Tunisia - Energy Conservation Demonstration Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16817 PERFORMANCE AUDIT REPORT TUNISIA ENERGY CONSERVATION DEMONSTRATION PROJECT (LOAN 2735-TUN) July 2, 1997 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Currency Equivalents (annual averages) Currency Unit = Tunisian Dinar (TD) 1990 US$1.00 TDO.87 1991 US$1.00 TDO.92 1992 US$1.00 TDO.88 1993 US$1.00 TD1.00 1994 US41.00 TD1.01 1995 US$1.00 TDO.94 1996 US$1.00 TD1.00 Abbreviations and Acronyms AME Agence de Maitrise de 'Energie GNP Gross National Product GOT Government of Tunisia GTZ Gesellshaft flir Tecnische Zusammenarbeiten (Germany) IBRD International Bank for Reconstruction and Development ICR Implementation Completion Report OED Operations Evaluation Department PAR Performance Audit Report PV Photovoltaic RE Rural Electrification STEG Soci6t6 Tunisienne de Gaz et d'Electricit6 SWH Solar Water Heater TOE Tons of oil equivalent Fiscal Year Government: July 1 - June 30 Director-General, Operations Evaluation Mr. Robert Picciotto Director, Operations Evaluation Department : Mr. Roger Slade for Ms. Elizabeth McAllister Division Chief Mr. Yves Albouy Task Manager Mr. Richard Berney FOR OFFICIAL USE ONLY The World Bank Washington, D C. 20433 U.S.A June 25, 1997 Office of the Director-General Operations Evaluation MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Tunisia Energy Conservation Demonstration Project (Loan 2735-TUN) Attached is the Performance Audit Report (PAR) prepared by the Operations Evaluation Department (OED) on the above project, approved in FY87. The total amount of the Loan was US$4 million, which was almost fully disbursed. The objectives of the loan were to: (i) develop an effective institutional and policy framework for a national energy conservation program; (ii) strengthen local capabilities for implementing energy efficiency audits and conservation actions; and (iii) finance short-term actions to demonstrate energy conservation in three energy-intensive sectors of the economy. In addition to strengthening Agence de Maitrise de l'Energie's technical capacity, the project components included energy audits, feasibility studies and the financing of short term energy savings actions in the industrial, transport and commercial buildings sectors. Implementation suffered extensive delays and the energy conservation and training objectives were only partially achieved. Because funds were not being used for energy conservation investments, a solar water heating demonstration project in a hospital was added, and the project scope was changed, informally, to support Tunisia's rural electrification program using solar photovoltaic (PV) technology for over 1500 isolated farm households with PV systems. About half the loan amount was allocated to the new project component. Project closing was delayed by four years to ensure the full utilization of the loan amount. The solar water heating demonstration was poorly executed, and has not met its demonstration goals. The solar PV component was broadly successful in technical terms, but, because the user pays only 10 percent of the investment cost and none of the maintenance costs, it is not replicablefinancially and is not in conformity with Bank policies on cost recovery in the electricity sector. OED rates the overall project outcome as unsatisfactory because the major part of the loan was spent on project components that did not meet the Bank's requirements for sector policy objectives or economic viability. The institutional development impact is rated as modest, sustainability as likely, Bank performance as highly unsatisfactory and Borrower performance as unsatisfactory. The ICR rates the outcome as satisfactory, sustainability as uncertain, institutional development impact as partial, and Bank and Borrower Performance as deficient. 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. 2 The significant lessons from the project are: (i) small loans are justified only for projects with clear cut, limited objectives that can be implemented in a short time frame; (ii) when executing agencies have no financial responsibility for cost recovery and repayment of Bank loans, there is a high risk of sub-optimal use of these funds; (iii) solar PV systems are a viable technical and economic alternative to other energy sources for a disbursed rural population; and (iv) rural solar PV programs should provide consumers with choices in system capacity and cost. Attachment Contents Preface ...................................................... 3 Basic Data Sheet ............................................... 5 Evaluation Summary ....................................... ..... 9 1. Project Background and Description ....................... ........ 13 Introduction .......................................... ....... 13 Project Objectives ..................................... ........ 13 Project Description ..................................... ....... 13 2. Implementation .............................................. 15 ICR Summary ......................... ................ 15 ICR Assessment ................................................. 15 Rural Electrification .....................6...... ...............16 Solar Water Heating Component ........................... ........ 16 3. Audit Assessment ..................9.........................19 Energy Audits ......................................... ...... 19 Energy-saving Measures and Investments ............9.... .............19 Solar Water Heating ...................... ............... 20 Solar Photovoltaic Systems for Rural Electrification .................. 20 Financing Arrangements for Solar PV Systems .................. ........ 21 4. Issues ..................................................... 23 Use of Loan Funds to Finance Subsidies .............................. 23 Procurement .................................................... 23 Bank Performance ..................... ................ 23 5. Overall Assessment .............................................. 25 6. Lessons .............................................. ..... 27 Annex Comments from the Borrower .......................................... 29 This report was prepared by Mr. Richard Berney (Task Manager) and Mr. Sunil Mathrani (Consultant) who audited the project in February 1997. Ms. Helen Watkins provided administrative support.  3 Preface This is a Performance Audit Report (PAR) on a loan to Tunisia for an Energy Conservation Demonstration Project (Loan 2735-TUN), in the amount of US$4.0 million, which was approved on July 3, 1986 and became effective March 17, 1987. The original closing date of June 30, 1990 was extended by four years, until June 30, 1994. The loan was fully disbursed. This PAR is based on the Project Implementation Report prepared by the Middle East and North Africa Region, issued April 28, 1995, the President's Report, loan documents, the transcript of the Executive Directors' meeting at which the project was considered, on a study of project files, and on discussions with Bank staff. An OED mission visited Tunisia in February 1997 and discussed the effectiveness of the Bank's assistance with the Agence de Maitrise de I'Energie (AME), the project implementing agency. The kind cooperation and valuable assistance of AME management and staff is gratefully acknowledged. The Implementation Completion Report (ICR) provided only a limited account of the project experience, and lacked any discussion of the implementation of the solar photovoltaic rural electrification component, for which half of the loan was allocated. The PAR provides an account and assessment of the project experience for this component, as well as the experience for a hospital solar water heating demonstration project, both of which were added on to the original project scope. Following standard OED procedures, this draft PAR was sent to the Borrower for comments on April 30, 1997. The comments received from the Ministry of Industry are reproduced as Annex 1 of the PAR.  5 Basic Data Sheet ENERGY CONSERVATION DEMONSTRATION PROJECT (LOAN 2735-TUN) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as percent estimate current estimate of appraisal estimate Total project costs 6.25 5.48 87.68 Loan amount 4.00 3.97 99.25 Cofinancing 0.00 0.00 0.00 Closing Date 6/30/90 6/30/94 Economic rate of return NA NA NA Cumulative Estimated and Actual Disbursements FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 FY95 Appraisal estimate 1.00 2.20 3.20 4.00 4.00 4.00 4.00 4.00 4.00 (US$M) Actual (US$M) 0.00 0.01 0.13 0.41 0.74 0.83 0.92 1.41 3.97 Actual as percent 0.0 0.5 4.1 10.3 18.5 20.8 23.0 35.3 99.3 of appraisal Date of final disbursement: 11/07/94 Project Dates Original Actual Identification 03/84 Preparations 12/84 Appraisal 01/86 Negotiations 05/86 05/86 Board presentation 06/86 07/03/86 Signing 07/86 07/86 Effectiveness 09/86 03/17/87 Project completion 06/30/90 06/30/94 Loan closing 06/30/90 06/30/94 6 Staff Inputs (staff weeks) FY86 FY87 FY88 FY89 FY90 FY91 Preappraisal 15.3 Appraisal 21.5 Lending Dev. 10.9 .8 Negotiations 5.8 .6 Supervision .3 8.1 7.8 12.0 4.6 3.3 Completion Other .2 .7 Total 54.0 10.2 7.8 12.0 4.6 3.3 Staff Inputs (staff weeks) FY92 FY93 FY94 FY95 Total Preappraisal 15.3 Appraisal 21.5 Lending Dev. 11.7 Negotiations 6.4 Supervision 5.5 7.4 .2 49.2 Completion .3 .3 Other .9 Total 5.5 7.4 .2 .3 105.3 7 Mission Data Performance rating' 2 Date No. of Staff days Specializations Imple- Develop- Types ofproblems (month/year) persons infield represented mentation ment Identification/ 10/84 3 10 Preparation 12/84 2 15 02/85 1 10 Limited technical and 03/85 2 10 managerial capabilities 09/85 1 8 of SME/AME 12/85 1 2 Appraisal 01/86 4 21 Supervision 09/87 1 4 E NA NA Delays in implementation 09/88 4 14 E (3), F 1 1 Lengthy approval procedures 07/89 1 4 E 1 1 No TOR or BTO in files. 