Mexico - Second Fertilizer (Lazaro Cardenas) Project

Document of The World Bank FOR OFFICIAL USE ONLY MICROFICHE COPY Report No. 10259-ME Type: (PPR) Report No 10259 ANDERSON, / X31676 / T9 111/ OEDD2 PROJECT PERFORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (LOAN 1686-ME) DECEMBER 31, 1991 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise he disclostd without World Bank authorization. ABBREVIATIONS DAP Diammonium Phosphate Fertilizer FERTIMEX Fertilizantes Mexicanos S.A. LAC 50 Initial Fertilizer Project LAC 93 Duplication project not financed by the Bank NAFINSA Nacional Financiera S.A. NPK Complex Fertilizer PEMEX Petroleos Mexicanos S.A. TSP Triple Super Phosphate Fertilizer THE WORLD BANK FOR OFFICIAL USE ONLY Washington. DC 20433 USA. Office of D'.ecto*Ceneral Operatwins Evaludition December 31, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECTS Project Performance Audit Report on Mexico Fertilizer II Project (Loan No. 1686-ME) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Mexico, Fertilizer II Project (Loan No. 1686-ME), prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only In the performance of their oicial duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (Loan 1686-ME) TABLE OF CONTENTS Page No. Preface .......................................................... i Basic Data Sheet..................................................... iii Evaluation Summary .......... .................................... vii PROJECT PERFORMANCE AUDIT A. Project Background......................................... I B. Project Objectives and Components.................. ........... 2 C. Appraisal History - Financial Covenants............. ......... 2 D. Implementation Experience...... ......5.......................5 E. Project Cost and Procurement.................................. 10 F. Operational Experience and Outcome............................ 11 G. Environmental Issues.......................................... 12 H. Financial and Economic Results................................ 12 I. Sustainability..................................... ......... 14 J. Role of the Bank........................................ ..... 15 K. Findings and Lessons.......................................... 15 ANNEXES Annex I - FERTIMEX Income Statement.............................. 19 Annex II - FERTIMEX Income Statement.............................. 20 Annex III - MEXICO Actual Production of Lazaro Cardenas Fertilizer Complex...................... ........... 21 Annex IV - Comparison of Actual and Forecast Input and Output Prices.................................... 22 PROJECT COMPLETION REPORT I. Introduction........................................ ...... 25 II. Project Background and Objectives............................ 26 III. Project Implementation, Institutional Development and Management.............................................. 28 IV. Operation Performance......................................... 38 V. Financial Performance and Covenants.......................... 47 VI. Economic Analysis............................................. 50 VII. Bank's Role................................................... 53 VIII. Conclusion, Sustainability and Lessons Learned............. 53 This document has a restricted distribution and may be used by recipients only in *he performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. TABLE OF CONTENTS (Continued) page No.- ANNEXES I Organization Chart of Lazaro Complex............ ........ 57 II Programmed and Actual Implementation Schedu?e............ 58 III Actual Capital Cost Expenditures Schedule (LAC 50)....... 59 IV Project Financing Plan................................... 60 V Actual Capital Cost Expenditure Schedule for All Lazaro Cardenas Plant...................... ........... 61 VI Working Capital Estimation for 1992...................... 62 VII Implementation Summary................................... 63 VIII Initial Monthly Production..................... ........ 64 IX Expected Performance of Process Plant.................... 65 X Projected Financial and Economic Prices...... ........... 66 XI Estimated Production Costs and Output Value for 1993..... 67 XII Costs and Benefits Streams for Financial Rate of Return Calculations - Total Lazaro Cardenas Plant... 68 XIII Costs and Benefits Streams for Financial Rate of Return Calculations - Total Lazaro Cardinas Plant...... 69 XIV Estimated Profitability and Cash Flow of the Lazaro Cardenas Complex................................ 70 XV Costs and Benefits Streams for Economic Rate of Return - Calculations Project LAC 50................... 71 XVI Costs and Benefits Streams for Economic Rate of Return Calculations Total Lazaro Cardenas Plant............... 72 XVII Foreign Exchange Savings................................. 73 XVIII FERTIMEX Balance Sheet - Historical 1982-86.............. 74 XIX FERTIMEX - Profit and Loss - Historical 1982-86.......... 75 ATTACHMENT I - PROJECT COMPLETION REPORT PREPARED BY FERTILIZANTES MEXICANOS S.A. PLANNING MANAGEMENT OFFICE............. ........... 77 i PROJECT PERFORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (Loan 1686-ME) PREFACE This is the Project Performance Audit Report (PPAR) for the Mexico Second Fertilizer Project (FERTIMEX II) for which a Bank loan in the amount of USS80 million was approved on April 12, 1979 to Nacional Financiera, S.A. (NAFINSA) and Fertilizantes Mexicano, S.A. (FERTIMEX). The loan was totally disbursed and helped increase Mexico's nitrogenous and phosphatic fertilizer production capacity by 795,000 tons per year. Following the closing of the loan on December 31, 1983, the Bank staff visited Mexico in November 1987 tor the preparation of the Project Completion Report (PCR). A PCR, dated June 1988, was also prepared by FERTIMEX (Attachment I). These PCRs provide a sAtisfactory account of the project experience. They discuss adequately the performance of the Bank and FERTIMEX. The PPAR elaborates on particular aspects such as the appraisal history and financial covenants and draws attention to the factors that marred the successful implementation of this project. Both the draft PCR and PPAR were sent to the borrower for comments. Correspondence was received from NAFINSA's Office of Multilateral Financing reflecting the views of the Federal Government of Mexico, NAFIN and FERTIMEX. The most relevant comments have been incorporated in the texts of' PCR and PPAR and, in some cases, as footnotes. - iii. - PROJECT PERFORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (LOAN 1686-ME) BASIC DATA SHEET (US$ mi;l.' n) LOAN POSITION As of 09/30187 Original Amended Disbursed Cancelled Repaid Outstanding Loan No. 1686-ME 80.0 80.0 80.0 - 29.1 50.9 CUMULATIVE LOAN DISBURSEMENT FY79 FY80 FY81 FY82 FY83 FY84 (i) Planned 24.0 68.0 80.0 80.0 80.0 80.0 (ii.) Amended - - - - - - (iii) Actual 4.2 29.6 40.7 60.9 77.2 80.0 (iv) (iii) as % of (i) 17 44 50 76 96 100 PROGRAM/PROJECT DATA Total LC Original loan/ Actual or project with credit date(s) loan agreements dglication Board approval April 12, 1979 Loan agreement May 18, 1979 Effective date August 20, 1979 - Loan closing October 31, 1982 Dec. 31, 1983 - Date of physical completion Dec. 1986 Dec. 1986 Completion time (in months) 36 105 108 Time overrun %-range - 191% - 69 Date of start up of opertions September 1981 June 1987 October 1987 Total Project Cost (US$ million) 301.3/a 507.8 802.8 Cost overrun (%) 68.5 - Financial rate of return 8.3 -0.3 +0.3 Economic rate of return 12.3 -1.0 -0.9 Ia Includes working capital for US$24.9 million and US$23.9 million for interest during construction. - iv - PROJECT PERORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (Loan 1686-ME) Mission Data Mnth/vear No. of weeks No. of persons Manweeks Raoort data Appraisal 09/78 0.6 3 1.8 03/21/79 prvision 06/79 0.4 2 0.8 07/05/79 Supervision 02/80 0.6 3 1.8 03/11/80 Supervision 09/80 0.6 2 1.2 09/25/80 Supervision 07/81 1.2 2 2.4 07/16/81 Supervision 11/81 0.8 1 0.8 11/11/81 Supervision 07/82 0.6 2 1.2 08/02/82 Supervision 01/83 2.0 3 6.0 02/15/83 Supervision 09/83 0.8 1 0.8 11/16/83 Supervision 0.4o4 0*4 1 0.4 12/28/84 Supervision 06/85 0.2 1 0.2 06/25/85 Completion 11/87 2.0 2 3.0 Dec 1987 OTHER DATA Borrover: NAPINSA/EERTIME2 Ezecuting Agency: FERIME Fiscal Year of Borrovert January 1 to December 31 PROJECT PERFORMANCE AUDIT REPORT MXICO FERTILI2ER II PROJECT (Loan 1686-ME) Project Scove Appraisal Actual Total Lazaro Cardenas expectation estimates plant with duplication (LAC 50) (LAC 50) (LAC 50 + LAC 93) (in '000 tpy) Sulfuric acid 660 660 1,320 Phosphoric acid 198 198 396 Nitric acid 215 215 215 Ammonium nitrate 270 270 270 DAP/NPK 525 525 1,050 /a CURRENCY EOUIVALENTS End of period Average (pesos) At appraisal US$ 1 * 22.5 1980 US$ 1 * 23.3 22.9 1981 US$ 1 * 26.2 24.5 1982 US$ 1 a 96.5 57,4 1983 US$ 1 a 143.9 120.2 1984 US 1 - 191.6 167.8 1985 US$ 1 a 371.5 256.9 1986 US$ 1 - 923.0 611.4 Time of physical completion US$ 1 a 923.0 - September 1987 US$ 1 * 1,545.0 1,515.0 October 1987 US$ 1 a - 1,650.P Nov. 1987 at time of completion US$ 1 a * 1,750.0 mission February 1988 US$ 1 * 2,261 - /a A section is being added to the second granulation plant to allow TSP production instead of DAP at a nominal capacity of 380,000 tpy. - vii - PROJECT PERFORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (Loan 1686-ME) EVALUATION SUMMARY 1. Introduction substitution program to ensure that Mexico would have sufficient capacity Since '-he late 1940s, the Bank to meet expected increased demand for has made agriculture the leading domescic phosphatic and nitrogenous sector for it3 lending to Mexico. By fertilizer through the early 1980s. the end of 1978, 19 loans (constitut- There were also other objectives ing some 40% of Bank lending) had paiticularly the creation of employ- been approved for agricultural and ment in the western region of the rural develipment. To support this country, the saving in foreign ex- strategy and to help improve the change and the transfer of up-to-date productivity of agriculture, the Bank fertilizer production technology to also began assisting the fertilizer Mexico. sector by approving a first loan (Loan 1112-ME) of US$50 million in The project consisted of five 1975. production units and included storage handling, utilities and infrastrue- All fertilizer production in tural facilities. The production Mexico, except sulfate and specialty units are: fertilizers is in the public sector. PEMEX (the national petroleum compa- * a sulfuric acid unit with a ny) produces all ammonia used in production capacity of about 660,000 Mexico and FERTIMEX produces all tons per year; other fertilizer intermediaries and all finished solid fertilizers. In a phosphoric acid unit with a addition, FERTIMEX is the primary production capacity of about 198,000 agency responsible for marketing and tons per year; distribution of fertilizers in Mexi- co. * a nitric acid unit with a pro- duction capacity of about 215,000 By the late 1980s, FERTIMEX '-ad tons per year; 53 plants in 12 different locations. N.:.rly half of these plants started a fertilizer granulation unit cperating before 1970. Many plants with a production capacity of about producing low concentration fertiliz- 525,000 tons per year of diammonium ers are economically non-viable. The phosphate fertilizer and/or NPK com- fertilizer subsector is undergoing pound fertilizer; and restructuring through a Fertilizer Sector Adjustment Loan approved in * a solid ammonium nitrate unit 1988. with a production capacity of about 270,000 tons per year. 2. Project Objective and Scope A major change in the project The second fertilizer project ecope during implementation was was part of the Government's import FERTIMEX's decision to double the - vii phosphatic fertilizer prod-ction earlier years isome equipment failures capacity by duplicating the aulfuric were e.-perienced and the operation of and phosphoric acid plants and the the phosphoric acid unit was affected fertilizer granulation unit for the by the mediocre quality of the indig- production of DAP/NPK. This decision enous phosphate rock from Baja Cali- was not communicated to the Bank. fornia. 3. Implementation Experience Despite the provision of sever- ai loan covenants, the unsatisfactory There were . number of factors financial situation of FERTIMEX, which negatively and seriously af- starting in 1982, remained unchanged fected the implementation and caused throughout the period of project a very large overrun and a delay of implementation and after the commis- rearly six years in the completion siouiing of the plant. Based on actu- and commissionlig of the plants. The al prodiction and prices for the last 1979 earthquake necessitated the several years and assumptions regard- revision of many structural design ing the future performance of the p..neters. The change of scope plant and international fertilizer placed an enormous burden on project and raw material prices, the finan- management structure that was both cial rate of return is 11kely to weak and unsuitable. It was not remain negative. Reestimation of the until early 1983 that FERTIEEX even- economic rate of returr also result tual37 put into effect a project in negative numbers: this may be management organization that it had traced to the long delay in implemen- agreed to at the time of appraisal. tation and enormous cost overruns, The onset of Mexico's financial cri- though the current price6 of inputs sis in 1982 also seriously affected and outputs and the poor quality of this project and the shortage of the domestically supplied phosphate funds especially dur3ng 1982 and 1983 rock have also been serious impedi- is considered to be responsible for ments to viability. at least a delay of 1 years in the implementation. Just when many 5. Environmental Issues plants were approaching start up, the 1985 earthquake struck and caused Althcugh the environmental large damage to the plants creating impact of this project was clearly both financial loss and delay in recognized from the beginning and completion. suitable measures introduced to re- duce the gaseous and solid effluents, 4. Results the project ran into solid waste disposal problems immediately after The commissioning and start up commissioning. In effect the gypsum of the offelte and utilities com- disposal system (designed to dis- menced by mid-year 1985. Subsequent- charge gypsum produced by the phos- ly, othei lants and units were com- phoric acid plant into the sea at a missioned during 1986 and 1987 with depth of 30 meters some 1.5 kilome- the duplicate phosphatic fertilizer ters from the shore) has suffered train coming i.nto production in Au- from a series of mishaps such aa the gust 1987. By the fourth year of blockage and rupture of pipes and operation (1990) capacity utilization malfunctioning of the discharge pump. had reached between 80% and 85% for This Droblem has been largely allevi- different units even though in the ated in recent times through remedial action and tests show no aJartiing The major loesonLi that: come out marine degradation. of this project. are: 6. Sustainability * The Bank should make certain at appraisal that the implementation From a purely technical point capability exists and that the Bor- of view, production has approached rower is not alrearv heavily charged the design capacity and FERTIMEX's with other investment activities, technical and operating personnel clearly have the capacity to operate * Mechanistas should be put in the complex efficiently. The fertil- place in order to avoid the crooping izer units are of modern design and of new large investment commirtments have the correct scale. The finan- beyond the technical and financial cial and economic sustainability will capability of the Borrower. depend, to a large extent, on the diligent implementaotion of the provi- * The establishment of a suitable sions of the FSAL and a government project management structure dedicat- pricing policy that would not penal- ed to the task at hand should become ize FERTIMEX and the course of Input 'a condition sine qua non' of pro- and output prices in the future. ceeding with financing. Calculations for 1990 show that even though domestic prices are allowed to * The Bank needs to ensure strict approach international prices, the compliance with the covenants and the depressed market for fertilizers will spirit of the loan agreemen:s instead result ir. losses for FERTIMEX II. of bending back-ward at times to save Under these conditions, the sustaina- the project. bility of the project is uncertain. 7. Findings and Lessons Although factors outside the control of FERTIMEX did materially affect the outcome, the experience with this project illustrateb, above all, how a series of missteps and the lack of a strong project managemen, organization can inflict irreparable damage to an otherwise highly benefi- cial and desirable project. The approval of an FSAL coming in the wake of this project is a clear indication that policy ,-ues needed to be addressed if the Mexican fertilizer industry is to provide a meaningful contribution to the Mexi- c.n economy. The Bank's initial insistence on the resolution of some policy issues was both necessary and timely. PROJECT PERFORMANCE AUDIT REPORT MEXICO FERTILIZER II PROJECT (Loan 1686-ME) A. Project Background 1. Since the late 1940s the Bank has made agriculture the leading sector for its lending to Mexico. By the end of 1978, 19 loans (constituting some 40% of Bank lending) had been approved for agricultural and rural development. To support this strategy and to help improve the productivity of agriculture, the Bank also began assisting the fertilizer sector by approving a first loan (Loan 1112-ME) of US$50 million in 1975. Thia loan wab given for the construction of two large scale urea plants with a combined capacity of 825,000 tons as well as the expansion of an existing insecticide plant. 2. Shortly after the approval of the first fertilizer loan, the Government of Mexico submitted to the Bank, through its fertilizer company, GUANOMFX, another study for the implementation of a large fertilizer project aimed at reducing further the imports of substantial quantities of nitrogenous and phosphatic fertilizer. Initially, the project was to be located at Guayamac (northwest Mexico) where it would be close to a planned new copper smelting facility of Mexicana de Cobre S.A. (supplying the sulfurous smelter gas for sulfur production) and a phosphate rock deposit under development in Baja California. Even though other raw materials namely ammonia and potassium salts were to be supplied by PEMEX and imports, Guayamas had the advantage of being close to the center of gravity of phosphate fertilizer and ammonium nitrate consumption. In addition, it enjoyed the existence of fairly developed infrastructural services and a pool of skilled and unskilled workers. At this stage, the total cost of the project was estimated at around US$140 million and project completion was projected by mid-1979. 3. At the pre-appraisal mission which was mounted in April 1976, it became clear that since the north-western reaion of Mexico is water deficient and the authorities would probably not issue permits to drill new wells, the industrial water might have to be pumped over a long distance from Yaqin Agricultural District. The timely availability of sulfuric acid from Mexicana de Cobre's smelter was also becoming an issue. Moreover, even though it was agreed that initially the phosphoric acid plant would use imported phosphate rock, the plant needed to be designed with the quality of the Baja California phosphates in mine. AUl this obviously had important implications for the design of the complex and its capital cost. While these matters were being pursued by the Bank, the senior management of GUANOMEX was changed at the end of 1976 with the inauguration of a new Presidency. Following these events, the project was initially postpone6 as a part of the austerity program, but the Government's approval was reissuer in mid-1977. The site of the complex was also changed to Lazaro Cardenas. )r November 1977, the Government of Mexico further decided to merge another comp%Ay producing phosphatic fertilizers (Fertilizantes Fosfatados Mexicanos - FFM) intc GUANOMEX to form a new entity called Fertilizantes Mexicanos S.A. (FERTIMEX), 2 4. The decision to relocate the project at Lazaro Cardenas instead of Guayamas necessitated a fresh reappraisal by the Bank. Although the relocation obviated the linkage with Mexicana de Cobre for the supply of sulfurous gases, the potential use of Baja California phosphate rock remained a design consider- ation (despite the fact that it had already been resolved that the project would go ahead based on imported phosihate rocks). The reappraisal, thus, proceeded throughout 1977 and 1978 for a somewhat enlarged project. The project was presented to the Board in April 1979. B. Proiect Objectives and Components 5. The objective of the project was to ensure the availability of sufficient domestic capacity to meet expected increases in demand for nitrogenous and phosphatic fertilizers up to the mid 1980s. The project consisted of plants which would produce 275,000 tons per year of diammonium phosphate (DAP), 250,000 tons per year of different grades of compound fertilizers (NPK), 270,000 tons per year of solid ammonium nitrate (AN). To produce these fertilizers, the manufacturing complex included a sulphuric acid plant (660,000 tons per year), a phosphoric acid plant (198,000 tons per year as p2o5) and a nitric acid plant (215,000 tons per year). The project also included offsite facilities for storage of raw materials, intermediate and finished products, weter supply, steam generation and power. FERTIMEX also constructed its own berth at the port of Lazarc Cardenas. 6. In addition to meeting the objective of relative self-sufficiency in fertilizer supply the project's location was in line with the Government's policy of moving industry away from the heavily industrialized eastern coastal areas. The project was expected to result in considerable foreign exchange savings (estimated at $125 million per year in 1978 dollars after provisions for principal and interest on foreign loans in the first year of full capacity utilization) for the Mexican economy. The project's direct and indirect employment creation would be modest, but still important for the region. More importantly, the project was expected to give a chance to the local engineering and equipment and manufacturing firms, thus deepening their know-how, expertise and technological capability. C. Appraisal History - Financial Covenants 7. Prior to the approval of the loan and based on the experience with the first fertilizer project (Loan 1112-ME), the Bank expended much effort in tackling the financial problems of FERTIMEX. At the time of the reappraisal of the project, during 1977 and 1978, it had become very clear that the Government's fertilizer subsidization policy had landed FERTIMEX in serious and continuous financial difficulties. The deterioration of FERTIMEX's financial situation was due largely to inadequate sales prices, increasing raw material prices, and heavy debt service arising from the 1975 devaluation of the peso. 3 8. The Bank's initial position was to insist on the financial restructuring of FERTIMEX and to have the fertilizer prices increased in such a way as to enable FERTIMEX to become financially viable and earn a financial rate of return of at least 10% annually by 1981. This was in line with the Bank's accepted policy of: (a) not expecting its borrowers to be subsidized; (b) insisting on the financial viability of projects in lending to revenue earning entities which would also ensure credit worthiness; (c) insisting that public enterprise borrowers would have sufficient autonomy from the government as to enable them to manage their effairs, allocate their resources and operate their facilities efficiently; and (d) requiring a pricing policy for major industries that would not preclude the entry of the private sector into such industries in the future. 9. In this respect, the Bank demanded compliance with the financial covenants of the first fertilizer loan (Loan 1112-ME) which obligated the Government to take the necessary steps by 1979 to achieve a current ratio of 1.3:1.0, a debt equity ratio of 60/40 and a debt service coverage ratio of 1.4 times. As for fertilizer prices, the Bank took a more sympathetic view of the Government's agricultural policy and accepted the proposition that they may need to be raised gradually. It, therefore, suggested a two tier pricing mechanism whereby FERTIMEX's marketing and manufacturing activities would be separated. In this way, the authorities would have more flexibility in pursuing their agricultural development policies (given their argument that increased fertilizer prices may not only affect their utilization, but gave rise to increased food prices and inflation rates) while permitting FERTIMEX to have ex-factory prices which enabled it to earn a reasonable return on its investment and to safeguard its financial viability. 10. While the Mexican Government was prepared to financially restructure FERTIMEX (by converting its short term debts to long term, revaluing its assets to reflect current value, improving its working capital and paying compensatory amounts for foreign exchange losses), it opposed the two tier pricing system on the grounds that: (a) the pricing mechanism would create a precedence for all other public enterprises thereby creating an avalanche of demands; (b) it would unduly restrict the Government's discretionary power in respect of pricing; (c) the automatic nature of the price increases would have a budgetary implication beyond one year while budgets were approved annually by the Mexican Congress; and, (d) the financial return of the public sector enterprises was, in any case, of little importance to the Government. Other related objections that surfaced later were in respect of the need to stop the public enterprises from making unauthorized investment and the possibility of having to pay a part of the revenue in profit sharing schemes. 11. After much internal discussion whether to continue or discontinue with the appraisal, the Bank eventually modified its position in September 1978 and suggested that provided the Government of Mexico took the necessary actions to financially restructure FERTIMEX in such a fashion as to make it meet the IDecision Memo dated February 10, 1978. 4 financial covenants contained in the first fert ( c,e lY. oc'3y 1981, (referred to in para. 9 above), and provided thu joii undej- 1i. un ,n would include the same financial covenants relating to the financ.aJ -,-A : .( of the company as were already agreed to under the first fertilizer loat :n( provided that the Government undertook to establish a tight budgetary system -f nupplying the company with the requisite funds, then it could agree to a fert - pricing scheme that would allow FERTIMEX to earn at least an 8% return -xLoots, by 1983. 12. In the ensuinq discussion, it became clear that the Go"er, Mexico was not prepared, in any way, to constrain its discret iunary' powei espect of pricing and, moreover, it regarded FERTIMEX (along with two - public enterprises) virtually as a branch of the Government, performi.. " E I i zed services and not as a fully independent entity whose purpose would b h achieve targets expressed in financial terms. In the end, the Mexican (overnment accepted a financial rate of return formula on the condition that no nmmitment be given on fertilizer price increases at consumer level and that o :<)tilizer price adjustment if contemplated, would take into account the res of cost benefit studies for various crops. This final understanding obviuus:v fell far short of the Bank's position as enunciated initially based on the Banli'i vrrepted policy stance. It was particularly inadequate given the Mexican Governent's repeated scatements regarding the role they assigned to FERTIMEX (a branch of Government performing specialized services). FERTIMEX remained in ietail?t of most of its financial covenants especially those relating to financial zatioS and rate of return throughout the implementation period. The financial weakness of FERTIMEX continued to affect its decision-making power with respect to the implementation of the project.3 In the event, the Bank and Government ol Mexico later found it necessary to prepare a Fertilizer Structural Adjustmect L'oan in Section 3.03, Guarantee Agreement dated May 18, 1979 as follo-s: "Unless the Guarantor and the Bank otherwise agree, the Guarantor shall take all action required on its part to enable FERTIMEX to obtain revenues in each fiscal year sufficient to cover operating, administrative and distribution expenses, including adequate provisions for maintenance, depreciation, taxes and other levies to provide FERTIMEX with a net income to yield, operating efficiently, a rate of return, before interest and after taxes, on its average net fixed assets in operations during the year (valued in accordance with Section 5.08 of the Loan Agreement) plus working capital requirements for normal operations, of not less than 3.5% per annum for each of the years ending December 31, 1980, 1981 and 1982 and not less than 8% per annum for the year ending December 31, 1983 and for each year thereafter." 