Chile - Copper Sector Project

Document of The World Bank FOR OFFICUIL USE ONLY Report No. 5383 PROJECT COMPLETION REPORT CHILE COPPER SECTOR PROJECT (LOAN 1200-CH) December 28, 1984 Industrv Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be diaclsed without World Bank authorization.  FOR OMCUIL USE ONLY PROJECT COMPLETION REPORT CHILE: COPPER SECTOR PROJECT LOAN NUMBER 1200 CH TABLE OF CONTENTS Page No. PREFACE ...................................................... i BASIC DATA SHEET ..................................................... ii HIGHLIGHTS .......................................................,.iv I. INTRODUCTION ................................................... 1 II. PROJECT BACKGROUND .. ........................................... 1 1. Project Origin, Preparation, Appraisal, Approval and Loan Effectiveness ............................. 1 2. Project Description and Objectives .......................... 2 3. Amendments in Project Scope . .......... 3 4. Loan Cancellations .......................................... 5 III. PROJECT IMPLEMENTATION AND MANAGEMENT .......................... 5 1. Achievement of Project Objectives ........................... 5 2. Project Management .......................................... 7 3. Use and Performance of Consultants .......................... 7 4. Implementation Schedules .................................... 8 5. Procurement and Performance of Suppliers and Contractors .............................................. 10 6. Costs, Financing and Disbursements .......................... 11 IV. OPERATING PERFORMANCE ......................................... 17 1. Commissioning and Start-Up ................................. 17 2. Environmental Aspects ...................................... 18 V. FINANCIAL AND ECONOMIC PERFORMANCE ............................ 18 1. Financial and Economic Rates of Return ..................... 18 2. Financial Covenants ........................................ 22 3. Economic Benefits .......................................... 22 VI. INSTITUTIONAL PERFORMANCE ..................................... 23 1. Development of Management and Organizational Effectiveness ............................... 23 | 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.  VII. BANK PERFORMANCE .................... .. &.......... .. . ...... . . 23 VIII. CONLSONSC.........................N S.. 24 1. Overall Assessment ............................ ......... 24 2. Lessons for Similar Projects.............................. 25 TABLES Implementation Schedule ................................ .9 Sources of Procurement ......... .. .................. 11 Summary of Project Costs .................... *........... S. .. 12 Analysis of Cost Variance - El Teniente Mine Project ......... 13 Project Financing ... ................................. 15 Allocation of Bank Financing ................................. 16 Disbursement Experience ..16 Financial Rates of Return .. ....... ........ ........ *. 19 ANNEXES 1. Project Implementation Units 2. Implementation Schedules 3. El Teniente Project - IRR Calculations ATTACHMENT Comments from the Executing Agencies  -i- PROJECT COMPLETION REPORT CHILE COPPER SECTOR PROJECT IOAN 1200-CH PREFACE 1. This is the Project Completion Report (PCR) on the Chile, Copper Sector Project for which a Bank loan in the amount of US$33.0 million was approved in February 3, 1976 to the Government of Chile to be onlent to Corporacion Nacional del Cobre (CODELCO) and Empresa Nacional de Mineria (ENAHI). Later on in September 1978, an amendment to the contract was proposed by the Government modifying the project scope to include a subproject to be executed by Empresa Nacional de Electricidad S.A. (ENDESA) and reducing the loan from US$33.0 to US$27.8 million. Further cancellations reduced the Bank loan to US$25.5 million. 2. Following the closing of the loan in June 30, 1982, IND staff visited Chile and had discussions with the Government and the individual companies in charge of the implementation of the subprojects for the preparation of this PCR. The report reflects findings of this mission as well as a review of project files and related documents. 3. ENAMI offered specific comments and clarifications which were taken into account in finalizing the report and are reproduced in the attachment to the report. 4. This project has not been audited by the Operations Evaluation Department.  - ii - PROJECT PERFORHANCE - BASIC DATA SHEET CHILE, COPPER SECTOR PROJECT LOAN 1200-CH I. KEY PROJECT DATA Amounts (USS million) Cancelled After As of July 31. 1983 Original Amended Amended Disbursed a/ Repaid Outstanding 33.0 27.8 2.3 25.5 6.7 15.3 b/ Cumulative Loan Disbursement 1976 1977 1978 1979 1980 1981 1982 Planned 2.0 17.0 30.0 33.0 33.0 33.0 3i.o Amended - 0.6 7.0 19.3 27.4 27.8 27.8 Actual - 0.6 4.7 8.7 15.9 23.7 25.5 Actual as Z of Amended - 100.0 67.1 45.1 58.0 85.3 91.7 II. OTHER PROJECT DATA First Mention in Project Files May 3. 1974 Negotiations December 15-19, 1975 Board Approval February 3, 1976 Loan Agreement Date March 24, 1976 Effective Date June 15. 1976 Closing Date June 30, 1982 c/ Loan Amendment Date September 28, T978 Borrower: Republic of Chile Executing Agencies: Corporacion Nacional del Cobre de Chile (CODELCO) Empresa Macional de Mineria (ENAMI) Empresa Nacional de Electricidad (ENDESA) Fiscal Year of Borrower: Calendar Year Loan Number: 1200-CH Amount (Original) USS33.00 million (Amended) USS27.7S million (Actual Disbursed) USS25.47 million III. MISSION DATA No. No. of and Man Report Sent by Month/Year Weeks Persons Weeks Date Appraisal IND/LCI 8/74 2 3 6 10/02/74 Post-Appraisal IND/LCP/LCl 1/75 1 3 3 03/07/75 Post-Appraisal IND/LCI 7/75 1 2 2 08/04/75 Post-Appraisal IND 9/75 2 1 2 10/09175 Supervision LCI 2/76 1 1 1 03/08/76 Supervision IND 4/76 1 2 2 04/14/76 Supervision IND 11/76 1 2 2 11/12/76 Supervision IND/LCI 1/77 2 3 6 02/10/77 Supervision IND 5/77 2 2 4 06/14/77 Supervision LCI 11/77 2 1 2 12/01/77 Supervision IND 5/78 1 2 2 06/02/78 Supervision LCP 4/79 1 1 1 01/02/79 Supervision LCP 5/79 1 1 1 08/08/79 Supervision LCP 3/80 1 1 1 04/04/80 Supervision IND 6180 2 2 4 07/31/80 Completion IND 11/82 2 1 2 12/16/83 a/ Following loan cancellations were made: (i) as a result of contract amendment. USS5.21 million (Sep. 1978) (ii) ENDESA change of sub-project scope US$1.30 million (May 1982) (iii) at closing date USS1.02 million (Nov. 1982) b/ After minus $3.5 million foreign exchange adjustment. c/ Original closing date September 30, 1979, amended to December 31, 19BO and extended to December 31, 1981 and June 30, 1982.  - iii - CURRENCY EQUIVALENTS At appraisal US$1 - 1,050 Escudos (ED) December 1975 US$1 - 8.5 Pesos (Ch$) (In October 1975, the Escudo was replaced by the Peso, Ch$1 - E1,000) Average 1976 US$1 - 13.05 Pesos Average 1978 US$1 - 31.66 Pesos Average 1980 US$1 39.00 Pesos Average 1982 US$1 - 50.91 Pesos October 1983 US$1 - 80.80 Pesos WEIGHTS AND MEASURES 1 metric ton = 1,000 kilogram 1 metric ton 2,205 pounds 1 pound - 0.454 kilograsn 1 kilometer = 0.62 miles ABBREVIATIONS CODELCO - Corporacion Nacional del Cobre de Chile ENAMI - Empresa Nacional de Mineria ENDESA - Empresa Nacional de Electricidad S.A. Fiscal Year = January 1 to December 31  PROJECT COMPLETION REPORT CHILE - COPPER SECTOR PROJECT LOAN 1200-CH HIGHLIGIITS 1. Chile's copper mining industry has continued to play a dominant role in its economy. It is the single most important contributor to exports, accounting in 1982 for approximately 44% of total merchandise exports. During 1982 copper exports amounted to about 7% of GNP. 2. The Chile Copper Sector project was identified by sector missions which visited Chile in July 1973 and January 1974. By that time, CODELCO and ENAMI, which accounted for about 95% of the country's production of copper, had put together investment programs intended to solve some production bottlenecks, to modernize existing facilities, and also aimed to increase Chile's productive capacity from 750,000 tons to a level of 1.0 million tons by 1980. The project was appraised in October 1974 and four subprojects from the CODELCO and ENAMI investment programs and a technical assistance program were selected for Bank assistance. The Board approved on February 3, 1976 a Bank loan to the Government of Chile for US$33 million repayable in 15 years with a grace period of 3.5 years bearing an interest of 8.5% p.a. The Government was to onlend the proceeds of the loan to CODELCO (US$29.54 million) and to ENAMI (US$3.46 million) with the same maturity and grace periods as the Bank loan to Chile, but bearing an annual interest rate of 10%. The loan became effective on June 15, 1976. 