10/89 1 4 E 1 1 Bankruptcy of supplier; AME's complicated procedures 12/89 5 11 E 2 2 Problems as above 03/90 7 17 PE 2 2 Problems as above 10/90 2 4 E 2 2 Problems as above 05/91 1 4 E 3 2 Problems as above 01/92 2 4 E, PE 2 2 Problems as above 02/93 1 7 PE 2 2 Procurement, resolved Completion 11/94 0 0 U S 1. Prior to FY94, Performance Ratings are for Overall Status and Project Development Objectives. 1 = Problem Free or minor problems; 2 = moderate problems; 3 = major problems. 2. E = Energy specialist (includes demand management, energy planning, conservation); EC = Economist; PE = Power/energy engineer; F = Financial specialist; LO = Loan offier; NS = not specified.  9 Evaluation Summary Introduction 1. By the mid-1980s, when this project was designed, it was clear that energy demand would outstrip indigenous supply by the early 1990s. The desire to maintain energy self- sufficiency for as long as possible was the main motivating factor for the Government of Tunisia (GOT) decision to pass an energy conservation law in 1985. This law made energy audits mandatory for large users and led to the establishment of a public body, AME, in charge of energy conservation and the promotion of renewable energy. AME was the executing agency for this project. In 1996 it had about 70 staff and an annual operating budget of 1.3 million TD (US$1.3 million), funded entirely by GOT. Objectives and Description 2. The objectives of this project were to: (i) help the AME develop an effective institutional and policy framework for a national energy conservation program; (ii) strengthen local capabilities for implementing energy efficiency audits and conservation actions; and (iii) finance short-term actions to demonstrate energy conservation in three energy-intensive sectors of the economy. Simply stated, the project sought to put the Government's 1985 energy conservation policy into practice. As such, its objectives were clear and simple. 3. Initially the project had three components: (i) an institutional development component provided technical assistance to AME; (ii) equipment for and training in basic energy auditing and energy conservation; and (iii) demonstration and direct energy conservation activities, comprising energy audits, short duration energy savings actions and retrofitting investments, and on-the-job training in energy conservation for the audited enterprises. The enterprises were to be in (a) the industrial sector (about 12 manufacturing enterprises); (b) the transport sector (about 8 transport companies); and (c) the hotel/commercial building sector (about 8 hotels and buildings). 4. The project was explicitly described as a "demonstration" project, with a significant portion of its benefits to the economy expected to come as indirect impacts from educating and encouraging other energy users to undertake similar measures on their own. However, difficulties in utilizing the available funds led to a major change in project scope during implementation. Some of the loan was used for a demonstration project to use solar heating for hot water supply and about half the loan amount was allocated to a rural electrification program using solar photovoltaic (PV) technology. Although this component was unrelated to the original project scope and objectives, the project was never formally restructured to incorporate it. Implementation 5. The energy conservation objectives were only partially achieved. The number of energy audits conducted under the project was fewer than planned and some of the audited firms did not implement the audit recommendations. The institutional objectives were partially achieved since 10 training in energy auditing was carried out and the auditing program increased conservation awareness and helped to strengthen local energy auditing capacity. 6. When funds were not used as initially envisaged, for implementing the recommendations of the energy audits, the Bank, in December 1989, agreed to extend the closing date and revise the project scope to finance other demonstration actions, including the installation of a solar water heating system for a hospital and household solar photovoltaic system for rural electrification. Solar PV systems were installed in 1245 isolated households in rural areas far from the electricity grid. These households have been equipped with two fluorescent lights and a socket for small appliances. The systems have been in service for about two years and are working well. 7. The revised scope also financed the installation of a solar hot water system in the Gafsa regional hospital in late 1994. This was intended to be a demonstration project for large institutional buildings (such as hotels) to showcase the advantages of such systems. However energy savings have been far less than projected and the installation has suffered from technical problems arising from underutilization . 8. This revised investment program also suffered long delays and was only implemented in 1993 and 1994. Three more extensions were granted before the loan closed on June 30, 1994, almost fully disbursed. Findings 9. The Bank-financed audits collectively proposed energy-saving investments that exceeded the remaining loan funds. However, the audited firms had access to other funding sources with attractive lending terms and less cumbersome procurement procedures, with the result that AME had great difficulty in finding suitable outlets for the IBRD funds. Nevertheless, after much delay, AME succeeded in placing US$0.5 million of the Bank loan in the transport and public buildings sectors. Total annual energy savings of about 2000 tons of oil equivalent (TOE) can be attributed to these subprojects. 10. The homes which have been electrified are often 5-10 kilometers from the nearest village. Once the decision to electrify these households had been taken, solar photovoltaic systems were the most appropriate technical choice to serve small, isolated consumers. The 1245 systems installed are all operating satisfactorily and the beneficiaries obtained a net improvement in their living standards as a result. However, GOT decided to opt for a subsidized PV program, in which the beneficiaries would pay only a small initial down payment of about 10 percent of the investment cost. The large subsidy has led to rationing because demand for the systems now outstrips supply. GOT has also taken on the responsibility for the recurrent cost burden for maintenance of the system, which is to be provided free of charge. Overall Assessment 11. OED rates the overall project outcome as unsatisfactory, primarily because the major part of the project funding was spent on project components that did not meet the Bank's 3 In its comments on the draft of this report, AME states that these problems have now been resolved. 11 requirement for sector policy objectives or economic viability. GOT's approach to solar PV systems is not consistent with the Bank's policy on rural electrification. GOT's subsidies to the PV program should be financed from current income from taxation, not from borrowings, as was the case for this project. It would have been more appropriate for the beneficiaries to be required to pay for the maintenance costs in full. Bank guidelines for lending in the power sector require that rural electrification (RE) projects be structured to recover recurrent costs, even if they provide subsidies of the initial capital costs. The reason for this policy is that Governments should not commit themselves to recurrent financial burdens that they may not be in a position to support. Furthermore, the beneficiaries' share in the investment cost is extremely low and the high subsidies to them have resulted in a rationing system for new systems based on waiting lists, without a transparent allocation system. A smaller subsidy with less restrictive eligibility criteria would have allowed more unelectrified households to have benefited sooner from this technology. 