3In its letter NAFIN states that: "The Audit Report places great emphas5is on the Federal Government's refusal to accept the price schemes and rEchanlsms proposed by the World Bank and claims that this hindered FERTIHFX's fina-cia1 restructuring. This is untrue, however, since the Federal Government pro'-ded the company with the funds it needed in the form of budget transfers to otint the consumer price subsidies." 5 order to correct many of the inherent financial weaknesses plaguing the subsector. D. Implementation Experience 13. The Project Completion Report gives a satisfactory account of project implementation experience. There were a number of factors which negatively and seriously affected the implementation and caused a very large overrun and a delay of nearly six years in the completion and commissioning of the plants. Although the financial and debt crisis had some influence on the implementation outcome, but t!e experience with this project illustrates, above all, how a series of missteps and the lack of a strong project management organization inflicted irreparable damage to an otherwise highly desirable and beneficial project. Some of the more important issues are reviewed in the following sections. 1979 Earthquake 14. Prior to the presentation of this loan to the Board on April 12, 1979, a strong earthquake with a magnitude of 6.5 on the Richter scale had shaken western Mexico on March 14, 1979. The epicenter of the earthquake was located around 50 kms from the site of the fertilizer complex. This earthquake caused considerable damage to some of the partially completed structures on the site. The consultants to FERTIMEX had reviewed the project foundation issue in June 197S and based on the preliminary site work, they anticipated to encounter foundation conditions similar to those at the SICARTSA steel plant located in the vicinity of the complex site. The foundation designs were, therefore, based on the information available for the adjacent steel plant and the anticipated soil conditions. However, following the earthquake, the FERTIMEX site exhibited considerable evidence of liquefaction leading to longitudinal cracks. Further studies indicated that the liquefaction problem was of such a magnitude as to require a redesign of all foundations. FERTIMEX, however, failed to keep the Bank informed of these activities until the Bank posed the question in a telex dated November 7, 1979. Significantly, there is no mention of the need for any earthquake-related remedial steps in the report of tne first supervision mission carried out in June 1979.' The matter is later picked up during the second supervision mission carried out in February 1980 when it is established that the completion of the project is likely to be delayed by 12 months (October 1982) and the cost increased by 20% because of the need for extensive piling and anti- earthquake foundation supports. Chance of Scope 15. Immediately upon the return of the second supervision mission from Mexico in early February 1980, the Board of FERTIMEX made a decision on February 14, 1980 to add a second stage to the Fertilizer complex (LAC 93) by duplicating the phosphatic fertilizer manufacturing plants. This decision was not communicated 4Supervision report dated July 5, 1979: "Project implementation started in February 1979 and is moving smoothly. No major execution problems are expected at this time. Bank engineering and procurement activities have been commenced. Site fill is 80% completed..." 6 to the Bank and wa; only difcovered durin the third supervision mission undertaken in August of 19805 This duplication affected especial)y the donign of the offsite facilitiei aknd the solids handling area because FERTIMEX was intent on implementing the new project simultaneously at the complex site. 16. The decision to simu)taneously implement another large fertilizer prnect at the Lazaro Cardenas site was certainly in contravention of both the spirit and the wording of the Loan Agreement. Even though the financing for this expansion was to come from the Treasury and no new loans were, therefore, taken by FERTIMEX, Section 5.05 of the Loan Agreement provided that: "FERTIMEX may make expenditures or commitments for expenditures for fixed or capital assets (including investments in or loans to other business entities) only to the extent that such expenditures or commitments for expenditures will not impair the ability of FERTIMEX to carry out and operate the Project and meet its other obligations under this Agreement." Elsewhere, in Section 3.04(a) of the Loan Agreement FERTIMEX was required to: "furnish to the Bank, promptly upon their preparation, the plans, specification, reports, contract documents and construction and procurement schedules for the Project, and any material modifications thereof or additions thereto, in such detail as the Bank shall reasonably request." The project files show that prior to the loan approval, when discussing one of the financial covenants regarding new projects, the Mexican Delegation had objected to the Bank having a virtual veto power over new investments above a certain level but accepted that "FERTIMEX would send a letter to the Bank (Annex A, paragraph 2) informing the Bank of its decision to make capital investments above us$20 million, and giving the Bank reasonable opportunity to comment on such proposed expenditures." Despite a statement in the Staff Appraisal Report dated March 29, 1979 that FERTIMEX had agreed to a financial covenant "(to) inform the Bank of all new projects requiring capital expenditure in excess of US$20 million", neither the SAR provision nor the letter ever became a part of the Loan Agreement. 17. In retrospect, the decision to simultaneously implement a second phosphatic fertilizer train consisting of sulphuric acid, phosphoric acid and DAP/NPK plants was reached t:ithout taking into account the workload existing at the time and the fact that most projects under implementation,including the first Bank financed project (Loan 1112-ME) were running behind schedule.7 It further taxed an overly-stretched project management organization with extremely serious results for tqe project. At the time of the third supervision mission undertaken in August 1980, the combined ef-ect of the change in the scope of the project and the need to strengthen the structural design against earthquake had pushed the anticipated completion rate to the end of 1982, indicating then a delay of around 1 years. 5There is nothing in the Terms of Reference for this mission to indicate that the Bank was even apprised of the Mexican Government's decision to expand the complex ffid-course. Memo to files dated January 25, 1979--Summary Minutes of Negotiations. 'This project, scheduled for start-up in June 1978 was eventually brought into production in July 1982 with more than 100% cost overrun. 7 Pro_ectManagement Structure 18. Based on the experience with the first fertilizer loan, the Bank advocated throughout the appraisal period the need for a task force approach to project implementation management. Essentially, this approach called for an implementation organization composed of functional managers with sufficient authority for quick decision-making, all working under a senior project manager. Under this set up the lines of authority and responsibility would be well defined and only very important technical arid financial matters would be referred to the higher authorities in the enterprise. During the appraisal period it appeared that the Mexican authorities had both understood and agreed with this approach. The Bank was sufficiently confident about this understanding as to indicate in para. 5.07 of the SAR that "Responsibility for overall management of the Project and supervision of the contractors will rest with FERTIMEX Construction Department, which has appointed a full time project manager acceptable to the Bank to lead FERTIMEX' project management team." 19. After the start of implementation, it soon became apparent that the project management structure was far from what had been anticipated. The report of the second supervision mission undertaken in February 1980 dealt with the issue and pointed out the need for corrective action. It went on to say that "FERTIMEX, however, recognizes the inadequacies of such an arrangement and has already identified a senior FERTIMEX manager to take over total charge of the Lazaro Cardenas project in the next few months. All. functional groups working on the project will thereafter be coordinated and managed by him." Despite several attempts at changing project managers and the project implementation organization the matter was never satisfactorily resolved because usually the project manager was not adequately supported with staff and authority to effectively manage the project. Moreover, the site construction management team was often inadequately staffed and at times the project manager also carried another portfolio with its demand on his time and attention. 20. The problems of inadequate project management was exacerbated by the inability of FERTIMEX to submit the periodic progress report as provided for in Section 3.04b of the Loan Agreement. Regular attempts, especially at the time of the supervision missions, to help FERTIMEX with the problem remained abortive until the beginning of 1982 when it became possible for FERTIMEX to submit reasonably up-to-date progress reports. These monthly progress reports for the early months of 1982 clearly indicated that the unsatisfactory performance was primarily due to insufficient availability of funds. Not only was the budget allocation for 1982 below requirement (and a further cut was being contemplated). the fund releases were only around 30% of the allocation during the early part of 1982. As a result and due to FERTIMEX's large payables, local suppliers and contractors were not willing to ship equipment or render services without substantial advance payments. By mid-1982 (around the onset of financial and debt crisis), supervision reports indicated that the project was less than 50% completed. (The PCR states that "capital expenditure schedule for the total project including the duplication project....shows that FERTIMEX had already spent in 1982 the amount required for the original project including a large part of the cost overrun".) At this time, project completon wris anticipated for the end of 1985, i.e. some four years heyond the pp:fe) compi(-etion dcte. 8 21. The above problems were communicated in a letter from the Bank to the Mexican Government in August 1982 upon the return of a supervision mission. It was suggested in this letter that not only was the project in serious jeopardy (the reestimated economic rate of return had already dropped from 12% to 4%), but it was also essential, given the financial crisis encountered by Mexico, to give clear priority to the Bank-financed project over its duplication (LAC 93). The Bank further suggested that "If, however, feasible alternatives are to be evaluated, the one with comparatively favorable economic benefits should be clearly established. For this purpose, it is necessary that alternative options are identified and their economic impact carefully analyzed before a decision is made in this regard." 22. FERTIMEX took strong exception to the Bank's suggestions about the inadequacy of the project management structure and other points raised in the Bank's above-mentioned letter. It assigned the cost overrun to: (a) the change in the scope of the project and its displacement from a site at Guayamas to Lazaro Cardenas including the need to redesign foundations in view of the soil condition and the earthquake prone nature of the region; (b) higher equipment cost due to inflation in excess of what was anticipated at the time of appraisal; and (c) extension of some facilities such as the wharf and warehouses when a longer sea frontage became available to FERTIMEX. It strongly disagreed with the suggestion that the delay was responsible for part of the cost overrun, estimated at over 80 percent by the end of 1982. 23. In terms of project management, it put the blame squarely on the shoulder of the Manager of the Project, who, it inferred, was appointed at the suggestion of the Bank.9 It contended that the Project Manager had been given all the support, but that he was incapable of making decisions and that FERTIMEX was making an evaluation of the situation. It further contended that FERTIMEX did not need any outside assistance (as suggested by the Bank) to improve its implementation and construction management effectiveness "since most of the personnel assigned to this project had an average of 15-20 years' experience of project execution, quite enough time to reaffirm their professional competence." The unavailability of sufficient funds, especially in 1982, was mentioned as a debilitating constraint. FERTIMEX also pointed out that the question of a phased implementation schedule and the prioritization among the original project and its duplicate was being studied by a commission. Thus, problems remained unresolved throughout 1982 at which time it was anticipated that the delay in project completion would extend to at least four years. 24. With a change of Administration and the appointment of a new top management at FERTIMEX came a change of approach. Given the Mexican Government's request for an extension to the loan closing date, the Bank made a further 8Letter dated August 18, 1982 to Subsecretario de la Industria Paraestatal, Secretaria de Patrimonio Nacional y Fomento Industriel. 9Letter dated November 8, 1982 from FERTIMEX to Subdirector de Organismes Financieros Internar-ionales - Direction de Financiamento Externo de la Secretaire de Hacienda y Cr6dito Publico. 9 attempt at resolving the project management issue. By February 1983 (the occasion of the Banks seventh supervision mission), "necessary arrangements had been made more in line with the Bank's recommendations and the Project Manager had been confirmed in his position to head a task force team removed from their functional line managers and dedicated to the Project." Thus, it took almost four yeats after the approval of the project for FERTIMEX to put into effect a project management organization that had been agreed upon at the time of appraisal. Financial Constraints 25. As noted above, Mexico's financial crisis began to seriously affect FERTIMEX from the beginning of 1982. This was in sharp contrast to the earlier availabMlity of substantial funds in 1979/80 which came in the wake of the oil price incease and resulted in the approval of the duplicate project (LAC 93) by FERTYMIX. At the time of approving the duplicate project FERTIMEX already had a hoct of qmaln and medium si7ed projects in hand. 26. By early 1982 both the cut in budgetary allocations and the slow disbursement by the Government began to affect payments to contractors and suppliers and produced a backlog of accounts payable. Given the uncertainties surrounding the availability of funds, FERTIMEX found it very difficult to plan expenditure on an array of projects which were under implementation. In the event, FETIMEX probably decided to allocate more funds for projects that it consideted to be nearer completion. Although the Bank had suggested in August 1982 that this matter be reviewed with FERTIMEX, no serious discussions took place. FERTIMEX remained heavily dependent on Government transfers to finance all categories of expenditures to sustain its operations, investments and to meet its debt repayment obligations. 27. As a res.ult, the uncertainty and irregularity in fund receipts during 1982 ard 1983 administered a severe blow to the implementation of the project causing a further delay of at least 1 years in the completion of the project. This, of course, necessitated some reordering of equipment and rearranging of contracts, with attendant cost increases. In fact, according to the PCR the Bank's staff consider that at least one third of the delay in project completion can be traced back to the financial problems faced by FERTIMEX, mid-way through implementation. The problems associated with the budgetary cycle and the annual allocation of funds remain a feature of the Mexican financial scene with no clear solution. FERTIMEX's income statements show very clearly that the corporation remains cash strapped with the government reducing its financial transfers (even in nomiria) terms) to FERTIMEX over the last few years (Annex I and II). 10With no financial autonomy FERTIMEX had to depend for 54% of its cash expenditure on Government transfer in 1983. In 1984 alone, budgetary transfer for direct operating subsidies amounted to US$210 million. In the same year, the budgetary transfers for capital investments, debt repayment and recapitalization was US$96 million, US$98 million and US$5 million respectively. This was in addition to an indirect subsidy of US$150 million from PEMEX on account of the low irmuricmnia prices. 10 1985 Earthquake 28. The project implementation picked up considerable momentum in 1984 and 1985. This was the result of the restructured project management team, the atter,tion given by FERTIMEX's top management to the need for speedy completion and the fact that most other investment projects had already been completed. Just as the project was approaching completion, another earthquake struck the region in September 1985. This second earthquake resulted in extensive damage to the plants and utilities as well as the pollution control system. In addition, an important part of the technical information (especially drawings and purcha3e orders) were destroyed in Mexico City. It was estimated that this earthquake caused a further 10 months' delay in the final completion of the project. E. Project Cost and Procurement 29. The project had an overall cost overrun of 68%. The final cost of the project including interest during construction amounted to Us$507.8 million against an estimate of USS301.3 million at appraisal. (This is similar to the previous project financed by the Bank (Loan 1112-ME) in which the cost overrun amounted to 112%--from US$150.2 million at appraisal to an actual of US$318.8 million.) Given the lack of separate accounting for the funds allocated by the Government, the PCR finds it difficult to trace the origin of financing. The foreign costs are, however, estimated to have risen from US$124.5 million at appraisal to US$211.9 million at completion while the increase in local costs is from US$176.8 million to US$295.9 million. Long-term loans other than the Bank loan are assumed to have financed the remaining part of the foreign exchange costs of the project. 30. All the issues discussed under project implementation above had their counterpart in rlelays and cost overruns. The need for a new foundation design, the change in the scope of work, currency realignments, price escalations because of prolonged delay in implementation, repairs after the second earthquake and other equipment problems, and the relative inefficiency of the capital cost monitoring system resulted in a serious cost overrun. In the event, the concurrent implementation of the duplicate project did not lead to any cost economy because of the long delay in bringing the whole complex into production. 31. Procurement of equipment and services financed under the Bank loan followed the Bank guidelines in respect of International Competitive Bidding. Mexican machinery manufacturers and consultants also participated in the supply of a considerable part of the equipments and engineering services. Although the ICB procedures were applied successfully and without much difficulty, problems such as delay in placing purchase orders and reorders, furnishing technical design information to suppliers, slow expediting and inspection were in evidence through the implementation period. Much of these problems had their roots in the inappropriate project management structure obtaining until 1983. Upon the change in thio structure and the delegation of more authority to the project team, the procurement performance was vastly improved. 32. lihc pe formance of engineers and consultants for the five process plants (both Mexica,n ad expatriate) were generally satisfactory. But FERTIMEX faced 11 considerable difticulty with the utilities and solid handling components which were under its own sole supervision. This was due to the substantial work load carried by the Mexican engineering firms during this period which resulted in wastage and delay. The construction of the plants which was carried out by some 25 large and 10 small indigenous contractors also suffered from many causes including the absence of local skilled manpower, difficulty of procuring construction material and shortage of funds. FERTIMEX was indeed facing a difficult coordination problem with so many erection and civil works contractors. F. Operational Experience and Outcome 33. The commissioning and start up of the offsite and utilities commenced by mid-year 1985. By the end of that year, water treatment, compressed air and steam generation plants were operational. While the cooling tower was mechanically completed in December 1985, it started commercial operation in May 1986. The process plants namely sulfuric acid, phosphoric acid, DAP, nitric acid and ammonium nitrate plants came on stream in May 1986, August 1986, October 1986, April 1987 and July 1987, respectively. The duplicate phosphatic fertilizer train (LAC 93) also came into production in July/August of 1987. 34. The PCR deals in depth with the start up problems in the different plants. FERTIMEX encountered many problems at commissioning because of the mechanical troubles with equipment damaged during long storage periods. The use of Baja California phosphate rock caused some start up problems and reduced the phosphoric acid plant output with the result that FERTIMEX now uses a mix of imported and indigenously-mined phosphate. The disposal of gypsum from the phosphoric acid plant into the sea still remains an unresolved problem as both discharge lines have ruptured and become unusable. This matter will be further discussed under environmental impact below. 35. Annex III shows the production build up for the fertilizer complex since commissioning of the plant. The actual production over the last three years has been considerably below what was expected in the PCR. In addition to the start up problems enumerated above the quality of Baja California phosphate rock remains somewhat of a bottleneck in increasing the phosphoric acid plant production to capacity. Equipment failure also remained a large factor in shutting down plants and reducing capacity utilization. Much of the shortcomings have now been rectified and production is approaching design capacities. 36. The expanded project (original FERTIMEX 17 and the duplicate plant) generated around 2,000 jobs, albeit at a very high cost of about $400,000 per job. This is not unusual in such highly capital intensive projects. There was not an inconsiderable technological impact in that much design work including practically all detailed design of plants and offsites were carried out in Mexico as well as the fabrication of many components. There will also be substantial foreign exchange savings, even though the figures will not match those anticipated at appraisal. 12 Environmental Issues The environmental impact of this project was clearly recognized from the gJtining and measures were introduced to reduce the pollution effects to repted levels. By selecting modern processes such as the double absorption Ifuric and nitric acid plant design, low emission of sulfurous and nitrogen ides was ensured. Initially, the project design included a 60 hectare marshy for the disposal of gypsum produced by the phosphoric acid plant. The !qned site was, however, located at a distance of 8 kilometers from the plant -t. This arrangement was revised along with the first design modifications 1'ingq into account the earthquake results. The studies conducted at the time ummended that the gypsum (after dilution) be discharged into the sea at a tance some 1.5 kilometers from the shcre. This change in the method of disposal never received the approval of the 1opriate department in t1e Bank. Late in 1983, the Office of Environment -'icated that the plan to discharge the diluted gypsum into the sea was in !radiction to the World Bank Environmental Guidelines (which were available the fertilizer industry at the time of the appraisal) because it represented unacceptable environmental hazard." After qu-stioning some of the conclu- iQl of an environmental impact study carried out for FERTIMEX, the Bank -uggested that FERTIMEX retain an independent expert consultant to review the anign and operation of the proposed system. In addition, the Bank recommended hat "An environmental program should be implemented now in the diffusor outfall tea to determine baseliie conditions for both benthic and pelagic species, as el1 as the commercial fishing potential. During plant operation, a continual ,Z)h1pling and analysis program should be maintained to determine, inter alia, >Rpnic, phosphorous and nitrogen, both in water column and bottom sediments, pH .-itoring in the water column should be performed as well." These suggestions '- offered, not as an endorsement of the adopted design, rather as a means of v;iing minimal assurances of environmental protection. Q Since the commissioning of the plants, this system of gypsum disposal has .en plagued with a series of mishaps ranging from the blockage and rupture of -ipes, to vibration problems associated with the mixing pumps (the gypsum is iluted by effluents from the evaporation condensers and fertilizer plant -rrubbers). While the Bank's worst fears have come to pass, it appears that the qecretaria de Desarollo Urbano y Ecologia (SEDUE) in Mexico is keeping a careful 'ye on this matter and is pursuing the issue for a satisfactory outcome. The 'IV i ronmental question is further being pursued through the implementation of the 1prtilizer Sector Adjustment Loan. U. Financial and Economic Results '0. At aiscussed in para. 12 above, and despite the provisions of several loan covenants, the unsatisfactory financial situation of FERTIMEX remained ,1;j-hanged throughout the period of project implementation and after commissioning ofT the plant. The PCR gives a detailed account of FERTIMEX's financial "Memo dated October 17, 1983 from Environment Office to Industrial Projects Nepartment. 13 performance and its compliance with the financial covenants. Given the fact that FERTIMEX continues to depend on the Government for subsidies under a very tight liquidity control, any attempt at reestimating a new financial rate of return would be largely meaningless. Even the PCR's financial rate of return calculations which was based on the price assumptions made in the Fertilizer Sector Adjustment Loan (FSAL, approved in March 1988) have Leen overtaken by events. In practice, price increases were not effected as expected in the FSAL while actual production proved substantially short of forecast. It is, therefore, nxpected that the revised estimate of -0.3 percent as given in the PCR would be on the optimistic side unless radical changes in the relative price of inputs and outputs take place and FERTIMEX is allowed to take advantage of such a situation. 41. The ERR calculations carried out in the PCR (pirre. 6.3) again assume figures for the performance of the plants in 1988, 1989 and 1990 that are materially above the actual production. Furthermore, while the economic prices assumed for the output remain very close to actual, the economic nrices of major inputs (phosphate rock, potassium chloride, ammonia) were between 10-20 percent higher during the last three years. This would indicate that, based on actual performance of the plant since start-up and the economic prices, the ERR is likely to be considerably lower than -1.0% as calculated in the PCR. While it is true that price forecasts at appraisal for inputs and outputs were higher than those obtained in practice (ANNEX IV) and that the poor quality of the domestically supplied phosphate rock has adversely affected the operational efficiency of the phosphoric acid plant, the major reason for the disastrous drop in ERR must be sought in the long delay in implementation and the enormous cost 12 overrun. 