3. The subprojects originally selected for Bank financing were the following (para. 2.02).1/ i. Construction of an underground ore crushing system at CODELCO's El Teniente mine; ii. Rehabilitation of CODELCO's Barquito thermal power plant; iii. Construction of a Selenium Recovery Plant at CODELCO's Chuquicamata mining complex; iv. Modernization of ENAMI smelters including the replacement of a cathode starter sheet machine at Ventanas smelter, the replacement of a blister copper casting machine at Paipote smelter and the modification of the burners in both ENANI's smelters to enable them to operate on coal as well as petroleum products; and v. A technical assistance for CODELCO and ENAMI which included: development of planning systems, project preparation and evaluation procedures, specific technical and feasibility studies and the contracting of specific experts. 1/ Refer to paras. in main report.  v - 4. During the early months of project implementation good progress was made by CODELCO in undertaking the major corporate reorganization agreed with the Bank, but execution of CODELCO and ENAMI subprojects was slower than expected. In mid-1978 the Government proposed a change in the project scope (paras. 2.03 and 2.04). Instead of rehabilitating the Barquito thermal power plant as originally envisaged under the project, the Government proposed that the supply of the El Salvador mine's future electricity needs be made by linking the mine with the Central Interconnected System (CIS), the country's main power grid which had been extended to Northern Chile in 1977/78. The cost of the interconnection subproject was estimated at US$22.9 million with a foreign exchange component of US$10.5 million. The Government proposed to cover th-1 foreign costs by onlending to ENDESA US$10.5 million of the US$16.3 million originally earmarked for CODELCO to carry out the Barquito subproject. Of the US$5.8 million in foreign exchange savings realized by this modification, the Government requested that US$5.21 million be cancelled and that the remaining US$590,000 be onlent to ENAMI to finance cost increases in equipment and the acquisition of a new molding furnace. 5. The Government also requested a relatively minor change in the project's scope. It consisted of eliminating from the project, the construction by CODELCO of the Selenium Recovery Plant which is going to be built and owned by private interests. With these changes the amended project scope was defined as follows: i. CODELCO's El Teniente underground crushing system; ii. Modernization of ENAMI, Ventanas and Paipote smelters; iii. ENDESA interconnection; and iv. CODELCO and ENAMI technical assistance. 6. Total cost of the original project was estimated at appraisal (March 1976) to be US$76.8 million. After the amendment of the project scope (September 1978) this estimate was US$84.0 million. At project completion (December 1982) the cost of the project was estimated at US$147.7 million; twice as much as the amended estimate (para. 3.16). Major changes in cost occurred in the local components of the subprojects mainly due to extended delays in project implementation and the higher than projected local inflation which in terms of US dollar equivalent was magnified by the fact that a fixed exchange rate of 39 pesos to the US dollar was maintained for a period of almost three years (1979-81). Other factors such as the poor ground conditions encountered at El Teniente crushing system and required changes in design and specifications also contributed to the increase in costs (para. 3.10). 7. The loan had three different partial cancellations. The first, already mentioned, was a result of the amendment to the contract in September 28, 1978 for an amount equivalent to US$5.21 million. The second in May 4, 1982 for US$1.3 million was the result of ENDESA changes in the scope of the project. The third, in the amount of US$1.018 million in November 30, 1982, was the undisbursed balance at the time of closing the  - vi - loan. Total cancellations amounted to US$7.5 million reducing the original loan from US$33.0 to US$25.5 million. All other funds required to complete the project (US$122.2 million) were provided by CODELCO (70%), ENDESA (25%) and ENAMI (5%) (para. 3.20). 8. During appraisal of the project it was estimated that all subprojects would be executed within about three years from early 1976 through early 1979 (para. 3.09). Changes in scope and unforeseen obstacles delayed the project till the end of 1982 taking in total seven years for completion. Project management was organized by the individual companies and staffed mainly with their own personnel (para. 3.06). This arrangement has proved to be successful and has permitted, in particular with CODELCO, to confirm its technical capabilities in handling large ari complex projects. Contractors and consultants performed well and with a few exceptions, there were no major problems with suppliers. 9. With the exception of some initial problems to start-up the new burners in the ENAMI's smelters, the commissioning of all other subprojects went reasonably well and no major problems were encountered (paras. 4.01 to 4.05). Minor problems caused by equipment failure or lack of experience were rectified with minimum delay by the staff of the individual companies. Despite the fact of considerable cost overruns and delays occurred during construction, overall, the project was successfully completed and the basic technical and economic objectives were achieved. 10. The rates of return calculated at project completion are as high as estimated at appraisal except in the case of the El Teniente underground crushing system where due to much lower copper prices than originally projected the return is lower than expected but still in the order of 21% before tax (paras. 5.01 and 5.02). II. The Bank's relationship with the Government and CODELCO was varied reflecting shifts in Government policies and priorities and changes in CODELCO's managerial staff. As a result, discussions during appraisal to define the project scope were lengthy and time consuming. Later on, the need for changes to the project scope remained as one of the major issues. 12. Bank involvement in this project was justified on the basis that this loan would be the first of a series of operations, which would enable establishing a dialogue on policy matters. However, since the late 1970s Bank operations in Chile were considerably reduced and no follow-up operations were considered. More specifically, on project grounds, a lesson learned is that dealing with only one of the departments of a large corporation such as CODELCO and financing only a small portion of a large investment program, proved to be an ineffective way of Bank participation. In the future, when a large corporation having a dominant role in the sector is involved, the Bank should seek to establish a more direct linkage, through traditional project lending with adequately sized loans to enable a constructive dialogue on sectoral policy issues.  PROJECT COMPLETION REPORT CHILE - COPPER SECTOR PROJECT (LOAN 1200-CH) 1. INTRODUCTION 1.01 Chile's copper mining industry has continued to play a dominant role in its economy. It is the single most important contributor to exports, accounting in 1982 for approximately 44% of total merchandise exports. During 1982 copper exports amounted to about 7Z of GNP. The country's total copper production, which remained stagnant during the period 1977-1981 at a level slightly exceeding 1.0 million tons of copper content, reached a level of 1.25 million tons in 1982, with the coming on stream of a few small projects implemented to solve bottle necks in the operations. About 85% of the copper output comes from the large state mines run by the Chile National Copper Corporation (CODELCO), which has become the world largest copper producer outside the USSR. The balance of the copper output is produced by medium and small-scale mines, and most of this production is purchased and processed by the state-owned National Mining Company (ENAMI) in its ore processing plants, and smelting and refining operations. 1.02 The Copper Sector Project is the first and only Bank involvement in the Chilean copper industry, excluding IFC. The project originally consisted of several subprojects, to be undertaken by CODELCO and ENAMI, directed primarily to maintain copper output and to reduce operating costs. In addition to the specific subprojects, the loan included financing for technical assistance aimed at strengthening the management of the two state enterprises in the sector. -I. PROJECT BACKGROUND 1. Project Origin, Preparation, Appraisal, Approval and Loan Effectiveness. 