Tunisia - Public Enterprise Reform Loan Project

Documnt of The World Bank FOR OFFMCLAL USE ONLY Report No. 14037 PROJECT COMPLETION REPORT REPUBLIC OF TUNISIA PUBLIC ENTERPRISE REFORM LOAN (LOAN 3109-TUN) MARCH 9, 1995 Industry and Energy Country Department I Middle East and North Africa Region 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.  CURRENCY EQUIVALENTS Currency Unit: Tunisian Dinar (DT) 1989 US$1 = DT 0.95 1990 US$1 = DT 0.88 1991 US$1 = DT 0.92 1992 US$1 = DT 0.88 1993 US$1 = DT 1.00 ABBREVIATIONS CAREPP Commission d'Assainissement et de Restructuration des Entreprises Publiques CPG Compagnie des Phosphates de Gafsa GC Groupe Chimique GDP Gross Domestic Product ICB International Competitive Bidding IM Initiating Memorandum IMF International Monetary Fund PCR Project Completion Report PE Public Enterprise PERL Public Enterprise Reform Loan SCG Societ6 des Ciments de Gabes SNCFr Societe Nationale des Chemins de Fer Tunisiens SNT Societe Nationale de Transport SOE Statement of Expenses SORETRAS Societe Regionale de Transport de Sfax STEG Societe Tunisienne d'Electricit6 et Gaz STIR Societe Tunisienne des Industries de Raffinage STS Societe de Transport du Sahel USAIID United States Agency for International Development  Foa oMcUL USE THE WORLD BANK Washington, D.C 20433 U.S. A. Office of Director-General Operations Evaluation March 9, 1995 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: TUNISIA - Public Enterprise Reform (Loan 3109-TUN) Project Completion Report Attached is the Project Completion Report (PCR) on Tunisia - Public Enterprise Reform Loan (PERL), (L3109-TUN) prepared by the Middle East and North Africa Region. The PCR does not contain Part It. The PERL was part of a series of loans in support of broader structural adjustment in Tunisia toward deregulation, trade liberalization, and greater reliance on market forces. Specifically, the PERL supported the introduction of policies affecting public enterprises to promote increased efficiency and profitability and to support private sector development through privatization. Reduction of excess employment and restructuring was planned for targeted public enterprises. Privatization was brought into the immediate government agenda. Performance contracts were to be undertaken to improve management autonomy and enterprise efficiency. The above objectives were partially achieved, but there were substantial delays in the preparation, adoption and implementation of performance contracts among the preselected enterprises. The privatization program had a promising beginning, but it failed to accelerate. Limited progress in privatization was partly due to inadequate performance criteria linking privatization to objectives in the loan design. The outcome of this project is rated as marginally unsatisfactory. The sustainability of the project is rated as uncertain. The institutional development impact is rated as modest. The PCR quality is satisfactory. An audit is planned. This document has a restricted distribution and may be used by reciplents only in the performance of their official duties. Its contents may not otherwise be disdosed without World Bank authorization.   FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT REPUBLIC OF TUNISIA PUBLIC ENTERPRISE REFORM LOAN (Loan 3109-TUN) TABLE OF CONTENTS PREFACE ..............i EVALUATION SUMMARY ..................................... iii PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE ......1.......... 1. Project Identity ...................................... 1 A. Economic and Sector Background ................ 1 2. The macro-economic setting .............................. 1 B. Public Enterprise Reform Loan (PERL) .............. 3 3. Loan Origin and Genesis ................................ 3 4. Loan Objectives and Description ........................... 6 5. Loan Implementation ................................. 10 6. Implementation of Individual Components .................... 12 C. Impact and Performance Assessment Under the PERL ...... ......... 17 7. Loan Impact and Sustainability ........................... 17 8. Borrower Performance ................................ 19 9. Bank Performance ................................... 21 10. Monitorability and Realism of Timetables ...... .. ............ 23 1 1. Catalytic Effect of the PERL on Mobilizing Additional Funds ... ..... 24 D. Lessons Learned ...................................... 24 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ... ....... 27 PART III: STATISTICAL SUMMARY ............................ 29 Table 1: Related Bank Loans .................................. 29 Table 2: Loan Data ........................................ 30 Table 3: Loan Timetable: Original and Actual Dates ................... 31 Table 4: Cumulative Loan Disbursements .......................... 31 Table 5: A: Use of Bank Resources .............................. 32 B: Mission Data .................................... 33 Table 6: Status of Loan Covenants .............................. 34 Table 7: Data to be furnished under the PERL ....................... 36 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. I   PROJECT COMPLETION REPORT REPUBLIC OF TUNISIA PUBLIC ENTERPRISE REFORM LOAN (Loan 3109-TUN) PREFACE This is the Project Completion Report (PCR) for the Public Enterprise Reform Loan (PERL) to the Republic of Tunisia. The Loan was approved by the Board on July 11, 1989 and consisted of Loan 3109-TUN in the amount of US$130 million equivalent. The original closing date of December 31, 1991 was extended two times to June 30, 1993. The final disbursement was made on June 28, 1993, and the loan was fully disbursed. This PCR was prepared by the Industry and Energy Division, Country Department 1 of the Middle East and North Africa Region (Preface, Evaluation Summary, Parts I and III). It is based on the Initiating Memorandum, the President's Report, minutes of negotiations, Loan Agreement, supervision reports, internal Bank memoranda, and correspondence between the Bank and the Borrower. The Government was sent Parts I and III on May 12, 1994. The Bank requested the Borrower on January 27, 1994 to prepare Part II by April 1, 1994, and as of the date of this report no reply had yet been received.   iii PROJECT COMPLETION REPORT REPUBLIC OF TUNISIA PUBLIC ENTERPRISE REFORM LOAN (Loan 3109-TUN) EVALUATION SUMMARY Background Tunisia's strong economic performance of the 1970s began to falter in the early 1980s, with the fall in world oil and phosphate prices. Necessary economic adjustments were slow in coming and by 1985 a balance of payments crisis was threatening. Government responded by beginning a process of policy reforms that broadened into a wide-ranging program of structural adjustment that included measures to reduce administrative controls, liberalize the trade and price regime, improve the fiscal and incentives systems, and strengthen the financial sector. The aim was to encourage a more outward-oriented growth and to rely increasingly on the private sector as the engine of growth. The Bank and IMF supported the Government's adjustment program (paras. 