Madagascar - Petroleum Sector Reform Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 19399 IMPLEMENTATION COMPLETION REPORT MADAGASCAR PETROLEUM SECTOR REFORM PROJECT (Credit No. 2538-MAG) June 10, 1999 Energy Team Infrastructure Group Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Currency Unit = Malagasy Franc (FMG) Year US$1.00 = FMG 1989 1603 1990 1494 1991 1835 1992 1864 1993 1914 1994 3067 1995 4266 1996 4061 1997 5091 1998 5273 WEIGHTS AND MEASURES Metric System FISCAL YEAR OF BORROWER January 1 - December 31 ABBREVIATIONS AND ACRONYMS AFD Agence Francaise de Developpement ESMAP Energy Sector Management Assistance Program ICB International Competitive Bidding LGP Liquefied Petroleum Gas MEM Ministry of Energy and Mines OMH Office Malgache des Hydrocarbures SOLIMA Solitany Malagasy Vice President Mr. Callisto Madavo Country Director Mr. Michael Sarris Sector Manager Mr. Mark Tomlinson Team Leader Mr. Angel Baide  IMPLEMENTATION COMPLETION REPORT FOR OFFICIAL USE ONLY MADAGASCAR PETROLEUM SECTOR REFORM PROJECT (Credit No. 2538-MAG) Table of Contents Page No. Preface Evaluation Summary ................ - iv Part I: PROJECT IMPLEMENTATION ASSESSMENT I. Project Background ............................... 1 II. Statement and Evaluation of Objectives ...............................4 III. Achievement of Objectives ...............................4 IV. Major Factors Affecting the Project ...............................8 V. Project Sustainability ...............................9 VI. Bank Performance ....9..........................9 VII. Borrower Performance .............................. 10 VIII. Assessment of Outcome .............................. 10 IX. Future Operation .............................. 10 X. Key Lessons Learned .............................. 10 Part II: STATISTICAL ANNEXES Table 1 Summary of Assessments .............................. I I Table 2 Related Bank Credits .............................. 13 Table 3 Project Timetable .............................. 14 Table 4 Credit Disbursements .............................. 14 Table 5 Key Indicators for Project Implementation .............................. 15 Table 6 Studies Included in Project .............................. 16 Table 7A Project Costs .17 Table 7B Project Financing .17 Table 8 Status of Legal Covenants .18 Table 9 Bank Resources: Staff Inputs .21 Table 10 Bank Resources: Missions .22 Part im: APPENDICES A. Aide-Memoire of ICR mission ....................... 23 B. Borrower's Contribution ....................... 26 Map: IBRD No. 30305 This document has a restricted distribution and may be used by recipients only in thie performance of their official duties. Its contents may not otherwise be disclosed without World Bank authiorization.  IMPLEMENTATION COMPLETION REPORT MADAGASCAR PETROLEUM SECTOR REFORM PROJECT (Credit No. 2538-MAG) Preface 1. This is the Implementation Completion Report (ICR) for Madagascar's Petroleum Sector Reform Project, for which Credit 2538-MAG in the amount of SDR 36.7 million was approved on July 29, 1993. An amendment to the Credit and Project Agreements was approved on October 18, 1996, and became effective on October 7, 1997. The amendment cancelled SDR 9.4 million, which had been destined to finance a rehabilitation of the Borrower's refinery at Toamasina. 2. The credit closed on December 31, 1998, the original closing date. It disbursed only SDR 3.3 million or 9 percent of the original credit amount and the last disbursement took place on May 21, 1999. No co-financing was involved. 3. This ICR was prepared by Angel Baide (AFTG 1), and reviewed by Michael N. Sarris (AFC08), Patricio Millan (LCC5F), Thuvara S. Nayar (EMTOG), and Noureddine Bouzaher (AFTG1). The Borrower reviewed the draft ICR and, during the ICR mission in April 1999, provided oral comments which were taken into account in preparing the final version. The report was processed by Lily Wong, Team Assistant (AFTG1). 4. The ICR is based on materials in the project files, as well as on the ICR mission's findings. The Borrower contributed its own evaluation of the project's design and execution, which is included as Appendix B to the ICR.  IMPLEMENTATION COMPLETION REPORT MADAGASCAR PETROLEUM SECTOR REFORM PROJECT (Credit No. 2538-MAG) Evaluation Summary Introduction 1. The concept of the Petroleum Sector Reform Project evolved from the findings of an ESMAP energy sector assessment published in 1987, and a World Bank study of the petroleum sector completed in 1989. These studies identified the following petroleum sector problems: (a) deteriorated state of installations and equipment; (b) inefficient import and refining operations; (c) inefficient pricing and taxation of petroleum products; and (d) non transparent and inefficient procurement of crude oil and refined products. Solitany Malagasy (SOLIMA), the national petroleum company, was in charge of all petroleum operations and installations in the country. The government had created SOLIMA to take over the operations of foreign oil companies when it nationalized the sector in 1976. 