2.01 The project was identified by sector missions which visited Chile in July 1973 and January 1974. By that time, CODELCO and ENAMI, which accounted for about 95% of the country's production of copper, had put together investmert programs intended to solve some production bottle-necks, to modernize existing facilities, and also aimed at increasing Chile's productive capacity from 750,000 t.p.a. to a level of 1.0 million t.p.a. by 1980. The project was appraised in October 1974 and four sub-projects from the investment programs were selected for Bank assistance, based primarily on the need to maintain existing capacity, improve production efficiency and reduce operating costs. The appraisal was followed by three post-appraisal missions in January, July and September 1975. The major issue was CODELCO's and ENAMI's wishes to constantly redefine the sub-projects to be included in the loan. Finally agreement was reached on project scope, and the Board approved on February 3, 1976 a Bank loan to the Government of Chile for US$33  -2- million repayable in 15 years with a grace period of 3.5 years bearing an interest of 8.5% p.a. The Government was to onlend the proceeds of the loan to CODELCO (US$29.54 millions) and to ENAMI (US$3.46 millions) with the same maturity and grace periods as the Bank loan to Chile, but bearing an annual interest rate of 10%. The loan became effective on June 15, 1976. 2. Project Description and Objectives 2.02 The project consisted of several distinctly separate sub-projects including some technical assistance. Total project cost was estimated at US$76.8 million. The original subprojects, which are self contained and independent of the rest of the investment programs, were the following. CODELCO Projects (a) Construction of an underground ore crushing system at CODELCO's El Teniente mine to handle the different and bulkier type of ore which needs to be mined in order to maintain the production levels of fine copper (total cost: US$41.7 million). (b) Rehabilitation of CODELCO's Barquito Power Plant which supplies the El Salvador mining, smelting and refining facilities. This sub-project was intended to reduce the risks of power failure, and the consequent loss of production, as well as result in fuel and maintenance cost savings (total cost: US$20.4 million). (c) Construction of a Selenium Recovery Plant at CODELCO's Chuquicamata mining complex aimed at the recovery of this metal from the tailings of the Chuquicamata and El Salvador copper refineries (total cost: US$2.3 million). (d) A technical assistance program for CODELCO which included: (i) development of planning systems, project preparation and evaluation techniques, (ii) specific studies for planning purposes, (iii) specific experts, (iv) organization and structuring of the CODELCO group, and (v) optimization of internal procedures (total cost: US$8.0 million). ENAMr Projects (a) Modernization of ENAMI smelters including: (i) replacement of a cathode starter sheet machine at ENAMI's Ventanas smelter, (ii) installing a new anode casting machine at ENAMI's Paipote smelter and (iii) modification of the burners in both ENAMI's smelters to enable them to operate on coal as well as petroleum products (total cost: USS3.9 million).  -3- (b) A technical assistance program for ENAMI, which included: ti) planning and project preparation procedures, (ii) specific technicai and economic feasibility studies, and (iii) assistance to diversify its mineral production (total cost: US$0.5 million). 3. Amendments in Project Scope 2.03 Up to mid 1978, project execution was slower than expected. Good progress had beer, made by CODELCO in undertaking the major corporate reorganization agreed with the Bank, which was successfully completed by end 1976. However, the short term effect of the reorganization, which was accompanied by important personnel changes, was to slow down temporarily physical execution of the project. By that time, the Government proposed a major modification of project scope which consisted of (i) replacing the rehabilitation of the Barquito Power Plant subproject by a transmission line to the country's main power grid, (ii) eliminating the selenium recovery plant subproject,'/ and (iii) some minor modifications to the scope of the technical assistance components. These modifications were agreed with the Bank and an amendment was approved by the Board on September 1978. This meant the inclusion of a third beneficiary, ENDESA and the loan was restructured as follows: Bank Loan (US$ million) Before After Amendment Amendment CODELCO 29.5 12.5 ENAMI 3.5 4.8 ENDESA - 10.5 33.0 27.8 a/ a/ US$5.21 million was cancelled (para. 2.04). 2.04 The cost of the ENDESA interconnection subproject, net of taxes and duties, was estimated at US$22.9 million with a foreign exchange component of US$10.5 million (versus the original Barquito Plant total cost and foreign exchange of US$20.4 million and US$16.3 million, respectively). The Government proposed to cover the foreign costs by cn-lending to ENDESA US$10.5 million of the US$16.3 million originally earmarked for CODELCO under Loan 1200-CH to carry out the Barquito subproject. ENDESA was to finance the local cost of the interconnection, equivalent to about US$12.4 million, from internally generated funds and by other borrowing. Of the US$5.8 million in foreign exchange savings realized by this modification, the Government requested that US$5.21 million be cancelled and that the remaining US$590.000 be on-lent to ENAMI to (i) finance a complementary molding furnace for the Paipote smelter and {ii) to cover increased dollar costs of equipment. The new furnace enables 1/ Due to a change in Government policy, this project was now to be undertaken by the private sector.  -4- ENAMI to produce higher quality processed copper at the smelter and to realize important savings in fuel and electricity. This and other adjustments, meant that the net allocation to ENAMI was increased by US$1.31 million. 2.05 The ENDESA Power Interconnection subproject consisted of the following works: (i) acquisition of fixed and portable telecommunication equipment for the Cerro Navia - Maitencillo transmission line; (ii) construction of about 280 km of a single circuit 220 kV transmission line from Maiteucillo to Diego de Almagro; (iii) construction of a new substation at Diego de Almagro 220/110/80/23 kV with a transformer capacity of about 200 MVA; and (iv) constructioa of a new substation at Chanaral with a transformer capacity of about 8 MVA. The objectives of the ENDESA Project were: (i) to provide reliable electric power to the El Salvador mine, and io substitute hydro and coal base generation for costly oil-based gi-r-ration, by linking the mine to the Central Interconnected System (CT--); and (ii) to improve electricity service to other users in the region between Maitencillo and El Salvador by constructing a new supply point at the Chanaral substation. 2.06 Interconnecting the mfne with the Central Interconnected System (CIS) was not feasible at the time of the project appraisal because the northern most terminus of the CIS was located about 900 km south of the miane. In 1977, however, ENThSA decided to extend its grid 630 km north with a 220 kV transmission line from Cerro Navia, near Santiago, to Maitencillo, located 280 kt south of the El Salvador mine (Diego de Almagro substation). The investme.-- cost of connecting the mine to the CIS was estimated to be US$2.5 m. ion (12%) higher than the investment cost originally foreseen for Ba-quito's rehabilitation, however much lower operating and maintenance costs made the interconnection a more attractive alternative. Specifically, tapping into the CIS' low cost electricity- much of it generated in large hydro plants in central and southern Chile- would permit considerable operating and maintenance savings. Apart from the advantages to CODELCO, the proposed interconnection has the added benefit of permitting ENDESA to provide reliable electric power to other users in the region between Maitencillo and El Salvador, 'ncluding some 15 small- and medium-sized mines. Most of the region was served by isolated and inadequate small thermal units--or not at all. 2.07 In 1980 ENDESA decided on increasing the tension in the Diego de Almagro Substation from 80 kV to 110 kV representing a reduction in investment costs in the order of US$0.7 million. However the subproject scope was later increased by adding two 22mw gas turbines as a back-up to  -5- the transmission line to supply energy to the El Salvador mine. This represented additional costs of US$6.5 million. In March 1982, ENDESA decided to cancel the construction of the Chanaral Substation because of lower than expected electricity demand in the area to be served. This cancellation represented an additional US$0.7 million reduction of the ENDESA investment plan. 4. Loan Cancellations 2.08 The loan had three different partial cancellations. The first one was a result of the amendment to the contract in September 28, 1978 for an anount equivalent to US$5.21 million. The second cancellation in May 4, 1982 for US$1.3 million was the result of ENDESA changes in project scope (para. 2.07) and lower than expected cost of major equipment (US$1.7 million) as explained in para. 3.19. The third cancellation in the amount of US$1.018 million in November 30, 1982, was the undisbursed balance at the time of closing the loan. Total cancellations amounted to US$7.528 million, reducing the original loan from US$33.0 to about US$25.5 million. III. PROJECT IMPLEMENTATION AND MANAGEMENT 1. Achievement of Project Objectives 3.01 All sub-projects comprising the Chile Copper Sector Project have been satisfactorily completed and its objectives, as described in paras. 