2.1-2.4 of main report). Within this framework of creating more competitive conditions, the Government also undertook reform of the public enterprise sector. The objectives were to: (a) divest Government ownership in all public enterprises (PEs) in competitive sectors; and (b) for all PEs likely to remain in the public sector for the foreseeable future, strengthen PE management autonomy, clarify the relationship between the PEs and Government, and streamline Government supervision (paras. 2.5-2.8 of main report). Loan objectives and design In support of the Government's public enterprise reform program, the three key objectives of the loan were to: (a) support system wide reform, including legal and institutional reforms, performance contracts, PE management improvements, and budgetary and financial reforms; (b) reinforce the Government's program of divestiture and restructuring; and (c) sub-sectoral action programs for selected PEs: CPG, SNCFT, and SNT', that were to have acceptable performance contracts, and Groupe Chimique. The loan had two tranches, the first of US$70 million to be released at effectiveness, expected in December 1989, and the second tranche of US$60 million, expected to be released around December 1990. The loan was to close on December 31, 1991. 1/ CPG: Compagnie des Phosphates de Gabes; SNCFT: Societe Nationale des Chemins de Fer Tunisiens; SNT: Societe Nationale de Transport.  iv The four conditions for effectiveness were: (a) Submission to the Bank of satisfactory performance contracts for CPG. SNCFT. and SNT. approved by the enterprises concerned and the supervising ministry; (b) Submission of a program satisfactory to the Bank for the preparation of performance contracts for additional PEs selected in accordance with criteria satisfactory to the Bank; (c) Presentation for sale of PEs selected on the basis of criteria satisfactory to the Bank; provision to the Bank of a satisfactory program for the divestiture of additional PEs selected in accordance with these criteria; (d) Allocation by Government to the Public Enterprise Restructuring Fund of an amount at least equivalent to US$70 million. The four conditions for second tranche release followed from the conditions for effectiveness: (a) Approval by the CAREPP2, signature of and satisfactory implementation of key actions in the performance contracts of CPG, SNCFT, and SNT; (b) Approval by Government and the enterprises concerned of performance contracts of an agreed group of additional PEs; (c) Government to bring to the point of sale a group of additional PEs as agreed under the condition for effectiveness; (d) Commitment by the Public Enterprise Restructuring Fund of at least US$35 million equivalent and allocation of at least an additional US$60 million equivalent to the Fund. Other actions to be monitored for release of the second tranche included: (a) satisfactory progress in the public enterprise reform program, including implementation of the new legislation on PEs, streamlining Government supervision, and increasing the autonomy of the PEs; (b) development of satisfactory data collection systems and provision to the Bank of annual data on budgetary and extra-budgetary transfers to PEs and the arrears situation of PEs; and (c) satisfactory progress on an action plan for Groupe Chimique. In addition, a Japanese grant was put in place in association with the PERL in early 1990 to finance technical assistance to be hired to develop management tools in priority PEs. A second Japanese grant was put in place in 1992 to finance technical assistance to study privatization prospects for the cement, dairy, and ceramics sectors (para. 3.7). 2/ CAREPP: Commission d'Assainissement et de Restructuration des Entreprises Publiques  V Loan implementation The PERL became effective on February 22, 1990, about two months behind schedule. The Bank had problems with draft performance contracts for SNCFT and SNT submitted after negotiations which had a number of parameters that differed considerably from those agreed at negotiations. While some evidence exists that Government made some of the changes as requested by the Bank, there is no evidence in the files that the Bank ever received fully satisfactory performance contracts that had been approved by the supervising ministry, although this was a condition of effectiveness. Nevertheless, on February 5, 1990 the Bank informed the Government that once the legal opinion was received from Government on the legal documents, the loan could be declared effective. The Bank seemed willing to overlook the fact that the Bank had never actually received acceptable documents in order to declare the loan effective. No waivers to the legal agreement were required (para. 5.1). The second tranche release was delayed by about twenty five months, mainly because of problems with the performance contracts. Initially expected in December 1990, the second tranche was released in January 1993. A July 1991 mission found satisfactory progress on institutional reform and privatization, and the use of the Enterprise Restructuring Fund far exceeded the target for second tranche release. In addition, the mission obtained much of the data required for monitoring the public enterprise reform program. Evidence of the commitment of over US$60 million equivalent to the Enterprise Restructuring Fund was received in August 1991, and at that point an audit for 1990 was also required, received in September 1991 (para. 5.2). The performance contracts on the first group of PEs, CPG, SNCFT, and SNT, had been approved by the CAREPP only in April 1991 (more than one year after they were supposed to have been approved by the supervising ministry). Sometime in early 1992, the Bank discovered that the first group of performance contracts had never been signed. As a solution, the Bank sought evidence that the performance contracts for CPG, SNCFT, and SNT had in fact been substantially implemented; if this proved not to be the case, the Bank insisted on satisfactory performance contracts for the period 1992-96. In October 1992, the Bank determined that the elements in the performance contract for CPG had been substantially met, but not for SNCFT and SNT, essentially because of Government's failure to meet its financial obligations under the contracts. The Bank then required that the performance contracts for 1992- 96 be predicated on adequate financial restructuring of the two transport companies to enable them to meet their targets under their contracts. The Bank was satisfied with the performance contracts drawn up, with the help of Bank consultants, and on receiving evidence of their approval and signature in December 1992, released the second tranche in January 1993 (paras. 