Turkey - Energy Sector Ajustment Loan (ESAL)

Document of The World Bank FOR OFFCLAL USE ONLY Report No. 13766 PROJECT COMPLETION REPORT TURKEY ENERGY SECTOR ADJUSTMENT LOAN (ESAL) (LOAN 2856-TU) DECEMBER 9, 1994 Industry, Trade and Finance Operations Division Country Department I Europe and Central Asia Regional Office This document has a restricted distribution and may be used bv recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  PROJECT COMPLETION REPORT TURKEY ENERGY SECTOR ADJUSTMENT LOAN (ESAL) (LOAN 2856-TU) LIST OF ABBREVIATIONS BOD - Board of Directors BOT - Build Operate Transfer BOTAS - Turkish Pipeline Corporation CIP - Core Investment Program DSI - State Hydraulic Works EIE - Electric Power Resources Survey Administration ESAL - Energy Sector Adjustment Loan EXIM Bank - Export-Import Bank GDP - Gross Domestic Product GDPA - General Directory for Petroleum Affairs GNP - Gross National Product GOT - Government of Turkey GUS - Gas Utilization Study IRR - Internal Rate of Return LA - Loan Agreement LOP - Letter of Development Policy LRMC - Long-Run Marginal Cost MENR - Ministry of Energy and Natural Resources PR - President's Report PSBR - Public Sector Borrowing Requirement SECALs - Sector Adjustment Loans SOE - State-owned Enterprises SPO - State Planning Organization TEK - Turkiye Electrik Kurumu (Turkish Electricity Authority) TKI - Turkish Coal (Lignite) Enterprises TTK - Turkish Hard Coal Enterprise TPAO - Turkish Petroleum Corporation TRP - TEK Restructuring Project TUPRAS - Turkish Refineries Corporation  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 9, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Turkey Energy Sector Adiustment Loan (Loan 2856-TU) Attached is the Project Completion Report on Turkey - Energy Sector Adjustment Loan (Loan 2856-TU) prepared by the Europe and Central Asia Regional Office. No Part II was contributed by the Borrower. The loan was a hybrid operation which provided $325 million from the Bank and $500 million from Japanese co-financing institutions. The Bank's loan consisted of a $175 million quick-disbursing component in two tranches for general imports, and $150 million project financing for physical investments, studies and technical assistance to eleven energy sector State Economic Enterprises (SEEs). The Japanese co-financing consisted of a $200 million equivalent B-loan for general imports, provided by a consortium of 14 Japanese commercial banks with IBRD and $300 million equivalent from the Export Import Bank of Japan (JEXIM) as joint co-financing for the energy sector investment program eligible for IBRD financing. The second tranche of the quick disbursing component was released about 9 months after Board presentation, but deterioration in the macroeconomic environment led to a subsequent inability to maintain key policy conditionalities. The investment component suffered from considerable procurement delays, and as a result, the project closing date was delayed by three years. This investment component was imprecisely defined in the President's report and legal documents, making the monitoring of the use of the IBRD and JEXIM funds, which were supervised by the Bank, difficult and less than fully successful. The reforms in the energy sector sought by the ESAL, including enhanced private sector financing, energy conservation, improved financial performance of energy enterprises, and a rationalization of the public investment program for energy were not achieved during the implementation of this project. However, the Government of Turkey did deregulate the procurement and marketing of crude oil and linked the pricing of oil product to world markets. In addition, further efforts to strengthen the power sector are being made under the related TEK restructuring project, which is still under implementation. Tariffs have been maintained at or above US cent 6/kWh since February 1990. The majority of the energy sector subprojects financed under the loan were not subject to economic evaluation. The project outcome is rated as unsatisfactory, and its sustainability as unlikely. Institutional development was negligible. The attached PCR is devoted primarily to the policy aspects of the operation. It only superficially covers the almost $450 million of physical investments made by the SEEs under the IBRD and JEXIM loans. An Audit is planned. Attachment 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.   FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT TURKEY ENERGY SECTOR ADJUSTMENT LOAN (ESAL) (LOAN 2856-TU) Table of Contents Page No. Preface ....................................................... i Evaluation Summary . ................................................ ii PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE A. Country Context .....................1....................... Macroeconomic Issues . ........................................ 1 Reforming the SOE Sector . ...................................... 2 B. Energy Sector Adjustment Loan ................................... 3 Origins and Objectives . ........................................ 3 Results of Key Actions under the ESAL .............................. 5 Institutional Reform . .......................................... 6 Investment Policies . .......................................... 8 Energy Conservation and Demand Management ......................... 10 Environmental Aspects ......................................... 13 C. Assessment of the Energy Reform Program . . ......................... 13 1. Energy Policies and Macroeconomic Links .......................... 13 Energy Policy .......................................... 13 Energy Planning ......................................... 15 Import Dependence ....................................... 15 2. Overall Economic Impact ..................................... 16 D. Borrower's and Bank's Performance ................................ 18 Borrower's Performance ........................................ 18 Quality of Bank Supervision ..................................... 19 E. Main Lessons Learned ........................................ 19 PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE 21 PART III: STATISTICAL INFORMATION 22 Annex 1: Note on the Project's Economic Impact 27 This document has a restricted distribution and may be used by recipients only in the performance of their g official duties. Its contents may not otherwise be disclosed without World Bank authorization.l   - i - PROJECT COMPLETION REPORT TURKEY ENERGY SECTOR ADJUSTMENT LOAN (ESAL) (LOAN 2856-TU) PREFACE This is the Project Completion Report (PCR) for the Energy Sector Adjustment Project in Turkey, for which Loan 2856-TU in the amount of US$325 million was approved on June 25, 1987. The loan package included a quick-disbursing component of US$175 million in support of policy reforms and an investment component of US$150 million for financing priority investments in the energy sector. The closing date of the loan expired on September 30, 1993, three years behind the original schedule. Total disbursements reached about US$312.0 million, with US$13.0 million expected to be cancelled as soon as the Special Account balance is fully recovered. The loan will then be closed as of the date of the last disbursement, March 1, 1994. The Project was cofinanced with the Export-Import Bank of Japan, which contributed additional US$300 million for financing priority investments in the sector. Their loan was to be disbursed concurrently with the Bank investment portion of US$135 million at a ratio of 69/31. Their loan was closed on January 31, 1994; about US$274 million equivalent has been disbursed and it is expected that the remainder will be cancelled. In addition, US$200 million of cofinancing to support the implementation of policy reforms was provided by a consortium comprised of the Bank (US$30.1 million) and 14 Japanese commercial banks under a B-loan arrangement, which were fully disbursed at the effectiveness of the B-loan on September 24, 1987. The PCR was prepared by the Industry Trade and Finance Division, Country Department I of the Europe & Central Asia Region (Preface, Evaluation Summary, Parts I and III). Part II of the PCR is being prepared by the Undersecretariat for Treasury and Foreign Trade for the Borrower, Republic of Turkey. Preparation of this PCR is based, inter alia, on the President's Report; the Loan Agreement; the Letter of Development Policy; supervision reports; project progress reports; correspondence between the Bank and the Borrower; and internal Bank memoranda.   PROJECT COMPLETION REPORT TURKEY ENERGY SECTOR ADJUSTMENT LOAN (ESAL) (LOAN 2856-TU) EVALUATION SUMMARY Objectives L. The Energy Sector Adjustment Loan aimed at supporting the Government in realizing its sectoral policy intentions to: (a) improve the efficiency of energy production and consumption in the public sector; and (b) to facilitate private sector participation in the sector. Hence, the ESAL was conceived to be implemented under three interdependent set of policy reforms in the areas of: institutional reform, an investment program, and energy conservation. This support was to be given through a foreign exchange Bank loan of US$325 million. The Project was co-financed by Japan's EXIMBANK and a syndicate of Japanese Banks for US$300 million and US$200 million respectively. The policy actions were to be implemented in two stages and the Loan was to be released in two tranches, the first tranche of US$75 million was to cater for a set of measures underway before Board Presentation which were related to Turkey's energy sector Fifth Five-Year Plan (1985-1989), while the second tranche was to be released on the fulfillment of the set of policy actions. About US$15 million of the Loan was allocated as technical assistance for training and consultancy services for undertaking studies. Implementation Experience ii. The loan became effective one month after Board approval. There were no special conditions of effectiveness and the first tranche of US$75 million was released immediately. Initially supervision was mainly concentrated on procurement, disbursement and physical implementation given the complexity of the investment component. There was a large number of beneficiaries (11 State-owned Enterprise/Institutions) each with their own investment programs, and as a result, the level of goods and services to be procured under Bank guidelines more than doubled from US$150 million to US$450. Cofinancing arrangements with Japan's EXIM Bank were agreed upon to cater for the additional US$300 million, about 2 months after Board Presentation of the Loan. Since most of the beneficiaries were not familiar with standard Bank procurement practices, procurement was delayed and the investment component did not advance as fast as anticipated. This resulted in funds being reallocated among the 11 beneficiaries twice, and the closing date had to be extended three consecutive times, from September 30, 1990, to September 30, 1993. Furthermore, policy-oriented staff was not included on a routine basis in the supervision teams to assess the implementation of the policy reform program. iii. The release of the second tranche was subject to the fulfillment of policy actions concerning institutional reforms; energy investments; and energy conservation measures. In February 1988, when the required conditionality had been fulfilled, the second tranche of US$100 million was released. Subsequently, a supervision mission in April 1988, noted the