Document of The World Bank FOR OFFICIAL USE ONLY MICROFICHE COPY Report No. 10501 Report No. 10501-TUN Type: (PPR) NAMISATO, / X31678 / T9105/ OEDD2 PROJECT PERFORMANCE AUDIT REPORT TUNISIA ELECTRICAL AND MECHANICAL INDUSTRIES PROJECT (LOAN 2113-TUN) APRIL 9, 1992 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 Currency Unit: Tunisian Dinar (TD) (annual averages) 1980 US$1= TD 0.42 1986 US$1= TD 0.79 1981 US$1- TD 0.49 1987 US$1- TD 0.83 1982 US$1- ID 0.59 1988 US$1- TD 0.86 1983 US$1- TD 0.68 1989 US$1= TD 0.95 1984 US$1- TD 0.78 1990 US$1- TD 0.88 1985 US$1= TD 0.83 11/91 US$1- TD 0.92 ACRONYMS AND ABBREVIATIONS API Agence de Promotion des Investissements BCT Banque Centrale de Tunisie BDET Banque de D6veloppement Economique de Tunisie BTKD Banque Tuniso-Koweitienne de D6veloppement CEPEX Center for Export Promotion CETIME Centre Technique des Industries M4caniques et Electriques CMT Complexe M6canique de Tunisie CNEI Centre National des Etudes Industrielles CPO Centre Pilote de 1'Outillage EMI Electrical and Mechanical Industries EPR Effective Protection Rate FSS Financial Sector Study IEQ Institut d'Economie Quantitative INNORPI Institut National de la Normalisation et de la Propriet4 Industrielle ITPAL Industrial and Trade Policy Adjustment Loan MNE Ministry of National Economy NPR Nominal Protection Rate PCR Project Completion Report PE Public Enterprises QR Quantitative Restrictions SAL Structural Adjustment Loan SAR Staff Appraisal Report SECAL Sectoral Adjustment Loan STIA Soci6td Tunisienne d'Industrie Automobile STUSID Soci6t6 Tuniso-Saoudienne d'Investissement et de D6veloppement Fiscal Year January 1 - December 31 THE WORLD BANK FOR OFFICIAL USE ONLY Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluatim April 9, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report on Tunisia - Electrical and Mechanical Industries Project (Loan 2113-TUN) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Tunisia - Electrical and Mechanical Industries Project (Loan 2113-TUN)" prepared by the Operations Evaluation Department . Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT TUNISIA ELECTRICAL AND MECHANICAL INDUSTRIES PROJECT (Loan 2113-TUN) TABLE OF CONTENTS Page No. Preface................... ...... . . . . . . Basic Data Sheez.. .............. .. ... . .. . . . iii Evaluation Summary.................. ..... . . vii I. INTRODUCTION................. ..... . . . 1 II. OBJECTIVES, CONCEPT AND DESIGN........ .... . . . . 5 THE FIRST EMI PROJECT.......... ...... . . . . . 5 A. Objectives . . . . . . . . . . . . . . . . . . . . . . . . 5 B. Components . ... . . . . . . . . . . . . . .. .. .. 6 C. Volume of Lending Target . . . . . . . . . . . . . . . . . 7 D. Special Eligibility Criteria and the Protection Issue . . . 7 THE SECOND EMI PROJECT.......... .... . . . . . . 10 III. IMPLEMENTATION EXPERIENCE.......... ...... . . . 11 A. Financing of Electrical, Mechanical and Other Industries 11 B. Conditionalities Related to Competitiveness...... . . . 17 C. BDET Performance............. ..... . . . . 17 a. BDET's Structure, Operations and Portfolio . . . . . . 18 b. Profitability........... ...... . . . . 22 c. Loan Conditionalities related to BDET as an Institution 23 D. Technical Assistance............. .... . . . 25 a. CETIME........................ . . *.... . . . . . . 27 b. INNORPI........... ...... . . . . . . . 31 E. Effective Protection Study......... .... . . . 33 F. Sectoral Studies............ .... . . . . . . 37 a. The Platework Subsector Study. ....... . . . . . . 38 b. The Foundry Subsector Study. ....... . . . . . . . 38 IV. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . 39 A. EMI Projects Financing . . . . . . . . . . . . . . . . . . 39 B. Non-EMI Projects Financing . . . . . . . . . . . . . . . . 40 C. Technical Assistance . . . . . . . . . . . . . . . . . . . 42 a. CETIME . . . . . . . . . . . . . . . . . . . . . . . . 42 b. INNORPI . . . . . . . . . . . . . . . . . . . . . . . . 42 D. The Protection Issue . . . . . . . . . . . . . . . . . . . 43 E. The Subsector Studies . . . . . . . . . . . . . . . . . . . 44 F. Institutional Building Efforts and BDET Performance . . . . 44 G. Overall Results . . . . . . . . . . . . . . . . . . . . . . 45 V. SUSTAINABILITY............. ....... . . . . . 45 This document has a restricted distribution and may be used by recipients only in the performance of their ofitcial duties. Its contents may not otherwise be disclosed without World Bank authorization. Table of Contents (Cont'd) Page No, VI. LESSONS OF EXPERIENCE.......... ...... . . . . . 46 ATTACHMENT: Banque de Ddveloppement Economique de Tunisie's Comments 53 ANNEXES 1. BDET Cumulative Disbursement of Funds. ........ . . . . . 57 2. Value Added, Employment and Investment in EMIs . . . . . . . . 58 3. Value Added, Employment and Investment in EMIs (in percentage) 59 4. Value Added and Nominal and Effective Protection Rates of EMIs 60 5. Value Added and Nominal and Effective Protection Rates of EMIs (selected Items).............. ..... . . . . . 61 6. Output and Investment in Electrical and Mechanical Industries 62 7. EMI Investments in the 1982-1986 Plan. ...... . . . . . . . 63 8. Number of Firms and Employment by EMI Subsectors . . . . . . . 64 9. Number of Firms and Employment in Exporting Electrical and Mechanical Industries........... ...... . . . . 65 10. EMI Imports and Exports......... ..... . . . . . . 66 11. BDET Pipeline of EMI Projects...... ..... . . . . . .67 12. EMI Projects Financed by BDET . . . . *...... .. . .. . . . 68 13. Project Analysis by Manufacturing Subsector - Loan 2113-TUN 69 14. BDET - Projected and Actual Industry and Tourism Operations (1981-1986)........... ..... . . . . . . . . . . 71 15. BDET - Analysis of Approvals....... .... . . . . . .72 16. BDET - Total Approvals, Commitments and Disbursements . . . . . 73 17. Loan and Participations Approvals in Manufacturing and in EMIs (BDET, BTKD, STUSID)........ .... . . . *...........74 18. BDET: Electrical and Mechanical Industries - Approvals, Commitments, Disbursements and Arrears. ........ . . . . 75 19. BDET - Financial Indicators........ ..... . . . . .76 20. BDET - Trend in Arrears .... 77 21. BDET - Foreign Exchange Losses* ...78 22. BDET - Forecast and Actual Financial Indicators..... . . . . 79 23. BDET - Structure of Borrowings..... .... . . . . . . .80 24. Centre Technique des Industries M6caniques et Electrique (CETIME)......... .... . . . . . . . . . . . . . . . 81 25. Institut National de la Normalisation et de la Propridtd Industrielle (INNORPI).......... .... . . . . . . . 82 PROJECT PERFORMANCE AUDIT REPORT TUNISIA ELECTRICAL AND MECHANICAL INDUSTRIES PROJECT (Loan 2113-TUN) PREFACE 1. This is a Project Performance Audit Report (PPAR) on the Electrical and Mechanical Industries Loan 2113-TUN. This Loan, in the amount of US$30.5 million, to the Banque de D6veloppement Economique de Tunisie (BDET), was approved in March 1982 and the last disbursement was in June 1987; the Loan was closed in December 1987, as scheduled. Following several cancellations, only 68.6% of the Loan was disbursed. 2. The PPAR was prepared by the Operations Evaluation Department (OED), and based on the Project Completion Report (PCR) prepared by the Europe, Middle East and North Africa Region (EMENA) of the Bank and issued in 1991,1' the President's Report, sector and economic reports, the summary of the Board discussion, study of the project files, and discussicns with Bank staff. An OED mission visited Tunisia in November 199U and discussed the effectiveness of the project with Government officials and representatives of the business and financial community. Their kind cooperation and valuable assistance in the preparation of this report is gratefully acknowledged. 3. The PCR provides a good account and assessme.t of the project experience with regard to the implementation of the various action programs, achievement of the components, and the Bank's role. The PPAR discusses the evolution of the electrical and mechanical industrial sector in Tunisia, portrays the critical issues of concern that led to the genesis of these indIstrial sector loans, evaluates the progress made in creating and implementing the various components of the projects, determines the purposefulness and effectiveness of conditionalities and technical assistance measures, ar. ascertains the key factors that determined the loans' outcome and the effectiveness and sustainability of the projects. The PPAR then draws lessons from the project experience. 4. The draft PPAR was sent to the Borrower for comments. The comments received from the Banque de D6veloppement Economique de Tunisie (BDET) are reproduced as Attachment to the PPAR. 11 Project Completion Report, Tunisia - Electrical and Mechanical Industry Project (Loan 2113-TUN), April 19, 1991. PROJECT PERFORMANCE AUDIT REPORT TUNISIA FIRST ELECTRICAL AND MECHANICAL INDUSTRIES PROJECT (Loan 2113-TUN) BASIC DATA SHEET LOAN POSITION (amounts in US$ million) As of December 31. 1991 L2aD Ori&Lnal Disbursed Cancelled ReRaid Outstanding 2113 30.50 20.92 9.58 14.58 6.34 CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS FY83 FY87 Appraisal Estimate (US$ million) 11.80 30.50 Actual (US$ million) 11.50 20.92 Actual as % of Appraisal (%) 97.4 68.6 Date of Final Disbursement June 9, 1988 PROJECT DATES Original Actual Project Brief 07/21/81 07/21/81 Yellow Cover 10/30/81 11/13/81 Negotiations 12/07/81 02/09/82 Board Approval 02/09/82 03/25/82 Signing 03/01/82 05/14/82 Effectiveness 07/31/82 12/02/82 Loan Closing 12/31/87 12/31/87 - iv - $TAFF INPUTS (Staffweeks) EiR FY82 FY83 FY84 EY8 Y86 FY87 ESA FY9 .T&TAL Preparation 45.4 62.2 7.0 114.7 Appraisal 33.8 1.2 34.9 Negotiations 4.9 4.9 Supervision1 1.6 15.6 5.5 2.5 6.5 2.8 34.5 Completion 8.7 8.7 -TAL 45.5 102.6 23.8 5.5 2.5 6.5 2.8 -.- 8.7 197.7 MISSION DATA MonthfYear No. Weeks No. Persons Staffweeks Date of Report Preparation 06/80 1 3 3 07/24/80 09/1 3 3 9 07/21/81 Appraisal 06/81 3 5 15 03/02/82 Supervision I 10/82 3 1 3 10/26/82 Supervision II 01/83 2 3 6 03/01/83 Supervision III 05/83 1 1 1 Supervision IV 05/84 1 1 1 05/25/84 Supervision V 09/84 1 3 3 Supervision VI 12/84 0.5 3 1.5 Supervision VII 12/85 2 2 4 Supervision VIII 09/87 2 2 4 Supervision IX 02/88 0.2 1 0.2 Supervision X 06/88 1 3 3 1 Supervision was carried out increasingly in conjunction wit', the second Loan to Electrical and Mechanical Industries (Loan 2554-TUN) app.oved in 1985. Staffweeks spent on supervision under that Loan totalled 25.6 in FY85-87. They amounted to 33.9 in FY88-91. -v - OTHER PROJECT DATA Related Projects Borrower: Government of Tunisia Project: Industrial and Trade Policy Loan Loan No.: 2781-TUN Amount: US$150.0 million Approval Date: 08/05/87 Project: Second Electrical and Mechanical Industries Loan No.: 2554-TUN Amount: US$54.0 million Approval Date: 05/23/85 Project: Export Industries Loan No.: 2522-TUN Amount: US$50.0 million Approval Date: 04/25/85 Borrower/Executing Agency: SOFOMECA Project: SOFOMECA Foundry/Modernization and Expansion Project Loan No.: 2301-TUN Amount: US$16.80 million Approval Date: 06/02/83 vii - PROJECT PERFORMANCE AUDIT REPORT TUNISIA ELECTRICAL AND MECHANICAL INDUSTRIES PROJECT (Loan 2113-TUN) EVALUATION SUMMARY Introduction 1. This is an audit of the Electri- product quality by establishing stan- cal and Mechanical Industries Loan, dards and quality control of products approved in March 1982, for US$30.5 through the establishment of an Insti- million. The circumstances that led tute for Standardization and Quality to the sanction of this project are Control of industrial products; and detailed in paras. 1-19. This pro- (ii) delivering technical assistance ject, which represents the first at- to EMI firms and providing on-the-job tempt to deal with industrial policy training, using the services of a issues, was followed by a second loan Technical Center to be established to electrical and mechanical indus- (para. 20). tries in May 1985 for an amount of US$54 million. Subsequently, an in- 4. The project was also intended to dustrial and trade sector operation help in implementing changes in incen- (ITPAL) in 1987 and a Structural Ad- tives to EMIs in the area of protec- justment Loan in 1988 provided the tion, including Lustom tariffs and foundation of Tunisia's medium-term quantitative restrictions. Special adjustment program. eligibility criteria were devised to limit the degree of protection granted Objectives to EMIs and an Effective Protection Study was to be undertaken as a step 2. The basic objectives of the pro- towards pursuing the dialogue between ject was to foster the development of the Bank and Tunisia on industrial and EMIs, improve their efficiency and trade issues (paras. 27-33). produc:ivity, and increase their pro- duction of competitive capital and 5. The provision of financing under intermediate goods. This sectoral that loan for general purpose lending approrch was an important departure to industries at BDET's request for from 1:revious operations i 'th con- dive-sification and flexibility rea- sisted either in financiny .astries sons and also as safeguard against across the board or hori.' wal opera- possible failure of tbe EMI sector tions such as small-sca; ..ndustries approach was, however, assumed to be (para. 21). secondary an objective of the Project (paras. 21-23). 3. Development of EMIs was to be supported by financing, through a line 6. Another traditional but also of credit to a development bank less important objective of the loan (BDET), new and modernization/reha- was to continue strengthening BDET as bilitation projects in intermediate an institution and conditionalities and capital goods subsectors and by were included regarding interest rate delivery of technical assistance in structure, government subsidies, prof- two complementary ways: (i) fostering itability and company's management. - viii - Implementation Experience als and commitments. Arrears accumu- lated and profits declined. The Gov- 7. On the whole, progress in imple- ern4ent did not fully compensate BDST menting the basic objective, i.e. for the high losses it incurred Jue to fostering EMI development, has not Zluctuations in foreign exchange risk. been satisfactory in most respects. BjET had also to meet growing competi- As a result of a general slump in tion from newly setup Arab-backed economic performance, of persisting development banks which were original- structural, technical and managerial ly conceived to support investments problems, and of a radical change in which the State was increasingly un- Government policy in 1985-86 regarding able to finance directly. To the industrial strategy and in particular extent that EMIs fixed investment was the EMI subsector, more than half of sharply r - from 1984 to 1988, funds available under the EMI-I Loan BDT felt t, rur; like the other new had to be cancelled. A second larger banks. Althougl BUT intensified its EMI-II project was hastily approved in project supervision efforts to recover 1985 to provide BDET with additional very large arrears, it 'icked suffi- funds for EMIs at lower interest cient expertise, and was not fully rates. But the demand for financing equipped, to evaluate and follow up on EMIs proved much smaller than antici- specific EMI projects which presented pated and, again, a substantial por- difficult technical and managerial tion of the second loan had to be problems. Bank supervision efforcs cancelled (paras. 41-68). were marginally effective in bringing about much needed specific improve- 8 The portion of the line of cred- ments in the EMI subsector. Such IL for general industrial lending supervision was in fact very limited, purpose was also only partially dis- and was further weakened by the multi- bursed, although to a larger extent plicity of objectives that had to be than for EMIs. This was mainly due to met under the Loans (paras. 70-104). the slowdown in economic activity during the project implementation 11. The Effective Protection Study period (1983-1987). had basic methodology defects and was implemented with delays. Given the 9. Ceilings on import duties for reluctance of the Tunisian authorities intermediate and capital goods fi- to discuss industrial policy issues in nanced under the loans were apparently the context of a limited subsectoral respected. However, quantitative operation financed through a DFC, the restrictions remained in force, and Bank agreed to pursue the dialogue on the Tunisian authorities refused to industrial and trade issues through dismantle this key element of their other channels, and in particular on protection system. !- Bank's condi- the occasion of the US$150 million tion that locally-prvduced EMI goods Industrial and Trade Policy Adjustment protected by import licensing maintain sector loan negotiated in 1986 (paras. reasonably competitive prices in com- 154-176). parison with similar foreign products could not be effectively implemented 12. The technical & istance compo- due to the practical difficulty to nent was implemented with the creatlon obtain cost and price data on foreign of a Quality Control Institute products (para. 69). (INNORPI) and a technical center to assist EMIs (CETIME). However, these 10. BDET evidently suffered from the institutions had to start practically sharp decline in the level of approv- from scratch, and time was naturally - ix - required before they could have some bursed balance is being cancelled impact on EMI development. Problems (paras. 65-67). were encountered in particular regard- ing private sector's acceptance of 15. The poor results obtained have CETIME interventions in providing to be undezstood in the context of lirect advice and suggestions to EMIs, declining economic conditions affect- since CETIME was seen mainly as a ing in purticula e the vulnerable to regulatory agency with powers to ap- imports 'eMI sector. Growth in value prove or disapprove private investment added in EMIs decelerated from 12.5% applications for EMI projects on tech- per unnum in 1980-82 to 5.6% in 1983- nical grounds. Over the years, this 85 and 0.2% in 1986-87. Only in 1989- role has been reduced but, in the 90 did the s-.tor perform somewhat meantime, CETIME had to center its better. Employment rose very slowly activities on manufacturing prototypes (i.e., about 3.5% in 1982-88) and and even selling some equipment to fixed investment in EMIs actually industries rather than to provide declined by 37% from 1984 to 1987 (at direzt, in-depth in-plant technical current prices) (para. 193). assistance to EMIs. CETIME's real impact on industry is hard to quanti- 16. Most EMIs are still basically fy (paras. 105-153). oriented toward the domestic market (in the old line of import substitu- Results tions) and have been hit by the liber- alization of the economy in 1986-1990. A. Financial Assistance A number of EMIs have had to reorient their production towards exports since 13. Only about 40% of the amount the domestic market remains too small earmarked for EMI was disbursed under to absorb their output and since the Loan 2113. Only eight projects were new, more liberal industrial policy financed and they are all i.n arrears aims at promoting efficient exports. except one. BDET contribution to Despite recent efforts, the share of their total costs was only 8.7% in exports in EMI output remains rela- 1983-87. The number of Jobs created tively small and exports are concen- was ridiculously low: 752 against trated in a few subsectors (para. 61). 6,000 originally envisaged. The aver- age investment per job was five to 17. The EMI sector has proved to be seven times higher than estimated at a difficult one, technically as well appraisal even after taking into ac- as commercially. Its market is pri- count the 1986 devaluation of the manly local, but even the local mar- dinar. Lending for general purpose ket is not easy to secure because the manufacturing investments was lower appeal of imports is high. Technical- (16%) than expected. Employment cre- ly, many EMIs still suffer from insuf- ation was larger than in EMIs although ficient standardization, low quality, investment cost per job was even high- lack of supervisory personal and qual- er (paras. 183-193). ified engineers and technicians, lit- tle Research and Development (R&D), 14. To date, some US$30.0 million of etc. Finally, there appears to be the approved US$54.0 million Loan 2554 significant over-capacity as well as has been cancelled, due to the avail- over-integration. Many entrepreneurs ability of less expensive resources seem reluctant to trust other local from bilateral sources and other Bank entrepreneurs for producing some of funds. The remaining small undis- the inputs in their production line. -x - Subcontracting is far from widespread C. The PrQtection Problem (paras. 62-63). 20. The Effective Protection Study, 18. The EMI project underestimated despite its defects, was useful in the time and the amount of assistance keeping alive the debate between the needed to overcome these problems, Bank and the Tunisian authorities on despite efforts to include technical the need to liberalize the economy by assistance measures in both projects. decreasing tariffs and reducing quan- titative restrictions and, thus, give B. Techniral Assistance impetus to industrial competitiveness and exports (paras. 200-201). 19. Achievements since 1983 by the Standards and Quality Control Insti- 21. This result was, however, a tute (INNORPI) are quite commendable side-effect of the EMI projects. The and, by any measure, Bank financing Tunisian authorities consistently has been well used. The Technical objected to full-fledged conditionali- Center (CETIME) which started from ties aiming at truly removing quanti- scratch needed time to be credible tative restrictions on EMIs across the vis-a-vis private industrialists. board. The timing of EMI-I was not CETIME feels it has provided jobs and good since in 1982 the Tunisian policy career opportunities to a number of was still de facto heavily protection- engineers and technicians, including ist. Even in 1985 when EMI-II was those who moved from CETIME to become approved, little progress was achieved entrepreneurs. It also claims to have regarding liberalization of EMI prod- helped a number of small and medium ucts. Thus, effective policy results companies who could hardly obtain obtained under EMI-I and II were ex- assistance elsewhere. But CETIME's tremely limited, suggesting that a anomalous position as part manufactur- subsectoral approach cannot be an er (through its own specialized tech- efficient vehicle for basic policy nical centers) and part R&D operation chpnges when real issues are much remains to be clarified, as well as broader than those pertaining to a the degree to which it is funded by particular subsector (para. 202). the Government. At another level, CELIME's overall impact on the avail- D. Institutional Building AsRects: ability and development of technology BDET's Performance for EMIs is inevitably still very small (in particular for large compa- 22. BDET's operations and portfolio nies), as the industries easily obtain did not perform well under EMI-I and whatever technology they want from IT, at least until 1988-89, when abroad at modest relative costs, with BDET's situation finally seemed to faster delivery, and more tailored to improve after a number of particularly their needs. What was underestimated difficult years. was not only CETIME's ability to solve the EMIs technical problems in a few 23. BDET failed to develop in time a years but, above all, the time needed full awareness of the fragility of the to transform :ssembly-type industries demand for investment financing for into a truly modern and efficient the EMI sector. This might have been electrical-mechanical subsector due to efforts to identify appropriate (paras. 