Tunisia - Structural Adjustment Loan Project

Docamn of The World Bank FOR OMCLL USE ONLY MICROGRAPHICS Report No. 12135 Report No: 12135 Type: PPAR PERFORMANCE AUDIT REPORT REPUBLIC OF TUNISIA STRUCTURAL ADJUST1ENT LOMAN (SAL 1) (LOAN 2962-TUN) JUNE 30, 1993 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. Official Exchange Rate: Dinar (TD) per US$ Pvriod End of Period Period Ava 1988 0.8985 0.8578 1989 0.9046 0.9493 1990 0.8368 0.8783 1991 0.8645 0.9246 ABBREVI[IT ASAL I Agricultural Sector Adjustment Loan CGC Caisse Generale de Compensation EFF Extended Fund Faciliy EFRSL Economic and Financial Reform Support Loan EM2CO Country Operations Division of the Maghreb Department of the MENA Region FPC Fonds de Perequation de Cahgnes GDP Gross Domestic Product ITPAL Industrial and Trade Policy Adjustment Loan MNA 1 Country Division of the Middle East and North Africa Regional Office PCR Project Completion Report PER: Public Enterprise Reform Loan SAL Structural Adjustment Loan VAT Value Added Tax QRs quantitative restrictions SMIG Salaire Minimum Interprofessionnel Garanti FISCAL YEAR OF BORROWER January 1 - December 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.8A Ofib* of Director-General Operations Evaluation June 30, 1993 1U ORAIJM TO THE EXECUTIVB DIRECTORS AND T[I PRESIDENT SURTECT: Performance Audit Report on Tunisia Structural Adiustment Loan (SAL) (Lan 2962-TUN Attached is the "Performance Audit Report on Tunisia - Structural Adjustment Loan (L2962-TUN)" prepared by the Operations Evaluation Department. The Government of Tunisia commented on the PAR (Annex II). The SAL, approved in 1988, was the third of a series of adjustment operations in response to balance of payments difficulties encountered in the mid-1980s. Four areas of reform were supported: (a) trade liberalization; (b) price liberalization; (c) financial sector reforms; and (d) taxation. Macroeconomic stabilization has been achieved with a combination of tough decisions dealing with public expenditure retrenchment, exchange rate adjustment, realignment of real wages, and other domestic measures to improve exports. The core issues for the SAL dealt with deepening the liberalization of the economy, which linked it directly with the previous industrial and trade adjustment loan (ITPAL). Even though the liberalization objectives set in the SAL were achieved, progress was slower than expected. The sequencing of trade reforms was also inappropriate. Since the main program of reducing tariffs was undertaken under ITPAL, the objective of the SAL was to make progress in the removal of quantitative import restrictions and of price controls. These measures proved to be much more difficult to implement in view of domestic pressures against them. The comprehensive reforms on taxation were implemented smoothly. They covered the adoption Vf a value added tax, extended up to the wholesale stage, and a revamping of the direct tax system, covering both personal income taxation and company taxation. Overall, this operation is rated satisfactory. In spite of the limited accomplishments of the liberalization program, the SAL facilitated focus on other important structural policies on public enterprise reforms, privatization and financial policies. Program sustainability is likely, with partial impact on institutional development. Attachment This&doumnahas a restictd disriutionwan may be use bytwcipiwntsoMnl teperfornmeof thirofficia dutie. Itoontnts asy no otrwis be discosed withut Worl BawkWatorzton. FOR OFFICIAL USE ONLY PERFORMAN(E AUDIT REPORT STRUCRIAL AD,STMENT IWAN (SAL D (Loan 2962-TUN) Table of Contents Eage No PREFACE ........................ ........................ i BASIC DATA SHEET ....... ............................... iii EVALUATION SUMMARY ..... ............................. V I. BACKGROUND ....................................... 1 II. ECONOMIC REFORMS SUPPORTED BY STRUCTURAL ADJUSTMENT LOAN ..................... 2 III. IMPLEMENTATION OF REFORMS AND OUTCOME ................ 3 Macroeconomic Developments ............................... 11 Trade and Industry Liberalization: Tariffication of Protection and Elimination of Quantitative Restrictions and Price Controls .......... .12 Slowed Pace of Liberalization.................................. 12 Tariff Protection Rates .................................... 15 Competition Law, Price Controls, and Financial Issues: Relationship with Industry and Trade Liberalization .................17 Fiscal Reforms ......................................... 19 Reform of Indirect Taxation: the VAT............................19 Direct Taxaticn ........................................ 22 Controlling subsidies ............................ .......... 23 Other Issues............................................ 24 Social Aspects of Adjustment and Employment.................... 24 Disbursements. ........................................ 25 IV. ASSESSMENT AND LESSONS OF EXPERIENCE................. 26 General Assessment and Sustainability........................... 26 Role of the Bank .... ................................... 27 Lessons of Experience .................................... 28 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. PERFOlBANCE AIT REPEOI STRUCTURAL ADJUSTMENT LOAN (SAL I) (Loan 2962-TUM Table of Contents (Cont.) age No. Table 1. Matrix of Reforms Supported by SAL and Assessment of Implementation of Action Program .................... 4 Figures Figurel: MacroeconomicEvolution .............................. 11 Figure 2: Macroeconomic Ratios ................................ 11 Figure3: EliminationofQRs .................................. 14 Figure 4: Price Decontrol..................................... 14 FigureS: Removal: DistributionMargins............................... 15 Figure6: NominalProtectionRates............................... 16 Figure 7: Effective Rates of Protection - Export Industries ................. .. 16 Figure 8: Effective Rates of Protection - Domestic Market Industries .......... 17 Figure9: First Years ofVAT .................................. 21 Figure10:ConsumerSubsidies.................................. 23 Figurell:RealWages, 1973; 1980-91 ............................. 25 PERFORMANCE AUDIT REPORT TUNISIA STRUCTURAL AD,USTMENT LOAN (SAI 1I (Loan 2962-TUN) Table of Contents (Cont.) PE No. Table 1: Macroeconomic Evolution During the VIITH Plan Period SALForecastsandRealizations ........................... 30 Table 2: Export Earnings in Millions Dinars at 1990 Prices ................ . 31 Table 3: Real Return to Labor Outside Agriculture ..................... .. 31 Table4: Macroeconomic Indicators............................... 32 Table 5: Trade and Industry Liberalization: QRs and Price Decontrol ......... 33 Table 6: Indicators of Nominal Protection Rates, Disaggregated by Source of Effects 34 Table 7: Effective Protection Rates for Domestic and Export Industries ......... 35 Table 8: Value Added Tax and Other Consumption Taxes Since VAT Introduction in 1988 ........................... 36 Table 9: Value Added Tax: Composition of Taxpayers .................. 37 Table 10: Expenditures of the Caisse Generale Centrale .................. 37 ANNEXII Letter from the Government of Tunisia .. .......................... .. 38 i PERFORMANCE AUDIT REPORT TUMISIA STRUCTURAL ADJUSTMENT LOAN (SAL D (Loan 2962-TUN) PREFACE This is a Performance Audit Report (PAR) on the Structural Adjustment Loan SAL, Loan 2962-TUN) involving a loan of US$150 million. The loan was approved on June 15, 1988 and became effective on March 31, 1989. It was fully disbursed and closed on June 30, 1991, after an extension. Disbursement was in two tranches, and the second tranche was approved for release in January, 1991. The PAR was prepared by the Operations Evaluation Department (OED). It was based on the President's Report, sector and econor * reports, special studies, the loan documents, study of the program files, the Program Completion Report (PCR), and discussions with Bank staff. An OED mission visited Tunisia in January 1993 and discussed the effectiveness of the Bank's assistance with Government officials, representatives of the business community, and the financial community. Their kind cooperation and invaluable assistance in the preparation of this report is gratefully acknowledged. The PAR discusses briefly the background of adjustment in Tunisia; the various reform measures supported by the SAL; the implementation of the reforms and their outcomes; assessment of the implementation of the action program; and the lessons of experience. The PCR made a keen analysis of the macroeconomic situation during the adjustment period but left a gap in the discu!sion of the implementation problems. This PAR contains a detailed assessment of the implementation of the adjustment program, and delineates the policy design issues which were critical for the ultimate success of the trade liberalization program. It also analyses the comprehensive tax reforms undertaken under the SAL. ECRFOMAM A=DI REOR T-UNISA SMEUCTURA ADISTBM LWAN (JAL 1) (LWAN M%-UMI BASIC DATA SHEET LOADA Loan Position Amounts (US$ Millionsl As of March 31, 1993 Original Disbursed Cancelled Repaid Outstanding 150.0 150.0 - - 150.0 Project Dates Original Loan Date Actual cr Re-estimated Initiating memorandum 2/10/88 2/10/88 Letter of Dev. Policy 5/9/88 5/9/88 Negotiations 515188 5/5/88 Board Approval 6/15/88 6/15/88 Loan Agreement 9/26/88 9/26/88 Effectiveness 1/26/89 3/31/89 Loan Closing 4/30/90 12/31/90 & then 6/30/91 Actual Completion 6/30/92 CUMULATIVE ESTIMATED AND ACTUAL LOAN DISBURSEMENT FY90 FY92 (i) Planned 150.0 150.0 (ii) Actual 93.3 150.0 (iii) (ii) as % of (i) 62.2% 100.0 iv MISSION DATA Mission Date of Month/Year No. of Weeks No. of Personw Staff Weeks Report Preparation I 4/22/87-5/7/87 2.5 3 7.5 6/19/87 Preparation II 11/7//87-11/20/87 2.5 9 22.5 1/27/88 Appraisal 2/10/88-2/27/88 2.5 9 22.5 3/15/88 Supervision I 2/16/89 1.0 4 4.0 3/30/89 Supervision II 10/27/89 1.0 6 6.0 11/2/89 Supervision 111 6110/90 1.0 5 5.0 9/8/90 Supervision IV 12/10/90 2.0 2 4.0 1/3/91 Completion 6/7/92 1.0 2 2.0 N/A STAFF INPUTS Fiscal Yes Preappraisal Appraisal Negotiation Supervision Other Total 1987 17.4 17.4 1988 117.0 62.2 20.9 2.3 202.4 1989 16.2 16.4 0.1 1990 29.6 29.9 0.3 1991 22.7 22.7 1992 7.8 