disruptions caused by a rapidly deteriorating macro-environment. However, once the second tranche was released the Bank had relatively little leverage to ensure that the policy reform  - iii - program provided under the project was fully implemented. Furthermore, a supervision mission in July 1991 reported that key policy conditionalities were not being met. The mission, in consultation with GOT, worked out a set of conditions for the extension of the closing date. This constituted a turning point for the ESAL, and these in particular redefined and strengthened the reform agenda for the power sector. The Loan Results iv. A second reallocation of the furnds from the investment component was carried out in October 1992 with a view to accelerate loan disbursements. Following this second reallocation, the combined share of the major beneficiaries, TPAO (US$35.5 million), TEK (US$30.8 million), TUPRAS (US$24.7 million), TTK (US$21.1 million), and TKI (US$14 million) represented about 93 % of this component. Similarly, as regards the technical assistance component, TEK, GDPA and EIE's allocation accounted for about 79% of the funds. The loan was 95% disbursed by end-1993 and the bulk of the investment program was concluded. The preparation of subsector plans for power, lignite, natural gas and refineries was completed. GOT deregulated the procurement and marketing of crude oil and the pricing of oil products by linking these prices to international levels. The hydrocracker refinery was commissioned and TUPRAS is now under a privatization schedule. A new tariff structure was implemented for the power subsector, and the energy conservation program was partially implemented. The first stage of the study on the Regulatory Framework for the power sector helped TEK's institutional reform and paved the way for the utility's privatization presently underway. v. The general objectives of the ESAL were only partially achieved as witnessed by the persistent weaknesses in the sector's policy framework. The major SOEs in the mining sector (TKI, TTK, TPAO) have seen their revenue base contract in real terms over the 1987-93 period raising concerns about the financial and/or economic viability of some of their operations. Private involvement did not materialize to any significant extent; progress in improving the efficiency of energy production was not forthcoming given the almost intractable character of SOE's inefficiencies, and growing domestic inflation made havoc of pricing arrangements and of investment and financing plans. Sustainability vi. Macro economic instability and the Government's attempts at stabilization seriously affected the reform program in 1987-89. This forced the adoption of a new tariff policy under the TEK Restructuring Project (TRP) which improved significantly TEK's financial performance. Nevertheless, Turkey's persistent economic instability raises doubts about the sustainability of such improvement. Although in 1993 the generation, transmission and distribution parts of TEK were separated into two joint stock companies, unless the sector is restructured to promote a competitive electricity industry and TEK privatized, in addition to continued operational inefficiencies, such instability could regress into highly subsidized power tariffs, resulting in a misallocation of resources. vii. Private sector participation is critically needed to help mobilize massive amount of resources for Turkey to expand its generation capacity to be able to keep up with the rapid pace of electricity demand, estimated at US$14 billion in 1994 prices during 1995-1999. A deep fiscal crisis will prevent the public sector from providing these resources, as was the case during  - iv - ESAL implementation. Likewise, government commitment needs to be stronger and communicated clearly at all levels as it is vital to implement the reforms which are still required to be able to attract such amounts of capital. The establishment of an appropriate regulatory and incentive framework for the private sector is still pending. Main Findings viii. The impact of ESAL's policy reforms on the economy will take time to evolve. From a long-term perspective, the ESAL paved the way for: (a) further specific institutional and policy reforms which would contribute to increase the efficiency of energy production; and (b) participation of the private sector through changing the market structure toward a more competitive market-oriented power subsector. ix. However, sector planning deficiencies are still apparent: lack of economic pricing in energy sector planning; a cumbersome investment selection process subject to political interference; lack of realism in assessing project implementation capacity; and weak environmental assessment capability. The collaboration of the Bank to achieve greater economic efficiency in the sector is still needed. Inadequate implementation of the energy conservation program involving major industrial consumers of energy increased the need for power capacity. x. Despite the existence of legislation to attract the private sector, it was ineffective due to TEK's vertically integrated monopoly and the inadequate financial and security arrangements. Both constituted a formidable barrier not only for the participation of the private sector but also to improve the efficiency of Turkey's power sector. xi. TEK's investment in the high voltage transmission network was partially completed and the remainder is being financed under the TRP. This is a high priority investment which will permit the utilization of considerable hydropower generation recently commissioned (2400 M). Economic gains here are estimated to be substantial. xii. During implementation, several weaknesses in project design became evident: (a) no economic criteria were systematically employed for the allocation of investment funds to the beneficiaries; (b) TEK, by far the largest SOE in the sector, was not adequately analyzed during appraisal; (c) insufficient analysis was undertaken of the macro linkages as well as of the possible impact of economic instability on investments, SOE financial performance and pricing policies; (d) a large investment component confronted a large number of inexperienced beneficiaries (11) with Bank procurement guidelines; (e) absence of comprehensive sector conditionalities for the release of the first tranche; and (f) the conditionality to release the second tranche did not encompass the complex policy issues of the sector, nor was the macroeconomic environment properly assessed. xiii. The policy framework and investments launched contributed to reduce the energy sector's dependence on imported oil in favor of domestic lignite and hydropower. On the other hand, the Govemment's new environmental stance manifested by the greater use of imported natural gas has led to increased dependence on imported energy, highlighting the trade offs between Turkey's extemal trade and environmental policies.  - v xiv. TEK's vertically integrated monopoly has compounded the inefficiencies of the centralized power system planning setup. Nevertheless, ESAL with the TRP have paved the way for further institutional reform. The financial performance of TEK has improved and its burden on the PSBR reduced in 1992-93, and the Government is now fully committed to its privatization through a Law enacted by Parliament in February 1994. The bout of economic instability in early 1994 will severely test the utility's financial position, however, as well as that of TUPRAS, whose prices have been temporarily frozen under the ongoing stabilization program. xv. The Borrower's supervision was weak. Treasury, the agency in charge of coordinating the operation, in supervising the beneficiaries' performance, and remitting to the Bank the audit and quarterly reports, had limited involvement. During the implementation period only two progress reports were received. During the initial two years the delays in procurement were significant, and TEK's and TTK's performance in procurement unsatisfactory. xvi. These complicated the work of the Bank which had to concentrate on procurement to facilitate disbursements and the implementation of the investment component, on top of the urgent need to organize the execution of the policy reforms. Bank's supervision efforts had mixed results: they were instrumental in helping the Borrower implement the investment component, but were far less effective in monitoring the policy component. The lack of continuity affected seriously project implementation, more on the Bank's than the Borrower's side, the Bank having had 6 task managers during project implementation. Main Lessons Learned xvii. The major lesson is that it is premature to launch an energy sector adjustment program when the macroeconomic environment is turning around and Government's commitment to the reform program is ambiguous. The Bank's quick disbursing funds aimed at supporting the policy based component were used to finance critical energy imports at a time Turkey was suffering from foreign exchange shortages. This was done without Turkey being able to reach an internal consensus on the overall reform program, and the Bank should have delayed the Loan until such time. Furthermore, the Bank did not fully assess the macroeconomic risks, probably because Turkey was still being perceived as one of the star performers in the developing world, and because it was unwilling to factor in important political economy aspects. The Bank could have called upon outside expertise, including the top international credit rating agencies, to validate its own macroeconomic stance. xviii. Another major lesson refers to the overly ambitious project design which involved a multitude of recipient institutions against the limited local absorptive capacity and a complex program of policy and institutional reforms. The number of institutions involved should be reduced to a few core ones and the policy framework streamlined. The absence of a clearly agreed core investment program and of a monitoring mechanism for each SOE was very damaging to project implementation. Supervision should focus on the SOEs' core investment programs as influence on policy can be exercised most constructively via this mechanism. The policy agenda should be the main responsibility of the SOEs (rather than the Government) with which the dialogue is generally more decentralized, fluid and sustainable. Militate against, it has to be noted, the relative lack of autonomy of SOEs and their vulnerability to the political cycle. In any event, to this effect greater priority should have been given to sector work without  - vi - which the Bank's ability to provide sound policy advice suffers, including, as was the case under ESAL, after the Loan has been approved. xix. The Loan should have been closed earlier and not dragged out for three years without strategic direction. It was only because a new investment operation was prepared and eventually approved by the Bank that it was possible to carry out some basic reform in the power sector (TRP). ESAL should, however, have been able to stand alone. Bank lending could instead have helped Turkey mobilize private sector resources more aggressively, which are urgently needed for this country to be able to cope with the rapid pace of energy demand, and to help arrest environmental degradation from the use of fossil fuels.   