194-199). ways to develop Tunisia's capabilities - xi . in this sector, and to the fact that 26. INNORPI and CETIME are increaa- investors were not fully convinced of ingly active institutions, and there the prospects of the sector (para. is hope that they will improve their 206). relationship with EMIs once they clearly separate their technical as- E. Overall Outcome sistarce functions from their regula- tory role. Consequently, those insti- 24. Simply stated, the EMI projects tutions, if properly managed, are tried to accomplish too much. They likely to remain sustainable in future aimed at developing the sector and at years. expanding and strengthening BDET, while bringing about major policy 27. After several years of stagna- changes. The projects clearly failed tion, fixed investment and output in to develop the EMI sector, although EMIs have resumed growth in 1988-1990 they succeeded in developing INNORPI but private investment in EMI still as a quality control institution to remains fairly unresponsive to liber- encourage exports of EMI products. alization measures undertaken in the The projects did not play a major role context of ITPAL and the SAL. in strengthening BDET, as project conditionalities in this respect were 28. One major reason for the lack of rather limited. And they did not sustainability of EMI I and II was the succeed in convincing the Tunisian impossibility to solve through a pure- authorities to change their protec- ly subsectoral approach the whole tionist industrial policies. However, question of incentives, customs du- discussions initiated under EMI I and ties, and quantitative restrictions. II indirectly provided an opportunity The Effective Protection Study results to pursue a dialogue between the Bank could not be implemented because cal- and Tunisia which eventually led, culations were not based on direct starting in 1986 and continued until price comparisons and could not in- now, to a significant liberalization cude the effect of quantitative re- of the economy and of the industrial strictions which were of paramount sector, including EMIs (para. 209). importance. Sustainability 29. It could be argued, that the part of the loans which was used for 25. The EMI I and II projects are general lending to industry has proba- not sustainable. EMIs did not develop bly helped strengthen the manufactur- because of changing economic condi- ing sector. In this case, the tions and, more importantly, because projects' effects, in some sense, are the drastic shift in policy led to the likely to be sustainable. Also, BDT cancellation of a number of large benefitted from strengthened project scale EMI projects and, by extension, supervision, if only because it was to the elimination of adjunct ancil- determined to recover large arrears. lary industries. This "ripple" effect BDET's recently improved performance was compounded by potent structural is, however, not directly connected to and technological constraints through- the outcome of EMI I and II. out the sector, and the much slower- than-expected progress in removing 30. The spirit of the two loans was these constraints which the EMI pro- to foster the development of a sus- jects were not able to foster. taiable EMI sector. But a large part - xii - of the funds earmarked for EMI financ- around to adopt a new, more liberal ing had to be cancelled, a number of industrial and trade policy could not subprojects a-e in arrears, and EMI be done through limited financing of export performance, although improv- selected priority EMIs. Such projects ing, remains limited to a few export- could not provide a broad enough vehi- ing companies benefitting from a com- cle for substantial and lasting chang- parative advantage stemming from the es in the overall industrial incen- high share of low-cost labor in value tives framework. added. For these reasons, the sustainability of the EMI projects is Development banks (specially unlikely (paras. 210-217). newly established oneF such as those created in 1982-1983) rarely possess Lessons of Experience the technical and managerial capabili- ty to intervene efficiently in entire 31. The lessons of experience from subsectors where special and thorough the EMI projects are detailed in paras. knowledge of technological problems is 218-221 and summarized below. required. Similarly, the establish- ment of technical assistance institu- 32. Regarding the conceptualization tions to provide in-plant advice and process of the project: help increase quality control con- sciousness, although justified, usual- * It is difficult to pursue con- ly takes a very long period of time to currently objectives aiming at foster- achieve the desired results. The time ing the development of an entire in- needed frequently goes beyond the dustrial subsector, and objectives normal implementation period of Bank aiming at strengthening a particular financed lines of credit. financial development institution, particularly when the latter lacks the Before embarking on an industri- necessary technical competence to al subsector operation, it is impor- positively participate in subsector tant to clarify with the Borrower development. their detailed development strategy for this subsector. Early disagree- * Subsector studies should be ments between the Bank and the Tuni- initiated when there is a genuine sian authorities on the economic jus- desire to contribute to the develop- tification of some large EMI complex- ment of the subsector, and not by the es, which later on collapsed, were need to justify at any cost an addi- partially ignored due to pressure to tional line of credit to a DFC. lend. Clarity of objectives and true commitment of the Borrower to agreed * A subsector study such as the detailed objectives are essential to one which provided the base for the the sustainability of subsectoral EMI project, must have enough vision projects (para. 219). and be broad enough to comprehend policy linkages between the subsector, 33. As regards imRlementation of the the industrial sector, and the economy project, useful lessons have been at large. Should overall policy con- learned: straints prove too important to be dealt with through a subsectoral ap- Fixing too detailed subsectoral proach, lending should be avoided. output, employment and investment targets is illusory and dangerous. * The EMI I and II projects demon- Directed sectoral lending cannot suc- strated that bringing the Government ceed when the financial institution in - xiii - charge of approving subprojects is in It is quite difficult to provide no real position to influence develop- effective technical assistance to a ment in a particular subsector, and ubsector when there are, at the start has no direct leverage on policy and of the project, no proper institutions technical constraints affecting this in place able to perform this task. A subsector. very careful analysis of the possible link between project financing and * Project implementation can only technical assistance has to be carried suffer from the need to achieve widely during project identification and different objectives. In this case, evaluation. If this is not done, fostering development of the EMI sub- risks of &aiure are compounded and sector (this approach having not been must not be hidden. tried before through a DFC) while trying to also provide general purpose Specialized technical centers lending to industry and strengthening should be fully autonomous from the the concerned DFC resulted in a di- State when dealing with the private chotomy very difficult to handle. industrial sector. When they have to perform regulatory functions, the * Project implementation suffered latter must be clearly defined and be from the failure to have put in place performed separately. a proper supervision and surveillance mechanism into the enterprise, linked When devising an industrial to the short/medium-term fluctuations subsector operation dealing with com- of the economy. A subsectoral ap- plex technological and quality control proach to industry lending can succeed issues, it is not enough to set up or only if warning signals can be estab- revitalize a few domestic technical lished and function properly. It is centers. A much broader and systemat- not feasible to develop a so called ic review of these issues should be "new approach" while continuing using undertaken, and problem solving mecha- traditional procedures, such as rou- nisms should be introduced, including tine supervision of a DFC. Supervi- access to transfer of technology and sion must be particularly intensive R&D from abroad (para. 221). when new systems are introduced (para. 220). 34. The experience with technical assistance tt the EMI sector provides some interesting insights: PROJECT PERFORMANCE AUDIT REPORT TUNISIA ELECTRICAL AND MECHANICAL INDUSTRIES PROJECT (Loan 2113-TUN) I. INTRODUCTION 1. A Bank loan for US$30.5 million was made in 1982 to the Government of Tunisia, to provide financial and technical assistance to the Electrical and Mechanical Industries (EMIs) and introduce a number of sector policy improvements and institution building efforts. This loan was followed by a second one in the amount of US$54 million in 1985. 2. The Bank's decision to provide financing to EMIs was due to the belief that the continued rapid expansion of other major industrial subsectors was no longer possible because of market and resource constraints. Export outlets for textile/clothing and phosphate products were assumed to become limited; local resources for food products and chemicals (other than phosphates) were said to be insufficient or uncertain; and the output of construction materials was soon to meet domestic demand. Further growth in the imports of EMI goods could have imposed a heavy burden on the balance of payments and constrained the expansion of the overall manufacturing sector; it was, therefore, necessary to develop a well-balanced EMI sector that could provide standard intermediate and equipment goods. For these reasons, the Government intended to give first priority to EMIs in the industry sector and requested Bank's assistance. 3. In 1980, the sector remained small, amounting to 1.53% of GDP and 13% of value added in manufacturing. EMIs represented 3.8% of total employment and 17% of manufacturing employment in the country.- The sector was largely inward-oriented. Exports of EMI goods amounted to less than US$80 million, representing 8% of manufactured exports and 3% of total merchandise exports (Annex 10). EMIs enterprises were largely in the public sector. Public and semi-public enterprises accounted for most of the sector's large enterprises, and for about one-third of the sector's employment; they generated some two-thirds of the sector's output. 4. After 1970, there was a concentration of public investment and new enterprises in assembly industries with lower investment costs per worker. These were import-substituting industries for consumer durable goods (automotive equipment, household appliances). As-embly industries relied primarily on imported components. Consequently, the growth of value added and of labor productivity slowed down, averaging respectively 8.8% and 1.7%1' per annum in 1970-78. The shift towards assembly industries had also led to insufficient sectorial integration and development and increasing pressures on the balance of 1 See Annexes 2 and 3. - As compared with 4.3% for labor productivity in manufacturing as a whole during the same period. -2- payments because of growing import of EMI intermediate and equipment goods. The ratio of value added to output decreased and the share of EMI goods in total imports increased. 5. The domestic market was small. To exploit its potential fully, Tunisian thus had to export. The potential for exports had already been harnessed in a few areas by foreign enterprises (assembly of electronic and electro-mechanical goods and components) as well as by some local firms (electrical transformers, ship repair, cutlery). But it was felt that exports could be expanded further. 6. There was also a potential for further efficient import-substitution of intermediate and capital goods with simple and intermediate technologies. A first group included goods with domestic markets large enough to sustain economic operations (ferrous castings, industrial boilers and furnaces, steel structures, agricultural implements, cranes, pumps, medium-power electric motors and transformers). A second group of goods comprised some capital goods with a small domestic market, making it necessary to find export outlets to absorb 30% to 50% of production while foreign partners were necessary to provide the necessary know-how. This group included mainly power engines, their components and parts, automotive equipment, agricultural machinery and tractors (Annexes 8 and 9). 7. Local manufacturing of consumer end-products based on assembly operations was thought to be generally uncompetitive, weakly intpgrated in the sector and not labor-intensive, due to the inter-related disincentives resulting from protection and price control polici3s. 8. The Tunisian authorities believed, as early as 1978, that the expansion of EMIs was justified in view of the subsector's moderate development at that date and of its development potential. The availability of islatively cheap labor was considered a comparative advantage in EMIs which could be characterized by a relatively high degree of labor intensiveness, provided that appropriate technologies be selected. Another advantage was Tunisia's proximity to potential export markets (including, but not exclusively. Europe) and the growing, though modest, domestic demand. 9. Detailed analyses of the subsector's development prospects, however, had not been carried out. Preliminary mar!3t studies for the major projects already under discussion in 1978 (diesel engines, tractors and agricultural machinery, automobile and truck assembly, foundry) had been prepared, but not to a sufficient depth. The newly created Industrial Development Finance Division (IDF) of the EMENA Region held the view, supported by the Tunisian authorities as well as by the conclusions of the Industrial Projects Department's Reconnaissance Mission which visited Tunisia in June 1978, that an in-depth study of the entire sector's development potential was needed. 10. In 1978, the question of a possible eighth line of credit to Banque de Ddveloppement Economique de Tunisie (BDET) was raised by the Tunisian authorities. Given the maturity of BDET as a Development Finance Company (DFC), a new loan to BDET to help it meet its financing requirements was believed to have been difficult to justify in the absence of clearly defined, specific economic objectives. A Bank mission's attention was consequently attracted by the EMI sector, because, at that time, the thinking of the Tunisian authorities, -3- of BDET, other Tunisian banks, and private investors appeared to be distinctly oriented toward the EMI sector. 11. Obviously, a decision to involve the Bank in the EMIs through an industrial sector loan (which would have been the first of its kind for the new IDF Division) could only be taken on the basis of improved knowledge of the sector by the Bank. In August 1978, it was decided to send an industrial sector mission to Tunisia. The latter visited Tunisia in March 1979 and submitted its final report in June 1980."i 12. The mission concluded that, before potential EMI development could be fully exploited, various constraints had to be removed. The first set of constraints originated from the inadequacy of the policy and institutional framework to promote higher factor productivity and efficiency, to encourage export competitiveness and orientation, and to generate efficient sectoral strategies. The second set, related to the first one, was largely internal to the enterprises and resulted from the weak design and management of production operations (product-mix; scarcity of firm specialization and sectoral integration, capacity planning and capital intensity, labor productivity). 13. Price controls, of a cost-plus type combined with high protection, did not encourage enterprises manufacturing end-products for the local market to increase efficiency and expand exports. The incentive framework for export- oriented production (Law 1972-38) did not provide incentives for increasing the competitiveness of exports and for identifying new markets for new products with substantial local value-added. 14. The biases of the Investment Code (Law 1974-74) favoring large enterprises irrespectively of their efficiency resulted at times in substantial excess capacity (including in some EMI subsectors), insufficient specialization of product-mix and excessive horizontal and vertical integration within large and medium firms, which were encouraged to expand their overall size rather than resorting to subcontracting to small scale industries when this was economically appropriate. The latter shortcoming had affected particularly the EMIs, which were thought to have failed to achieve their key role of providing the necessary linkages between the manufacture of parts and of end-products. 15. The general constraints on manufacturing, which affected particularly the EMIs, were primarily: a. the structure of protection favoring consumer and product manufacturing; b. the weakness of incentives for establishing partnership or association with foreign firms; c. the shortcomings of the mechanisms and institutions in charge of planning strategies and project identification/preparation. V Review of the Electrical and Mechanical Industiles. June 4, 1980, Report No. 2666-TUN. -4- 16. The main issues and constraints specific to the EMI sector were: a. the insufficient product standardization and the excessive diversification of product-mix resulting from small domestic markets and hampering the exploitation of economies of scale in many production lines; b. the scarcity of specialized machinery and the low productivity of labor (due in particular to the shortage of specialized mechanical workers and production engineers), relatively to Europe and other competitors in some developing countries; and c. the low and irregular quality of products. 17. The Bank's Report made the following recommendations, in order of importance: a. the Government should revise the chapters of the protection and indirect taxation systems affecting the EMIs, with a view to remove the negative effective protection on intermediate and capital goods industries and to induce cost reduction through lower protection in the firms producing EMI end-products; b. it should also revise the objectives of the Investment Code to encourage EMIs to export, create employment at lower cost, improve quality, specialize, and intensify inter/intra-sectoral linkages between local and export-oriented firms; c. the Ministry of Industry should establish a "Technological Center" for EMIs to be in charge of standardization, quality control and technological assistance (in technological choice and adaptation of equipment and production processes, operations management, selection of foreign associates, etc.). 18. In 1980, agreement was reached between the Tunisian authorities and the Bank that the EMIs were indeed a priority subsector and that a loan focussed on this subsector for BDET merited high priority. In addition to providing funds for financing projects in this subsector, such a loan was to include two important components. The first was the formation of a Technological Center in the lines proposed in the Bank's report. The Ministry of Industry strongly supported the formation of such a Center for the EMIs (later on, a second institute to specialize on quality control was to be proposed). The second component was to include a review of tariff policies (as they related in particular to EMIs) during the course of project preparation. That review was to focus on how to foster the development of EMIs by removing the prevalent negative protection rates resulting from existing tariffs in inputs and also on reviewing how to protect effectively the output of that sector. 19. One disquieting factor was, however, the Tunisian's decision to move ahead with the establishment of a new mechanical complex to produce mainly -5- tractors and diesel engines and also to set up other projects such as several new automobile assembly plants. The Bank's view was rather to concentrate on the expansion of labor intensive enterprises with an actual or potential comparative cost advantage (e.g., foundries, steel platework and construction, some electrical products such as transformers and small motors, cutlery, ecc.). The Tunisian authorities, although in favor of such enterprises expansion, also insisted that the large, new industrial complexes should be set up. Their view was that it was appropriate to proceed with these operations on the ground that various externalities (training of labor force, backward linkages, etc.) would justify the higher production cost. Despite such disagreements between Bank staff and Tunisian authorities on some of the broader issues of the development strategy of EMis and because of the deep earlier involvement of the Bank in the sector through a basic review of the electrical and mechanical industries, it was decided vo go ahead with the financing of an industrial sector loan to EMIs through BDET. II. OBJECTIVES, CONCEPT AND DESIGN THE FIRST EMI PROJECT A. OBJECTIVES 20. The main objectives of the project were the following: a. to promote the efficient development of EMIs and other industries by financing, through a line of credit to BDET, new and modernization/rehabilitation projects in priority subsectors; b. to increase the delivery of technical assistance to EMIs in two complementary ways by: i. fostering product quality by establishing standards and quality control of products through the establishment of an Institute for Standardization and Quality Control of industrial products; and ii. delivering technical assistance to EMI firms and providing on-the-job training, using the services of a Technical Center (CTM) to be established. c. to help implement changes in incentives specific to EMIs in the area of protection, exports and financing. 21. The basic objective of the project was to foster the development of EMIs, to improve their efficiency and productivity, and to increase their production of competitive capital and intermediate goods (PCR, para. 8). The provision of financing to other industries at BDET's request [because BDET wanted to retain some flexibility in managing its borrowed resources, many of which were said to be tied to specific types and sizes of subprojects (i.e., small-scale -6- industries) or to particular bilateral sources of procurement] was assumed to be secondary an objective, since the Bank loan was to be considered as financing a first genuine sectoral-oriented operation with particular sector objectives. This sectoral approach was an important departure from previous operations which consisted either in finaL!cing industries across the board or horizontal operations such as small-scale industries. 22. Another less important objective of the loan was to continue strengthening BDET as an institution. It was agreed that: a. BDET would commit itself to review annually its interest rate structure and make necessary adjustments to ensure a reasonable spread of, at least, 2 percentage points, starting 1982. b. The Government would make necessary arrangements so as to help BDET maintain the required spread after rescinding its financial assistance to BDET before the end of 1986. c. The foreign exchange risk on BDL foreign exchange resources would continue to be borne by the .vvernment. d. BDET would hire additional staff to deal with project promotion and project supervision. B. COMPONENTS 23. The project aimed at meeting the above objectives through the following components: a. A loan to BDET (US$30 million), composed of a US$28 million credit line subdivided in two parts earmarked to finance respectively EMIs (US$14 million) and other industrial subprojects (US$14 million), and of a US$2 million subloan to the Government to finance part of the technical assistance components. b. A technical assistance component for EMIs, through a Technical Center (CETIME) to be set up, its foreign exchange operating costs (US$1.5 million) for the first two years were to be financed under the proposed subloan to the Government. c. An Institute for Standardization and Quality Control of industrial product (INNORPI) to be set up. The foreign exchange cost of technical assistance to the Institute (US$300,000) was to be financed under the proposed subloan to the Government. d. A study of effective protection in manufacturing, with priority to EMIs, in the context of an incentive package for EMIs (changes in EMIs tariff protection, incentives to exports, etc.); the foreign exchange cost of technical assistance for the study (US$200,000) was to be financed under the proposed loan to the Government. e. A front-end fee of US$0.5 million. -7- f. Two studies were also to be undertaken to define long-term development strategies for the foundry and platework subsectors. C. VOLUME OF LENDING TARGET 24. Although the basic project objective was to finance EMIs as indicated by the title of the Staff Appraisal Report, "Electrical and Mechanical Industries Project," almost half of the loan proceeds was to be used for general purpose lending to industry. The official reason given was that BDET wanted to retain some flexibility in managing its borrowed resources, many of which were said to be tied to specific types and sizes of subprojects (such as KfW loan, which was small-scale industries oriented) or to particular sources of procurement (such as French and Italian credit lines). 25. The Bank agreed to BDET request because, although the amount for financing EMI projects was relatively small (US$14 million), it would finance projects with a total cost of US$80 million (about TD 40 million), equivalent to about 20% of total EMI's projected investments in 1982-83. Moreover, it was assumed that BDET would finance additional EMI projects through other available lines of credit, the main objective being that a maximum amount of funds be made available for EMI investments. 26. In addition, no free 1.mit was granted in order to allow Bank review of all EMI subprojects and a feedback to BDET on its lending orientation in EMIs. It was e:timated, based on BDET's pipeline (Annex 11), that up to twelve EMI priority projects could be submitted to the Bank, with most of the subloans ranging from US$500,000 to US$1 million. In addition, all other priority EMI projects to be financed by BDET were to be reviewed by the Bank and one out of three subprojects financed by BDET in non-priority EMI subsectors (i.e., consumer goods) was to be sent to the Bank for ex post information and review. This indicated that the Bank was concerned about the future direction taken by BDET in lending to the EMI sector and also that special subproject eligibility criteria imposed on EMI projects would be effectively applied. D. SPECIAL ELIGIBILITY CRITERIA AND THE PROTECTION ISSUE 27. These special eligibility criteria for EMI projects originated from the Bank's desire to ensure the competitiveness of the EMI subsector. Not only EMIs were protected by import dutii. but there were many quantitative restrictions. The Operations Senior Vice President had insisted that quantitative restrictions on imports of priority EMI products should be applied only in duly justified "infant industry" circumstances and should be abolished after one year. Import duties should not exceed 18%. However, under exceptional circumstances, a tariff increase to up to 26% would be acceptable provided it was accompanied with an ! Total Sixth Plan projected EMI investment was TD 480 million, of which TD 220 million in 1982-83 subdivided between TD 90 million in priority sectors and TD 130 million in nor-priority sectors (see Annex 7). -8- agreed rogram to reduce the tariff to no more than 18% over a five-year period.