7.8 1993 1.4 1.4 TOTAL 134.4 62.2 20.9 77.7 2.8 297.9 FORM 590 RATINGS Evaluation Development Legal Management Year Overall Objectives Covenants Performance 1989 2 2 2 2 1990 2 2 2 2 1991 1 1 1 1 1992 1 1 1 1 V PERFORMANCE AUDIT REPORT TUNISIA STRUCTURAL ADJUSTMENT LOAN (SAL 1) EVALUATION SUMMARY 1. The Structural Adjustment growth of the nominal wage below the Loan (SAL), approved in 1988, is th third of inflation rate. a series of adjustment operations supported by the Bank, following balance of payments 3. The SAL supported difficulties experienced in Tunisia in the mid- macroeconomic stabilization, as did the first 1980s. The first two adjustment operations -- two adjustment operations. Four areas of Agricultural Sector Adjustment Loan (ASAL, structural economic reforms specifically approved in September 1986) and Industry assisted by SAL were: (a) foreign trade and Trade Policy Adjustment Loan (ITPAL, liberalization; (b) price liberalization; (c) approved in February 1987) -- supported financial sector; and (d) taxation. sector reforms. Each of these three quick disbursing loans made available $150 million, 4. The core issues of the structaral for a total of $450 million. reforms in the SAL centered on the liberalization of trade and industry. Foreign 2. Tunisia had strong GDP growth trade liberalization was focussed on the in the 1970s, averaging 7% per year. This elimination of quantitative restrictions, and on was propelled by oil prosperity and by high further improvement of tariff reform already investment. But this growth could not be begun under TAL. This also included the sustained in the 1980s. Oil prices declined liberalization of price and distribution and domestic oil production was expected to controls, the promotion of domestic be limited by exhaustion of reserves. The competition, and the reduction of consumer Government recognized these problems in the subsidies. Financial sector measures included Sixth Plan (1982-86). Faced with these the institutionalization of a money market, the economic difficulties, a program of reduction of some interest subsidies, the macroeconomic stabilization was undertaken. adoption of a foreign exchange risk Adjustment required reduction of aggregate mechanism sought to transfer risk taking to demand: cutting government expenditure and domestic enterprises from state institutions, reduction of investment, accompanied by a and the introduction of measures designed to depreciation of the dinar and a tightening of improve the financial position and import restrictions, and a freeze of minimum competitiveness of Banks. wages for two years and containing the A VI 5. The reforms under the SAL value being liberalized, was not easy to were, on balance, satisfactorily achieved. monitor. The components of production The macroeconomic stabilization was value depended on the relative weights of undertaken with a combination of tough production among industries in the sectors. decisions taken to restore external and internal Ther were differences between the balance. Public expenditures were held back Government and the Bank with respect to the in relation to budgetary resources, and the levels of disaggregUons of imported public sector deficit was contained under 4% commodities and their industrial classification. of GDP. Strong measures were undertaken to Final agreement on the percentage attained realign real wages, by initially freezing wages within the liberalization process with respect for two years and later on allowing nominal to second tranche release took a longer wage increases in line with productivity amount of time than needed. A more explicit growth. This enabled the Government to hcild agreement right at the outset would have been down the public payroll and thereby the preferable. control on the budget deficit. The real exchange rate was devalued by around 30% 8. The actual effeivenem for ;e between 1980 and 1990. These measures loan was delayed by about six months from made exports and tourism more attractive. the original date because of delays in the overall implementation of the liberalization 6. The original program of program. The slippage was caused by liberalization, however, was slowed down by opposition within the Government (as a result a number of factors. The foremost of these of industry pressure) principally against the was that the Government did not accept the removal of quantitative restrictions. In spite pace of liberalization as determined during of this, shortly before the second tranche negotiations. The industries threatened with release (originally planned for September loss of traditional protection had powerful 1989 but finally released only January 1991), channels within the Government. Just as the targeted liberalization ratios for important, there was a design fault in the quantitative restrictions, for price decontrol, sequencing of the import liberalization and for decontrol of margin distribution were aspects. By undertaking to reduce tariff rates reached by the Government. ahead of the quantitative restrictions in an earlier adjustment loan, the opportunity was 9. Two points should be noted lost in removing the quantitative restrictions about the stage a which economic more quickly and also, as a consequence, the liberalization in Tunisia presently stands. elimination of rents derived from protection. First, the adoption of steps towards the This would have temporarily put the burden liberalization targets were seriously delayed (a of protection on tariffs, which also implied a total of 14 months from the planned date of revenue gain for the Government. effectiveness). More significant, the libralzatonobjectives were modest. At the 7. Another design weakness was closing of the SAL, the Tunisian economy the methodology selected for monitoring the had achieved only a modest goal in the progress of liberalization. The formula, while liberalization of the trade and industry conceptually attractive, in that it announced regimes. After three adjustment operations the proportion of the domestic production that focussed on import and trade Vii liberalization, a substantial amount of commodities such as petroleum and a few quantitative restrictions on imports and price luxuries. The VAT was successfully put in controls remained, place without raising the inflation rate, which has remained moderate. 10. A major objective of the SAL was to encoura- the growai of competition 13. Tax reform also covered in domestic commerce. A law was passed to thorough revamping of the direct tax system, promote competition, and implementation of i.e. personal income taxation and company the provisions is beginning to take place. The taxation. In general, the rates of taxes were present emphasis of -he implementation of the brought down, the number of marginal rates competition law, however, is on protection of reduced, and single taxation of dividend consumer interest and on the price controls income. that are in effect. So long as trade restrictions and price controls are present in 14. In spite of the limited the magnitude that they exist in the Tunisian accomplishments of the liberalization economy, the prorotion of competition is program, SAL facilitated the initial focus on likely to be restrained. other structural policies, especially enterprise reforms, privatization, and financial policies. 11. Fiscal reforms supported by the Without Bank support, the liberalization SAL included a comprehensive reform of the program would probably be in an even more tax system, both the direct and indirect taxes. rudimentary stage. Given a highly In addition, the control and rationalization of regimented economy, Bank assistance was subsidies improved the allocation of fiscal instrumental in focussing the Government's revenues and helped to reduce the fiscal early attention to this issue. If appropriate deficit. sequencing had been pursued and the monitoring system had been designed better, 12. Tax reform was implemented the degree of liberalization might have been very satisfactorily. The reformed system of higher. taxation was designed to be simpler, more transparent, and economically efficient. The Lessons of changes covered both direct and indirect taxation. The adoption of a value-added tax 15. The implementation of SAL (VAT) was successfully put in place under reveals important issues of design and SAL. The VAT was applied initially to the adjustment strategy. The major lessons of production level and was extended to the experience are discussed below: wholesale stage by the time of the second tranche release. The extension to the retail 16. Political COnm of refrm. stage has not yet been adopted. Adoption of Difficult structural reforms can be carried out the VAT proceeded relatively smoothly, and speedily only if sufficient national consensus revenues from it have risen. The temporary in support of them is present. This is the consumer tax on luxuries was adopted as a main lesson from the slowed trade and parallel measure to prevent a fall in revenues. industry liberalization Program. The This has now been regularized as a tax Government was unable to implement its covering only a limited number of essential commitment to undertake the programmed Viii liberalization measures, because there was not 20. Impo ce of monitorabift sufficient natifnal consensus to back it up. prges in liberalixWin. A methodology The technocratic team which promoted the that is not easy to monitor induces friction agenda of economic liberalization was not during economic policy dialogue, because of successful in conwncing industry and the rest lack of understanding. An example was the of Government of the benefits to be derived methodology designed to determine progress from liberalization. of the liberalization program. Although conceptually attractive, its complexity 17. In connection with the hindered its use. The use of a positive list of transitional costs of liberalization, industries commodities under quantitative restrictions - whih are grossly inefficient would not to be reduced progressively in stages -- would survive competition. But there are industries hve been a more practical form of that could survive the change in regime. The conditionality. failure to identify those industries and to provide them with some temporary support to 21. Price liberalization c proceed survive the changing situation strengthened more effmcively under a r prom resistance against reform and made it more internal co m . The price liberalization difficult to win support from industry. process, as negotiated, was modest. A large degree of price controls remained within the 18. Sequencing the liberalization economy. Because economic agents operated process: tariffication of import protection is under a "dirigiste" regime of government made moredifficult if quantitative restrictions administered pricing policies, competitive on imports are not removed ahead of tari pricing was institutionally hampered. So long rate restructuring. The rer val of as price controls remain a feature of domestic quantitative restrictions should be undertaken commerce, the implementing machinery for ahead of tariff reduction. Doing it the other promoting competition will remain hampered way -- reduce tariffs ahead of eliminating by price regulatory pressures. quantitative restrictions - was the approach attempted in Tunisia. This sequence of policy 22. Tax reform. The experience change meant that the potential gains in tariffs of Tunisia in instituting two major sets of tax from protection were lost by Government (to changes within a reasonably short period of the extent that the protective rate exceeded the time indicates that, with adequate preparation new, lowered tariff rate). and careful study, it is possible to underAe a comprehensive reform of the tax system. 