PROJECT COMPLETION REPORT TURKEY ENERGY SECTOR ADJUSTMENT LOAN (ESAL (LOAN 2856-TU) PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE Project Identity Name: Energy Sector Adjustment Loan (ESAL) Loan Number: 2856-TU RVP Unit: Europe and Central Asia Country: Turkey Sector: Energy A. COUNTRY CONTEXT Macroeconomic Issues 1. Following a serious external crisis in the late 1970s, Turkey embarked on a far- reaching program of stabilization and reform marking a major break with past policies which had favored an inward-looking and state-dominated pattern of development. Considerable progress was made in reducing macroeconomic imbalances and in liberalizing the external sector during the first half of the 1980s. GNP growth averaged about 5 percent per year, inflation was reduced from three-digit levels to 25 percent, and exports as a share of GNP increased threefold, substantially improving Turkey's creditworthiness. After 1986, there was further progress in integrating Turkey into the world economy. Import restrictions were eased, export subsidies rationalized and controls on capital transactions reduced substantially. The process culminated in a reform of the exchange rate system which left it largely free of restrictions on external transactions. Turkey reaped significant benefits from these improvements in the external policy regime. Export growth remained strong and, aided by receipts from tourism and workers' remittances, kept the external current account deficit at comfortable levels despite the rapid growth of domestic demand. Foreign investment flows also surged, and Turkey was able to diversify its sources of foreign financing while repaying the last of its IMF and rescheduled commercial debt. 2. These achievements have been undermined in recent years by a resurgence of public sector imbalances. The public sector borrowing requirement increased from 6.5 percent of GNP in 1988 to almost 17 percent in 1991. GNP growth slowed sharply and inflation accelerated to an annual rate of 66 percent. Despite growing recognition of the fiscal problem, the public sector borrowing requirement (PSBR) has stayed high, averaging an estimated 16 percent of GNP in 1992-93. Although there has been a recovery of growth during these years, inflation remained above 60 percent and, more recently, the external position has begun to suffer  as well. The external current account deficit increased from US$900 million in 1992 to an estimated US$5.5-6 billion in 1993. The increased vulnerability of the Turkish economy has caused its creditworthiness to come under renewed scrutiny. The persistent macroeconomic difficulties will also complicate Turkey's planned accession to the European Customs Union in 1995. 3. At the root of Turkey's financial difficulties is the persistently large public sector deficit. In recent years, this deficit has been driven primarily by four factors: (a) the poor performance of the state-owned enterprise (SOE) sector--the consolidated operating balance of the sector fell from a surplus equivalent to 3.6 percent of GNP in 1985 to a deficit averaging over 5 percent of GNP in 1991-92; (b) rapidly growing interest payments, fueled both by the rapid growth of debt and growing macroeconomic uncertainties; (c) large wage increases for public sector workers; and (d) costly agricultural support policies. The magnitude of Turkey's fiscal imbalances and the structural nature of the underlying factors indicate that a comprehensive medium term program of stabilization and reform measures is required to restore macroeconomic balance and provide the basis for sustainable growth. In particular, any effort to bring about a sustainable reduction in the PSBR has to include fundamental reform and privatization of the SOEs. 4. The cyclical behavior of the Turkish economy during the Post War period was again very recently made painfully evident. The inadequacy of the macroeconomic framework led to a severe currency crisis during the early months of 1994. The market value of the lira depreciated by about 50 percent against the major currencies during the first quarter of 1994. In an attempt to stem the depreciation of the lira, the central bank intervened massively in both foreign exchange and money markets. The foreign exchange reserves of the central bank declined by over US$3 billion during the first quarter of 1994, while overnight money market rates averaged about 500 percent. There have been successive downgrading of Turkey's credit rating by the international rating agencies, and access to external financing has been constrained severely. These developments compelled the Treasury to resume its recourse to central bank financing, further exacerbating excess liquidity and inflationary expectations. On April 5, 1994, the Government unveiled its long-awaited adjustment program. The primary focus of the "April 5 Program" is on stabilizing the economy. However, it aims also to strengthen the country's growth prospects and competitiveness by introducing structural reforms in several key areas. Reforming the SOE Sector 5. The process of economic liberalization during the 1980s exposed many of the weaknesses of Turkey's SOE sector, which includes major energy SOEs (TEK, TKI, TTK, TUPRAS, TPAO, BOTAS). Originally established within an economic planning framework for the purpose of industrial development and the provision of infrastructure, few of these enterprises, which account for about 10 percent of GDP, are equipped to face the challenges of a dynamic market economy. The reform agenda for Turkey's SOE sector needs to focus on improving productive and allocative efficiency. Divestiture (sales and liquidation) must be brought to the forefront as the primary instrument for achieving this objective. Other,  - 3 - complementary elements of the reform agenda include labor restructuring, commercialization of SOEs to be retained in Government hands and the strengthening and rationalization of the Govemment's regulatory role. 6. Past efforts to reform the SOE sector through legal and institutional reforms as well as privatization were extremely limited in scope, with correspondingly meager results. This has most recently included the Bank financed TEK Restructuring Project approved in May 1991 with mixed results to date. Between 1986 and 1991, a few small SOEs and minority shareholdings were sold, representing less than 1 percent of SOE fixed assets. The overall pace of privatization was constrained by a lack of political commitment, unclear privatization objectives, inadequate guidelines on buyers, sale instruments and valuation, and the shallowness of the Turkish capital market. The pace of SOE privatization accelerated in 1991, with the sale of a few cement plants and other smaller SOEs, but for the most part privatization remained confined to the sale of minority shareholdings as means of raising revenues. 7. Since 1992, there has been a significant acceleration in the pace and scope of SOE privatization. A number of sizeable transactions are on their way to completion and a substantial pipeline of SOEs has been identified for privatization over the next five years, including major energy sector SOEs such as TEK and TUPRAS. Building the necessary momentum into the process will require that the Government promote the growing public awareness of the need to downsize the SOE sector, forge a political consensus, and broaden popular support for privatization; strengthen the legal framework as well as the capacity of implementing agencies; provide an effective social safety net for displaced employees; and assist with the recovery of the regions likely to be affected by concentrated layoffs. 8. The Privatization Implementation Assistance Project which was approved on May 3, 1994 will support the implementation of the divestiture program, and help accelerate and assure early privatization successes. The project will help develop an effective institutional framework for privatization, strengthen the capacity of the implementing agencies, and provide a social safety net for affected employees. In so doing, the project will lay the foundation for a more comprehensive and far-reaching privatization program which will contribute to increased efficiency and productivity in the economy, sustained fiscal adjustment, and the further development of Turkey's dynamic private sector. In parallel, the ongoing TEK Restructuring Project is being recast with a view to establishing a competitive and privatized electricity industry in Turkey. The onset of macroeconomic instability in 1994 might, however, constrain the pace of privatization in the short-term (para 4). B. ENERGY SECTOR ADJUSTMENT LOAN Origins and Objectives 9. The international oil price increases of the mid-1970s precipitated a dramatic change in energy policy in Turkey. While domestic oil production was falling significantly during the 1970s, oil imports, which were only US$200 million in 1970, jumped to US$3.5  - 4 - billion in each of the years from 1979 to 1984, contributing heavily to a serious external crisis (para 1). The response by Turkish policy makers was to curtail, through substitution, the level of energy imports by embarking on a massive program of indigenous resource development, especially lignite and hydropower, and also oil, gas and geothermal resources. Nevertheless, by the time of the second oil shock, Turkey's dependence on imported oil was still growing. Of the country's net primary consumption of about 32 million toe in 1980, petroleum accounted for as much as 60%, compared to only 20% in 1960. Electric power accounted in 1980 for 23% of all commercial energy used 5% of which was imported from Eastern Europe; the industrial sector consumed then 61 % of all electricity produced. Lignite accounted in 1980 for only 15% of commercial energy consumed, almost a third of which went for power generation. Non-commercial energy (primarily fuelwood, but also other biomass) represented in 1980 only one fifth of the energy used in the economy as a whole, but was the primary energy source for the household sector, accounting for about 57% of its total final consumption, which helped to cushion the population from the oil shocks. 10. Energy investment showed a dramatic increase in the post 1975 period and by 1984 had peaked at about 40% of total public fixed investment and 3.0% of GDP. By the mid- 1980s, however, Turkey's energy strategy was seen as having failed to meet its ambitious targets. Various Bank reports in the early 1980s had concluded that "... resources, human and financial, have been spread too thinly over too many projects in the investment program", and this resulted in "... major supply interruptions, project delays, frequent and protracted plant breakdowns and unacceptably high losses". Furthermore the policy of import substitution by indigenous resources entailed high risks since shortfalls in electricity supply could not be fully met, for technical reasons related to system stability, by electricity imports. This lack of flexibility had caused, by the early 1980s, a substantial energy deficit. The sector's problems had reached crisis proportions - problems that were deep rooted, and almost all linked to inefficient practices of the public sector energy agencies in planning, project implementation, financing and operations. The public sector had, since the end of the 1970s, an almost total monopoly in energy supply. The lignite sector had been nationalized in the early 1970s and statutory barriers to private electricity investment assured TEK's monopoly. 