42. The PCR also calculates the effect of the project in terms of foreign exchange savingo (para. 6.5). While it was anticipated that, at full production, the foreign exchange payback period would be around one year, the reestimation shows a period in excess of two years. this would still appear to be the most important positive outcome of the project in addition to the availability of large quantities of fertilizer products to the Mexican agricultural sector. The plant uses much of its inputs from the indigenous sources, but still needs to import all its potassium chloride and a considerable proportion of its high grade phosphate rock requirements because of the mediocre quality of the indigenous phosphate supplied by ROFOMEX. Its output is estimated at US$150 million in 1993 (at 1987 constant prices) with the foreign exchange saving approaching US$80 million. In its letter NAFIN states that: "One of the aspects most responsible for the nil return on the investment and not analyzed in sufficient depth in either the Audit Report or the Project Completion Report is the low prices of phosphate fertilizers compared to the high prices of the raw materials, this being the prevailing situation on the international market and one that has tended to worsen in recent years. This is a substantially different situation from the one reflected in the original assumptions on which the project's initial rate of return was based. Added to this is the fact that in the present circumstances no phosphate fertilizer project based on phosphate rock purchased on the international market would offer any prospect of economic viability." X. Aut a i nnhi 1 itv~ 43 In the cout sn ef Ix u u-t np3 eint at 1 oi t hp Batk repest i mated t.he economic rate of 1,<*Lun in Auut 108 i which t imp dolaye had pushed the then expected com.pl on rlt, toh end A 1981 with a uot ovoi run of around US$200 million. Thin valcolia on BhoNnd tt under the asnuption made for the prices of inputs awd outputs, Uhu evtmluntlOWd ERR uould diop to 5./ percent. It was then argued: "houuvc.>, I t he pi oje(t in considered on a "sunk--cost" basis, with capital o p aiorpq through end Juno J98] Pxclurd, the tevisecl base ERR is a more Ov i - , \ 1 1 '' '* Vi q 1e:aslculation al o showed that out of the two major facLou aifecting the ERR, the capital cost increase had a much 3arger share in reduc i ng V he ERR tha inple.ientat ion s ippage. " 0 This prompted the Bank to conuidor ch projpct ;o nar.'ira) and to urge the Government of Mexico to atrnythen the ip1VmWi aWOP A1rnqPmHPW n a"d management of the ptoject. 44. By the PnO of 1982, given the fact that the reestimated ERR had dropped to nrounfd 4% nro the implementation prowlems were nowhere near resolution the Bank .heO 1o oot'"n of caoeing e a vndipbarsed loan balance and, thus, no longot being agsociated with ih, ptoject. Again in February 1983, a supervision mission Uck-to-Officp Roputi ndunted chat "although the project has lost part of it, economic vitrautivLnes. it is still mauginally economic and would yield a satiunfactory return on v qonk--ront basis." The Supervision Mission report, therefore, concluded that, Hence, in economic terms, continuation of the project execution 10 junstiKed and deserves Rank support, provided FERTIMEX takes immediate StCpf to reL'ti$y the presently unatisfactory project management arrangements." Thin judgement was based on the assumption that the plant would opurat.o at 70%, 851, 95 and 1001 of uapacity during 1986, 1987, 1988 and 1989, reepectively. Beyoyd thin point, there was no furtter attempt to reca3culate the ERR until the Project Completion Report was prepared. In fa .t, the project implementation had picked up spepd by late 3983, tne judgement remained the same. Nonetheless, the mid-1985 Supervision Mission Report suggested that "on the basis of current. prices and imported phosphate rock it seems unlikely that the phosphate ferti)Wzer operation, bulk of the output, will do better than cover c.sh conts. 45, In view of the above background, the sustainability issue can be approached from two viewpoints: from a technical point of view, it now appears that production has, of late, approached the design capacity and FERTIMEX's technica) and operating staff clearly have the capability of operating the complex efficiently. Furthermore, the nitrogenous and phosphatic fertilizer units have been constructed with the right gcale and up-to-date technology. The operating personnel has been able to debug the units and much of the previously damaged equipment has been repaceO. 46. From a financial point of view, even though FERTIMEX's financial Hitution has nut improvod enormously, the Government is moving fairly rapidly in th diroction of correcting fertili7e- ptices and bringing them in line with intcr&tonn) prievu a q \un W-3-w; .Memo on Xe-.n 4.', na"e nf ;Irin dated August 6, 1981. 1 8 VEillTI I EI PR I (Hu)h -U $/.n Domest ic to Dms c Internati onzal Do-., rtie DIes i in at ioa ) 111lteCrnatII ial Price Ra lo P1ice Ict Vrico P ic, Rati o Fertilizer i i:1 . 9I End Hi : ': :q M dI-Dec 1990 . Dec. 1990 Ura 9 110 1(/ 175 93 Ammonitu Nitrate '3 12 ;92 Ammnonic . ,;4 165 99 Amonium Sulfato 11 60 1. 90 93 Triple SuperphoophaLe (9 Il 11 205 98 Single Superlhoaphate 36 r , A2 121 65 Potatai- C) orido 81 125 127 100 Potaoitum S"Ifito :08 2 3 2 ?,3 106 WiArmonium Phoaphat 13 19 '24 225 100 Moreover most Input pl-ic". '. or. I p; it ; e c o kl1ur lhave been dec ontrolled. This trend indi1ate -haT :n yeprs to 0C01110 A0fld vith the restructin-qng succe8sful- ly comp)eted, UERTIMEY will be )C3.,; dependent o. the Government than at the present time. Miuch, of cou-isr, I depend oii the pricing policy of the government and the course of ini-c.-: icnail pr cesC (which are al the moment depressed) rendering the sust alnaho. ity of S 1h1 pi-oJi-ct uncertain. But assuming that the FSAL covenants are corp1. :d vith, the Lazao Cardenas complex will remain a sustainable -po 'I':m fe, 2W ' *bC int': n 3largely considered ar sunk co;t. 1?h, ) (ptn c ,p .:a.t"' ly, I fh e I'! N Ii C be . poFitive cash margin to defray a paii cf the finauC carts rnd debt service. J. Role of the ;onk 47. The Bank's role was por:., nla LY I i v:s a in the choice of process technology, offsites, infrastructure and cnov e;-io( e) aspet: of the project and the institution of the procurmiaer' proccd.t -. lank Jurthe. co, 'buted to the decision to carry out a fervriL.:zer distrit. arid transportation study and the need for an operational staff training prog-a,. 48. Despite the Bank's initial insistence or measures to be taken to improve FERTIMEX's financial situation especially the need to insulate it form the debilitating influences of the Government's fert: ar prJeIng policy, it had to resign itself to a covenant providing for a graduated upward rate of return target. Most financial covenants were not adhered -o and the Bank was unwilling to enforce covenants through suspension of disbursements, Nor did the Bank seriously follow-up on the findings of the supervision missions except with the exchange of correspondence. It should also be noted that the Bank'b forecast of the strength of the fertilizer market was somewhat optimisti as prices collapsed in the 1980s. K. Findings and Lessons 49. The PCR brings out many important point in the Conclusions section. While confirming chose conclusions, it can be summarized that the implementation of this project suffered from the three basic factoror (a) The project came at a time when FERTI1EX vpp rlveacy heavily charged with other investment activities some of -i-cl )cie fdling seriously behind schedule. This was exacerbate:J -hen IUTi TLX ducided to duplicate the phosphatic fertilizer capa - ' aeo CArdcnas in 16 1980. the company just did not have the capacity to deal with the work load and was unwilling to set up an efficient and workable project management organization. Moreover, the Bank and FERTIMEX's management did not fully realize the seriousness of the decision to enlarge the scop, of FERTIMEX's investment activity and took no measures to cut it down to a manageable level. (b) Perhaps, in a desire to lend, the Bank weakened its position in not insisting upon more assurances especially in respect of pricing and other policy issues, financial transfers from the Government and the need to clear new investments with the Bank. This is especially troublesome when a covenant discussed in the SAR about new investments exceeding US$20 million is dropped form the loan agreement. The Bank's unwillingness to take a strong stand vis-A-vis non-compliance and in the face of unsatisfactory implementation performance (not to mention surprises such as the approval of LAC 93) must be construed to have helped the disappointing outcome of this project. (c) The implementation hit a snag in that the Mexican financial and debt crisis came when the project was still in its earlier stages of implementation. This and other external factors (two earthquakes, overloading of contracting and consulting services, etc.) 7:ertainly affected the implementation outcome. But inspite of the Bank's entreaties for a prioritization exercise, FERTIMEX decided to expedite implementation of other capital investments at the expense of the Bank- financed project. The logic of these decisions was never discussed with the Bank and the dialogue with FERTIMEX continued through very infrequent exchange of letters. Cost overruns are but one of the consequences of this type of practice. 50. The PCR draws some very useful lessons from the experience with this project. There are other lessons to be learned in order to avoid the problems encountered in projects of this nature. (a) The Bank should make sure at appraisal that when the borrower's implementation capability is limited or just sufficient to carry oat the project, that no other major capital expenditure is allowed to be carried out in parallel. (b) The Bank needs to insist on being consulted in earnest when a new capital investment is being contemplated. In case the Bank's views are not respected, the Bank needs to discontinue the project if in its opinion, the execution of its financed project would be unfavorably affected. A problematic project of this nature also needs more frequent supervision for early detection of shortcoming and the putting in place of remedial actions. (c) The establishment of a suitable project management structure dedicated to the task at hand should become a 'condition sine qua non' of 17 proceeding with the financing. Time and time aga!n, this has b- shown to be the difference between success and failure. Even FERTII) finally accepted that "project administration was not adequa, (attached PCR prepared by FERTIMEX, Section 7). (d) Whenever there is need for outside consultants to support the pro) implementation team, especially in areas such as planning, conti expediting, inspection, reporting, etc., their appointment should insisted upon. (e) Of special importance is the putting in place of a cost monitoring control organization which can provide accurate and timely informatI to the project management and the Bank project officers. (f) The Bank needs to ensure strict compliance with the covenants and . spirit of the loan agreements instead of bending backward at t-ire. save a project. 19 ANNEX I MEXICO LAZARO CARDENAS FERTILIZER PROJECT FERTIMEX INCOME STATEMENT (in billions of current pesos) 1987 1988 1989 1990 Production In thousand tons of product 4,110 4, 4,362 4,756 Sales Revenue 539 1,212 1,288 1,813 Total Operating Cost 806 1,745 2,081 2,791 Cash Cost of Sales 531 1,175 1,207 Distribution Cost 169 322 450 Administration Cost 44 93 107 Depreciation 62 155 314 Financial Charges 150 242 253 405 Grand Total Coste 956 1,987 1.334 3.196 of which cash total costs 894 1,832 1,020 2,508 Cash Deficit (355 (619 (728 (695 Government Transfers 401 508 459 431 Operations 277 327 299 149 Interest Payment 124 181 160 272 Cash Margin After Transfer 46 111 (269 (273 Total Cash Cost/ton ('000s pesos) 218 455 462 527 Total Cost/ton ('000s pesos) 233 494 535 672 20 ANNEX II MEXICO LAZARO CARDENAS FERTILIZER PROJECT FERTIMEX INCOME STATEMENT (in millions of $US) 1987 1988 1989 1990 Exchange Rate (pesos/$US) 1,378 2,273 1,461 2,850 Production (in '000s of tons) 4,110 4,025 4,362 4,756 Sales Revenue 391 533 524 636 Total Operating Cost 585 768 846 979 Cash Cost of Sales 385 517 490 Depreciation 45 68 129 242 Distribution Cost 123 142 183 Administration Cost 32 41 43 Financial Charges 109 106 103 142 Grand Total Cost 694 874 949 1,121 of which Cash Total Cost 649 806 820 879 Cash Deficits (258) (273) 296) (244) Government Transfers 291 224 186 148 Operations 201 144 121 52 Interest Payment 90 80 65 96 Cash Margin After Transfer 33 (49) (L110) (96) Total Cash Cost/ton ($US ) 158 200 186 185 Total Cost/ton ($US ) 169 217 218 236 21 ANNEX III MEXICO LAZARO CARDENAS FERTILIZER PROJECT ACTUAL PRODUCTION OF LAZARO CARDENAS FERTILIZER COMPLEX (Z of rated capacity) Annual Rated Unit Capacity 1987 1988 1989 1990 ('000 tons percent percent percent percent per year) Sulfuric Acid 1,320 56.6 55.5 60.9 79.5 Phosphoric Acid 396 56.0 46.6 62.4 85.1 DAP/NPK 1,050 39.9 35.6 53.1 81.2 Ammonium Nitrate 270 22.2 65.9 71.6 85.2 22 ANNEX IV MEXICO LAZARO CARDENAS FERTILIZER PROJECT COMPARISON OF ACTUAL AND FORECAST INPUT AND OUTPUT PRICES ($/ton - 1978 constant prices) 1981 1982 1983 1984 1985 1986 1987 1 1989 1990 INPUTS Phosphate Rock (Appraisal) 49 49 50 52 54 54 54 54 54 54 Popteoct(Actual) 40 34 31 32 29 24 20 22 25 24 Potash (Appraisal) 83 84 86 88 91 91 91 91 91 91 (Actual) 90 67 62 72 71 49 45 54 60 56 DAP (Appraisal) 203 206 221 238 249 249 249 249 249 249 (Actual) 156 149 153 161 143 110 114 120 105 99 Soure*s Staff Appratsl Report for forecast prices. Price Prospects for Major Primary Commodities, World Bank, Report 814/90 for actual prices. 23 PROJECT COMPLETION REPORT MEXICO FERTILIZER II PROJECT (Loan No. 1686-ME) April 19, 1988 Trade, Finance and Industry Division Latin America and the Caribbean Region Country Operations, Department II 25 I. INTRODUCTION 1.1 An important long-term policy of the Government of Mexico has been to promote agricultural development through irrigation projects, agricultural research and extension work, and by making agricultural credits and fertili- zers available to farmers under favorable conditions. In particular, the Government has sought to maximize local production of fertilizers, protect farmers from price fluctuations, through controiied fertilizer prices and provide fertilizers at highly subsidized prices. 1.2 The fertilizer industry is also viewed by the Government as a strate- gic industry in the context of its overall objective to make effective use of its energy resources and to achieve agricultural self-sufficiency in basic grains and cereals. With the exception of small quantities of ammonium sulphate and specialty fertilizers, all fertilizer production in Mexico is in the public sector. Petroleos Mexicanos (PEMEX), the National Mexican Petroleum Company, produces all the ammonia used in Mexico and FERTIMEX produces all other fertilizer intermediates and all finished solid fertili- zers. FERTIMEX is also the primary agency responsible for fertilizer market- ing and distribution in Mexico; its main objective during the past fifteen years has been to increase fertilizer production and deliver it to farmers at subsidized prices. 1.3 By the end of 1986, FERTIMEX had 53 plants (of which 25 were finished product plants and 28 intermediate product plants) in 12 different locations throughout the country, producing nitrogen as well as phosphate and complex fertilizers and their intermediates except for ammonia. Thirteen of the finished product plants and 12 of the associated intermediate plants started operating prior to 1970. Most of the plants are not economically viable. Most of the unviable plants are producing low concentration fertilizer (ammonium sulfate, single superphosphate) and represent 29Z of total product capacity.l/ 1.4 The Fertilizer II project was part of the government's import substi- tution program to ensure that Mexico would have sufficient capacity to meet expected increased demand for domestic phosphatic and nitrogenous fertilizer through the early 1980s. Through Fertilizer II, the government also intended to correct the imbalance in the Mexican fertilizer industry, which exported intermediate products such as ammonia and phosphoric acid, while importing at the same time substantial quantities of finished fertilizers to meet domestic demand. The project involved the building of a new fertilizer production complex at Lazaro Cardenas, a new industrial center on the Pacific coast, close to the main fertilizer consuming regions of the country. 1/ For further details, see Mexico. Fertilizer Sector Restructuring Program Report, Internal paper, Industry Department, April 29, 1986 and Mexico, President's Report on a proposed Fertilizer Sector Adjustment Loan of US$285 million to National Financial (NAFINSA) S.N.C. Report No. P-4639-ME February 1988 (referred in this report as Fertilizer Sector Adjustment Loan or FSAL). The project was approved on March 15, 1988. 26 1.5 The Bank had earlier supported the government effort with a first loan of US$50 million made in 1975 to NAFINSA/FERIMEX ("Fertilizer I project" Loan 1112-ME) implemented by FERTIMEX. The first project completed in MUy 1983. included (a) a 330,000 tpy urea plant at Bajio; (b) a 495,000 tpy urea plant at PAJARITOS; and (c) an expansion and modernization of an existing parathion plant at Salamanca. A PCR dated December 28, 1984 was prepared by the Industry Department of the World Bank. Some of the implementation problems, particularly management problems reported on this PCR were also experienced in Fertilizer II under review. 1.6 Fertilizer II had a difficult implementation, resulting in long delays and important cost overruns. This report describes the main problems that arose in the eight years between the beginning of construction (1979) and commercial operation (1987) focusing on implementation management, operation experiences and environmental issues. It will highlight some major events that have contributed to implementation problems: (a) an early FERTIMEX decision to duplicate the phosphatic fertilizer production of the plant and change offsites and utilities project scope; (b) 1979 and 1985 earthquakes; and (c) the deterioration of the economic situation in Mexico and of FERTIMEX's financial situation causing shortage of funds available for the project. These events had an effect on implementation schedule, project cost and initial operation. A separate section deals with environmental aspects and the temporary malfunctionning of the pollution control system. II. PROJECT BACKGROUND AND OBJECTIVES A. Project History and Objectives 2.1 -The Project was first identified in October 1975 and was initially planned for the site at Guaymas, Sonora, adjacent to a planned new copper smelting facility of Mexicana de Cobre S.A. However, following a decision to delay the smelting project and the change in Government in December 1976, FERTIMEX (formerly known as GUANOMEX) reappraised the Project. Based on FERTIMEX recommendation the Government granted top priority in mid-1977 to the construction of the project at a new location (Lazaro Cardenas), with a 20% increased production capacity. 2.2 Following a request from the Government of Mexico and FERTIMEX, the revised project (in its new scope and site) was pre-appraised in November/ December 1977 and appraised in August 1978 on the basis of a feasibility study prepared by FERTIMEX.2/ The engineering contracts for the implementation of the Project were already awarded and registration of equipment suppliers and preparation of bidding specifications were in progress with a view to start commercial operations by October 1981. 2.3 On April 12, 1979, the Board approved a loan of US$80 million equiva- lent to NAFINSA (a financial agent for the Government) and FERTIMEX at an interest rate of 7? p.a. and maturity of 15 years including 4 years of grace. The Government guaranteed the loan and agreed that NAFINSA would charge 2/ A first study was developed by FERTIMEX with the help of an international consulting firm. 27 FERTIMEX a 32 p.a. fee, bringing the effective interest rate to 10% p.a. FERTIMEX would carry the foreign exchange risk. The loan became effective on May 20, 1979. 2.4 The main objective of the project was to provide the Mexican agricul- ture sector near a major consuming region with an assured supply of high quality fertilizer produced from domestic raw materials; other objectives were (i) to improve the balance of payment by reducing fertilizer imports; (ii) tc build a major industry in Lazaro Cardenas, a new industrial center close to the market and away from the already heavily industrialized gulf coast and major urban centers; (iii) to create employment and (iv) to provide technology transfer opportunities beneficial to Mexican firms during the project imple- mentation. B. Project Description 2.5 Fertilizer II Project was designed to include the following main facilities on a 120 ha site at Lazaro Cardenas: (a) a fertilizer granulation unit with a production capacity of about 525,000 tons per year of diammonium phosphate fertilizer and/or NPK compound fertilizer; (b) a solid ammonium nitrate unit with a production capacity of about 270,000 tons per year; (c) a sulfuric acid unit with a production capacity of about 660,000 tons per year; (d) a phosphoric acid unit with a production capacity of about 198,000 tons per year; and (e) a nitric acid unit with a production capacity of about 215,000 tons per year; (f) storage and products handling facilities for all raw materials, intermediates and finished products (except for amonia to be delivered from PEMEX' adjacent facilities); and (g) utilities and infrastructure including (i) generation of all steam and critical power requirements (13 MW), the remaining power needs being supplied by the local power authority; (ii) water intake pumping station and pipeline; (iii) road and rail extensions to the site; and (iv) a berth at the port of Lazaro Cardenas. 2.6 All process units were to include facilities adequate to protect the environment from pollution. 2.7 The main raw materials needed annually in the project were (a) 200,000 tpy of sulfur (at full production) supplied from the Azufrera Panamericana S.A. Mine at Jaltipan, Veracruz and transported to Lazaro Cardenas by railroad; (b) 680,000 tpy phosphate rock initially imported from Florida and/or Morocco to be replaced as domestic Baja California rock of 28 -, tisfactory quality produced by ROFOMEX becomes available; (c) 205,000 tpy m!unonia supplied by Pemex from Salina Cruz and to be stored in a 20,000 tons .;,orage tank to be built by PEMEX in its existing storage terminal adjacent to iie project site; (d) 63,000 tpy imported potassium chloride for NPK formula- on; and (e) domestic diatomaceous earth to be used as coating agent. The rstic rock by ROFOMEX has become available in limited quantity, currently )mt 500,000 tpy. YTT. PROJECT IMPLEMENTATION, INSTITUTIONAL DEVELOPMENT AND MANAGEMENT 1 The Fertilizer III project was completed and commissioned in June 4R7, 69 months behind schedule. While commercial start up of DAP production ook place in November 1986, ammonium nitrate production started only in optember 1987. Project Scope Changes .;2 Major changes in project scope took place during implementation of ihe project and had an important effect on implementation delays and cost ,-verruns. The main change in the project (the original project is known as LAC 50) resulted from the FERTIMEX decision to duplicate the phosphatic ferti- izer production by implementing simultaneously at the project site twins :PK/DAP fertilizer and upstreams sulfuric and phosphoric acid plants (LAC 93). he decision in turn required: (a) the expansion and redesign of the offsites Itilities-and services; and (b) a very important expansion of solid materirls borage, handling and bagging to take into account the increased plant produc- 'ion and 750,000 tpy of urea to be imported in a first phase and produced at ,ite later. ".3 Other important changes were: (a) need for higher seismic standards 4or civil work after the 1979 earthquake; (b) the shift from a pond for the disposal of the gypsum produced by the phosphoric acid plant to an outfall dilffusor to dump it in an existing marine deep at 1.5 km from the cost; ,c) the late shift from pneumatic instruments to electrical instruments (1982); (d) the addition of two 3,500 tonnes capacity each ammonia spheres. ). Project Management 3.4 The implementation responsibility including procurement, construc- tion, management, project scheduling, cost control supervision and control of the enCineering firms, etc. rested entirely with FERTIMEX. FERTIMEX felt that it had adequate capability to implement the project on its own and the Bank ws in agreement with FERTIMEX. 3.5 During implementation the project went through the following main p-oject management structures: (a) up to mid-1983 management of the project was organized in functional lines; each function (construction, engineering, accounting, procure- ment, etc.) was carried out from the respective functional divisions at head office. While this can be effective for small projects, it 29 is ineffective for such a major and complex project. The problem of coordination was exacerbated by (a) the absence of a project manager with overall responsibility for the project. Contrary to the state- ment in (para. 5.07) of the SAR that a full time project manager had been appointed; and (b) during this period planning, programming, management function and reporting were totally inadequate partly because of the lack of experienced and skilled manpower. (b) In late 1983, following repeated urging by the Bank for the adoption of a task force maragement concept, the new administration of FERTIMEX organized the management of the project in a "limited task force" headed by a project manager, reporting directly to the planning and development (P&D) general manager. The new setup included: (i) a construction group; (ii) programming, planning, cost control specialists; and (iii) accounting and administration person- nel. An engineering group was also created specifically for the project, but it was headed by the "engineering division" Deputy Manager and not reporting to the project manager. Also, procurements and disbursements rested respectively with the administration and finance divisions at head office. Following the Bank's recommenda- tion, FERTIMEX strengthened the ta. '- force management team by hiring a consulting firm to assist in scheduling, cost control and report- ing. The organization of the team remained the same. (c) By the end of 1984 engineering design and most of procurement were completed. FERTIMEX began to concentrate more on the Lazaro Cardenas projects (both LAC 50 and LAC 93) as it had completed all other projects; the Planning and Developing Department already reorganized in the present form, virtually became the task force for the project, working full time for its completion. 3.6 The impact of inadequacies in the project management in terms of implementation delays and costs was exacerbated by other major factors includ- ing the lack of counterpart funds (see paras. 3.24 below). For the current functioning of the plant, the management organization (Annex I) could be improved by the creation of a technical department which would act as support for planning, control, problems solving, development, etc. without being involved in the day to day production and maintenance problems. This depart- ment could include the following divisions: (a) laboratory for quality and pollution control; (b) design; (c) process; (d) production programming and reporting; and (e) materials and equipment control. There is no need for an increase in the total staft as these functions already exist, but are dispersed in the operation and maintenance departments. C. Employment and Training 3.7 The project's operating manpower requirements were estimated at appraisal at about 630 including professionals and skilled workers. Key staff was to be transferred from other FERTIMEX units and new staff trained with the assistance of the foreign engineering firms. 