2.02 and 2.05 fully achieved. 3.02 The underground crushing project at CODELCO's El Teniente mine was initiated in 1975 with the selection of the consulting firm to develop the engineering of the project. After six years of construction, including delays attributable to unforeseen poor ground conditions, the project was completed in March 1982 at a cost of US$89.5 million or Shout 2.1 times higher than the estimate at appraisal (US$41.7 millitn). The cost overruns which occurred in the local component were mainly due to delays in construction and to an underestimated local inflation. The main objective of the project, to avoid a drop in production and maintain its level of 250,000 tons per year of fine copper output was achieved. Over a period of more than 70 years, the El Teniente mine had produced from the secondary enriched zone using highly efficient block caving techniques, although it had been evident that the mine would have to produce eventually less efficiently from the underlying primary ore zone, a point which now is being reached. This is expressed in a drop of productivity from 250 tons per man shift to 25 tons in the primary zone, and would, in the absence of actiozn to introduce the underground crushing system, increasingly translate into both costlier mining and loss of production. In terms of ore and metals production, CODELCO estimates that the direct benefits of the El Teniente project for the period 1983-1992 range from 34,464 to 71,504 tons of incremental copper in concentrates produced per year as shown in the table below.  -6- El Teniente Project Incremental Copper Production Ore Production (Hillion tons/year) Incremental Yearly Production With Without Tons of Copper Tons of Molybdenum Year Project Project in Concentrates a/ in Concentrates 1983 22.8 19.9 34,464 364 1984 22.8 18.6 48,929 536 1985 22.8 18.4 52,704 550 1986 22.8 17.8 65,259 608 1987 22.8 17.6 71,504 625 1988 22.4 17.2 55,484 636 1989 22.4 17.4 42,233 606 1990 22.4 17.4 43,924 606 1991 22.4 17.0 40,426 657 1992 22.4 16.7 45,520 691 a/ Average incremental copper in concentrates production, 50,045 tons per year. Source: CODELCO 3.03 The re'&abilitation of the ENAMI's smelters consisted of the replacement of antiquated and obsolete equipment components at ENAMI's Ventanas and Paipote Smelters - including (i) a cathode copper starting sheet machine for Ventanas, (ii) an anode copper casting machine for Paipote and (iii) the modification for both smelters' burners to enable them to operate on coal as weU as petroleum products. Replacement of the equipment was needed to avert breakdowns and production losses and to lower maintenance costs. Modifications of the burners enabled important energy saving though the substitution of coal for petroleum products. The Ventanas' starting sheet machine component was completed in March 1980, the Paipote's casting machine and a new molding furnace were completed in September 1980 and the modification of the burners in December 1980. All components are operating as planned. The total final cost of the rehabilitation of ENAMI's smelters and refinery amounted to US$10.3 million, a value substantially higher than the original estimate of US$5.5 million. This was due to important changes in project scope. Major installations, such as a complete fire-refir.ing and anode casting plant for Paipote and a cathode starting sheet facility for Ventanas, led to significant extensions and other improvements in process capabilities. Major investments derived in higher freight, import duties and related costs, as well as larger construction projects; all this required extended implementation schedules. Costs were also affected by unpredicted high local inflation, which was not offset by adequate currency adjustments. 3.04 The main objective of the ENDESA Project, to interconnect the El Salvador's mine with the CIS, was achieved. In fact, the interconnection was commissioned in mid-1981, and in 1982 energy supplied by the CTS  -7- replaced about 85% of the thermal generation in the EL Salvador wine. Construction of the Chanaral substation was postponed as electricity demand in the region between Maitencillo and El Salvador did not increase as originally envisaged, and a new supply point at Chanaral was no longer necessary. Microwave and telephone equipment included originally as part of the communication facilities for operation and maintenance of the transmission system were not acquired, as the Government decided that ENDESA should procure those services from telecommunication companies serving the region. Installation of the power line carrier equipment is not complete yet. 3.05 Under the technical assistance component of the project both CODELCO and ENAMI have made extensive and satisfactory use of consultants. An important contribution to CODELCO was the implementation of a planning group to develop new capacities in medium and long range planning not only for the mining operations but for the whole corporation as well. 2. Project Management 3.06 In view of the available expertise in CODELCO, ENAMI and ENDESA the arrangements for implementing the subprojects came under the direction of the engineering, construction, and operating staff of the individual companies. In all cases, foreign experts or consultants assisted in project implementation. The technical assistance programs were administered through detailed terms of reference and reporting requirements acceptable to the Bank. The concept of assigning teams made up of foreign and local counterpart staff, under the direction of a project manager, was used. 3.07 At El Teniente, the project management function was initially divided between the project administrator responsible for engineering, consultants and foreign procurement, and the mine superintendent responsible for mine engineering, local procurement and construction in addition to production. This division of responsibilities was not conducive to an efficient management of the project. With the experience gained, El Teniente administration decided to modify the organization structure centralizing all functions in the project administrator, reporting to a development/engineering function created separately from production. This resulted in a more satisfactory arrangement. At ENAMI, the subprojects were centrally managed by the engineering department, with coordinators appointed by the smelter's administration to facilitate the interaction between project management teams and production activities. This scheme worked well. ENDESA, having a long tradition for implementing large and complex projects, used its standard project management arrangement, with reponsibilities centralized in its engineering construction departments. Annex 1 shows details of the organigrams of the Project Implementation Units for CODELCO and ENAMI. 3. Use and Performance of Consultants 3.08 Foreign consultants played a significant role in the successful implementation of the project. The following specific studies were completed with IBRD financing; in all cases the performance of consultants has been fully satisfactory.  -8- - Corporate, Organization and Planning study to make CODELCO and lts mining companies a single and fully integrated corporation. - A long-term mining plan for CODELCO's EL Teniente mine, taking into account the difficult mining conditions being encountered with the present methods as production moves into harder rock. - A long-term mining plan for the CODELCO Chuquicamata mine, and a feasibility study for the expansion of the concentrator. - Modernization and Expansion Study of ENAMI's Ventanas and Paipote smelters. - Technical and economical feasibility of (a) the use of timber, natural gas, or other hydrocarbons in the reduction of copper oxides during the process of fire refining at ENAMI's Ventanas smelters; tb) the agglomeration of copper cement for use in converter feed in ENAKI's Paipote and Ventanas smelters; and (c) the use of coal-oxygen burners in the ENAMI's reverbatory furnaces. CODELCO's foreign expenditures in consultants amounted to US$5.7 million, ENAMI's foreign cost for technical assistance costs amounted to US$0.2 million making US$5.9 million the total amount financed by IBRD. ENDESA did not have a technical assistance component. 