19. At the same time, the rents The preparatory work prior to the shift to from quantitative restrictions were retained VAT was significant in making VAT accepted and, in turn, strengthened those whose by the general public. The presence of a interests were to oppose the liberalization countervailing set of temporary consumption program. This had the consequence of taxes, within the tax structure during the shift further strengthening the position of those to VAT, was significant in avoiding revenue holding quotas and had the indirect effect of losses. strengthening the already imbedded resistance to reform coming from quota holders. ix 23. Imortance of monitorability administered pricing policies, competitive progress in liberalization. A methodology pricing was institutionally hampered. So long that is not easy to monitor induces friction as price controls remain a feature of domestic during economic policy dialogue, because of commerce, the implementing machinery for lack of understanding. An example was the promoting competition will remain hampered methodology designed to determine progress by price regulatory pressures. of the liberalization program. Although conceptually attractive, its complexity 25. Tairfom The experience hindered its use. The use of a positive list of of Tunisia in instituting two major sets of tax commodities under quantitative restrictions -- changes within a reasonably short period of to be reduced progressively in stages - would time indicates that, with adequate preparation hve been a more practical form of and careful study, it is possible to undertake conditionality. a comprehensive reform of the tax system. The preparatory work prior to the shift to 24. Price liberalization can pi ;ed VAT was significant in making VAT accepted more effectively under a regime promoting by the general public. The presence of a internal competition. The price liberalization countervailing set of temporary consumption process, as negotiated, was modest. * A large taxes, within the tax structure during the shift degree of price controls remained within the to VAT, was significant in avoiding revenue economy. Because economic agents operated losses. under a "dirigiste" regime of government PERFORIMAINCE AUDITf REPORT STRUCTURAL ADJUSMN LOAN (SAL B (LAnm 2962-TUN) I. BACKQROTID 1. The Structural Adjustment Loan (SAL), approved in 1988, is the third of a series of adjustment operations supported by the Bank, following baldnce of payments difficulties experienced in Tunisia in the mid-1980s. The first two adjustment operations -- Agricultural Sector Adjustment Loan (ASAL, approved in September 1986) and Industry and Trade Policy Adjustment Loan (ITPAL, approved in February 1987) -- supported sector reforms. Each of these three quick disbursing loans made available $150 million, for a total of $450 million. The Government's adjustment efforts were also supported by the IMF in the form of a Stand-by arrangement that ended in May 1988 and an Extended Fund Facility, which was approved in July 1988. In addition, cofinancing by bilateral donors, notably by Japan, provided additional financial assistance. 2. The Project Completion Report (PCR) provides a keen analysis of the macroeconomic developments in Tunisia during the adjustment period covered by SAL. Moreover, the recently finished Performance Audit Reports (PARs) for ASAL and ITPAL have also discussed the background and the developments leading to the adoption of the adjustment program. A brief summary of these developments is made below. 3. The strong growth of GDP exceeding 7% per year in the 1970s, which was propelled by oil prosperity and high investment, fell to about one-half of that rate of growth during the early 1980s. The growth of the previous period could not be sustained. Oil prices had declined and domestic oil production was in decline due to limited reserves. Investment had been highly capital intensive, due to ambitious public sector investments and low returns. Wages (maidated by ever rising minimum wages) rose faster than productivity, and the public sector budget deficit had been rising. But the signs of an unsustainable path were strong, with I Project Completion Report; Tunisia: Structural Adjustment Loan (SAL I), Loan 2962-TUN, Report No. 11354, November 16, 1992, Country Operations Division, Country Department I, Middle East and North Africa Regional Office. 2 Performance Audit Report. Tunisia: Agricultural Sector Adiustment Loan (Loan 2754-Tun), 1992 and Performance Audit Report. Tunisia: Industry and Trade Policy Adiustment Loan (Loan 2781-Tun), 1992, World Bank, Operations Evaluation Department. 2 the debt service ratio rising to 21%. The trade deficit had grown to 11% of GDP and the fiscal deficit had reached almost 7% of GDP. 4. The Government recognized these problems and articulated them in their medium term economic plan, the Sixth Plan (1982-86). A program of macroeconomic stabilization was implemented reducing aggregate demand. Government expenditure and investment were cut. This was accompanied by depreciation of the dinar, tightening of import restrictions, and a freeze of minimum wages for two years. The growth of the nominal wage was held below the inflation rate thereafter. Adjustment was complicated by the drought in 1986, which led to poor crop harvests, and by the further collapse of oil prices. 5. The Bank responded to the Government's request for adjustment assistance by undertaking initially the sector loans, first ASAL and then ITPAL. Even though the first loans were mostly designed to deal with structural measures in the sectors of agriculture and, later, trade and industry, the quick disbursing assistance made available were designed to relieve the balance of payments pressures at a time when major macroeconomic stabilization was taking place. Even though IMF support was made at about the same time, the Bank's role in restoring macroeconomic balance in Tunisia was critical and timely. H1. ECONOMIC REFORMS SUPPORTED BY STRUCTURAL ADJUSTMENT LOAN 6. The SAL supported macroeconomic stabilization, as did the first two adjustment operations. Four areas of structural economic reforms were specifically assisted: (a) foreign trade liberalization; (b) price liberalization; (c) financial sector reform; and (d) taxation. The specific elements of these reform programs are fully summarized in the presentation of Table 1. This table sketches all the components of the policy actions supported by the SAL, providing a background of earlier accomplishments, reform objectives, and the specific programs of actions agreed upon. This table will be referred to again in connection with the discussion of implementation. 7. To attain macroeconomic stability with growth, the Government would undertake the appropriate policies with respect to the wage rate, exchange rate, and the budget, which would restore internal and external equilibrium. In practical terms and in relation to the problems of the past, this meant the reduction of aggregate demand. 8. The core issues of the SAL reforms, however, centered on the liberalization of the Tunisian economy. The foreign trade liberalization highlighted the removal of quantitative restrictions and further improvement of triff ieform already begun under ITPAL. Price liberalization issues were related to the dismantling of price and distribution controls, the promotion of domestic competition, and the reduction of consumer subsidies in favor of market 3 liberalization. Financial sector measures included the institutionalization of a money market, liberalization of interest rates, the removal of interest subsidies, the adoption of a foreign exchange risk mechanism which increased entrepreneurial risk taking, and the introduction of measures designed to improve the financial position and competitivity of Banks. 9. All these reforms represented a slice of the overall adjustment process initiated in 1986 with the adoption of ASAL I. Some of the measures that were dealt with in SAL were further continued in the succeeding adjustment operations dealing with reforms in public enterprises and in the financial sector. Ill, IMPLEMENTATlON OF REFORMS AND OUTCOME 10. The reforms under the SAL were, on balance, satisfactorily achieved. The macroeconomic stabilization was undertaken with a combination of tough decisions taken to restore external and internal balance. Public expenditures were held back in relation to budgetary resources, and the public sector deficit was reduced to under 4% of GDP. Strong measures were undertaken to align real wages to economic reality, by initially freezing wages for two years and later on allowing nominal wage increases with productivity growth. This enabled also the Government to hold down the public payroll and thereby the control on the budget deficit. The real exchange rate had been devalued by around 30% between 1980 and 1990.3 Coupled with good luck such as tourism (which was sensitive to political developments at the Libya frontier) and good weather, these measures helped to improve export receipts and, hence, the external balance. They further enabled Tunisia to overcome the adverse external shocks that accompanied the 1991 Gulf crisis. 