definitiva. 0 es- tudo de tarifas nao chegou a ser contratado com a mesma con sultora (julghda vencedora da ccncorrencia), devido as suas dificuldades financeiras que a impediram de apresentar rela t6rios finais definitivos de dois estudos. O estudo de tarifas foi realizado pelo GEIPOT, com ass?ssoramento da CETU e TRENSURB. Foi concluldo em sua forma definitiva em setembro de 1987, tendo sido entregue em maos a missao do Banco que visitou a TRENSURB nos dias 5 e 6 de outubro/1987. Pig. 7 - Paragrafo 3.16 Outros dois estudos foram realizados por un.cons6r- cio consultora (TRANSPLAN-Bourscheid): o estudo do desen- volvimento de areas de comercio e servigos junto asestagoes, e o estudo da politica de restrigao ao uso do autom6vel, am bos concluldos em 1982. O PCR nao menciona o estudo do desenvolvimento ur- bano ao longo da area servida pelo trem metropolitano. 0 relat6rio de ccnclusao elaborado pela TRENSURB (pag. 60)faz esclarecimentos a respeito do estudo mencionadc, e que foi elaborado pela METROPLAN em 1982. Pig. 7 - Paragrafo 4.01 . A TRENSURB esta operando hoje, 17,5 horas. (5:30 hs as 23:0 hs); . Consideramos que embora a oferta de trens seja inferior aquela prognosticada no "Appraisal",ela esta adequada a demanda, proporcionando um bom nivel de servigo ao usuario. T4 .? RUA ERNESTO NEUGEBAUmR, 1965. FONE PFAX (0512 43.4111 - PORTO ALEGRE ARS - 27 - AM=E 1 Pase 4 of 9 TRENSURB EMPRESA DE TRENS URBANOS DE PORTO ALEGRE S. A. CGC. 90176853000 1-56 Pag. 8 - Paragrafo 4.04 Ap6s conclusao dos trabalhos de modificagao do perfil de velocidade em julho de 1987 a veloclda de media aumentou de 41 km/h para 46 km/h. 0 tempo de viagem entre Mercado e Sapucala foi reduzido de 42 minutos para 35 minutos. Com estas modificag6es, reduziu-se o nCuero de trens em operagao de 9 para 7 no horario entre pico, e de 14 para 12 no horario de pico matinal. Sem alterar o "headway". No pico da tarde fo- ram mantidos os 14 trens, a fim de aunentar a oferta diminuindo-se um pouco o "headway". Pag. 8 - Paragrafo 4.05 Em 1987 a demanda media de dia util passou para 115.000 pass./dia (periodo de Janeiro aagosto de 1987). Pesquisa realizada no dia 28.09.87, 2' feira, acusou uma quantidade total de 124.138pas sageiros. Pag. 10- Paragrafo 5.06 0 total de receitas apresentado no "Appraisal"e de US$ 18,14 milhoes. Pag. 10- Paragrafo 5.07 0 "Appraisal" previa a operagao em 1985, ccm um ni vel de demanda de 260.000 pass./dia. Os trens adquiridos (25 TUE's), emfbora -calculados com base nessa demanda (26C.000 pass./dia), nao teriam con- dig8es de atende-la, visto que, seriam necessarios 33(TUE's (vide explicagao da pig. 81 do RCP-TRENSURB). Pag. 13- Paragrafo 6.08 As restrigoes de uso do automovel foram procedidas somente junto as estaq6es Rodoviaria e Mercado. RUA ERNESTO NEUGEIAUER, 196 - FONE PAUX (06121434111 - PORTO ALEGRE * RS -28 - ANNEX 1 Page 5 of 9 t G ~rRENSURil , EMRESA D TRENS UNRANOS DE PORTO ALEGRE S. A. COC. w0710630loao s Pa8. 13- Paragrafo 7.05 A TRENSURB entende que o numero de pessoal nao es- ta elevado, principalmorite, porque se ressente de profis- sionais na area de rranutengao e opera9ao. TABELA 3 Apresentamos, em. anexo, a tabela 3, com dados de ! 