4. Implementation Schedules 3.09 During appraisal it was estimated that all sub-projects should be executed within about three years from early 1976 through early 1979. Changes in scope and unforeseen obstacles delayed the project till the end of 1982 taking in total seven years for completion. Original and actual project implementation schedules are as follows:  -9- Implementation Schedule Completion Dates Appraisal Delays Sub Project Estimate Actual (months) CODELCO Projects: (a) El Teniente Crushing Plant July 1978 Sept. 1982 50 (b) Barquito Power Plant Early 1979 Cancelled - (c) Chuquicamata Selenium Plant July 1978 Cancelled - (d) CODELCO - Technical Assistance Sept. 1978 Dec. 1981 39 ENAMI Projects: (e) Smelters modernization June 1978 Jan. 1980- 19 (f) ENAMI - Technical Assistance Nov. 1978 Sept.1982 46 ENDESA Projects: (g) Interconnection System - Diego de Almagro Sub Station June 1981 August 1981 2 - Maintencillo Transmission Line June 1981 August 1981 2 - Communications June 1981 June 1982 12 a/ a/ Partial completion, about 40Z of already purchased equipment has not been installed. 3.10 Major delays occurred during the construction of the El Teniente underground crusher mainly due to (i) unforeseen weak ground conditions encountered when excavating the main chambers and ore passes and (ii) changes in the basic design. The weak ground conditions required much more than expected steel and concrete support and slowed considerably the speed of construction. Serious difficulties were also encountered in the supply of construction materials from the surface stations to the underground working areas since operators gave higher priority to the handling of ores and materials related to ore production. Major changes in the basic design of the system were required to accommodate a larger than originally planned crusher. The installed crusher produces blocks of about 10 inches instead of 7 inches as originally planned. The ore bins and ore passes were also enlarged. The rail tracks were changed from 30" to 56 1/2" gauge to provide better stability to the ore wagons. The feeder system was changed from vibratory feeders to pan type feeders (APRON). In total, the El Teniente underground crushing system required the excavation of some 310,000 cubic meters of rock, 5,000 tons of steel and 23,000 cubic meters of concrete. Details of construction schedule are shown in Annex 2.  -10- 3.11 The modernization of ENAMI's smelters was delayed because of difficulties in formalizing the contract with the supplier of the casting machine, unexpected start-up problems in the new burners at Paipote and Ventanas and the extension of the Paipote subproject with the addition of a fire-refining furnace not included in the original project. 3.12 The major components of the ENDESA interconnection system project were completed just two months behind schedule (as compared to schedule included in ENDESA's Progress Report No. 1). The communication system was partially installed about one year behind schedule due to delays in procuring goods and services and technical problems with the power line carrier. About 40% of the telecommunication equipment already purchased has not been installed yet, awaiting a solution of the problems mentioned above. 5. Procurement and Performance of Suppliers and Contractors 3.13 Equipmertt packages financed under Bank loan were procured by ICB and limited international tendering. Consultant services were selected in accordance with the Bank's guidelines. In the early stages, procurement was slow mainly due to the fact that CODELCO and ENAMI were not familiar with Bank guidelines (para. 3.23-ii). Suppliers from the United States, with US$11.9 million or 46.7% of the total loan funds, were the most successful suppliers of enuipment and services. In particular, they were awarded most of the consultants' services, about US$5.5 million out of a total of US$5.9 million. Equipment supplied by the US comprised mainly mining equipment including railway accessories (about US$0.8 million), hopper installation for the El Teniente crusher (about US$0.7 million), electrical equipment such as transformers, switchgears and substations, and diverse auxiliary equipment. The F. R. Germany supplied railroad cars (about US$3.1 million) and auxiliary mining equipment and Japan supplied the cathode starting sheet machine. Eleven other countries, including Chile, which supplied electrical equipment, make up the balance equivalent to about US$8.0 million or 31.3% of total loan funds. A detail breakdown of sources of procurement is shown in the following table.  -11- Sources of Procurement a/ Amount Country of Origin USS million % United States 11.9 46.7 F.R. Germany 3.6 14.2 Japan 2.0 7.8 Spain 1.4 5.5 Finland 1.4 5.5 Brazil 1.2 4.7 Italy 1.2 4.7 Chile 0.8 3.1 Sweden 0.8 3.1 France 0.5 2.0 Canada 0.3 1.1 United Kingdom 0.2 0.8 Switzerland 0.1 0.4 South Africa 0.1 0.4 25.5 100.0 a/ Based on awards and respective disbursements of Bank loan. 3.14 In general, the performance of suppliers was acceptable with the exception of the supply of railroad cars and ventilation fans. The original contract to supply the cars had to be cancelled because of the manufacturers'financial problems and imminent bankruptcy. After being prequalified, the firm presenting the lowest evaluated bid faced a rapid deterioration of its financial position to the point of not being able to comply with the required guarantees. Under these circumstances, CODELCO decided to award the contract to the second lowest evaluated firm. The supply of ventilation fans for the underground works faced an 8 month delay in manufacturing. 3.15 Most of the implementation of the project was executed by the operations departments of the individual companies and the few contractors that participated in the project performed satisfactory. 6. Costs, Financing and Disbursements Costs 3.16 At appraisal, the Chile Copper Sector Project was estimated to cost US$76.8 million including physical and price contingencies and working capital. At the time of the Amendment in September 1978 the full completion of the project was estimated at US$84.0 million, reflecting the change in scope as well as a revision of ENAMI's subprojects (para. 2.03 to 2.07). At closing date in June 1982, when the project had been in practice fully completed, the total cost was calculated at US$147.7  -12- million, representing a 76Z cost overrun over the revised estimate. A summary comparison of the project cost at these three stages is given in the following table. Summary of Project Costs (US$ million) a/ Appraisal After Estimate Amendment Actual (March 1976) (Sept. 1978) (Dec. 1982) CODELCO Projects: (a) El Teniente Crusher System 41.7 41.7 89.5 (b) Barquito Power Plant 20.4 - - (c) Chuquicamata Selenium Plant 2.3 - - (d) Technical Assistance 8.0 8.0 8.7 Total CODELCO Projects 72.4 49.7 98.2 ENAMI Projects: (e) Ventanas/Paipote Smelters Modernization 3.9 5.5 10.3 (f) Technical Assistance 0.5 0.5 0.3 Total ENAMI Projects 4.4 6.0 10.6 ENDESA Projects: (g) El Salvador, Power Interconnection - 28.3 38.9 TOTAL Copper Sector Project 76.8 84.0 147.7 a/ Includes physical and price contingencies and working capital. 3.17 Major overruns occurred in all subprojects with the exception of the rechnical assistance component. The CODELCO's El Teniente crusher subproject actual cost (US$89.5 million) is about 2.1 times the original estimate. The following table summarizes the major changes and modifications and their effects in the cost of this subproject. The most significant increases were: (i) in the construction of the underground primary crushing station (ii) additional engineering and project management and (iii) problems of higher than expected local inflation rates applied over an extended period and an overvalued Chilean peso.  -13- El Teniente Underground Crusher Analysis of Cost Variances Cost Estimate (USS million) Item Change Original Actual A. Rail Transportation System (Level 1) Exchange station 0.7 1.9 added. B. Dumping Station (Level 1 Retram) New 3,500 ton ore 0.4 3.8 bin was required. C. Vertical Ore Shoots (to Crusher) Additional concrete 2.1 7.5 steel reinforcing works was required (240 mts). D. Rail Transportation System (Level 5) 6.0 km of additional 5.4 15.7 lines. E. Dumping Station (Level 5) Ore bin added. 0.6 2.9 F. Underground Primary Crushing Station Larger chamber and 7.8 17.4 additional ground support works. G. Ore Shoot System (to Level 6) Additional concrete 2.4 4.5 and steel support works. R. Feeding System (to Level 8) Larger than estimated 8.0 15.1 ore bins. I. Various Additional Engineer- 3.8 20.7 - Construction and Electrical ing and Project Equipment Management. New - Ventilation ventilation system. - Consulting Services Larger facilities (workshops, warehouses, and offices). J. Contingencies 9.5 - Total El Teniente Subproject Cost 40.7 89.5 3.18 Total actual cost of the ENAMI subprojects in the Ventanas and Paipote smelters amounted US$10.3 million or 1.9 times the amended estimate. Foreign cost of equipment procured to replace the cathode starter sheet machine, the new molding furnace, the anodes casting machine  -14- and the burners were in lire with original estimates. Cost overruns are the result mainly of high local inflation, overvalued Chilean peso, longer than expected construction period and some problems to start up the burners (para. 4.03). 