12. The Gafsa Solar Water Heater project was a poor candidate for demonstrating the advantages of investing in solar water heating systems. It is highly unlikely even to cover its initial costs during its economic life. The fact that the subproject was approved just weeks before final loan closure suggests that the Bank tacitly agreed with AME's approach of maximizing disbursements from the loan, essentially without regard to financial and economic criteria. The ICR rates the outcome as satisfactory, and it assumes that the solar PV and the Gafsa SWH components were appropriately designed. 13. The project's energy audit component was a partial success and there has been substantial follow-up action taken by the audited enterprises to save energy, even though the Bank directed credit for this purpose went unutilized. The potential for savings is far from exhausted, particularly in the industrial sector, but low real energy prices and high interest rates have lengthened payback periods and discouraged further investments. 14. The loan should have been canceled at the end of the energy audit program in 1991. By then, AME had benefited from the institution building and training aspects of the project and local consultants had also gained useful skills and experience in energy auditing. As it turned out, the bulk of the energy savings attributable to the project were achieved without any Bank participation in their financing. The repeated extension of the closing date enabled AME to pursue non-project related objectives without the need to seek budgetary resources from GOT, and cost the Bank four more years of supervision and loan management expenses. 15. The solar PV component was, for the most part, successfully implemented and sustainable in technical terms, despite the equipment problem experienced (and as yet unresolved) with one consignment. However, this component is not financially replicable without substantial direct subsidies for both the investment and recurrent costs. 16. The project contributed directly to the institutional strengthening of AME, to local capacity building in energy auditing and conservation techniques, and indirectly to the creation of small enterprises to service the needs of the solar PV program. The audit, therefore, assesses the institutional development impact of the project as modest, which is consistent with the "partial" rating of the ICR. Project sustainability is rated as likely. 17. The overall performance of the Bank is rated as highly unsatisfactory. The non- observance of several important Bank operating practices under this project is disturbing. The 12 solar PV component, which consumed half the project funding, was neither an energy conservation operation nor a demonstration project, and was not evaluated with the thoroughness normally required. The Audit could find no evidence that senior Bank management reviewed the redesigned project. The ICR has a similar rating, but for reasons related to supervision and design of the original project. Borrower performance is rated as unsatisfactory, as it was in the ICR. Lessons Learned The key lessons from this project are: * Small loans are justified only for projects with clear cut, limited objectives that can be implemented in a short time frame. * In revenue earning sectors, executing agencies should have some financial responsibility for cost recovery and repayment of Bank funds. Otherwise there is a high risk of sub-optimal use of these funds. * Solar PV systems are a viable technical and economic alternative to other energy sources for a dispersed rural population. However, these systems are expensive, and affordable only to middle-income populations. * Rural solar PV programs should provide the consumer with a choice of more than one system capacity (watts), at different costs. This would enable systems to be sized according to the varying needs and incomes of end-users. * Subsidies high enough to generate waiting lists lead to an inefficient allocation of resources. Providing smaller subsidies to a larger group of people would increase economic efficiency and equity and also lead to a faster dissemination of PV systems. 13 1. Project Background and Description Introduction 1.1 Tunisia is a middle-income economy with a population of 9 million and a per capita Gross National Product (GNP) of $1720 in 1995. The industrial sector accounts for over 30 percent of the Gross Domestic Product (GDP) while the share of agriculture has declined to 18 percent. Primary energy consumption is 0.6 TOE per capita, with the industrial sector accounting for the largest share (35 percent) of consumption. 1.2 In 1995, Tunisia's oil and gas production fell to 4.3 million TOE. With a primary commercial energy demand of 5.4 million TOE, Tunisia became a net energy importer for the first time since oil was discovered in the 1960s. Total primary energy demand has grown at an annual average rate of 3.5 percent between 1985-95, and is projected to grow at an average annual rate of 4 percent p.a. for the coming decade. 1.3 By the mid-1980s, when this project was designed, it was already clear that energy demand would outstrip indigenous supply by the early 1990s. The desire to maintain energy self-sufficiency for as long as possible was the main motivating factor for the GOT decision to pass an energy conservation law in 1985. This law made energy audits mandatory for large users and led to the establishment of a public body, AME, in charge of energy conservation and the promotion of renewable energy. AME was the executing agency for this project. In 1996 it had about 70 staff and an annual operating budget of 1.3 million TD, funded entirely by GOT. Project Objectives 1.4 According to the President's Report, the objectives of this project were to: (i) help the AME develop an effective institutional and policy framework for a national energy conservation program; (ii) strengthen local capabilities for implementing energy efficiency audits and conservation actions; and (iii) finance short-term actions to demonstrate energy conservation in three energy-intensive sectors of the economy. Simply stated, the project sought to put the Government's 1985 energy conservation policy into practice. As such, its objectives were clear and simple. The stated objectives did not envisage any initiatives in rural electrification using solar PV technology. Project Description 1.5 Initially the project had three components: (i) an institutional development component provided technical assistance to AME; (ii) equipment for and training in basic energy auditing and energy conservation; and (iii) demonstration and direct energy conservation activities, comprising energy audits, short duration energy savings actions and retrofitting investments, and on-the-job training in energy conservation for the audited enterprises. The enterprises were to be in (a) the industrial sector (about 12 manufacturing enterprises); (b) the transport sector (about 8 transport companies), and (c) the hotel/commercial building sector (about 8 hotels and buildings). However, while the initial energy audits were to be subsidized by AME, the financial terms for use of the Bank funds onlent to finance the energy-saving investments were on 14 commercial terms and conditions. They had to be repaid in full over seven years at the prevailing Central Bank rate. 1.6 The project was explicitly described as a "demonstration" project, with a significant portion of its benefits to the economy expected to be come as indirect benefits of educating and encouraging other energy users to undertake similar measures on their own. 1.7 Difficulties in utilizing the available funds (paras. 