11. With respect to structural reforms, the greatest degree of difficulty was encountered in connection with the liberalization of trade and industry. The issues giving rise to these difficulties are discussed in detail later. They focus on core issues of process and design of economic reforms. Under the section "Assessment of Implementation", of Table 1, column 3, a brief description of the implementation of the action program is summarized. At a glance, this table provides a convenient representation of the achievements and/ or difficulties encountered during the implementation phase of undertaking the reforms, which were contained in the policy reform matrix. The nominal depreciation of the dinar between 1985 and 1986 of 25% was large by any measure. Yet, contrary to experience in many countries, the inflationary effects were contained by the complementary measures undertaken by the authorities to reduce aggregate demand. Table 1. MATRIX OF REFORMS SUPPORTED BY SAL AND ASSESSMENT OF IMPLEMENTATION OF ACTION PROGRAM Polley Objective, including achievements prior to SAL Activity Supported by the SAL Conditionality, Action & As ment of 1. Macroeconomic stability Tax, credit, income and exchange Satisfactory review of macroeconomic performance required for second tranche release. with growth rate policies should continue to These reviews will focus on the attainment of three objectives: growth, external equilibrium be kept in check. Budget deficit with emphasis on reducing the amount of debt and domestic equilibrium (employment and to be cut to 2.2% of GDP or less inflation) through: wage policy, exchange rate policy, budgetary policy. Limitation on by 1991. transfers from the state to public enterprises for operating subsidies and for restructuring. Economic justification for all major public sector projects. Control of the growth of real Assessment of Implementation: wages in the public sector. The measures taken to establish macroeconomic balance were satisfactory. The actual results of macroeconomic evolution to SAL forecasts were close, and on the whole better than expected. Ensure that sufficient credit is available for the private sector while controlling the growth of the money supply and of prices. 2. Foreign trade reform Preparation of a program to At least a 15 points reduction (relative to June 30, 1988) in the weighting of restricted achieve the objectives in stages, imports, measured in terms of production before second tranche release. Quantitive Restrictions in step with the decontrol of Elimination of all QRs on prices. Asment of Implementation: all imports, except for a Original timetable for the liberalization of quantitative restrictions was stretched resulting in limited number of goods. delay of effectiveness and of second tranche release. Hence the target dates for liberalization Progress before SAL: had been lengthened. In terms of overall delays, from planned effectiveness to actual 9]s- QRs eliminated for (a) effectiveness and the date of second tranche release was a slippage of close to one year and a most raw materials and semi- half. At effectiveness, freely importable goods corresponded to 6% of domestic production. manaufactures except for The objective of reducing at least at least 21% of production (meaning an additional 15% poorly integrated industries; increase from 6%) was not realized at original timetable. At second tranche release (in Jan. (b) capital goods, except for 1991), this target was achieved. To achieve the target of elimination of QRs, the Bank infant and poorly integrated agreed to the imposition of temporary compensatory duties ranging from 10% to 30% of industries; (c) semi- value. Nevertheless, the removal of scheduled QRs is moving forward, although much more manufactures and raw slowly. materials for eqterprises exporting more than 15% of their output. Policy Objective, including achievements prior to SAL Activity Supported by the SAL Conditionality, Actions, & Assessment of Implementation Customs Tariffs Preparation of a program to (Program monitored by Bank as Maintenance of a maximum achieve the objectives for 1991. part of the semi-annual reviews) nominal rate of 35% and a minimum nominal rate of Assessment of Implementation: 15% with average effective The program of improving the tariff regimc has continued and has achieved basically the rate of protection of 25% by goals began under ITPAL. There is no direct conditionality on tariffs, but the monitoring of 1991. the developments along this front was considered important in achieving the program of tariffication of protection. However, so long as QRs (and additional trade taxation cum (a) Integration of the 5% compensatory taxes) are in place and are significantly high, the tariff regime fails to become customs formality tax (TFD) a transparent measure of protection. in the tariff structure; (b) Reduction in the maximum rate from 236% to 41% and rationalization of the tariff structure. Safeguard Measures A study on transparent anti- Employment of consultants before effectiveness. Review of the study and implementation of dumping measures. its conclusions before second tranche release. Assessment of Implementation: Study on safeguards was delayed and caused slippage in loan effectiveness. This issue attracted considerable discussion, in connection with the use of "reference prices' as basis for determining harmful pricing. Safeguards were potentially a weapon for raiskg nontariff protection. The passage of a law on competition has helped provide a basis for this. But Tunisia's accession to GATT in 1990 made the anti-dumping issue fall under GATT rules. Policy Objective, including achievenents prior to SAL Activity Supported by the SAL Conditionality, Actions, & Assessment of Implanentation PRICE LIBERAUZATION Progressive reduction in price The percentage of fully decontrolled goods to be raised to 70% by second tranche release. Ed Pic controls until totally decontrolled See pam. 53 Elimination of all controls prices represent between 75% by 1991, except for and 80% in terms of production Asesment of Implementation: subsidized products. Eqrgai in 1991. Like the issue of QRs, this phase of the program suffered slippage. At effectiveness, the before SAL Simplification of Only two categories of goods benchmark for decontrolled goods was 60% of locally produced goods were subject to price price control system (by (fully decontrolled and controlls. At original date of second tranche release, a proposal to achieve the target was merging homologation and controlled) to be used for price short of Bank expectations. But the proportion achieved by the actual second tranche release autohomologation with libert control purposes by 1991. date (Jan.1991), the price liberalization exceeded the 70% target, exceeded the target. controlde). The percentage However, this included commodities in the list which were subject to monopoly enteqises of products whose prices (sulphuric and phosphoric acid) and over which liberalization was meaningless. have been totally decontrolled is now 60%. IMPROVEMENT OF Preparation of a law on domestic A draft law agreed with the Bank will be presented for Parliament's approval kfore second DOMESTIC COMPETITION competition at the production and tranche release. Ensure the efficient distribution stages. operation of market Assessment of Implementation: mechanisms. The preparation of draft law took center stage in the debate between those seeking further protection and those seeking to to regulate competition. A draft law prohibiting noncompetitive pricing activities was presented to the Chamber of Deputies in November 1990. This became law in July 1992, hence action on this law took much time. This law also created a Commission on Competition and provided for legislation covering competition and consumer rights. Timing of tis draft law was delayed together with other actions related to se:ond tranche release conditionality, but the condition was finally met even though there were considerable slippages. DISTRIBUTION MARGINS Progressive reduction in price Margins on goods representing at least 10% of production would be liberalized at the time of Elimination of 50% of controls, so that by 1991 fully presentation to the Board, and at least an additional 10% fPr the second tranche. controls by 1991, except for decontrolled products will subsidized products. represent more than 50% of Assessment of Implementation: production. The liberalization of controls on margins proceeded at the same pace as QR liberalization. By the time of the second tranche release in 1991, goods which were not subject to controls on margins had been estimated at 30%, but this included commodities which were subject to monopoly (sulphuric and phosphoric acide). Goods not covered by margin controls was 22.7% of production. The targets on margins were met, therefore. Policy Objective, including achievements prior to SAL Activity Supported by the SAL Conditionality, Actions, & Assessment of Implementation CONSUMER SUBSIDIES Reduction of subsidies with the (a) Before second tranbe release, the Bank would review performance in 1988 and the Limit the cumulative smallest redistributive effects. measures contemplated to achieve an agreed ceiling in 1989- and volume of subsidies during the period 1987-91 to (b) Semi-annal reviews will take place with Bank staff of progress towards objectives. between D855 million and D955 million. Assessment of ImVfematatim: Before SAL, reduction in Ceilings on subsidies in 1989 came to the level of the fiscal deficit of 3.8% of GDP. This the share of subsidies in GDP wan considered satisfactory by the Bank. That year was one of drought and high import from 3.5% to less than 3%, prices for foods. The overall level of subsidies had been put under control; 1989 level of in part by raising prices. subsidies were the highest during a five-year period beginning 1987. The prospects of achieving a ceiling of D190 million in 1988 were reviewed by the Bank and found satisfactory.