3.19 Total cost for the ENDESA interconnection subproject was US$38.9 million, representing a cost overrun of US$10.6 million or about 37% over the original estimate of US$28.3 million. The foreign exchange cost was US$12.6 million, about 20% over the original estimate. About US$4 million of the cost overrun was caused by a net increase in base costs, either increased work quantities or unit costs; and the balance, US$6.6 million was caused by higher than expected price escalation, mainly under- estimation of price escalation provisions compounded by delays in certain activities during project execution. This was partially offset by some reduction (US$1.7 million) in the cost of major equipment as shown in the following table. ENDESA - Reduction in Cost of Major Equipment (US$ million) Original Actual Item Estimate Cost Savings Structures 1.4 1.0 0.4 Switches and Accessories 1.0 0.9 0.1 Transformers 2.6 1.4 1.2 5.0 3.3 1.7 _= == The main reason for a net increase in the base cost was a new component added to tne project in 1980: the installation of the two 22 MW gas turbines (para. 2.07) at Diego de Almagro Substation. This solution was adopted as ENDESA had available at that time, two gas turbines that were procured in 1975 for other purposes. The cost of this equipment was charged to the project US$6.5 million of which US$4.4 million was foreign exchange. Financing 3.20 The financing plan prepared at appraisal is compared with both the financing plan after the amendment of the contract (Sept. 1978) and the actual sources at completion (Dec. 1982) as follows:  -15- Project Financing (US$ million) Appraisal After Actual Estimate Amendment Sources Source (March 1976) (Sept. 1978) (Dec. 1982) CODELCO Cash Generation 42.9 37.2 85.5 ENAMI Funds 0.9 1.2 6.0 ENDESA Funds - 17.8 30.7 IBRD 33.0 27.8 25.5 76.8 84.0 147.7 At appraisal, Bank financing (US$33.0 million) was estimated to be 43% of total project cots, CODELCO and ENAMI financing at that time, the balance or US$42.9 and US$0.9 million respectively. At the time of the amendment of the project scope (Sept. 1978), Bank participation was reduced to US$27.8 million or about 33% of total financing. CODELCO's funding was also reduced from US$42.9 to 37.2 million, ENDESA's funding was estimated to be US$17.8 million, and ENAMI's small participation was increased from US$0.9 to 1.2 million. At closing date (June 1982) the actual Bank participation amounted to US$25.5 million or 17% of total project costs. 3.21 ENDESA financed US$4.4 million in foreign exchange (much more than the cost overrun of US$2.1 million in foreign exchange) representing the cost of gas turbines mentioned earlier (para. 3.19); consequently about US$2.3 million of the loan amount originally allocated to this subproject was not used. 3.22 Apart from the Bank loan in the amount of US$25.5 million all other funds required to complete the project (US$122.2 million) were provided by CODELCO (70%), ENDESA (25%) and ENAMI (5%). Except for the changes of scope, agreed in 1978, the allocation of Bank financing to the subprojects was generally in line with estimates prepared during appraisal as shown in the table below.  -16- Allocation of Bank Financing (US$ million) Appraisal After Estimate Amendment Actual (March 1976) (Sept. 1978) (Dec. 1982) CODELC3 Projects: `a) El Teniente Crusher System 7.5 7.5 7.0 (b) Barquito Power Plant 16.3 - - (c) Chuquicamata Selenium Plant 0.7 - - (d) Technical Assistance 5.0 5.0 5.7 Total CODELCO Projects 29.5 12.5 12.7 ENAMI Projects: (e) Ventanas/Paipote Smelters Modernization 3.1 4.4 4.4 (f) Technical Assistance 0.4 0.4 0.2 Total ENAMI Projects 3.5 4.8 4.6 ENDESA Projects: (g) El Salvador, Power Interconnection - 10.5 8.2 Total Copper Sector Project 33.0 27.8 25.5 Disbursements 3.23 Withdrawals under the IBRD loan were made as shown in the following table. The appraisal forecasted the loan to be fully disbursed by the end of 1979. The actual Bank disbursement was not completed until the end of 1982, about three years behind the original schedule or about one year behind the amended schedule. Disbursement Experience (US$ million) Appraisal Amendment Actual Calendar (March 1976) (Sept. 1978) (Dec. 1982) Actual as X of Year Annual Cum. Annual Cum. Annual Cum. Appraisal Amendment 1976 2.0 2.0 - - - - - - 1977 15.0 17.0 0.6 0.6 0.6 0.6 3.5 100.0 1978 13.0 30.0 6.4 7.0 4.1 4.7 15.7 67.1 1979 3.0 33.0 12.3 19.3 4.0 8.7 26.4 45.1 1980 - - 8.1 27.4 7.2 15.9 48.2 58.0 1981 - - 0.4 27.8 7.8 23.7 71.8 85.3 1982 - - - - 1.8 25.5 77.3 91.7 The main reasons for slow disbursement of Bank loan compared with the profile estimated at appraisal were:  -17- (i) Changes in the scope of work such as cancellation of the Barquito Power Plant and the Chuquicamata Selenium Plant, and the amendment of the agreements to include the ENDESA interconnection system. (ii) Delays in procurement of goods and services which were due in part to the fact that CODELCO and ENAMI had not worked with the Bank before; therefore they were not familiar with Bank Procurement Guidelines and disbursement procedure.. This situation was reflected in delays in the prequalification of vendors. Another factor affecting the expediency of procurement and disbursement of Bank loan was the rather low limit set for international shopping US$50,000. It should be noted that CODELCO's own procurement guidelines, which are very well in line with the Bank's, set this limit at US$300,000. 3.24 Another issue with regard to disbursements relates to the difficulties in adjusting to the mechanics of Bank disbursements. These were charged to one single Government account for latter transfer to individual accounts of CODELCO, ENAMI or ENDESA. This procedure and the nature of Bank disbursement in different currencies, originated confusion and constant consultation on outstanding amounts and subsequently in interest calculations. IV. OPERATING PERFORMANCE 1. Commissioning and Start-up 4.01 With the exception of some initial problems to start-up the new burners in the ENAMI's Paipote smelters, the commissioning of all other subprojects went as planned and no major problems were encountered. There were of course the usual start-up problems caused by equipment failure or lack of experience but they were rectified with minimum delay by the staff of the individual companies who worked well to overcome these problems. 4.02 Tests for CODELCO's El Teniente underground crushing system were carried out during the April-August 1982 period; 80% of the designed capacity was reached. Production for 1982 was estimated to be 22.8 million tons of ore per year representing a 12% increase over the estimated production *without" the project. 4.03 The conversion of burners from oil to coal at ENAMI's Paipote and Ventanas smelters was completed in January 1980. The coal burners did not work satisfactory if operated at low capacity (the flame was too short). A modification of the burners was necessary at a cost of about US$100,000; this modification was completed by the end of 1980. 4.04 The cathode starter sheet machine at the Ventanas refinery and the new furnace and anode casting wheel at the Paipote smelter were commissioned on March 1980 and March 1981, respectively, without major start-up problems.  -18- 4.05 The ENDESA Interconnection project had no major start-up problems other than delays in commissioning the telecommunication equipment (para. 3.12). The Maitencillo-Diego de Almagro trausmis bio line was completed in the last half of 1981 and was energized (220 KV) in August 1981. 2. Environmental Aspects 4.06 Close attention was given to environmental aspects during the design and implementation of the several subprojects. At CODELCO's El Teniente underground crushing project two ventilation systems have been installed to monitor fresh air to the working areas and to coatrol dust within internationally accepted limits. 4.07 As a result of the study to modernize and expand the ENAMI's Paipote and Ventanas smelters, the consultants recommended the construction of a sulphuric acid plant to eliminate the emission of toxic gases (SO2). ENAMI has received authorization for construction of a 525 tons per day sulphuric acid plant for the Ventanas smelter. This project has been bid upon and will be awarded for construction early in 1985. V. FINANCIAL AND ECONOMIC PERFORMANCE 1. Financial and Economic Rates of Return 5.01 Because of the nature of the individual subprojects, only in three cases, out of seven project components, an attempt to calculate financial rates of return was made during appraisal; they were for CODELCO's El Teniente crusher system, the Barquito Power Plant and the Selenium Plant at Chuquicamata. Of these three subprojects the last two were cancelled and not implemented. 5.02 The following table summarizes the financial rates of return estimated during appraisal and those calculated at project completion using information provided by CODELCO, ENAMI, ENDESA and staff estimates.  -19- Financial Rates of Return Appraisal Project Completion Estimates Estimates CODELCO Projects: El Teniente Crusher Sys'em Greater than 502 16Z Barquito Power Plant 20-35% Project cancelled Chuquicamata Selenium Plant 8-35Z Project cancelled Technical Assistance N/A N/A ENAMI Projects: Ventanas/Paipote Smelters Modernization N/A Greater than 50% Technical Assistance N/A N/A ENDESA Project: El Salvador, Power Interconnection N/A 20% a/ a/ The El Salvador Power Interconnection project is a better alternative than the original Barquito Power Plant project for discount rates up to 20%. Economic rates of return were estimated to be only marginally above the financial returns since transfer payments to the government, apart from income tax, are not significant. The product is priced at border prices, and no conversion factors are warranted for other inputs. CODELCO Projects 5.03 Without the installation of the underground crusher and related infrastructure at CODELCO's El Teniente mine, the installed capacity would be insufficient to maintain present levels of production. The underground crusher was the bottleneck in the production chain and is the single most important element limiting the ore production. 5.04 To take full advantage of the new underground crusher, additional complementary investments have been made at El Teniente mine. Therefore, to measure the economic impact of the combined, crusher and other installations, the total investments required has been compared wi-h the net benefits resulting from incremental copper production. Follows a summary of the crusher and other related investments. El Teniente Mine Project Investment Investment Components US$ million) Underground crushing system 89.5 Modification of exploitation method 8.2 Large holes drilling equipment 0.3 Expansion of concentrator plant 15.4 Mine drainage system 1.1 Loading equipment, LHDs 1.9 Total 116.4 Source: CODELCO.  -20- Based on the incremental copper production (para. 3.02), the investments shown above, and detailed data on cost, depreciation and income tax provided by CODELCO, the financial rate of return has been calculated assuming copper prices projection as follows: between US$0.65 to US$0.90 per lb. in constant 1983 terms, during the period 1983-1997.2/ The financial rate of return thus calculated is, 21.9% before taxes and 16.1% after taxes, indic-ating that the subproject has a high return, although somewhat lower that. the T71? estimated at appraisal. Details of calculations are show in Annex 3. ENAMI Proiects 5.05 The objective of the ENAMI's smelters rehabilitation project was primarily to assure the continuation of production from smelting operations as well as to bring about a moderate increase in installed production capacity; significant production losses have been caused by plant breakdowns which had been occurring with increasing frequency. Quantifiable savings, mainly in fuel consumption, are estimated at USS 11.9 million per year as detailed below. This amount compared with total investment in ENAMI projects (US$10.3 million) confirms the high return expected at appraisal (greater than 50%). (i) the replacement of an obsolete cathode starting sheet preparation machine at ENAMI's Ventanas refinery reduced operating and mai.ntenance costs and increased overall productivity. The tankhouse is now able to work consistently at high current efficiencies (98%) at least, 3% to 4% over past performances, allowing the operation at current densities of over 300 A/M2, as compared to previous levels of 215 A/M2. This increase is reflected in about 12,000 tons of incremental refined copper production that previously was exported as blister copper. Quantifiable incremental benefits are estimated at US$1.2 million per year since refined copper sells at a higher price than blister yielding a net benefit in the order of US$100/ton. (ii) at the Paipote smelter, two obsolete blister casting units of limited capacities were eliminated, extending the smelting process capability to fire-refining and anode casting. A complete installation wa- provided including a tilting furnace and an anode casting wheel with advanced pour and weight controls. This is an on/line facility handling molten copper which eliminated blister casting at Paipote and the subsequent and expensive blister melting operation at Ventanas of over 60,000 tpy of copper. Quantifiable incremental benefits are estimated at US$0.7 million per year due to savings in the order of 3,000 tons p.a. of fuel. 2/ Copper price projections are based on recent Bank staff estimates.  -21- (iii) a complete pulverized coal preparation and handling Installations were provided, including dual oil/coal burner systems for the reverberatory furnaces of the Paipote and Ventanas smelters, thus extending and providing new capabilities for using coal, a less expensive fuel. F*AMI estimates that savings resulting from this modification are in the order of US$10.0 million p.a.; US$7.5 million in the Paipote smelter and US$2.5 million in the Ventanas smelter, the difference is the result of fuel costs in Paipote being twice as high than in Ventanas. The modified burners consumption rates are 1.93 tons of coal instead of one ton of diesel fuel at Paipote and 1.73 to one in Ventanas. The cost per ton of coal at both smelters is estimated to be US$73/ton and the cost of diesel fuel, US$447/ton at Paipote and US$207/ton at Ventanas. ENDESA Project 5.06 Substitution of the Barquito Power Plant project by the ENDESA Interconnection project was justified by showing that it was a more attractive alternative than the original rehabilitation and expansion of the existing thermal system for all discount rates up to 20% as shown in the Table below. Although the investment in the transmission system was higher than the investment in rehabilitating and expanding the generation plant, the operation and maintenance costs were lower, as it taps from hydropower and coal-fired generation instead of burning fuel oil. ENDESA - Power Interconnection Project Economic Justification (US$ million, 1983 terms) Incremental Barguito Thermal Plant Interconnection Project Costs Investment a/ 0&M b/ Total Investment c/ O&M d/ Total 1980 7.0 - 7.0 13.8 - 13.8 6.8 1981 7.0 - 7.0 16.6 - 16.6 9.6 1982 13.2 - 13.2 13.1 - 13.1 -0.1 1983-2012 - 15.3 15.3 - 11.2 11.2 -4.1 Equalizing discount rate .. . *.20X a/ 20 MW oil fire unit plus heat-recovery boiler. b/ Operating and maintenance costs include fuel cost at US$185.5/tonne (Bunker C). c/ Includes investment in the interconnection project and in conversion of mining operations to 50 HZ. d/ Includes energy purchases (255 GWh/year) to CIS valued at long run marginal costs: US$88.3/KW-year and 24.1 mills/kWh.  -22- Most of the benefits to the economy derived from the Interconnection proj ect are related with fuel savings. It is estimated that as a result of the project about 86,000 tonne/year of fuel oil and diesel oil are being substituted by 82,000 tonne/year of coal. Priciag these fuels at international prices, the net effect is savings in foreign exchange in the ordear of US$16 million per annum. 2. Financial Covenants 5.07 The loan and project agreements including the amendment signed on Sept,mber 1978, have in practice, no specific financial covenants other than those standard to maintaining sound accounting practices and to furnish to the Bank annual audited financial statements.31 In the case of ENAMI, additional covenants relates to pricing policies and cost reduction programs. With the exception of ENAMI's pricing policy to purchase copper concentrates which was modified4/ in mid 1982 without consulting the Bank. All other covenants were fully complied with. 3. Economic Benefits 5.08 The copper sector plays a major role in the Chilean economy as the major foreign exchange generator. Its continued competitiveness on the critical copper world market is of paramount importance to the country. In this respect the different individual subprojects have made a contribution by (i) reducing operating and maintenance cost, (ii) saving considerable fuel costs, (iii) allowing more flexibility to the operations by increasing machine availability, and (iv) maintaining the required levels of copper production. Other economic benefits are those related to the project's impact on the net foreign exchange savings follows: (i) incremental copper production in concentrates at El Teniente mine estimated at an average of 50,000 tons per year (para. 3.02), equivalent to about US$60 million p.a.I5/; (ii) about US$1.2 million p.a. as a result of incremental production of refined copper that previously was exported as blister, and (iii) indirect foreign exchange savings in the order of US$10.7 million p.a. as a result of savings in fuel oil at ENAMI smelters. 3/ Available in project files. 