11. In response to these problems, the Government adopted a set of policy initiatives aimed at adjustment in the sector over the medium term. The Government assigned highest priority to the energy sector and to investments that would reduce the energy deficit and costs of supply. The strategy for the development of the sector was developed during the preparation of the Fifth Five Year Plan (1985-89). In line with the priorities of the structural adjustment program (import liberalization, investment rationalization and an enhanced role for the private sector) the Development Plan identified the key areas to be addressed in the energy sector, as part of a sector policy framework to be supported by the Bank under ESAL: (a) continued priority would be given to domestic sources of energy, especially hydro and lignite, when economically justified;  - 5 - (b) the economic justification of projects based on imported energy, especially gas and coal, would be subject to close scrutiny; (c) private sector financing, both local and foreign, would be sought for participation in energy development; (d) conservation through indirect (pricing) as well as direct (non-pricing) measures would be supported; (e) energy SEEs would be given greater financial autonomy especially in setting output prices; and (f) the public investment program for energy would be rationalized by focusing on priority projects and priority subsectors. 12. By 1986, the initial achievements of the new energy policy were already apparent. The rapid expansion of installed power generation capacity and lignite production, in conjunction with a successful economic stabilization and structural adjustment, and a significant fall in the world market price of oil, had beneficial impact on the economy. Turkey's import substitution program and aggressive energy pricing policies allowed it to overcome the energy shortage which had plagued the economy in the early 1980s (para 10). The first half of the 1980s was characterized by a rising trend in energy prices in real terms for most sectors (the exceptions being lignite prices to the electricity sector and electricity prices to the household sector). Moreover, sector planning, particularly medium and long-term energy demand and supply forecasting, improved substantially with the adoption of sophisticated computer models. Most critically, economic deregulation was initiated which facilitated private sector investment in power generation under the Build-Operate-Transfer (BOT) arrangement. Results of Key Actions under the ESAL 13. Against this backdrop, the ESAL was appraised in February 1987. Negotiations took place in Washington from April 30 to May 15, 1987 and became effective on July 7, 1987. Closing date was set at September 30, 1990 and was extended three consecutive years to September 30, 1993, mainly due to delays in procurement and physical implementation by the 11 beneficiaries of the investment program. 14. The Energy Sector Adjustment Loan aimed at supporting the Government in realizing its sectoral policy intentions to: (a) improve the efficiency of energy production and consumption in the public sector; and (b) to facilitate private sector participation in the sector. Hence, the ESAL was conceived to be implemented under three interdependent set of policy reforms in the areas of institutional reform, an investment program, and energy conservation. 15. The loan amount was US$325 million, US$175 million to finance general imports in support of policy reform and the remaining US$150 million to cover the foreign exchange cost  of equipment, material, and services needed for the implementation of investments. Project implementation was organized in two tranches. The first tranche of US$75 million together with the US$200 million in cofinancing from Japanese commercial banks (B-Loan) was released and disbursed right after loan effectiveness and was not contingent on other than standard Bank conditions. The following US$100 million second tranche was released in February 1988. Additional cofinancing of US$300 million from the Export-Import Bank of Japan (JEXIM) was agreed for financing the investment component. 16. The loan was 95% disbursed by end-1993. The investment program was concluded. The preparation of subsector plans for power, lignite, natural gas and refineries was completed. The GOT deregulated the procurement and marketing of crude oil and the pricing of oil products by linking these prices to international levels. The hydrocracker to be installed at the Karikale refinery was commissioned and TUPRAS is now under a privatization schedule. A new tariff structure was implemented for the power subsector, and the energy conservation program was partially implemented. The first stage of the study on the Regulatory Framework for the power sector helped TEK's institutional reform and facilitated the utility's privatization. 17. The general objectives of the ESAL were only partially achieved, however, given the persistence of deeply rooted institutional and policy weaknesses in the sector. Private involvement did not materialize to any significant extent; progress in improving the efficiency of energy production was not forthcoming given the almost intractable character of SOE's inefficiencies, and growing domestic inflation made havoc of pricing policies and of investment and financing plans. 18. Key provisions of the Loan Agreement (LA) concerned the release of the second tranche for policy action. These are described in detail below against the backdrop of more recent developments in the three main reform areas: First, Institutional Reform (paras 18-23), second, Investment Policies (paras 24-30) and third, Conservation Policies (paras 31-39) and their compliance assessed (Schedule 5, of the LA: (a), (b), (c), (d), (e), (f), and (g); and LA covenants 4.03 (a), (b), (c), and 4.04 (a), (b), and (c)). Section C (paras 40-53) includes an overall Assessment of the Energy Reform Program agreed under the Government's "Statement of Energy Policy" of May 22, 1987. Institutional Reform (a) "Carrying out the reform program of the regulatory body of MENR, including the development of computerized procedures for pre-feasibility screening and appraisal of projects; and introduction of procedures for analyzing, leasing and operating agreements." (Schedule 1, para 4(b) of LA, Schedule 5(a)). This covenant was partially met. Plans were spelled out to prepare the power subsector for privatization, following the agreements with the Government and TEK in September 1991 to monitor TEK's financial performance and auditing requirements agreed under ESAL to the TEK Restructuring Project (para 33).  - 7 - 19. The "regulatory body" was to be created within MENR to implement an ambitious program for restructuring the sector and mobilizing private sector investments in infrastructure, mainly for the generation of electricity. A Regulatory Framework Study for the power subsector was supposed to propose regulations to enforce the intent of the set of existing parent laws. However, the study suffered delays due to MENR's inability to build an intemal consensus on clearly a highly sensitive matter. 20. In July 1993, a Bank supervision mission substantiated that TEK's vertically integrated monopoly (which comprises the generation, transmission and distribution of electricity in Turkey), and the absence of competition constituted a formidable barrier not only for the participation of the private sector but also to improve the efficiency of Turkey's energy sector.' Major constraints to reforming TEK have been: the lack of continuity of top management; lack of consensus within TEK's middle level management; and lack of a clear government commitment to the autonomy of its management. This, in conjunction with inadequate security and financial arrangements, explains why in spite of the existence of a complex set of laws designed to facilitate the participation of the private sector, private participation failed to materialize to any important extent under ESAL2. It was not only that the existing regulations were inadequate but also that the MENR was initially uncooperative or unable to set up an independent Regulatory Body to unbundle the highly monopolized electricity industry. 21. By mid-1992, MENR resumed the Study on the Regulatory Framework which first stage was concluded only in September 1993. The recommendations of the Study helped brealdng up TEK's vertically integrated monopoly into two separate joint stock companies: a generation and transmission company, and a holding of distribution enterprises. The process of TEK's privatization was launched on the basis of Government Decree-Law of September 15,1993, and a Law enacted by Parliament in February 1994. Most recently, however, the Constitutional Court has suspended this legislation raising considerable uncertainty about the scope and pace of privatization in the power sector. 22. Although a step in the right direction, the recent breaking up of TEK into two companies does not yet ensure a level playing field in the power sector. Because of this, after the passage of the above legislation (para 18), in late 1993 agreement was reached in principle with the Government on the need to overcome the hypercentralized nature of the power subsector by "unbundling" further TEK into: (i) a set of independent generation companies; (ii) an independent transmission company acting as a common carrier allowing the generation 1/ The major issue is the need for mobilizing massive amount of resources for Turkey to expand its generation capacity to be able to keep up with rapid pace of electricity demand, estimated at US$14 billion in 1994 prices during 1995-1999. The public sector is in no position to provide most of the resources as was the case in the past given the seriousness and ramifications of the current fiscal crisis (para 3). 2/ Decrees No. 233 to reorganize state energy enterprises (SEE), Law No.3291 to give up the monopoly of SEE, and Laws No.3154, 3096, 3613, which deal with the electric sector.  companies to have direct access to electricity user and compete among themselves to satisfy market demand; and (iii) independent distribution companies. The natural monopolies, namely, the distribution companies and the transmission/load dispatching company, would be subject to "hands-off" regulatory oversight. This agreement has been most recently put into question by MENR and work on the second stage of the Study has been temporarily suspended (para 18). (b) "Carrying out an independent financial audit of TEK's records and accounts starting with its 1988 fiscal year." (Schedule 1, para 4(b) of LA; Schedule 5(b)). This condition was met, although the international auditing company has been so far unable to issue an opinion on TEK's accounts. The audits highlighted the deficiencies in the utility's financial management practices and this allowed the Bank and TEK to agree on a comprehensive financial management improvement program (FMIP) currently being carried out under the TEK Restructuring Project with assistance from a team of Deloitte, Touche, Tomatsu (DRT). Results so far have been disappointing, however, as there exists disagreements within the utility as to how to proceed with the FMIP. To this day the utility cannot assess properly operating costs per plant, nor the costs of transmission, or that of its 72 distribution enterprises. (c) "Strengthening TEK's Board of Directors (BOD) to enable it to assume its role as an effective policy making body and overseer of TEK management." (Schedule 1, para 4(b) of LA; Schedule 5(c)). This covenant was partially met. Although outside appointments were made to reinforce the BOD, the results have been mixed and the utility still lacks a corporate strategy. 