3.8 The total work force for all Lazaro Cardenas plant including the duplication project and the expansion of utilities and solid handling is presently alfost .,30 and i.l rench about 2000 whet all the recruitment is completed The la),ge variation between appraisal estimate and new estimate y FERTIXYX ,; due to (A) mnupowv requiremnt for the duplication project and (b) sime undo:estiuation at appi:otsia. But the actunti umiber may turn out be less than 2,000. The organization chart of the plant is in Annex I. 3.9 Apart from some training for key operation 3taff at supervision rel iv 'ther fretiliver siwilar plants, most of the basic training took plac >n 1he i am ;1 Iricoiuzeno ing , couamissioning and start up with the he-d of procoss .icensorn to'oJIs. In goiw.al t aining of 6 months to one year was provided for operators and superv,soY-s and of about three months for workers. Programs of training arc' in place and improving. 3 .10 Personnel turn over has been high for the lower level staff and hampers more maintennce than production operation. The high turpover is due to scarce housing and inedequate soc . services, such as transport and schools in Lazaro Cardenns. To -?duce the housing problem FERTIMEX has rented 100 houses for the higher )ove ztvff mnd a number of apartments for supervi- sors; but there are no othec plans to :Imp)ove social services. D. Use of Engineer.iig Firms and Consultants 3.11 For the five process p3nuts, FERTI1ME.X obtained licenses, basic engineering, foreign procurement, supervision of detailed engineering, and construction services from reputed foreign firms. The performance of these companies uas generally satisfectory. 3.12 Basic engineering for the solid handli.ng area was obtained from a foreign consortium whose proposal was suitible only for the smaller scope of LACSO. When FERTIMEX changed the scope of the system a new contract was awarded to another foreign firm which provided a proposal that can accommodate the enlarged scope, i.e., both LAC50 and LAC93 requirements. For the utilities services a local engine.:ring firm provided the basic engineering. In September 1982, while the interconnections were not completed, this fiim asked for contract cancellation as payments were delayed and the remaining work was not significant; F2RTIX awarded the interconnections work to a second local firmi. 3. .3 Detailed engineering was mainly provided by local firms. While three firms provided the services for the five process plants. FERTIMEX has used more than 14 firms for the solid handling and utilities areas and this has increased coordination problems which contributed to implementation delays. 3.14 Fhkihi&A assumed overall responsibility for general design and super- vision of the project. However, because of the inadequacies in project management, the pyoject was not provided with effective time contrcl, cost control, and rcporting system during very long periods. Apart from licensors' teams and assistance in scheduling from local firms, FERTIMEX did not make use of any externl techricel lstance during implementation and commissioning. The project did not have J, -oipounut probably because the Bank found at apprAiSa1) thAti wRT ' m erYno nrr-ngemcnt for the project was satisfac- tory. 31 3.15 The detailed engineering design faced serious problems including: (a) overbooking of work in engineering firms during the construction boom (CANGREIERA and the Petrolchemical Complex of TABASCO); (b) delays in the delivery of equipment design; (c) lack of qualified personnel; and (d) loss of an important part of the technical information during the 1985 earthquake. E. Procurement and Performance of Suppliers 3.16 Procurement of equipment and services financed under the Bank loan were done through ICB procedures in accordance with Bank guidelines except for items that had been agreed as proprietary or critical and small items. 3.17 Procurement of all local equipment and materials and award of contracts were through FERTIMEX procedures which are in agreement with the Mexican laws ("Ley de adquisiciones" and "Toy do contrates de serviclos"). These procedures seem to provide foi uffective compctition. 3.18 There were, however, several problems: (a) delay in the delivery of equipment design information. The delay in delivery of equipment. to the site during the first phase of project implementation was caused by the overbooking of suppliers as a result of the "Boom" in construction in Mexico. Later on, delays in delivery were caused by delays in payment to suppliers due to shor- tage of funds; (b) delays in placing purchase orders. In some instances delays were due to reordering. In other cases, FERTIMEX had not taken deliveries due to lack of counterpart funds or to large price escalations beyond what was agreed; (c) late reordering of equipment to replace old ones damaged by long storage and by the 1985 eerthquake, resulting in extra delays for the start up of the solid handling facilities. This has also caused delays for the duplication project (LAC 93) from which equipment was taken to start the initial project (LAC 50), i.e. the project under review. (d) insufficient inspection of materials and equipment prior to shirment. Initially procurement services were subcontracted .o the engineering firms, since 1984 they were handled by FERTIMX. 3.19 For about three years thie cquipm..' purchasing division was not effi- cient. Reporting to the administrative department, the division was organized functionally by categories of itoL, chis system created coordination problems as the same agent purchased a category of items for any project and any plant in operation. Since 1985 the division has been reorganized into a better structure: (a) specific subdivisions for plants in operation and projects in construction; and (b) in the latter subdivisio., groups are organized by project in "mini procurement task forces". In Yebruary 1987 FYRTIMEX intro- duced a fully computerized system for its equipment procurement Frtivity and control. 32 3.20 In the appraisal report it was anticipated that US$56.4 million, i.e., 54Z of the equipment would be imported and US$9.4 million, i.e. 70% of the engineering costs would be paid in foreign exchange. The actual cost of the imported equipment (US$73.7 million) and of the engineering (US$8.9 mil- lion) amount to 77 and 66 percent of the respective totals. The percentage of engineering cost in foreign exchange is lower than estimate! for equipment FERTIMEX had to import such materials and equipment as structural steel, instruments etc. which can be produced in Mexico but were not available due to overbooking of Mexican suppliers during the first phase of implementation. F. Performance of Erection and Civil Work Contractors 3.21 The Construction of the entire Lazaro Cardenas project (LAC 50 and LAC 93) was contracted to about 25 local major contractors. By 1985 another 30 smaller contracts were awarded tor finishing discrete areas of construction which have been left aside or where progress was behind schedule. The large number of contractors involved created considerable coordination problems. 3.22 The Construction supervision rested with FERTIMEX with the assistance of specialists from the process engineering firms and equipment vendors. Several problems were experienced during the construction; the main ones were: (a) higher seismic standards for piling and civil work were required after the 1979 earthquake; (b) lack of local specialized manpower due to insufficient housing infrastructure and poor living conditions in Lazaro Cardenas; (c) difficulties in procuring structural steel from local suppliers; (d) FERTIMEX shortage of funds and budgetary constraints resulting in delays in payments and in the slowdown of the construction work for long periods; (e) repairs required due to the 1985 earthquake; and (f) some misalignments in equipment installation and inefficient sequencing of activities have also been reported. G. Implementation Schedule 3.23 Estimated and actual implementation schedule is in Annex II and summarized below. 33 PROJECT IMPLEMENTATION SCHEDULE Appraisal Terminai Estimate Actual Dela-s Begin End Begin End (month, Evaluation of Bids and award of foreign contracts 11/77 10/78 11/77 10/78 Basic engineering process plants 10/78 12/78 10/78 05/79 Basic engineering off sites 12/78 01/79 12/78 04/80 Bids evaluation and award contracts to Mex. Eng. Firms 06/78 12/78 10/78 07/79 Detailed engineering process area 01/79 12/79 02/79 10/81 Detailed engineering off sites 03/79 12/79 09/79 09/83 Foreign procurement 10/78 12/80 04/79 12/83 Local procurement 02/79 03/81 11/79 11/86 Site preparation and civil work 01/79 04/80 02/79 12/84 Erection and construction 12/79 06/81 02/81 12/86 Commissioning of process plants 03/81 09/81 01/86 06/87 Start up of commercial production 09/81 06/87 3.24 The overall delay in completion of the project was 69 months bcn'' the appraisal schedule. While the 36 months considered at appraisal was rather optimistic for the implementation in Mexico of such a large proje. this very long delay was due to a combination of many different reasons dxt_': the project implementation period. An attempt is made to analyze them fox each main phase: Redesign, 1979 to 1981 the project already took almost two years dela-. The main factors were: (a) redesign of piling and foundations follou-_):P the 1979 earthquake; (b) FERTIMEX decision to implement simultaneousl. duplication (LAC 93) of the phosphatic fertilizer and upstream plants the resulting changes in the scope of utilities and solid handling wh. have been completely redesigned and expanded substantially; (c) the 0' booking of the local engineering and equipment supplies due to the in trial construction boom; and (iv) deficient project and construction management arrangement. In effect, FERTIMEX had embarked on a large project without having the organization and manpower needed for impli- 34 tation; at the same time it had to implement many other important projects such as the urea plant of Pajaritos, the sulfuric acid plants of Queretaro and Guadalajara, etc. Shortage of funds, 1983. FERTIMEX faced acute shortage of funds as a result of government's limited budgetary resources during one of Mexico's most difficult economic situations. The shortage affected both FERTIMEX's capital budget and its liquidity; the devaluation in February 1982 resulted in an increase in local funds requirements for capital expendi- ture. Any available scarce funds for capital expenditures were used to complete other FERTIMEX projects considered urgent that needed less funds than LAC 50 for completion (e.g. the urea plant of Pajaritos and sulfuric acid plants of Queretaro and Guadalajara.) As a result, project LAC 50, although not mothballed, progressed at a very slow pacet FERTIMEX was often unable to pay contractors and suppliers in time, procurement was stopped, in some cases orders were cancelled, resulting in delays in delivery of design information and the construction activity was practi- cally halted. FERTIMEX had to stop the duplication project (LAC 93) and considered the possibility of phasing completion of the various units of the project giving priority to the phosphatic fertilizer and upstream plants. The slower plan proposed was not implementable as the funds allo- cated were not sufficient and served only to protect erected equipment. Reorganization of Project Management 1984-1985. By this time, FERTIMEX had completed other projects and gave priority to the Lazaro Cardenas plant (LAC 50 and LAC 93). The project management was reorganized and limited "annual* funds are available to the project allowing procurement orders to resume and field construction work to pick up. However, further delays were caused by (a) delays in delivery of equipment design informa- tion; (b) delays in delivery of equipment; (c) difficult coordination and increased interfaces problems with the large number of contractors and engineering firms for the utilities and the solid handling systems; and (d) late placement of orders, coordination problems and deficiencies of the procurement division. The September 1985 earthquake resulted ins (i) important damages to the plant while the project was about to be completed and some of the utilities systems (water, steam, power) were under start- up; and (ii) the loss of an important part of the technical information in Mexico City (mainly drawings and purchase orders). The earthquake also caused damage to the pollution control system. Final Phase Completion: 1986-1987. FERTIMEX concentrated on repairing and completing the plant. There were again delays due to repairing or reordering of equipment and materials damaged during the earthquakes and because of the long storage. 3.25 Part of the reason for delays over the years was beyond the control of FERTIMEX (Mexico's economic situation, 1979 and 1985 earthquakes) but FERTIMEX could have reduced the delays by (a) setting up, before the beginning of the project implementation, an effective project management organization; (b) early appointment of a project manager, adequate organization of manpower; (c) early use of outside help to improve its planning, programming, reporting, time control and cost control and recording capabilities; (d) reducing the impact of the duplication LAC 93 project by phasing or changing early the 35 - scope of the solid handling system of which at least most of the important bagging system is not needed; and (e) not giving high priority, (without any sensible criteria) to projects such as the sulfuric acid plants of Guadalajara and Queretaro (identical to the sulfuric acid plant of the project) to produce a low concentration fertilizer as ammonium sulfate from sulphur. 3.26 FERTIMEX's decision to simultaneously build a phosphatic fertilizer DAP/NPK fertilizer and upstream plants (LAC 93) and consequently to expand the scope of the utilities and to increase more than needed the solid handling storage and bagging system, had the following main impacts on the initial project implementation: (a) dilution of the scarce and already insufficient manpower; (b) dilution of available funds. As indicated (see para. 3.33 below) by 1982 FERTIMEX had already used up for Lazaro Cardenas site more funds than anticipated in the total financing, estimated at appraisal for the original project; (c) increase of total financing required to start the initial project due to the important changes in the scope of the off sites; and (d) increased cost and delays due to the need to redefine and redesign the offsites. 3.27 Main factors responsible for the implementation delay are rated as follows: Rated Rated by by FERTIMEX Bank (in months) 1. Optimistic implementation time at appraisal - 10 2. 1979 earthquake, civil work redesign and increase in piling 8 8 3. Decision to implement LAC 93, scope changes and redesign of utilities and solid handling 6 6 4. Project organization and management - 5 5. Overloading of Mexican engineering and supplies firms in 1979-1981 13 4 6. Economic problems, lack of funds 20 18 7. Reordering of equipment due to high price escalation and consequent delay in delivery/ procurement problems 12 5 8. 1985 earthquakes 10 10 9. Start-up problems due to damaged equipment during long storage 3 Total 69 69 36 H. Capital Cost, Financing and Loan Disbursement 3.28 Table III-1 summarizes capital cost estimates at the time of appraisal and actual costs. Annex III gives the actual capital cost expendi- tures schedule based on information provided by FERTIMEX for main areas of the project. Part of the shared utilities and solid handling facilities is allo. cated to the duplication project LAC 93 on the basis of production capacities and scale factors 0.6 for utilities, 0.8 for solid handling. 3.29 The cost overruns excluding interest during construction is US$159.4 million, equivalent to 632 of appraisal estimates. Cost overruns were parti- cularly significant for construction and equipment costs. The overall cost overrun is about 68% of total financing estimated at appraisal (Actual US$507.8 million against US$301.3 million at appraisal, see Table III.1). COST VARIATION BY CAUSE Z of US$ million total overuns Construction cost 81.9 51.4 Equipment and supplies 69.3 43.5 Project implementation 8.2 5.1 159.4 100.0 3.30 The main factors responsible for cost overruns were: (a) increase in civil %-ork scope (additional piling and foundation work) following the 1979 earthquake; (b) project implementation delay of 69 months resulting in high equipment and construction contracts' prices escalation and interest accumulated during construction; (c) important changes in project scope; (d) currency realignment which in many cases, combined with plice escalation, obliged FERTIMEX to renegotiate contracts or rebid for equipment; (e) repairs after 1985 earthquakes; and (f) equipment failures at start-up due to long storage period. 3.31 FERTIMEX did not have an efficicnt capital cost monitoring system specific to the project and it is impossible to obtain more details on cost allocation; for example no separate cost accounting for utilities and solid handling facilities items was established. Several items had to be estimated (prorated for LAC 50 and LAC 93) by the PCR mission. 3.32 Also, FERTIMEX does not have separate accounting for funds by origin earmarked for the project, however, according to FERTIMEX, in agreement with government policy, the total project cost of US$411 million was met by 50Z share capital and 50% Bank and other long-term loans; projected financing plan at appraisal time and estimated actual for the project and the total plant including duplication is given in Annex IV. TABLE III-1 MEXICO SECOND FERTILIZER PROJECT PROJECT COMPLETION REPORT Capital Cost and Financing Required (US$ million) Appraisal Estimate La Actual Overruns Local ForeignLb Total Local Foreign Total US$ (Z change) Direct Indirect Total Site preparation, civil work erection, infrastructure 91.2 10.2 191.4 165.0 18.3 18.3 183.3 81.9 80.8 Equipment Process plants 18.4 38.6 57.0 37.7 44.7 9.4 54.1 91.8 34.8 61.0 Utilities & solid handling 23.1 31.9 55.0 48.4 29.0 12.1 41.1 89.5 34.5 62.7 Subtotal equipment 41.5 70.5 112.0 86.1 73.7 21.5 95.2 181.3 69.3 62.0 Licenses and services 4.7 11.1 15.8 4.6 8.9 - 8.9 13.5 (2.3) (14.6) Administration & supervision 9.0 3.4 12.4 28.1 - - 28.1 15.7 126.6 Preoperating Costs 7.9 2.0 9.9 4.1 0.6 - 0.6 4.7 (5.2) (52.5) Project Cost 154.3 97.2 251.5 287.9 83.2 39.8 123.0 410.9 159.4 63.4 Working Capital 19.0 6.9 25.9 23.8 /c - - - 23.8 (2.1) (8.1) Interest during construction 3.5 20.4 23.9 6.2 66.9 66.9 73.1 49.2 205.8 Total financing Required 176.8 124.5 301.3 317.9 150.1 39.8 189.9 507.8 206.5 68.5 /a After allocation of physical and price contingencies. /b Include 20.3 million of indirect foreign exchange estimated at 1OZ of civil work and erection and 202 of equipment. Ic At maximum output in 1992. Source: FERTIMEX for actual. 38 3.33 The cumulative disbursement pattern for the Bank loan as estimated at appraisal and the actual pattern is summarized below: the actual disbursement as the actual capital expenditures schedule (Annex IV) differs widely from that projected at appraisal as a result of the important completion dela No attempt was made for amending the disbursement schedule. However, c: tal expenditure schedule for the total project including the duplication pre *ct given in Annex V and VI shows that FERTIMEX had already spent in 1982 - amount required for the original proje-.-, including a large part of th cost overrun. This suggests that FERTIMEX cc.. ,- -ave completed and startec up the initial project (LAC 50) by this time, if it had been decided to implement the duplication project after the completion of the initial project, not at the same time. CUMULATIVE LOAN DISBURSEMENT (US$ million) 1979 1980 1981 1982 1983 1984 (i) Planned 24.0 68.0 80.0 80.0 80.0 80.0 (ii) Amended -*- - - - - (iii) Actual 4.2 29.6 40.7 60.9 77.2 80.0 (iv) (iii) as I of (i) 17.0 44.0 51.0 76.0 96.0 100.0 * no amendments to schedule throughout the year. IV. OPERATION PERFORMANCE A. Commissioning and Start-Up 4.1 As mentioned earlier project facilitie were mechanically completed and the last procass plant was started in June 1987. The commissioning of the plants was carried out generally with greater problems than normal due to mechanical troubles with damaged equipment during long storage. The mechani- cal completion and commissioning dates of the various facilities are listed in Annex VII. Individual pieces of equipment were tested during start up, while the process plants were not formally tested through acceptance tests. How- ever, since the sulfuric acid, phosphoric acid and DAP/NPK plants have operated more than 90 consecutive days at 801 of their rated capacity, they can be considered as completed according to schedule 2 of the loan agreement. The nitric acid and ammonium nitrate plants have started only in June 1987 and have not yet reached the required levels; in October 1987 they produced respectively 702 and 60% of their rated capacity. 4.2 The Sulfuric Acid Plant started operation in May 86 and after five months reached a monthly production equiva4ent to 93? of the rated capacity. For 1987 production was expected (the PCR mission was in the field in November 1987) to reach more than 802 of the rated capacity. At present the factors limiting the production are (a) the capacity of the main air blower; and 39 (b) difficulties in the sulfur melting, due to the imported sulfur quality and to the melting pit design. FERTIMEX is building a second melting pit, similar to the one they already added in Queretaro and, if required,3/ the main blower problem can be overcome by using modern low pressure drop catalyst. 4.3 The Phosphoric Acid Plant, started operation on July 15, 1986 and after three months reached a monthly production of 88Z of the rated capacity. In 1987 the plant was expected to produce more than 148,500 tons equivalent to 75% of the nominal capacity with an overall P205 recovery exceeding 90?. At present, the main problems are: (a) the use of local ROFOMEX phosphate rock from Baja California (some 500,000 tpy) instead of higher quality Florida rock increases production problems, which results in a somewhat reduced plant capa- city, and a lower filter acid concentration which in turn overloads the evapo- ration section; (b) operation of the filter slurry feed pumps; and (c) impor- tant vibrations of the effluent pumps which forced FERTIMEX to discharge to the Balsa river, through the rain water canal of the effluents from the evapo- ration condensers and from the final scrubber of the DAP plant. While the plant is designed to dump these effluents in the sea together with the gypsum through undersea lines equipped with diffusors this is one problem related to environmental aspects (see below). 4.4 FERTIMEX is currently testing the Baja California rock in a pilot plant and in the industrial plant to improve its use. The main tests involve (a) use of floculants; (b) mixtures with Florida rock; and (c) use of unground rock. Operations with Baja California rock have already been improved. However, if ROFOMEX production can be increased, FERTIMEX will probably have to increase its evaporation capacity. 4.5 -FERTIMEX is resolving the effluent pump problems i.. two phases: (a) a tempoary solution to be completed within one month which requires the modification of the pumping pits (adding baffles to avoid water cycloning and vortex formation) and strengthening of the pump's structure; and (b) a final solution including the redesign of the pumping pit and the use of new pumps. (See Environmental Aspects, Chapter IV). 4.6 The NPK/DAP Plant started operation in September 1986 and after three months reached a monthly production equivalent to 72% of nominal capacity. The main difficulties experienced were: (a) mechanical problems due to long storage (mainly with the dryer bearings); and (b) ammonia scrubbers did not function well. These problems are now solved and in 1987 the plant is expected to produce about 310,000 tons per year equivalent to 65? of its nominal capacity in nutrients. When only DAP is produced the production of this unit is limited by the phosphoric acid Lpstream plant production (198,000 tpy in tons of P205). The annual requirement of the unit at full rated capa- city is more than 250,000 tons. This caiL be a bottleneck if the two fertili- zer units (LAC 50 and LAC 93) simultaneously produce DAP during long periods. The PCR mission was in the field in November 1987 and learned that the produc- tion of NPK Fertilizer (17-17-17) will start by January 1988 and no major problems were expected. 1/ The sulfuric acid plant has a stream rate capacity higher than needed by the phosphoric downstream acid plant. 40 4.7 The Nitric Acid plant started operation in April 1987 and within three months, reached a monthly production of about 68% of nominal capacity. The start up was reportedly carried out smoothly. However the production has been limited by storage capacity and insufficient consumption of the ammonium nitrate plant. 4.8 Ammonium Nitrate Plant. The plant started in May 1987 and reached about 60% of monthly rated production in October. The main difficulties experienced were: (a) mechanical problems with some main equipment (e.g. bucket elevator); and (b) excess of dust on the prilling tower exhaust air which hampers operation and increases maintenance. Many of the mechanical problems have been solved and production is expected to reach 60,000 tpy in 1987 equivalent to 63Z of the prorated nominal capacity. 4.9 Solid Handling started with important instrument problems. The interlock system has been simplified to start operation and the whole system is expected to be in operation in early 1988. Because of the very large capa- city of the system, this delay had little effect on operation. Phosphate, sulfur and ammonia have been all received smoothly in the system and fertili- zer is shipped in bulk and bagged without major problems. 4.10 The duplication project (LAC 93) units have started in June and July 1987. Some equipment of these units has be taken as spare parts to replace damaged equipment (during low storage and ea.,Lquake) for the initial project (LAC 50) units. The commissioning and start up of the duplication project benefitted from the experience gained in the start up of the initial project and were carried out smoothly during a relatively short period (only about 7 weeks) for the three plants. B. Build up of Production 4.11 After a difficult and long implementation period the performance of the overall complex is satisfying and the plants will probably operate (some already have been) at a higher capacity utilization than the 70Z expected at appraisal for the first year of production. Actual production of the various products achieved in 1986 and 1987 for the Project (LAC 50) and the duplica- tion project (LAC 93) are given in Annex VIII. 4.12 Annex IX gives the expected build up of production of the different units in percent of rated capacity. Ammonium nitrate production is expected to reach its rated capacity in 1991, a large part being produced as solution for NPK production, while DAP/NPK will reach their maximum capacity only in 1993. C. FERTIMEX Market 4.13 The SAR (paras. 