6.1-6.8). The Government has used both Japanese grants, albeit at a slower than expected pace, for the purposes intended. About 40 % of the grant to strengthen PE management has been disbursed and another 23 % committed. The second grant for privatization studies has also been largely committed, although the funds proved sufficient to finance only two of the three sector  vi studies intended (cement and ceramics), and the Government plans to request another Japanese grant to fund further privatization studies. Loan Impact and Sustainability In support of legal and institutional reforms, the loan probably had the biggest impact during its preparation and appraisal. Bank missions made specific recommendations on various aspects of the institutional framework and there is some evidence that a number of the Bank's recommendations were followed. The loan contained only a general condition, however, requiring an exchange of views on progress on the public enterprise reform program. Since the the Bank at all times was satisfied with Government action in the public enterprise reform program, it is difficult to assess any direct impact the loan had on this aspect (para. 7.1). On privatization, counterpart funds from PERL disbursements were used to finance the Public Enterprise Restructuring Fund, which in turn was used to finance compensation for worker lay-offs and PE liabilities to the social security fund for PEs undergoing either privatization or restructuring. Allocations to the Public Enterprise Restructuring Fund exceeded the equivalent of the loan amount. Thus the PERL made a direct short term contribution to the privatization process. On the other hand, the PERL played a passive role in terms of affecting either the scope or the speed of privatizations in Tunisia, accepting the Government's position that committing itself to a specific timetable with a list of PEs to be privatized would be politically impossible. In spite of the slow pace of privatizations,3 the Bank was satisfied with progress to the extent that the process never stopped throughout the life of the loan. It could be argued that the long-term impact of the PERL on the privatization program was quite limited (paras. 7.2). The loan's most direct impact was on the elaboration of performance contracts on the basis of financially and economically sound principles. In spite of Government reluctance to face up to its financial responsibilities, the Bank insisted that the Government undertake fundamental financial restructuring of the PEs concerned prior to starting the implementation of the performance contracts. The Bank also insisted that a transparent mechanism, based on sound economic principles, be included in the performance contract for annual compensation from Government for the social services performed by the PEs. The release of the second tranche was held up for more than two years essentially because of the issue of satisfactory performance contracts. Thus the PERL has contributed to the elaboration, approval, and signature of performance contracts for at least six PEs, which are based on the principles of on-going responsibility of the State and the respective PEs. To the extent that these contracts are now being satisfactorily implemented, the PERL can be credited with improving the efficient and economic performance of at least six relatively important PEs (para. 7.3). 3/ In the couise of one year, from end-1990 to end-1991, Government sold US$40 million of its assets in PEs, accounting for less than 0.3% of total assets of PEs as of end-1987; at end 1991, after four years of privatizations, only about 1% of the assets of PEs had been divested.  vii The contribution of the PERL to the long-run sustainability of the public enterprise reform program is weakest on the institutional and privatization aspects. Here the PERL played a relatively passive role. The Bank endorsed the Government's narrowing of the definition of public enterprises, because the Bank felt that it facilitated the Government's supervision role. The real meaning of the exclusion of 400 formerly public enterprises from the definition of "public enterprise" was never examined. At a minimum, the Bank should have requested the Government, under the PERL, to commission a study on the status, performance, and access to subsidies, services, and bank financing of these no-longer-public enterprises, with a view to defining whether a more pro-active approach to these enterprises (such as divesting Government's share) was necessary. In addition, the Bank should have questioned the Government's approach to privatizing only those PEs in "competitive" sectors, by examining the definition of competitive and questioning the scope for privatizing even in non-competitive sectors through the establishment of an appropriate regulatory framework. While reference was made in the PERL's legal documents to acceptable "criteria" to be used for selecting PEs for privatization, there is no evidence in the files that such criteria were ever explicitly discussed or agreed. Thus the process of PE privatization remains to this day relatively slow and non- transparent (paras. 7.4-7.5). On the other hand, the contribution of the PERL to the long-run sustainability of improving PE performance can be considered more substantial. By supporting the elaboration of clear performance contracts for a number of priority PEs, the PERL reinforced the notion of Government's explicit financial responsibility for up-front financial restructuring as well as continued annual support of the PEs to prevent the recurrence of an unsound financial structure. By insisting on specific performance targets, the PERL also reinforced the notion of PE accountability to reaching monitorable efficiency goals. Whether these performance contracts will in fact be satisfactorily implemented, and whether the use of them can be sustained and generalized to other priority PEs remains to be seen. Evidence from many other countries across time has shown that Governments are reluctant to meet their obligations to PEs when faced with resource constraints, and performance contracts in the final analysis have proven to be a third or even fourth best solution to the problems of loss-making public enterprises. The question of long-run sustainability is therefore still open to the test of time, as is Government's commitment and willingness to face its financial responsibilities (para. 7.6). Conclusions and lessons learned In summary, the loan of US$130 million represents a substantial part of Tunisia's lending program. Conditionality for an adjustment loan of this magnitude should have been concomitantly substantial. Instead, the Bank established unmonitorable criteria, relying on "satisfactory progress" in two areas critical to the reform process: institutional reform and the privatization program. Where the criteria were monitorable, they were not met. Satisfactory performance contracts to be approved by the supervising ministry, a condition of effectiveness, were apparently never received by the Bank; the Bank nevertheless declared the loan effective. These same performance contracts were never signed or implemented; the Bank had to find a constructive and creative way to substitute conditionality to release the second tranche.  