3.3-3.4) led to a major change in project scope during implementation. Some of the loan was used for a demonstration project, funded by AME, to use solar heating for the hot water needs of a hospital, replacing the hotel solar heating demonstration component. However, most of the unutilized funds (about half the loan amount) were allocated to a rural electrification program using solar PV technology. These funds were used to provide over 1500 isolated farm households, located far from the conventional grid, with PV systems for electric lighting and small appliances. Although this component was unrelated to the original project scope and objectives, the project was never formally restructured to incorporate it. 15 2. Implementation ICR Summary 2.1 The ICR notes that training in energy auditing was carried out, and the auditing program increased conservation awareness and helped to strengthen local energy auditing capacity. It also notes that the number of energy audits conducted under the project was fewer than planned and most of the audited firms did use the project's credit line to implement the audit recommendations. 2.2 When funds did not get used, as initially envisaged, for implementing the recommendation of the energy audits, the Bank agreed, in December 1989, to extend the loan's closing date and to revise the project scope to finance other demonstration actions supported by AME, including the installation of household solar photovoltaic systems in rural areas. Three more extensions were granted to accommodate further delays before the loan was closed on June 30, 1994, almost fully disbursed. ICR Assessment 2.3 According to the ICR, the energy conservation objectives were partially achieved, through the audit program. The institutional objectives were partially achieved since training in energy auditing was carried out and the auditing program increased conservation awareness and helped to strengthen local energy auditing capacity. The key factors that interfered with achievement of the project's major objectives were the poor performance of the Government, the Bank, and the implementing agency: Government commitment and support for the project were lacking, AME was not always well-managed, and was insufficiently staffed and trained; and procurement processing was too time-consuming; in addition, the Borrower did not comply with key covenants, particularly relating to the operation of AME. 2.4 The ICR also states that the implementation plan devised at appraisal was unrealistic, given Tunisia's lengthy disbursement profile and AME's inexperience. It states that supervision quality was generally unsatisfactory, that supervision missions consistently overstated the achievement of the objectives and the quality of AME's management and rated the project's progress more highly than warranted, that a shift in task management following the 1987 reorganization adversely affected continuity, and that the lack of a critical eye perpetuated the pattern of delays and extensions. 2.5 However, the ICR, which was based on a desk study of available Bank files, was unable to provide any information on the outcome of the added project components (the solar PV subproject and the hospital solar water heating subproject). Although it opined that the program was expected to have a substantial development impact, it never considered the implications of changing the project scope from energy conservation to rural electrification, and was unable to throw any light on the implementation process, technical success or economic merits of this program. This information was gathered by the Audit Mission, and is discussed below. 16 Rural Electrification Component 2.6 The Bank project financed the installation of solar PV systems in 1245 isolated households in rural areas around the town of Kef. These 100 peak-watt capacity systems were sufficient (in this region) to provide a net daily average usage of 350 (Wh) of electricity. The selected battery (90 amp hours) allows about three days of "autonomy", sufficient for most periods of cloudy/rainy conditions. The cost of these 1,245 systems (including installation) was US$1.2 million, which at about $1,000 per unit, or about US$ 10/watt installed, is at the lower end of the cost range for these sized units.4 2.7 The selected households have been equipped with two fluorescent lights and a socket for appliances like radios, cassette players or TVs. Average daily electricity consumption is estimated to be about 300 Wh, or 100 kWh annually. AME has estimated the average cost of production5 to be about 1.5 TD/kWh (US$1.50 kWh). The systems have been in service for about two years and are working well. 2.8 The Bank loan was also used to purchase a second consignment of 1000 solar PV systems from a different source. The contract (US$0.8 million) was signed a few days prior to the loan's closing date and covered equipment delivery only. AME contracted with a local firm for installation of the systems using its own resources. Installation was halted by AME after only 380 systems had been put in place because it found that the systems had stopped working due to faulty charge controllers. These controllers have been modified by the supplier, but still do not operate satisfactorily. Even though the equipment is under guarantee, AME does not expect that the supplier (who was not the manufacturer) will be able to resolve the problem. It is, therefore, proposing to purchase new regulators from a different source in order to complete installation of the remaining solar PV systems, which have now been in storage for two years. Solar Water Heating Component 2.9 This subproject financed the installation of a solar hot water system in the Gafsa regional hospital in late 1994. It was intended as a demonstration project for large institutional buildings (such as hotels) to showcase the advantages of such systems, in preparation for the implementation of a US$7.3 million GEF grant for solar water heating projects which was approved in October 1994. The implementation of this project is discussed in detail here because it is not covered in the ICR. 2.10 The subproject consisted of a 240 m2 of solar collectors linked to storage tanks with a total capacity of 12,000 liters and a remote monitoring system connected to AME's office in Tunis. The total cost was US$0.2 million, which is much higher than comparable water heating systems elsewhere,6 even after adjusting for "luxury" features such as remote monitoring. 4. See Table 2-1 in World Bank Technical Paper # 324 for international comparisons. 5. Including actual capital costs, estimated 0 & M costs and based on an average 100 kWh annual consumption. 6. According to the GEF Project Document (# 13485-TUN), Section 2.3 of Annex II, this project was 50 percent more expensive (on a cost per sq. m. basis) than comparable projects in Europe. 17 2.11 Energy savings have been far less than projected and the installation has suffered from technical problems arising from underutilization. These problems are the result of oversizing the system relative to demand. First, only 60 percent of the hospital has been hooked up to the SWH system, because funding for connecting the new wing was not part of the Bank-financed project and the Ministry of Health has not yet budgeted the requisite amount. This under- utilization is further aggravated by a second factor: the hospital's daily hot water requirements were seriously overestimated by the feasibility study. The hospital had no hot water meters, so the design was based on average consumption per bed taken from French hospital norms. This has proved to be about double the actual consumption recorded since the SWH system was installed. 7. In its comments on the draft of this report, AME states that the system has "operated perfectly since it was installed", and adds, "It is entirely to be expected that a few minor technical problems would occur, and be immediately resolved; these problems, however, have had no impact on the actual technical performance, which has considerably exceeded the standards guaranteed by the engineer." The problems mentioned in this paragraph refer, primarily, the fact that at the time of the Audit mission (February 1997) the water/glycol mix in the primary circuit was overheating because there is insufficient user off-take of hot water. AME was planning to disconnect some of the solar collectors until hot water consumption increased to an adequate off-take level, and has, apparently, done so by June 1997. 8. The system is designed to deliver 23,000 liters/day. Taking 60 percent for the old wing would mean a daily demand of 14,000 liters, but consumption has averaged under 7000 liters. 