4.11-4.14) used FERTIMEX's projections (1978-85) for its analysis as they seemed realistic at that time (1979); the projections were based on 8Z and 13? annual growth rates for N and P205 respectively, which were much lower than the historical growth rates of 14? and 15? recorded for the previous 20 years. As shown in the table below, FERTIMEX's demand projections for P205 at appraisal seemed in line with actual lupward] trends (although not with .ctual numbers) until 1982; in 1983 actual consumption of 41 P205 dropped and it began to regain the 1982 level only in 1987. (The Appraisal Report presents projections for five years, up to 1985). The supply estimates made at appraisal were much higher than actual reports and domestic production combined; the delay in implementation of LAC 50 project is certainly a factor for the estimate variation. (Data were not available for comparing appraisal estimates and actual figures <f demand/supply of nitrogen). Table IV.l: DEMAND FOR AND SUPPLY OF P205 APPRAISAL ESTIMATES AND HISTORICAL FIGURES ('000 tons) 1977 1978 1980 1982 1995 Appraisal Estimate Demand 217 286 341 403 521 Supply 282 265 310 390 415 -35 21 31 13 106 Historical Supply/Demand Balance Demand 217 259 289 445 437 Production 282 227 201 250 273 Balance 65 -30 -79 -195 -164 Source: Historical Supply/Demand Supply: FERTIMEX Table IV.2 presents comparative figures of demand/supply projections for each nutrient made at appraisal for 1977-85 and new projections for the next five years 1987-95, beginning with the first year of production (1987). If demand continues to grow at historical rates and with the production build up of all the Lazaro Cardenas plant including the duplication project, Mexico will be about self-sufficient in phosphatic fertilizer by 1990. After that deficit will increase again and by 1995 import requirements will reach a level equiva- lent to the initial project production. D.Fertilizer Prices 4.14 Current Pricing Policy4/ has remained essentially the same for the 10 years since the project was appraised; both producer and farmer prices are controlled by the government and they remain below international prices. Prices received by FERTIMEX as projected at appraisal were made on the 1/ Policy prior to any actions to be taken to increase prices in accordance with the FSAL. Table IV.2 42 MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPI .JN REPORT Comparative Suply and Demand Projections (000 tons) Actual as Actual projected at appraisal /a Actual Current projections /b 1977 1978 1980 1982 1985 1986 1987 1990 1992 1995 Demand Nitrogen Anhydrous ammonia 186 216 237 257 285 264 280 335 340 343 Nitrogen fertilizer 609 776 900 1,059 1,395 1,023 1,168 1,326 1,525 1,827 Total Nitrogen 795 992 13 1 448 1,661 1 865 2 7!0 P205 217 286 341 403 521 406 446 545 608 730 K20 /c 58 66 80 100 135 94 113 138 153 178 Total 100 1,344 12558 1,819 2,336 _ 2 2.345 2,626 3.078 Production with Projects Anhydrous ammonia /d 186 216 237 257 285 264 280 335 340 343 Nitrogen fertilizer 423 421 750 870 1,153 1,000 1,027 1,011 1,016 1,043 Total Nitrogen 609 637 987 11127 Ji43 1-264 130711 36 I 356 1 386 P205 Le 282 265 310 390 415 257 394 535 535 535 Total /c 891 902 1,297 1,517 1,850 1.521 1,701 1 881 1,891 1921 Demand/Supply Balance Nitrogen fertilizer -186 -355 -150 -189 -245 -23 -141 -316 -509 -784 P205 65 -21 -31 -13 -106 -149 -52 -10 -73 -195 Subtotal -121 -376 -181 -202 -351 -172 -193 -326 -582 -979 K20 -58 -66 -80 -100 -135 -94 -113 -138 -153 -17e Total -179 -462 -261 -302 -486 -266 -306 -464 -735 -1,15* Projected Output of Totai Lazaro Cardenas Plant P2C5 222 356 376 38 Nitrogen 164 206 218 22( /a Includes initial Project output. 7t' Assume uneconomic plant closure, total Lazaro Cardenas plant output, and no new capacity addition except for the completion of the Pajaritos Ammonium Nitrate Plant and that existing plants operate at 90Z capacity. /a All potassium chloride including NPK fertilizer content is considered imported. 7-d Ammonia for export have been deducted from SAR projections. 7e Does not include 150,000 tpy technical phosphoric acid produced at Pajaritos which was and remains a noafertilizer product, mainly for export. Source: FERTIM for data on demand and supply of P205 and Bank estimates for current projec- tions. . 43 assumption that an increase of 25Z in real terms over five years starting 1979 (SAR paras. 7.04 and 7.07) would be allowed by the Government. No specific action program was committed )y the government to implement this crucial plan apart from a government's recognition of "the need to develop a pricing policy which would allow FERTIMEX to become financially self-sufficient" (SAR para. 4.30) and an agreement by the government to take action required to provide FERTIMEX with revenues which would enable it to earn before interest and after tax a return on assets specified for each year (SAR, para. 9.01(c) and Loan Agreement). Prices received by FERTIMEX were generally much lower than appraisal estimates as shown in Table IV.3. To have a realistic estima- tion of profit, new price projections being used for financial reestimation in this report are also lower than appraisal estimates for the most part. Table IV.3: EX-WAREHOUSE /a AND IMPORT PRICES AS X OF APPRAISAL ESTIMATES b FOR 1985 AND BEYOND (All prices in US$ of 1987) Actual 1987 Projected 1980 Projected 1995 Ex-Ware- Ex-Ware Er-Ware house Import house Import house Import DAP 50 63 65 62 76 72 NPK 59 56 90 64 103 70 Ammonium nitrate 31 57 51 81 101 79 a See Annex X. ]b Appraisal estimates are in Annex 7-1. Appraisal Ex-Warehouse prices include 25% increase. Appraisal import prices are CIF prices. 4.15 Ex-Warehouse Prices or "Itnancial Prices".5/ As shown on the previous table, ex-warehouse prices of finished products have decreased signi- ficantly in real terms between appraisal (1978) and 1987. In constant 1987 terms, ex-warehouse prices in 1987 are equivalent to 48% to 73Z of those in 1978, and only 30% to 60% of prices projected at appraisal for the years beyond 1985. 4.16 World prices of fertilizers have fluctuated around a downward trend since 1974 when they reached an unusually high peak (e.g. US$500 level in bulk for both TSP and DAP, for FOB Florida). Prices received by FERTIMEX (ex-ware- house prices) decreased, however, faster than import prices. As a result prices projected at appraisal that were assumed to approach import prices in 1985, have in fact moved further away from import prices; ex-warehouse prices were supposed ro be in the range of 78% - 962 by 1985 to reach parity beyond 1985 in fact were down to a range of 42? to 76% in 1987. (Table IV.4 below), I/ In the SAR, this term was used to indicate essentially ex-warehouse prices modified for IRR calculation. /44 Table IV.4: EX-WAREHOUSE PRICES AS Z OF IMPORT PRICES Appraisal projections for 1985 and beyond Actual 1987 1990 1995 - ---------------- ---------- -------------------------- AP 96 76 100 100 'J"K 68 70 95 100 iF,nonim Nitrate 78 42 49 100 muort Prices (or "Economic Prices").6/ S17 Import prices in 1987 were only 56% to 63% of those projected at ppraisal for 1985 and beyond, reflecting the depressed situation worldwide ir le fertilizer industry since the early 1980s. Based on current (September 987) Bank projections (Commodities Division), international prices are ,.xpected to recover slowly, still only reaching 70% to 80% of appraisal esti- oates by 1995. The revised economic analysis of the project under review is Ased on these assumptions. Distribution Study .18 FERTIMEX had agreed to complete (SAR, para. 9.01(c)), by August 31, 1980, a study of its current distribution system, including transportation Leeds, as well as the required steps to be taken to improve the system. The 4tudy was carried out by FERTIMEX and completed by February 1982. It was sent o the Bank for review. The study concluded that there was a need to develop a distribution infrastructure based on about 80 primary warehouses built and operated by FERTIMEX at strategic Jocations, serving secondary warehouses operated by the various distributors' agents. Increased use of coastal shipment and unit trains for large volumes over large distances was also recommended. 4.20 The study served as the basis for FERTIMEX's decision to establish a network of primary warehouses, which in 1987 numbered about 45. A second phase of this program, together with the construction of maritime terminals, would be financed as part of the proposed FSAL. 4.21 The study did not contemplate the possible economic advantages of bulk transport to, and storage in, primary watehouses. To this date, the entire primary network was based on transport and storage of bagged products, which explains the large bagging capacities installed in Lazaro Cardenas and Pajaritos. Recently, FERTIMEX has started transporting bulk materials by ships ".id rail, and future warehouses (expected to be financed by the proposed FSAL) will be for bulk handling and storage. It is also recommended that 6/ In SAR, the term 'Economic Price" was used for ERR estimation. 45 . FERTIMEX should study the possibility of equipping the existing primary ware- houses to handle bulk material. F. Environmental Aspects 4.22 Except for the disposal system to discharge Gypsum, Phosphuric Acid evaporation cooling water and final scrubbing fertilizer water (see below), the Lazaro Cardenas plant facilities have been constructed in conformity with the environmental regulations norms agreed to during appraisal. The double absorption sulfuric acid plant and the nitric acid plant are both modern process plants which in normal operation have respectively low emission of sulfur and nitrogen oxides and met high standards of pollution control. 4.23 The Phosphoric Acid and the fertilizer plant are equipped with suffi- c'ent GA7 scrubbers aid should meet also the levels of fluorine, ammonia and dust given as in the guaranty by the process vendor (USA). 4.24 However, the Ammonium nitrate plant which started in May 1987 has presently a high dust emission in the prilling tower exhaust air. This inireases pollution near and inside the unit, increases maintenance and makes operation more difficult. FERTINEX will have to solve this problem jointly with the process vendor. 4.25 While it was stated (para. 61) in the President's Report (P.2497-ME) that the plant site would include a 60 ha site for the disposal of gypsum produced by the phosphoric acid plants, according to the SAR (para. 5.10) the gypsum produced by the phosphoric acid plant was expected to be pumped at some 8 kms from the plant site for disposal. On the other hand, for economic and seismic reasons FERTIMEX had analyzed early on, other solutions jointly with the Universidad National Autonoma de Mexico (UNAM). The studies (1979-1980) recommended to discharge the gypsum after dilution to the sea, taking into account the submarine topography (high slope and depth) near Lazaro Cardenas and the important currents toward south. An oceano/hydrographic study was also prepared by a foreign consultant. 4.26 In NovLmber 1983 the Bank discovered during a supervison mission of the Port Project III, that FERTIMEX had decided to dispose of the gypsum to the ocean through two plastic sea lines. This proposal generated important discussions in the Bank which urged FERTIMEX to seek additional assistance from consultants, specialized in design and operation of ocean outfalls, particularly for the fertilizer industry, to examine the proposed design, study the outfall area and assess environmental impact of the proposed scheme. FERTIMEX did not accept outside consultants. It requested instead assistance from various departments of the Secretariat de la Marina to review the design of the sealines, and to assist in the supervision of its construction. After the 1985 earthquake the Bank did not follow up on this matter. There was no more supervision after June 1985. 4.27 In the actual design, the gypsum was to be discharged in the sea, at 1.5 km from the coast after dilution with water effluents from the evaporation condensers and the NPK/DAP fertilizer plant final scrubber. This was to be done through two plastic undersea lines each with three diffusers discharging at the recommended depth of 30 meters. The basic and part of the detailed engineering were obtained from an international firms (USA). FERTIMEX 46 requested approval of the proposed system from the relevant government entities including Secretaria de Desarollo Urbano y Ecologia (SEDUE), Secretaria de la Marina, Instituto National de Pesca, Fondo National de Desarrollo Portuario etc. To take into consideration some of the Bank's concerns, each sea line had been designed to handle the total gypsum (and water effluent) produced at site by the two phosphoric acid units at the UNAM recommended dilution. A connection system equipped with a valve allows the switch frow one sea line to the other. The plastic pipes are protected over 500 meters through the surf zone by a steel -ell and are covered by some two meters of rocks. 4.28 At start up the main problems were (a) the project sealine have been broken probably by an hydraulic ram; (b) the duplication project sealine plugged; and (c) strong vibration of the effluent pumps. 4.29 During a fe-d months Fertimex discharged the gypsum at about 100 meters from the coast and it seems that no accumulation occurred which confirms the appropr-ateness of the water current direction. 4.30 Presently cne sealine has been repai!ed and is being used to dis- charge the gypsum p:oduced by both trains and because of the pumping problems, the water effluents are eJected to the Balsa river through the (open) rain canal. 4.31 FERTIMEX is resolving the pumping problem in two phases: (a) a tempo- rary solution to be completed within one month consisting in the modification of the pumping pits, and the strengthening of the pump's structure; and (b) a final solution including the complete redesign of the pumping pit and new pump. 4.32 FERTIMEX is also monitoring the different water and solid effluents of the plant, in particular the following samples are reportedly taken monthly and analyzed: (a) water effluent to the Balsa river; (b) gypsum effluent to the sea; (c) sea water at different depths et gypsum discharge point; and (d) sea water at 15 km from discharge point, etc. 4.33 The results of this monitoring are reportedly sent each month to SEDUE for approval but the PCR mission was not able to obtain any of the results for review. (The mission requested but has not received a copy of the analysis). 4.34 It remains essential that (a) FERTIMEX complete urgently the imple- mentation of its program to solve definitively the effluent's pumping problem and avoid the disposal of acidic water effluent to the Balsa river; (b) unplug the second sea line to reduce the risk of having again problems with the two lines simultaneously; (c) the phosphoric acid plant operation should be tight- ly controlled to optimize raw materials recovery, reduce rejected pollutants and consequently improve production costs; (d) a domestic market study for flousilisic acid may be done to assess the feasibility of adding a fluosilisic acid recovery in the concentration section, which will reduce the fluorine level in the condenser effluents; and (d) FERTIMEX maintains the effluents monitoring and frequently assess their effect on the environment. In a letter dated March 31, 3988 sent to the Bank prior to the approval of the FSAL, FERTIMEX is committed to take steps to resoJve the pollution control problem including the installtion of a new pump by August 1988 and a thorough reeval- uation of the sealines to determine the appropriate measures to be taken to ensure their proper functioning. V. FIMANCIAL PERFORMANCE A. Financial Situation of FERTI-MX and Covenants U 5.1 In general, FERTIMVX'c financia) situation has not improved since 1978, and had, in the main, mintained some of the saLie features of the period prior to appraisal (SAR paKas. 3. -2-3.16) among other things, FERT:MEX continues to have a tight liquidity polition, and difficulty to service debt from internal cash generation. Some important financial ratios are inade- quate. The Company still has ineff'.c'iive control of working capital. It continued to operate at a loss for uost of the years from 1978 to 1986, even with large transfer payments fcom vh. Governmeit. in 1987, it had a negative working capital; the Company depended on Government budget transfers for 50Z of its cash needs and it was unable to service large debts. Other recurrent events of which FERTIMEX has no contrnl %ere sevcrri vltins of the Deso such as the one in 1982 which resulted in increasc!; in locra funds require- ments for FERTIMEX's projects including LAC 50. (i) The Debt/Equity ratio has been generally loer tha)-, 60/40 (1.50) as required and, since 1980 lower than unity. The equity" as presented in FERTIMEX's Financial Statement every yeax includes tvo big items, the "surplus due to re-expression" (superavit po, reexpresion) or reserve resulting from asset revaluation and "cont"hutions for future capital increases" (aportaciones para fut'%os avymentoo de capital). In this sense the "equity" base may have been overstated as the second item represents the sum of all goverinient transfers, i.e., including operation subsidies. The share capital of the Company remains at pesos 21.397 million. FERTIMEX: CURRENT RATIOS AND DEB2/EQUITY 1ATIOS (1977-87) 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 Current ratio 1.29 1.16 1.04 0.83 0.67 0.33 1,03 0,99 0,79 0.88 Long term Debt/Equity 1.89 1.28 1.05 0.77 0.65 0.88 0672 0.42 0.51 0.86 Source: Computed from Balance Sheets 7/ For Covenants See Loan Agreement, Guarantee AgreamE:- n- , Ai rs. 7.10 and 9.01. 48 (ii) The current ratio covenant of 1.3 or higher was not met as the actual ratio has been less than one since the project was appraised in 1979 except for 1983 when it reached 1.08 but was still below the required ratio. (iii) The government was to take all action required to provide FERTIMEX with revenues which will enable it to earn before interest and after tax a return on fixed assets of at least 3.5Z for 1980, 1981 and 1982 and 82 for 1983 and thereafter. Over the years since 1979 the return to fixed assets has been far below 3.5Z.8/ The government has continued to provide subsidies to FERTIMEX but has not taken other necessary actions such as an increase in prices (suggested but not covenanted) of 25Z in real terms over five years, etc. This contri- buted to FERTIMEX's persistently low and often negative return. (iv) FERTIMEX continues to have high inventories and accounts receivable which raise its Working Capital 9/ above 3.5 months of sales, the maximum required in the Guarantee Agreements. Under the consignment system with payments being in cash, accounts receivables were reduced but inventories remain high. During the PCR mission it was reported that in 1986 FERTIMEX succeeded in converting part of its short-term loans into long-term obligations 10/ for other years however the Company has excessively high short-term debts. (v) Although regular reporting is required, the Bank has received irregu- lar progress reports and financial statements during implementation of the project. (vi) The SAR mentions (para. 7.10) an important covenant which requires FERTIMEX to inform the Bank of all new projects requiring capital expenditures in excess of US$20 million. The Bank was not informed of the duplication project LAC 93; the project became known to thi Bank only during a supervision mission. From a legal point of view, it should be noted that the Loan Agreement, does not include this covenant. 8/ For many years the ratio was less than 1% or negligible when the actual ratio was based on net profit plus financial changes as implied in the term "return after tax" in the Agreement. 9, Gross Working Capital, i.e., Current Assets in this case although it was not defined as such in the Loan Agreement or Guarantee Agreements. 10/ Long-term borrowings include US$300 million from Canada and US$850 million from the USA. 49 - B. Reestimation of Internal Financial Rate of Return (IRR) and Economic Rat- of Return (ERR) 5.2 As the project first started up in 1987, financial data are not available for even a full year of productionll/ to permit a 'post evaluatiow of project performance especially since all the plants will build up produc. tion to full capacity only in 1993. However, following the Guidelines for Preparing PCRs 12/ an attempt is made to reestimate the IRR and ERR with two main new elements being introduced (1) the start up year is 1987 (instead of 1981 as in the SAR) and (2) price assumptions are based on price policy refo as proposed in the Fertilizer Sector Adjustment Loan (FSAL) approved March 1988 as a Obase case". Pricing has been a major issue in Mexico's fertilize. industry and the direction of price reform under the FSAL (supported by an earlier subsector study) seems a reasonable and realistic framework for the reestimation of financial and economic rates of return. A sensitivity anal ysis will show the effects of changes in prices and changes in capital costs 5,3 Under the base case ex-warehouse prices would increase by an averay of 17.12 in real terms per year over the period 1988-1992, so that, by the e of 1992, FERTIMEX would become financially self sufficient (with input price at or close to opportunity cost), e.g. cover all its costs, including depre- ciation and financial costs, and earn a return of 8% on equity invested in I projects. The resulting ex-warehouse prices or "financial prices" for the life of the project are given in Annex X, together with import prices or ,economic priceso. Compared to appraisal projections, ex-warehouse prices projected in the FSAL for 1990 still amount only to 50% to 90% of appraisal estimates .for the years beyond 1985. Only in 1995 would prices reach ap- praisal estimates for Ammonium Nitrate and NPK, but would still be less tha. 80% of projected DAP prices. In the FSAL it is anticipated that, by 1995, L prices will have reached import parity. The revised financial analysis of t project under review is based on this assumption (see Financial Analysis below). 5.4 Annex XI summarizes projected financial and economic production co and revenues for 1993 (the first year of full development in 1987 terms. By 1993, the project will have an estimated gross profit of US$37 million (US$56.7 million for the entire plant). This margin compares with an econoD gross margin of US$31.4 million (US$45.2 million for the entire plant). Tht difference is attributed to the import price differential 13/ and subsidized ,1/ Since the project just started up and the cost accounting system was only recently developed at FERTIMEX, only very few data on actual costs (operating costs) were given to the PCR mission. They are not sufficient for a meaningful analysis for comparing appraisal estimates and actuals. These data (particularly wages and utilities expenses) were nevertheless used to a limited extent for production costs estimates in this report. 12/ General Guidelines for Preparing PCRs September 1983. Para. 8 requires new estimates for IRR even when the PCR comes early in the operation stage. 13/ The ex-warehouse price of ammonia is based on total cost plus an 8Z return on ammonia plant investment while the import price is derived from the international export value. 50 fuel and power, while it is estimated that output prices will have practically reached import parity by that time. 5.5 Financial Rate of Return. Costs and Benefits streams for the finan- cial rate of return calculations are presented in Annex XII and XIII respec- tively for the project and the entire Lazaro Cardenas plant. 5.6 The revised financial rates of return estimates, compared with appraisal estimates, are as below: Project Entire Plant Appraisal estimate .3Z a Revised es-4mate -0.3% +0.3Z /a The duplication project was not included in the appraisal estimates. 5.7 The significant shortfall relative to the appraisal estimate is essentially due to (a) higher capital cost due to large overruns; and (b) much lower new price projections for FERTIMEX. At appraisal, it was assumed that the Government would implement a long-term pricing policy which would allow FERTIMEX to cover its costs and earn a return on its assets of about 8Z, implying a total 25% real price increase at 5Z p.a. during a 5-year period starting in 1979. This increase did not take place, in fact, real prices actually decreased during the period. 5.8 The revised estimate of IRR shows that financially the project is not viable and this raises a question on sustainability (see the Conclusion Chapter). 5.9 Annex XIV presents the estimated financial profitability and cash flow analysis of the entire Lazaro Cardenas complex, under assumptions that the pricing policy under discussions in the framework of the proposed Fertilizer Sector Adjustment Loan (FSAL) will be implemented. Net pre-tax profit would remain negative until 1993, reflecting the pace of progressive price adjustments and the initial burden of interest charges on long-term loans. The net cash flow also is projected to remain highly negative until 1990, unless a substantial portion of the debt is written-off, as proposed under the financial restructuring package presently under discussions under the FSAL. VI. ECONOMIC ANALYSIS A. Economic Rate of Return (ERR) and Sensitivity Analysis 6.1 Estimated economic costs and benefits for the project and the entire complex (both LAC 50 and LAC 93) are summarized below for the first year of 51 commercial operations of the entire complex (1988) and for the first year of full development (1993): NET ECONOMIC COSTS AND BENEFITS (in 1987 terms - US$ million) 1988 1993 Project Entire plant Project Entire plant Fixed costs 18.7 27.6 19.6 28.7 Raw materials 64.0 116.4 95.3 169.0 Other variable costs 3.4 6.2 4.3 7.9 Total Costs 86.1 150.2 119.2 205.6 Revenues (at import prices or "economic prices") 101.0 164.0 150.3 250.7 Net Benefits 14.9 13.8 31.1 45.1 6.2 The above table indicates that the plant shows positive net benefits in 1988. Despite this, the Economic Rate of Return (ERR) is still negative for the "base case". Estimated economic operating costs and output values for 1993 are further detailed in Annex XI. 6.3 Revised Costs and Benefits Streams for economic rate of return calcu- lations are presented in Annexes XV and XVI for the project and the entire plant, respectively and summarized below. ECONOMIC RATE OF RETURN (Z) Appraisal Revised Estima.es Estimate Project Entire plant Base Case 12.3 -1.0 -0.9 With appraisal Report Price assumptions - 7.2 8.3 6.4 Most of the shortfall of appraisal estimates is attributable to lower existing and projected international prices than assumed at appraisal, and the remaining to cost overruns and implementation delays. Further details on sensitivity analysis are presented below: 52 SENSITIVITY ANALYSIS IRR ERR LAC 50 LAC 93 LAC 50 LAC 93 Base case -0.3 +0.3 -1.0 -0.9 -2OZ capital cost +1.0 1.7 0.1 0.5 +10% revenue 4.0 2.8 1.5' 2.0 +40% revenue 7.0 8.0 7.0 8.0 +100% revenue 13.0 14.2 13.6 15.0 B. Foreign Exchange Savings and Employment Creation 6.5 Annex XVII gives estimates of foreign exchange savings in 1978 and 1987 terms: (a) as per appraisal for 1985 when production reached the highest level as assumed at appraisal, (b) as per appraisal for 1985 adjusted to take into account the value of foregone sulfur exports, (c) new estimates for 1993 when production will reach its highest level. 6.6 By 1993 most of the foreign debt, except for the Bank loan is assumed to have been repaid, the annual net exchange savings in 1987 terms is only US$82.6 million compared with the estimated saving of US$174 million at appraisal and after adjustment for the foregone sulfur exports. The main reasons for this decrease are: (i) the lower existing and projected import prices (used for economic analysis) of fertilizers; (ii) at appraisal all the phosphate was to be supplied from domestic sources (Baja California rock from ROFOMEX). Presently, ROFOMEX production is only about 500,000 tpy, of which 250,000 tpy are allocated to the project in the present estimate. 6.7 Since the foreign exchange component of the project has been re-esti- mated at US$189.9 million (versus US$124.5 million at appraisal) the project's foreign exchange pay-back period is now estimated at 2.3 years, versus less than one year at appraisal. 6.8 As expected in capital intensive industries, the project will gene- rate a relatively small number of jobs; a total of possibly 2,000 at full production (presently 1,530 jobs, see para. 3.8) for both project LAC 50 and the duplication project LAC 93 for a total investment of US$802.8 million, which means an investment cost per job of about US$400,000. C. Transfer of Technology 6.9 As in the case of the Fertilizer I Project, the project under review has been implemented with substantial participation of Mexican Engineering firms and manufacturers and in the process substantial transfer of technology was made. This is particularly significant for Fertilizer II where a very 53 large phosphoric acid plant and a very large sulfuric acid plant were built using modern processes. Before the project. Mexico's Fertilizer industry used rather old technology. For example, the country had also another large phosphoric acid plant; it was built in 1969, using a process that is now about 19 years old. VII. BANK'S ROLE 6.10 The Bank has played a% mportant role in addition to providing financing for the project. The process technology, offsites, infrastructure and utilities requirements, environmental and safety aspects, procurement procedures were extensively discussed with the Bank during appraisal. Bank missions also discussed the need for (a) a fertilizer distribution and trans- port study; (b) an operational staff training program; (c) the design, erec- tion and operation of the project facilities in conformity with accepted environmental standards; (d) reduction of FERTIMEX's working capital and improvement in its financial planning and control system; (e) financial cove- nants including government action, as required, to provide FERTIMEX with sufficient revenues and funds needed to bring the facilities to full produc- tion; and (f) progress and financial reporting requirements. 