1986 devidamente corrigidos, tomando como referencia o rela t6rio encaminhado pela TRENSURB. Informamos, tambem, numa terceira coluna, os dados referentes ao periodo de Janeiro a setembro de 1987. -MAPA Foram executadas algumas observagoes na copia ane- xa, RUA ERNESTO NEUGESAUER, 1965 * FONF PABX (0512 434111 - PORTO ALEGRE - RS tJ :2103.02 - 29 - ANNEX 1 Page 6 of 9 COMMENTS FROM THE BORROWER (ENGLISH VERSION) T R E N S U R B Our ref: OF-PRES-148/87 Porto Alegre October 16, 1987 Mr. Rene A. Ribi Operations Evaluation Department IBRD Washington, D.C. Dear Mr. Ribi: Subject: IBRD Loan 1839-BR Project Completion Report Thank you for the copy of the draft PCR prepared by the Bank, kindly forwarded for our perusal. As our project team members have now examined it, their comments are attached, together with a copy of the study on tariffs in the north corridor of Porto Alegre Metropolitan Region. These comments are designed to supplement a number of the points the Bank makes in the Report and to update operating figures with information on the period up to August 1987. This probably brings to a close the series of reports and commentaries exchanged between us in correspondence and in the course of successive Bank missions over the period Loan 1839-BR has been in effect, although we would of course be happy to provide any further information the Bank may require. We shall look forward to receiving a copy of the final version of the PCR. Yours, etc. /s/ M. de Castro Serafini, Jr. Managing Director -30 - ANNEX 1 (-2-) Page 7 of9 Comments on draft Project CompletIon Report dated September 11. 1987 Page 2, para. 2.02 The assumption suggested by the consultants refers to mode transfer by 10X of automobile and 50X of bus passengers. Page 2, para. 2.04 The project objectives were land-use development in the north-south corridor of the metropolitan region. Page 3, para. 3.02 In Table 1: - the figures listed are in millions of dollars and therefore do not correspond with the heading; - the correct figure for Item 2 (Train Units) is US$89,370,000; - the correct figure for Item 1 (Superstructure) is US$10,530,000; - the figure produced by the January 1983 review continues to be US$159,000,000. Page 4, para. 3.03 Ticket vending machines were not purchased. Page 7, para. 3.15 The studies commissioned from the consulting firm SAPSA were on Phase II feasibility, re-routing of bus lines, and satellite parking areas. Although the first two were completed only up to the draft report stage, the report on the third was presented in final version. The contract for the study on tariffs was not awarded to this firm (actually the successful bidder), since its financial difficulties had prevented it from submitting final reports in the two instances mentioned. The tariff study was instead carried out by GEIPOT, with support from CBTU and TRENSURB. The report in final version, presented to us in September 1987, was handed to the Bank's October 5-6 mission team. Page 7, para. 3.16 Two further studies, both conducted jointly by TRANSPLAN and Bourschied and completed in 1982, dealt with the development of commercial and service areas adjacent to stations and policy restricting automobile use. -31 - ANNEX I ( 3) Page 8 of9 The PCR does not mention che 1982 study conducted by METROPLAN on urban development along the corridor served by the metropolitan railway. Clarification of it is provided in the completion report drawn up by TRENSURB. Page 7, para. 4.01 - TRENSURB is currently operating 17.5 hours a day, from 5:30 a.m. till 11:00 p.m.; - in our view, although fewer trains are run than projected in the Appraisal Report, there is a good service level consistent with demand. Page 8, para. 4.04 After completion in July 1987 of the works to improve speed performe-ce, mean velocity increased from 41 to 46 km/hour and travel time between Mercado and Sapucaia dropped from 42 to 35 minutes. This