23. Sections (b) and (c) above were designed to help TEK to achieve an acceptable degree of financial credibility in order to mobilize resources through its access to capital markets including its association with private investors. The strengthening of the Board of Directors with an outward-looking policy-making body was expected to streamline TEK's organizational, management, financial structure and its capacity to set broad policy through the appointing senior government staff with experience in both the private and public sector. In fact, three outsiders were appointed to TEK's BOD from SPO, MENR, and the Treasury. TEK has not been able, however, to develop a corporate strategy and its management is still heavily dependent on Government decisions, including for tapping the capital markets. Investment Policies (d) "Preparation of subsector plans for electric power, natural gas and refineries" (Schedule 1, para 4(b) of LA; Schedule 5(d)). This condition was fully met by September 1993. Economic efficiency criteria, however, were not systematically adopted in the design of the investment component and subsequent loan reallocations among the institutions involved.  24. Supervision was instrumental in assuring the execution of the proper resource allocation policy for the investment programs to ensure that indigenous resources were exploited according to strict economic principles. During the supervision conducted for the second extension of the closing date, following GOT's request of a major reallocation of loan proceeds among the beneficiaries, it was agreed that regarding new projects (under the 1991 program) the economic analysis together with the financial and technical justification would be the prerequisite for the use of funds. Previously, no strict economic efficiency criteria were employed for such allocation, particularly critical in respect to TTK, the hard coal SOE. 25. There were no feasibility studies to examine the economic viability and efficiency of TTK's coal mines. Additionally, since TTK was not fully appraised, the equipment needs of each of the mines was not known a priori, and therefore, it was not possible to determine which of the equipment procured would be going to which mine. This was clarified during the project implementation. Nevertheless, one cannot establish the linkage between investments made via the procurement of equipment, such as locomotives for a particular mine, and the associated increase in productivity of that mine. Some TTK mines are so inefficient that it would have made economic sense to shut them down rather than to keep them operating. In fact, coal produced from such mines costs twice as much when compared to import parity at the 1991 exchange rate. 26. Assistance to TTK for mine rehabilitation was therefore, given because it was believed that investments on the modernization of the mines would increase efficiency and thus help reduce the size of subsidies that the GOT was providing to the mines every year in order to keep them in operation since these mines provided employment to fifty thousand people. 27. The oil refineries modification study was completed. An updated Petroleum Product Supply Option study was also concluded in order to firm up the medium term investment program for the petroleum subsector. Construction of the hydrocracker for upgrading TUPRAS refining capacity toward higher value added petroleum products started in 1990 and was concluded before project completion. Also, was concluded the pipeline engineering study a precursor of an overall gas utilization study. The Gas Utilization Study (GUS) to optimize gas development and address complex supply and demand issues was never launched, however, as BOTAS management decided not to conclude the bidding process and select consultants in spite the recognition of the Study's importance by the company's own technical staff. 28. TEK formulated a long-term least cost plan for the period 1993-2010 covering generation, transmission and distribution. This was undertaken with the assistance provided by the International Atomic Energy Agency under financing by the UNDP. A revised least cost program was also agreed under the TRP in 1991, which also showed an adequate internal rate of return for Turkey's long term power investment program. In July 1992, agreements were reached with TEK to integrate the least-cost approach to power system planning with cost benefit analysis techniques applied to ranking investments from an environmental standpoint.  - 10 - 29. It is expected that in the long-run, a more market oriented, decentralized, and transparent approach would permit more realistic investment plans based on cost benefit analysis techniques and procedures. In the meantime, energy sector planning in Turkey is still overly centralized and weak in critical areas of policy analysis and formulation. Deterministic long- term planning methods at the Government level giving limited consideration to pricing, risk and uncertainty are exacerbated by TEK's, until recently, vertically integrated monopoly, likely to persist under the new transmission and generation company (paras 20 and 22). 30. In short, the core medium term energy investment program for the Fifth five-year Plan (1985-89) was completed during the ESAL implementation. Nevertheless, the rate of investments proceeded very slowly, as reflected in the extension of the closing date for three consecutive years due mainly to delays in procurement and physical implementation (para 13). The need to accelerate disbursements led to successive reallocation of funds which in certain instances were not fully supported by in-depth economic analysis. Energy Conservation and Demand Management (e) "Introduction of a new oil refinery formula designed to provide economic incentive to optimize operating and investment decisions by linking product prices to their international market value. " (Schedule 1, para 4(b) of LA; Schedule 5(e)). This policy covenant was initially met, indeed, the reforms went beyond the initial agreements. Most recently, there has been some backsliding as the Government has reinstated some degree of control on ex-refinery prices. 31. Prior to 1989, retail prices for petroleum were set by the government and frequently did not adjust to cost variations caused by inflation, exchange rate devaluations, and changes in prices of other energy substitutes. The GOT had the monopoly to import oil and market oil derivatives. The serious misallocation of resources from this policy was underscored by a Bank supervision in April 1988. In 1989, the GOT in a bold measure, deregulated the procurement of crude oil as well as the pricing of oil products at the ex-refinery and retail levels by linking these prices, as a minimum, to their import prices. Subsequently, imports and marketing of oil derivatives were liberalized and in 1993 TUPRAS (refineries) and Petrol Offisi (the distributors) were slated for privatization. More recently, under the 1994 stabilization program, however, ex-refinery and retail prices have been again subject to Government controls and the frequency of adjustment thus reduced, raising concerns about the financial position of these SOEs. (f) "Introduction of a new electricity tariff structure on the basis of voltage level in order to simplify the tariff structure and the introduction of time-of-day pricing." (Schedule 1, para 4(b) of LA; Schedule 5(f). This condition was met. Initially, the policy fulfilled its conservation and economic efficiency objectives, but a deterioration in the macroeconomic environment caused a significant erosion of the tariff in real terms by 1989-90. Only when the tariff was set in US cents and  - 11 - adjusted periodically under the TRP since 1991 was this erosion reversed through 1993. 32. On December 1987, three months after loan effectiveness, and right after the national elections, tariffs were raised by 36% to reflect voltage level usage and time-of-day pricing for industrial consumers. Moreover, taxes on tariffs were reduced from 3. 1% to 1%. TEK set in motion a promotional campaign to encourage industries to maximize their use of the new time-of-day tariff which represented a further step towards more rational use of electricity. 33. However, the new tariff policy proved to be ineffective as tariffs declined significantly in real terms in the wake of increasing domestic inflation, which, in turn, significantly deteriorated TEK's financial performance and led to non-compliance of the ESAL financial covenants. One major design problem was that the financial performance criteria did not provide enough incentives for compliance as it combined two different sector entities with separate mandates and management, namely, DSI (the State Hydraulic Works), responsible for the construction of hydropower plants, and TEK, responsible for building thermal power plants (both entities were being asked to comply jointly with a sector-based internal cash generation criteria based on the sum of their respective investments), rather than defining separate performance criteria for each organization to make them individually accountable. As a result, in September 1991, in connection with the second extension of the closing date of ESAL it was agreed that the then recently negotiated agreements under the TEK Restructuring Project would be used to monitor ESAL's covenants in regard to TEK's financial performance and auditing requirements as these were designed to take into account the radically new set of circumstances. This arrangement was construed to be temporary until such time the ESAL financial covenants/requirements could be met. To this end, it became crucial to overhaul again the existing tariff policy. 34. Under the TRP it was agreed to maintain net average tariffs in real terms at US$0.06 per Kwh net of taxes, to be adjusted periodically (monthly as was the case through December 1993).3 As regards energy conservation policies the actions to improve TEK's financial performance were essential to establishing the efficient electricity tariff. LA covenants 4.03 and 4.04 for ESAL dealt with this objective but were not being met by TEK. These 3/ The new tariff covered LRMCs which, in turn, was consistent with international utility comparators. This interim arrangement was used in lieu of explicit indexation which was perceived as feeding domestic inflationary expectations. Also, to support the tariff policy, major agreements were reached with TEK to improve its financial performance, financial management, accounting and auditing, TEK's medium term Core Investment Program (CIP) and its financing; the procurement process was streamlined; a comprehensive commercialization agenda was established, which among other things, intended to identify TEK's operating costs and devise a program to increase operational efficiency. An annual Corporate Performance Plan was also introduced to improve corporate planning. The new corporate approach, away from the original sectoral approach to financial performance under ESAL, also made TEK's management less vulnerable to Government interference and in principle more accountable.  - 12 - modifications not only prevented TEK's financial collapse but also permitted compliance with the new financial covenants under the TRP in 1993. TEK's 1994 Corporate Performance Plan (December 1993) shows that the internal cash generation ratio, current ratio, and average A/R ratio improved dramatically in 1992/93 over 1990/91. Two key measures were instrumental: (a) the above new tariff policy which has led to tariffs maintaining their real levels; and (b) a rationalization and reduction of the volume of TEK's investments. (g) "Carrying out an energy conservation program including the preparation of energy audits of industrial plants for potential investments to improve energy efficiency." (Schedule 1, para 4(b) of LA; Schedule 5(g)). This covenant was met, although, it did not lead to any major retrofitting because of the absence of an adequate incentive and institutional framework for energy conservation. 35. The GOT entrusted the responsibility for a national conservation program to EIE, the Electricity Survey Institute. EIE's Conservation Department gathered energy consumption data on the industrial sector and on the basis of cost/benefit analysis identified 21 industries which were audited with a view to identifying investments and policy changes needed to improve energy efficiency and substitute, where economically feasible, for lower value fuels and retrofitting. These industries cover the six largest industrial consumers of energy: steel, cement, glass, fertilizers, pulp and paper, and textiles. 36. Although the studies showed the economic feasibility of retrofitting 21 industries only four were actually carried out and in SOEs alone. In the absence of a law attributing to EIE responsibility for conservation activities, it had little authority over this matter and thus assigned relatively low priority to this activity. Potential private sector borrowers also manifested their unwillingness to bear the foreign exchange risks and go through the Bank's cumbersome procurement process. As a result, in mid-1991, the Bank agreed to a reduction of EIE's loan proceeds from US$27.5 million to US$6.75 million. EIE, on the other hand, implemented with success the technical assistance envisaged for the Energy Conservation Program. In October 1992 a supervision mission took advantage of the awakening interest of Turkey for energy conservation to reignite discussions with MENR and EIE for the preparation of a study which would assist them in setting up the institutional framework for the Energy Conservation Center and an Energy Conservation Master Plan for Turkey, for which TORs were prepared by EIE and accepted by the Bank. 37. Finally, by end-1993, before the closing date, an energy optimization study was carried out on the steel production complex of ASIL CELIC to analyze, among other things, the energy consumption of the complex to determine its level of environmental pollution and compare results with internationally acceptable standards and assess investment needs together with savings and pay-back periods for energy efficiency projects and retrofitting programs.  - 13 - Environmental Aspects 38. The ESAL was designed to improve the economic efficiency of the power industry aimed at reducing the environmental externalities it imposes on the economy. The increasing use of indigenous but highly contaminating lignite and coal, in particular impose these externalities (through increasing amount of pollutants sulfur dioxide and particulates) on population health, and production in agriculture; forestry; fishing; and tourism. 39. A major trade-off for Turkey is that between increasing the share of environmentally more benign imported fuels and the reduction of pollutants via the retrofitting of indigenous lignite-based power plants. To resolve this problem, TEK with Bank assistance developed a methodology to estimate the damage costs of lignite generated electricity and to rank investments based on environmental considerations. This method also paved the way for the imposition of pollution charges on polluters, one of Turkey's strategic options to bring the environmental dimension into energy sector development. TEK has also created a Department of Environment to carry out environmental impact assessments of its major investments. The sustainability of the effort, however, will depend on stronger Government commitment and further Bank assistance. C. ASSESSMENT OF THE ENERGY REFORM PROGRAM 40. This section evaluates the energy sector reform program under ESAL on the basis of the Government's "Statement of Energy Policy" of May 22, 1987 (P.R. pp. 51-58). It focuses first on three broad reform areas: energy policy, energy planning, import dependence, and assesses trends in the energy sector against the deterioration in the macroeconomic environment which took place during project implementation (para 41-48). Subsequently, the section examines the pattern of sector adjustment during 1987-93 with a view to assessing the economic impacts of Turkey's overall energy policy stance during project implementation (paras 49-53). 1. Energy Policies and Macroeconomic Links Energy Policy 41. The very same year ESAL was being approved, economic instability in Turkey reappeared which ultimately would prove to be highly detrimental to ESAL's basic objectives. High and variable inflation had a particularly deleterious impact on the energy sector during 1987-89, which was compounded by the Government's efforts to reduce public sector expenditure to fight inflation. The turnaround in performance took place in 1987, although this became apparent only in 1988. Both annual investment budgeting and the setting of energy prices were put under serious stress as a result of accelerating domestic inflation and the continuous devaluations of the exchange rate, which also made more difficult the financial situation of some key energy SOEs. Moreover, the significant, although short lived, increase of the world market price for oil in 1987 highlighted once again the vulnerability of the economy  - 14 - to potential external shocks. The Government's attempts at stabilization restrained energy sector investments--only ongoing power projects were included in the 1988 and 1989 budgets. In the end, investments in the energy sector fell substantially in real terms each year during 1987-89, with power investments falling by 16% and mining investments (which include lignite, oil and gas) by over 60%, leading to significant delays in physical implementation/procurement of ESAL's investment component. 42. Most of the fall in real energy prices took place during 1987-89, when inflation accelerated. Domestic fuel prices decreased by about one third in real terms in 1987 alone. With the increase of domestic demand, imports of crude oil and other petroleum products shot up by an unprecedented 18% in real terms to about US$3 billion in 1987, representing almost one-third of the increase in the country's commodity imports. The Government was eventually forced to draw down the oil stabilization fund in order to cover the cash shortfall of TUPRAS in the wake of rising crude oil prices, domestic operating costs and fixed (nominal) retail prices for fuel products. The setting of electricity tariffs suffered from analogous problems. From February 1986, tariffs were allowed to be eroded by inflation, and eventually fell by about 24% in real terms between December 1986 and October 1987; by December 1987, however, significant increases authorized by the Government in the nominal level of tariffs, which were called forth under ESAL, had fully reversed that situation. But the rapid pace of inflation in 1988 eroded those gains, and tariffs in real terms again fell rapidly. As in the case of petroleum, the fall in the real tariff, in conjunction with the rapid growth in domestic demand throughout the economy led to a rapid increase in electricity demand (13% in 1987), although not sufficiently to prevent a deterioration of TEK's financial situation. Internal cash generation for the power subsector reached negative levels in 1988 and 1989 in face of an agreed 35% ratio under ESAL. 43. The evolution of investment in the power subsector, the largest in the energy sector, is also largely explained by the unstable macroeconomic dynamics and the Government's stabilization attempts. Since the mid-1980s, the Government increasingly financed the capital account deficit and covered the requirements of the consolidated public sector by placing financial claims on the domestic bond market. As a result, public expenditure shifted from investment financing to debt servicing, no longer sustaining the large power investments with long gestation periods which had been financed in the previous years. Private savings were, to some extent, channelled to fuel the requirements of an inefficient public sector. Public investment in fixed capital peaked at TL 3.448 billion in 1986 (growing by 14% in real terms from the previous year). This level of public investment was, however, eroded by almost 11 % in real terms during 1987-88, dropping again by approximately 22% in 1990 after a slight recovery in 1989. 44. In contrast to the power subsector whose tariffs were subject to Government controls, the financial situation of the three SOEs operating in the oil and gas subsector TPAO, TUPRAS and (initially) BOTAS did not suffer, mainly because their pricing arrangements were essentially on a cost-plus basis. The Government's deregulation of the industry in 1989 (para 29), in particular, increased TUPRAS' autonomy, since then fully responsible for pricing its own  - 15 - product slate, whereas formerly it was the role of the Government's Directorate of Petroleum Products (although with some recent backsliding; para 31). Reacting swiftly to world market developments in the wake of the Gulf crisis, domestic retail petroleum prices were raised in three separate occasions in August and September 1990, for a cumulative 63 % for gasoline, 38 % for motor oil and 24% for fuel oil. This paved the way for the privatization of the refinery and oil distribution industry launched by the Government in 1993. BOTAS financial situation, on the other hand, suffered considerably as a result of the closure of the oil pipeline from Irak to Turkey's major Eastern Mediterranean seaport following the sanctions imposed to Irak by the UN during the War. 45. The aggressive energy pricing policies of the first half of the 1980s were relaxed in the second half, reflecting a loosening of Turkey's demand management policies at the macro level, but also the softening of energy prices worldwide. Real prices fell quite sharply to virtually all sectors from 1985 to 1988. By 1991 most energy prices had, however, increased in real terms, the exceptions being heavy fuel oil and natural gas for power generation. On a heat equivalent basis, lignite is the cheapest form of fuel in all three sectors (electricity, household heating and industry). Lignite prices for electricity generation are only about 30% of those of heavy fuel oil and about 41 % of those of gas. Conversely, the prices of lignite and natural gas are comparable for industry on a heat equivalent basis. For households, lignite prices are only 59% of natural gas prices. There is cross subsidization of lignite prices as a result of Government intervention, which has led to prices for power generation being below financial costs. Lignite prices do not cover long run marginal costs; in particular, despite increasing development costs at the mines during the last decade, lignite prices for all consumers have fallen significantly in real terms during 1980-90, and still remain much below 1980 levels for the industrial (20%) and power (42%) sectors. Energy