(a) BeforesecondtrancerelesethBankwouldeviewerforancei198 measrescontmplted o ahiev anagred cilin in1989 an Policy Objective, including achievements prior to SAL Activity Supported by the SAL Conditionality, Actions, & Assessment of Implementation IV. ALLOCATION OF (a) The effective coat of (a) Circular announcing the introduction of the measures as of ]an. 1, 1989 to be issued FINANCIAL RESOURCES preferential loans to be raised to before loan effectiveness. Market determined interest (b) Circular announcing the introduction of the measure as ot Sept. 1, 1989 to be p'iuished rates prior to release of the second tranche. Before SAL. (b) One third reduction in the gap between the effective cost of (c) Consultants would be hired before effectivemes. Review would take place prior to (a) Money market: 7% and the money market rate. release of second tranche. Reorganize money market for the free exchange of liquidity (d) Prior to release of second tranche. for treasury bonds or (c) Bank gpreads- Review of certificates of deposit. interbank competition to evaluate Assessment of IMlementation: the possibility of raising or (a) Met; by an. 1989, preferential rates had risen to 7%. (b) Substantially met: Even (b) Interest rate strMture: eliminating the cap ont the spread. though the evolution of interest rate differentials between money market rate and preferential Except for deposits for less interest rates had varied over time, the objectice reducing the differential hod been atained. than D500,000 with less than (c) Efforts to improve competition in the banking sector for deposits by developing the 3 months tcn and priority (d) Taxation system on treasury money market through the development of certificates of deposits and commercial paper. activities, controls on lending bonds will be rationalized and the Problem is one of longer term development. The study (conducted by IMF) recommended rates has been replaced by a yield of the bonds would be lifting of cap on spreads between money market rates and lending rate. Substantially met. maximum 3 points avove the increased to compensate for the (d) Met: Tax advantage to bons d'gQWp-men removed through the new tax on income. money market rate. elimination of their fiscal advantages. (c) Preferential rates: Limited to agriculture, exports and small/medium enterprises, and to artisanal and craft industries. (d) Treasury bonds. Interest rate raised from 5.5% to 6.5%. Policy Objective, including achievements prior to SAL Activity Supported by the SAL Conditionality, Actions, & Assessment of Impleentation Foreign Exchange Risk (a) Adoption of a new foreign (a) By August 15, 1988. exchange system based on market Expose entrepreneurs to mechanism. events on international currency markets and the (b) Review and, if necessary, (b) Prior to release of the second tranche. impact of fluctuations on revision in agreement with the exchange rates. Bank, of the operation of the new Assessment of Implementation: Progress: exchange risk system. (a) Met: The Circular was issued December 18, 1987. (b) Generally, the experience is not Preliminary version of the fully satisfactory, but shortcomings of the system are related to the inadequate experience and study agreed to in ITPAL the absense of appropriate instruments, e.g., a sufficient indicator of the true cost of funds. reviewed by the Bank. Reliance on the use of the money market rate does not indicate a long-term reference rate. Financial System Elimination of ex.ante approval Undertaking measures, satisfactory to the Bank, to increase competition within the banking for loans and introduction of new sector, including the increase and eventual elimination of the ceilings on bank spreads over Improve efficiency, procedures for classifying loans, the money market rate. promote competition, and aciounting treatment of accrued ensure better credit interest, procedures for Assessment of Implementation: managemen. provisions for bad debts and This is basically a developmental issue that requires much more time to assess. But the losses as well as risk instruments for increasing competition within the banking system are being introduced, e.g., concentration and capital the development of certificates of deposits and commercial paper. One foreign bank branch requirements as a function of had been allowed to operate. The cap on interest rate spread, recommended by the study, risk. has not yet been implemented, but its removal prospectively is anticipated by the central bank. Substantially met. The first foreign bank would be opened shortly to operate in Tunis. Policy Objective, including achievements prior to SAL Activity Supported by the SAL Conditionality, Actions, & Assessment of laplementation V. TAX REFORM Introduction of law providing Introduction of VAT before loan effectiveess. for: (a) Replacement of indirect Indirect Taxes: taxes by VAT (with three rates); Extension of VAT coverage to wholesale ta4e with a turnover of over D500,000 per year introduction of VAT and (b) As a transitional measure, before second tranche release. improvements in the system supplementary consumption taxes of indirect taxation. (at three rates) for certain goods Study on extending VAT coverage to retail trade would be started before loan effectiveness. taxed at the highest VAT rate; Its results would be discussed and a program for their implementation would be agreed with Final draft of law providing (c) Permanent excise duties only Bank prior to second tranche release. for replacement of the three on special items (luxury goods, turnover taxes by VAT as of tobacco, alcohol, autonobiles, Timetable for implementation of the reduction of oconsumption taxes to be agreed with July 1. 1988 presented to fuels); (d) Extension of VAT to Bank prior to second tranche release. Chamber of Deputies for wholesale trade and possibly to approval in May 1988. retail trade. Submission of bill to the Chamber of Deputies befort, loan effectiveness. Preparation of a public Entry into force of new law on income earned in 1988 prior to second tranche release. Direct Taxes: information program and training of tax officials. Assessmet of Implementation: Tax on individuals: The tax reforms were implemented with minor revisions. In the case of the VAT, coverage Replacement of schedular is only up to wholesale. Extension to retail was not undertaken. The direct taxation reforms taxes by a single tax. on individual and company income was more thorough than had been originally planned. These are discussed in greater detail in the text. Preparation of final legal texts Tax on legal entities and implementation of a new, (company tax): Replacement single law covering taxes on of corporation tax and individuals an on legal entities. business license tax by a single tax. 11 12. The discussion of this Performance Audit Report will deal more fully with issues of reforms in trade and industry, after summarizing the macroeconomic developments in Tunisia. Trade and industry reforms included the following: policy changes affecting tariff rates, removal of quantitative restrictions, decontrol of prices, and promotion of commercial competition. These issues represented the core structural reforms of the SAL. The Audit also pays particular attention to the tax reforms undertaken in connection with SAL. Such Fig 1 MACROECONOMIC EVOLUTION reforms suggest that it is possible to . institute comprehensive tax reforms within a significantly brief period of time. Macroeconomic Developments 13. The evolution of the macroeconomic framework since 1986 is summarized by referring to Figs. 1 and 2. Both figures provide a comparative view of performance as against targets in the Government's medium term economic plan. Comparing the forecasts made by the Bank during the period of SAL preparation with the actual performance of the economy ouring the Seventh Tunisian Plan period, an improvement in Fig. 2. MACROECONOMIC RATIOS the overall macroeconomic performance is evident. The actual results had been better than the base case scenario projected in the SAL. GDP had performed better as a result of a better outcome for the energy sector and for agriculture. At the same time, consumption was slightly below forecast, but investment was higher than expected. And a higher growth of exports outpaced the overshooting of the import projections. The percentage point g-m differences from the 1986 base year ratios to GDP also reveal these patterns. The ratios of certain indicators of performance did not worsen during the period of &djustment as expected. For instance, reductions in the investment to GDP ratio did not get realized. With respect to fiscal and trade deficits, the performance was better than+ expected. The debt service position also improved, partly as a result of improved performance in the trade sector. 14. Improvement in the trade balance was brought about by strong measures involving the exchange rate and real wages, which made exports more competitive. By taking strong actions holding down the rise of wages, by cutting overall expenditure, including investment and public spending, the Government was able to deal with the fiscal deficit and to hold back inflation, which on an annual basis was around 7%. 