- 28. The Tunisian authorities' position was that licenses were designed to provide the necessary protection for the growth of the country's nascent engineering industry for the following reasons: (i) Tunisian consumers tended to prefer foreign products, even when these were more expensive than local products of similar quality; (ii) in view of the existing foreign exchange controls, imports of capital goods often provided importers with foreign exchange abroad (through "commissions"); (iii) foreign products were sometimes sold in Tunisia at "dumping prices"; and (iv) the absence of import controls would discourage joint ventures with foreign partners, especially in export compensation projects which the Government was particularly keen to encourage at that time. Furthermore, the Tunisian authorities indicated that their system of import controls was an important tool for balance of payments management, which they wished to retain. Finally, the Tunisian authorities indicated that they did not want to dismantle that system. 29. In view of the Tunisian position, it was agreed, during negotiations, that: a. import duties on priority EMI goods should remain below 18% (with the exception of steel structure and platework goods for which the Tunisian authorities agreed to reduce import duties from 26% to 21%), and not to raise tariffs above these levels without prior consultation with the Bank; b. the import licensing system would continue, and for projects involving a foreign partner, the Government would ensure (via temporary prices controls and, if necessary, subsidies) that the ex-factory prices, during the first three years of operation, would not exceed 118% of the ex-factory price of the project's foreign partner in his domestic market, plus an estimate of insurance and transportation costs (or 121% for steel structure and platework goods). After these first three years, any project producing at prices above the 118% threshold would be exposed to gradually "increasing import competition," which would force the project either to increase its efficiency or become unviable, and withdraw from the market; c. in the EMI projects without a foreign partner, agreed price levels (118% and 121%) would also be applied, and the comparable foreign Locally produced capital goods were protected by duties ranging from 0 to 16%, with the major exception of steel structure and platework products which enjoyed a 26% nominal protection. But, because of administrative difficulties, most local capital goods industries had de facto received no nominal protection and suffered from a negativ, effective protection ranging between -10% and -15%, thus hampering severely the development of these industries as well as of exports. However, to compensate for the negative effective protection, import licensing procedures had been used for EMIs. -9- prices be determined by the Government on the basis of the ex- factory prices in the domestic market of the major traditional commercial partners of Tunisia. Import licensing would continue when needed; d. export-oriented EMI subprojects in non-priority subsectors would also be eligible for Bank financing if based on buy-back arrangements with foreign partners covering at least 70% of the subproject output and if appropriate evidence of the buy-back arrangements was submitted. 30. As a result of these agreements, the Government was to. a. apply existing tariff duties on imports of EMI capital goods and supplies which could be produced in Tunisia. As indicated above, in fact, tariff duties were not previously applied effectively. A list of EMI products was established to be used systematically by the authorities responsible for granting tariff exemption and import licenses; b. maintain below 18% the tariff duties referred to in para. 30.1. above, and consult with the Bank before it increased any of these duties above 18%; and c. take appropriate measure to ensure that priority EMI products protected by import licensing procedures be charged reasonably competitive prices in comparison with similar foreign products. 31. The 1980 Bank report on EMIs had led to believe that existing EMI enterprises were producing capital goods at reasonably competitive prices, on average 120% of European FOB prices. However, no detailed study had been made of this and it was also agreed with the Government that the appropriate levels of protection for EMIs would be subsequently discussed through an Effective Protection Study for manufacturing to be carried out in the context of the project. 32. The practical implication of this negotiated agreement was that, since the Tunisian authorities did not wish to dismantle the protection system, existing quantitative restrictions on EMIs would remain in place for three years (the original Bank position was to abolish them in one year) and then be gradually removed, but without specifying the measures to be taken. In addition, protection was given to EMIs through effective enforczment of tariffs which, until then, had never been really applied on capital goods for administrative reasons. This indicated the difficulty to negotiate a full-fledged trade liberalization reform in the context cf a lending operation limited to one single subsector (EMIs) and for a modest amount (US$14 million). 33. Since it seems that there have been some doubts on the practical application of the negotiated agreement, it was decided that for each subproject, eligibility criteria (along those described in para. 29) would have to be met. However, this was to be a very challenging conditionality since, for example, the borrower had to prove that its proposed products would be competitive, in 1rice - 10 - and in quality, with similar foreign products. It was specifically stated that "in the case of subprojects protected by import licensing and quotas, the comparable foreign prices should be quoted and compared to the projected prices of the subproject's output, and the non-confiuential features of arrangements made (technical assistance or licensing contract with foreign firms, other local arrangements) to ensure acceptable specifications and quality standards of the products." (SAR, para 6.09). THE SECOND EMI PROJECT 34. The main objectives of the Second EMI Project approved in 1985 built upon those pursuel under the EMI-I Project. The main difference vas that EMI financing was provided not only through BDET, but also through two newly created development banks, BTKD and !,TUSID, which were interested in financing electrical and mechanical industries. 35. Main objectives were: a. to promote the development of EMI's by financing throtgh BDET, BTKD and STUSID, the creation, expansion and modernization of viable enterprises in priority (capital and intermediate goods) subsectors; and b. further foster efficiency in the EMI development in three complementary ways: i. assist CETIME to expand the areas where it provided tecknical assistance, guidance, and on-the-job training to EMI firms; ii. help INNORPI initiate its certification activities (in order to make effective the technical standards it had already formulated) and establish its technical assistance and audit program in quality control, in order to facilitate local integration and improve exports; and iii. introduce further improvement to the EMI protection policy framework to incite firms to achieve more promptly competitive levels of efficiency and productivity. 36. The EMI-II project was supported by a loan of US$54 million to the Government of Tunisia, which passed on the loan proceeds to: a. BDET (US$20 million), BTKD (US$15 million) and STUSID (US$15 million), through subsidiary loans to cover part of their resource needs for the financing of eligible EMI projects; and b. CETIME (US$2.1 million) and INNORPI (US$1.9 million) as grants to finance the extens:on of foreign technical assistance and the creation of new laboratories and operational sections. - 11 - 37. Eligibility criteria for subprojects were similar to those agreed upon under the EMI-I Project (para. 27 above). 38. No basic change was introduced to the conclusions reached in 1980 on the relative positions of different production lines (by nature of the goods produced) vis-A-vis the levels of protection, domestic resource cost and financial and economic rates of return.- Priorities adopted under the EMI-II Project, therefore, remained basically the same as those under the EMI-I Project, i.e., capital and intermediate goods for which tariff duty protection was limited to 18% (or 21% in the case of steel structure and platework products). 39. In the case of goods protected by quantitative restrictions (QRs), it was understood that QRs would not be allowed for new EMI goods (not produced in Tunisia) except under very special circumstances and for a minimum period of five years from the date of first production. 40. The Government and the Bank were to jointly review Tunisia's trade and industrial policy on the basis of the work undertaken by an Industrial Sector mission in January 1985. This meant that wider policy reforms were not to be addressed in the context of the EMI-II Project. III. IMPLEMENTATION EXPERIENCE A. FINANCING OF ELECTRICAL, MECHANICAL AND OTHER INDUSTRIES 41. The EMI-I loan was declared effective in December 1982, seven months after signature. The delay in effectiveness was modest; it was due essentially to administrative steps which had to be taken before formally creating the two technical institutions (CETIME and INNORPI) financed by the Loan and for nominating their respective directors (PCR, para. 14). 42. Five subprojects (of which three EMIs) were sent to the Bank for approval immediately after effectiveness and by March 1983, BDET's pipeline for industrial projects was strong, projecting a total commitment of about US$15.5 million by June-July 1983 of which US$6 million for EMIs. By October 1983, the equivalent of about US$6 million had been committed for each of the two components of the BDET loan. t Namely: i) the more protected the production, the higher its domestic resource cost and the lower its rate of return; (ii) the productions primarily oriented to the domestic market (consumer/durable goods, as opposed to capital and intermediate goods) were the most protected; and (iii) enterprises in public ownership were involved in productions that were significantly more protected than others. - 12 - 43. One year later, i.e. by March 1984, it was felt that BDET's pipeline for general industrial lending was adequate and commitments of funds made available to BDET were US$13 million. However, it already appeared that a 6-month postponement of the project commitment date (December 31, 1984) was needed in order to fully commit the US$14 million part of the loan earmarked for EMIs, since commitments had remained at ab.ut US$6.65 million at that time. BDET's pipeline for EMI projects only totalled V $4 million. This was confirmed in November 1984 when BDET requested that the original completion date be postponed to September 30, 1985, i.e. by nine months since it became clear that the EMI line of credit would not be fully used in time. 44. In October 1985, due to delays in preparation of EMI subprojects and to the then stated "need" to provide additional funds for EMI lending while a second line of credit to EMIs (Loan 2554-TUN) which had been approved in May 1985 was not yet effective (effectiveness was expected by the beginning of 1986), the final commitment date was again extended, this time to June 1986. 45. The levels of commitments and disbursements in 1983-1984 were not maintained in 1985-86. By March 1986, disbursements had reached about US$20.6 million only. However, the situation was particularly difficult for EMIs as shown below (in 000 US$): Original Disbursed Cancellations Revised Amount Amount EMI 14.025 6.510 5.1802' 8.845 Other Industries 106 2.040 1' Sub-Total 28.050 18.130 7.220 20.830 Technical Assist. 2.000 1.990 -- 2.000 (CETIME, INNORPI) Front-end Fee 450 450 -- 450 Total 30.500 20.570 7.220 23.280 46. The loan was closed on December 31, 1987 and an unused amount of US$2.4 million was again cancelled at that date. Such cancelled amount was added to previous cancellations for US$7.2 million, thus totalling US$9.6 million. As a result, only about 70% of the loan was disbursed (Annex 1) of which about US$6.5 Z Of which US$345,000 for approved projects and US$4.835 million for unallocated funds. 1i Of which US$216,000 for unallocated funds. - 13 - million for EMIs, i.e. less than half of the amount originally anticipated. It is to be recalled that the basic objective of the Project waq to finance EMIs. 47. It is quite interesting to note that nobody seems to have anticipated the difficulties that started to plague the EMI sector in the course of 1984 and subsequent years. In fact, based on 1982-1983 commitment data, which led to believe that the loan could be fully committed rather quickly, discussions were 'nitiated in early 1984 between the Bank, the Government, BDET as well as two new development banks (BTKD and STUSID) on a second line of credit to EMIs as a follow up to that extended under the project. Preparatory work on the new line started in mid-1984, leading to an appraisal in the fall, negotiations in early 1985 and Board approval in March 1985 (PCR, para. 19). 48. It seems that despite an evident slowdown in the level of EMI project approvals by end-1983 (Annex 17), BDET was interested in obtaining a second line of credit for EMIs as the financial terms attached to this line were more favorable than those of the first (interest rate was 7.75% against 11.6% for EMI- I). BDET requested--and the Bank accepted--that the uncommitted part of the first line of credit be cancelled when the second line of credit would become effective, i.e. in June 1986. Then, it appears that at that time, the situation of under-commitment under the EMI-I line was explained by BDET's wish to keep upcoming EMI projects for commitment under the new line soon to be established (PCR, para. 20) and not by the feeling that demand for investment financing in the EMI sector was drying out. 49. However, as a result of the general slump in economic growth, EMI performance in 1982-1983 had fallen short of expectations: growth in value added was 8.9% on average during these two years and declined further to 7.5% in 1984. This compared with an average annual growth of 15% in 1980-1981 and a plan target of 18.7% per annum in 1982-1986 (see Annex 6). 50. However, there were more positive signs: employment in EMIs had grown from 35,500 in 1980 to 42,500 in 1983 and 44,300 in 1984. Also, in comparison to TD 198 million of projected outlays for the 1982-84 period, actual EMI investments had amounted to TD 177 million. It represented, however, only about 17% of manufacturing investment against a targeted 22%, because of delays in the implementation of particular projects in the EMI sector (e.g. vehicle assembly and electronic components projects). 51. EMI investments planned for 1985 amounted to a total of TD 106 million of which TD 84 million concerned projects under implementation (e.g., extension/modernization of the SOFOMECA foundry, renewal of equipment at the El- Fouledh steel mill, vehicle assembly and components) with the remaining TD 22 million covering new projects, mainly in basic metals (e.g., steel wires and metal rolling) and mechanical industries (vehicle assembly and components, and diesel engines). It was thus believed that the Plan's investment targets (1982- 1986) were likely to be met. 52. In addition, the implementation of the EMI-I loan was judged "satisfactory" at the time of EMI-II appraisal. As of January 31, 1985, eight EMI subprojects had been approved by the Bank and involved metallic construction (leaf springs, steel and aluminum pipes, insulated vehicular containers, wheel - 14 - frames), mechanical equipment (engines for export, wood-working machinery), and electrical equipment (transformers and water-heaters, accessories for electric lines). (See Annex 12). 53. However, what was apparently not emphasized at that time was that BDET project approvals in the manufacturing sector and in the EMI subsector had already slowed down substantially in 1983 as compared with the peak reached in 1982 (see Annexes 16 and 18). Even worse, approvals of EMI projects had been estimated at TD 11.7 million for 1984 while in reality TD 3.7 million only were approved during that year, the lowest level in years. Such a "statistical" discrepancy is revealing of the state of mind at that time. One must also note that a similar 1984 fall in EMI approvals took place in 1984 for BTKD and STUSID and was, thus, not purely coincidental (see Annex 17). 54. At that time, economic performance was already deteriorating due to a number of exogenous factors and Government policies. Serious macroeconomic imbalances emerged. World oil prices started to soften and the depletion of Tunisian oil reserves resulted in a permanent downward trend 4n oil exports. Manufacturing export and tourism also slowed down*as the Tunisian currency was allowed to appreciate vis-A-vis its competitors and as EEC demand fell with recession. Other exogenous factors also contributed to the increasing difficulties and worse-than-average agricultural performance in 1984 led to increased food imports. 55. The Government did not resnond quickly to stabilize growing macroeconomic imbalances and to adjust to the change in Tunisia's resource position. Instead, expansionary policies continued until 1985. As indicated already above for EMIs, investment remained high, with the greater part coming from the public sector, and wages increased significantly faster than productivity, depressing demand for labor. Consequently, the balance of payments detexiorated rapidly and, in 1984, the current account deficit reached a peak of 11% of GDP. These policies, while helping to achieve an average growth of 4.5% p.a., also led to a rise in inflation to 10% p.a. The budget deficit increased to 6.7% of GDP on average. The Government resorted to continued external borrowing, increasing the debt ratio to GDP to 46%. 56. In 1985, when the EMI-2 loan was being appraised, Tunisia began experiencing difficulties in obtaining external financing for its high current account deficit and, with very low reserve levels, it became evident that the policies pursued were no longer sustainable. Import restrictions were extended and heightened and customs tariffs were raised, increasing the anti-export bias of the trade regime. Government investment expenditures were also substantially cut, reducing total investment and the resource gap, real wage increases were frozen and a more flexible exchange rate policy was adopted, leading to a depreciation of the dinar by 5% in nominal terms. Through these stabilization policies succeeded in reducing the current account deficit from its 1984 level to 7.1% of GDP in 1985 (and 8.3% in 1986), import restrictions caused shortages of spare parts and raw materials, hampering industrial activity. Manufacturing value added grew by about 5% only in 1985 and EMI output, by 5% also compared to 7.5% in 1984. In 1986, EMI value added growth was 0.4% and even fell by 0.1% in 1987. Fixed EMI investment (at current prices) declined from TD 71.6 million in 1984 to TD 69.6 million in 1985, TD 60 million in 1984 and TD 45.8 million in - 15 - 1987. In constant dollar terms, the decrease was even greater, as the dinar real effective exchange rate declined by about 30% in 1986-1987. It was only in 1988 and mainly in 1989 t it EMI output and investment growth was resumed. 57. The problems affecting the EMI sector were particularly acute in the car assembly subsector. By end-1984/early 1985, the Government decided to withdraw from four previously envisaged ventures in the automobile assembly subsector for lack of resources and realization that automobile assembly was basically uneconomic. At the same time, the Government decided to allow Tunisians working abroad to import up to two new or second-hand vehicles each year in an attempt to maintain the flow of remittances. Imports of second-hand vehicles thus rose from 6,500 in 1984 to 21,400 in 1985 and 17,600 in 1986. In parallel, the number of vehicles produced in the country went from 12,000 in 1984 and 13,700 in 1985 down to 6,300 in 1986. Also, the poor performance of agriculture had a considerable impact on the purchase of tractors, which fell from 1,900 in 1985 to only 500 in 1987, at the time when the new diesel engines and tractor assembly (CMT) was coming into production, with disastrous consequences for this project. 58. This crisis had serious consequences for the many EMI industries producing components for the transport equipment and agricultural machinery subsector. A well known example is the Bank financed SOFOMECA foundry expansion which could not sell castings as expected to CMT and STIA because of this crisis. 59. A key element of the Tunisia industrial strategy had precisely been to center investments in the transport equipment and agricultural machinery subsector. Assembly plants were to be set up and be progressively integrated with a growing local production of components and parts. A typical case was the STIA car/truck assembly plant which was to buy diesel engine from the new CHT mechanical complex which itself was to be supplied with castings from the SOFOMECA foundry. 60. The Government's plans had led private investors to develop a number of projects to produce components and inputs for the automobile assembly subsector. In 1985/86, these projects were dropped, with the result that overall the sector became significantly depressed and in serious need of reorientation. Some enterprises began reorienting their operations (especially towards export). Other enterprises, however, could hardly encourage such shifts in their operations because their production was too directly and restrictively linked with the car assembly lines which the Government, in 1983-84, was planning to launch. In the latter enterprises, the only hope was in a reorientation of a significant part of the whole EMI sector but this proved time consuming and difficult, particularly when the Tunisia economic policy started to change in 1986 and became oriented towards liberalization and increased international competitiveness. 61. Today, most EMIs are still basically turned toward the domestic market (in the old line of import substitution) and have been hit by the liberalization of the economy. However, an increasing (although relatively limited) number (operating as subcontractors, or not) have been able to achieve levels of efficiency that permit them to be competitive on the international market and to work increasingly for export. While the share of EMI exports in total EMI output remained low in 1982-85, i.e. 10/11%, and averaged 16% in 1986-87, it jumped to - 16 - 25% in 1988-89. The main mechanical industry exports are castings, spring leaf, small electric transformers and motors, welded tubes, and small diesel engines. Main electrical products exported are electric wires, circuit breakers, measurement instruments, bobbins, etc. Obviously, a number of EMIs have had to reorient their production towards export since the domestic market was too small to absorb their output and the new industrial policy was to promote efficient exports. EMIs would appear to be competitive when products require a large relatively low cost manpower and thus attract foreign contractors who buy all or part of the EMI plant output. Other EMIs with low labor intensity and high imported machinery and capacity requirements are evidently poorly placed to meet foreign competition and eventually export. Interestingly, EMI exports represented about 60% of the subsector increase in output during 1988-89. 62. The EMI sector has proved to be a difficult one, technically as well as commercially. Its market is primarily local; but even the local market is not easy to secure because the appeal of imports remain high. Technically, many EMIs still suffer from insufficient standardization, lack of quality control, lack of supervisory personnel and qualified technicians and engineers, and insufficient applied research. The EMI I and II projects definitely underestimated the time needed to overcome these problems, despite efforts made to provide technical assistance through CETIME and INNORPI (see below). 63. Finally, there appears to be significant over-capacities as well as over-integration. Many entrepreneurs seem reluctant to trust other local entrepreneurs for producing some of the imports in this production line. Subcontracting remains insufficiently widespread. 