Planning 46. Sector planning deficiencies are still apparent: (a) forecasting methods remain mechanistic in their approach; (b) the economic implications of greater private sector involvement in future power generation, as well as of increased reliance on imported energy products (coal and/or gas) are not being fully examined; (c) annual investment decision-making authority remains diffused in spite of the central role played in principle by the core planning agencies; and (d) the capacity to prepare a medium-term rolling investment program is lacking. The efficiency gains envisaged have not been forthcoming, nor have the high expectations placed on private sector investment to develop power generation capacity. Over the long-term, major successes were achieved, nonetheless, as summarized below. Import Dependence 47. The policies and investments launched during 1975-1990 in response to the two major oil shocks were largely successful not only in overcoming the serious energy shortages of the 1970s - early 1980s, but also in changing significantly the structure of the energy sector toward less dependence on imported oil. By 1990 Turkey's overall energy intensity had  - 16 - increased considerably, since net primary energy consumption grew by about 72% during the 1980s, as compared to a 49 % increase of real GDP. With the supply of electricity having grown faster than net primary energy consumption (gross supply more than doubling in 1980-90), the country's dependence on petroleum products had fallen by as much as 5 percentage points to 55% of total commercial consumption. Consequently, in 1990, electricity, accounted for a somewhat larger proportion (27.5 %) of all commercial energy supplied than in 1980 (23 %), with industry's share having declined over the past decade to the benefit of residential consumption. In parallel, the use of domestic lignite expanded significantly during the 1980s (by as much as 6 percentage points) representing in 1990 about 19% of commercial energy consumed. 48. Notwithstanding the vigorous policy of indigenous resource development undertaken by the Government since the 1970s, Turkey's dependence on imported energy was exacerbated over the last decade. The share of energy imports in the country's overall net supply of primary energy increased from 44% in 1980 to 46% in 1990. This has, however, been due to the country's growing pro-environmental stance, leading to greater reliance on imported natural gas and coal with lower sulphur content. This trend clearly reveals the trade- offs between Turkey's trade policies and its environmental objectives. 2. Overall Economic Impact 49. This section assesses whether the broader adjustment objectives were realized by examining key sectoral trends during the 1987-93 project implementation period in reference to: (1) intrasectoral growth patterns and relative price movements; (2) subsector investments and their financing; and (3) efficiency of energy production. This is further examined in Annex 1 which includes a set of indicators for that period. 50. Sector Growth and Pricing Patterns. There were widely divergent output trends over the 1987-93 period: whereas electricity output grew by 58% in real terms, the coal and hydrocarbon sector did so by only 13%. Output in the lignite and hard coal subsectors actually contracted considerably during 1987-93. Moreover, over that period there was a significant fall in real coal/hydrocarbon prices and, albeit to a lesser extent, real electricity prices as well. The combined decline in output and prices have in particular impacted the coal/hydrocarbon sector whose revenue base fell by roughly 42 percent in real terms during 1987-93, with lignite, hard coal and crude oil suffering real revenue declines. In contrast, the electricity subsector's revenue base expanded by about 9 percent in real terms during 1987-93. To some extent, therefore, ESAL has contributed to financing investments in subsectors with contracting outputs and declining profitability, raising concerns about their prospective economic and/or financial viability, and this applies to TKK, TKI, and possibly TPAO. 51. TUPRAS is an exception in so far as the ESAL financed a clearly identifiable subproject, the hydro cracker, for which was estimated an ex ante IRR above the opportunity cost of capital in Turkey. TEK's high voltage transmission investments are also expected to yield high economic returns as they will allow the use of new hydro generation capacity commissioned in the recent past (2,400 MW) in what is a rapidly expanding subsector.  - 17 - 52. Energy Sector Investments. There was a drastic fall of energy sector investments in real terms during 1987-93. The fall in coal/hydrocarbon investment was due to the contraction of the lignite and hard coal subsectors mentioned above, increased competition from imported coal from the former USSR, and the Government policy of subsidizing power generation via lignite prices. On the other hand, the significant reduction in electricity subsector investments is temporary as it is due to the existence of excess generation capacity during that period, the rationalization of TEK's investment program under the TRP, but also, the disappointing performance of the private sector. In spite of a significant increase in the volume of private sector investment for power development during 1987-93, it still amounted to only US$106 million in 1993, compared to the sector's US$1.1 billion in public sector investment that same year (1988 prices). SOE (mostly TEK) self-financing capacity declined as its contribution fell from 68% of power subsector investment in 1987 to 59% in 1993, highlighting the need for mobilizing private sector resources. In sum, investment performance was mixed with the sizable fall in coal/hydrocarbon investments signalling the long-term decline of the coal and lignite industries, whereas the fall of power subsector investments during 1987-93 was of a temporary nature as the subsector's growth prospects are very good. 53. Energy Efficiency. Overall energy intensity increased from 428 to 491 toe/GNP in Turkey during 1987-93. Turkey has thus so far not yet been able to undertake the major structural shifts in energy production and use which characterized the most developed economies in the wake of the two oil shocks leading to significant declines in their energy intensities. Efficiency in the transformation of net primary energy supply (NPES) into secondary energy also seems to have decreased in Turkey from about 18% of NPES in 1987 to 20% in 1993. There are significant variations, nonetheless, within subsectors: (a) whereas overall power generation became more efficient, this was not the case of thermal power generation whose transformation process became less efficient in energy terms; (b) the oil refinery subsector maintained its transformation efficiency; and (c) own energy use/losses, on the other hand, increased over the period reflecting, in part losses in the transmission and distribution of electricity. 54. In sum, from the point of energy production and distribution there seems to have been a decline in efficiency during 1987-93 with thermal power plants and transmission/distribution network being the major culprits. This is due to a variety of factors, in particular, the continued repression of lignite prices for power generation (which offers little incentive to improve the quality of this fuel), the relative obsolescence of some of TEK's thermal plants, and the utility's own operational inefficiencies. These results have to be taken with some caution as there are other factors which could increase the amount of energy required for the transformation process, such as the shift from hydro to thermal power generation and the increase in the capacity of the refineries to produce higher value added fuels. Sector work was not carried out to examine the critical issue of energy efficiency and conservation in Turkey.  - 18 - D. BORROWER'S AND BANK'S PERFORMANCE Borrower's Performance 55. The Borrower's performance was mixed as evidenced by its inability to implement the loan in the agreed timeframe and by the lack of a sustained and intensive supervision effort. Originally, the closing date was established as September 30, 1990, nevertheless the ESAL's closing date was extended three consecutive years up to September 30, 1993. The operation was designed to be implemented in three years from 1987 through 1990. In fact, it took six years to conclude the investment component of the ESAL. Although, disbursements accelerated during the last years, initially they were very slow. By end-1992, 87% of the investment component of US$435 million had been utilized. The rate of physical progress after the major reallocation of funds in August 1991, proceeded close to expectations in the case of most of the beneficiaries with the exception of TEK. By the last closing date, 95% had been committed. 56. Most of the beneficiary agencies had never worked with the Bank before, and therefore were not familiar with the standard Bank procurement practices. Furthermore, in 1987, after the second tranche release, the economic performance of Turkey deteriorated and this had a dampening effect on disbursements (para 41). In this environment, the utilization of ESAL's funds slowed down even further. 57. It took some of the agencies time to receive adequate training and proper exposure to Bank's procurement procedures. There were other agencies such as TTK who continued to resist adopting Bank procurement practices. This caused both delays and embarrassing situations where international private companies which had submitted bids for procurement awards more than once complained to their Government representatives challenging the fairness of TTK's procurement practices and requested Bank's assistance to normalize these procurement practices in line with the guidelines. Bank's files indicate that its staff spent considerable time in attempting to assist several beneficiaries and encouraged them to follow the agreed guidelines. Procurement contracts sent to the Bank for "no objection" of the award in almost all cases were found to be unsatisfactory. 58. Since the date of ESAL effectiveness, Bank staff had alerted management that the project could encounter problems if corrective measures were not taken upfront, given the significant increase of the investment component which jumped from US$135 million as envisaged in the LA to US$435 million when the cofinancing from JEXIM was obtained. This increased greatly the complexity of loan administration, the uncertainties surrounding disbursement profiles and the risks associated with misprocurement and subsequent cancellations. 59. The limited involvement of the Treasury, the agency in charge of coordinating the ESAL and in supervising the 11 beneficiaries, exacerbated the difficulties encountered. Because of this, the Bank had to concentrate unduly on procurement, disbursement and physical  - 19 - implementation, particularly, during the first two years of the program. Quarterly reports and auditing reports of the beneficiaries were also remitted to the Bank with delays. On occasion, the Bank requested the Treasury to improve the quality and timing of the Quarterly Progress Reports and audit reports of the beneficiary agencies with mixed results. In fact, the Progress Reports were discontinued by Treasury in 1992 in spite of repeated Bank requests. Quality of Bank Supervision 60. On the whole Bank supervision had mixed results: instrumental for the implementation of the investment component, but to a much lesser extent the policy component. The supervision team of July 1991, in the wake of the second extension of the closing date reached important agreements with the GOT concerning tariff policy, institutional reform of TEK, ciiteria for the allocation of investment funds and enhanced participation of the private sector. This was done more than three years after second tranche release, however, and moreover, focussed mainly on power sector issues. The environmental dimension was strengthened during the last two years but under a different operation (TRP). 