6.11 During implementation, the Bank conducted supervision missions every year since 1979. During these visits, Bank staff tried to raise important issues including the duplication project and its impact on the implementation of the original project. The most intensive supervision mission was done in 198", other missions averaged about one week. There are, however, other areas where the.Bank could have played a more adequate role (see below): VIII. CONCLUSION, SUSTAINABILITY AND LESSONS LEARNED A. Conclusion and Lessons Learned 7.1 From an engineering point of view, the plants of LAC 50 are well designed; the limited objective of implementing the project was achieved as all plants have been completed and started production; it is too early, however, to determine whether the overall technical and financial objectives of the project, including reaching full capacity production to meet demand will be achieved, particularly as the sustainability of the project is open to question (see below). 7.2 Actual production for 1986, production estimates for 1987 (the first year all the project components operate) and the capacity utilization in Z of rated capacity for the first 12 months of commercial operation, is sunmarized below: 54 Production Capacity 1986 1987 utilization ----- tons -------() Sulfuric acid 249,008 534,332 82 Phosphoric acid 61,309 148,511 76 DAP 56,028 309,400 60 Nitric acid - 55,794 70 (*) Ammonium nitrate - 60,000 63 (*) (*) For the first 3 months of commercial production. The performance of phosphoric and sulfuric acid production was better than the 70Z capacity utilization expected in the appraisal r3port while the production of DAP was limited by phosphoric acid availability. After three months of commercial production, the capacity utilization of nitric acid and ammonium nitrate plants is improving and will probably reach the appraisal report expectation of 70Z for the first year of commercial production. It is antici- pated that full capacity production in all plants will be reached by 1993. 7.3 Some of the secondary objectives of the project have largely been attained. About 1,500 jobs have been created for the total plant including the duplication of the phosphatic fertilizer output and many Mexican engineer- ing equipment and construction firms have benefitted from their involvement during the nine years of project implementation. 7.4 Overall the project's technical objectives will probably be met but with high cost as a result of delays in implementation. The design of the project was generally well conceived but the limited implementation capacity of FERTIMEX was not sufficiently understood, particularly concerning manage- ment aspects although FERTIMEX's financial difficulties were discussed in details (SAR paras. 3.12-3.13). 7.5 There were other shortcomings in the Bank's approach to this project, which was common to many similar projects at that time, and which reflected in part the then prvailing attitude of the Mexican Government towards policy reform; i.e., an unwillingness to discuss policy reform in the context of the oil boom during the late 70s when resources were abundant. These shortcomings in the approach to the project are: (a) unwillingness to tackle basic policy issues, even when resulting in sectorial inefficiency and financial dependence. At that time, the 55 Bank did not discuss the justification for fertilizer subsidies, but it could have insisted on a proper ex-factory pricing system to be set in place and covenanted. (b) insufficient monitoring of the investment program of FERTIMEX; the provision requiring FERTIMEX to inform the Bank of new investments exceeding US$20 million - discussed as a covenant in the SAR should have been included in the Loan Agreement. (c) unwillingness to enforce covenants through suspension of disburse- ments when major problems arose and lack of follow-up on findings of supervision missions except through exchange of letters. (d) as in some other major commodities, e.g., cement and clinker, the Bank had over-optimistic price assumptions, particularly interna- tional prices of fertilizers and failed to predict that World markets for fertilizer were collapsing. 7.6 A few other useful lessons of more general nature can be learned from the project experience: (a) In large fertilizer projects such as LAC 5C, the scope of the off- site, the utilities, storing capacity, etc. should be clearly defined to facilitate implementation and avoid the "improvisation" of a duplication project being built on the same site. It should be noted than an earlier fertilizer project (Mexico Fertilizer I, Loan No. .1112-ME) for which a PCR (Report No. 5382, December 1984) was recently prepared, had also a duplication project issue. (b) The general problem of project management by FERTIMEX was identified as early as Fertilizer I project (see PCR op. cit. para. 7.02 Lessons Learned). For future projects this persistent problem once identi- fied can be avoided by having the management structure in place before the project can be approved. The management arrangement can be based on a task force concept which is the usual practice for large projects, or a contractual arrangement with a general contrac- tor who should be a reputed industrial "generalist" firm,14/ not necessarily a fertilizer industry firm. In cases like FERTIMEX where project management had been a problem, a general assessment in SAR, that the company will assume project management and has appointed a project manager is an inadequate arrangement. B. Sustainability 7.7 The main question with sustainability of the project is its financial viability. In the reestimation, considerable price increases were required to reach World prices but the IRR is still negative; in this sense the project is not sustainable. It should be noted that even at appraisal the project carried 14/ i.e., a firm experienced in the implementation of large industrial projects, such as BECHTEL (US), UHDE, LURGI, KREBS (Germany), SPIE Batignole (France), COPEE (Belgium) etc.). The names are indicated only for clarifying the term generalist, not for recommendation. 56 an IRR of only 8.3%. 15/ The situation is not helped by the fact that the project was completed at a time when World fertilizer supply was in surplus 4nd prices depressed with io sign for upward changes in a foreseeable future. 7.8 As indicated in Sensitivity Analysis. it would require treating a very substantial amount of investments made (capital costs) as sunk cost ! raise the financial rate of return by any significant magnitude. The pr ct can only be considered as financially sustainable by treating part or al if the investments made as sunk cost. As in all capital intensive projects. shadow pricing of labor cost would not change ERR substantially; furthermore, relatively small reductions in capital cost are offset by the introduction of "economic prices" (import prices) which are higher than financial prices (ex- warehouse prices); it would take more than 40Z increase in revenue to bring ERR higher than IRR (see para. 6.4, Sensitivity Analysis Table). At 40% increase in revenue, the IRR and ERR are still low at 8%. 7.9 On the other hand, FERTIMEX as a whole can, in principle, sustain itself as an institution following the financial company reform proposed in the FSAL, a reform that would include the closing down of inefficient plants. The company has its strength in its engineering capability, high quality staff whose productivity can be further improved with continuing reform. But the impact of LAC 50 and LAC 93 on the future and overall sustainability of FERTIMEX should not be underestimated as the potential production of the two projects represent about one-fourth of FERTIMEX activities. 15/ The SAR mentioned indirectly that this is a low return but did not explain explicitly why the project should be approved despite its low return (see SAR para. 7.09). FERTI間EX Organlzation Chart of the Lazaro Cardenas Complex 抑 58 A," EX I I mæco SECOND FERTILIZEQ PROJECT-PCR PrOgraM~l & "al Imp~taflon Scheduýe Q^ 4 2 314 1 2 3 4 1 2 3 d 1 2 3 4 4 2 3 4 4 12 ýl, 121314 v£ W~ 2 No~ Ame 3 Aø~ 5~ hnq~ Cff~ mm 9 PI~ ~Cm 12 A~ Coni ýýci m~ EnGr~ I ý Dømma Enow~ %~ ~m lå Do~ b.,QV~ C"~ 16 kow~, (D~ 17 <O~) 21 EM~ P~ 22 b~ UNNW & o~ 23 C~n~ 2A Ccrr m d~ø ~ %m ýu Stan COM~ P Qøx 0 LAC 2 Roe~ (iw) 3 A~ Cc~ w~ erg~,q A" cwaffia".~ A p~ 6 taeg Aom~ (Del~ pou~ 3 cm W~ 10 b~ .1 2 N~ øcq- ......... A~ Acr~ 1979 E*'?~ A~~Gampou A~ ,,ffiffløffiffiffi\ em 0 -c c^m Eam~ N~ I cm W0N m~ *om~ 4 l mk~r~ 2 Q~ cø ~" ý0 tcke =ma~ fil* DU~m A~ 5 ch~ « to LAC 93 p ø~ 3 o~=ur9 et ØG~ K~ & m~ c( q~~ & m~ 0~ ojm «rha~ Fw~ wow km d13m ANNEX III MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Actual Capital Cost Expenditures Schedule (LAC 50) (Million Current Dollars) 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Total Process Plants Sulfuric acid plant 0.026 0.441 3.090 5.511 8.612 9.346 3.907 4.894 4.000 2.177 0.649 42.653 Phos. acid plant 0.024 0.401 2.813 5.016 7.838 8.506 4.353 4.353 2.893 1.777 0.583 38.557 Nitric acid 0.024 0.406 2.847 5.078 7.935 8.611 2.324 2.395 2.046 1.692 1.325 34.683 Ammonium nitrate 0.017 0.;97 2.014 3.51 5.6:.2 6.091 3.264 3.738 2.516 2.323 1.999 31.452 NPK/DAP 0.021 0.362 2.535 4.520 7.069 7.660 2.869 3.559 2.425 2.561 0.684 34.265 Subtotal 0.112 1.897 13.299 23.716 37.066 40.214 16.717 18.939 13.880 10.530 5.240 181.610 Ln Solid handling a 0.038 0.643 4.506 8.036 12.556 13.626 8.581 18.410 7.738 4.725 1.872 80.741 Utilities lb 0.073 1.236 8.659 15.441 24.128 26.184 10.545 23.858 17.670 9.963 6.740 144.497 Total 0.223 3.776 26.464 47.193 73.750 80.034 35,843 61,207 39.288 25.210 13.872 406.848 ja Prorated between LAC 50 and LAC 93 taking into account (i) the plant's production and the handling of 700,000 tpy of urea; (ii) a scaling factor of 0.8. lb Prorated between LAC 50 and LAC 93 taking into account the plant's production and a scale factor 0.6. Source: FERTIMEX ANNEX TV MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Project Financing Plan (US$ million) Actual /b Total Plant At time of appraisal Project (LAC 50) (LAC 50 + LAC 93) /b Foreign Local Total Foreign Local Total Foreign Local Total Equity Share capital 95.4 95.4 - n/a n/a - n/a nla FERTIMEX internal cash 6.9 19.0 25.9 - n/a n/a - nla n/a Subtotal 6.9 114.4 121.3 - 295.9 295.9 - 479.8 479.8 Long-Term Debt IBRD 80.0 80.0 80.0 - 80.0 80.0 - 80.0 > Exim Banks 4.6 4.6 - - - - - - Long-term loans 33.0 62.4 95.4 131.9 - 131.9 243.0 - 243.0 Subtotal 117.6 62.4 180.0 211.9 - 211.9 323.0 - 323.0 Total Financing 124.5 176.8 301.3 211.9 295.9 507.8 323.0 479.8 802.8 /a Long-tern debt has been estimated at 50Z of PRoject cost. Lb Total plant at Lazaro Cardenas including duplication and increase in off-site facilities. n/a: Not available. Source: Bank estimates. ANNEX V 61 HEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Actual Capital Cost Expenditures Schedule for all Lazaro Cardenas Plant (Current US$ million) Per year Cumulative 1977 0.29 0.2; 1978 5.03 5.32 1979 45.90 51.22 1980 73.20 124.42 1981 113.90 238.32 1982 123.80 362.22 1983 55.10 417.32 1984 99.00 516.32 1985 62.00 578.32 1986 38.10 616.42 1987 23.50 639.92 Subtotal (1) 639.92 Preoperating cost /a 6.62 Project cost 646.54 Working Capital Lb 41.30 Interest during construction.c 115.00 Total Financing Required 802.84 /a Estimated at 3 months of fixed costs. /b When maximu output is reached in 1991. /b Estimated. Source: (1) FERTIMEX. 62 ANNEX VI MEXICO SECOND FERTILIZER PROJE.CT (LOAN 1686-ME) PROJECT COMPLETION REPORT Working Capital Estimation for 1992 (1987 US$ million) Project Total plant (LAC 50) (LAC 50 + LAC 93) Finished products inventory (60 days @ pro- duction cost) 16.0 28.5 Raw materials inventory (30 days' consump- tion) 6.2 11.4 Minimum cash (1-month fixed costs) 1.5 2.4 Accounts payable (60 days' raw materials and 30 days' utilities) (12.7) (23.6) Short-term borrowings (38% receivables) (7.8) (13.8) Total Working Ca2ital 23.8 41.3 63 ANNEX VII MEXICO SECOND .FRTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Implementation Summary Start of Mechanical Start-up commercial Completion date operation Utilities Water treatment 06/20/85 06/25/85 06/27/85 Compressed air 10/19/85 10/21/85 10/21/85 Steam generation 11103/85 11/06/85 12/03/86 Cooling water 12/15/85 12/18/85 05/20/86 Project Process Plants Sulfuric acid 05/17/86 05/20/86 08/07/86 Phosphoric acid 07/13/86 07/15/86 07/24/86 DAP plant 11/07/86 09/09/86 11/09/86 Nitric acid 04/02/87 04/04/87 04/21/87 Ammonium nitrate 08/17/87 05/05/87 08/20/87 Duplication (LAC 93) Process Plants Sulfuric acid 07/07/87 07/10/87 07/25/87 Phosphoric acid 06/08/87 06/09/87 07/10/87 DAP plant 07/29/87 07/30/87 08/27/87 Source: FERTIMEX ANNEX VIII MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Initial Monthly Production '(Tons) LAC 50 LAC 93 /a Sulfuric Phosphoric DAP Nitric Ammonium Sulfuric Phosphoric DAP Year/Month Acid Acid Acid Nitrate Acid Acid Monthly Capacity 1986 May 4,771 - - June 11,575 - - July 9,127 953 - August 42.832 7,811 - September 40,537 11,855 3,254 October 41,807 14,485 9,047 November 50,879 14,014 12,312 December 47,480 12,189 31,4i5 Subtotal 1986 249.008 61,307 56,028 1987 January 37,112 7,184 12,465 February 51,300 16,633 20,681 March 41.002 14.042 24,465 April 47,844 13,426 22.095 3,046 May 49.316 11.860 36,082 - June 49,573 10,099 20,900 - 1,373 5,393 July 45.266 11.973 31,707 3,290 2,936 16,469 13,616 August -9,866 12.161 30,104 1.886 3,258 29,700 12,625 7.686 September 44,677 13.213 23,922 10.361 9.775 39,692 11,809 18,543 October 41,376 13,900 24,730 12,211 13,363 40,323 11,209 18,725 November /b 42,000 12,000 31,249 13.000 14,295 42,000 9,500 31.000 December Ib 45,000 12,000 31,000 13,000 15,000 45,000 10,000 33,300 Subtotal 1987 534332 148,511 309,400 55,794 60,00 213,j84 74152 19,j4 Capacity Utilization 812 75Z 60Z 70Z 63Zc 72Z 63Z/c 63Lc 1a Dtiplication, not financed by this project. /b Estimate. L Commercial production started in October. Source: FERTIMEX 65 ANNEX IX MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Expected Performance of Process Plants (Z of rated capacity) Annual rated Unit capacity 86 87 88 89 90 91 92 93 (tpy) (Actual) Sulfuric acid 660,000 78 80 90 95 95 95 100 Phosphoric acid la 198,000 75 80 90 90 95 95 97 DAP/NPK 525,000 57 /b 80 90 90 95 95 100 Nitric acid 215,000 70 80 90 100 100 100 Ammonium nitrate 270,000 70 80 90 100 100 100 /a Based on the use of 500,000 tpy of Baja California domestic rock and the remaining needs imported from Florida. Ib Fertilizer plant production is limited by phosphoric acid availability when DAP is produced. Source: Bank estimates. 9f) ANNEX X SECtOMO RTTLIR PROJECT <LOAN 1066-~E' PROJECT COLETtON *EPORT Pro.ecte, fin7ctal and teonoelc Prices (US/ton. .987 tare)> 1987 s98 989 990 1991 _.99 1992 .993 1e 2000 Sep Apr mo. Åpr mo- Apr qo. Apr 5ov Apr 'o, ýactua1) Finished Prod.ct la Bulk )AP: ex-fet=E (ten 8.2% Increqsel) 157.4 170.3 184.3 199.4 215.7 233.4 252.6 273.3 295.7 319.9 346.2 Les: 5(ctrbucton coatt 34.0 33.3 32.3 )1.8 31.0 30.3 29.5 28.8 28.0 27.3 26.5 Fquito46. ex-factory 123.4 13.1 151.8 167.6 184.7 203.2 223.1 244.5 267.7 292.7 319.7 Seport parity rnocKIc pricO) /b 181.5 182.5 182.5 183.5 183.5 185.5 185.5 194.0 194.0 202.0 202.0 210.0 06.0 219.0 E-factory Iftnanc1l prica) 2.4 137.1 15!.8 167.6 183.5 185.5 185.5 194.0 194.0 202.0 202.0 210.0 426.0 235.0 SPK: 17-17-17 'ten 1.22 Licreaee) c 132.0 142.8 154.5 167.2 180.9 195.8 211.8 229.2 248.C 268.2 290.3 Laes1: letrib1t,on coate 3.0 33.3 32.5 31.8 31.0 30.3 29.5 18.8 28.0 27.3 26.5 Eq4Ivalt Re-tactory 98.0 L0.6 122.0 935.5 149.9 165.5 182.3 200.4 220.0 241.0 263.8 Seport partiy <ecoonmc prire) L58.4 167.5 167.5 172.7 72.7 192.6 192.6 197.9 197.9 202.7 202.7 207.9 217.2 229.9 E-fUctory (finaL pre.? 98.0 109.6 122.0 135.5 149.9 165.5 162.3 197.9 197.9 202.7 202.7 207.9 217.2 229.' p O:' ex-FERTIMEK (ten 8.22 (ncrea***) ,d 132.7 143.6 155.4 168.L 161.9 196.8 212.9 230.4 249.3 269.7 291.8 303.7 325.8 3&2.7 Srt parlt. 1*concIte pric.> 260.0 267,1 267,1 274.2 274.2 281.4 281.4 293.8 293.8 306.3 306.3 311.8 341.9 359.7 t-factory (ftinnel pricel 132.7 143.6 155.4 168.1 181.9 196.8 212.9 230.4 249.3 269.7 291.8 303.7 32.8 342.7 TSP: (econoate priec 146.0 158.0 158.0 192.0 202.0 A~.& itract de-FEttmu /e 65.4 79.0 75.7 82.7 91.4 102.1 L15.2 131.2 150.7 174.5 205.5 146 ,ostri,button cost. 40.0 39.3 36.5 37.8 37.0 36.3 35.5 34.8 34.0 33.3 32.5 *qu (..lent ee-facory le (fin. orice) 25.4 30.7 37.2 45.0 54.4 65.9 79.7 96.5 156.7 141.2 173,0 173.0 171.1 181.4 tort parlcy (ecooi7c prtc*> It 15.0 133.0 133.0 14L.C 141.0 186.0 t86.0 186.0 186.0 183.0 585.0 185.0 å83.0 194.0 0APt e-fITM (ta 8.21 introaee 178.6 M*3.2 209.1 226.2 244.8 264.9 286.6 310.1 335.5 363.0 392.8 5ase Diseribution costs 40.0 39.3 38.5 37.8 37.0 36.3 35.5 34.8 34.0 33.3 a2.5 uluveoeot x-factory 138.6 154.0 170.6 188.5 207.8 228.6 251.1 275.3 301.5 329. 360.3 beoa past • 6u1 (economel prc) 11.5 182.5 182.5 1583.5 .83.5 185.5 185.5 295.0 194.0 202.0 202. 00 20.0 226.0 235 0 • 6egged (ecofo.Ie prlce) 193.5 154.? 194.5 '95.5 195.5 597.5 597.5 206.0 206.0 256.0 254.0 222.0 2'8.0 255.0 Uo-factory (finnetal prIes 138.6 554.0 170.6 58.S 195.5 197.5 197.5 206.0 206.0 214.0 214.0 222.0 2J8.0 251.0 I98: 57-57-57 (ten 8.21 o.rease.) 147.6 k59.7 172.8 L87.0 202.3 218.9 236.8 256.3 277.3 300.0 324.6 ..: DIutribution . c.t 40.0 39.3 38.5 37.8 37.0 36.3 35.5 34.8 34.0 33.3 32.5 esane ex-factory 107.6 120.5 534.3 L49.1 &65.3 t82.6 201.3 221.5 243.3 266.8 292.1 Igport perity - bulk (eceowei pricet 158.4 167.5 167.5 172.7 172.7 192.6 192.6 197.9 197.9 202.7 202.7 207.9 217.2 229.9 - bagged (econccite prico) 170.4 1'9.S 179.5 184.7 164.7 204.6 204.6 209.9 209.9 214.7 216.7 219.9 229.2 141.9 f-tectory (finanetal prte) 107.6 120.5 134.3 149.2 165.3 182.6 201.3 209.9 209.9 214.7 214.7 29.9 229.2 24.9 01 et-FtTIM (ta. 8.2 Ineresaee 132.7 L43.6 155.4 168.1 181.9 196.8 212.9 230.4 249.3 269.7 291.8 303.7 325.8 342.7 t partiy <cono.Se pric., 260.0 267.1 267.1 274.2 274.2 281.4 281.4 293.8 293.8 306.3 306.3 318.8 341.9 359.7 ex-f(.torv (f.nancial prCe) s32.7 543.6 555.4 568.1 585.9 96.8 22.9 230.4 249.3 265.7 291.8 503.7 325.8 342.7 MO 1f-u " 117.0 120.0 L20.0 LA6.0 124.0 127.0 L27.0 129.0 129.0 1.0 131.0 133.0 537.0 140.0 Phosphate rock • FlorLda 36.5 37.0 37.0 38.0 38.0 38. 38.5 39.5 39.5 å0.5 40.5 .2.0 44.5 .1.5 - hajt Callformla 30.5 3M.0 31.0 31.5 31.5 32.5 32.5 33.5 33.5 34.5 34.5 36.0 38.5 -5.5 #oLa • Fnasectai1 proce 35.0 40.6 67.1 54.6 63.6 13.5 85.3 98.9 154.7 133.1 155.0 - tmono~Oc prce 4t 120.0 141.0 561.0 161.0 161.0 207.0 207.0 187.0 187.0 186.0 186.0 185.0 184.0 200.0 Poeasluf eblorIde 76.0 83.0 83.0 90.0 90.0 96.0 96.0 99.0 99.0 102.0 102.0 105.0 112.0 123.0 l All flnaattal prIces are pased on ten 6.21 seantal Increass ta GR-RLTUM list prIces «til 1952 or låport par ty It loner thaa rekultint prtce. Th1 reta Is *ettnd to cover produCtIon Costa. tacludlng deprantattoo and fåanelal chargs and mnu a 81 recurn no equicy. ror äwotu attratt. the projeteud leare* ta higher becagse of plaot adjugtmote ta an5.a prIcos. /b concauc prIae are calculated by jagn Må0k projectonm for Long-tr VO1 Internattonal price plus freight and hanaitng cost dttfterntial (08811.3 for PAP). for 1987 the FOB prtic ts an average proce banad on Green Market publication and öéwä'# estLestes (ro. Prices of fortilserq and Chelcaluw puateriala, ad occasSemal sattestea dono by DCI. Actual E!aMle: DAP 567 190 ,955 2000 MalU CUtlf (1987 O8) 170.0 174.0 215.0 227.0 Tranport to (... sribble ...) 15.0 15.0 L5.0 15.0 inus trelgat fron LC to Top. for 70Z at prodttlöw 8 USIS/to (3.5) (3.5) (3.5) (3.5) total t8t.I ma5. 226.1 238.3 e rice tof 8 17-57-17 kg basad on equivalet bulk bland COat from a^an Gtrate, 0AP. poetLUa ehlortde and Ts plus 101 preiw. /d P 0 cocOt prlen te coMAtderod ~WGaIsiet to FOB U$ Gulf Ccot and Cal.uated from actual 1987 versge prica, ansfg thet st ¥M fotllge a a itr porice pattera Le the fetre an TU9. a EU-fauterr prte caicuute to c~ver prodbctiof Costa Includiet depretettne sad Loteses charee. pIus an 8 retur on 91g1t7 (501 ge tetit Ce9sc94 with -i.a price Ieoreast8 fron US$35 Lo 1987 to U18155 in 1992. /f t eEn ettste *e£c price la caleulated nder the teeuftln that 6 ctalnd ta 45 ha* a 10% prentäm compared to a contalind La Atea du* to lis bettet eros reepoea. The ftemetal mrftes et saltur. ph~8phate roek and potasGLO chlortde are elready et Iter *CoMnOC Lenele. Lh a§"fur te cmrretlr retaived r Vaaever and prtee IncIudes 057 transport roar. I # A st s cowaiderad te bo recaleud troe Slln Cru uot&I 1990. Fro 1991 tt tIl be produced at hae ea pemnsa plant ta Saro Crdeana. 67 ANNEX XI 'AEXICO SECOND FERTILIZER PROJECT (LOAN 1686-mE) PROJECT COMPLETION REPORT Estimated Production Costs and Nutput Value for 1993 Total Economic Financial quantities 'osts iosts Total Total Total Project plant Project plant Project plant ('000 tons) (1987 constant '2SS million) Cost of Production Raw Materials Sulfur 183.6 359.7 24.4 48.8 24.4 48.8 Phosphate Baja California 250.0 500.0 9.0 18.0 9.0 18.0 Florida 421.9 816.0 17.7 35.4 17.7 35.4 Ammonia 200.7 282.0 37.1 52.8 31.1 44.3 Potassium chloride 65.7 116.8 6.9 13.8 6.9 13.8 Coating agent 6.2 7.9 0.1 0.2 0.1 0.2 Subtotal 95.2 169.0 89.2 160.5 Other Variable Costs Power and water 'c 1.7 3.1 0.7 1.2 Fuel oil Id 0.8 1.5 0.4 0.8 Bags 0.9 2.1 1.7 2.1 Other 0.9 1.1 0.9 1.1 Subtotal 4.3 7.8 3.6 5.2 Fixed Costs Labor /b 4.6 6.7 4.6 6.7 Maintenance materials /f 9.2 12.0 9.2 12.0 Power and water /c 2.3 4.3 0.7 1.4 Insurance/overhead /e 2.3 3.3 2.3 3.3 Fuel /d 1.2 2.4 0.6 1.2 Subtotal 19.6 28.7 17.4 24.6 Total Cost of Production 119.1 205.5 110.2 190.3 Value of Output /a DAP 300.0 600.0 64.1 128.2 64.1 128.2 AN 189.0 108.0 35.0 20.0 32.7 18.7 NPK 225.0 450.0 47.5 95.1 46.9 93.1 P205 sales 11.6 23.2 3.7 7.4 3.5 7.0 Total 150.5 250.7 147.2 247.0 Gross Margin 31.4 45.2 37.0 56.7 /a Thirty percent of NPK and DAP and 100% of AN are assumed bagged. 7b Based on present salaries adjusted for 2,011 and 1,410 employees, respec- tively, for total plant and project. /c Power and water are considered, respectively as 402 and 602 variable cost. Financial power cost is considered at USS0.016/ Wh versus USSO.065 for economic cost, Water is considered at USSO.1/m . /d Fuel is considered at a financial price of USS60/ton and at an economic price of USS120/ton. Consumption is 12 kg/ton for fertilizer drying and 20,000 Lpy for steam production. /e Fifty percent of labor cost. 7 Two-and-a-half percent of fixed assets cost of process plant and utili- ties, and 1.5% of fixed assets cost of solid handling facilities. Source: Bank estimates. 68 ANNEX XII MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Costs and Benefits Streams for Financial Rate of Return Calculations Project (LAC 50) (1987 USS million) Incremental Opera Ang costs Capital working Raw Other Sales Net cost capital materials variables Fixed Total revenues benefits 1977 0.4 - - - - - - (0.4) 1978 5.6 - - - - - - (5.6) 1979 34.9 - - - - - - (34.9) 1980 56.6 - - - - - - (56.6) 1981 88.5 - - - - - - (88.5) 1982 96.8 - - - - - - (96.8) 1983 44.4 - - - - - - (44.4) 1984 77.7 - - - - - - (77.7) 1985 49.1 - - - - - - (49.1) 1986 26.7 - - - - - - (26.7) 1987 18.8 8.6 38.5 0.9 16.5 55.9 12.7 (70.6) 1988 - 7.0 48.7 2.9 16.5 68.1 68.7 (6.4) 1989 - 2.6 58.1 3.0 17.4 78.5 83.9 2.8 1990 - 2.5 63.8 3.4 17.4 84.6 102.6 15.5 1991 - 3.1 74.8 3.6 17.4 95.8 123.5 24.6 1992 - 2.0 83.9 3.6 17.4 104.9 135.9 29.0 1993 - 1.6 89.2 3.6 17.4 110.2 147.1 35.3 1994 - 0.4 90.5 3.6 17.4 111.5 150.5 38.6 1995 - 0.5 92.1 3.6 17.4 113.1 153.9 40.3 1996 - 0.3 92.8 3.6 17.4 113.8 156.2 42.1 1997 - 0.2 93.4 3.6 17.4 114.4 157.7 43.2 1998 - 0.3 94.1 3.6 17.4 115.1 159.8 44.4 1999 - 0.2 94.5 3.6 17.4 115.5 161.3 45.6 2000 - 0.2 95.4 3.6 17.4 116.4 162.8 46.2 2001 (50.0) (29.5) 95.4 3.6 17.4 116.4 163.5 126.6 Financial rate of return (base case): -0.3% Source: Bank estimates. 69 ANNEX XIII MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Costs and Benefit Streams for Financial Rate of Return Calculations Total Lazaro Cardenas Plant (1987 USS million) Incremental Capital working Operating Sales Net costs capital costs revenues benefits 1977 0.5 - - - (0.5) 1978 7.4 - - - (7.4) 1979 60.6 - - - (60.6) 1980 87.8 - - - (87.8) 1981 136.7 - - - (136.7) 1982 149.8 - - - (149.8) 1983 68.3 - - - (68.3) 1984 125.7 - - - (125.7) 1985 77.5 - - - (77.5) 1986 40.3 - - - (40.3) 1987 28.3 17.4 84.1 60.8 (69.0) 1988 - 9.8 124.2 120.7 (13.3) 1989 - 6.4 142.7 166.1 17.0 1990 - 2.9 151.6 187.0 32.5 1991 - 4.8 169.7 218.3 43.8 1992 - 2.1 102.6 228.0 43.3 1993 - 2.3 190.1 247.0 54.6 1994 - 0,9 192.6 253.9 60.4 1995 - 0.8 194.7 261.1 65.6 1996 - 0.5 196.0 265.1 68.6 1997 - 0.4 197.3 267.5 69.8 1998 - 0.5 198.6 271.0 71.9 1999 - 0.3 199.5 273.5 73.7 2000 - 0.4 201.3 276.0 74.3 2001 (78.3) (49.5) 201.3 276.0 202.5 Financial rate of return (base case): 0.3% Source: Bank estimates. ANNEX XIV SEC~ nETILIZEE m9<ECT (L~48 06~6-te) PU<JECT OJÆPLETI0rn RE10BT E&t-R«td r.fig.bility -nd VIe F-ofe t.areCres C-spe 1987 1988 1919 1990 199 1992 l993 194 §99S 1996 1991 998 Ø999 T.t.l a 60.446.26 120.6n1.44 166,164.44 t8.952.62 2a8.263.92 22.912.81 2.967.03 253.930.t8 268.058.18 261.069.96 2? ,544.10 2)$.019.95 2469.< £3.9. . Cash optrottag co..& k409z. 24.190.16 142,l42.13 .1f,93.21 169,624.M9 12,600.99 t90,j32.37 192.604.3 191*460.01 151.a198.61c, 99.1 3..4? 2Mi.?20.1 2ø2.4.5.1i 20!2.4N.< i 06ep.c..toW 30.700.00 50.100.00 0.000.00 00. 700.0 00.100.00 . 00,.00 sti.00 00.700.00 5o.bo0.00 S0.00.00 50.J00.00 50.100.00 ,o100.00 W0)co.o > S.blot41 14.9.65 174,690.4 193,442.1 3 202,293.21 220 424.19 233 30099 240,832.3 241.104 . 24,w00.? 247,8s.ø1 249,161. 6 220.41 z,< 420.! ZN i. I r -3s Cro.s RirLG (13,949.62) (1 213.12) (27.271.69) (OS-40.61) (2,860.81) ( t.126.<t 4-1,4.66 10.626.21 .,.us I1 86. .09 3D.42 .22&26 loter*.90 22.000.00 l6.200.00 14.300.00 t0,300.00 1,00.00 .100.00 2,900.00 300.00 - - IIet neraa (93,949.42 (12.413.12> (41,111.69) (25.640.00) (9~0.01) (10.624.18) 3.234.66 1 326.21 i410S-.o _jJ. . 36i 20 33.1< 2 avaU .. .e C..b Fla6 Cash <.fia. Gro* ~rg&. (13.949.62) (14.213.12) (2.2171.69) (15.340.61) (2,160.8W) (1.31.>8 6.34.66 10.626.21 14,108.1 1.18. 9 18. '.«.42 2u.54.al 2.069.0< 22,8#.49 / D.rton $0.100.00 10.100.00 10,100.00 10.100.00 0.100.00 10.100.00 10.100.00 $0.700.00 30.000.00 50.)0.00 50.00.00 0.00.00 S0.00.00 5.m..w10.100. B.btt (21249.62) 3.M3.42) 23.422.3< 33 .39.19 4 . 61.324.21 6,._ 6 . 9. 69,060.42 ?1!243. _? 8 1 < Zl.8 L, C*ob outilo~ (L.T debt O.rvsce) trtctpøi 1,100.00 1m00.00 1,300.00 I,100.00 1,3<.00 1. 3.300.00 - I.merest 1,400.00 4,700.00 4,000.00 3,300. 2.sm.00 2.k i 300.O- 0<bar 31.00.00 31.700.00 31.700.00 I1.700.00 17,800.00 7.10.,. . -' -- <at81es< 16,600.00 <3100.06 O,30.0 l 0.90 m000.00 1,30.00 1"00.00 I.00 - - 7.enetpei 39,000.00 39.000.00 39.000.00 25.000.00 25.000.00 25.000.00 2,000.00 3,100.00 - 22,000.00 16.200.00 14.100.00 t0. 30.00 1.ø0.00 1.00.00 2.900.00 300.00 1o6000.00 i200.00 1 . 32. 100.00 3,00. v .00 - m et Cst hfio ("4.249.62) (60,13..I2 (29,8. .69) 19.39 IS.19. I S.Ø11S62 2893..6 SI,?26.21 6n.2. é,1,8.09 69i61.42 442243..1 ff I of G o 6aeb s (60,099.62) (38.113.62) (.8;?.69) <2,409.39 2;.219.11 38.64.6 ,.;26z.2< &3.206.51 67.885.09 69.m60.42 /1,14 .81 /1,/69.<. 7I (48.024.42) (26,13.12) <,622.3< <6.i4.39 32.989.13 30.9p1.82 43.359.46 sp.21.2b U.208.11 41.888.09 49,060.42 t1.4.6 12.169.u 1.3%2m.6 t 1.14.,.. 1001 (3.,949.62) <<1,$<3.2> 2,22.13.1< 24,1"9.39 3,19. 3 36.211.62 46.434.6" 1.126.21 63.208.1< 41.888.09 49.060.42 :.241.B? 2.169.01 ?1.328.69 Ja Sourc.I al~ ogsttet. 