2. At the time the project was identified, the government was particularly interested in rehabilitating the country's refinery at Toamasina. In 1982, SOLIMA had completed, at a cost of about $20 million, a major reconfiguration of the refinery -- built in 1966 as a simple hydro-skimming facility -- by adding a visbreaker, a desulfurization unit, and a bitumen unit. But the new section had been disabled by an explosion shortly after commissioning. Moreover, while this extension was being built, the rest of the refinery had deteriorated rapidly, because the heavy investment program had used up all of SOLIMA's foreign exchange, leaving little for maintenance needs. As a result, the refinery suffered frequent accidents, including fires, which forced it to shut down for months at a time. 3. Although the Bank studies expressed doubts about the feasibility of refinery rehabilitation, mainly because its justification depended on the price differential between crude and refined products, which was determined by external factors, consultants financed by a project preparation facility (PPF) showed the rehabilitation to have a very high rate of return. The same consultants also pointed to the serious environmental and safety risks posed by inadequate existing loading-unloading facilities at Toamasina, and by the old, leaky pipelines connecting those facilities to the refinery, and recommended building a new maritime petroleum terminal and replacing the pipelines. The Bank accepted to finance a first stage rehabilitation of the refinery, as well as construction of a new petroleum terminal and pipelines at Toamasina, as part of a project that would liberalize downstream petroleum sector operations, and prepare the way for a possible private participation in the refinery. Following a recommendation by the Bank, the project also would rehabilitate storage and transport infrastructure around the country.  - ii- 4. Within the framework of the dialogue leading to project approval, the government adopted a series of measures to improve petroleum-product pricing and taxation, as well as the procurement of crude oil and refined products. In 1991, the government introduced pricing based on the concept of import-parity, reflecting actual costs, with monthly adjustments by SOLIMA. In 1992, SOLIMA adopted international competitive bidding (ICB) as the method to procure crude oil and refined products. In 1993, shortly before credit negotiations, the government issued Decree 93-002, declaring that the purchase, transportation, storage, transformation, and distribution of all petroleum products were liberalized. 5. The project called for market liberalization and the introduction of competition in the petroleum sector. For this purpose it advocated separation of the refinery from the rest of SOLIMA. After rehabilitation, the refinery would have to compete with imports of refined products. Essential infrastructures would be managed by a joint-venture of SOLIMA and private operators under an "open-access" logistic system. SOLIMA would no longer enjoy the monopoly of distribution either, it would have to compete with new private entrants. In retrospect, this scenario was unrealistic. It was, for example, unlikely that the government, after investing in refinery rehabilitation, would allow the refinery to close in case it could not compete against imports. Most probably the government would have offered protection to the refiner. This strong possibility, and the presence of state-owned SOLIMA in logistics and distribution, discouraged private oil companies from entering the sector even after legal liberalization of petroleum activities was completed in 1995. 6. A number of disbursement conditions, making rehabilitation works dependent on progress in the reform component, became a source of delays, especially when, during project implementation, the Bank changed its position on reform, and began pressing for outright privatization of SOLIMA. Agreement on privatization took a very long time to materialize. In the meantime, disbursement conditions were not being complied with, and execution of infrastructure components was therefore blocked. To complicate things further, the project implementation period was a time of great institutional uncertainty as evidenced by the many government changes the country went through. Between project identification in 1989 and credit closing at the end of 1998, there were eight Ministers of Energy and Mines, and SOLIMA's top management team changed twice. Project Objectives 7. The objectives of the project were: (a) to open up the petroleum sector to private participation and create the framework for a competitive environment; (b) to support basic infrastructure investments so as to increase operational efficiency and attract private operators and investors; and (c) to develop institutional capacity for the efficient regulation of the petroleum sector. To achieve these objectives, the project included the following components: (A) SOLIMA's restructuring and corporatization, including creation of a refining subsidiary; establishment of an open-access system for the use of essential infrastructures; and  - iii - capacity building for regulation. (B) Rehabilitation of the refinery and technical assistance for its management. (C) Rehabilitation and expansion of storage and transport infrastructure. Replacement of pipelines. (D) Construction of a new maritime petroleum terminal at Toamasina. And (E) An LPG promotion program. Implementation Experience and Results 8. The project achieved partially its reform objectives. Reform, in the end, was deeper than originally proposed, with the full privatization of SOLIMA. The project prepared a draft new legal framework, and also a strategy to privatize SOLIMA. The proposed petroleum law was adopted by the National Assembly in March 1999, after credit closing. The regulatory agency is being set up. The privatization of SOLIMA is being implemented, basically following the strategy developed by the project. The project had initially helped prepare a legal framework for liberalization, based on Decree 93-002, followed in 1995 by Decree 565-95, which introduced modem principles of sector operations. In 1996, the government issued an international invitation to private oil companies to enter Madagascar's liberalized petroleum sector, but with no results. The international oil companies expressed concerns about the dominant position SOLIMA would enjoy as a state-owned operator. 