allowed a reduction in the number of trains operated during non-rush hours from nine to seven, and during the morning rush hour from 14 to 12. without loss of headway. Durirng the evening rush period, 14 units are still run, so that there is increased service but a slight loss of headway. Page 8, para. 4.05 In 1987, mean demand per business day rose to 115,000 passengers/day (for the period January-August). A survey carried out on September 28, a Monday, showed a total of 124,138 passengers. Page 10, para. 5.06 The total estimated revenue figure given in the Appraisal Report was US$18.14 million. Page 10, para. 5.07 The Appraisal Report expected the system to have to meet an operating demand figure of 260,000 passengers/day in 1985. [It was estimated that] the number of trains purchased (25 TUE's), although geared to that figure, would not be large enough to meet it, and that the right number of TUE's would actually be 33. (For explanation, refer to page 81 of TRENSURB PCR.) Page 13, para. 6.08 Restrictions on automobile use were introduced only in the vicinity of the Rodoviaria and Mercado stations. Page 13, para. 7.05 TRENSURB takes the view that the number of staff is not high, mainly because of the shortage of professionals in the maintenance and operations sphere. - 32 - ANNEX I (4) Page 9 of 9 Table 3 We attach a version of Table 3 showing 1986 data corrected as per the report forwarded by TRENSURB. In addition, a third column provides data for the period January-September 1987. Ma A number of comments have been made directly on the attached copy of the map. - 33- ANNEX 1 Tabl* - Xo TRENSURB EMPRESA DE TRENS VRSANOS DE PORTO ALEGRE S. A. CGC. 90976853/000156- PROJECT COULPLETION REPORT BRAZIL SECOUND URBN TRANSPORT PROJECT (WAN 1839-BR) OPERATING STATISTICS ; 3 AI'I'IAI:A L A"' IIJAL ACTUAL ( 2) 1. TRAIN OP'ERATIONS (T MATE 14861 (19871 A. General Information Origin M'-n Irn StntloiF Mrr'ado Station Merefao Destination S IpUe i1i nnpueoia Sapucals Route Length :.1 ki ' 7fi.7 kim 26.7 km Average train speed 4? kn/br 41 km/hr 46 km/hr Opereting day IN hore 1i7 tioura 30 min 17 horao 30 gin Average operating days prr yrar 14no .Igyx 1113) days 300 die Unit train 4 '-in 4 en1ra 4 Cerra: Unit train passenger rnpnrtIy 1,2(1U 1 u18 1080 (6 psea/rn') Unit train energy consumption 62.4 wntt/lhr 64.7 uatt/hr 68.5 wat/hr (86) per gross ton-km B. Passeiger Traffic (per dny) 260.00(1 iw13.00 11.000 : Number of Trains Peak hour II 10 10 Off-pS k hour 4 S D. Headuay Between Trains (minutes) Peak hour r, 6 Mid-peak hour h.f 1> 12 Oft-peak hour 15.o 17 12 E. Fleet Requirements (unit trains) Average peak hour ?? 17 ll In repair/mainterinac 87 Total pr, 75 25 Avalability fi,% 7(0% 72% F. Personnel Requiremernts (starf members) Directors q 17 12 Managers I( 22 23 Supervisors 1 2828 Drivers and inspectors -rl 110 Adminlitration f.0 7'8 319 Statton Staff 4rq 438 Workshops ;^94 347 Track Maintenace an 77 Total t IjI94 1/ 1.354 Operating Statistics Number of daily train tripes 1:17194 Number of dolly unit traini trips 197 194 Daily train-km 4S.4 5.lfiS 5165 Daily unit train-km (6.'0? ' 17168 5.165 Daily gross ton-k1 (100) I1176 .140 1/ T ius figure ccimpares with a total of 1,790 staff reciorwended by the prmject consultant. The appraisal nuibers were considerably underestimatec. Source: TRENSURB March 1987 (2) Dados de Janeiro a setembro/1987 . 3 37 RUA ERNESTO NEUGEBAUER, 1985 - FONE PABX (05121434111 -PORTO ALEGRE. 1437.03.0037 \ \ fe~P~.eIeCae,asDSoI 5-.- ..~ SANTA TI~ SOT NOVHAMBURGO-r4. INDUTRIAL - k