71 ANNEX XV MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Costs and Benefits Streams for Economic Rate of Return Calculations Project (LAC 50) (1987 USS million) Incremental Operating costs Capital working New Ocher Sales Net cost capital materials variables Fixed Total revenues benefits 1977 0.4 - - - - - - (0.4) 1978 5.6 - - - - - - (5.6) 1979 34.9 - - - - - - (34.9) 1980 56.6 - - - - - - (56.6) 1981 88.5 - - - - - - (88.5) 1982 96.8 - - - - - - (96.8) 1983 44.4 - - - - - - (44.4) 1984 77.7 - - - - - - (77.7) 1985 49.1 - - - - - - (49.1) 1986 26.7 - - - - - - (26.7) 1987 18.8 10.7 47.0 1.1 18.1 66.2 23.0 (72.7) 1988 - 9.9 64.0 3.4 18.7 86.1 101.0 5.0 1989 - 1.7 75.9 3.6 19.6 99.1 105.5 4.7 1990 - 3.2 87.5 4.0 19.6 111.1 126.9 12.6 1991 - 1.6 90.8 4.2 19.6 114.6 139.9 23.7 1992 - 0.5 92.3 4.2 19.6 116.1 143.1 26.5 1993 - 1.0 95.3 4.3 19.6 119.2 '50.3 30.1 199.4 - 0.4 96.3 4.3 19.6 120.2 53.7 33.1 1995 - 0.5 97.9 4.3 19.6 121.8 157.1 34.8 1996 - 0.4 99.2 4.3 19.6 123.1 159.4 35.9 1997 - 0.3 100.4 4.3 19.6 124.3 160.9 36.3 1998 - 0.3 101.7 4.3 19.6 125.6 163.1 37.2 1999 - 0.2 102.8 4.3 19.6 126.7 164.9 37.7 2000 - 0.3 104.4 4.3 19.6 128.3 166.2 37.6 2001 (50.0) (31.0) 104.4 4.3 19.6 128.3 166.2 118.9 Economic rate of return (base case): -1.0% ERR with appraisal report price assumptions: 7.2% Source: Bank estimates. ANNEX XVI 72 MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-ME) PROJECT COMPLETION REPORT Costs and Benefit Streams for Economic Rate of Return Calculations Total Lazaro Cardenas Plant (1987 US$ million) Incremental Capital working Operating Sales Net costs capital costs revenues benefits 1977 0.5 - - - (0.5) 1978 7.4 - - - (7.4) 1979 60.6 - - - (60.6) 1980 87.8 - - - (87.8) 1981 136.7 - - - (136.7) 1982 149.8 - - - (149.8) 1983 68.3 - - - (68.3) 1984 125.7 - - - (125.7) 1985 77.5 - - - (77.5) 1986 40.3 - - - (40.3) 1987 28.3 21.7 98.1 92.7 (55.4) 1988 - 12.0 150.3 164.0 1.7 1989 - 4.7 172.4 189.9 12.8 1990 - 3.2 189.3 207.2 14.7 1991 - 2.9 196.1 229.5 30.5 1992 - 0.9 198.3 236.4 37.2 1993 - 2.1 205.5 250.7 43.1 1994 - 0.9 207.7 257.7 49.1 1995 - 1.0 211.0 264.8 52.8 1996 - 0.6 213.1 268.8 55.1 1997 - 0.4 215.3 271.3 55.6 1998 - 0.5 217.5 274.9 56.9 1999 - 0.5 219.2 277.4 57.7 2000 - 0.5 222.1 279.9 57.3 2001 (78.3) (51.9) 222.1 279.9 188.0 Economic rate of return (base case): -0.9% ERR with appraisal report price assumptions: 8.3% Source: Bank estimates. ANNEX XVII MEXICO SECOND FERTILIZER PROJECT JLOAN 1686-HEI PROJEOT COMPLETION REPORT Foreigr Exchanee Savings Present Appraisal estimates Appraisal restimates estimates for 1985 for 1985 for 1993 1978 US$ m 1987 US$ m 1978 US$ m 1987 US$ m 1987 US$ m Total Revenues 165.0 246.0 165.0 246.1 150.3 Less - KCL Imports 5.7 8.5 5.7 5.7 6.9 Foregone sulfur exports - - 12.7 18.9 24.4 Phosphate rock imports - La - - - 17.7 La Debt Service Schedule Interests 19.4 28.9 19.4 28.9 1.8 Principal repayment 10.7 15.9 10.7 15.9 16.9 Subtotal 30.1 44.8 30.1 44.8 18.7 Net Forelan Savings 129.0 192.8 116.3 173.9 82.6 La Assuming 250,000 tons only from domestic sources (Baja California). versus all requirements at appraisal. b Almost all foreign exchange debt (except Bank loan) assumed to be repaid by 1993. Source: Bank estimates. ANMX XVill MEXICO SECOND FERTILIZER PROJECT (tOAN 1686 ME) PROJECT COMPLETION REPORT FERTIMEX Balance Sheet - Historical 1982 80 to current USS million) 1992 1983 1984 1986 1911 1982 1983 1984 1986 21U8 Assets Liabilities and Equity Current Assets Current Liabilities Cash and Banks 38.2 22.0 64.9 44.1 8.9 Short term loans 546.1 100.8 201.7 334 6 133 1 A/C Receivable 39.8 110.9 189.8 126.9 70.9 A/C Payable 249.1 158.0 247.2 267.6 178.9 Inventories 146.6 151.8 262.0 800.6 208.8 Other 28.8 4.4 8.3 2.7 3.0 Total 209.6 284.7 446.2 470.F; 278.1 Total 818.6 283.6 462.2 694 9 313 6 fixed Assets Long-Term Liabilities Property, Plant & Equipment 351.3 788.5 1,714.5 1,304.4 1,583.8 Long-term Debt 392.8 763.3 700.8 636.6 801.7 Work in progress 941.8 1,256.6 1,080.0 1,000.8 520.4 Other 8.8 8.0 - - 2.0 .ese: Acc. Depr. - (386.8) (493.9) (406.3) (339.3) Total 339.1 769.3 106 8 836 83 7 Net Fi sed Asseta g .1 g ~Q3 ~ Lj Total Liabilities 1,217.7 1029 1,23. 1,111 3 (ther Asset* ShareholdereEquity Invest. In affiliates A Rec. 105.7 302.9 92.5 38.8 16.4 -Pa-in caplia 221.7 148.8 111.1 61.6 23 2 Other 3.2 2.8 2.6 1.6 3.7 Contrib. to f. . co. incr. 117.6 624.0 832.1 622.8 336.3 Surplus from somet,revaluation 209.4 684.3 1,117.3 764.7 410.4 Total Other 108.9 106.7 95.1 40.3 20.1 Retained earnings (94.7) (202.3) (177.1) (94.0) (33.6) Total Assets 1,671. 2,07._§ 2,042. 2L470.7 2 062.6 Total Equity 463.9 LR604.0 I,883.6 .240 7 9363 Total Liabilities & EgitZ 1,871.6 2A077. 2,82.8 2470:7 2,05? 8 fnd of Period ea. Rate USS12 98.5 144.0 192.6 311.7 923.6 .ource: Based on FERTIMEXv Balance Sheets in Spanis* f.nd in a different format. The present format was propered during the preparation of the fSAl 75 ANNEX XIX MEXICO SECOND FERTILIZER PROJECT (LOAN 1686-me) PROJFCT COMPLETION REPORT FERTIMEX - Profit and Loss - Historical 1982-86 (in current US$ million; 1982 1983 1984 1985 1986 Sales 474.5 367.8 453.4 451.5 377.5 Cost of goods sold 382.0 424.5 570.5 581.1 573.7 Gross Profit/(Loss) 92.5 (56.7) (117.1) (129.6) (136.2) Less: Selling, General and Admin. Exp. 99.2 47.2 58.5 85.3 64.1 Cost of plant stoppage 9.8 6.3 8.1 13.7 - Operating Profit/(Loss) (16.5) (110.2) (183.7) (228.6) (200.3) Less: Financial expenses /a 65.9 88.3 74.8 57.4 50.7 Other expenses/(income) 83.3 (4.3) (11.4) (5.8) (18.3) Net profit/(loss) before Extraordinary items (165.7) (194.2) (247.1) (280.2) (232.7) Less: Exchange losses Lb - 11.9 - - - Plus: Surplus realized - - 37.9 81.7 42.4 Plus: Government subsidy for - Operational losses and finance charges - 38.7 179.5 193.9 197.3 - Exchange losses - 0.6 0.3 - - Net Profiti(Loss) (165.7) (167.0) (29.4) (4.6) 7.0 Av. ex. rate used US$1 = 56.4 120.1 167.8 256.9 611.7 a Net of financial income. /b Surplus realized on depreciation of assets an inventories as a result of asset revluation for inflationary impact. Source: Based on FERTIMEX's Balance Sheets in Spanish and in a different format. The present format was prepared during the preparation of the FSAL. 77 ATTACHMENT I PROJECT COMPLETION REPORT FERTILIZER II PROJECT (LOAN 1686-ME) PREPARED BY FERTILIZANTES MEXICANOS S.A. PLANNING MANAGEMENT OFFICE June 1988 78 1. PREFACE Fertilizantes Mexicanos (FERTIMEX) is a parastatal company, whose main objective is to comply with Governmental policy to increase production and the use of fertilizers, making them available to the farmer in a timely fashion and at low prices. This support has contributed to the development of the agricultural sector, stimulating a continuous increase in the use of fertilizers. In the 1950s, the average rate of nutrient consumption growth was 33%; in the last 18 years the increase has been 7.4% annually, reaching a consumption of 1,86 million tons of nutrients. Of this figure, 72% is nitrogen (1.34 million tons) and 23.5% to phosphorus (438,000 tons of P205). Due to the significant increase in fertilizer consumption, the Mexican Government considers the national fertilizer industry as an area of strategic importance to achieve the necessary agricultural development to ensure food production for the population. During the mid-1973s, the national fertilizer industry was in a process of integration and had a series of ongoing projects which were not enough to cover the increasing demand for fertilizers required by the agricultural sector. The development strategy devised at that time was to speed up completion of those prcjeets and the creation of new ones, while at the aame time improving the other sectoral aspects: production, distribution, marketing and financial health. The Government, through its Industrial Development Plan, has placed the fertilizer sector in a priority category, along ,ith food, energy, and other industries that provide machinery and equipment to those sectors, with a view to using public companies as developmental foundations, promoting those capable of exporting, and substituting imports in a efficient manner. Thus taking advantage of local natural resources and promoting the geographic decentralization of the economic activity towards tne coastline, the borders, and other alternative locations to traditional industrial centers. In turn, through the expansion program of the fertilizer sector, the objective is to achieve self-sufficiency on fertilizers and have a sustained access to the international market, since there is a growing petrochemical industry and important reverves of phosphate rock and sulphur. Within this expansion program, the project of the Lgzaro CArdenas industrial complex in Michoacdn, in its two phases named 50 LAC and 93 LAC, represented an important part, since it implied increasing the installed capacity for finished products by 83% (in terms of nutrients), with respect to the existing installed capacity in 1977. The LAzaro CArdenas Project, along with others of urea, ammonium nitrate, and ammonium sulfate, were expected to achieve self-sufficiency of nitrogen and phosphate fertilizers by 1982, as well as to provide a production surplus to cover any une.ipected increases in demand. However, the delay in the implementation of the 50 LAC and 93 LAC projects prevented tht achievement of the projected goal with respeit to phosphorus. The main causes for delay were: the change in project scope, cwo high intensity earthquakes in 1979 and 1985, long delays oi the delivery of equipment 79 due to work overload of local suppliers during the time of the oil boom and the economic recession. In spite of these delays, the start-up of one part of the Lizaro Cirdenas phosphate fertilizer industrial complex at Lizaro Cirdenas in 1986 and its final completion in 1987 constituted a great achievement for FERTIMEX under adverse conditions and for its strategic importance in attaining objectives such as: achieving self-sufficiency in phosphorus during the 1987 second semester; a decrease in delivery and handling costs, as state of the art facilities became available closer to the market and pvoducing concentrated products. 2. BACKGROUND 2.1 Project Preparation and Feasibility Project 50 LAC was originally planned for Guaymas, Sonora, and it was projected that it would produce 1,750 tons daily of NP and NPR complex formulas. This location would permit taking advantage of the sulfuric acid to be obtained by a new steel company (Mexicana de Cobre) which was planning its establishment close to the site, as well as its proximity to the Northeast, an important agricultural zone with a significant demand for fertilizers. In view of the uncertainty in materializing the steel project and the lack of water in Guaymas, it was decided to change the location of the fertilizer complex towards tne new development center of Lizaro CArdenas in Michoacin. In- 1977 FERTIMEX assessed anew the 50 LAC Project, which was approved. Negotiations began with the World Bank to finance the project and quotations were requested from process and engineering bidders. In August 1978, the World Bank appraised the project on the basis of the feasibility study prepared by FERTIMEX and approved a USS80 million loan in April 1979. The Loan Agreement was signed on May 18, 1979, and it was agreed that the loan was destined to Fertilizantes Mexicanos, but with NAFINSA acting as guarantor, with a 15-year term for payment of the loan, including a 4-year grace period. The interest rate charged by NAFINSA to FERTIMEX for financing services is 3Z per year on the unpaid balance and the interest rate charged by the World Bank to NAFINSA ib 7% on unused balances. Therefore, +-he global financing rate paid by FERTIMEX is 10%. The loan was divided among the following categories: "A" LICENSES, ENGINEERING AND SUPERVISION USS9 million "r" EQUIPMENT PROCURED THROUGH INTERNATIONAL US$32.644 million BIDDING "C" EQUIPMENT AT A COST OF UNDER US$100,000 US.5.2 million "C-1" CRITICAL AND SPECIALLY-MANUFACTURED EQUIPMENT US$24.9 million "D" INTEREST US$8.256 million 80 Such a distribution differs slightly from that considered in the Staff Appraisal Report. The table in Annex 2 includes a breakdown of payments made by FERTIMEX to the World Bank during 1979-1987. 2.2 Proiect Description The 50 LAC Project consists of two finished fertilizer plants and three intermediate products plants. It also includes the necessary facilities to load and unload products and raw materials by land and sea, and all auxiliary services to operate the complex. Listed below are the main ard auxiliary facilities: a) Sulfuric acid plant of 660,000 tons per annum, Lurgi process. b) Phosphor acid plant of 198,000 tons per annum of P205, Gulf Swenson process. c) Nitric acid plant of 215,000 tons per annum, Uhde process. d) Ammonium nitrate solution and solid ammonium nitrate plants with a capacity of 200,000 tons per annum of solid product, Stamicarbon process. e)- DAP/NPK plant with a capacity of 525,000 tons per annum, Gulf process. f) Auxiliary services composed by different systems, the most important being the following: - Steam generating system. - Water treatment and distribution. - Electric power system. - Acid storage and handling. - Compressed air system. - Water cooling system. - Gypsum disposal into the sea. - Ammonia distribution and storage system. - Product storage. - Fire-fighting system. - Seawater network, and - Fuel system. g) A solid handling system which includes reception, storage, and handling of phosphate rock, sulfur, potassium chloride, and diatomite earth. Additionally, this system includes the handling, screening and shipment of final products. By the end of 1979, FERTIMEX decided to expand the Lizaro Cirdenas complex, by approving project 93 .AC which consisted, basically, of duplicating the 83 sulphuric acid, phosphoric acid, and DAP/NPK plants, which in turn required the expansion of the auxiliary services and solid handling systems. As to the supplies of raw mterials (sulphur, ammonia, phosphate 3ck, potassium chloride, and diatomite earth), it is estimated that the Unit w use 370,000 tons of sulphur per year purchased from the parastatal company Aa .rera Panamericana, which, through a trade agreement signed with a Canadian c 4pany, agreed to send to Lizaro Cirdenas sulphur coming from Vancouver, in exck .ge for sulphur coming from Jiltipan, Veracruz, exported from the Atlantic sit . Ammonia is supplied by PEMEX, and is produced in the southeastern region of the country (Cosoleacaque, Veracruz) from which it is pumped to Salina Cruz, Oaxaca, where it is stored until it is transported to Lizaro CArdenas. This situation will last until 1992, when PEMEX expects to start operations of an ammonia plant that will produce 445,000 tons per year. It is estimated that use of this material will reach 304,000 tons per year. As to the supply of phosphate rock, a total consumption of 1,344,000 tons per years is expected, of which it is estimated that 500,000 tons will be purchased from ROFOMEX. The remaining amount (844,000 tons) will be imported from Florida, USA. Annual requirements of potassium chloride for the production of NPK complexes will be covered by imports, whereas diatomite earth used as a coating agent will be procured by local suppliers whose development centers are located in the Central region of the country. It is important to note that consumption of these raw materials are estimated assuming that the Complex would only produce DAP and ammonium nitrate. 3. PROJECT IMPLEMENTATION AND ADMINISTRATION The Lizaro CArdenas industrial fertilizer complex in Michoacin, in its first stage (50 LAC Project), started commercial production in June 1987, i.e. 69 mGnths after the scheduled date of October 1981. Commercial start-up for DAP production was in November, 1986, whereas commercial production of ammonium nitrate started in August, 1987. 3.1. Changes in the Scope of the Project The main change in the scope of the project was that once implementation of the 50 LAC complex started, FERTIMEX deemed it convenient to simultaneously build a second phase, which it called 93 LAC Project, which included the duplication of the sulfuric acid, phosphoric acid, and DAP/NPK plants. This meant the redesign of auxiliary and general services, as well as the solid handling system, which additionally to this, was expanded to receive bulk fertilizer (urea or other), store it, package it, and deliver it by sea, railroad, or truck within the Mexican Republic, without affecting the operation, storage, bagging and delivering of products manufactured at the Complex. On the other hand, once piling started, damage caused by the 1979 earthquake forced the strengthening of the type of concrete foundation and 82 structures originally recommended, resulting in modifications such as the linkage with stoppers of the shallow foundations with shoes, uniform piling at a depth of 12.50 meters, and increase their cross section to 45 x 45 centimeters as well as placement over piles of the pipes' support. With respect to disposal of gypsum produced in the phosphoric acid plants, the original decision of its transfer to a pond was changed to sea disposal based on recommendations from the Engineering Institute of the National Autonomous University of Mexico after it carried out comprehensive studies to prevent environmental damage. Finally, another change worth mentioning occurred in 1982 from pneumatic to electronic instrumentation. 3.2 Project Management FERTIMEX was responsible for project implementation. The Projects and Construction Management unit was in charge of implementation. Initial organization (see Annex 3.1) included the formation of working groups in the field (Construction Residencies) and at headquarters (Project Chiefs, Procurement and Contracting Departments). This type of organization had difficulties due to lack of coordination and of cost and implementation control because, as priorities changed, the staff assigned to the project were given other tasks, or due to lack of sufficient material and human resources. In 1983, the organization for planning and implementing the Lizaro CArdenas project was modified. A Project Manager was appointed with responsibilities for construction, coordination, and programing areas. He also had the support of the engineering project chiefs (see Annex 3.2). Finally, FERTIMEX hired consultants to plan construction in each area in which the project was divided and formed internal areas to control works, costs and supplies coordination (see Annex 3.2.1), achieving greater progress and efficiency in project implementation. 3.3 Employment and Training The organizational chart on Annex 3.3, shows that the total number of workers at the Lizaro Cirdenas Complex would reach 2,000. At present, a little more than 1,500 people are in the site. It should be noted that with the change in project scope, the work force needs were increased by almost 220% more than what had initially been foreseen with only the construction of the 50-LAC Project (630 posts), filling most key positions (Superintendency, Area, and Shift Chiefs) with staff transferred from other FERTIMEX Units with experience in similar facilities. Tf the tLzAro CArdenas city's current status is compared with that when the project began, it can be determined that substantial improvements have taken place in the provision of services and infrastructure with a view to promoting regional development allowing for social and economic decentralization. However, 83 it is necessary to note that in the short term important investments should be made in those areas to add to those already in place. As to staff training, although a training program had been developed for the L4zaro C&rdenas complex since 1984, it was not until the beginning of 1986 that some actions were taken based on an actual identification of needs. T.is resulted in in-house training courses carried out during 1986 and 1987, even though some of these courses were given by institutions outside FERTIMEX (CABLESA, CONTROLMATIC, NTN, CONACYT and others). Training was primarily based on the training of supervisors as internal instructors, taking advantage of their previous experience and the knowledge acquired in start-up of the new facilities. It should be noted that special emphasis was placed on training newly-hired blue-collar workers, especially those in charge of the maintenance and industrial safety areas. 3.4 Utilization and Performance of Engineering Firms and Consultants Basic engineering, licenses, foreign procurement services and the detailed engineering and construction supervision for the processing plants and the solid handling area was handled by foreign firms, whose performance is considered to have been satisfactory. The original design of solid handling had to be modified because of project expansion. A new competitive bid took place for the new design, awarding the works to another foreign company. The design of auxiliary services was contracted with a local firm, which did not complete the works, making it necessary to contract another firm to complete them and making use of FERTIMEX own resources for the interconnections among the Complex's areas and for other activities stemming from the larger scope of the project. Detailed engineering was assigned to local firms (see Annex 3.4). The large number of companies participating in the areas of solid handling and auxiliary services caused coordination problems. Besides these problems, FERTIMEX faced other setbacks such as: (a) the overbooking of suppliers during the oil boom, which caused longer delays on delivery, on technical information, and on equipment; (b) shortage of funds caused suspensions, cancellations, and renegotiation of contracts with engineering firms and suppliers; (c) loss of technical information in 1985 due to the earthquakes; and (d) during 1984-85 there was a drain of qualif-ed technical staff due to lack of economic incentives. On the othet hand, it can be stated that the contacts developed by national firms with supervising staff from foreign firms provided valuable experience to the Mexican techniciane that could be applied to future projects. 84 3.5. Procurement and Performance of Suppliers Procurement of equipment and foreign-made materials financed with proceeds from the World Bank loan was carried out following for the selection of suppliers by International Competitive Bidding procedures (ICB). National and foreign (financed by sources other than the World Bank) procurement was done following guidelines established under the National Procurement Law for the Public Sector which also implies the selection of suppliers through competitive bidding. The Law also allows for a simplified bidding or bidding by invitation in cases where there are few specialized suppliers for a specific good. Direct contracting is allowed for small amounts. These procedures allowed for quality and cost advantages, however, procurement suffered from lack of implementation of a good system where procedures could be more agile and permit a good control of purchases. Currently, a computerized system is already in place to order, control, and record purchases; in addition, the reorganization of the procurement area which regrouped staff by project has brought about a better performance. The real cost of procurement for the 50 LAC Project can be compared with the estimates of the appraisal report, and only in the case of process plants, since for the auxiliaty areas the records include the expansion cost (93 LAC Project) without distinguishing the corresponding part of each project. Foreign procurement costs for process facilities in the 50 LAC Project resulted in overruns of 29% above the estimates in the appraisal report (Annex 3.5), with the major differences in the nitric acid and DAP/NPK plants. The supplier countries with the biggest share in the purchases made with the World Bank loan were the U.S., Germany, Italy, France, and Sweden (Annex 3.6). The cost of local procurement for the process plants in the 50 LAC Project was far greater (by 196%) than estimated in the appraisal report (Annex 3.5), due to escalations and the need to place purchase orders, because of lack of funds at FERTIM1EX, the overbooking of local suppliers, and cancellations of procurement services contracts. 3.6 Performance of Civil Works Contractors Several contractors worked in the construction of the Lizaro Cardenas Complex, and were directly supervised by Fertilizantes Mexicanos, who assigned specialized engineers in the different areas of construction and additionally contracted external local and foreign supervisors. The pe:formance of the contractors was satisfactory. Delays in project construction were due to reasons both internal and external to FERTIMEX, among them: - Changes in the type of concrete foundations and structures after the 1979 earthquake. 85 - Changes in the scope of the project, inclusion of LAC 93 Project. - Delays on detailed engineering caused by overbooking of local engineering firms during the oil boom. - Low availability of skilled work force given the poor social infrastructure at Lizaro CArdenas. - Shortages of structural steel at the national level. - The economic crisis suffered by the country beginning in the 1981 second semester, which forced the suspension of construction activities in most facilities, and the implementation of only those activities necessary to prevent damage to equipment and others related to erection, in an attempt to optimize the few available resources. - Setback of the progress achieved due to damages caused by the 1985 earthquakes. - Finally, during the last stage of the erection, testing, and start- up of the main equipments and process areas, technical assistance from foreign specialists was required but their timely arrival to the site of the works was not possible in part due to the lack of hard currency for immediate payment of services. 