64. As a result of all these macroeconomic and sectoral EMI problems, Loan 2113 was only partially disbursed and more than half of the funds available for EMI investments had to be cancelled. Although, theoretically Loan 2554 was approved in 1985 to pursue Bank financing for up to US$54 million in the EMI sector (of which US$20 million through BDET), the continued deterioration of demand for EMI projects in 1985-1988 also led to low disbursements, i.e. about one third only of the original loan amount. BDET's approval of EMI projects amounted to TD 21.1 million in 1985-88 against TD 59.5 million projected under Loan 2554 for the same period. Similarly, only TD 12.4 million were approved by BTKD although loan approvals were assumed tq reach TD 46.3 million under Loan 2554. Approval rates were also particularly low for STUSID, the third Development Bank involved in the second EMI project loan. 65. The amount of the EMI-II line of credit was reduced in April 1988 by 40% and the revised loan amount became US$32.5 million. The largest cancellation related to BDET's share of the loan (reduced from US$20 million to US$11.5 million, i.e. by 57.5%), because of the availability to this institution of less expensive resources from bilateral assistance (France and Italy). These funds were tied but this did not constitute any significant handicap, since a large fraction of equipment imported in Tunisia for investment purposes originated from these donor countries. 66. In the investment climate prevailing in 1988 in Tunisia, there was an over-abundance of long-term funds available. There was also redundancy between three Bank lines of credit: EMI, Small and Medium Industries and Export - 17 - Industries. Of them, the EMI sector was the least likely to undertake significant recovery in the foreseeable future. The Bank, therefore, recommended that the uncommitted part of the EMI-II loan be cancelled. 67. In May 1989, another US$8.1 million were cancelled which brought the total amount of cancellation to US$29.6 million. Out of the remaining US$24.4 million, only US$18.5 million had been utilized by May 1991. 68. In 1989 approvals of EMI projects by BDET have significantly increased (such approvals were not only related to Loan 2554) while approvals by other development banks (BTKD and STUSID) did not follow. Some revival may also h.ve taken place in 1990 but at any rate at a much lower level than at the beginning of the 1980's. The sharp fall in the growth of the EMI sector (from 10.3% in 1980-84 to 2.9% in 1984-86 and 1.9% in 1986-88) may have been stopped in 1989 (+8.8% growth) and in 1990 mainly due to increased exports by several EMI enterprises. But overall, the 1980s did not witness at all the rapid expansion originally envisaged by the Tunisian authorities and supported by Bank staff. B. CONDITIONALITIES RELATED TO COMPETITIVENESS 69. EMI subprojects were sent to the Bank for review and as feedback to BDET on its lending orientations in EMIs. However, it would seem the condition that comparable foreign prices should have been quoted in BDET's approval requests for EMI projects and compared to the projected prices of the subproject's output was not effectively implemented. One reason given was the practical difficulty to obtain cost and price data on foreign products. The idea of requesting evidence of economic competitiveness at the subproject level through questionnaires while tariff and quantitative restrictions would remain largely in place, proved impractical. Identifying international prices turned out to be operationally difficult, as problems of comparability arose, along with the need to identify dumping practices. The Government and the Bank could only agree in January 1985 [Issues Paper on Second EMI Project, para. 3 (ii)] that more resources would be needed to be spent on the Tunisian side to ensure that the covenant under EMI-I be effectively met. Specific measures to ensure adequate administration of the control and assessment of the international reference prices were to be firmed up with the help of a Bank industrial sector mission whose report was, however, not to be available before the second half of 1985. As a result, the EMI-II project was approved without conditions related to evidence of price competitiveness at the subproject level. C. BDET PERFORMANCE 70. BDET's financial situation started showing signs of a slight deterioration in 1984-85, as a result of (a) significant delays in the Government payments of BDET's losses on account of foreign borrowing, and (b) increasing client arrears. Because of its cash flow problem, BDET had to start borrowing expensive short-term funds, which led to a reduction of its profitability (PCR, para. 22). - 18 - a. BDET's Structure, Operations and Portfolio 71. BDET was the only financial intermediary for this sector loan. It was believed that an experienced DFC like BDET could be relied upon to carry out effectively the complex financing assistance program described above. Since 1966, the Bank had channelled most of its lending to industry through BDET, which received seven Bank loans between 1966 and 1977. The last loan for US$30 million was fully committed in June 1981. Until early 1980, BDET had been the only development finance intermediary in Tunisia oriented towards long-term lending to the industrial private sector. In 1977-80, BDET had approved loans for 61 EMI projects for a total amount of TD 17.3 million (US$34.6 million), i.e., 18.5% of BDET's total lending to manufacturing and 13.9% of BDET's total lending. In 1980, approvals for EMI projects amounted to TD 6.3 million and rose to TD 9.7 million in 1981, the highest level ever recorded. BDET was considering favorably to lend TD 22 to 36 million in 1982-83, representing 28-46% of its total projected lending (a percentage much higher than in 1977-80) and 10-16% of total EMI planned investments during the same two-year period (see Annex 11). 72. Although two new development banks had been established (STUSID and BTKD) in 1980, they had just started operations and it was believed that time would be needed before they could match BDET's experience in appraising and promoting project ideas into tangible and viable projects. Moreover, the new development banks did not require financial resources in 1981 at least until they used up their initial share capital. However, STUSID and BTKD became co- borrowers with BDET in 1985 under the Second EMI Loan in response to a Government's request since the latter intended to maximize financing of EMIs through those banks (partly financed with Arab funds) rather than through scarce direct Government financing. By 1982-1983, BTKD and STUSID had already approved loans for EMIs for TD 30 million and TD 29.4 million, i.e. more than the TD 23.8 million financed by BDET during those two years. 73. BDET was owned in majority by shareholders (including IFC) other than the Tunisian public sector and was said to be free of major interference from the Government in its lending and borrowing policies. Share capital which had increasrd from TD 6 to TD 10 million in 1978-81 was to be raised to TD 20 million in 1982. The purpose of this capital increase was to allow BDET to match, by using its gre9ter leverage, the resources of the new development banks. 74. BDET's total approvals had remained practically constant in 1977-1980 and had even declined as regards industry. However, they had jumped by 90% i 1981 and there were prospects that they would remain at a high level in 1982, partly due to resumption of industrial investment (total manufacturing investment rose from TD 133.9 million in 1980 to TD 216 million in 1981 and TD 290 million in 1982). 75. BDET had a well diversified and sound portfolio, which amounted to TD 111.7 million in 1980, including 89.2% in long-term loans and 10.6% in equity participations. Arrears represented only 4.4% (including overdue interest) of the outstanding loan portfolio although some deterioration of the portfolio was expected. Finally, out of the 126 companies in BDET's equity portfolio, 31 were in difficulties or unprofitable. - 19 - BDET Share Capital, Approvals and Financial Indicators (Forecast by EMI I and II and Actual Results) 1980 1982 1983 1985 1986 1987 1988 1989 BDET Share Capital (TD million) EMI I 20 20 30 30 EMI II 30 30 30 30 Actual 8 20 20 30 30 30 30 30 Loan and Equity Participation Approvals (TD million) EMI 1 55.0 53.0 57.5 60.5 EMI II 46.5 48.3 44.3 46.7 Actual 32.1 62.8 40.5 44.7 32.5 27.8 68.3 123.4 Return on Average Net Worth (1) EMI 1 9.3 7.4 10.0 10.0 EMI II 6.1 8.1 8.8 9.0 Actual 9.0 7.0 9.0 6.6 7.0 5.3 6.7 13.0 Long-Term Debt/Equity Ratio EMI 1 5.1 6.4 5.6 6.1 EMI II 6.2 5.4 5.6 5.7 Actual 5.6 4.2 5.1 5.4 5.5 5.8 5.9 5.7 Margin on Borrowings EMI I 1.4 1.0 1.6 2.0 EMI II 1.5 1.7 2.0 2.1 Actual 1.4 2.7 2.6 2.2 2.8 2.3 2.6 3.2 Sources Annex 19 - 20 - 76. BDET's net profits had stagnated in 1977-1980 and the return on average net worth had remained around 9%. Cost of term debt in percentage of average outstanding term debt averaged 7.1% in 1978-1980 while income from loans as percentage of the average loan portfolio was 8.7% leaving a modest average spread of 1.6% for those three years (excluding the Government subsidy given to BDET to theoretically ensure a 3-point interest spend on loans financed with foreign exchange borrowings). In 1981, the spread rose to 1.85 point and BDET was requested by the Bank to maintain a minimum spread of 2 percentage points in following years. 77. Although not part of Loan 2113 conditionalities, financial assumptions were made regarding BDET projected operations. Projections were updated in 1985 when the Second EMI Loan was made and can thus also be compared with actual results. 78. The above data show that, in 1982-1988, although BDET's capital increase took place as expected, total project approvals rose much more slowly than expected. Total approvals were substantially lower than assumed under both EMI I and EMI II and recovered by 1988-89 only. This was also true for BDET's approvals in manufacturing and in the EMI subsector. The latter amounted to TD 23.8 million in 1982-1983 ag-inst a higher TD 36 million target anticipated under EM1 I. Approvals of EMI projects, projected under EMI II, were TD 59.5 million for 1985-1988, while actual BDET financing was TD 21.1 million, i.e. about one- third only (see Annex 18). 79. BDET's financial position was marred by three detrimental factors during the EMI project implementation period: (i) overdue payments from the Government; (ii) short-term debts; and (iii) arrears on loans. BDET's need to contract short-term debts (together with other short-term liabilities, they rose from TD 19.3 million in 1980 to TD 53.6 million in 1986 but decreased somewhat in 1989) was largely related to delays in Government payments, which themselves grew over time and developed into a self-perpetuating phenomenon. i. Overdue Government Payments 80. The Government had a statutory obligation to repay BDET losses on foreign exchange borrowings, and has been doing so but that with very significant delays. As shown in Annex 21, a backlog of payments due to BDET increased from TD 2.4 million at end-1981 to TD 34.1 million at end-1987. Despite a large TD 13.2 million payment to BDET in 1987, during 1988 the dinar depreciated further and BDET's losses increased again. In 1985, the Foreign Exchange Risk Fund was reactivated. But the Fund depended on a constant income from taxes on bank overdrafts, and the latter proved to be insufficient in covering the losses. To help solving this problem, a study of the foreign exchange risk coverage was to be completed by March 31, 1986 (Ref. 1985 Export Loan), but this was not done until end-1988, first in the context of a SECAL and then of a SAL. A more comprehensive system that previously envisaged, combining financing with hedging, was then designed. The Foreign Exchange Risk Fund's operations have been limited to its obligations as of August 15, 1988 and the Central Bank remained reluctant to fully cover foreign exchange losses for financial operations (including BDET). The proposed mechanism is for the State to assume the risk for a fee that would - 21 - make foreign exchange funds cost a bank the same as its other resources of similar maturity. This fee has been temporarily put a 3% above the cost of the foreign exchange loan. In the meantime, the combined backlog of Government payments has been detrimental to BDET's liquidity (see para. 95). ii. Increasing Short-term Debt 81. Due to the shortfall in Government payments, BDET has had to continue expensive short-term borrowings to meet its debt service obligations. This had an obvious detrimental impact on BDET's profitability. A significant reduction in BDET's short-term borrowings would be achievable only following a substantial decrease in payments due from the Government. iii. Large Arrears 82. BDET has had a serious problem with arrears since 1984. Total arrears represented 4.4% of the loan portfolio in 1980 but rose to 8.2% in 1984 to reach a peak of 16% in 1987 when Loan 2113 was closed. The situation has improved in 1988 and 1989 (14.3% in 1989) but the situation remains serious. Arrears were not seen as a problem when EMI I was appraised. At the time of EMI II appraisal (1985), the fast increase in arrears which had occurred in 1984 was seen as being mainly due to a poor economic conjuncture and significant improvement was expected "by the end of 1985 or mid-1986 at the latest" (SAR No. 5487-TUN, para. 5.11, May 1, 1985). This judgment proved incorrect and BDET's arrears kept worsening for a number of years. This resulted from unfavorable, structural economic conditions, strong competition from other banks and also from specific organizational weaknesses within BDET. 83. In 1980-83, many investments made were evaluated under over optimistic assumptions. They were often too large in scope and their competitiveness was not always assured. Companies and banks could invest in those projects because of considerable oil resources, the inflow of large amounts of capital from abroad (particularly from petroleum exporting countries around the Persian Gulf) and the Government's decision to advance its industrial development with new large-scale projects in the EMI, fertilizer, construction materials and tourism sectors. The creation of several new development banks provided an environment where project promoters were offered easy access to finance and where the banks were competing with each other to finance projects with marginal economic, financial and technical justification. 84. When the petroleum revenue diminished and general economic conditions deteriorated in 1984/1985, many of those projects: (i) found themselves with insufficient working capital because of generally underestimated financial requirements; (ii) operated with an unfavorable cost structure because of low capacity utilization and high financial and amortization charges; (iii) were frequently unable to earn the foreign exchange (through exports) required to pay for raw materials and intermediate products imports; and (iv) were affected by foreign currency controls instituted by the Government from mid-1985 to late 1986 to overcome its foreign exchange shortage, the effect of which was to hamper enterprises- -especially new ones--in their purchase of imported equipment or inputs, and then to prevent them from operating as planned. As a result, many borrowers were unable to repay their loans. - 22 - 85. Most arrears are in the industrial sector, including EMIs. It is interesting to note that seven out of the eight EMI projects financed under Loan 2113 are in arrears but that only one out of eight EMI projects financed under the Second EMI project (Loan 2554) is now in arrears which seems to indicate that BDET has been more careful in financing projects in the later years. 86. A large proportion of arrears (about 50%) is in public enterprises (PE). Attempts to reduce arrears from PEs, however, had very limited results, as payments of arrears was dependent on the Government's difficult budgetary situation (PCR, para. 24). 87. Since 1985 BDET has initiated general and specific actions to address its portfolio deterioration by increasing supervision activity and the number of staff assigned to it and by working more closely with management of some larger companies to come up with restructuring plans. BDET also increased its legal actions against borrowers in arrears !n order to protect its own interests and oblige borrowers to make every effort to repay their loans (the percentage of arrears in litigation went up from 41.5% in 1984 to 60.2% in 1990). It also used the rescheduling of loans for projects which have suffered from Government policies and measures. 88. Although efforts have been made to increase its legal action against borrowers in arrears, BDET is careful not to liquidate borrowers' assets at too low a price in order to minimize losses, and thus, is reluctant to act too precipitously. It would appear that, in a majority of cases, BDET has been able to realize gains in selling assets, as compared with the accounting value of its financial claims on the borrowers. For arrears not in litigation, penalties have been established and in 1991 amounted to about 16% final rate. Provisions have also been made to cover losses and are considered sufficient by external auditors. However, arrears management may benefit from better computerized data management. Despite increased supervision efforts, BDET's arrears problems remain quite serious for the reasons explained above. b. Profitability 89. Although BDET's interest margin on borrowing remained above 2% throughout project implementation, BDET's profitability has remained at a low level with a return on net worth averaging 7.4% in 1982-1987, substantially below EMI I and EMI II projections. This low profitability resulted from losses on foreign exchange borrowings and the backlog of payments due to BDET and from the non-performing assets (low or negative return of many industrial projects and large accumulated arrears). (See Annex 23). 90. The interest margin was lower than it could have been due to substantial changes in the structure of BDET's borrowings (see Annex 19). Borrowings in the international capital market have now become BDET's main source of funds, which haF resulted in an increase of the cost of its resources. The average interest rate on borrowings increased from 6.4% in 1982 to 8.3% in 1987 and increased further to 8.6% in 1989. 91. One must note, however, that interest rates on lending regularly increased from 1982 to 1987 and particularly in 1982-1984. Under Loan 2113 it - 23 - has been assumed that average lending rates woul* increase by 0.5% per annum and this effectively took place in 1982 (9.5% against 9.0% in 1982), in 1984 (10% against 9.5% in 1983) and still 10.3% in 1985. 92. After several years of relatively poor profitability performance, the situation improved in 1989 when the return on iverage net worth reached 13% against 5.3% in 1987 and 6.7% in 1988. This was due to higher lending rates (11.75% in 1989 against 10.75% in 1988), increased lending and large profits on participations sales (from TD 1.56 million in 1988 to TD 3.30 million in 1989). 93. Due to BDET's low profitability, dividends as a percent of par value share remained (with the exemption of 1984) much below projections. Only in 1989 and for the first uime in five years did this percentage reach the 8% level, considered as "minimum" when the project was appraised. c. Loan Conditionalities related to BDET as an Institution 94. These conditionalities have been summarized in para. 22 and numbered essentially four: (i) ensure a minimum 2% spread between interest income on average loan portfolio and interests payments on average term debt; (ii) obtain Government assistance, including subsidies, to help achieve a 2% spread; (iii) obtain Government guarantee that it would bear foreign exchange risks on BDET's foreign exchange resources and cover eventual losses; and (iv) strengthen project promotion and supervision. i. Minimum 2% spread 95. It was agreed that BDET would review annually its interest rate structure and make necessary adjustments to ensure a reasonable spread between its interest income on average loan portfolio and its interest payments on average term debt, of at least 2 percentage points, starting 1982. ii.. Government Assistance 96. Assurances were obtained during negotiations from the Government that the necessary measures would be taken to ensure BDET a minimum 2% spread through lending interest rate adjustment and Government assistance when necessary with a view to rescind such assistance before end of 1986 (Guarantee Agreement). In particular, BDET was to ask from the Government a new form of assistance by subsidizing interest rate exceeding 8% on foreign resources. Such assistance was to be required until 1985 to ensure a minimum spread of 2% and to allow BDET to remunerate its increasing capital (up to TD 30 million) at the minimum 8% of par value required by its shareholders. Finally, an annual review of BDET's interest rate structure would take place in order to limit the magnitude of any Government assistance to BDET. This agreement was confirmed during negotiations for the Second Loan to EMIs of 1985 (Loan 2554-TUN). iii. Foreign Exchange Risk 97. The foreign exchange risk on BDET's foreign exchange resources was to continue to be borne by the Government. Up to 1981, this practice had resulted in profits for the Government because in nominal terms the dinar appreciated in - 24 - relation to the basket of currencies involved in these borrowings. After 1981, the situation was reversed (particularly vis-&-vis the dollar), though the result over a ten-year period was still positive for the Government. The question of foreign exchange risk coverage thus assumed increasing importance as BDET and other, newly created development banks, were to have to meet a significant part of their resource requirements through borrowing abroad: designing and adcpting an appropriate foreign exchange coverage scheme was, therefore, becoming increasingly important. A Financial Sector Study (FSS) nompleted by the Bank in December 1984, proposed that sub-borrowers be charged an increasing cost to cover the foreign exchange risk and that this cost be related to the curl.ency borrowed in order to reflect more accurately the risks involved. It was agreed in 1985, during negotiations of an Export Industries Project (Loan 2522-TUN), that the Central Bank would undertake a study for designing and instituting a suitable foreign exchange scheme along the principles recommended in the FSS. The study was expected to be discussed with the Bank by March 31, 1986 at the latest. In the meantime, under the Second Loan to EMI, the Government was to continue bearing the foreign exchange risk, as it had under previous loans to BDET. iv. Project Promotion and Supervision 98. An agreement was reached between the Bank and BDET, according to which the latter would hire additional staff to deal with project promotion and project supervision. The Promotion Department had been active only since 1978. Moreover, no more than half a dozen projects could be promoted per year with the staff of seven professionals available in 1981. Total staff was to be increased by about 50% by mid-1982 and the commitment to further strengthan this department was confirmed during negotiations of Loan 2113. Moreover, since in 1982 the staff of four professionals was insufficient to handle the amount of work required in order to systematically follow-up the large number of loans to and equity participations in enterprises in their start-up period (161 out of a total of 454 clients), a commitment from BDET was obtained to hire additional staff for supervision purposes. By 1985, the staff of professionals had been increased from seven to nine, while three additional staff had been recruited in 1984 to participate in project supervision. However, due to the fact that BDET had experienced a worsening of its arrears position in 1984, a new reporting system was agreed upon in 1985 between the Bank and BDET at the occasion of negotiations of the Export Industries Loan and the Second EMI Loan. This system aimed at enabling BDET's management to monitor staff involvement in supervision, the number and type of projects visited, the kind of problems encountered by subprojects, and the actions taken by BDET and the project sponsors to overcome these problems. Progress in Meeting Conditionalities During Project Implementation 99. Conditionalities have been partially met. The minimum 2% spread was achieved due to higher lending rates which compensated the higber cost of borrowing particularly on international capital markets in 1982-1985. Lending rates increase slowed down in 1985-1987 but this was compensated by a relative decline in interest rates on borrowings (including on international market) since 1985 and as a result margin on borrowing remained over 2%. - 25 - 100. Government assistance was provided to subsidize interest rate exceeding 8% on foreign resources until the end of 1984. However, the subsidy was never high enough to allow dividends to be maintained at 8% of par value. Such dividends averaged 5.6% in 1982-87, 6% in 1988 and only reached 8% in 1989. 101. The biggest disappointment resulted from the large backlog of Government payments to BDET as compensation for large foreign exchange losses. The study, which was to be completed by 1986 to help solving their problem, was delayed until 1988, i.e. after Loan 2113 was closed. The issue was to create mechanisms that, with the absence of a true foreign exchange market, encourage enterprises to cover against foreign exchange risk in a way that is not distortionary and does not cause systematic gains or losses to the Central Bank. The foreign exchange risk fund which had proved unsuitable and ceased to take on new obligations after August 15, 1988, has been replaced by a new mechanism for borrowings by banks. At issue are loans contracted by the banks at the behest of the State, usually official assistance from multilateral agencies (thus including loans to BDET). The terms are advantageous to the economy, but the banks are reluctant to pass the foreign exchange risk on to the final borrower because they lack the skills to cope with variable interest rates and few enterprises know how to hanale the risk. 102. The proposed mechanism is for the State to assume the risk for a fee that would make these funds cost the Bank the same as its other resources of similar maturity. But, since there was no market determined long-term interest rate, the money market rate was used. This is a one-week rate and proved too volatile. As a result, the authorities have temporarily put the fee at 3 percentage points added to the cost of the loan. The authorities judge that this system is the best suited to the present conditions and it will evolve as the banks acquire some of the necessary skills and markets for long-term instruments develop. This view can be shared but, in the meantime, a 3% margin may seem high and the time needed for enterprises (and not only banks) to know how to handle the foreign exchange risk may be longer than now anticipated. 