viii Ultimately, it could be argued, that the sizeable adjustment loan brought remarkably few long- run tangible benefits to the reform program it was designed to support. The five main lessons that can be drawn from the PERL: * Given the importance of the public enterprise sector in Tunisia, which represents a drain on Government's scarce human and financial resources and may be crowding out the private sector in several important areas (such as access to credit), the PERL was an important and justifiable operation. * Objectives under an adjustment loan of this nature should recognize the long-term nature of the process of restructuring and privatization of public enterprises, and set concomitantly modest targets. At the same time, however, conditionality under an adjustment loan should include quantified and/or clearly monitorable targets that demonstrate Government's progress in carrying out the reform program. The PERL was weak in this respect, and had a limited impact, on two of the three objectives it set for itself: institutional reform and the privatization program. * With respect to performance contracts, estimates should be made, early on in the cycle, of the budgetary implications to Government of the measures to be contained in the contracts; in addition, the public enterprises involved should be present at negotiations when the details of the contracts are agreed in principle, and the Ministry of Finance should be associated with the process. These measures may ensure greater willingness by all sides of the contract to honor their obligations. - Adequate resources for supervision are critical for ensuring timely and thorough follow up of the reform program. In the case of the PERL, had supervision missions been carried out earlier and with more adequate resources, the failure to respect an important condition of second tranche release would have been recognized at least one year earlier than it was. Government compliance and release of the second tranche might have occurred as much as six months earlier than it did, and Government's commitment to performance contracts might have been reinforced considerably earlier as well. * Firmness by the Bank on conditionality, combined with a willingness to help Government to meet its obligations on monitorable conditionality, is an effective way to achieve good results on the implementation of an adjustment program, without harming the relationship between the Bank and Government (para. 12.1).  PROJECT COMPLETION REPORT REPUBLIC OF TUNISIA PUBLIC ENTERPRISE REFORM LOAN (Loan 3109-TUN) PART 1: LOAN REVIEW FROM BANK'S PERSPECTIVE 1. Project Identity Project Name: Public Enterprise Reform Loan (PERL) Loan Number: 3109-TUN Region: MENA Country: Republic of Tunisia Sector: Public Enterprise Sector Loan Amount: US$130 million A. ECONOMIC AND SECTOR BACKGROUND 2. The macro-economic setting 2.1. Despite modest natural resources, Tunisia's economic performance has been generally strong. During the 1970s, the Tunisian economy grew at over seven percent p.a. on average, aided in part by a brief oil boom. Non-oil exports grew at over ten percent p.a., however, and all sectors did well, especially manufacturing, whose share in GDP and exports increased substantially. The rapid growth of GDP was due to high levels of investment, about 30 percent of GDP, financed in part from oil export earnings (accounting for 17.5 percent of budget revenue), while levels of external financing remained modest, at about 42 percent of GDP by 1979. Domestic inflation remained moderate, at about six percent p.a. 2.2. Difficulties began in the early 1980s, with the fall in world oil and phosphate prices and the depletion of oil reserves. Necessary economic adjustments were slow in coming. The Sixth Plan (1982-86) proposed a reduction in the investment rate, a shift to labor-intensive investments, increased emphasis on exports, and containing the growth of Government's recurrent expenditures. Instead, from 1980 to 1984, investment remained high, at 31 percent of GDP, mostly in the public sector, and concentrated on heavily capital intensive investment projects that were slow-maturing and, in the event, yielded low returns. Wages increased in real terms and outstripped productivity increases, dampening the demand for labor, inflation rose to an average of ten percent p.a. over this period, and exports slowed. The continued expansionary policies kept growth high, at about 4.5 percent p.a., but also involved continued external borrowing, so that by 1984 external debt had increased to 46 percent of GDP.  2 2.3. The Public Enterprise (PE) sector mirrored the developments in the rest of the economy. The sector was quite large and included 600 public enterprises in which the state held at least 10% ownership, accounted for about 25-30% of value added, 30% of formal sector employment excluding agriculture, about half of total enterprise investments, and 75 % of export earnings. During the early 1980s, large public sector investments were made in heavy industry (fertilizer, cement, sugar, car assembly, steel) and in infrastructure (particularly the railway). PE investment represented 57% of total investments during the Sixth Plan (1982-86). Overall indebtedness increased markedly, and with the increase in real wages, the wage bills of the largest PEs increased, between 1981 and 1982, from 38% to 45% of value added. With the economy-wide increase in unemployment, there was considerable political pressure on PEs to increase employment and considerable social resistance to reducing employment, even where warranted. Government attempted to reduce the oversight burden on its PEs by passing a law in 1985 that redefined public enterprises as those in which the state held 34% or more of the shares; the number of PEs defined as such decreased to 300. In 1989, the definition was changed again to reduce further the resources necessary for PE supervision, to enterprises in which the state owns at least 50% of the shares either directly or through wholly-owned state subsidiaries. This reduced the number of public enterprises to 200. Companies defined as PEs were subject to legal and administrative controls, including Government supervision. Macroeconomic and sectoral adjustment programs 2.4. By 1985 it was clear that a balance of payments crisis was threatening. Government responded by beginning a process of policy reforms that broadened into a wide-ranging program of medium term structural adjustment. The adjustment program aimed at encouraging a more outward-oriented growth, reducing administrative controls, reforming the incentive structure including the fiscal system, and promoting the private sector to take a lead role in the economy. The adjustment measures that have been undertaken on a consistent basis since 1986 include trade liberalization, which involve reducing quantitative restrictions and tariffs, liberalization of domestic margins and producer prices, reduction of across-the-board