3.7 Implementation Schedule Real and estimated implementation schedules are shown in Annex 3.7. Total delay for completion of the 50 LAC Project was 69 months. With a view to clarify the reasons that caused the delays for completion of the Lizaro CArdenas Complex, the project has been divided into four phases, as follows: Phases Corresponding Period a) Initial phase 1979 to mid-1981 b) Economic crisis period Mid-1981-1983 c) Major progress phase 1984-late 1985 d) Final phase late 1985-1987 Following are the factors responsible for the delays in each phase: a) Initial phase (1979-mid 1981). During this period, FERTIMEX was implementing a great number of development projects, to which the duplication of the 50-LAC Project was added under the name of 93-LAC, bringing about changes in basic engineering for the areas outside battery limits and gradual changes in the organization of the company itself to make it adequate to the requirements of the projects executed in this phase. The explanation for each of those occurrences is as follows: 86 - In March 1979 when the Lizaro Cirdenas Staff Appraisal Report was prepared, an optimistic program envisioned project completion in 36 months. This period was to start with the awarding of engineering contracts (October 1978) and end with commercial production by all facilities (October 1981). It is considered that this calindar of activities was underestimated by 12 months. - FERTIMEX decided was convenient to build a second phase, known as 93-LAC Project, which included the duplication of the sulfuric acid, phosphoric acid and DAP/NPK plants, thus forcing the change and expansion of the engineering of auxiliary and general services and the solids handling system. - In 1978, the soil mechanics studies to determine the appropriate type of concrete foundations and structures for the Complex's facilities and equipment were carried out; however, in 197 , when piling had begun, an earthquake with an intensity of /.6 in the Richter scale took place, forcing the modification of the detailed engineering for the civil works because of reconsideration of the seismic facter criteria with which the original design was made, which considerably altered the areas already built. - Due to the economic boom experienced in Mexico during 1978- 1981, when GDP growth rates ranged from 7.95% to 9.15%, local engineering firms and equipment manufacturers were overbooked, causing engineering and equipment production delays and consequently delays in the development of detailed engineerirg and in construction. b) Economic crisis period (mid 1981-1983). This phase corresponds to a strong economic recession period reflected in the GDP, which in 1982 decreased by 0.5% compared to 1981 and by 5.3% in 1983 compared to 1982. During this period, the Government budget for parastatal companies, among them FERTIMEX, was restricted causing the Company's expansion programs to be limited to available resourcea. The Company chose to invest the limited funds on those projects showing considerable progress and those that could be completed with relatively few resources. The main factors during this period were as follows: - By the end of 1981, important FERTIMEX debts with suppliers and contractors, originated by budgetary restrictions, caused a nearly total suspension of expansion projects. - The local financial situation reached a point that FERTIMEX was forced to cancel purchase orders, given the unacceptable price escalation from suppliers. This lack of funds prevented obtaining the manufacturers' drawings to assist with engineering development and the opening of construction fronts especially for the areas outside battery limits. Likewise, 87 tha lack of funds prevented acquiring information for process areas to permit the total definition of the electrical engineering and instrumentatior areas, and for the outside battery limit areas procurement of main equipment had to be cancelled. This s.tuation caused delays both in engineering and in construction. - There was a critical shortage of production of structural steel at the local level, causing delays on the one hand on the erection of equipment in process areas, and on the other hand, making it necessary to import a large quantity of structures from Europe to avoid further delays. - With the limited financial resources available during this stage, priorities were established in the sense of defining which projects in the investment program would be suspended and which would have allocations. It was opted to invest on those projects which could be concluded with relatively small expenditures, i.e., the sulfuric acid plants in Quer6taro and Guadalajara and the urea plant I in Pajaritos Nitrogenados. - Consequently, construction activities in the Complex were stopped in most facilities, only carrying out activities to prevent weather damage to equipment and some erection works, in an attempt to be-;ter utilize the limited resources available. - Because of budgetary allocations, late in 1983 (June), procurement activities were again started as well as with specific restrictions and priorities construction activities, resulting in substantial progress up to the last quarter of the year. All process plants suffered from lack of materials both imported and local. In the areas outside battery limits, work was carried out where equipment and materials availability permitted (auxiliary services), and until September of the same year, significant activities were initiated in some working areas (handling of solids). Because of a labor dispute, terracing activities of a great part of the areas outside the battery limits were affected. In the case of 93-LAC Project, budgetary allocations were provided only for pending payments and equipment protection, however, taking advantage of available materials and equipment, greater progress than scheduled was achieved. c) Major progress phase (1984-end of 1985). In this phase, FERTIMEX had already concluded a large number of important projects and its investment program gave priority to the 88 Lizaro Cirdenas project. The main factors causing project completion delays during this phase were the following: - With procurement activities picking up during the second half of 1983, reception of instrumental and electrical control equipment for process areas and purchase of main and secondary equipment for the areas outside the battery limits was initiated. The latter caused the issuance of bid invitations for electromechanical erection for said areas, especially for solids handling; the results of building progress became apparent only in 1984. Even though the engineering for the process areas was practically completed by the end of 1981 and for the areas of outside battery limits in 1984, engineering activities were undertaken because of additions not originally contemplated in the projects as well as improvements to optimize operation of the facilities. - Finally, the earthquakes of September 19-20, 1985, provoked differential settlements on facilities over superficial foundations, causing serious damage to equipment already installed but mainly to raw materials and finished products warehouses; consequently, it was necessary to update design aspects, disassemble the affected facilities, demolish damaged superficial foundations, drive piles where originally they were not considered, and reinstall equipment. d) Final Phase. In this stage, it was necessary to overcome the physical damage suffered in the Lizaro Cirdenas Complex caused by the 1985 earthquakes. Material and equipment damaged by earthquakes were repaired or replaced in the case of process facilities with supplies originally purchased for the 93-LAC Project facilities, however, in the areas outside battery limits, repairs or replacements caused delays. The Project's main delaying factors are summarized below: 89 FACTORS RESPONSIBLE FOR DELAYS IN COMPLETING THE INDUSTRIAL FERTILIZER COMPLEX AT LAZARO CARDENAS, MICHOACAN FACTORS ESTIMATES (MONTHS) 1. 12 months underestimation in construction schedule established in the 1979 Staff Appraisal Report (poor program feasibility) 12 2. Modification of engineering of auxiliary, general, and solids handling services due to the incorporation of the 93-LAC project 6 3. 1979 Earthquake forcing modification of general arrangements and detailed engineering that was already prepared, considerably altering the facilities already built 10 4. Construction delays due to overbooking of local equipment manufacturers, causing long delivery delays 13 5. Delays in building and procurement caused by cancellation of purchase orders because of unacceptable price escalations and the local economic recession 20 6. Damages caused by the 1985 earthquakes 8 TOTAL PROJECT DELAYS 69 3.8 Investment Costs Estimated investments in the initial assessment and that obtained from FERTIMEX records are shown in Annexes 3.8 and 3.9, respectively. Table 3.8.1 is a comparative summary of both investments. Total Complex cost overruns, without considering the process plants for the 93 LAC project, adds up to 119.7% broken down by discipline as indicated below: 90 LAZARO CARDENAS COMPLEX COSTS OVERRUNS (As a Percentage of the Initial Estimated Investment) EngiMtrin Procurement Construction Administration Tota 50 LAC Project plants 49.0 50.2 96.2 35?.8 72.8 93 LAC Project plants Not considered in initial assessment Auxiliary Services 49.4 308.3 87.2 35!.7 152.7 Solids Handling 50.2 77.8 295.0 355.2 154.1 Total j/ 49.6 100.7 130.3 354.0 119.7 11 Without considering the 93 LAC Project plants. INVESTMENT COSTS AND REQUIRED FINANCING FOR LAZARO CARDENAS COMPLEX (USS Million) REFERENCE INITIAL EVALUATION I/ ACTUAL FIGURES ?/ ORIGIN/PLANT LOCAL FOREIGN TOTAL LOCAL FOREIGN TOTAL 50 LAC PROJECT SuLphuric Acid 14.85 12.08 26.93 32.509 10.144 42.653 Phosphoric 17.03 15.84 32.87 27.958 10.599 38.557 Acid Nitric Acid 7.88 10.14 18.02 21.830 12.853 34.683 Ammoniun Nitrate 7.47 6.83 14.30 26.826 4.626 31.452 DAP/NPK 8.76 4.25 13.01 24.928 9.338 34.266 93 Lac PROJECT Sutphuric Acid 23.604 7.640 31.244 Phosphoric Acid 23.189 10.663 33.852 DAP/NPK 14.802 5.750 20.552 AUXILIARY SERICEIS 59.71 17.79 77.50 174.964 20.910 195.874 SOLIDS OANDLIG 34.63 34.21 68.84 133.273 41.655 174.928 TOTAL 150.33 101.14 251.47 503.883 134.178 638.061 1/ SOURCE: Staff Appraisal Report. Mexico, Second Fertilizer Project (Lazaro Cardenas). g/ SOURCE: Subdireccln de Administraci6n de Proyectos. Gerencia de Integraci6n y Programacidn de Inversiones. The main factors responsible for cost overruns at the Complex were: more expensive cement foundation and structures after the 1979 earthquake; 69-month 91 delay on project completion with respect to initial plans, resulting in exe eive escalation; modification and expansion of auxiliary services and solid ha ling areas, caused by an 80% increase in the capacity of intermediate products ld of 66% in finished products, besides the increase in the solid handling are o be able to receive, store, bag, and distribute approximately 750,000 tons year of some imported solid fertilizer. On the other hand, regarding actual figures for recorded investmer ,, there were problems since FERTIMEX's cost records did not breakdown the ac. .nting of expensee paid in hard currency trom those paid in local currency. Zhere was confusion in assigning costs due to the inclusion of the 93-LAC Project facilities (especially those related to auxiliary services and solids handling) and there were difficulties in dividing administrative expenses for each project (50 LAC and 93 LAC). 4. OPERATIONAL PERFORMANCE 4.1 Prolect Implementation and Start-Up Start-up for the facilities at the LAzaro CArdenas industrial complex took place under difficult conditions caused by mechanical problems and equipment deterioration due to long storage. In spite of this, the present operation of the Complex can be considered satisfactory. Mechanical completion dates for producing facilities both for 50-LAC as well as 93-LAC, besides those corresponding to some auxiliary services and the dates for start-up and beginning of operations for each one, are shown in the Table below: UNIT DATE OF START-UP BEGINNING OF MECHANICAL DATE COMMERCIAL COMPLETION PRODUCTION FACILITIES: SULPHURIC ACID 50-LAC 05-17-86 05-20-86 08-07-86 PHOSPHORIC ACID 50-LAC 07-13-86 07-15-86 07-24-86 NITRIC ACID 50-LAC 04-02-87 O't-04-87 04-21-87 DAP/NPK 50-LAC 07-11-86 09-09-86 11-09-86 AMMONIUM NITRATE 50-LAC 05-05-87 06-17-87 06-20-87 SULPHURIC ACID 93-LAC 07-07-87 07-10-87 07-25-87 92 UNIT DATE OF START-UP BEGINNING OF MECHANICAL DATE COMMERCIAL COMPLETION PRODUCTION PHOSPHORIC ACID 93-LAC 06-08-87 06-09-87 07-10-87 DAP/NPK 93-LAC 07-29-87 07-30-87 08-27-87 SERVICES: WATER COOLING 12-15-85 12-18-85 05-20-86 STEAM GENERATION & DISTRIBUTION 11-03-85 11-06-85 12-03-85 POWER GENERATION & DISTRIBUTION 01-30-86 02-04-86 02-11-86 WATER TREATMENT 06-20-85 06-25-85 06-27-85 COMPRESSED AIR 10-19-85 10-21-85 10-21-85 Although tests were not directly made to verify warranties agreed with the different licensing companies, individual operational tests were performed on control equipment and systems to guarantee that the facility as a whole was in optimal operational condition. On the other hand, the resulting installed capacity was necestarily verified and as of today its results are satisfactory. The main problems found during start-up of the facilities were attributed to weather deterioration of equipment and instruments and long storage periods. Thus, it was necessary to perform preventive maintenance and corrective activities to maintain the equipment in satisfactory operational condition. In some other cases, it was necessary to purchase new eqtkipment and instruments because of the possible malfunctions on some equipment and instruments in the medium term due to the refurbishing carried out on them. The refurbishing carried out in all cases gave good results. Production records of all process plants installed at the Lzaro CArdenas Unit during the first months in operation are shown on Annex 4.1. 93 As can be seen the sulphuric acid facility reached 78% of its design capacity in its fourth month of operation. The main problems hampering production are: turbo-blower restrictions and difficulties in sulphur fusion due to the low quality of this raw material. These problems are being corrected and it is expected that a yield of 95% can be attained. The r'.osphoric acid unit reached 88% of its design capacity in its fourth month of operation. Problems faced are: the use of ROFOMEX rock which is of lesser quality than that available in the international market and difficulties with the filter slurry feed pumps. This is being corrected with the use of ROFOMEX-Florida rock mixtures. The DAP/NPK plant reached 72% of its design capacity in its fourth month in operation. Problems faced are: equipment damaged by long storage (dryer). This situation is being corrected. It is important to note that if the plant only produces DAP, its operation is limited by the production capacity of phosphoric acid (198,000 tons per year, P),O), since at full capacity DAP production would consume 277,000 tons per year of P20s The nitric acid plant reached 60% of its capacity in the sixth month of operation. Low productivity is due to the later start-up of the amonium nitrate plant, which is the consumer of this intermediate. The ammonium nitrate plant reached 80% of its capacity in the fifth month in operation. Difficulties faced are: mechanical problems with some equipment and too much dust thrown into the air from the prilling tower. Some mechanical problems have been already solved. 4.2. Production Normalization and Future Prospects Bearing in mind the satisfactory manner in which production in the different plants forming the LAzaro CArdenas Complex have been developing, it can be forecast that plants capacity utilization rates will have the following behavior starting 1987, year in which all plants should have initiated commercial production: POTENTIAL OPERATION OF THE LAZARO CARDENAS UNIT FACILITIES (% of Annual Nominal Capacity) FACILITY/YEAR 1987 1988 1989 1990 1991 1992 1993 SULPHURIC ACID 78 80 90 95 95 95 100 PHOSPHORIC ACID 75 80 90 90 95 95 97 NITRIC ACID 44 70 80 90 100 100 100 AMONIUM NITRATE 38 70 80 90 100 100 100 DAP/NPK L/ 57 75 90 90 95 95 100 jf The granulation plant has an overcapacity with respect to the phosphorie acid plant when it only produces DAP. 94 It should be noted that both the financial evaluation and the economic evaluation for .e Project were prepared, based on the above capacity utilization rates. 4.3. Fertilizers Market In the appraisal report for the 50-LAC Project it was estimated that demand for nitrogen and phosphorus (P205) would grow by 7.8% and 13.3% per annum. respectively, during the period 1978-1985. In both cases forecasts were above the actual data, as shown in the following Table: CONSUMPTION OF NITROGEN AND PHOSPHORUS (in thousands of tons of nutrient) YEAR NITROGEN PHOSPHATES Projection Actual Data A/B Projection Actual Data A/B 1/ (A) (B) 1 (A) (B) 1978 992 777 1.28 286 249 1.15 1980 1,137 953 1.19 341 298 1.14 1982 1.316 1,181 1.11 403 408 0.99 1985 1.680 1,310 1.28 521 426 1.22 1987 1,345 - - 438 - [/ Estimate of the 50 LAC Project Appraisal Report. As to solid fertilizers, summarized below are production and demand estimates compared with actual data: PRODUCTION AND DEMAND FOR NITROGEN SOLID FERTILIZERS (thousands tons of nutrients) YEAR ESTIMATE ;/ ACTUAL DATA Demand Production Demand Production 1978 776 421 587 421 1980 900 750 732 548 1982 1,059 870 963 883 1985 1,395 1,150 1,054 1,116 1987 - - 1,090 1,141 / Estimate of the 50 LAC Project Appraisal Report. 95 As can be inferred, on the one hand the lower growth of nitrogen demrnd in the last five years as compared to estimates; and on the other, the start-up of new urea and ammonium sulphate plants, allowed to achieve a balance between production and demand, not withstanding that the LAzaro CArdenas fertilizer Complex had not yet begun operations. With respect to solid phosphate fertilizers, a comparison between actual data and estimated values is shown below: PRODUCTION AND DEMAND OF SOLID PHOSPHATE FERTILIZERS (in thousands of Ps05 tons) YEAR ESTIMATE L/ ACTUAL DATA a/ Demand Production Demand Production 1978 286 415 249 227 1980 341 460 289 201 1982 403 540 408 252 1985 521 565 419 267 1987 - - 418 399 Estimates from 50-LAC Project Appraisal Report; figures include phosphoric acid. U/ Excludes phosphorics acid. In this case, because the LAznro CArdenas Complex encompasses mostly the production of phosphatic fertilizers, delays in its implementation caused a significant phosehorous deficit situation during 1980-1986. It is expected that within the next three years with normal operation of all LAzaro CArdenas plants, a balanced phosphorus supply-demand situation will be achieved, however, beginning in 1992, demand growth will cause an increasing deficit, and that is why FERTIMEX is planning the construction of a new ammonium phosphate facility (117 PAJ Project) which would make it possible to maintain self-sufficiency. 4.3.1 Price of Fertilizers The general policy followed in the last 16 years has been to keep prices low to promote its use in local agriculture. On average, local farmers have purchased fertilizer at 50% of its international price. In 1987, local prices at railroad storage points of CAP, complex and ammonium nitrate registered levels ranging from 67% to 53% of international prices (CIF Western coast of Mexico). See Annex 4.4. Consequently, this policy has caused large budgetary transfers from the Government to FERTIMEX. 96 1,e Government has recognized the need to implement a price policy that would allow FERTIMEX to achieve financial self-sufficiency. Thus, there is a plan underway to increase L cal prices by 1992 to such a level &a to allow cost recovery and obtain some return on investments. This will obviously go hand in hand with restructuring measures to reduce costs and optimize operations. 4.4 Environmental Pollution Control During construction of the LAzaro CArdenas Complex's process plants and auxiliary services, local environmental laws and existing international standards were taken into account with respect to permissible maximum ceilings of undesirable compounds such as sulphur dioxide, nitrogen oxides, fluorides, and dust particles. However, in spite cr the efforts made there are some small problems, such as: - The ammonium nitrate plant, which has maintained high dust emission levels through the exhaust air of the prilling tower; however, it is a problem that can be solved jointly with the supplier. - On the other hand, given that the area is earthquake-prone, FERTIMEX modified the original concept with respect to the disposal of gypsum in an area located some 8 km. away from the Unit, to its disposal at sea, diluted with the aqueous effluents from the evaporation condensers and the final washer from the DAP/NPK plant. When the latter started-up the following difficulties appeared: a) One out of the two existing ltnes broke down. b) The other discharge line became blocked. C) The effluent pumps presented high vibration. FERTIMEK repaired as soon as was possibla the broken line, while the other is still Llocked. At present, it is implementing an emergency program consisting of modifying the effluent pumping pits and reinforcing the structures of the pumps to avoid wastewater from being disposed into the Balsas river through an open channel. Additionally. FERTIMEX is planning to restore the blocked line, redesign the pump pit as well as to purchase a new pump. 5. FINANCIAL PERFORMANCE FERTIMEX's financial situation has been deteriorating for the last ten years, during which subsidies and debt have reached very high levels, as shown inAnnexes 5.1 and 5.2. In 1984-85, direct operational subsidies represented 30% to 47% of costs while in 1986 the documented debt grew to more than 800 billion Pesos. These financial problems are partly caused by factors beyond the company's control, such as: the low fertilizer prices policies and the continuous devaluation of the Peso. However, the bad financial situation is also due to inadequate project administration and control of the Company's projects and working capital. Delays on project implementation increase construction costs 97 and interests payable on project financing -- this obviously increases budgetary transfers and debt. Besides, FERTIMEX has required a very high working capital because of high inventortes and accounts receivable and although this offsets in part the high accounts payable, the working capital is still too high. It is hoped that FERTIMEX's financial performance improves with the implementation of the Industry Reconversion Program which promotes a structural change establishing the basis for improvement of operational efficiency and support development with financial self-sufficiency. 5.1 Estimated Financial Rate of Return To determine the internal financial rate of return, the existing local prices in November 1987 for raw materials were used, except for ammonia, the price ol which is expected to increase in real terms gradually as a percentage of PEMEX's export price, FOB Coatzacoalcos, so that for 1989 it will be 54%, for 1990 63%, for 191 71%, and finally for 1992 and thereafter, 80%. As to fertilizer prices, it was agreed that they will be adjusted according to the inflation rate with a view to maintaining up to March 1, 1989 the value in real terms reached on March 1, 1988, and from that date on prices will increase 12.3% annually during four years so that FERTIMEX achieves financial self-sufficiency. Raw materials' and fertilizer prices used in the financial evaluation are shown on Table 5.1. On the other hand, production costs determined on the basis of the above raw material prices used in the financial evaluation, are shown on Annex 5.3. The resulting internal financial rate of return was negative (Annex 5.4), which differs from the 4.3% rate (rate afte -axes) computed in the initial appraisal. To explain the reasons behind that difference it is necessary to go back to the resultE obtained in the Staff Appraisal Report of which the following are highlighted: CASE INTERNAL RATE OF FINANCIAL RETURN 1 Base 8.3% Revenues decrease by 5% 5.7% Investment increase by 10% 7.1% Production costs increase by 10% 4.3% Delay on completion (one year) 7.4% Ammonia at international price level 3.4% Fertilizer prices at 1978 level -4.4% j/ Values defined before taxes. 98 PRICES OF RAW MATERIALS AND FERTILIZERS USED IN THE FINANCIAL EVALUATION USS/TONS YEAR/PRODUCT 1987 1988 1989 1990 1991 1992 1993 SULPHUR 132.00 132.00 132.00 132.00 132.00 132.00 132.00 AMMONIA 1/ 36.60 54.80 72.90 110.30 128.70 149.90 154.90 PHOSPHATE ROCK 2/ 33.68 34.51 34.69 34.69 34.78 34.78 34.82 AMMONIUM NITRATE 61.z1 79.83 89.65 100.68 113.06 126.97 126.97 DAP 167.27 218.88 245.80 276.04 310.00 348.12 348.12 The ammonia price considered for the following years was: 1994 -159.9; 1995 - 164.8; 1996 - 166.7; 1997 - 168.6; 1998 - 170.5; 1999 - 172.4 and for the yeat 2000 - 174.4. i/ Average price weighted according to local and imported amounts consumed. As can be observed, the financial IRR before taxes was very sensitive to decreases in revenues, to production costs increases, and to the price of ammonia used, on the other hand, it had some sensitivity to increases in investment and to project completion delays. Considering the above, actual production costs in constant terms resulted higher for the first years -- between 9% and 13%, compared to those of the Staff Appraisal Report. The price of ammonia used was assumed to be at 80% of the international price, well above the highest price considered in said report; the project was delayed in its start-up by 69 months, the project scope was expanded, and there was a significant cost overrun. It is obvious that the simultaneous combination of all these negative factors had a significant effect in the difference obtained as to the original financial return of the project. 6. ECONOMIC PERFORMANCE 6.1 Economic Rate of Return The internal economic rate of return was calculated considering economic prices for materials ana final products, which were determined based on the international price projections assumed by the World Bank (Annex 6.1). In the case of ammonia, it is expected that from 1917 to 1991, surplus ammonia from Cosoleacaque will be used, and that beginning in 1992, that being produced in the PENEX plant at LAzaro CArdenas will be utilized. With respect to sulphur, there is a trade agreement with Canada in which part of the sulphur exported by that 99 country is supplied with Mexican sulphur in the eastern coast of Mexico and in turn replaced by Canada in the L&zaro CArdenas port. As to the phosphate rock, some will be of local origin and the other will come from Florida, the opportunity price having been calculated at under this pr#nise. There are no international price projections for ammonium nitrate, thus its economic price was determined from the urea international price, calculating its equivalent in terms of nutrient and add.ng a premium factor of 10% due to its better agricultural response. Detailed calculations of raw material economic prices and finished prodacts are shown on Annex 6.2, and are as follows: ECONOMIC PRICES OF RAW MATERIALS AND FINISHED PRODUCTS (USS/Ton) 1987 1988 1989 1990 1991 1992 1995 200u AMMONIA 108.6 123.6 143.6 183.6 189.6 167.1 185.7 197.7 SULPHUR 107 115 125 130 132 134 140 140 PHOSPHATE ROCK 25.6 33.8 36.9 38.5 39.2 39.2 39.5 39.5 AMMONIUM NITRATE 116.1 120.1 136.1 168.1 173.1 178.1 189.9 199.5 DAP 190.6 189.6 202.6 226.6 238.0 249.4 283.6 289.6 On the other hand, production costs determined by using the above raw material prices and used in the economic evaluation are shown on Annex 6.3. The cash flow in the economic evaluation (Annex 6.4), was calculated at present value, resulting in a rate of return of 0.01%, very inferior to the 12.3% forecast in the initial evalua,-tn. The difference in the project's rate of return from the planning phase -o the implementation stage has been attributed to project cost overrun, implementation delay, and change in the scope of the project; however, the failure to materialize the original projections of international prices for raw macerials and finished products had a significant impact, since globally, if these projections had been maintained at 1987 prices, additional favorable economic cash flows would have been achieved, as follows: 100 YEAR ADDITIONAL POSITIVE BALANCE IN STATEMENT OF RESULTS (US$ million) 1987 19,380 1988 48,338 1989 56,237 1990 46,373 1991 43,777 1992 25,872 1993 4/ 18,720 1/ It is the last available projection in the Staff Appraisal Report. The above would cause at least a two percentage poiiLt increase in the internal economic rate of return. 6.2. Foreign Exchange Savingg The value, in foreign exchange, of the production of the LAzaro CArdenas Complex at its highest capacity utilization rate and at 1987 prices adds up to US$174.0 million. On the other hand, annual debt service paymants of US$52.7 million will have to be budgeted during the first ten years; likewise, US$40.1 million will have to be assigned to import 844,000 tons per year of phosphate rock; thus, net foreign exchange savings at current prices and highest capacity utilization amounts to US$81.2 million per annum. 6.3. Transfer of Technology Engineering firms and national equipment manufacturers played an important role in the construction of the LAzaro CArdenas Complex as shown by the data below: ENGINEERING ENGRG. % CARRIED OUT % CARRIED OUT FIRMS CARRIED OUT IN AMOUNT IN PHYSICAL WORK LOCAL DETAILED ENGRG. 41 70 FOREIGN BASIC ENGINEERING 59 30 PROCUREMENT ORIGIN AMOUNT (US$ Million) PERCENTAGE LOCAL 165,316 57.81% FOREIGN 120,647 42.19% TOTAL 285,963 100.00% As can be seen, based on the amounts spent for contracts with engineering firms, 59% was carried out by foreign firms and 41% by local firms; however, 101 considering the physical development of the works, 70% was performed by local firms. Consequently, most of the technology was assimilated and left Mexican technicians wit' enough experience. As relevant aspects, the use of better quality materials and the use of electronic processors, may be mentioned. The innovation of the isothermal reactor of phosphoric acid plant and full utilization of energy in the ammonium nitrate process should also be mentioned as two good examples. The experience acquired will be applied in other projects depending on the protection granted to the original processes according to contractual clauses. 7. Findings and Lessons Learned There is no doubt that the completion of the fertilizer industrial complex at LAzaro CArdenas is an important achievement for the company, since the installed capacity for the production of nutrients of the 50-LAC y 93-LAC Projects represent, approximately, cne third of the national total. HorTever, this achievement is limited by lack of economic and financial profit9bility, basically a result of cost overruns and implementation delays. A feasible project, such as -he 50 LAC Project, usually has no difficulties in obtaining implementation financing. However, changes in the s.ope, such as expansions and modifications in the original design which imply higher capital costs, and have no secure or properly planned financing, ruin the profitability perspectives of the project. Nevertheless, it is important to take into account that the cost overruns and delays in project implementation were also caused by factors beyond the control of the company, such as earthquakes and the country's economic crisis. On the other hand, deficient information of real costs in each area of the Complex and the difficulty in obtaining daca on problems arising during implementation, indicate that there was lack of control over costs and program of the Project. The various changes made in the organization for project implementation and the enormous cost overruns in administration in relation with that planned, indicate that project administration was not adequate. NOTE: Thn Annexes attached to this Report have not been translated and are available in the Project's files.