103. The emergence of severe arrears problems has led BDET to strengthen supervision to reduce such arrears. However, although results have been obtained, the situation is still unsatisfactory and further improvements are needed. 104. Overall, BDET's performance would now seem to finally improve after several d.fficult years. However, conditionalities under Loan 2113 (and also Loan 2554) were only partially met during the implementation period, mainly due to the fact that the economic and financial "scenario," originally designed for the Electrical and Mechanical Industries sector, did not materialize as expected. D. TECHNICAL ASSISTANCE 105. Under EMI-I, the need to accompany promotional efforts with technical support services to EMIs was addressed. The Bank loan included a US$1.8 million technical assistance component to create two semi-public autonomous institutions under the aegis of the Ministry of National Economy (MNE), the Centre Technique des Industries Mdcaniques et Electriques (CETIME) and the Institut National de la Normalisation et de la Propridtd Industrielle (INNORPI). - 26 - 106. The Bank's Review Report on Electrical and Mechanical Industries (Report No. 2666-TUN, June 4, 1980) had recommended that the Ministry of Industry establish a "Technological Center" for the mechanical industries with the aim of overcoming the weaknesses in engineering adaptation and in standardization and quality control which were a serious handicap for the development of efficient operations and inter-enterprises linkages. The Center was to fulfill the following functions: * establishment of norms and standards for Tunisian products; * quality control for local mechanical products; and * provision of technical assistance in technological choice and adaptation of equipment and production processes, operations organization and management and identification and selection of foreign partners. 107. The Tunisian Ministry of Industry strongly supported the formation of such a Technological Institute which was to be focussed initially on the mechanical subsector. Its major functions, as visualized by the Ministry of Industry, were to develop norms for the subsector; provide quality control for the products of the subsector; advise entrepreneurs on the most suitable technology to meet Tunisia's conditions; and provide specialized training and other forms of technical assistance to these industries. However, BDET felt that norms and quality control were specialized tasks to be kept from the proposed Technological Institute. 108. In 1980, the Ministry of Industry decided the creation of a Sub- Direction for Standardization and Quality Control to be staffed by engineers and to be in charge of the standardization work in the EMIs. Thus, a distinction was being made between the quality control and standardization tasks on one hand, and technical assistance to EMIs on the other hand. In nine months, and with limited means, a dozen standards were prepared in cooperation with the manufacturers concerned. In the process a consensus emerged among manufacturers on the need to set up a National Institute to develop standards on a large scale and at a faster pace, and ensure that those standards were adhered to. 109. A Bank mission offered to help define the profile of a foreign expert in standardization of metal products that could help the Tunisian authorities to speed up the standardization process. 110. In June 1981, the Bank felt the need to design a delivery system of technical assistance which would cater the specific needs of EMIs. A Bank mission visited several existing technical centers (such as Centre Pilote de l'Outillage, CPO, in Sousse) and a sample of EMI firms to assess their TA needs and the best way to fulfill them. Separate from the idea of a Technical Center for EMIs, the Bank mission assessed the Government proposal for setting up an Institute for Standardization and Quality Control. 111. The Bank's mission concluded that technical and technological assistance were needed to: (i) bring about improvements in technical management and in labor productivity at the shop level, particularly in the steel - 27 - structure/platework and mechanical works subsectors; (ii) helD enterprises to restructure and modernize their facilities and to adapt and develop more technologically advanced products (e.g. industrial boilers, heat exchangers, condensers, and furnaces for the platework sector); and (iii) spread within the sector the use of widely accepted standards and of quality control procedures. Moreover, detailed reviews were recommended for two specific subsectors-- foundries and platework which were thought to have potential for efficient ioport- substitution, but required rationalization. All these actions were seen at that time as meeting the Bank's emphasis in linking policy changes (through sector studies and technical centers) to lending operations, thus moving from the DFC approach to an Industrial Development Finance (IDF) approach. 112. The Tunisian authorities accepted to include funds for technical assistance in Loan 2113 provided grant funds from other foreign sources were not available. Allocations were included in the 1982 budget (TD 240,000) for the Institute of Standardization and the latter was expected to use the facilities of an existing "Laboratoire Central." UNDP/UNIDO were requested to provide US$165,000 for the Institute. 113. Budgetary allocations (TD 101,000) were also made in 1982 for Centre Technique de la M6canique (CTM) and UNDP was requested to provide US$930,000 to the Sousse Centre d'Outillage to produce plastic moulds, starting in 1984, for three years. Tunisia also requested DM 7 million from Germany for an electrical training center. 114. It was agreed, at negotiations, that the Government could seek alternative sources of financing for all or part of technical assistance elements under Loan 2113 provided, however, that such financing could be identified and the related agreements signed at least one month before effectiveness. 115. After some delays (which resulted in a three-month extension of Loan Effectiveness limit), the law creating the CTM and its organization decree were published (May 25, 1982). The CTM became CETIME (Centre Technique des Industries M6caniques et Electriques). The creation law of the National Institute for Standardization and Quality Control was published in August 6, 1982. a. CETIME 116. In a first phase, the proposed Center was to deliver technical assistance to two priority subsectors (steel structure/platework and mechanical works) where Tunisia was thought to have a comparative advantage with low wages and increasing competence and quality. The range of technical services offered to enterprises in these two subsectors was to comprise initially: i. In-plant guidance and troubleshooting to resolve production problems (e.g., reorganize plant layout and work flow; remove bottlenecks; improve product quality; demonstrate to workers and technicians appropriate and more efficient practices with regards to machinery and tools utilization as well as their maintenance). - 28 - ii. Assistance to firms in their efforts to rationalize existing product mix, to adapt/adopt new and appropriate products, to plan and prepare new or expansion investments. iii. Advice to firms in the design and improvement of cutting tools and die metals. 117. To avoid duplication of efforts (e.g., the Bank Fourth Education Loan approved in May 1981 was to finance expended training facilities with annual output of 3,300 workers with EMi skills), Loan 2113 was to finance training of EMI workers only in the form of on-the-job and shop floor level guidance of workers in EMI enterprises, at least during the first two years. 118. It was expected that in the first two years the Center would provide in-depth assistance to about 20-30 small- and medium-size firms in the two priority subsectors, not counting numerous spot assistance actions in other EMI subsectors. Few actions were expected in large firms because the latter often obtained technical assistance directly from their foreign technical partners. 119. A Centre Pilote de 1'Outillage (CPO) located in Sousse had been created in 1971 with the objectives of training, development and fabrication of tools and dies needed by enterprises. In order to revitalize CPO, it was decided to incorporate it as a Tools and Dies Department of the new CETIME. 120. CETIME was an autonomous semi-public institution created under MNE. To encourage private EMI firms to accept more readily CTM's services, four of the nine members of CETIME's Board were to be representatives of the private sector. 121. CETIME was to have a minimum core of staff comprised of eighteen Tunisians (six engineers and 12 technicians) and of eight foreign experts for two years. The team of foreign experts was to be headed by a senior engineer, whose task was to assist CETIME's management and formulate its program. The other foreign experts included two production engineers (steel structure/platework and mechanical works); one product development and design engineer; and four senior technicians (welding; mechanical works; tooling; dies). 122. CETIME's operational costs in foreign exchange were estimated at US$1.5 million equivalent for the first two years to cover the foreign experts' salaries (192 man/months at about US$7,500 per month), the expenses for additional external consultancy services as needed, and some training abroad for the Tunisian counterparts. 123. In 1983 and 1984, CETIME carried out rather successfully some 45 technical assistance interventions. By May 1984, it had eleven engineers and six technicians, in addition to eight French experts (US$1.27 million contract). CETIME's development, however, was limited by its still rather ineffective integration in the whole process of EMI projects identification, design, development and operation. EMI firms were not sufficiently informed of the existence of CETIME, nor of the scope and type of services that this agency could provide. CETIME sat on the Board of API which delivered authorization to invest in particular projects (CETIME's role was suspended later on); however, the - 29 - project profile on the basis of which this was done, was far too preliminary (as regards production technology and process) for this examination to have provided a real opportunity for CETIME to offer its sources and demonstrate its competence. Moreover, CETIME's regulatory role to judge the technical aspects of EMI proposals was highly undesirable since it confused the agency's technology promotion role with the role of administrative controls. This is no longer the case, and CETIME's services are generally provided only to clients who seek them. On the other hand, CETIME remains responsible for certifying the degree of integration of some industries in the mechanical sectors, e.g., in connection with the administration of local sales agreements for export-oriented companies. CETIME still suffers from this regulatory role which necessarily undermines its "service" relationship with the private sector. 124. There is little doubt that the Tunisian private sector was reluctant to use CETIME services in the early years since CETIME was largely seen as a Government-controlled organization. CETIME's management policy was thus to build its reputation through the p: omotion of high quality technical centers which would have had a positive demonstration effect on the industrial community. The most important center was the Sousse Center which makes molds and tools. 125. For this reason, in October 1984, CETIME requested the Bank to use about US$260,000 of a remaining unused balance of US$392,000 allocated for technical assistance to CETIME to install a line of specialized ovens for heat treatment of mechanical parts and components, in particular for the tools industry. The treatment was said necessary to bring about the desired mechanical quality of industrial tools produced, thus contributing to the development of this priority subsector. It was o benefit mostly small- an medium-sized firms in the industry which, because of their limited size, were unable to finance such equipment. Loan 2113 included only the financing of expert services to CETIME btt the Bank agreed to an amendment of the Loan Agreement to satisfy CETIME's request. 126. Following the same policy to develop technical centers before emphasizing direct in-plant technical assistance to EMIs, CETIME prepared an expansion program, which was financed under the Second EMI Loan in 1985. This program involved: a. the opening of a Mechanics Laboratory; b. the creation of a Metal Tools for Foundries Section; c. the opening of a small Electronics Division; and d. an extension of foreign technical assistance services financed by the EMI-I Loan. 127. The Mechanics Laboratory's objectives were to provide firms with control services on their process and final products and to help adapt production technologies and processes. 128. The smaller foundries depended on imports of metal tools. The later were to be produced by CETIME at the old CPO. - 30 - 129. The Electronics Division was to design prototypes of electronic implements needed for production controls and assist enterprises in the design and fabrication of custom-made circuits and chips. 130. Finally in 1985, CETIME expressed the wish that the technical assistance received under EMI-I be extended but reduced from a team of eight foreign experts for two years to a team of four for one extra year at a cost of US$288,000. 131. Total cost of assistance to CETIME was increased from US$1.5 million under EMI-I to US$2.1 million under EMI-II. Priority was given to purchase of equipment (see Annex 24) while mostly expert services were financed under EMI-I. 132. From the beginning CETIME charged firms for the services that they received (essentially for the time spent by the staff on assignments and a small fraction of overheads) and was able this way to cover 40% of its 1984 operating costs. When the budgetary and economic crisis deepened in 1985, CETIME made additional efforts to increase revenues. In 1990, receipts covered 60% of its costs. The CPO operation alone accounts for 73% of the receipts, 64% of the costs and 71% of the personal of CETIME. 133. The Bank financed the Metal Tools for Foundries Section and the Electronics Division which are now operational. The latter produces and tests electronic printed circuit boards (PCBs), including computer aided design (CAD) and computer aided manufacture (CAM) systems. The ACB equipment appears to be under utilized, probably because CETIME does not produce PCBs for sale as, unlike for tools and moulds, it cannot yet compete with imported PCBs (see Tunisia - Industrial Sector Note, February 21, 1991, Industry and Energy Division-EMENA). 134. A large mechanical workshop and laboratory financed under EMI-2, with extensive facilities for non-destructive testing is being built. Finally, an Electric Test Division (which until end-1988 was to be implemented under INNORPI management) is under construction. 135. Technical assistance, financed under EMI-II for one year, has been extended through bilateral financing. For example, ten man/years were financed by Belgium. 136. In 1987-88, to supplement technology dissemination among entrepreneurs, CETIME stressed enterprise diagnostics and provided advice on the type of process to be selected, on the kind of materials and equipment to be used, on the need to increase quality control to facilitate exports, etc. In 1988, a total of about 200 interventions had been made since CETIME started operations in 1982. 137. Despite achievements of most objectives set under Loan 2113 and 2554, CETIME's overall impact on the availability and development of technology for EMIs is small, as the industries easily obtain whatever technology they want from abroad at modest relative costs, with faster delivary, and more tailored to their needs. CETIVE feels it has been fulfilling its mission to provide advanced technical services to small and medium industries which do not have an easy access to foreign technology as the larger companies. It also feels that it has now a staff of 140 of whom 100 work in the high tech Sousse plant that makes - 31 - molds and tools, and points to the fact that a number of former CETIME engineers have been able to start their own firms. 138. However, CETIME also recognizes that the current dependence on Government subsidies is not desirable, although CETIME would not like to depend solely on commercial success if it is to act as an industrial incubator and pioneer. One possibility would be to move away from sole Government ownership, but establish a close sub-contract relationship with the Government for clearly defined and budgetized Research and Development to meet to national goals it considers important. 139. In conclusion, it is still difficult to measure CETIME's success and impact on EMI's real performance since official reports continue to point to persistent, serious technical and production problems. Relations with the EMI private sector have been improving but a clearer definition of CETIME's role (non-regulatory) seems needed after almost ten years of operations. b. INNORPI 140. Under the EMI-I project, US$300,000 were provided to finance the foreign exchange cost of technical assistance to the newly created Institute for Standardization and Quality Control of Industrial Products (INNORPI). 141. The Institute was a semi-public, autonomous agency under the aegis of MNE. It was to be responsible for preparing and issuing standards in cooperation with manufacturers, and for ensuring that goods were produced in accordance with them. It was to be assisted by the existing laboratories to perform the necessary tests and controls. 142. During its first two years, INNORPI was to start off with a minimum core of about ten technical staff and was to be able to produce each year about 30 standards while adapting a greater number of existing international standards. EMIs were to be among the first priorities of the Institute's work program. 143. Three foreign experts in standardization were to be hired. The first expert was to advice MNE on the standardization program and make recommendations on INNORPI's scope of activities, staffing and budget. The two other experts (one in standardization, one in quality control) were to come subsequently for one year each. 144. A French consulting firm was hired by INNORPI. By end-1983, five French experts were in Tunis. All together 32 commissions had been set up, out of a total of 108, to define the norms. An elaborate flow chart had been prepared to follow the normalization procedures of each product. There were 66 employees of which 26 higher-level staff members. 145. INNORPI did not initially succeed in evaluating and coordinating testing laboratories activities due to lack of authority and recognition since the Institute was still very young. As a result, it concentrated in 1984 in preparing further work on delivering "certificates" to products made in Tunisia and meeting INNORPI's technical standards. - 32 - 146. In May 1984, INNORPI had a staff of 55 professionals and the assistance of two long-term experts provided by the French consulting firm under a Bank- financed contract of US$245,000 for 26 man/months of services, signed in October 1983. In May 1984, INNORPI had prepared 600 standards/norms. The objectives for 1984 were to produce 500 other norms (mainly for EMIs for which 63 norms only were produced in 1983) and finalize a decree on product certification, and establish a network of homologated laboratories. 147. The Institute was expected to operate on a commercial basis only to a limited extent. INNORPI hoped to eventually raise some revenues, once it started delivering certificates of conformity. In 1984, its operational budget came to about TD 300,000, of which only 15% were covered with its earnings from services provided to producers, registration fees and sale of the institute's periodical. 148. In 1985, INNORPI intended to provide control services (starting with the control of measuring instruments in use in the country) and increasingly deliver certification to products of the EMI sector. For this purpose, it submitted to the Bank (for financing under the EMI-II project) a program of development for its facilities which included the creation of a service of "metrology" and the opening of a laboratory to test EMI products. 149. The "metrology" section aimed at verifying the accuracy of scales and other weighing and measuring systems. Often with time instruments had lost accuracy and it was necessary to have good measurements in order to improve quality. This project, however, was never implemented since the existing "Weights and Measures" Division vetoed it due to potential duplication of activities. However, funds allocated for this purpose were diverted to the establishment of a small laboratory to develop mold standards for metal cans and cardboard packaging, which are important for ensuring minimum quality benchmarks to Tunisia's production an exports of agro and food industries. 150. The proposed EMI laboratory to be financed under Loan 2554 included three divisions (basic electric tests, industrial power test for transformers and mechanics divisions) at a cost of US$1,427,800 (see Annex 25). This subproject was transferred to CETIME at end-1988 at the Government request. It seems that costs had been underestimated. Moreover, Tunisian authorities felt that it would have been confusing to have enterprises pay a fee to use the EMI laboratory for testing purposes while enterprises would also seek INNORPI's certification. Another explanation is that according to British Standard Institute tradition, it would have been considered quite normal for INNORPI to have its own laboratory while in French AFNOR or German DIN traditions, independent laboratories were considered the norm. And in the Tunisia context, AFNOR-DIN customs prevailed. 151. As a result, EMI-II project objectives were not met. CETIME was put in charge of the EMI laboratory and INNORPI set up a packaging laboratory instead. It would appear that many still see INNORPI as being in charge of producing standards and issuing certificates of conformity, these tasks being considered as largely administrative and formal. As a result, there has been a tendency to remove laboratories and testing facilities from INNORPI, which does affect INNORPI's staff morale and reduces operational ambitions. - 33 - 152. The sizable work force (121 of which 97 professionals) and budget reached by INNORPI by 1990 have been partly due to its secondary functions (translations, publications, data collection and processing partly for international audiences and customers, external cooperation with Ara standardization institutes, etc.). The core functions of INNORPI, i.e., standardization, certification, industrial property and promotion, comprise about 55 professionals out of its total 97 professionals. The other, above mentioned functions carried out by INNORPI for the Government are not always reimbursed at full cost by the Government subsidy which has decreased in recent years (except for a subsidy to reimburse INNORPI for non-payment by some cement companies of annual fees for cement products certification). 153. INNORPI, considering the peculiar type of public service it provides-- preparing and edicting standards, and delivering certifications--is not expected to generate revenue beyond 15% of its financial needs. At present, only 8% of its budgetary needs have been raised by INNORPI. E. EFFECTIVE PROTECTION STUDY 154. As indicated above (para. 29) in order to introduce protective measures as well as sufficient incentives for the development of local capital goods industries and for competitiveness and efficiency improvement, agreement was confirmed during negotiations of Loan 2113 that tariff on EMI capital goods would not increase above 18% (or 21% for steel structure and platework products). Measures were also to be taken to ensure that priority EMI products (non-consumer goods) protected by quantitative restrictions (QRs) be charged "reasonably" competitive prices in comparison with similar foreign products. 155. As early as 1979, the Bank informed the Tunisian authorities that it intended to raise the protection issue which had a negative impact on the EMI sector and that it wished that an Effective Protection Study be carried out by the Quantitative Economy Institute in Tunis (also called the Bach Hamba Institute). 156. During preliminary discussions with BDET and the Tunisian Ministries (Industry, Planning), a review of the tariff policies (as they related in particular to EMIs) during the course of project preparation was again mentioned. The Bank Regional Project Director later on mentioned the need to condition financing of an EMI sector project on the "adjustment" of existing tariffs. 