subsidies, introduction of targeted subsidies, simplification and improvement of the fiscal system, and gradual liberalization of the financial sector through removal of a priori controls and deregulation of the structure of interest rates. The Bank and IMF have supported this adjustment program through a series of Structural and Sectoral Adjustment Loans (Bank) and a Stand-by arrangement, purchase of funds under the compensatory financing facility, and an extended arrangement (IMF). Government strategy and performance in the public enterprise sector 2.5. It was in the context of this medium term structural adjustment that the Government started a program in 1987 of disengaging from its activities in the public sector in certain areas and ensuring greater efficiency and accountability from the public enterprises in others. Within the framework of creating new, competitive conditions, the PE reform program was designed to enable the sector to adapt by rationalizing the relations between the state and the enterprises and by divesting, restructuring or liquidating PEs.  3 2.6. Government's stated objective was to disengage itself from all PEs in competitive sectors where the private sector was able to assume the role of operating enterprises, to regulate effectively public sector monopolies, and for those PEs likely to remain in the public sector for the foreseeable future, to strengthen the management autonomy of PEs and to clarify the relationship between the PEs and Government. In competitive sectors, viable PEs were to be transferred to the private sector as fast as possible, while non-viable PEs were to be liquidated. This objective was driven by practical concerns: the financial and managerial resources necessary to support unsuccessful PEs and the resources required to enable efficient PEs to expand and exploit new opportunities strained scarce budgetary resources. In the non- competitive sectors and for those PEs unlikely to be divested in the medium or long term, regulations were to be codified in performance contracts (contrats-programmes) so as to establish specific performance targets for the PEs and tariff or other forms of support to be supplied by the Government, with a view to ensuring that the PEs would operate on a financially sound and sustainable basis. 2.7. Prior to 1989, Government's performance in divesting itself of PEs through sale or liquidation and in establishing stronger regulations for those PEs to remain under state ownership was modest. Aside from passing the law to decrease the number of PEs under its supervision, Government had sold part or all of its shares in 6 PEs and had taken decisions to do so on 8 more; it had liquidated a small number of firms and had plans to liquidate 6 more. The Government's ability to carry out the decisions was stalled, however, by the lack of funds in the Public Enterprise Restructuring Fund, which was set up to finance the cost of staff lay-offs and debt repayments associated with the liquidations and divestitures. In addition, the process of privatization was lengthy and complex, with the initiative coming from individual PEs and/or their supervising ministry, and with a number of inter-ministerial committees involved. 2.8. Progress was also slow on strengthening the supervision of PEs, shifting from ex ante controls to ex post evaluation of performance, and clarifying the respective roles of the state and the PE through the establishment of performance contracts. Good data were required to monitor performance and forecast budgetary requirements for PEs; the availability of such data was limited and in turn, depended on inadequate PE management information and accounting systems and internal controls. Thus, PEs had to be strengthened internally in all these areas before realistic corporate plans or performance contracts could be drawn up. The few efforts at restructuring that had been made, for two cement companies and a paper mill, concentrated mainly on alleviating short-term problems. B. PUBLIC ENTERPRISE REFORM LOAN (PERL) 3. Loan origin and genesis 3.1. An Initial Project Brief for a Public Enterprise Restructuring Loan was issued in March 1987. At that stage, the support envisaged was in the form of a fast-disbursing loan to accompany reorganization and restructuring of a selected number of heavily loss-making PEs in the mining, chemical and transport sectors. The PEs were inter-related: la Compagnie des  4 Phosphates de Gafsa (CPG) mined the phosphates that were used by the Groupe Chimique (GC, consisting of five fertilizer companies) to produce fertilizers; both the phosphates and the fertilizers were in turn transported by the Societe Nationale des Chemins de Fer Tunisiens (SNCFT -- the railway). Together they had received, during the 1982-86 period, about 33% of the total transfers to PEs from the Government. 3.2. With the concurrence of Government, the design of the proposed operation was soon broadened by the Bank to include: (i) addressing reforms in the Government's institutional framework and organization for monitoring the performance of the PE sector; and (ii) support for Government's privatization and restructuring efforts of PEs.4 3.3. Progress on processing the loan was based on a public enterprise sector note, sent to Government in November 1987 and a sector mission in December 1987 to further the discussions with the Tunisian Government on Bank assistance in this sector. Preparation of the loan started in earnest with a preparation mission in February 1988. At that time, the main objectives and components were established that are to be found in the final design of the PERL. They covered: (i) the institutional and regulatory framework for the process of both privatization of PEs and restructuring of those PEs that were to remain in the medium or long- term in the public sector; (ii) support for the process of privatization and restructuring of PEs; this included the use of counterpart funds from Bank funds to finance the Public Enterprise Restructuring Fund (Fonds de Restructuration des Entreprises Publiques) to cover payment of liabilities and compensation packages for laid-off staff; and (iii) specific action plans for the largest loss-making enterprises mentioned above: CPG, GC, and SNCFT. 