9. The project did not achieve its original objectives concerning infrastructure. Refinery rehabilitation was dropped after the government, with some encouragement from the Bank, requested an amendment to the Credit and Project Agreements canceling the component. One reason for abandoning refinery rehabilitation was that the facilities had continued to deteriorate, making economic viability increasingly doubtful. Other rehabilitation works could not be executed due to the disbursement conditions linking them to the institutional component, which was progressing very slowly. Also, once the government agreed to privatize SOLIMA, the idea of public investment to rehabilitate infrastructure lost its justification. With the decision to privatize, modernization of installations was left to be implemented by the new private owners. 10. The project could not complete even the engineering design for the new petroleum terminal at Toamasina. During project preparation, an engineering firm had identified a new site within the harbor, and completed a feasibility study, including an environmental analysis. But an environmental expert later hired to verify the conclusions of the feasibility study rejected the site due to safety and environmental concerns. A second engineering consultant tentatively selected a new site located directly in front of the refinery. This minimized the length of pipelines, but the site was on open seas. A bathymetric survey, and wind, current and wave measurements were carried out. A simulation study based on these measurements showed the new site was also inadequate, because of the rough sea conditions. Concerning LPG promotion, the project financed procurement of 47,000 LPG bottles, but did not execute any of the envisaged investments in expansion and rehabilitation of storage and transport facilities, because of disbursement conditions and the decision to privatize.  - iv - 11. During the period of project execution, the Bank kept pushing for liberalization and for privatization, not only within the framework of the project itself, but also as part of the dialogue for two Structural Adjustment operations and also a private sector development study that was followed by a private sector development project. When the government finally agreed to put SOLIMA on the list of public enterprises slated for privatization, the consultants hired by the project to study SOLIMA's restructuring were asked to prepare a privatization strategy for the company. The financial advisor and other assistance required for the privatization is being financed by the Private Sector Development and Capacity Building Project (Cr. 2956-MAG). SOLIMA was split into several companies which were awarded to private bidders in April 1999. The privatization is expected to be completed by end of July 1999. Summary of findings, future TA and training, and key lessons learned 12. The project was one of the vehicles the Bank used to press a reluctant government to open the sector to private participation and to divest from SOLIMA's installations and operations. This process involved a difficult struggle, particularly since the Bank de facto changed the scope of reform. The reform program as defined in the Credit Agreement did not include the privatization of SOLIMA. Political instability in the country made things more difficult. Progress was very slow, because the government was not interested in privatization. Cancellation of refinery rehabilitation reduced the interest of the project for SOLIMA and the government. The infrastructure components were blocked by disbursement conditions. 13. The lesson that can be drawn from this experience is that the Bank should have either stuck to the original project objectives, showing more patience, or, once it decided to go for privatization, should have drawn the full consequences and proposed a radical restructuring of the project. Following the decision to privatize, most of the investment components should have been cancelled. The project could have instead supported investments required as part of the privatization process, such as environmental cleanup operations. Investments in infrastructure would have been left to the private sector.  IMPLEMENTATION COMPLETION REPORT MADAGASCAR PETROLEUM SECTOR REFORM PROJECT (Credit No. 2538-MAG) PART 1: IMPLEMENTATION ASSESSMENT L. PROJECT BACKGROUND 1. The project concept evolved from the findings of two studies done between 1984 and 1989: an ESMAP energy sector assessment, and a report on the petroleum sector'. The situation found was: (a) substantial deterioration of most petroleum installations and equipment; (b) inefficient import and refining operations; (c) inefficient pricing and taxation2 of petroleum products; and (d) non transparent and inefficient procurement of crude oil and refined products. Both reports paid much attention to the refinery at Toamasina. The refinery has continued to be at the center of the debate and at the center of the difficulties to implement sector reform, even now, at the beginning of 1999, when the state-owned petroleum company SOLIMA is being privatized. This reflects the government's reluctance to give up control of petroleum imports, explained as a concern to guarantee security of supply. 