4 19 3. Audit Assessment 3.1 This section of the PAR presents information on and analysis of project components that was lacking in the ICR. The coverage of the components outside the original project scope presented here contains material that cannot be found in Bank files on the project and appears never to have been submitted to Bank management during the course of project implementation. Energy Audits 3.2 As part of the project's initial phase, 16 energy audits (instead of 28 projected at appraisal) were carried out at a cost of US$1 million, 80 percent of which was funded by the Bank loan. However, this record must be judged in the context of Tunisia's overall progress in energy conservation activities after the passage of the energy conservation law in 1985. In Tunisia over 150 mandatory energy audits were carried out by large energy users during the 1986-96 period as result of the 1985 energy conservation law. Thus, the Bank-financed audits represent barely ten percent of the total. The audit program financed by this Bank loan was constrained by the condition imposed that the audited company should pay at least 35 percent9 of the cost of international auditing consultants. Only the very largest energy users were prepared to pay their share (and some of these large firms had already been audited under a previous Bank project [Ln. 2197-TUN]. Most Tunisian firms preferred to use local consultants, particularly those who had been trained and had gained experience working as the local partners of the foreign consultants hired under this and the previous Bank projects. Energy-saving Measures and Investments 3.3 The Bank-financed audits collectively proposed energy-saving investments that exceeded the remaining loan funds. AME was unwilling to commit the bulk of these funds to large projects because it felt that this was a demonstration project and, therefore, important to cover as wide a range of energy users as possible. However, lending for smaller projects was very difficult because at that time the local banking system also had excess liquidity. GOT directives on local bank lending policy required them to lend for productive investments or else put the surplus funds on low-interest deposit at the Central Bank. Hence, firms had access to funds on terms as good or better than IBRD's onlending conditions. Finally, suppliers' credits were also available on more attractive terms. Consequently, AME had great difficulty in finding suitable outlets for the IBRD funds. 3.4 The Bank loan was thus both inadequate in size relative to the investments identified by the energy audits, its terms were unattractive for potential borrowers who had access to sufficient funds in the local capital market, without the cumbersome requirements of satisfying both AME and the Bank's procurement procedures. The appraisal team do not appear to have anticipated these difficulties in utilization of the Bank funds, even though the GOT lending directives existed at that time. 3.5 After much delay, AME succeeded in placing US$0.5 million of the Bank loan in the transport and public buildings sectors. The borrowers are all state-owned entities. Total annual energy savings of about 2000 TOE can be attributed to these subprojects. Other organizations, 9. Later reduced to 25 percent to increase the attractiveness of the Bank-financed audits, but with little effect. The exclusion of local-cost financing from the Bank loan also reduced its utilization. 20 including the oil refinery, a cement producer, a dairy and two hotels also undertook energy- saving measures recommended by the audits, without seeking Bank financing. Total energy savings achieved as a result of the project are over 11,000 TOE/year0, somewhat below the 14,000-27,000 TOE expected at the time of appraisal. 3.6 The ICR argues that a grant rather than a Bank loan would have been an appropriate financial instrument to have used to finance a demonstration project. The Audit does not share this view. The energy audits had identified investments with attractive payback periods, some of which were implemented without recourse to Bank funds. The real obstacle was the inability and/or unwillingness of AME to provide potential borrowers with their full financing needs. And borrowers were unwilling to process two separate loans when they could get the financing they needed from a single institution on terms as good or better than the IBRD loan. Solar Water Heating 3.7 The Gafsa hospital hot water system failed to meet the project's own technical, economic and financial criteria for acceptability. According to the President's Report, investments in energy conservation or renewable energy in the hotel/buildings sector were eligible for Bank financing provided they had been preceded by an energy audit and the proposed investments had a maximum payback period of six years. Neither of these conditions were respected in the case of the SWH system for Gafsa hospital. There was no prior energy audit of the hospital, and the feasibility study itself demonstrated a fifteen-year payback, which was far too long to have been acceptable. 3.8 Under the project, Bank funds were to be onlent to project beneficiaries and no grants for investments were envisaged. Yet neither Gafsa hospital, the local government, or the Ministry of Health agreed to make any financial contribution to the SWH project. Maintenance so far has been covered by the supplier because the equipment is still under guarantee, but in future the hospital should be expected to finance this from its own budget. Whether it will do so is open to question, even though its fuel bills have fallen. Solar Photovoltaic Systems for Rural Electrification 3.9 The national electricity utility, STEG, has achieved an electrification rate of about 90 percent of households. Virtually all villages have been electrified, leaving about 200,000 isolated, rural households (about 10 percent of the population) without access to grid-supplied electricity. STEG aims to achieve a 97 percent electrification rate by 2010. GOT envisages that those households without access to the grid will be equipped with solar photovoltaic (PV) systems. 3.10 Tunisia's experience with solar PV systems began in the late 1980s, by means of a German-funded pilot project in the Kef region. The positive results of the pilot project laid the basis for the installation of 1000 PV systems in rural homes and primary schools in 1991-92, also with German aid. The PV program received a further impetus from 1994 onwards, as a result of the Bank's involvement (see below), and the subsequent GOT decision to include an expanded PV program in its Ninth Economic Plan (1997-2001). At present, nearly 3000 PV systems are in 10. Over 50 percent of which are located in the cement mill. 11. By Studi (Tunis, 1993). 21 service and it is expected that a further 2500 systems will be installed annually over the next five years. 3.11 The underlying principle in GOT's approach to electrification using PV systems is to start with those areas furthest from the grid while at the same time continuing to push the grid outwards towards the unelectrified areas. The reasons for the decision to provide access to electricity to such households are explained in para. 3.14 below. The project site is in a hilly region in north-west Tunisia with a lower than average household electrification rate due to a predominantly dispersed rural population. The homes which have been electrified under the project are often 5-10 kilometers from the nearest village. Even those homes closer to the STEG grid are unlikely ever to be connected, due to the costs involved in a hookup for a single low- voltage customer. Once the decision to electrify these households had been taken, solar photovoltaic systems were the most appropriate technical and economic choice to serve small, isolated consumers. Such households spend on average 12-15 TD/month on kerosene and battery power, which would be sufficient to amortize a solar PV system. Financing Arrangements for Solar PV Systems 3.12 The ground rules for GOT's solar PV program were devised on the basis of the experience gained with a 1989 German-financed pilot project, prior to the installation of 1000 systems under the GTZ-financed project. The absence of documentation indicating that the Bank participated in, or commented on, the financial arrangements for the solar program, suggests that the Bank played a very passive role in this important area, even though it appears to have discussed funding the PV program as early as 1989 and suggested that it was willing to consider financing in mid-1990.12 3.13 Given the monetary outlay involved, it is impossible for most rural households to buy a PV system outright. Purchasing on credit is also impossible, because of the lack of access of such households to long-term consumer loans from banks.14 Leasing by private equipment suppliers or a state agency is impractical due to the high risk of non-payment5 and the cost of collecting funds and/or removing the equipment in the event of non-payment. These factors led to GOT's decision to opt for a subsidized system, in which the beneficiaries would pay only a small initial down payment. This system thereby avoided costly and recurrent administrative overheads involved in running a credit scheme. 