3.4. Between preparation in February 1988 and appraisal in February 1989, there were several pre-appraisal missions that focussed on identifying and understanding the strengths and weaknesses of the regulatory, legislative, institutional, financial, and technical framework of: (i) Government monitoring of PEs; (ii) the privatization and restructuring process; and (iii) the three large PEs to receive particular attention under PERL. At pre-appraisal in July 1988 the basic conditionality was set for appraisal, negotiations, Board presentation, effectiveness, and second tranche release of the loan, which were formalized in the Initiating Memorandum (IM) of December 1988. The basic condition proposed for negotiations was the passage of a new law on PEs, incorporating a number of features, such as reducing the number of PEs under the direct supervision of Government, clarifying the nature of Government's supervision of PEs, reforming Boards of Directors of PEs, and simplifying the procedures for approving restructuring and divestiture proposals. Board presentation conditions covered adequate budgetary provisions for 1990 for carrying out an agreed program of PE divestiture and restructuring, agreement with Government on the overall strategy for the three PEs, CPG, GC, and SNCFT, and signature by Government and the concerned PE of performance contracts for CPG and SNCFT. The IM proposed an unusual approach for second tranche release: gradual releases of small amounts of 4/ In recognition of these broader objectives, the name of the loan was changed slightly to Public Enterprise Reform Loan (PERL).  5 funds were to be based on overall progress on divestiture and restructuring of PEs. The amount of the proposed operations was between US$100 and US$130 million. 3.5. The review of the proposed operation by the Operations Committee in January 1989 raised a number of questions on the adequacy of the conditionality, since the release of the first tranche was conditional only on passage of a law and agreement on strategy, rather than any actions. In addition, the Committee's reaction to the proposal for gradual release of the second tranche was not positive, with comments to the effect that such a design no longer had the impact of a fast-disbursing loan, and leaving the impression that Government would be taking piecemeal steps toward reform. The committee noted that it would be preferable to set a minimum percentage of assets in the PE sector to be privatized for release of the full amount of the second tranche. 3.6. Between appraisal in February/March 1989 and negotiations in May 1989, Government passed the new law as well as all the key implementing decrees on the PEs, which established, inter alia, a new, narrower definition of public enterprises, thus reducing the burden to Government of supervision and oversight of PEs. Since the passage of this law had been originally proposed as a condition of negotiations, the Bank considered that no further specific conditionality on the legal/regulatory aspect of public enterprise sector reform was necessary. On the second major component of the PERL, divestiture, and in spite of the highly politically sensitive nature of privatization, the Government did bring to negotiations, as requested, an indicative list of PEs to be privatized, with budgetary implications for their privatization. The Bank did not set, however, minimum quantitative targets for privatization as a condition for tranche release: the reason given by the Bank was that progress realized on privatization did not depend exclusively on Government's actions, but also on the response of private sector interests. The third major element of the operation involved developing satisfactory action plans for major loss-making PEs. Given the complexity of establishing realistic performance objectives, however, the condition for Board presentation of signature of two performance contracts was changed to a condition of effectiveness of approval of three performance contracts by the supervising ministry and the concerned PE (para. 4.4). Their signature and satisfactory progress in implementation was a condition for second tranche release. Agreed performance indicators to be included in the performance contracts for the three PEs were contained in an annex to the minutes of negotiations, as were elements of an action plan for a fourth PE. The loan amount finally agreed was fixed at the upper end of the range, US$130 million, with two tranche releases. Cooperation with other Agencies 3.7. The Bank requested, in August 1988, a Japanese grant for about US$2.5 million equivalent, to finance technical assistance whose main objective would be to carry out diagnoses of priority PEs5 and to enable them to prepare performance contacts or business plans, as t/ Throughout the report, there are refezences to "priority PEs", or "selected' PEs. The terms "priority" and 'selected" are not defined in the documentation for the loan or in the files.  6 necessary. Ultimately, a Japanese grant for 270 million Yen, which at the time was the equivalent of about US$2.0 million, was granted in February 1990 for this purpose. In addition, during the implementation of the PERL a second Japanese Grant was put in place, in 1992, for US$800,000 equivalent, to finance technical assistance to study privatization prospects in three sectors: cement, dairy and ceramics industries. 3.8. Both the African Development Bank and the European Investment Bank provided loans for US$40 million equivalent for CPG and ECU 17 million for SNCFT, respectively, to finance priority investments and maintenance. These loans were put in place in coordination with the Bank and the timing of the PERL. 3.9. The United States Agency for International Development (USAID) financed technical assistance to the Directorate for Public Enterprises, one of the key Tunisian agencies involved in privatization, to assist in the privatization program, as well as to assist in the development of capital markets. 4. Loan Objectives and Description 4.1. As presented in the President's Report, the PERL aimed to help Government in its program of public enterprise reform. The program consisted of privatization of public enterprises operating in areas where the authorities no longer saw a need for public ownership, and the establishment of a legal and regulatory framework to lead progressively to quasi- commercial operating and management conditions in those enterprises which were to remain in the public sector. The underlying motivation for the reform program was to enable the Government to allocate its scarce human and financial resources more efficiently; to clarify the expectations and obligations of Government vis-a-vis its public enterprise sector; and to introduce greater transparency in the relationship between the Government and its public enterprises. 4.2. The three key areas of PE reforms targeted under the PERL, as presented in the President's Report, were: (i) system-wide reform, including legal and institutional reforms, performance contracts, PE management improvements, and budgetary and financial reforms; (ii) a program of divestiture and restructuring, including progressive divestiture of Government holdings in PEs in competitive areas of the economy, restructuring priority PEs in the public sector, and addressing the social costs of divestiture and restructuring; and (iii) sub-sectoral action programs for selected priority PEs: CPG, Groupe Chimique, and SNCFT.  