61. Because the ESAL dealt with broad policy and institutional reform, investment selection, use and conservation of existing energy capacity, and indirectly the environmental dimension, the supervision teams should have involved more economists. Supervision of the policy component should have been stronger and sustained. E. MAIN LESSONS LEARNED 62. The major lesson drawn from the ESAL experience is that it was premature to launch an energy sector adjustment program when the macroeconomic environment is turning around and Government's commitment to the reform program is ambiguous. The Bank's ability to make available quick disbursing funds was of considerable importance to Turkey as it suffered at the time critical foreign exchange shortages. The Bank was seemingly subject to pressure to make the loan as early as possible with a view to financing critical energy imports, and it did so without Turkey being able to reach an internal consensus on the scope of the reform program. The design of the project suffered accordingly, and the risks involved were not fully assessed (paras 63-64). 63. The Bank was in particular unable to assess properly the macroeconomic risks, probably because Turkey was still being perceived as one of the star performers in the developing world, and because it was unwilling to factor in important political economy aspects. The Bank could have delayed the Loan to help reach such consensus and reduce the risks of backsliding. It could have also called upon outside expertise, including top international credit rating agencies, to validate its own reading of Turkey's macroeconomic stance. 64. Another major lesson refers to the overly ambitious project design which involved a multitude of recipient institutions against the limited local absorptive capacity, particularly in terms of procurement and project coordination. SECALs should be the object of intensive  - 20 - supervision and critical importance given to the policy component, which was not the case under ESAL. The Bank is in a weak position to bring to fruition complex undertakings such as the ESAL program, given Turkey's political and institutional realities. Until this situation persists it should not get involved in this type of operation except in very exceptional circumstances. If it does, the number of participating institutions should be reduced to a few. The absence of a clearly agreed core investment program for each SOE and of an operational mechanism to monitor them adequately was damaging to project implementation. Supervision should focus on the institutions' core investment programs as leverage can be exercised most constructively via this mechanism. The policy measures should be mainly the responsibility of the SOEs (rather than the Central Government) with which the dialogue is more decentralized, fluid and sustainable. A sector investment loan would have probably been more effective and manageable. To this effect, greater priority should have been given to sector work, without which the Bank looses intellectual leadership and credibility, as was to some extent the case during project implementation. 65. As Turkey was unable to launch a credible SOE reform and privatization program to confront systemic problems, ESAL was ultimately derailed by macroinstability and persistent sector inefficiencies. The Loan should have been closed earlier and not dragged out for three years without direction. It was only because a new investment operation was prepared and approved by the Bank that it was possible to carry out some reforms in the power sector (TRP). This should not be the case, however, and ESAL should have been able to stand on its own feet, which it did not. Bank lending should instead help Turkey mobilize the private sector resources needed to be able to cope with the rapid pace of energy demand, for which the establishment of an independent adequate regulatory and incentive framework should have been given greater priority; and to help arrest environmental degradation from the use of fossil fuels, for which a specific environmental program should have been agreed upon.  - 21 - PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE No comments have been received from the Borrower.  - 22 - PART III: STATISTICAL INFORMATION BASIC DATA SHEET (US$ Million) ___________________ Original Disbursed Cancelled Repaid Outstanding Loan No. 2856-TU 325 312.0 62.01 249.96 BLoanLB-02201 1/ 30.1 30.59 - j- 30.1 Original Date Actual Date Initiating Memorandum 10/86 11/5/86 Appraisal 11/86 11/5/86 Negotiations 04/87 04/30 - 05/15/87 Board Approval 06/87 06/25/87 Loan Agreement 06/87 06/29/87 Effectiveness 07/87 07/02/87 Loan Closing 09/30/90 09/30/93 CUMULATIVE LOAN DISBURSEMENT 1988 1989 1990 1991 1992 1993 (i) Planned 88.5 265 325 (ii) Actual 246.72 2/ 258.09 297.74 332.89 337.39 342.59 (iii) (ii) as % of (i) 244 86.0 82.2 1/ US$30.59 million disbursed at B-loan effectiveness date (September 24, 1987) 2/ US$175 million released for BOP support. (US$75 million at Loan effectiveness and second tranche US$100 million released in February 1988. Also includes B-loan.  - 23 - MISSION DATA Month/ No. of No. of Staff Date of Year Weeks Persons Weeks' Report 1st. Preparation2 10/85 4 3 12 12/12/1985 2nd. Preparation 6/86 3 13 36 7/6/86 3rd. Preparation 9/86 1 3 3 3_ Appraisal 11/86 3 9 27 1/5/87 Post Negotiations 3/87 1 2 2 3/17/87 Supervision I 9/1/87 1.3 6 7.8 9/30/87'4 Supervision II 12/6/87 2 8 2.7 4/6/88 Supervision III 4/88 4 2 8 5/16/88 Supervision IV 10/18/88 1.3 1 1.3 12/2/88 Supervision V 11/28/88 2 8 2.7 2/28/89 Supervision VI 5/25/89 1.1 2 2.2 6/30/89 Supervision VII 7/89 2.1 6 2 9/89 Supervision VIII 11/28/89 2 5 2 10/10/89 Supervision IX 2/7/90 .6 1 .6 4/5/90 Supervision X 7/5/90 .5 1 .5 8/21/90 Supervision XI 7/25/90 2.5 3 1.3 9/25/90 Supervision XII 6/18/91 2.1 9 6.3 8/16/91 Supervision XIII 8/92 2.6 4 6.8 10/30/92 Supervision XIV 6/93 1 1 1 7/9/93 5 Supervision XV 7/93 2 2 1 7/15/93 6 Estimates. (mission staff weeks)/(No. of projects visited) 2 Figures includes a supplementary mission whose report was presented on 6/30/86 Back to office report not available in files. This mission presented two BTORs each on key segments of the ESAL This mission supervised only the refinery project. 6 This mission was fielded to participate in the review of the regulatory framework study for the power subsector under the project.  - 24 - STAFF INPUT Staff-Weeks FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 Preparation/ 12 39 Appraisal l Negotiations/ 27 2 Post Negotiations Supervision 16 7 6 2.1 6.3 10.8 Completion 5. Total 12 66 18 7 6 2.1 6.3 15.8 TURKEY ESAL CLn 2e56-T) 70 50 40 3 tn u 20 86 87 89 89 90 91 92 93 YQarr 19E6-93 Sevi gion  - 25 - OTHER PRO.TECT DATA Related Projects FOLLOW-ON INVESTMENT OPERATIONS COUNTRY: Turkey. TEK Restructuring Project. LN 3345-TU, effective on Apri, 27, 1992 in the amount of US$300 million. COUNTRY: Turkey. Berke Hydropower Project. LN.3476, effectve on October 29, 1992 in the amount of US$270 million.  - 26 - TURKEY Energy Sector Adjustment Loan (Ln. 2856-TU) Compliance with Loan Covenants Agreement Clause Covenant Performance Loan 3.05 Review by October 31 each year These actions were partially the composition, size, and met particularly with regard to financing of the public investments the power sector; in 8/92 the for energy for the next year. last power sector investment assessment for 1993 was completed. Loan 3.06 Send to the Bank for review and This covenant was not met, comment, not later than May 1, primarily due to TKI's 1988, the study on the operating reluctance to use borrowed practices and policies of TKI and funds for technical assistance. the measures to be taken to In addition, it was unable to improve the operational efficiency mobilize grant funds to carry of TKI. out the study. Loan 4.03abc Set of actions to improve TEK's These were not complied with financial performance. under ESAL. After 1991 TEK's financial covenants were monitored under the TRP. A new adjustable tariff policy set in US$ terms not only prevented TEK's financial collapse but also permitted compliance with the new financial covenants in 1993. esal-cvn  - 27 - Annex I Page I of 3 NOTE ON THE PROJECT'S ECONOMIC IMPACT 1. The investment component was designed as a three year time slice expenditure program and not as a series of discrete and clearly identifiable subprojects. As such, with the possible exception of the oil refinery (TUPRAS), it was not feasible to estimate internal rates of return for the individual project items financed. One can nonetheless attempt to assess whether the broader adjustment objectives were realized by examining key sectoral trends during the 1987-93 implementation period: (1) the first pertains to the nature of the intrasectoral growth patterns and relative price movements (Table 1); (2) the second refers to the evolution of subsector investments and their financing (Table 2); and (3) the third to the overall efficiency of energy production at the country and subsector levels (Table 3). (The indicators in Table 1 are presented in indexed form with 1987=100.) 2. Subsector Growth and Pricing Patterns. The energy sector in Turkey for purposes of the ESAL comprises two major subsectors: coal and hydrocarbons-which includes lignite, hard coal, oil and gas- and electricity. There were widely divergent output trends over the 1987-93 period: whereas electricity output grew by 58 % in real terms, the coal and hydrocarbons sector did so by only 13 % (Table 1 below). Output in the lignite and hard coal subsectors actually contracted and considerably during 1987-93. Moreover, over that period there was a significant fall in real coal and hydrocarbon prices and, albeit to a lesser extent, real electricity prices as well (Table 1). The combined decline in output and prices have in particular impacted the coal and hydrocarbon sector whose revenue base fell by roughly 42 percent in real terms during 1987-93, with lignite, hard coal and crude oil all suffering real declines. In contrast, the electricity subsector's revenue base expanded by about 9 percent in real terms during 1987-93. 3. In sum, although one cannot draw a final conclusion on this matter, it is clear that the financing of investments in subsectors with contracting outputs and declining profitability raises concerns about their respective economic and/or financial viability, and this applies to TKK, TKI, and possibly TPAO too. TUPRAS is an exception in so far as the ESAL financed a clearly identifiable subproject, the hydro cracker, for which these was estimated an ex ante IRR above the opportunity cost of capital in Turkey. TEK's high transmission investments are also expected to yield high economic returns in a rapidly expanding subsector. Table 1: Energy Sector Adiumtment Indicators (1987-100) Output Prices in Nominal Prices in Terms Real ____________ Terms