3.14 GOT's existing rural development program already contained an annual budgetary allocation for electrification, which till recently had been used exclusively by the local authorities to subsidize grid connections for rural households. In rural areas, STEG's cost ceiling for a new hookup is 1400 TD (US$1,400). This limits the grid supply to grouped settlements of a certain minimum size, leaving isolated households with no prospect of a connection, because extending the network costs over 13,000 TD per km. The difference between the consumer contribution (200 TD) and the real cost is met by GOT and/or local government. GOT decided that PV technology offered a way of improving the living conditions of the remotest rural household who would probably never have access to grid supply. This was 12. Letter to AME from the Acting Division Chief, Industry & Energy Division, 9 July 1990. 13. The Bank-financed systems, which were exempt from taxes & duties and did not include a dealer's margin, cost about US$950 each (installed). 14. The vast majority of such households would not even have bank accounts. 15. A pilot project in the village of Ouled Nouaouia had tested the willingness to pay 5 TD/month instead of an initial down payment. The repayment rate was very poor. 22 seen as a way of slowing the rural exodus. Given that other rural families had benefited from state subsidies for a grid connection, GOT was ready to extend similar subsidies to these households for reasons of equity. However, GOT decided to limit user participation in the PV program to 100 TD per system on the grounds that service offered by a PV system was inferior to that of grid supply, which requires a 200 TD contribution to hookup costs by the consumer. The balance of the required funding for the PV systems was to be shared between the central and regional authorities, in line with the existing practice for grid connections. The low level of the initial consumer contribution can be judged by comparison with the cost of a replacement battery, currently 145 TD, which households will have to bear about once every three years. 3.15 The selection of households to be electrified is done by the local administration out of the pool of eligible households, using socio-political criteria. Clearly, the high level of subsidy limits the number of systems GOT can install annually, while stifling any possible private supply. The Audit mission's field visit to a sample of households with PV systems suggests that many beneficiaries have the financial capacity to contribute a larger share to the investment cost (as do many of those receiving grid hook-ups). The current long waiting list for PV systems confirms this impression. Demand for such systems has taken off and there is now a waiting list of eligible households (i.e. those that STEG has confirmed it will not electrify in the next decade and who have deposited the down payment). 3.16 The subsidy is thus having the perverse effect of operating as a rationing device. Apart from those on the waiting list, many other potential users are also shut out of the solar PV market because they cannot afford to purchase a PV system outright and at the same time are "ineligible" for a publicly subsidized system because in the medium to long term they will be connected to the grid. A smaller subsidy with less restrictive eligibility criteria would have allowed more unelectrified households to have benefited sooner from this technology. 3.17 The Government also plans to subsidize the operational costs of the PV systems. Households with PV systems are not required to pay any maintenance charges even though they make a monthly saving in avoided energy consumption of as much as 12-15 TD, depending on their previous usage of kerosene for lighting, dry-cell batteries for radios and charge-up fees and transport costs of car batteries used for TVs. However, beneficiaries are expected to replace the battery out of their own resources. 3.18 AME is now finalizing its maintenance policy for PV systems, which have been under guarantee till recently. It has convinced the Government to meet maintenance costs from the national budget because of the expected high collection cost and poor results of getting beneficiaries to contribute, and because rural households connected to the grid also continue to benefit from state subsidies. Accurate estimates of maintenance requirements are still unavailable. However, AME has to date, obtained an annual allocation of 100 TD per system17 from GOT's budget for this activity. The maintenance will be carried out by private firms selected by competitive tendering and will be supervised by STEG, who will be reimbursed by AME from the funds its receives from GOT. 16. The tariff for consumers using less than 50 kWh/month is 6 cents/kWh, much less than the full marginal cost of supply, particularly in rural areas where demand is almost entirely at peak time and where system maintenance costs are also above average. 17. About 10 percent of the initial capital cost. Dispersion of sites and poor access makes servicing costly. 23 4. Issues Use of Loan Funds to Finance Subsidies 4.1 GOT's justification for the subsidies to the PV program have been stated in paras. 3.14 and 3.18. These are legitimate concerns for any government. However, such subsidies should be financed from tax revenues, not from borrowings, as was the case for this project. GOT's financial management was poor because this project has given rise to a recurrent cost burden for maintenance on top of the debt service on the loan, without generating any revenues for GOT in return. Normally beneficiaries should be required to pay for the maintenance costs in full. Bank policy on rural electrification is to support only those projects that can cover recurrent costs, even where it accepts the concept of subsidies towards the capital costs on equity grounds. Power sector investments with Bank participation also require a substantial domestic local contribution to the financing plan,-typically 50 percent in non-IDA countries-not the 30 18 percent that was the case here. Neither AME, which is a non-revenue earning body, nor the power utility, STEG, had any financial stake in the project outcome. Procurement 4.2 The case of the late procurement of additional solar PV systems demonstrates the importance of a single contract for both supply and installation. AME could not use a single contract because of the limited time for disbursements after the closing date. Installation only began after the equipment had been received and the supplier had been paid in full out of the Bank loan. The purchase of this second set of PV equipment was motivated by a desire on the part of AME to use up the remaining undisbursed balance of the Bank loan. The Bank should not have approved their procurement at such a late stage, knowing the potential risks involved. It is symptomatic of the poor management of the loan by both AME and the Bank that the quantities procured under the first consignment (supply and erect), were not increased when rebidding took place (ICR para. 25), less than a year before the final closing date. This would have enabled AME to avoid the problems that later arose from the equipment-only contract. Bank Performance 4.3 As explained in the ensuing paragraphs, the performance of the Bank was highly unsatisfactory. First, the substantial change in the objectives and scope of the project, represented by the addition of a new rural electrification component, accounting for over half the loan amount, was not subject to senior management review or Board scrutiny. This was a serious deviation from the normal practice of carrying out a formal project restructuring (that is submitted to the Board for approval), and a formal change in the project scope as described in schedule 2 of the Project Agreement. In addition, this component was not subject to the usual standards of project evaluation. No financial and economic analysis was undertaken because it was described as a "pilot" project, but in fact it was the third (and fourth) in a series of rural household PV projects going back to 1991, and should have been analyzed as a component of an ongoing program. It is highly unlikely that the Bank's senior management would have agreed to ' The local authorities contributed 20% and the beneficiaries 10%. 