7 4.3. The loan had two tranches: the first of US$70 million, to be released at loan effectiveness, expected to be in December 1989, and the second of US$60 million, expected to be released around December 1990. Loan conditionality 4.4. The four conditions for loan effectiveness and release of the first tranche were: (i) submission to the Bank of satisfactory performance contracts for CPG, SNCFT, and Societe Nationale de Transport (SNT -- the Tunis bus company), approved by the supervising ministry and the enterprise concerned; (ii) a list, satisfactory to the Bank, of additional PEs to prepare and sign performance contracts; (iii) presentation for sale by Government of a group of PEs or Government equity holdings in PEs selected on the basis of criteria satisfactory to the Bank; and provision to the Bank of a satisfactory program for a further group of candidates proposed for subsequent privatization; (iv) allocation to the Public Enterprise Restructuring Fund of an amount at least equivalent to US$70 million. 4.5. The conditions for the release of the second tranche release followed from the conditions on the first: (i) Approval by the Inter-ministerial Commission for Restructuring and Divestiture (Commission d'Assainissement et de Restructuration des Entreprises Publiques -- CAREPP), signature by the responsible ministry, and satisfactory implementation of key actions in the performance contracts of CPG, SNCFT, and SNT; (ii) Approval by the CAREPP and the signature of the sectoral ministry of performance contracts of an agreed group of additional PEs; (iii) Government will bring to sale a further group of PEs or holdings satisfactory to the Bank;6 and (iv) Commitment by the Public Enterprise Restructuring Fund of at least half of the first allocation (US$35 million) for purposes of divestiture and restructuring PEs, and allocation of at least an additional US$60 million to the Fund. 6/ This does not imply that the sale of the first group of PEs would necessary have been completed.  8 The expectation was that the fulfillment of these conditions would reduce the burden on Government of PE monitoring through the implementation of the new PE legislation and through divestiture; put in place the appropriate framework for PE monitoring; and have an important demonstration effect for development of clear and monitorable performance contracts throughout the PE sector. The implementation of this phase of the reform program was to set the stage and provide the momentum for continued privatization and for formulating and supervising effective performance contracts on a regular basis with priority PEs. The impact would be a sustainable improvement in the efficiency and profitability of the PEs and a reduced burden on Government resources. Perfornance contracts and action programs for PEs Compagnie des Phosphates de Gafsa (CPG) 4.6. The major actions and specific indicators to be included in the performance contract for CPG over the period 1989-1991 included: (i) closure of the most uneconomic mines; (ii) a decrease in employment of about 630 workers (by end 1991) and re-deployment of another 725 (by end 1992); (iii) the principle that investments would be based on a least-cost approach; (iv) Government financing to avoid excessive short-term borrowing and to ensure that at least 40% of the investment program would be self-financed; and (v) a series of technical and financial indicators to be met and an indicative investment program for the years 1989-92. Societe Nationale de Chemins de Fer Tunisien (SNCFT) 4.7. The performance contract for SNCFT was to be established in the context of an overall policy for the transportation sector, as reflected in Government's sectoral development policy letter: the private sector was to be allowed to operate, users were to be allowed freedom of choice of transport mode, and tariffs were to be aligned over the medium term with long-term marginal costs of each mode of transportation. Agreements on performance criteria, changes in the tariff structure, and criteria for ensuring economic investments for SNCFT were the consequence of this sectoral policy. The performance contract for SNCFT was to include the following elements: (i) the investment program of SNCFT was to be limited over the Seventh Plan period (1987-92) to DT 163 million, and all investments were to have an economic rate of return of at least 10%; (ii) a reduction of the work force by 150 workers per year and an annual increase in productivity (quantified in a table) in 1989 and 1990; (iii) tariffs were to increase by at least 2% per year in real terms; (iv) the Government was to ensure compensation to SNCFT for the difference between the tariffs and the long-term marginal cost for each category of traffic. Several target financial ratios were also specified for 1989 and 1990. Soci&e Nationale de Transport (SNT) 4.8. Again within the context of agreement on a sectoral strategy for the transport sector, the performance contract for the Tunisian bus company, SNT, contained the following elements: (i) limitation on investments to about 800 buses; (ii) identification of bus lines to be ceded to  9 private companies; (iii) a decrease in employment to achieve a specified productivity standard; and development of a staff action plan for training, re-deployment, retirement and recruitment; (iv) an increase in specified productivity measures; (v) commitment to develop a cost accounting system; (vi) system of annual revision of tariffs to ensure, at a minimum, the 1989 rate of cost recovery, in real terms; (vii) Government commitment to subsidize SNT to ensure financial equilibrium and adequate funding of their investment program. Groupe Chimique 4.9. Groupe Chimique consists of five fertilizer companies, whose financial performance deteriorated with the precipitous decrease of fertilizer prices in the 1980s. Because of the importance of foreign governmental partners in Groupe Chimique, the need to consolidate the five fertilizer companies and to undertake major financial restructuring, no performance contract was envisaged under the PERL. In addition, although the measures mentioned below were all to be undertaken by the time of the release of the second tranche of the PERL, none of the actions were formalized as legal covenants. The Tunisian Government was to seek agreement from their foreign partners for the merger of four of the companies in order to rationalize the organization of Groupe Chimique' s operations. At negotiations the Bank was informed that such agreement had been reached with one of the partners and was being sought from the other foreign partner. Progress on the proposed merger was to be reviewed prior to second tranche release. In addition, agreement was to be reached by the second tranche release on a satisfactory financial restructuring plan for 1990-1991 to reduce short-term debt and to prevent future arrears accumulation. Third, because pollution from the fertilizer plants was considered an important issue, satisfactory progress was to be achieved on the implementation of anti- pollution investments for gypsum and sulfur dioxide effluent. Finally, because of the importance of the transfer price for phosphate between CPG and Groupe Chimique,7 agreement was reached (at negotiations) that the 1989 price would be based on CPG's export price, adjusted for (poorer) quality, as an interim measure while the new contract between CPG and Groupe Chimique was being negotiated. Other actions to be monitored for second tranche release 4.10. Aside from the legal conditions (paras. 