4/ Policy modified to pre-finance advances against future sales to alleviate the financial situation of medium and small scale miners due to depressed copper prices. 5/ Assuming a price of 70 cents per lb of copper and 15 cents per lb of direct and indirect foreign cost to produce one pound of refined copper (including the annualized foreign exchange component of capital costs). Metallurgical recovery from concentrates to refined copper assumed at 98%.  -23- VI. INSTITUTIONAL PERFORMANCE 1. Development of Management and Organizational Effectiveness 6.01 An important institution-building feature of the project was the reorganization of CODELCO aimed at reducing the excessive independence of the individual mining companies which CODELCO inherited from foreign firms at the time of the copper industry's nationalization. To accomplish this, the Government committed itself to take all necessary action to convert CODELCO and the mining companies into a single, fully integrated corporation with a central administration responsible for planning, sales, and purchasing. However, individual mines were to retain sufficient operational autonomy, to safeguard their efficiency. As established by Decree Law No. 1.350, on April 1, 1976 the state-owned enterprise Corporacion Nacional del Cobre de Chile, CODELCO-CHILE, was created replacing the following ex-government owned companies Compania de Cobre Chuquicamata, Compania Minera Exotica, Compania de Cobre Salvador, Compania Minera Andina and Sociedad Minera El Teniente, which were dissolved as of that date. CODELCO started implementing its reorganization immediately after the law was passed and since then is now operating successfully as a single, fully integrated corporation. 6.02 Another important feature is the fact that the project has proved that CODELCO, ENAMI, and ENDESA are technically able to undertake the implementation of projects of this nature; in particular CODELCO who successfully has implemented the large and complex El Teniente underground crushing system. Considerable improvements have also been achieved in the areas of planning and control in areas such as detail planning for project execution, long term production and cost planning, and overall corporate planning and development, areas that were hardly developed before the execution of the project. VII. BANK PERFORMANCE 7.01 The project was identified during sector missions in mid 1973 ard early 1974. Appraisal was carried out in October 1974 and the Board approved the loan in February 1976, about 16 months after appraisal. 7.02 Initially, the project received close supervision with three missions during 1976 and again three missions in 1977. After the amendment in scope and due to the progress reached in implementation, less frequent supervision was required, though, more supervision missions would probably have not resulted in better implementation. For project supervision, the Bank also relied on progress reports which were submitted regularly by the implementing agencies and provided a good overview; every three months in  -24- the case of CODELCO and ENAMI and almost biannual in the case of ENDESA. There was a considerable turn over on the Bank staff supervising the project which resulted in lack of continuity and familiarity with the project and with the executing agencies. 7.03 Because this was a sector loan comprising several individual subprojects, the information for each subproject, on capital and operating costs, implementation schedules, procurement packages, etc. was less detailed than for a conventional project. In particular, the information on expected financial and economic benefits was not prepared with sufficient detail to facilitate comparisons of actual performance with estimates at appraisal. 7.04 The Bank's relationship with the Goverrment and CODELCO was varied reflecting shifts in Government policies and priorities, and changes in CODELCO's managerial staff. As a result discussions during appraisal to define the project scope were lengthy and time consuming. Later on, the project scope needed to be amended several times during implementation. The Bank's relationship with ENAMI and ENDESA throughout the project preparation and implementation was excellent. In general, Bank's relationship was greatly facilitated by the provision by the companics of adequate teams of competent professionals. VIII. CONCLUSIONS 1. Overall Assessment 8.01 Despite the fact of considerable cost overruns (paras. 3.16 to 3.19) and delays which occurred during construction (paras. 3.09 to 3.12), overall, the project was successfully completed and the basic technical and economic objectives were achieved. The rates of return calculated at project completion are as high as estimated at appraisal, except, in the El Teniente underground crushing system, where due to much lower copper prices than originally projected the return is lower. 8.02 At completion, the project cost has been calculated at US$147.7 million, nearly twice as much as the amended estimate. Main reasons have been: (i) the delays in initiating the procurement of equipment (due to lack of familiarity with Bank guidelines), (ii) delays in the supply of railcars and fans, (iii) delays in the construction of the El Teniente underground crushing system mainly due to unforeseen poor ground conditions, (iv) more than expected engineering and administrative costs, and (v) the higher than projected local inflation which in terms of US dollar equivalent was magnified by the fact that a fixed exchange rate of 39 pesos to the US dollar was maintained for a period of almost three years (1979-80-81). 8.03 The need for changes to the project scope was one of the issues from the beginning; negotiations were also delayed for this reason. During project implementation, two of CODELCO's original subprojects were  -25- cancelled. To supply the requicsed electricity originally coming from the El Barquito power plant, a new subproject was introduced; the ENDESA's Interconnection project. Finally, ENAMI's subproject scope at the Paipote smelter was also changed replacing a blister casting wheel by a complete anode production installation, including a fire-refining furnace and an anode casting wheel with advanced controls, not considered during the initial appraisal. 2. Lessons for Similar Projects 8.04 At the time of appraisal, the individual mining companies that originated CODELCO had an efficient operating management, but lacked planning capabilities since all decisions on major investments have historically been taken abroad by the foreign owners. Within this framework, the project was regarded as the first of a series where the Bank's role could be significant. The strategy being that after winning the confidence of management and the operating people, and helping them improve their planning and implementing capabilities, the Bank would play an important role in helping frame sector strategy and corporate policies; however, owing to the fact that for various reasons lending to Chile became much smaller and more sporadic than expected, the Bank association with CODELCO did not work out. On project grounds, a lesson learned is that dealing with only one of the departments of a large corporation such as CODELCO, and financing only a small portion of a large investment program, did not provide a meaningful role for the Bank. In the future, when a large corporation having a dominant role in the sector is involved, the Bank should seek to establish a direct linkage, through traditional project lending or financing a slice of a multi-year program, with adequately sized loans to enable a constructive dialogue on sectoral policy issues. 8.05 Procurement and disbursement delays in the early stages of the project could be reduced with closer interaction with the Bank. In similar cases, when the borrower or executing agency is not familiar with Bank guidelines for procurement and disbursement, early missions should dedicate more time and effort to these matters. 8.06 Wherever possible, high turn over of Bank staff supervising a project should be avoided. Staff familiar with the project and the borrower should be maintained at least during appraisal and early stages of implementation. 8.07 The satisfactory completion of the individual suips.ojects and the achievement of the technical and financial objectives was mainly due to (i) appropriate funding of resources both local and foreign, till completion of the project, (ii) full time project implementation teams in the individual companies which appointed qualified and dedicated staff, and (iii) generally good relations with consultants, contractors and suppliers. 8.08 The project has introduced effectively, planning and project implementation capability, particularly within the operational staff of CODELCO. M17-Chile  - 26 - ANNEX 1 Page I of 1 C17LE - COPPFR SF.CTnR PROJF.T CODELCO A:D ENAWI lIrLrMHENTAT1ONl UNITS CODELCO 1I Phase: CODELCO General Mine Superintendent Vice General Project Mine Superintendent