12 Trade and Industry Liberalization: Tariffication of Protection and Elimination of Ouantitative Restrictions and Price Controls 15. The reforms concerning the liberalization of trade and industry in Tunisia formed the core of the issues supported by the SAL. This was not a one-time reform, but a continuing and intensifying effort. The initial studies and measures began with the ASAL and deepened with the ITPAL. Major reforms were undertaken under the ITPAL, notably tariff rate reductions. But this is also the area in which program slippage had occurred. The issues are deepseated and relate to the difficulty of promoting economic policy change in an environment that has been historically used to a different policy regime. In most countries, as in Tunisia, industry had been promoted not only by state sponsorship but with a heavy dose of protection. The intents of reformists in the Government cannot be easily translated into positive actions if the reformists fail to develop a national consensus on the appropriate reform to be undertaken. Slowed Pace of Liberalization 16. The difficulties that were encountered during the liberalization process were due to three factors. First, the technocratic team which maneuvered the agenda for liberalization did not have the full mandate to proceed with the program. The industries threatened with loss of traditional protection had powerful channels within Government. Faced with opposition, the reformers were unwilling to push for their liberalization. Also, they failed to sponsor their program of reform within a package of benefits and costs to help foster political consensus on liberalization. The reform program was perceived as a package of painful measures by its opponents. If efficiency and need to adjust are used as the principal arguments for reform, it would be difficult to persuade those who have for a long time benefitted from protected markets to change their position. There are transitional costs to liberalization, and even though for the most part, industries which are grossly inefficient would die a natural death when exposed to competition, there are industries that could survive the change in regime. The task was partly to identify those industries and to provide them with some temporary support to survive the changing sit.t.nion. There was a failure to instill this framework within the political sphere.' 17. Second, the design of the overall structural adjustment program had a wrong sequencing. The basic strategy to undertake the relatively easy part first of reducing tariff rates commenced with the ITPAL, and the more difficult task of eliminating quantitative import restrictions was assigned to the SAL. Such a strategy was fraught with risks, for it would strengthen the position of those who were against the liberalization process by making them hold on to import privileges. It would have been more easy to remove the quantitative restrictions, if tariff rates were seen as creating a protective wall. But once tariffs were reduced, the protection against imports were transferred to the maintenance of quantitative restrictions. Removing quantitative restrictions ahead of tariff reduction would have been more beneficial to 4 Bank staff ould have helped in this process, but they dealt only with Government representatives and did not seek the opinion of industry representatives. The opportunity to do this was present. It could have helped the liberalization process. 13 the fiscal coffers. The Government could have collected revenues from tariffs, especially as import restrictions were eliminated. Rents from quantitative restrictions bestowed on some elements of industry would be eliminated, hence causing industry to be exposed to competitive pressure. This would have strengthened the adjustment process by getting rid of imbedded resistance against reform. As a result of the wrong sequence in the liberalization process, the reduction of tariffs rates strengthened the rents deri-Vd from quantitative restrictions. 18. Third, the methodology to determine progress of the liberalization program was difficult to monitor. The procedure used was to measure the percentage of liberalization as a ratio of production value in a given year (pegged to 1984). This procedure depended on agreements on the calculation of the degree of progress that should have been pretested and fully understood by both parties, not only during negotiation but more specifically during the implementation process. Complications arose from lack of clear agreement on the specific degree of aggregations concerning commodity and industry classifications.5 As it turned out, the calculation of the percentage of liberalization became a touchy technical issue during the loan supervision phase, resulting in prolonged discussion of the resulting percentage of liberalization achieved. It took four supervision missions to deal with all the implementation issues, and a substantial part of these missions were related to getting agreement on the progress in liberalization measures. 19. A simpler alternative for reviewing progress in liberalization would have been to design a liberalization program based on a positive list of items subject to particular import restrictions. From this list, agreement could have been made to progressively eliminate specific commodities in stages from restrictions. This procedure has the virtue of easy monitorability and the added advantage of greater transparency. Under the calculation method chosen, the composition of goods under restrictions were not definitely known in advance, since only a percentage of production value was being measured. This lack of transparency was the artificial result of the method of calculation. Prospectively, the method put the Bank in the dark as to which commodities under restrictions would be removed from restrictions, although the list of items under restrictions were known in advance and appended in the minutes of negotiations. Only after the decision had been made would the list of commodities being liberalized be made public through the official government journal. 20. The Government's Letter of Development Policy had set the quantitative targets of the reforms: by 1991, at least 75% of prices in terms of the value of production would have been liberalized; the next phase, scheduled for the first quarter of 1989, would bring the level S A specific reference to the manner of computation was stated in an annex to the negotiation, but the exac formula was not stated. Different interpretations on how to compute the percentage of production being liberated arose from the lack of direct correspondence between the customs classification of (imported) goods, which is based on a higher degree of disaggregation and the classification of domestic production which is based on standard national accounts. Liberalization was to be measured as a proportion of domestic production and manufacturing (based on 1984 figures) comprising of goods whose imports were not subject to quantitative restrictions. Bank staff does not share the conclusion that the formula for measuring import liberalization was difficult to monitor. Their position is that the formula was straightforward and was clearly understood by the Tunisian negotiators. They add that those of the authorities directly responsible for implementing trade liberalization who did not properly understand the formula were influenced largely by their reluctance to come to terms with the prospect of reducing protection. The objectives stated were somewhat ambiguously phrased in that they referred to percentage of prices (weighted by production) being freed from price controls. Expressed therefore as commodities being subject to price controls, by the end of 1991, 25% of production would be still subject to price controls. This meant that 75% of production would be liberated from the controls. The interim period by first quarter of 1989 was to have only 30% of commodities being subject 14 of liberalized prices to 70%; at the distribution level, prices would be deregulated for at least 20% in terms domestic production, in two steps of Fig. 3. ELIMINATION OF lAs 10% each, one before May 1 COM t* AL. WT LAN,A 1988 and the second at the end of the first quarter of 7 1989. , 21. T h e implementation of these measures was undertaken at a much slower rate, however. Fig. 3 compares three 2 schedules related to the 1 * elimination of quantitative restrictions. The percentage ".& , '-1),, ,"I1 ,'1' . Tiaing ey 0"aterly Ronse targets in liberalization are a G . + W=IN. A,TA G shown in the vertical axis and the dates of action in the horizontal. Three schedules are compared: per SAL as stated in the Letter of Development Policy; per the medium term program of the Government which spans a longer period; and per actjal implementation achieved as confirmed from Bank supervision reports. This figure shows that the timing of actual implementation was accomplished later (i.e., farther out to the right of the figure) compared to the original schedule indicated in the SAL, or even as announced in the medium-term plan. For a given period, the actual percentage of Fig. 4. PRICE DECONTROL liberalization achieved is also r A 1'C lower than the target indicated for 0 a 0 . the time period. 22. Fig. 4 shows a I corresponding chart on the 40 objectives and accomplishments, I in terms of the degree of liberalization, of price controls. The target at the beginning of i SAL for the percentage proportion of the production base covered by "SJ,,E' ,it' 4,, Timing By OWrterY 11144" price control was 60% of SGAL 90O * N*VIM GW S . A production value. By second tranche release, the target would be 70% of production value. Even though these targets are seen as having been reached, the slippage in action was similar to those for quantitative restrictions. With respect to the commodities covered by distribution margins, Fig. 5 gives information on projected targets as to price controls. I Goverment of Tunisia Iotr of Develrmet Polikv. Annex M to Presidents Report, recommending a proposed "Structural Adjustment Loan" to tbe Republic of Tunisia, Report No. P-408-TUN, p. 50. 15 against actual accomplishments. Fig S. REMOVAL: DISTRIBUTION MARGINS The program projected that the Gooo tw-v to ZAL. awl "LACrUAL removal of distribution margins would be undertaken up to a level of 20% of production value. These measures were all linked to the release of the second tranche of the SAL. The emerging pattern, however, is that the actual so timing of the actions related to the SAL conditions took place somewhat later (i.e., the actual implementation was later than the schedule anticipated in the SAL). a SAL Agtm 8. ACTU*4 dUL1 The main conclusion from these figures is that the accomplishments of the liberalization program by the second tranche release were relatively modest. 23. Aside from the delay in effectiveness of the loan, the slippage in schedule was caused by problems regarding acceptance of the trade liberalization measures within the government (as a result of industry pressure). In spite of this, shortly before the second tranche release (originally planned for September 1989 but finally released only January 1991), the targeted liberalization ratios for quantitative restrictions, for price decontrol, and for decontrol of margin distribution were reached by the authorities, therefore signalling continued commitment to the objectives of the liberalization measures. 