24 the use of IBRD funds for direct subsidies to the solar PV program, had they been given the opportunity to review the proposal. 4.4 As mentioned in the ICR (paras. 21-23), the project suffered from poor and insufficient supervision throughout its life. The Audit was unable to trace any supervision mission Aide- Memoires or BTO reports on the project after early 1992 (the two and a half year period of maximum activity and disbursements) and as the ICR notes (para. 23), record keeping was haphazard and crucial documents such as closing date extensions were lost. No field visit was undertaken to inspect the new project components approved in 1992, even upon project completion, and the non-objection granted to AME to procure additional equipment at the end of the project was imprudent. 4.5 In the Audit's judgment, the original description of the project does not permit (Loan Agreement, Schedule 2) the proceeds of the loan to be used to finance the purchase and installation of solar PV equipment, or the purchase of vehicles for this component. However, no formal legal amendment was made to permit disbursements for these items, showing a surprising disregard for standard Bank procedures. It is also significant that the Loan Department processed these payments, despite their non-conformity with the project description. 4.6 Although as early as mid-1990, the Bank had indicated to AME9 that the loan management costs seemed disproportionately high compared to the benefits AME could obtain from drawing on the loan, there appears to have been no attempt by the Bank to cancel the loan. 4.7 The ICR is also unsatisfactory. It was unable to describe the outcome of the major project components because it did not make a field visit to see the results on the ground and meet with the project implementing entity. As a result its coverage was incomplete. It did not discuss the solar PV or SWH components; nor did it contain any discussion of the rationale for the decision to reallocate the unused loan funds to these activities. Given the lack of field supervision by the Bank of these components, the preparation of the ICR should have been used as an opportunity to assess the outcome of the project upon completion. The decision of the Region not to do so is characteristic of the superficial treatment of this project throughout its implementation. 19. Translation of letter in French to AME from the Acting Division Chief, Industry & Energy Division, 9 July 1990. 25 5. Overall Assessment 5.1 The project contributed directly to the institutional strengthening of AME, to local capacity building in energy auditing and conservation techniques, and indirectly to the creation of small enterprises to service the needs of the solar PV program. Over the past decade, AME has developed into a mature institution with competent and motivated staff. However, it suffers from the constraints arising from its status as a state body entirely dependent on GOT's budget for its revenues. It has to follow government procedures for procurement, accounting, etc. and has a high rate of staff turnover at junior levels due to its public-sector pay scales. The Audit therefore assesses the institutional development impact of the project as modest, which is consistent with the "partial" rating of the ICR. 5.2 OED rates the project outcome as unsatisfactory, rather than satisfactory as in the ICR. The reason that it received this unsatisfactory rating is that the revised project design, which used half of the loan for a PV component, was not consistent with Bank guidelines on lending for the energy sector, and would most likely have been rejected if it had been subject to the Bank's standard screening process. The major part of the loan was spent on project components that did not meet the Bank's requirement for sector policy objectives or economic viability. Neither did the project qualify as a demonstration project since it was a continuation of the existing RE program that incorporated may policies that were contrary to Bank guidelines, including the selection of beneficiaries discussed in para. 3.15, and the subsidization issues discussed in para. 4.1. 5.3 On the positive side, the project's energy audit component was a partial success and there has been more follow-up action taken to save energy by the audited enterprises than the ICR indicates. Although the potential for savings is far from exhausted, particularly in the industrial sector, low real energy prices and high interest rates lengthen payback periods and discourage potential investments. Furthermore, because many large energy users are state- owned enterprises (such as cement works) which may be privatized in the coming years, the present management sees little point in undertaking such investments. Experience here, as well as in other countries where the Bank has tried to support Government-controlled industrial energy conservation entities, suggests that such activities are probably best left to the private sector. 5.4 The solar PV component of the project can be considered as broadly successful and sustainable in technical terms, despite the equipment problem experienced with one consignment. The 1245 systems installed are all operating satisfactorily and the beneficiaries obtained a net improvement in their living standards as a result. Once the remaining systems are installed, over 2,000 families (10,000 people) will have obtained access to electric lighting for an investment of US$2 million. 5.5 However, the project is not financially replicable because GOT has to bear the full recurrent cost burden of about US$0.25 million annually, for maintenance. It is unlikely that the project could ever be made financially self-sustaining in an environment where community- based credit and savings schemes are unknown and where the banking system does not operate. Fortunately for GOT, the resulting proposed investment program and recurrent-cost subsidies to 26 rural households are relatively modest. However, to electrify 70,000 rural households,20 over the next 15 years could put a strain on the Government's budget.21 5.6 For the Gafsa SWH project it is difficult to see how it can serve as an effective and convincing demonstration of the advantages of investing in solar water heating systems, given its high cost and long-payback period. And poor experiences with locally made equipment have already damaged the reputation of SWH systems in Tunisia. This component was a poor use of Bank funds. It is impossible to defend borrowing for a project that is highly unlikely even to cover its initial costs during its economic life, quite apart from generating any surplus to service the additional debt. 5.7 A demonstration project with a hotel would have been much more useful, given the potentially large market for such systems in the hotel sector. The poor design and choice of a hospital for the hot water demonstration project is, at least partially, a consequence of AME having no financial responsibility for loan repayment, because the funds were never onlent to it. The lack of financial participation by the beneficiary of the demonstration project (the hospital) was also an error in project design, as can be seen by the hospital's failure to connect its new wing to the SWH system, more than two years after commissioning. The decision of the Bank to approve financing for this component was unjustified. The fact that this approval took place just weeks before final loan closure suggests that the Bank tacitly agreed with AME's approach of maximizing disbursements from the loan, essentially without regard to financial and economic criteria. OED considers Borrower performance under the project to have been unsatisfactory. 5.8 The non-observance of several important Bank operating practices under this project is disturbing. Outside the social sectors, IBRD funds should not be used to finance projects that generate no financial returns. The GOT approach to solar PV systems is not consistent with the Bank's guidelines on rural electrification. Overall Bank performance has therefore been rated as highly unsatisfactory. 