157. However, the overhaul of the tariff and protection structure was considered to be a "long and complex undertaking" and the Bank thought of assisting in this process through an Effective Production Study programmed "for the coming years" and not as previously envisaged during the EMI project preparation. Within the preparation of the 1982-86 Plan, a Commission had started to review the current tariffs with a general view to reduce the variations of tariff levels, and "if possible" adjust on an ad hoc basis some obvious cases of overprotection. 158. In view of this, Bank staff focused its attention on some issues where remedies could be undertaken relatively quickly and easily such as the procedures - 34 for duties exemptions on inputs used for exports. The Government agreed to a review of the procedures designed to allow industrial producers (in particular in EMIs) who exported infrequently or under short notice to recover the custom duties they paid on the imported inputs used in the manufacture of exports. Even though such procedures were provided for in a 1971 law, they were not being applied. This discouraged local firms from seeking contracts abroad. In 1981, a Government task force recommended to streamline existing procedures by requiring the refund authorization from one authority only (Customs). This was incorporated in the 1982 Finance Law. The text was reviewed during negotiations and was considered acceptable to the Bank. However, no provision was made for Bank's follow-up on actual implementation of refunding procedures. 159. The protection issue had led to heated discussions within the Bank since the Region had argued that protection was required as an incentive to the development of EMIs considered one of the main priorities of the 1982-1986 Plan. The Region also felt that a protection ceiling of 18% to 21% was acceptable for Bank-financed EMI projects. Such rates were considered reasonably low and of a temporary nature. 160. In view of these discussions and of the need to have a better knowledge of the actual protection levels among EMIs and other sectors, the Bank requested the Tunisian authorities to agree to a review on effective protection to be undertaken by the Bach Hamba Institute in 1982. 161. The Ministry of Planning and Finance agreed to include funds for this study in the proposed loan, provided this could not be financed under the technical assistance for macroeconomic research from Canada which was then under discussion. A Tunisian agency (CNEI) was to be involved later on for detailed subsector studies of effective protection. It was understood that the Effective Protection Study would cover the EMI sector as a first priority (Guarantee Agreement) and that arrangements would be made for a satisfactory comparison between prices of locally produced goods, and ex-factory prices in the partners/countries for similar quality products (supplemental letter to the Guarantee Agreement). All efforts were to be made on the part of the Ministry of Planning and Finance and the Bank to complete the terms of reference of the Study before April 30, 1982. This was effectively done in time. 162. The Bank was informed in September 1982 that a Canadian Institute was to co-finance the Study with the Bach Ham7a Institute as part of a wider study on the Tunisian economy. As a result, the US$200,000 provided under Loan 2113 were reallocated for financing of EMI subprojects. 163. In June 1983, a draft study was discussed with Bank staff. It was found that little use had been made of price comparisons (as previously recommended by the Bank) rather than customs tariffs, for estimating the nominal protection on industrial output and imports. The Bach Hamba Institute had not relied on visits to enterprises and had mainly used administrative dossiers of public enterprises. Moreover, the sample of enterprises surveyed was very small which reduced the credibility and usefulness of the study. 164. It was decided, at that time, that the EMI analysis would be completed by end of summer 1983 while three other subsector studies (food, textiles, chemicals) would be completed by June d84. Finally CNEI was to undertake by itself, but in cooperation with the Bach Hamba Institute, one of the remaining subsector studies in order to disseminate into other administrative units the know-how and experience acquired by Tunisia. 165. Despite these shortcomings (para. 59) and the feared lack of experience at CNEI, overall the work seemed to be proceeding very well and, in particular, it was encouraging to see how thoroughly the approach and methodology had been absorbed and was being used by the Tunisian experts with only periodic guidance by the foreign consultants (see para. 156). 166. A draft EMI study was ready in December 1983. The Draft was based on 1980 data and paid only lip service only to Bank comments on the lack of price comparisons. The Bach Hamba Institute decided to limit its further involvement to the overall Effective Protection Study of the whole economy and it delegated the other subsector studies to CNEI which had very little experience and resources (1.5 full-time staff) for this purpose. However, later on, CNEI proved unable to complete the studies on its own and the Bank asked the Government to transfer responsibility back to the Bach Hamba Institute. Nevertheless, it was also felt that the "Global" Study might provide an adequate substitute for the subsectoral studies transferred to CNEI. 167. One major drawback of the Study was that it could not take into account the effect of widespread quantitative restrictions on EMIs. In addition, since tariffs rather than actual price comparisons were being used, effective protection calculations may not have given a true picture of the situation. 168. In any caqe, results indicated that industries with the highest level of protection had imited abilities to export, served primarily the domestic market, and had the lowest economic and financial rates of return. The same study also showed that public enterprises had generally benefitted from the highest levels of protection, while showing the lowest rate of return. The overall effective protection level for EMIs was 96% in 1980 while the average nominal protection rate was 26% (see Annex 4). 169. Domestic oriented industries were shown as significantly less efficient than export-oriented ones (effective protection was 0.99 for 100% exporting firms and 1.63 for 100% domestic-market-oriented firms). A rationalization of the protection system was thus likely to be an important policy tool for further encouraging the efficiency of the sector and the growth of exports. 170. The whole issue of the protection system was raised again when the EMI- II project came under discussion. IDF Bank staff felt hesitant in opening discussions with the Tunisian authorities on this subject in the frame of a specific relatively small line of credit. They were helped by the fact that a separate Industrial Policy Bank mission visited Tunisia in January 1985, to design and submit recommendations on (i) reducing tariffs and quantitative restrictions, (ii) improving pricing policies, and (iii) fostering higher productivity and efficiency in public enterprises. 171. The Region argued that the restructuring of a system of economic incentives could not be limited to a single subsector (EMI or other), and had to - 36 - be integrated in the overall reform of the industrial sector,which was being reviewed by the Bank's Industrial Policy mission of January 1985. Consequently, the EMI-II loan simply aimed at updating and expending the measures agreed under the EMI-I loan. In addition to the condition that subpro'ects should only involve priority goods (non-consumer EMI products) for which customs tariffs did not exceed 18%, the EMI-II project proposed to gradually phase out quantitative restrictions. Tariff protection was only allowed for new EMI goods (i.e., manufactured for the first time in Tunisia) financed under the EMI-II loan, up to the maximum 18% level agreed under the first EMI loan. In exceptional cases, justified on an individual basis by a development bank in consultation with the Bank, other protection (i.e., QRs) may have been granted, but only for a maximum period of five years from the date of first production. 172. The Region was convinced that (i) there was a need to redirect the Tunisian strategy in the EMI sector away from the highly protected assembly-type operations and excessive emphasis on local content and (ii) there was also a need to revise the policy framework to encourage export orientation and concentrate Tunisia's EMI sector on products in the production of which it would, indeed, have a comparative advantage. This was said to be achievable only through a reduction of effective protection and removal of QRs. These wider issues were addressed by the above mentioned Bank mission in January 1985. It was hoped that the recommendations of this mission would provide the basis fDr an industrial sector loan (which indeed was approved for an amount of US$150 million in 1987 under the title "Industrial and Trade Policy Adjustment Loan-ITPAL") and that it was more appropriate that the conditionalities relating to ove;:all industrial policies (and not only EMI sector issues) be reserved for that loan. 173. Certainly the Tunisian authorities were still not inclined in 1985 to discuss broad policy issues in the context of a line of credit for the EMI subsector. Based on EMI-I implementation experience, Bank staff understood that and consequently EMI-II basically called for the same limited conditions regarding the protective system than EMI-I. The only addition, as mentioned above, was that for Bank-financed projects, QRs would not be allowed for new EMI goods (not produced in Tunisia) except under very special circumstances to be determined by careful analysis. Otherwise, it was finally realized that wider policy reforms could not be addressed in the context of this (or any other) limited subsector operation. 174. The Senior Vice President Operations indicated that, if no action was undertaken on trade policies, i.e., specifically, QR reductions, the Bank would be in no position in the future to remain active in industry lending (memo to files, EM 2 C, EMENA Region, April 1, 1985). This position was understandable since nominal protection on EMIs had on average increased from 26% in 1980 to 32% in 1984 and 36% in 1985, while effective protection had only slightly decleased from 96% in 1980 to 92% in 1984 and was increasing again to 104% in 1985 (see Annex 4). Increased protection in 1984/85 was related to growing balance of payments difficulties which also implied even more stringent QRs. However, in the end, senior management agreed with the Regional position that it was more appropriate (and realistic) to seek a Government commitment to introduce policy changes across the board through a general dialogue with the Tunisian authorities rather than through a limited EMI sector operations. In fact, at the time of negotiations of the second EMI loan, economic conditions were worsening in - 37 - Tunisia and there was a growing feeling among Tunisian authorities that liberalization measures were necessary to make the economy and industry more outward-looking and competitive. Partly for these reasons, the Government confirmed for the first time that after the completion of the Industrial Sector Report (due in July 1985), it would establish jointly with the Bank the measures to increase efficiency and competitiveness in the industrial sector. This took place in 1986 when a new more liberal economic policy was decided upon, supported by two SECALs in 1987 and a SAL in 1988. 175. In September 1988, the Tunisian Government logically requested that custom tariffs related eligibility criteria for subprojects financed under the EMI-II project be dropped, as superseded by the Government's commitments under ITPAL. From 1985 to 188, nominal protection on EMIs had decreased from 36% to 29% and effective protection from 104% to 63% (see Annex 4). There were prospects for further declines in 1989-1990 and, more importantly, elimination of QRs on a number of EMIs goods produced in Tunisia Of course, the issue remained of how insuring the survival of non-competitive EMIs during a transition period, e.g. through temporary tariffs increases. But, for all practical purposes, these were issues to be discussed in the framework of the SAL and not of the EMI-II loan. Eligibility criteria agreed under that loan in 1985 had lost their relevance and had become unnecessary and therefore could be waived. At any rate, few applications for EMI projects were being submitted by end-1988 to development banks and a side effect of the waiver may have been to encourage more applications for loans without protection-related conditionalities. 176. In 1985, the Bank had insisted on procedures to allow EMIs to recover customs duties they paid on the imported inputs used in the manufacture of exports (para. 151). Liberal procedures were effectively introduced by a 1987 amendment to the Custom Code to bring about an improvement of the reimbursement of taxes on imported goods used for exports. It must be noted that this was done as a follow-up of the 1985 Bank's Industrial Policy Report and in the context of negotiations for ITPAL in 1986, rather than as a direct result of the EMI projects. F. SECTORAL STUDIES 177. It was agreed under Loan 2113 that two studies would be undertaken to define long-term development strategies for the platework and foundry subsectors which had potential for efficient import-substitution ad were thought to require rationalization. These studies were to: (i) identify economically efficient EMI product lines based on a sufficiently large local market and on relatively simple technological processes; (ii) propose to rationalize and specialize productions between firms and outline investment programs for them; and (iii) define long-term development strategies for these two subsectors. - 38 - The Platework Subsector Study 178. In June 1981, the Ministry of National Economy (MNE) commissioned BTKD, a new development bank established in 1981, to do a study of the platework subsector. The Bank prepared draft terms of reference for the sectoral part of the study which was scheduled to be completed before July 1982. The Bank was, upon its completion, to review it and discuss with the Government its proposed strategy for the development of the platework sector and this before July 30, 1982 (Guarantee Agreement). An understanding was also reached with BDET that it would submit to Bank financing subprojects in this subsector (and also the foundry subsector) only after the Bank and the Government would have agreed on the development strategy to be implemented in this subsector (Minutes of Negotiations). 179. The full S-udy was delayed but a Market Study was completed by a consulting firm for BTYD by Pecember 1982. In October 1983, i.e. ten months later, Bank staff had the following observations under market studies: lack of analysis of production costs to establish competitiveness of the subsector; lack of economic justification for new productions and lack of strategy for conversion of capacity for plate thickness below 10 mm. towards new activities. 180. BTKD management apparently intended then to commence a second phase with the help of an Italian consultant and with the hope to take into consideration Bank criticism. However, by May 1984 the platework subsector study was stopped half-completed apparently for lack of foreign partners potentially interested in promoting new platework projects. BTKD stated its interest in completing the second half of the study (which would have been useful for sub- project identification and promotion) only provided BTKD could find such potential foreign partners. At that time, BTKD's approvals for EMI projects were already declining rapidly. b. The Foundry Subsector Study 181. In the context of both the SOFOMECA project (which benefitted from a US$15 million Bank loan in 1983) and the EMI-I project, the Bank carried out a sectoral study of the foundry subsector to propose a strategy for the development of foundries. The Study was completed in 1982 and discussed with the Government to reach an understanding on the desirable strategy for the development of the foundry subsector. Although Bank staff seems to have been genuinely interested in such broad discussions, tne main objective was clearly to justify financing of the rehabilitation/expansion of the SOFOMECA Foundry. Otherwise, there is no evidence that there was substantial follow-up on the findings and recommendations of the study. Moreover, the SOFOMECA project ended up as a failure (cf. PPAR on SOFOMECA Project, June 1991). - 39 - IV. RESULTS 182. Results can be judged according to the various components of the EMI-I and EMI-II projects. A. EMI PROJECTS FINANCING 183. By the time the EMI-I project was completed, eighteen projects (of which eight in the EMI sector) had been financed through the Bank line of credit. Overall Bank financing amounted to 19% of total investment (TD 98.4 million) (PCR, para. 28). 184. Project objectives and actual results compare as follows for EMI subprojects: Appraisal Actual Actual (Ln. 2113) (Lns. 2113 & 2554) (1983-1987) 1983-1989 BDET Lending for EMIs 14.01' 9.4 19.3 (TD million) % of BDET Total Approvals 14.0 4.9 5.0 Bank Lending for EMI 7.0 5.2 12.3 (TD million) Total Cost of EMI Projects Financed by BDET 40.0 27.1 51.5 (TD million) % of BDET Lending in EMI Project Cost n.a. 34.7 EMI Projects Cost Financed by BDET as a Percentage of Total Projected EMI Investments (Plan) 201 8.71i 4.61' Number of Jobs Created 6,000 752 1,005 Average Investment Cost 3, per job (TD) 6,750- 36,000 51,200 1 Assumed from mid-1982 to mid-1984. - Fixed investment in EMIs amounted to TD 313 million in 1983-1987 (cf. Annex 2). I Fixed investment in 1987-1989 amounted to TD 460 million (in current dinars). t Equivalent to US$13,500. - 40 - 185. Bank lending to BDET for EMIs amounted to TD 24 million (equivalent to US$20 million) under Loan 2554. EMI projects costs approved under that Loan amounted to TD 24.4 million of which TD 9.9 million were financed by BDET including TD 7.1 million from Loan 2554. Jobs created numbered 253 only. These data indicate that the situation over the period 1985-1989 did not improve due to the continued slackening of demand for EMI projects financing. One must remember that Loan 2554 was approved partly on the ground that, in order to benefit from reduced interest rates, the pipeline of EMI projects would be transferred by BDET from EMI-I to EMI-II and would continue to grow further after 1985. This did not take place. 186. As regards the EMI component, only about two-thirds of the expected demand materialized in dinars. However, in dollar terms (the dinar was formally devalued in 1986), disbursements amounted to US$6 million against US$14 million originally earmarked in Loan 2113, i.e., 43% of the amount earmarked for EMI by Loan 2113. 187. BDET financed only 8.7% of total EMI projects costs in 1983-1987 and even less (4.6%) when including 1988-1989 because BDET lending increased substantially in recent years while lending to EMIs remained relatively marginal. 188. The number of jobs created was very low: 752 only in 1983-1987 and 1,005 in 1983-1989. Even worse, the average investment cost per job was 5 to 7 times higher than what was estimated at appraisal, even after taking into account the 4evaluation of the dinar. 189. Most EMI projects financed under Loan 2113 were non-profitable and all but one are still in arrears to date. B. NON-EMI PROJECTS FINANCING 190. As already mentioned, it was decided (perhaps wisely with hindsight) to allocate half of the proceeds of the loan to non-EMI industries. Here the performance was better: Appraisal Actual (1983-1987) BDET Lending for non-EMI Industries 14.0 11.8 (TD million) % of BDET Total Approvals 14.0 11.4 Bank Lending for non-EMI Industries 7.0 9.0 Total Cost of non-EMI Industrial Projects Financed by BDET (TD million) n.a. 71.3 % of BDET Lending in Project Cost n.a. 16.5 Number of Jobs Created n.a. 1,305 - 41- 191. BDET lending for manufacturing under Loan 2113 was somewhat lower than expected at appraisal in current prices and also lower when expressed as a percent of total BDET lending to manufacturing. Bank loan disbursements amounted to TD 9 million equivalent to US$11.7 million, i.e., 16% lower than the original US$14 million target. But this was a much better performance than for the EMI subsector. Cancellations for non-EMI industrial projects amounted to US$2 million, while for EMI projects they amounted to US$7.6 million. 192. Employment creation was larger than for EMI projects but investment cost per job was higher with TD 54,626. The share of non-EMI projects cost financed by BDET was relatively low, i.e., 16.5%. However, when excluding a large type plant with total project cost of TD 51 million, to which BDET contribution was TD 3.5 million only, BDET share in industrial project costs was 39.1%, very close to BDET share in EMI project costs. Thus, overall, results obtained through this component of the EMI-I project were rather satisfactory, except perhaps regarding the small number of jobs created. This contrasted with the dismal picture of the EMI lending component. 193. The poor results obtained in the EMI sector have to be understood in the context of declining economic conditions. Growth in value added in EMIs decelerated from 12.5% per annum in 1980-82 to 5.6% in 1983-85. In the crisis years (1986-87) average growth was negligible (0.15%) and only in 1989-90 was the sector somewhat revived. Employment rose very slowly in 1982-88, i.e. by about 3.5% per annum and fixed investment in EMIs actually declined by 37% from 1984 to 1987 (at current prices). The share of EMI investments in total industrial investment increased, but this is because EMI investments are more capital intensive and the 1986 devaluation of the dinar substantially increased EMI important machinery costs. This was reflected in 1987-89 figures. Value Added, Employment and Investment (% of EMI in Manufacturing Sector) 1982 1987 1989 Value Added 14.9 14.1 13.7 Employment 16.8 16.1 n.a. Investment 13.7 17.7 18.9 Source: Annex 2. - 42 - C. TECHNICAL ASSISTANCE a. CETIME 194. The institution building of CETIME was only partly successful. CETIME has suffered from its regulatory role and the resulting lack of confidence of the Tunisian private sector which re.nained for some time reluctant to seek in-plant technical assistance. In view of this, CETIME has tried to develop various technical centers and obtained through Loans 2113 and 2554 substantial resources which were entirely utilized for this purpose. It thus became manufacturer of components and tools sold to industries. As a result, the original mandate to provide in depth technical assistance to EMIs became somewhat secondary. 195. The key issue facing CETIME is how to measure its success. At one level, CETIME provides Tunisia with technical systems that it would not, otherwise, have and thus is good as it enables Tunisian engineers to have access to modern equipment and facilities without the expense and difficulty of going abroad. Also, small and medium industries with modest resources can benefit from CETIME more than large enterprises can. But at another level, CETIME's overall impact on the availability and development of technology for EMI is very small, as the industries easily obtained whatever technology they want from abroad at modest relative costs, with faster delivery, and more tailored to their needs. 196. The fact remains that EMIs remain technologically rtlatively backward and poorly integrated. This implies that CETIME had little imi.act on industries over the last seven years. But what was overestimated was not only CETIME ability to solve EMI's technical problems in a few years, but above all the time needed to transform and "assembly-type" of industries in a truly modern and efficient electrical/mechanical subsector. 197. CETIME feels that it has provided jobs and careers opportunities to a number of engineers and technicians, including those who moved from CETIME to become entrepreneurs. It also claims to have assisted a number of small and medium companies who could hardly obtain assistance elsewhere. But CETIME's anomalous position as part manufacturer (e.g., the Sousse high tech plant which makes moulds and tools) and part R&D operations needs to be clarified, as does the degree to which it is funded by the Government. In its present Government- owned status, CETIME is vulnerable to being subjected to pressure to increase national technical independence--an obsolete concept in today's technically interdependent world. CETIME to more fully succeed may need to achieve greater autonomy and greater efficiency. b. INNORPI 198. INNORPI has achieved a considerable amount of standardization work, which had been long due, in order to introduce harmony and discipline in the area of product quality in Tunisia. Most standards, selected by consensus between the agency, industrialists and the Government, meet minimal international standards and cover a wide range of products (and some services) produced and used in Tunisia. The number of standards produced (cumulative) rose from 12 in 1982 to 2,420 in 1988 (of which 14% in EMIs). The number of patents/models/copying registered (cumulative) was 1,160 (of which 800 foreign) in 1988. Product - 43 - certifications number about 17 of which 14 products in the cement industry. Training has been provided for quality circles (workers meet to discuss issues and improve quality) in six industrial enterprises and two public agencies. 