24. Even as these targets for liberalization met the standards against which they were measured - the tranche release conditions - two points come out clearly. First, their adoptions were delayed seriously (a total of 14 months from the planned date of effectiveness). Second, and more significant, the liberalization objectives were modest. At the closing of the SAL, the Tunisian economy had achieved a modest liberalization of the trade and pricing regimes.$ Tariff Protection Rates 25. With respect to tariff protection, the basic thrust of the reduction of nominal protection rates under the program was achieved. The nominal protection rates were made to fall within a range of less than 40% ad valorem (for food imports) and about 10% (for energy imports). The average rate of nominal protection for the economy ("All Sectors") reached around 30% ad valorem. A decomposition of these rates is presented in Fig. 5 for 1990, based on calculations made by the authorities.' The decomposition indicates the various influences I This is further elaborated in the discussion of promotion of competition and price controls; sec ares, 33-34 below. 9 See"La Protection Effective en 1990," Notes et documents de travail de I'Institut d'Economie quantitative, No. 6, 1992. For a referenceon the earlier estimates of protection rates, see also 'Evolution de la protection et des incitations aux activirds dconomiques en 1977, 1980, et 1983 (resultats et mdthodologic)," Les Cahiers de IEO, Institut d'Economie Quantitative, No. 2, Mars 1986, pp. 48-107. 16 of tariffs with customs fees and the additional protective effects of Fiq.a. Nominal Protection Rates consumption taxes, tax Co O""" "". " 1 discriminations, and tax 061 deductions for each sector of " industry. Movements in this * decomposition of protection rates are also tracked between 1989 to :* - 1990. , Such changes indicate .04 -. significant reductions in the - components of average nominal -0- protection (see the negative bar -* for "All Sectors"), as shown in 0.0. , Fig. 6. For instance, the following changes in components SS TS. - - have contributed to the pattern: unfinished products (around -7%) and primary minerals (-4%), and agricultural equipment (-4%). 26. Effective rates of protection, which incorporate the protection of inputs as well as outputs in the measure of protection, have also been reduced. Effective rates of protection for industries serving the domestic market are markedly higher than their corresponding nominal protection rates. For export industries, the effective rates of protection are negative, as a result of fiscal incentives and zero nominal protection rates of inputs and outputs. The average effective rate of protection for the export sector is -4% ad valorem, and this is however heavily weighted by tourism, as shown in Fig. 7. The effective rate for tourism, the highest negative rate of protection of any export service, reached almost -20% ad Fig.7. Effective Rates of Protection valorem during 1987-1988. a Tourism subsidy is more than - . twice the average effective - 3 0 as protection rate for all exports. & Given the exchange rate * adjustments that already made li tourism attractive, these negative .012 protection rates indicate that the - issue still had to be addressed by * -0, the Government. I -a -0.2 27. In the case of " '. :"., ) 1, 50 . effective rates of protection for 160m 1 M a I a04 . domestic industries, the picture is very different. Fig. 8, which shows effective rates of protection from 1985 to the present, demonstrates the relative fall of the protection rate. But the effective rates are much higher than the corresponding nominal rate of protection. For manufacturing industry, effective rates of protection are about 1.5 times the average effective rates of protection for all domestic industries. On the whole, the level of effective rates of protection during the adjustment program up to the period of SAL had remained relatively stable since 1987. There was a slight increase in the protection rates in 17 1989, from the levels of the Fig.8. Effective Rates of Protection previous year, but this was not , OOMIC ET INMmIeS significant. 28. Even though effective rates of protection provide greater reality in the measure of protection, they do not take into account the impact of a binding set of quantitative restrictions. Quantitative restrictions provide scarcity rents for the owners of the privilege, but these rents of course cannot be M os 1M7 lees a og9 tracked down statistically. The scarcity rents derived from these restrictions would be high if there is actual unmet demand. The effective rates of protection are almost twice as much as the nominal rates, and for food industries, the effective rate is found to be as much as three times the nominal rate. 29. In conclusion, much still remains to be undertaken by the Government if it is to achieve its program of liboralizing trade and industry in the economy. The SAL on this account achieved a modest gain, largely because the targets of the reform in this context were also modest and the Government's implementation pace was slower than originally planned. Competition Law. Price Controls. and Financial Issues: Relationship with Industry and Trade Liberalization 30. A major objective of the SAL was to encourage the growth of competition in domestic commerce. The liberalization program in trade and industry represents an element of this thrust. A specific component of this objective was to encourage competition. As part of the conditions for second tranche release, a satisfactory draft law on the promotion of competition was to be introduced in Parliament. However, so long as trade restrictions and price controls are present in an economy, it would be difficult to foster competition. In view of the restrained targets to achieve liberalization, the credibility of any law designed to promote competition would be suspect. 31. There is therefore a problem of circularity. Industrial interests receiving rents from protection would be against trade liberalization so long as they face price controls. Since price controls would remain a permanent feature of some specific parts of domestic commerce, there would be continuing fear that trade liberalization and competition would not be fully achieved. The remaining controls on prices, especially on distribution margins, would remain extensive and the quantitative restrictions on imports would affect about 25% of production value. Domestic competition would be hampered under such a business environment. As long as the objectives of liberalization, and its speed of implementation, were not kept clearly delineated, the hindrances to competition would therefore remain. 18 32. In order to break this impasse, a number of measures could have been undertaken. One would have been to pursue a more rapid removal of quantitative restrictions and price controls. But by choice, the schedule slowed down. A second option was to create the conditions for domestic coiipetition by adopting a law on competition, to be discussed below. Another measure was related to strengthening the position of enterprises that had a potential to survive liberalization. 33. Law on competition and price controls. The draft law on promotion of domestic competition was presented to Parliament in November 1990 and became law in July 1991.1o The law deals with a wide range of market situations concerning competitive and noncompetitive behavior. It provided for regulations relating to price setting, affecting merchants and producers. It created a Commission on Competition which is a semi-judicial body designed to entertain questions brought to it by complainants related to anti-competitive behavior. It defined situations related to transparency of pricing to protect consumers and restrIctive commercial and professional practices, with an accompanying system of penalties. A wide ranging law such as this contains a number of compromises. While it could enhance competition, many provisions of the law could be turned around and interpreted as a means of stifling competition. This is possible in the case of the regulations related to consumer protection and those pertaining to price controls. How such an overreaching law would be implemented would determine, therefore, whether its impact would promote competition or help to derail it. The burden of proof on how it would function is critically dependent on the manner in which the Government perceived its strongest provisions to be. 34. The existing restrictions on domestic commerce demonstrates the difficulty of the policy agenda toward promoting greater competition. The presence of widespread price controls remains a major stumbling block. So long as they remained, competition would be hampered. This is especially made clear by reference to the implementing circular to the law on competition and pricesu which specifies the types of price controls and the regulations pertinent to their implementation. At present, there are essentially three types of price controls in existence: (a) goods whose prices are controlled at all stages of activity (homologation des prix i tous les ta4"); (b) goods controlled only at the production stage (homologation des prix au stade de la prQduction); (c) goods controlled by distribution margins (auto-homologation au stade de la distribution). Thus, even as the adjustment program had promoted a reduction of price controls, what remains is still a complicated system, affecting a significant part of domestic commerce. 12 to Law no. 91-64, approved on July 29, 1991, relating to competition and prices. As seen in Table 1, the presentation of an acceptable enabling bill was a condition of tranche release. This process was quite delayed and was one ground for the delay in the relase of the second tranche, in addition to the slow resolution of issues related to trade and price liberalization measures. "1 The decree passed by the Ministry of National Economy, no. 91-1996, dated December 23, 1991. 