5.9 The loan should have been canceled at the end of the energy audit program in 1991. By then, AME had benefited from the institution building and training aspects of the project and local consultants had also gained useful skills and experience in energy auditing. As it turned out, the bulk of the energy savings attributable to the project were achieved without any Bank participation in their financing. The Minister of Plan's recommendation in 1992 (ICR para. 27) to cancel the unused balance of the loan should have been approved. Instead, the repeated extension of the closing date enabled AME to pursue non-project related objectives without the need to seek budgetary resources from GOT, while costing the Bank at least two more years of supervision and loan management expenses. The recorded staff inputs for this operation were 110 SW, the cost of which far exceeds the income the Bank earned through commitment charges etc. on the loan. 20. Representing the 3 percent of households who will not have grid supply in 2010. 21. If GOT were to finance the current program, which aims to have 15,000 systems installed in five years' time, it would require US$15-20 million for capital expenditures and have to meet annual recurrent costs of about US$2-3 million However, in the event that GOT is unwilling to support the latter cost, it would be an insupportable burden on AME, whose current annual income and operating budget is only US$1.3 million. 27 6. Lessons The key lessons from this project are: * Small loans are justified only for projects with clear cut, limited objectives that can be implemented in a limited time frame. Prolonging small projects for several more years after their closing date should be avoided unless the arguments for an extension demonstrates a strong case that to do so will ensure that the project will meet its original objectives. Extensions should not be given just to allow the borrower to use the funds,. * In revenue earning sectors, executing agencies should have some financial responsibility for cost recovery and repayment of Bank funds. Otherwise there is a high risk of sub-optimal use of these funds. * Solar PV systems are a viable technical and economic alternative to other energy sources for a dispersed rural population. However, these systems are expensive, and therefore may only provide a viable alternative for middle-income populations. * Rural solar PV programs should provide the consumer with a choice of more than one system capacity (watts), at different costs. This would enable systems to be sized according to the varying needs and incomes of end-users. * Subsidies high enough to generate waiting lists lead to an inefficient allocation of resources. Providing smaller subsidies to a larger group of people would increase economic efficiency and equity and also lead to a faster dissemination of PV systems.  ANNEX 1 29 Comments From the Borrower (Translated from the original French) FAX MINISTRY OF INDUSTRY AME: Agence pour la Maitrise de 'Energie DATE: June 17, 1997 No.: 009 TO: Mr. Yves Albouy Division Chief Infrastructure and Energy Operations Evaluation Department, World Bank FROM: Mr. Salem El Golli CEO COPY: Multilateral Cooperation Unit, Ministry of International Cooperation and External Investment SUBJECT: Performance Audit Report on the Energy Conservation Demonstration Project (Loan 2735-TUN) Further to your fax of June 6, 1997, we enclose AME's comments on the PAR that you forwarded to us regarding the above project. Sincerely, Attachment 30 ENERGY CONSERVATION DEMONSTRATION PROJECT, COFINANCED BY IBRD LOAN No. 2735-TUN The purpose of this report is to provide an evaluation of the Energy Conservation Demonstration Project cofinanced by IBRD Loan No. 2735-TUN. The project had the following objectives: - to assist AME in designing an effective institutional and policy framework for the PNME (National Energy Conservation Program); - to strengthen local capabilities for both implementing the PNME and carrying out energy efficiency audits and conservation actions; - to finance actions identified by the above audits, so as to demonstrate the benefits produced by such investments. These objectives were to be achieved through activities relating to the following: - providing AME with: - technical assistance for preparing a demonstration program of energy conservation measures and related monitoring activities; - equipment, tools, and instruments for measuring and monitoring energy consumption, to enable AME to train consumers and implement demonstration activities; - organizing training in basic energy audits and energy conservation methods for AME staff, consumers participating in the demonstration activities, and selected specialists such as the faculties of engineering schools; - conducting energy audits and feasibility studies, implementing energy conservation activities, organizing on-the-job training in energy conservation, and supplying equipment, tools, and instruments for a number of industrial enterprises, road transport companies, and hotels and other buildings. Implementation began in 1986, and AME's activities were monitored throughout the project period. The support provided was focused on its programs and priorities, enabling readjustments to be made to the original plan, in light of the progress made and the new demands arising at particular stages. Consequently, the project was launched with the issuance of Tunisia's first energy conservation legislation (Decree-Law No. 85-8 of September 14, 1985 governing energy conservation, ratified by Law No. 85-92 of November 22, 1985, and Decrees No. 87-50 and No. 31 87-51 of January 13, 1987, the first introducing compulsory periodic energy audits and the second requiring prior consultation with AME for the implementation of projects entailing large- scale energy consumption). This legislation emphasized the need for energy conservation and the means for achieving it, and this accounts for the importance assigned to this component in the activities undertaken in order to fulfill the project objectives. Promulgation of Law No. 90-62 of July 24, 1990 marked a new stage in this area. It extended the field of energy-conservation activities in Tunisia by including the use of alternative energy, together with promotion of the use of renewable energy. In light of the findings of the energy audits carried out under the project (since the volume of investment required by the activities necessary for achieving energy savings was too large to be covered by the loan), amendment of the loan was essential. This being so, the amendments served to support energy conservation; thus, the objective of helping AME develop an effective institutional and policy framework for a national energy conservation program was achieved. The second objective -- namely, to strengthen local capabilities for implementing the energy conservation program and carrying out energy efficiency audits and conservation actions -- was also achieved. In fact, a considerable amount of activity has developed locally, supported by a constantly developing body of skills that was well and truly launched by the Energy Conservation Demonstration Project, though its training sessions and the active participation of local consulting firms in the audits performed under the project, since these provided a transfer of skills and know-how that is now enabling Tunisian firms to fulfill their functions under the best possible conditions. Even though the final objective (the financing of activities identified by the audits in order to demonstrate the benefits produced by such investments) was only partially achieved (the loan amount being insufficient to cover the volume of investment necessary), the activities that were in fact financed served as pilot projects. They were successful, and some were taken up by enterprises that had not been participants in the original project. The main activities receiving financing as a result of being added to the amended project (the rural electrification component involving the use of PV systems, the solar water-heating component, and the introduction of vehicle engine test benches) enabled AME to explore these areas and incorporate them into the national energy conservation program because of the convincing results they produced. The areas in question are now at the leading edge of energy conservation activities in Tunisia. NB: The PAR for the Energy Conservation Demonstration Project (Loan No. 2735-TUN) states that the solar water heating system at the Gafsa regional hospital has not operated satisfactorily since it was commissioned. It must be pointed out that the system has, in fact, operated perfectly since it was installed. It is entirely to be expected that a few minor technical problems would occur, and be immediately solved; these, however, have had no impact on the 32 actual technical performance, which has considerably exceeded the standards guaranteed by the engineer (i.e. guaranteed output = 151,716 kWh; actual = 157,829 kWh; difference = 6,113 kWh).  IMAGING Report No.: 16817 Type: PPAR