199. Those achievements since 1983 are quite commendable by any measure and Bank financing has been well used. One issue, however, is the scant knowledge about the role and the capabilities of the existing laboratories in Tunisia. INNORPI may have wished to develop its own laboratories but existing laboratories have expressed a wish to rather have their facilities used to the maximum extent (not always known). Some of these laboratories may be certified as capable to be consulted for efficient testing and experiments while INNORPI would essentially play a coordinating and advisory role among existing laboratories. D. THE PROTECTION ISSUE 200. The Effective Protection study was finalized with some delays and suffered from major defects such as: (i) the lack of direct price comparisons between locally made and EMI products manufactured abroad and (ii) the non- inclusion of QRs in the calculations which paused a difficult methodology problem. 201. Nevertheless, this study was useful in 1984 to keep alive the debate between the Bank and the Tunisian authorities on the need for decreasing tariffs and reducing QRs in order to liberalize the economy and thus give impetus to industrial competitiveness and exports. 202. The EMI-I project, and also later on EMI-II, tried unsuccessfully to deal with incentives and protective policies through an elaborate set of eligibility criteria for EMIs mixing conditionalities on the level of tariffs (18% to 21% over CIF prices) and some QRs reductions under well defined circumstances. For each subproject to be financed under the Bank loans detailed information had to be submitted on comparative prices of similar local and foreign products. This proved impractical. The possibility remained to calculate economic rates of return (which was done to some extent) but this was supposed to be standard practice in project evaluation. Consequently, the complicated and rather ineffectual criteria set under EMI I and II, stemmed only from the well-founded feeling in some quarters of the Bank, that it was wrong to finance the highly protected EMI sector without doing something about protection in general. The Region's answer was that financing EMIs was justified to provide employment and increase much needed exports. Since Tunisian authorities were basically opposed until 1986 to dismantle ti-e protection system, the Region felt that the only possible compromise was to set up a system by which Bank financed subprojects would remain relatively "competitive" with moderate tariffs and quotas. The result was a sector-oriented loan channelled through a DFC, mainly geared to provide funds to EMIs but with little or no impact on key sectoral policy and incentives issues other than through narrow eligibility criteria for projects. The outcome is known: not only EMIs proved to be the wrong target for Bank financing, but the Tunisia authorities never agreed to change existing protectionist policies until the 1986 crisis led them to shift to a basically new, more liberal approach, supported by two Bank financed SECALs and one SAL in 1987-1989. - 44 - E. THE SUBSECTOR STUDIES 203. The platework subsector study was never completed. Following a swift beginning, methodological and technical difficulties soon began hampering its execution. This was compounded by the unfavorable evolution of that branch of the sector, which led to the suspension of the study. Here again, the strategy supported by the Bank and the Tunisian authorities in the late 1970s and early 1980s to develop a particular subsector proved unrealistic and unworkable. 204. The foundry subsector study was seen essentially by the Tunisian authorities as part of the exercise to obtain financing for the SOFOMECA project. The latter failed largely because of the wrong, strategy adopted for the transport equipment and agricultural machinery subsector. The change in economic policy in 1986 led to the failure of the SOFOMECA project. F. INSTITUTIONAL BUILDING EFFORTS AND BDET PERFORMANCE 205. The EMI I and II projects included provisions regarding the need to increase BDET's efficiency in promoting and supervising operations. These provisions were partially implemented. However, assumptions made regarding volume of lending and profitability were found overoptimistic and little justified given basically wrong judgements regarding EMIs, other industries and tourism development. 206. BDET failed to develop in time a full awareness of the fragility of the demand for investment financing for the EMI sector. This might have been due to significant efforts that BDET had made identifying an appropriate way to develop Tunisia's capabilities in that sector, and the resulting difficulties that BDET then had acknowledging that investors were not fully convinced of the feasibility to enter that sector (PCR, para. 47). 207. As indicated in the section on Project Implementation, EMI-I conditionalities have been only partially met. The minimum 2% spread objective was achieved and Government subsidies were provided to subsidize interest rates. But a big disappointment was the large backlog of Government payments to BDET as compensation for foreign exchange losses. A study planned to deal with this issue was repeatedly delayed and only now some kind of system is emerging to cover BDET foreign exchange losses. 208. Vast arrears have accumulated over the years and seem to have reached a plateau only recently. Such arrears were linked to poor economic performance and financial results of borrowing companies. BDET also made insufficient efforts to recover arrears at least for some time. Profitability declined. Return on net worth much lower than expected at appraisal and, until 1989, dividends never reached the 8% of par value share originally anticipated. Then, overall, BDET's operations and portfolio did not perform well under EMI-I and also EMI-II at least until 1988-1989 when BDET's situation finally seemed to improve after several particularly difficult years. - 45 - G. OVERALL RESULTS 209. Simply stated, the EMI projects tried to do too much. They aimed at developing the EMI sector while expanding and strengthening BDET (and other development banks), and bringing about some rather major policy changes. The projects clearly failed to develop the EMI sector although they succeeded in fostering INNORPI development as an institution in charge of controlling quality to encourage EMI products export. They did not play a major role in strengthening BDET but after all project conditionalities related to the latter were rather limited and the Bank kept a "normal" relationship with BDET which led to recommendations to improve its performance. The projects did not succeed in convincing the Tunisian authorities to change industrial policies in the context of a purely sectoral operation. However, discussions initiated at the occasion of EMI I and II, indirectly gave an opportunity to pursue a dialogue between the Bank and the Tunisian Government which eventually led (after 1985) to a significant liberalization of the economy and of the industry, including the EMI sector. V. SUSTAINABILITY 210. The question is to what extent were the EMI Projects likely to maintain an acceptable level of net benefits throughout their economic life. 211. The primary objective of the projects was to encourage the development of the electrical and mechanical industries. Part of the loans was also to be used to finance technical assistance to EMI through two new institutions (CETIME and INNORPI). The project also aimed at putting ceilings on the level of protection offered to EMIs. 212. A secondary objective of the EMI Projects was to provide funds for general purpose lending to industry since the Borrower (BDET) and Bank staff hesitated in tying the entire amount of the loan to a single subsector. However, the first priority remained financing of EMI's and the Project was entitled "Electrical and Mechanical Industries Project." 213. A third objective was obviously to provide needed funds to BDET, BTKD, and STUSID, and use this opportunity to introduce some conditionalities such as reasonable spread, Government's assurances regarding subsidies to be provided to BDET and strengthening of promotion and supervision efforts. 214. As regards the first objective, i.e., assisting in building a stronger EMI sector, the projects have not been sustainable since EMIs have not developed as expected not only because of poor macroeconomic conditions but also because the whole strategy of promoting transport equipment and agricultural machinery industries failed. EMIs were also unable to develop technologically as quickly as expected. The Projects proved much too optimistic regarding the feasibility of transforming a relatively backward, non-integrated, assembly types of industry into plants technically advanced, integrated or well linked to a network of domestic and/or foreign subcontractors. CETIME and INNORPI made efforts to help - 46 - solving EMI'S engineering and quality problems, but the whole issue has proved much more difficult and complex than could be achieved though a limited financing operation using a DFC, itself without the necessary means to seriously contribute to the solution of problems much deeper than the lack of foreign exchange. Evidence of this is to be found in EMI-II whi-ih also failed because not only demand for financing of EMI projects dried out but also due to the fact that the period covered (1985-1990) was much too short to basically reform and expand an entire subsector with so many problems and constraints. 215. One major reason for the lack of sustainability of EMI I and II was the impossibility to solve though a purely subsectoral approach the whole question of incentives, customs duties, quantitative restrictions, etc., which hampered the development of the economy. The Tunisian authorities consistently resisted changing protective policies in the context of a limited EMI sector operation. It is only in the framework of a larger dialogue which started in 1986 and was concretized through two SECALs and one SAL that the Tunisian authorities have determined that the most effective way to promote competition and efficient export orientation was through trade liberalization, complemented by a set of incentives involving minimal administrative interventions. 216. It could be argued that the part of the Loan which was used for general purpose lending to industry and was better utilized than the EMI portion, has probably helped strengthening the manufacturing sector. And, in this case, the project has, in some sense, proved sustainable. It could also be said that BDET has benefitted from the Bank loan(s) as an institution since some strengthening of its promotion and supervision departments has taken place. After suffering serious financial difficulties in 1985-1988, BDET level of lending and profitability is now improving but this is not directly linked to the Bank financed EMI I and II projects. 217. The spirit of the two loans was to support the development of the EMI sector. But such development has not taken place to a significant extent and practically none of the growth in output, employment and investment original targets has been reached. A large part of the funds earmarked for EMI financing had to be cancelled, a number of the subprojects financed are in arrears and the quality of the institutions and their effectiveness in assisting EMI development remain uncertain. For these reasons, it cannot be said that the Project may be judged as likely to be sustainable in the end. VI. LESSONS OF EXPERIENCE 218. The review of the experience with EMI I and EMI-II, including the circumstances that led to their sanction, offers instructive lessons that may provide a better defined framework for shaping the Bank's position and approach to industrial lending and sectoral policy issues. 219. As regards the identification #nd the conceptualization process of the projects: - 47 - a. A clear distinction has to be made between objectives of industrial lending: it is difficult, if not impossible, to mix the objective of supporting the development of an entire industrial subsector and the objective of strengthening a particular financial development institution, particularly when it is not at all evident that the latter would prove to be the most competent channel for evaluating and supervising specific EMIs. b. The desire to pursue financial assistance to one or several development finance companies (the loan to BDET was to be the eighth) cannot alone justify the selection of particular new themes (such as EMI's sector development) using as an excuse that a new line of credit would not "fly" in the Bank without new "ideas." c. After a quick identification of a new "theme" has taken place, an in-depth "Sector Study" may be necessary but this might, later on, force the Bank to intervene in the financing of the sector. In this case, too deep Bank's involvement may reduce its degree of freedom in choosing to intervene. d. The EMI Sector Study was a good and thorough study but did not have enough vision to gauge the tremendous policy problems affecting not only the EMI sector but the entire Tunisian economy. Pressure of lending to BDET through using a new "theme," i.e., industrial sectoral development helped in hiding this basic problem, which led, in part, to the failure of the EMI project. e. The Sector Study concluded that the first set of constraints in the EMI sector originated from the inadequacy of the policy and institutional framework to promote higher factor productivity and efficiency, to encourage export competitiveness and orientation, and to generate efficient subsectoral strategies. The second set, related to the first one, was largely internal to the enterprises and resulted from the weak design and management of production operations (product mix, scarcity of firm specialization and sectoral integration, capacity planning and capital intensity, labor productivity). But these conclusions could not easily be translated into a clear agenda for action in the traditional framework of a DFC loan. The first set of policy constraints to be addressed resulted in protracted discussions within the Bank and with the Tunisian authorities. Experience showed that adopting a new industrial and trade policy could not be done through a single project financing selected priority EMIs. The lesson is thus that such type of project cannot provide a vehicle for substantial and lasting changes in the incentives framework, custom tariffs, quantitative restrictions, etc. - 48 - f. The second set of constraints resulted from deep weaknesses of EMIs at the management and technical levels. Experience has shown that one or even several development banks (and some of them under EMI-II were in any case very new) did not posses the technical and managerial capability to intervene efficiently in an entire subsector where specific knowledge of technological problems is required. Similarly, new institutions to provide ta-plant technical assistance and also to increase quality control consciousness, although justified, takes a long period of time to achieve quantifiable, positive results. To assume that such institutions can have a sizable impact on a whole sector during the short implementation period of a Bank line of credit, is to totally underestimate the depth and the extent of problems to be solved. g. When the decision was made to assist in financing the EMI sector, disagreements developed between the Bank and the Borrower on the approach to EMI development. The Bank's view was to concentrate on the expansion of labor intensive enterprises with an actual or potential comparative advantage. But the Tunisian authorities also insisted that large, new EMI complexes had to be set up on the ground that various externalities (training of labor force, backward linkages, etc.) would justify the higher protection. Despite such disagreements and because of the deep earlier involvement of the Bank in the sector through its in-depth sector study of EMIs, it was decided to go ahead with the financing of a subsector loan to BDET, which in fact included at least partial financing of such large complexes and/or plants producing components for these complexes. A few years later, these industrial complexes collapsed with dismal consequences on the Bank financed projects through a ripple effect. The lesson is that before embarking in an industrial subsectoral loan, it is important to elucidate the Borrower detailed strategy for the development of the subsector. In this case, this "thinking" was mainly to have some industrial development "poles" which would generate a number of smaller projects leading to progressive integration of the subsector. The approach was also to mainly serve the domestic market behind a high wall of protective measures. The whole operation would have been financed with Bank loans and other financial resources. The lesson is obviously that clarity of objectives and true commitment of the Borrower to such objectives, as agreed with the Bank, are essential to the success of such subsectoral projects. 220. As regards implementation of the projects, useful lessons have been learned: a. Fixing detailed subsectoral targets regarding output, employment and investment in the EMI sector proved illusory and dangerous. The lesson is that directed sectoral lending - 49 - through a DFC can hardly succeed when the DFC is in no real position to influence developments in a given entire sector, and has no direct leverage on policy and technical constraints affecting this sector. b. Funds which had been earmarked for general industrial lending were more largely utilized under EMI-I than funds for EMIs. This confirmed the above lesson that too specialized lending particularly in difficult to apprehend and technically more complicated EMIs should be avoided, unless much more attention is paid to policy and technical assistance issues. c. The fact that half of the EMI loan proceeds were diverted to finance general industrial lending proved a certain ambivalence on what the project was really to achieve. On one hand, the objective was to foster EMI development and liberalize industrial sector policies but, on the other hand, the objective was mainly to provide funds through an eighth line of credit to a DFC, with some reluctance to seriously raise policy issues and to mix too closely financial and technical assistance. The lesson is, here again, that project implementation can only suffer from the need to achieve too many objectives too quickly, particularly when such objectives are in practice almost impossible to implement over a period of just a few years. d. Although the first line of credit had been only partially committed, it was decided to finance a second line of credit for EMIs which was more than three times bigger than under the first operation. Such decision was made without taking into account growing signs of a deep deterioration in macroeconomic conditions which had to have an impact on the EMI sector. The lesson is that a purely subsectoral approach, particularly in industry, may obscure the need to see clearly what the overall economic situation is in the country. A narrow link between the subsector approach and the institution building aspects (in this case a DFC) uan only compound the problem. The Bank EMI Sector Study first took a long-term look at the EMI sector issues and it was then decided to go ahead with financing through a DFC. But neither the Sector Study nor the DFC had been or was in a position to analyze short-/medium-term fluctuations in the sector and to link these fluctuations with changes in general economic conditions. The lesson is that surveillance mechanisms have te be put in place to follow up on sectoral developments and that whichever institution is in charge of implementation of a subsector loan must be equipped to follow up on changing sectoral conditions. e. Close supervision must be made by the Bank for this type of operations. One important lesson from EMI I and II is that these subsector loans, which were a new type of operation in the Bank in the early 1980s, had to be supervised very closely, - 50 - particularly regarding the technical and management issues of specific EMI subsectors. Unfortunately, such supervision remained very limited. f. The implementation of detailed eligibility criteria for industrial subprojects is always difficult to follow on and supervise over time, often because the borrowers and subborrowers are not convinced or do not want to make the analytical effort to satisfy Bank imposed criteria. The fact that the conditionality to supply price comparisons between similar EMI products produced in Tunisia and abroad (mostly Europe) was not implemented is quite revealing of this practical difficulty. Too detailed conditionalities are not realistic and cannot substitute for broader reforms in policies and incentives applying to industry or the economy at large. g. BDET's operations and portfolio deteriorated during project implementation. Very few of the financial assumptions made at appraisal proved reliable. Volume of lending stagnated or declined in real terms and profitability fell due to large arrears and Government's reluctance to fully cover losses due to foreign exchange fluctuations. Bank staff devoted time to dialogue with BDET on these issues and suggested improvements in recovering loans arrears and negotiating payments by the Government. However, since few supervision resources were available, the impact may have not been too substantial. The lesson is that the dichotomy from which the project suffered (priority assistance to the EMI sector versus traditional financing of development banks) reduced effectiveness in dealing with different, non-complementary objectives. 221. The experience with technical assistance to the EMI sector, provides some interesting insights: a. It is extremely difficult to effectively support a sector such as the EMI sector from a technical and managerial viewpoint when there are no effective, nationwide institutions in place to provide technical assistance. CETIME and INNORPI received financial assistance under EMI I and II. It took naturally several years for these institutions to be set up and have some impact. In the meantime, the extent to which EMIs have been definitely and clearly helped by the new institutions is not entirely clear, although results were obviously obtained, particularly in setting quality standards and providing certifications. b. CETIME's regulatory functions made it necessarily anpear as a Government agent with resulting neg&tive connotations, since the private sector may prove reluctant to request in-pl.gnt personalized advice and assistance from such "agent." The lesson is that specialized "Centers" must be fully autonomous when dealing with enterprises and that, when they have to - 51 - discharge official functions, this must be done clearly and separately. c. The technical assistance institutions financed under such a subsector project could have only a very small impact on the availability and development of technology for EMIs, not only because they have had only a few years to operate but also because the number of industries easily obtain whatever technology they want from abroad at modest relative costs, with faster delivery, and more tailored to their needs. The lesson is that when devising an industrial sector operation which implies dealing with technological and quality issues, it is not enough to set up or to revitalize a few domestic technical centers. A broader and systematic review of these issues must be undertaken and problem-solving orientations presented including access to transfer of technology and R&D from abroad. -53 - Attachment (These comments are taken into account in page 22, para. 88) BANQUE DE DEVELOPPEMENT ECONOMIQUE DE TUNISIE EMETTEUR a B.D.E.T DATE : 15/10/91 IRECTION EXPLOITATION Nombre de Pages 1 ESTINATAIRE : B.I.R.D. Ng- APPEL TELECOPZIEUR--r COL 93314133 ATTETION MR. MARK BAIRD CHEF DE DIVISION POLITIQUES ECONOMIQUES, INDUSTRIE ET FINANCE DEPARTEMENT DE L'EVALUATION RETROSPECTIVE DES OPERATIONS NOUS AVONS L'HONNEUR D'ACCUSER RECEPTION DE LA VERSION PRELIMINAIRE DE VOTRE RAPPORT D'VVALUATION RETRDSPECTIvE D2S PRETS 23-13 ET 2554 ÿ?ROJETs IEM I & II) DONT LA LECTURE NOUS A PERMIS D'APPRECIER LA PERTINENCE DES CONCLUSIONS ETe ENSEIGNEMENTS TIRES DE CETTE ANALYSE ET PART:CULIEREMENT L'EXTREME DIFFICULTE D'EXECUTER DANS DE BONNES CONDITIONS UN PROJET DE FINANCEMENT TROP CIBLE ET ORIENTE VERS UN SOUS SECTEUR BIEN PARTICULIER. LES PERFORMANCES DE LA PARTIE A CARACTERE GENERALE DE LA DEUXIEME LIGNE CONFORTE CE CONSTAT ET DEMONTRE L'INTERET DES PRETS A CARACTERE GENERAL POUR UNE INSTITUTION DE FINANCEMENT DU DEVELOPPEMENT NGM'DU SIGNATAIRE ../.. DATE E I R - 54- BANQUE OE DEVELOPPEMENT ECONOMIQUE DE TUNISIE EMETTEUR : B.D.E.T DATE IRECTION Nombre de pages 2 ESTINATAIRE N*- APPEL TELECOPIEUR PERMETTEZ-MOI PAR AILLEURS DE VOUS FAIRE PART DANS CE QUI SUIT DE MES OBSERVATIONS*SUR CERTAINS POINTS DE L'NALYSE DES PERFORMANCES DE LA BDET : / ARR=eRES IL EST INDIQUE (PAGE 25 8 88) « QU'AUJOURD'HUI ENCORE LA BDET NE S'EMPLOIE PAS SYSTEMATIQUEMENT A OBTENIR INDIVIDUELLEMENT DE SES CLIENTS UN ENGAGEMENT FORMEL DE REMBOURSEMENT DES ARRIERES» EN FAIT LES ACTIONS DE RECOUVREMENT DE LA BDET S'OPERENT COMME SUIT t a) RECOUVREMENT CREANCES CONTENTIEUX z NONOBSTANT LA LENTEUR DES PROCEDURES JUDICIAIRES DE TELS RECOUVREMENTS LA BDET ADOPTE UNE APPROCHE PRUDENTE DANS LES REALISATIONS DE SES SURETES REELLES AFIN D'ASSURER UNE REUTILISATION DES ACTIFS ET MINIMISER SES PERTES DE CREANCES. A CE TITRE LES RECOUVREMENTS PAR ZETTE VOIE, ONT PERMIS DANS LA MAJORITE DES CAS DE REALISER DES PLUS VALUES EU EGARD A LA VALEUR COMPTABLE NOM DU SIGNATAIRE DE NOTRE CREANCE DATE ET SIGNATURE - 55 - BANQUE DE DEViLOPPEMENT ECONOMIQUE DE TUNISIE EMETTEUR : B.D.E.T DATE IRECTION Nombre de Pages : 3 ESTINATAIRE : N40 APPEL TELECOPIEUR b) RECOUVREMENT NORMAL NOTRE SYSTEME DE RECOUVREMENT DES ARRIERES PREVCIT L'APPLICATION D'UN TAUX PENALISANT LES CLIENTS EN IMPAYES : . POUR LE PRINCIPAL t JUSQU'EN 1990 LE TAUX DE PENALITE ýTe DE 1% A 2% EN SUS DU TAUX DU PRET, CE TAUX A DEPUIS ETE RELE SORTE QU'ACTUELLEMENT LES CREANCES IMPAYEES DES PRETS* ENAGES A. PARTIR_DE CETTE DATE SUPPORTENT UN TAUX FINAL DE 16%. . POUR LES INTERETS t CEUX-CI FONT L'OBJET D'UNE CAPITALISATION UNE ANNEE APRES LA DATE DE L'ECHEANCE. N.B. s LES INTERETS ETANT RECLAMES PAR SEMESTRIALITE ET D'AVANCE, LA CAPITALISATION INTERVIENT EN REALITE 6 MOIS APRES QUE CES INTERETS NE SOIENT RELLEMENT ACQUIS. NOM'DU SIGNATAIRE DATE ET SIGNATURE - 56 - BANQUE DE DEVELOPPEMENT ECONOMIQUE DE TUN;S!E EMETTEUR : B.D.E.T DATE IRECTION Nombre de Pages : 4 /_-> ESTINATAIRE NO= APPEL TELECOPIEUR : PAR AILLEURS ET POUR LES PROJETS JUSTIFIANT D'UNE GENE DE TRESORERIE PASSAGERE, DES CALENDRIERS DE REMBOURSEMENT SONT SYSTEMATIQUEMENT ETABLIS T MATERIALISES PAR DES ARRANGEMENTS. 