4.5-4.8) and progress on Groupe Chimique's action plan (para. 4.9) to be monitored for the second tranche release, the Bank was to review the impact of the new public enterprise legislation on a selected group of PEs to ensure that Government involvement in day-to-day operations had been reduced and that progress on preparing performance contracts was satisfactory. In addition, data were to be furnished to the Bank covering financial and economic performance of all PEs based on audited financial 7/ Eighty percent of CPGs phosphate production is transformed locally by Groupe Chimique, and the transfer price had been based on the international price until the agreement had lapsed in 1988. Since then, the foreign partners of Groupe Chimique were exerting pressure on the Tunisian Government to obtain a moie favorable transfer price. The objective of insisting on using the international price as a basis for calculating the transfer price was to ensure a commercial orientation of the PEs.  10 statements, budgetary and extra-budgetary support to PEs, PE debts and arrears to Government and arrears towards other PEs, a list of performance contracts approved the supervising ministries and the CAREPP, and all decrees and circulars related to the new law on PEs (See Table 7 in Part III for a complete list of information to be furnished to the Bank). In addition, technical assistance was to be hired to improve MIS, accounting, and internal controls in a list of priority PEs. Disbursement and Procurement Mechanisms 4.11. The total loan proceeds of US$130 million equivalent were to be disbursed in two tranches. The first one of US$70 million was to be released on loan effectiveness and the second one of US$60 million was to be released contingent on satisfactory progress in carrying out the public enterprise reform program as spelled out in the Government's sectoral development policy letter, the loan document, the minutes of negotiations, and the Bank's Memorandum of the President. Financing was provided for all imports, except those financed by other sources and those on a negative list that included military or para-military goods, nuclear reactors and parts, and luxury goods such as tobacco, precious stones, jewelry and gold. Petroleum products could be financed up to a total of US$26 million and foodstuffs up to a total of US$26 million. 4.12. International competitive bidding (ICB) was required for all contracts of US$5 million equivalent or more; all other imports were to be procured following the usual commercial practices of the purchaser. Disbursements were to be made against statements of expenditures (SOEs), except in the case of purchases made under ICB, which were to be fully documented. The Central Bank was responsible for the administration of the loan, with a Special Account to be set up at the Central Bank with an initial deposit of US$20 million. 5. Loan implementation 5.1. Effectiveness. The PERL became effective on February 22, 1990, about two months behind the schedule anticipated at the time of Board presentation. The reasons for the delay are not entirely clear from the files. The Bank had numerous substantive comments on draft performance contracts for SNCFT and SNT that had been submitted to the Bank after negotiations. Many parameters in these drafts differed considerably from those agreed at negotiations. A supervision mission in January 1990 reported that progress on the other conditions was good, however, and although the two performance contracts still had many weaknesses, their preparation was progressing. On February 5, 1990 the Bank informed the Government that the substantive conditions for loan effectiveness had been met and as soon as the legal opinion on the legal documents was received from the Government, the loan could be declared effective. No waivers to the legal agreement were required. 5.2. Second tranche release. The release of the second tranche, originally expected to be around December 1990, took place in January 1993, about twenty five months behind schedule. Several factors contributed to this delay, the most salient of which had to do with the  11 performance contracts. The other legal conditions for the second tranche release were met by September 1991.8 The performance contracts for CPG, SNCFT, and SNT were approved by the CAREPP only in April 1991, but never signed. For the additional group of PEs, satisfactory performance contracts were submitted to the Bank only in January 1992. In accordance with the conditionality spelled out in the Loan Agreement and elaborated on in the President's Report, the Bank insisted that these performance contracts be signed, and evidence of their signature was received in July 1992. At that point, the Bank sought evidence either that the performance contracts of the first three PEs, while never signed, had been substantially implemented, or that new acceptable performance contracts for these three PEs for the next period, 1992-94/96, were prepared and signed. For CPG, the Bank found that the elements in the performance contract had been substantially met except for certain areas, where exogenous factors had played a role (para. 6.3). For SNCFT and SNT, the Bank found that their performance fell short in critical areas, both because of the PEs' own performance and because of Government's unwillingness to undertake financial restructuring and provide adequate compensation for annual losses. After input from Bank-supplied consultants, satisfactory performance contracts for SNCFT and SNT for the 1992-96 period were approved by CAREPP and signed by the supervising ministry in late December 1992. The loan closing date was extended twice, by one year each time, to June 30, 1993, to allow the legal conditions on performance contracts to be met and disbursements to be completed. 5.3. Disbursements. The first tranche of US$70 million was fully disbursed by June 30, 1991, about six months behind schedule. The initial US$20 million disbursement was made into the Special Account, as foreseen. Because of the delay in releasing the second tranche, however, the loan was not fully disbursed until June 30, 1993, about two years behind schedule. 5.4. Audits. Audits were received for 1990, 1991 and 1992, with about three to six months' delay. The audit report for 1993, covering the special account and the use of the statements of expenditures, was received (ahead of schedule) in November 1993. All original documentation supporting the SOEs for the period 1993 was mistakenly sent by the Central Bank of Tunisia to the Bank, and no copies were available at the Central Bank, so the auditor was unable to issue an opinion. As of this writing, the Bank had returned the documentation to the Central Bank and requested that another audit be carried out for disbursements during 1993.