2. The refinery, commissioned in 1966, is a simple hydroskimming facility, with a processing capacity of 450,000 tons per year, which was increased in 1972 to 750,000 tons per year3. Until 1976, the refinery was owned by a group of foreign oil companies and operated by one of them, the French company Elf. Since nationalization in 1976, the refinery has been operated by SOLIMA. In 1982, SOLIMA undertook, at a cost of about $20 million, a major reconfiguration by adding a visbreaking unit, a gas-oil desulfurization unit, and a bitumen unit. Shortly after commissioning, this extension was disabled by an explosion. At the same time, the rest of the installations had deteriorated considerably due to inadequate maintenance, as a result of the strain the heavy investment program had put on SOLIMA's limited foreign exchange. There was a succession of shutdowns caused by operation and maintenance related problems, several involving fires. 3. Because of the large investment that had just been made, the reaction was to identify what additional investments were required in order to solve all the problems. In 1984, the French CCCE (now AFD) financed repairs, and also a feasibility study for the refinery's complete rehabilitation and upgrading. The study, updated in 1986, showed rehabilitation to ESMAP, Jan. 1987, Madagascar: Issues and Options in the Energy Sector, Report No. 5700 MAG. World Bank, July 1989, Madagascar - Petroleum Sector Report, Report No. 7986 MAG. 2 SOLIMA negotiated every year, ex-post, the amount to be paid in taxes to the State. This was necessary to ensure SOLIMA's financial equilibrium, given that the government kept prices too low. 3 Despite this expansion, the refinery's production has never exceeded 450,000 tons per year.  - 2 - be technically and economically feasible, but also that the rate of return depended on a parameter determined by the international market: the "refining margin", or price differential between crude oil and the products obtained from it. This parameter was influenced particularly by large, efficient, export-oriented refineries in the gulf region, and was uncertain. The margin was small, of the order of 5 percent of the cost of feedstock, and this resulted in an extreme sensitivity of the rehabilitation's rate of return to variations in the relative prices of crude oil and refined products. 4. This study, as well as the Bank studies, also noted that in addition to rehabilitation, efficient management was a necessary condition of the refinery's viability. The refinery needed quick reaction capabilities in the procurement of feedstock to take advantage of opportunities in the international oil market, a sizable foreign exchange fund, improved operation and maintenance, etc.. The recommendations were to invest in rehabilitating and provide money for spare parts, to improve operations planning, to train personnel, and to provide technical assistance. There was no mention of the unlikelihood that an efficient performance could be sustained by a state-owned enterprise. The studies on the refinery sought to find the least cost method to supply the country. The underlying assumption was that, in case rehabilitation failed to make the refinery competitive, the consequence would be a higher cost for the country's fuel supplies, not bankruptcy and closure for the refining company. It is very doubtful, in any case, that the government would have let the refinery close after having invested in its rehabilitation. 5. The ESMAP and Petroleum Sector studies recommended more detailed study of rehabilitation feasibility. After the ESMAP study, the Bank informed the government it would not finance refinery rehabilitation, but would not oppose it either. After the petroleum sector study, however, the Bank agreed to finance from a PPF the more complete study that had been recommended. This study showed rehabilitation to have a very high rate of return. The consultants also recommended building a new petroleum terminal at Toamasina, and replacing old, leaky pipelines running between the port and the refinery. Indeed, continued operation of the current installations presents serious public-safety and environmental risks. 6. The Bank agreed to finance a first stage of refinery rehabilitation, as well as the construction of a new petroleum terminal and associated pipelines at Toamasina, subject to the government's agreement to liberalize petroleum-product imports and distribution, as well as some form of private participation in the refinery at a later unspecified date. The project also would finance technical assistance for SOLIMA, particularly a management contract for the refinery to be run by a professional operator for a three-year period during and after rehabilitation. Finally, following a recommendation by the Bank, the project would finance rehabilitation of transport and storage infrastructure around the country. 7. During project preparation, and within the framework of the dialogue with the Bank, the government introduced adequate petroleum-product pricing in 1991, based on the concept of import parity and reflecting actual costs, with automatic adjustments by SOLIMA. The government also rationalized taxation, increasing rates and eliminating differences between products. Tax revenue increased dramatically as a result of these changes. Also, in 1992,  - 3 - SOLIMA adopted ICB as the method to procure crude oil and petroleum products. However, there were from time to time interruptions in the application of price adjustments, in one case following public demonstrations by taxi and truck drivers. Procurement based on ICB has been suspended since mid- 1996, because of the uncertainty of SOLIMA's remaining life following the decision to privatize. Since then, SOLIMA has signed successive