2/PROVISIONS LES PROVISIONS POUR RISQUE DE NON RECOUVREMENT CONSTITUEES PAR LA BDET ONT ETE TOUJOURS JUGEES SUFFISANTES PAR NOS AUDITEURS EXTERNES. RESTANT A VOTRE DISPOSITION POUR TOUT RENSEIGNEMENT COMPLEMENTAIRE. SALUTATIONS DISTINGUEES.- NON DU SIGNArAIRE DATE ET SIGNATURE - 57 - Annex 1 BDET CUKULATIVE DISBURSEMENT OF FUNDS (US$000) ForecastI Aatualt Actual/Forecast (SAR) (%) December 31, 1983 6.900 8.843 127.5 June 30, 1984 11.800 11.499 97.4 December 31, 1984 21.000 15.972 76.1 June 30, 1985 24.300 16.631 68.4 December 31, 1985 27.100 17.601 64.9 June 30, 1986 28.500 18.693 65.6 December 31, 1986 29.700 20.697 69.7 June 30, 1987 30.500 21.367 70.1 Source: World Bank Loan Disbursement System. V Includes credit line to BDET (US$28 million), subloan to Government (US$2 million) and US$0.5 million front-end fee. Li Cancelled funds amounted to US$9.6 million. ELECTRICAL AND MECHANICAL INDUSTRIES (EMI) Annex 2 VALUE ADDED, EMPLOYMENT AND INVESTMENT IN EMI. 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Value Added Value Added (in 1980 million TD) 54.30 63.10 68.70 74.80 80.40 84.80 85.10 85.00 88.30 95.00 Percentage Growth (1) 13.80 16.20 8.90 8.90 7.50 5.00 0.40 -0.10 3.90 8.80 Percentage Share in GDP!'(%) 1.53 1.60 1.65 1.67 1.75 1.80 1.92 1.85 1.90 1.97 Percentage Share in Manufac- turing Value Added (Z) 13.01 13.52 14.88 14.91 14.86 15.21 14.61 14.06 13.47 13.70 Employment Employment (in 000) 35.5 38.8 40.5 42.5 44.3 46.3 47.8 49.3 51.4 n.a. Percentage Share in Total Employed Population (Z) 3.77 3.95 3.98 3.83 4.05 4.11 4.16 4.20 4.28 r.a. On Percentage Share in Manufac- turing Employment (Z) 16.98 17.09 16.84 16.65 16.52 16.37 16.23 16.14 16.12 n.a. Investment Fixed Investment (at current million TD) 20.20 23.40 39.60 66.00 71.60 69.60 60.00 45.80 78.00 69.60 Percentage Share in Manufac- turing Investmen- (1) 15.08 10.83 13.68 18.21 19.65 25.39 20.69 17.70 29.30 18.90 Percentage Share in Total Fixed Investment (Z) 2.02 1.81 2.42 3.77 3.73 3.76 3.56 2.80 4.70 3.50 Sources: - Production and Investment: Ministry of Planning and Finance. - Employment: Institut d'Economie Quantitative, Tunis. i' GDP at current market prices. - 59 - ELECTRICAL AND MECHANICAL INDUSTRIES (EMI) Annex 3 VALUE ADDED. EMPLOYMENT AND INVESTMENT IN EMIs (in percentage) 1980-1989 1980-1982 1980-1984 1984-1986 1986-1988 1989 Value Added Percentage Growth Rate 6.4 12.5 10.3 2.9 1.9 8.8 Percentage Share in GDP 1.76 1.59 1.64 1.82 1.89 1.97 Percentage Share in Manufacturing Value Added 14.22 13.80 14.24 14.89 14.05 13.70 Employment Percentage Growth in Empl)yment 4.7-' 6.8 5.7 3.9 3.7 n.a. Percentage Share in Total Employment 4.01; 1.9 3.9 4.1 4.2 n.a. Percentage Share in Manufacturing Employment 16.51 17.0 16.8 16.4 16.2 n.a. Investment Percentage Share in Total Investment 3.2 2.1 2.8 3.7 3.7 3.5 Percentage Share in Manufacturing Investment 18.9 13.2 15.5 21.9 22.6 18.9 &urces: - Ministry of Planning and Finance - Institut d'Economie Quantitative, Tunis. 11 1980-1988. ELECTRICAL AND MECHANICAL INDUSTRIES (ENI) Annex 4 VALUE ADDED AND NOHIRAL AND EFFECTIVE PROTECTION RATES OF EMIs (in percentage) Subsector Value 1980 1983 1984 1985 1986 1987 1988 Added NPR E_ NPR EPR NPR La NPR EPR NPR EPR NPR EVE NPR EPR Output Iron and Steel 53 6 55 11 73 13 78 14 77 15 14 16 20 17 22 Steel Producrs 35 11 74 14 71 20 79 73 87 23 47 22 41 22 40 Won-ferrous Metals 34 22 94 27 114 46 118 49 115 50 109 36 78 33 68 Metallic Construction 48 31 75 33 75 35 74 41 93 41 89 36 70 33 61 Hardware 33 30 119 30 95 31 84 36 99 36 89 32 75 30 65 Metal Furniture 30 51 678 54 694 56 801 62 954 63 848 50 306 43 182 Metal Utensils and Apparatus 23 37 323 38 222 39 161 44 183 45 140 39 141 33 99 Electrical, Elec- tronical Products 30 27 76 17 19 40 125 46 143 46 132 38 104 34 87 Mechanical and Elec- trical Motors 16 30 390 34 347 32 184 41 944 43 1536 32 211 28 118 Agriculture Machin- ery and Equipment 21 14 24 22 41 21 46 24 47 26 48 24 45 23 35 Transport Equip. Assembly and Components Manuf. 11 26 -3492 31 235 34 219 38 249 39 251 33 158 30 150 Ship Construct. and Repair 60 4 0 9 13 2 -3 2 -5 3 -6 12 9 12 8 TOTAL XMI 24 26 96 25 67 32 92 36 104 37 88 32 73 29 63 TOTAL RAUUFACTURING 29 39 242 37 178 42 153 48 207 48 124 36 81 34 78 NPRs Nominal Protection Rate EPR3 Effective Protection Rate Sources Institut d'Economle Quantitative, Tunis. ELECTRICAL AND MECHANICAL INDUSTRIES (ED) Anne5 VALUE ADDED AND W0MINAL AND EFFECTIVE PROTECTION RATES OF EMIe (Selected Items) (in percentage) Item Value 1980 1983 1984 1985 1986 1987 1988 Added NPR EPR NPR EPR NPR EPR DPi P HPR EPR P RP D EPR Output Iron 52 5 35 11 49 12 51 13 50 13 12 16 19 17 21 Steel 56 7 136 12 177 14 195 15 187 14 17 16 23 17 25 Foundry Products 47 28 71 29 68 31 73 35 82 35 78 30 58 28 53 Platework 42 33 95 35 91 38 100 45 130 45 127 39 94 17 84 metal Furniture 30 51 678 54 694 56 801 62 954 63 848 50 306 43 182 Cooking Stoves 16 55 -1613 36 1883 57 374 44 728 42 327 36 281 33 201 Transformers 46 16 23 23 37 24 32 27 35 27 34 26 36 28 40 Radios 3 46 -119 28 -176 32 -181 37 -190 37 -204 31 -206 28 -226 Televisions 25 14 11 -10 -48 58 1298 64 1242 64 1028 51 397 44 246 Batteries 25 36 418 41 739 43 233 48 213 48 180 43 148 39 117 Meters 3 12 -76 23 49 24 45 21 -64 23 -57 23 -20 24 186 Electr. NotGre 25 10 0 24 53 25 57 28 78 28 72 26 66 16 22 Mechanical Motors 9 51 -161 45 -208 45 -208 51 -196 51 -106 37 -430 31 11129 Cars 7 65 -147 42 -269 44 -335 45 -575 47 -426 39 -1148 33 940 Trucks 11 28 228 22 29 27 40 29 29 29 25 25 20 26 60 Buses 6 29 87 30 68 31 21 37 77 37 85 37 85 28 56 NPRs Nominal Protection Rate EPR: Effective Protection Rate Source: Institut d'Economie quantitative, Tunis - 62 - Annex 6 OUTPUT AND INVESTMENT IN ELECTRICAL AND MECHANICAL INDUSTRIES 1982-86 1982-86 1987-89 (Forecast (Actual) (Actual) 6th Plan) Output Value added in EMls (%) (Percentage Growth p.a.) 18.7 5.3 5.7 Percentage Share of EMIs in Manufacturing Output (%) 19.0 14.6 13.7 Percentage Share of EMIs in Gross Domestic Product (%) 3.3 1.9 2.0 Investment Value of Investments (In million Dinars at current prices) 480 307 193 Percentage Share of Total Manufacturing Investment (%) 27.0 19.3 21.3 Source: Ministry of Planning and Finance - 63 - Annex 7 EMI INVESTMENTS IN THE 1982-1986 PLAN (TD million) 1. Projects initiated/undertaken before 1982 9 2. Maintenance and renewal of existing EMI projects 46 3. Projects to be undertaken in 1982-1986 425 4. Priority Projects a) Priority Projects (1982-83) 90 b) Other Priority Projectsy 130 Subtotal 220 5. Non-Priority Projects a) Non-Priority Projects (1982-83) 130 b) Other Non-Priority Projects" Subtotal 205 6. TOTAL (1 + 2 + 4 + 5) 480 Source: Ministry of National Economy It Project ideas or under study to be financed after 1983. - 64 - ELECTRICAL AND MECHANICAL INDUSTRIES Annex 8 NUMBER OF FIRMS AND EMPLOXMENT BY EMI SUBSECTORS Number of Firms Employment Foundry Products 23 2,000 Platework, Metal Furniture and Other Welded Products 210 15,000 Machinery and Equipment, Diesel Engines 150 12,000 Transport Equipment (Assembly, Components) 80 8,000 Ships, Boats 60 1,000 Electrical Products 109 7,800 Electronic Products _1 3,200 693 49,000 ' Excluding Iron and Steel Plant (2,000 people employed). Source: API, November 1990. - 65 - Annex 9 NUMBER OF FIRMS AND EMPLOYMENT IN EXPORTING BLECTRICAL AND MECHANICAL INDUSTRIES Number of Fizms Employment 1. Mechanical Industries Foundries and moulds 2 7C1 Platevork and metal construction 16 2,132 Agricultural machinery, construction and transport equipment assembly 21 4,892 Transport equipment components, cables and tool. 24 1,371 Hardware, nails, bolts and nuts 14 746 Metal container and household articles 10 1,874 Metal and plastic tables and pipes 5 575 Metal furniture 3 240 Subtotal 95 12,531 II. Electrical and Electronic Industries Electrical apparatus (lamps, batteries, cables, components for cars) 27 2,263 Household equipment (refrigerators, cooking stoves, heaters, washing machines) 10 1,736 Electro-mechanical equipment (motors, transformers, ovens, air conditioners) 25 1,662 Electronic equipment (televisions, radios, antennas, micro-processors, micro-computers) 14 765 Subtotal 76 6,426 III. Total I + II 171 18,957 Source: Center for Export Promotion (CEPEX), Directory of exporting or potentially exporting electrical and mechanical industries, Tunis. ELECTRICAL AND MECHANICAL INDUSTRIES Annex 10 EMI IMPORTS AND EXPORTS (TD million) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 bIports Mechanical Products n.a 581.7 753.6 651.0 483.9 687.7 657.7 603.9 797.6 1126.7 Electrical Products 65.7 84.0 97.6 130.0 158.0 116.2 129.0 136.5 163.8 216.3 TOTAL n.a. 665.7 851.2 781.0 641.9 803.9 786.7 740.4 961.4 1343.0 Exports .echanical Products 17.0 25.7 38.3 24.7 n.a. 38.2 57.5 65.0 103.0 128.9 Electrical Products 13.7 15.6 17.0 27.6 29.6 29.7 44.8 66.8 89.5 128.9 TOTAL 30.7 41.3 55.3 52.3 a.a. 67.9 102.3 131.8 192.5 257.8 Z ExportsImports Hechanical Products n.a. 4.4 5.1 3.8 n.a. 5.6 8.7 10.8 12.9 11.4 Electrical Products 20.9 18.6 17.4 21.2 18.7 25.6 34.7 48.9 54.6 59.6 TOTAL a.e. 6.2 6.5 6.7 n.a. 8.4 13.0 17.8 20.2 19.2 Source: National Institute of Statistics. - 67 - Annex 11 BDET PIPELINE OF EMI PROJECTS Investment BDET Potential BDET Lending Investmentnti CostEMLndn asXoToa1 (1982-1983) EM(1982-983) (TD million) (TD million) M I. Priority Pro Aluminum Foundry 0.46 Leaf Springs 4.50 Excavators 0.37 Wood Machinery 2.50 Compressors 0.30 Diesel Generators 0.50 Auto Radiators 1.50 Fork Lifts 0.36 Structure/Platework (modernization) 6.00 4 Export "Compensation" Projects in Auto Industry- 33.31 Subtotal 49.80 Other Priority Projects!' 8,45 Subtotal 58.25 10-23 11-26 II. Non-Priority Projects Foundry (bath tubs, iron pipes) 10.00 Diesel Engine Mechanical Complex (Phase I) 16.00 Car Rims 1.12 Auto Seats 0.50 Auto Wipers 0.31 STIA Extension (cars) 1.80 Auto-Electrical Accessories 0.94 Household Appliances, Common Hardware 1.21 Suototal 31.88 12.13 9-10 III. Total (I + I1) 90.13 22-361' 10-16 Source: PDET, 1981. N : Minimum amount of lending by BDET In mid-1982-mid-i984 was agreed under Loan 2113-TUN to be TD 15 million of which TD 10 million for priority EMIs. Project costs( financed were to total TD 40 million, i.e. about 20( of TD 220 million assumed for priority EMI's investments under the Sixth Plan (1982-1986). SSee Annex 7. VFour Joint ventures for buy-back exports (Renault, Peugeot, Citroan) SSmall priority projects agreed by API in 1980-81 and to be eventually financed by BDET. J Equivalent to 28-46% of BDET total potential lending in 1982-83 while lending to EMI priority projects would amount to 12 to 29% of the total manufacturing lending of BDET. Annex 12 EMI PROJECTS FINANCED BY BPET (Loan 2113-TUN) Subsector No. of Project BDET IBRD Arrears Projects Cost Employmcnt Lending Fundini (11115/90) (TD 000) (TD 000) (TD 000) (TD 000) Electrical Machinery (transformers, electric motors, electric water heaters) 1 7.272 309 3.167 1.049 1.774 Welded Tubes, Radiators 1 3.400 102 300 288 495 Wheel Steel Frames for vans, small trucks 1 1.170 23 636 630 1.025 Aluminum Accessories for Electric Lines 1 2.290 25 988 970 206 Wood Working Machinery 1 2.500 38 1.360 407 724 0% Metal Springs for Trucks 1 8.000 148 1.955 1.414 628 1 Refrigerated Cabins Mounted on Trucks 1 2.050 40 815 313 495 Motors Reconditioning 1 408 67 180 177 -- TOTAL EfI 8 27.090 7521' 9.401 5.248 5.347 TOTAL INDUSTRY 18 98.393 2.0570' 21.216 14.233 n.a. I EMI 44.4 27.5 36.6 46.3 36.8 Source: BDET 11 Average investment per job is TD 36.000 and 40.000 eccluding electrical machinery and metal spring large projects. 11 Average investment per job is TD 47.800. Annex 13 Page 1 of 2 PROJECT ANALYSIS BY MANUFACTURING SUBSECTOR - LOAN 2113-TUR Number of Investment Number of Jobs Cost per Job BDET of which of which Subsector Prolects Prolect Cost Created Created Loan!' IBRD LoanV IBRD loa! (TD 000) (TD 000) (TD 000) (TD 000) (US$ equiv- alent) Enl 8 27.090 752 36.0 9419.7 5247.0 6636.2 Other Industries 10 7. 303 1305 54.6 11796.1 8986.3 11828.9 Tourism - . - - - - - Transport - - - - -L- Total 18 98.393 2057 47.8 21215.8 14233.3 18465.1 Appraisal Estimate (1981) 30 105.000 6000 17.5 21000.0 14000.0 28000.0 Creation and Expansion EMI - Creation21 5 16.0101 2741' 58.4 5752.7 3733.4 4835.1 Expansion3' 3 11.080 478 23.2 3667.0 1513.6 1801.1 Total 8 27.090 752 36.0 9419.7 5247.0 6636.2 Source: BDET !' Disbursements. 2 One wheel-frame factory, one plant manufacturing accessories for electric lines, one wood-working machinery plant, one leaf spring for trucks factory, one plant producing refrigerated cabins mounted on trucks. 2t Of which 8,000 for a leaf-springs-for-trucks factory. ' Of which 148 for leaf-springs factory. ' One electric motor and transformer plant, one welded tubes and radiator factory and one motor-reconditioning factory. Annex 13 Page 2 of 2 PROJECT ANALYSIS BY MANUFACTURING SUBSECTOR - LOAN 2113-TUN (Cont'd) Number of Investment Number of Jobs Cost per Job BDET of which of which Subsector Prolects Prolect Cost Created Created Loan-'I IBRD Loany IBRD Loan-V (TD 000) (TD 000) (TD 000) (TD 000) (US$ equiv- alent) Otber Industries Creation 6 66.59531 10851 61.4 9173.1 7001.9 9235.5 Expansion 4 4.708 220 21.4 2623.0 1984.4 2593.4 Total 10 71.303 1305 54.6 11796.1 8986.3 11828.9 Total Creation 11 82.605 1359 60.8 14925.8 10735.3 14070.6 Expansion 7 15.788 698 22.6 6290.0 3498.0 4394.5 Total 18 98.393 2057 47.8 21215.8 14233.3 18465.1 Source: BDET c ' Disbursements. V Of which 51.000 for a tire plant. ' Of which 525 for a tire plant. Annex 14 BDET - PROJECTED AND ACTUAL INDUSTRY AND TOURIS! OPERATIONS (1981-1986) (TD thousand) 1981 1982 1983 1984 1985 1986 (Actual) (Proj.) (Actual) (Proj.) (Actual) (Proj.) (Actual) (Proj.) (Actual) (Proj.) (Actual) ApprovaleI Industry 28191.5 40500.0 46763.0 37500.0 28484.0 39500.0 33372.0 41000.0 37106.0 44000.0 26114.0 Tourism, Others 32429.5 14500.0 16001.0 15500.0 11986.0 16500.0 14645.0 16500.0 7594.0 16500.0 6400.0 Subtotal 60621.0 55000.0 62764.0 53000.0 40470.0 56000.0 48017.0 57500.0 44700.0 60500.0 32514.0 2 Industry/ Total 46.5 73.6 74.5 70.8 70.4 70.5 69.5 71.3 83.0 72.7 80.3 Commitments- Total 26353.( 5300.0 57486.0 50800.0 45050.0 46300.0 45088.0 47C00.0 33510.0 48700.0 32060.0 - Disbursements- Total 26531.0 33400.0 42930.0 43400.0 48340.0 47500.0 45900.0 43200.0 40980.0 43000.0 35380.0 Sources: Staff Appraisal Report No. 3689-TUN, Annex 11, March 2, 1982 and BDET. -t Loans and participations. Annex 15 BDET - Analysis of Approvalell (TD thousand) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 (eat.) Total All Se-tors 32.090 60.621 62.764 40.470 48.017 44.700 32.514 27.823 68.343 123.439 97.746 Industry 18.532 28.192 46.763 28.474 33.372 37.106 26.114 16.819 22.510 48.418 39.536 EMSI 6.259 9.747 17.575 8.188 3.669 7.446 8.115 .868 4.717 13.402 n.a. 2 Industry/Total 57.8 46.5 74.5 70.4 69.5 83.0 80.3 60.4 32.9 39.2 40.4 1 EMfl/Total 69.5 34.5 24.8 28.7 7.6 16.7 24.0 3.1 6.9 10.9 n.a. 2 Eil/Industry 33.7 16.0 33.3 20.2 10.9 20.0 31.1 5.1 2!.0 27.6 n.a. Source: BDET 11 Loans and equity participations. Annex 16 BDET - TOTAL APPROVALS, CONIThENTS AND DISBURSEMFWTS (1971-1989) 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Approvals Forecast EMI 1 55.000 53.000 56.000 57.500 60.500 Forecast EMI II 50.500 55.000 57.500 60.000 Actual 33.194 28.633 30.486 32.089 60.621 62.674 40.470 48.017 44.700 32.514 27.823 68.343 123.439 Comitments Forecast ENI I 53.000 50.800 46.300 47.000 4R.700 Forecast EN II 46.500 48.300 44.300 46.700 Actual 21.770 22.622 22.399 22.613 26.354 57.486 45.050 45.088 33.510 32.060 25.500 30.700 n.a. Disbursements Forecast EMI 1 33.400 43.400 47.500 43.200 43.000 torecast EMI 11 43.150 44.100 44.200 43.900 Actual 20.372 22.280 25.561 24.093 26.531 42.930 48.340 45.900 40.980 35.380 39.600 36.600 n.a. I Sourcess Staff Appraisal Reporte No. 3689-TUN, dated March 2, 1982, and No. 5487-TUN, dated Nay 1, 1985, and BDET. Annex 17 LAN AND PARTICIPATIONS APPROVALS IN MANUFACTURING AND IN ElIs (BDET. BTKD. STUSID) (TD thousand) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 I. Manufacturing BDET 18.532 28.191 46.763 28.484 33.372 37.106 26.114 16.819 22.510 48.417 BTED -- ---68.070--- 38.230 38.015 16.179 14.378 17.643 9.718 8.317 STUSID -- 8.388 40.745 31.398 Subtotal 18.532 98.112 II. EHIs BUDET 6.259 9.747 17.575 8.188 3.669 7.446 8.115 .868 4.717 13.402 BTKD -- ---27.356--- 11.801 7.775 6.905 2.108 2.300 1.092 2.057 1 STUSID -- 18.183 11.316 7.170 4.423 3.161 .876 4 Subtotal 6.259 31.305 18.614 7.591 8.970 16.335 Sourcess BDET, BTED and STUSID. 11 BDET had approved 12 EMI projects in 1977 for TD 1870 thousands, 20 in 1978 for TD 7305 thousands and 9 in 1979 for TD 1915 thousands. Annex 18 BDET - ELECTRICAL AND MECHANICAL INDUSTRIES Approvalst Commitments. Disbursements and Arrears (TD thousand) 1980 1931 1982 1983 1984 1985 1986 1987 1988 1989 1984-1989 Approval&V 6.259 9.747 17.575 8.188 3.669 7.446 8.115 .868 4.717 13.402 38.217 Commitments n.a. r-S. n.a. n.a. 2.051 3.126 3.460 .808 2.621 7.239 19.305 Disbursements n.a. U.a. n.a. n.a. 1.998 2.779 3.198 .669 1.401 4.596 14.641 Number of Projectall 20 20 22 20 13 9 8 5 6 10 41 Arreare3l n.a. U.a. a.a. n.a. .186 1.290 .478 .218 .185 .268 2.625 Number of Projects n.a. n.a. n.a. U.a. 8 4 5 2 1 2 22 with Arrears Source: B DET .' Approvals projected under EMI-I were TD 22-36 million in 1982-83. Approvals estimates under EMI-Il were TD 13.4 million in 1985, 14.4 in 1986, 15.4 in 1987 and 16.3 in 1988. It Commitments and Disbursements. ., .cincipal and interests. Annex 19 BDET FINANCIAL INDICATORS (TD thousand) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Loans Outstanding 91.921 101.185 118.255 143.630 167.886 185.413 196.406 205.746 213.457 223.187 Net Profit/Equity Ratio (2) 8.97 8.49 7.03 8.98 9.24 6.60 6.99 5.26 6.70 13.03 Net Profit (after tax) 1.390 1.440 1.440 2.390 2.540 2.400 2.594 1.971 2.532 5.299 Capital Gainal 270 360 410 170 260 233 150 435 1.591 3.323 Governmet Subsidyv 454 1.032 460 2.945 5.848 2.978 3.875 4.540 2.978 3.104 Average Rate of Lending (2) 8.83 8.79 9.06 9.54 9.99 10.31 10.54 10.60 10.75 11.75 Average Rate of Borrowing (Z) 7.39 6.94 6.39 6.93 8.44 8.10 7.78 8.33 8.14 8.55 Margin on Borrowings (2) 1.44 1.85 2.67 2.61 1.55 2.21 2.76 2.27 2.61 3.20 Administrative Expenses 1.143 1.151 1.408 1.617 1.700 2.016 2.172 2.243 2.339 2.656 Administrative Expenses/Outstanding Loans and Participations (M) 1.04 0.94 0.92 0.88 0.81 0.89 0.92 0.91 0.95 1.02 Dividends as 2 of Par Value Share!' 7.25 6.70 4.30 6.30 8.00 5.07 4.69 5.00 6.00 8.00 Long-term Debt/Equity RatioA 5.58/1 4.7811 4.21/1 5.07/1 5.23/1 5.4111 5.46/1 5.81/1 5.86/1 5.73/1 Source: BDET / Profits on sales of investments. * Contribution covers foreign exchange losses on interest on foreign borrowings as well as part of BDET losses on account of arrears from public enterprises. AI Share capital of TD 8 million in 1980, 10 in 1981, 15.55 in 1982, 20 in 1983-84 and 30 in 1985-89. At Equity is defined as share capital, reserves and government subsidies. Annex 20 BDET - TREND IN ARREARS 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 A. Loaa Portfolio (TD million) 76.8 88.6 99.9 112.0 139.4 168.7 194.2 210.6 227.6 239.8 247.1 266.5 B. Total Arrears -in TD million11 3.2 3.4 4.4 7.2 8.1 9.5 15.9 21.1 32.9 38.4 38.6 38.1 -as a percentage of A 4.2 3.8 4.4 6.4 5.8 5.6 8.2 10.0 14.4 16.0 15.6 14.3 C. Portfolio Affected by Arrears -in TD million 10.3 12.7 19.4 23.3 36.0 31.0 58.6 63.9 74.5 94.6 -as a percentage of A 13.4 14.3 19.4 20.1 25.8 18.4 30.2 30.3 32.7 39.4 Source: BDET 'II - Arrears over 3 months amount to about 75% of total arrears. It During negotiations of the Second EMI Proje-t (Loan 2554), in 1985, BDET had committed itself to reduce its arrears to the level of 1983, i.e. TD 7.1 million (Staff Appraisal Report, May 1, 1985). - 78 - Annex 21 BDET - FOREIGN EXCHANGE LOSSES (TD thousand) Backlog of Government Government Payments due Contributions Losses for Payments to BDET at to Interest Ratq the Year to BDET Year-end Differential' 1981 3490.3 1082.9 2407.4 1030.0 1982 3381.6 1552.9 4236.1 460.0 1983 10698.5 2220.7 12713.9 2946.0 1984 8377.8 2100.2 18991.5 5848.0 1985 11755.2 --- 30746.7 2978.0 1986 11312.4 10650.0 31409.1 3875.0 1987 15812.0 13150.0 34071.1 4540.0 1988 n.a. 19150.0 n.a. 2978.0 Source: BDET 1 This contribution only covers the additional financial costs incurred by BDET plus part of BDET losses on account of arrears from public enterprises as BDET has stopped recording the interests due on these loans. Annex 22 BDET - FORECAST AND ACTUAL FINANCIAL INDICATORS 1980 1982 1983 1984 1985 1986 1987 1988 1989 1. Return on Average Net worth (M; EKI-I Projections 9.3 7.41' n.a. 10.0 10.0 EMI-II Projections 6.1 8.1 8.8 9.0 Actual 9.0 7.0 9.0 9.2 6.6 7.0 5.3 6.7 13.0 2. Loni-term DebtfEquity Ratio BMI-I Projections 5.1 6.4 6.1 5.6 6.1 EMI-II Projections 6.2 5.4 5.6 5.7 Actual 5.6 4.2 5.1 5.2 5.4 5.5 5.8 5.9 5.7 3. Margin on Borrowings EMI-I Projections 1.0 1.1 1.6 2.0 EMI-Il Projections 1.5 1.7 2.0 2.1 Actual 1.4 2.7 2.6 1.6 2.2 2.8 2.3 2.6 3.2 4. Dividends as I of par value EMI-I Projections 8.0 8.0 8.0 8.0 EMI-II Projections 7.3 7.4 8.0 8.0 Actual 7.3 4.3 6.3 8.0 5.1 4.7 5.0 6.0 8.0 Source: BDET N Net income on share capital and reserves. I Decrease due to share capital increase from TD 8 million in 1980 to 10 in 1981, 20 in 1982-83, 25 in 1984 and 30 in 1985-86. 1 Before Government subsidy. - 80 - Annex 23 BDET - STRUCTURE OF BORROWINGS 192 1985 1987 1989 &mnMt ..har. Amuntm. Share Amount Share Amount Share (TD million) (2) (TD million) (2) (TD million) (1) (TD million) (2) - Local Resources 14.7 16.2 21.1 10.3 34.3 15.6 49.1 20.7 - Borrowings from Bilateral and Multilateral Sources 65.6 72.4 80.0 39.4 85.0 38.7 90.7 38.3 - Borrowings on the International Capital Market 10.3 11.4 101.9 5 100.3 45.7 9 40 TOTAL 90.6 100.0 203.0 100.0 219.6 100.0 237.0 100.0 Source: BDET - 81 - Annex 24 CENTRE TECHNIQUE DES INDUSTRIES MECANIQUES ET ELECTRIQUES (CETIME) (Tunisian Dinars) 1. Loan 2113-TUN a. Eight foreign experts salaries for two years (192 man/month at US$7,500 per man/month) 720,000 b. External consultancy services n.a. c. Training abroad for Tunisia counterparts n.a. TOTAL 750,00011 II. Loan 2554-TUN a. Mechanics Laboratory: - Materials composition 160,000 - Materials hardness 50,000 - Metrology 80,000 - Surface treatment 30000 Subtotal 320,00011 b. Metal tools for foundriest - Equipment purchase for the design stage 30,000 - Equipment for the fabrication 370,000 - Two foreign experts (foreign exchange cost) for two years 239.000 Subtotal 639,0001' c. Electronics Division: (i) Implements for industrial applications - Equipment (designing and testing) 150,000 - Components and accessories (stock) 25,000 (ii) Equipment for design and production of custom chips 250,000 (iii) One foreign expert for one year 75,000 Subtotal 500,0001' d. Technical Assistance: four experts for one year 239,00011 TOTAL (a+b+c+d) 1,698,00011 Sources: Staff Appraisal Report Nos. 3689-TUN and 5847-TUN 1 Equivalent to US$1.5 million. l Equivalent to US$385,500; rounded to US$400,000. A' Equivclent to US$788,000. 1 Equivalent to US$621,000. l Equivalent to US$288,000. i' Equivalent to US$2.097 million, rounded to US$2.1 million. - 82 - Annex 25 INSTITUT NATIONAL DE LA NQBMALISATION ET DE LA PROPRIETE INDUSTRIELLE (INNORFI) Tunisian Dinars US Dollars (equivalent) I. Loan 2113-TUN Three foreign experts in standardization for one year (30 man/month at about US$10,000 per man/month) 150,000 300,000 II. Loan 2554-TUN a. Metrology Section - Staff training 38,345 46,200 - Implements (scales, length/weight/ volume standards) 141.515 170.500 Subtotal 179,860 216,700 b. EMI Laboratory - Basic electric tests division 223,700 269,500 - Industrial power test division 332,300 400,400 - Machinery division 629,100 757.900 Subtotal 1,185,100 1,427,800 c. Technical Assistance - Foreign experts for 3 years (10 man/month per year over 3 years for seminars, audits and preparing manual on quality management) 205L425 247.500 TOTAL (a+b+c) 1,570,385 1,892,000 (rounded up to US$1.9 million)
World Bank Group · Project Performance Assessment Report
Tunisia - Electrical and Mechanical Industries Project
View original document
The full text is hosted by the publishing organisation. lawenc.com indexes the metadata and links to the official source.
Full text
Key facts
Organisation
World Bank Group
Document type
Project Performance Assessment Report
Country
Tunisia
Source
World Bank