addenda to the last contract awarded on the basis of ICB. 8. The project was to be a vehicle to open the downstream petroleum sub-sector to private participation and competition, but it proposed at the same time to maintain and strengthen the state-owned SOLIMA as refiner and distributor. Although liberalization would include freedom to import, the project would finance the rehabilitation of SOLIMA's refinery, thus creating pressures to give the refiner control of imports. To level the playing field, all "essential infrastructures", meaning loading-unloading facilities, large storage facilities, and bulk transport, would be operated under an "open-access system", by SOLIMA, in ajoint-venture with the future private distributors. To help ensure SOLIMA's competitiveness vis-a-vis private distributors, the project would support the company's restructuring and modernization. This model of sector reform proved to be unrealistic. The strong state presence discouraged private oil companies from entering the sector after legal liberalization was completed in 1995. 9. The project included a number of disbursement conditions making rehabilitation works dependent on progress on the reform component. These conditions became a source of delays because after effectiveness the Bank began pressing for outright privatization of SOLIMA. This represented an important change in the Bank's position on reform. The only mention of privatization in the Credit Agreement is of an "eventual private sector participation in the capital" of SOLIMA, an approach chosen after extensive internal debate at the Bank during project preparation. Despite its change of position, the Bank took no action to amend the Credit Agreement on this point. The discussion on privatization was difficult. The government began taking concrete steps to privatize SOLIMA only in 1997. In the meantime, disbursement conditions were not being met, and execution of infrastructure components was therefore blocked. 10. The Board approved the project in July 1993, just before a major political change after the country's first presidential elections in 20 years. The credit became effective only in September 1994, after several extensions of the deadline. Delays were partly the result of the government change, which entailed replacement of staff at the Ministry and SOLIMA. But also, the 90-day period allowed was too short for one of the conditions, later waived: the completion of bidding documents for the rehabilitation of the refinery. At the beginning of 1994, the new government expressed doubts about the project, particularly about refinery rehabilitation. Indeed, the idling parts of the refinery had continued to deteriorate, and the installations had been severely damaged by cyclone Geralda in February of that year. The Bank took advantage of the concerns expressed by the government to propose, even before the credit became effective, the cancellation of the refinery component.  - 4 - II. STATEMENT AND EVALUATION OF OBJECTIVES 11. The project had the following objectives: (a) to open up the petroleum sector to private participation and create an appropriate framework for a competitive environment; (b) to support the basic infrastructure investments required to increase operational efficiency and to attract private operators and investors; and (c) to develop institutional capacity for the efficient regulation of the petroleum sector. 12. In order to achieve its objectives the project included the following components: Part A Sector reform, comprising the following: (i) restructuring and corporatization of SOLIMA, and eventual private sector participation in its capital, including constitution of a subsidiary for the refinery; (ii) establishment of an open access system to key infrastructure; and (iii) establishment of a national regulatory authority to oversee the sector. Part B Rehabilitation of SOLIMA's refinery, and management assistance to SOLIMA's refining subsidiary. Part C Improvement of storage and transport infrastructure comprising: (i) acquisition of 18 rail tankers; (ii) rehabilitation of existing storage tanks and construction of new tanks; (iii) replacement of pipelines from the port of Toamasina to the refinery; (iv) improvement of unloading facilities at Mahajanga and Manakara; and (v) improvement of petroleum unloading operations and of transport and distribution. Part D Construction of a new maritime petroleum terminal at Toamasina. Part E LPG promotion program, including the acquisition of bottles, construction of storage tanks, rehabilitation of filling stations, and acquisition of transport equipment. 13. Although the objectives were clear and adapted to Madagascar's needs, they were undermined by the project components, in the sense that those components would have helped the State to maintain an overbearing presence in the sector. The investment in refinery rehabilitation, for example, would create strong pressures for the government to protect it against competition, thus giving de-facto control of imports to state-owned SOLIMA, and undermining liberalization. 111. ACHIEVEMENT OF OBJECTIVES 14. The project achieved partially the objective of opening the sector to private participation through preparation of a new legal framework. It did not achieve the objective of improving basic infrastructure. Investments to improve infrastructure, however, will be carried out by the new private owners. The project could not contribute to capacity building for regulation, as the regulatory agency could not be set up before credit closing.  -5 - Part A. Sector reform 15. During project preparation, the government had issued Decree N