12 Goods under price control at all stages belong to consumer essentials. They cover a list, which, as of December 1991, included 20 commoditics. The list includes not only bread (subsidized), flour, sugar, cooking oil, energy (fuel, electricity, gas), and medicines, but also school fees, apartment rentals, nonalcoholic drinks served in restaurants. As for the goods subject to price control at the production stage include 21 commodities. The price control regime operating through distribution margins (auto-homologation), cover a list of 139 items (classified under either a three or four digit level of disaggregation from the national commodity classification system). 19 35. Financial issues linked to trade and industry liberalization. As shown in Table 1, the SAL supported elements of financial sector reform, especially those affecting interest rates. Preferential interest rates were removed. 36. The present discussion sketches the relationship of trade liberalization with the financial position of enterprises. In view of the highly protected industrial sector, enterprises have high unit costs and are inefficient. Tunisia's internal market for industry is dominated by the public enterprise sector in the case of larger industrial plants, and much of the smaller scale of manufacturing belongs to the private sector. If trade and industry reforms would have the effect of exposing enterprises to more competition, financial reforms had the effect of increasing the cost of capital. 37. At the same time, the banking sector was experiencing a difficult financial position. This was the result of past lending. Poor performance of borrowers was caused by credit having been guided at behest of public policy and not necessarily by economic criteria. Further weakened by a high incidence of nonperforming loans, financial sector reforms required that the banks be restored to health. This meant that they had to improve the quality of their portfolio and to provision for loans losses on nonperforming assets. Concerning enterprises, a delicate problem existed, which was to identify those enterprises that could remain viable during liberalization, to identify their transitional problems, and therefore to assist them during such a transition. This problem was not attended to during the design of the SAL. It could have helped to reduce the alliance against the liberalization process. Fiscal Reforms 38. Fiscal reforms supported by the SAL included a comprehensive reform of the tax system, both the direct and indirect taxes. In addition, the control and rationalization of subsidies improved the allocation of fiscal revenues and helped to reduce the fiscal deficit. 39. Tax reform is one aspect of the adjustment program that was implemented very satisfactorily. The tax system in Tunisia evolved from a long history of reforms since independence. These isolated efforts had caused the overall system to be complicated in structure and highly distortionary in its impact. The reformed system of taxation was designed to be more simple, transparent, and economically more efficient. The system of changes would cover both direct and indirect taxation. And the reform was not confined to the alteration of the tax base and the rates, but would extend to a reform of investigation and judicial procedures in tax cases. The net result was a new system of taxation that would be fairer and simpler and would provide improved deterrence against tax fraud. Reform of Indirect Taxation: the VAT 40. The adoption of a value-added tax (VAT) was extensively discussed with the Bank. Its adoption was undertaken prior to SAL effectiveness. But further improvements in the coverage of the VAT were implemented prior to the second tranche release. The VAT was designed to replace an indirect tax system based on three turnover taxes and a group of excise taxes. The turnover taxes had a multiplicity of rates - a structure of 13 different rates. In addition to this, excise taxes were levied with 45 different tax rates, the highest of which was 20 at 500%. In contrast, the VAT that was adopted had only three rates. The principal rate was at 17%, but basic consumption items and professional services would pay 6% and luxuries would be levied a rate of 29%. Exports would be zero-rated and agriculture exempt. 41. As initially conceived, the strategy was to eliminate the complex turnover taxation with a simpler VAT without creating any drastic impact on revenues and on prices. The base rates for the VA. were therefore made to correspond to the customary rates for the turnover taxes. This would create a switch in the tax base without the necesbity of disrupting the familiar tax rate, even though the tax base had changed. This strategy resulted from the concern of the Government that the revenues from VAT would fall and thereby cause a fiscal crisis. To provide a cushion, a transitional consumption tax on commodities was introduced on luxuries and on important sources of tax revenue under the old system. Initially, a wide number of formerly taxed products were covered by this temporary ad valorem consumption tax, collected at import or factory gates, and levied at three rates of taxes. The coverage was progressively reduced until the products on which excise taxes were levied constituted only alcoholic beverages, tobacco, motor vehicles, petroleum fuels, and a few luxuries.3 42. To achieve a high enough initial yield for the VAT required preparations prior to its full adoption. It depended crucially on implementation procedures being anticipated fully. This included training and a public information campaign. The initial training of tax agents was confined to an elite group of inspectors, who then conducted their training of other agents, especially those at the regional tax collection centers. The Government's public information campaign included issuing a booklet explaining the VAT in simple terms; holding several seminars for businessmen and accountants; and undertaking radio and television broadcasts as channels of further information. In spite of these preparations, there were transitional problems encountered during the initial phase of introduction. 43. The VAT was origin?lly planned to apply to wholesale traders with a yearly turnover of at least TD 500,000 in sales. As it turned out, the Government decided to apply the VAT to the wholesale trade without the threshold level for sales. This objective was undertaken, in part, to create a trade-off on the issue of the exclusion of foodstuffs from the application of the VAT: the loss of revenues from the exclusion of food would be partly compensated by the wider coverage of the VAT on wholesalers. 44. A study on the advisability of extending the VAT to retail level in the case of large retail establishments was to be considered prior to the release of the second tranche. This study, undertaken by a government institute, however, did not recommend that such a step be undertaken, arguing that it would be premature. In fact, some of the issues related to the resistance of the tax authorities to extend coverage of the VAT to retail commerce resulted from difficulties of implementation that were being experienced even at the manufacturing and wholesale stages. There were enough delays experienced in extending the VAT to wholesale trade. The practical difficulties of enterprises encountered in shifting to a new administrative system were partly due to lack of preparedness on their part and to some fear that the Government would use this new tax to help in the administration of price controls. (In this latter U Before second tranche release, this objective of limiting the scope of this tax and relying more on the VAT as the source of commodity taxation was nearing completion. The budget law for 1991 provided for the folowing steps regarding this consumption tax: (a) elimination of the tax on all goods currently taxed at 11%, with the exceptions of tea, coffee, and pepper; (b) reduction of the 25% tax rate to 11% and elimination of the tax for goods taxed at the lower rate; and (c) reduction of the 35% rate to 25%. 21 context, the official list of wholesalers maintained by the authorities in connection with price controls had to be separated from the list maintained to implement the VAT. This allowed the tax procedures to be independent of procedures tied to price regulations.) But, in spite of the efforts at public education, commerce was not accustomed to the same tax arrangements as producers. As a result, the tax authorities were obliged to negotiate the conditions for implementing the VAT reform sector by sector. 45. Even though the shift to a new form of indirect Ffg,9. FIRST YEARS OF VAT taxation through the introduction of O'N" the VAT experienced some transitional problems, on the whole, the reform was a successful experience. The coverage in terms " of taxpayers has steadily grown. - When initially introduced in May 1 4 - 1988, the number of taxpayers were 53,000. In December 1991, there were 92,414 taxpayers in the VAT registry. Even though most of .. these taxpayers were under the reme that is, the list of taxpayers whose assessments were basically dictated by the tax authorities without need for validation through records,' the number of taxpayers under normal VAT regime has also increased. Among individual taxpayers, only 8% of them were under the VAT regime. However, corporations and professionals, together with the individuals under the VAT regime, account for 26% of all taxpayers. The number of corporations has increased in coverage from year to year, as has the overall coverage of the tax. 46. VAT revenues (and those from the consumption tax) are shown in Fig. 9. The revenues have not fallen, although they failed to keep pace with the increase in the consumer price index (CPI) during the first year of introduction. By 1990, the VAT and consumption tax revenues had outpaced the CPI rise. The fear that the introduction of the VAT would cause inflation did not materialize. The inflation rate in Tunisia had been modest, by international standards, at about 7% per year. In common with many countries, the import base of commodity taxation is high. This is observed by comparing the composition of the VAT derived from domestic activity and from imports. The increase in the VAT revenues from 1989 to 1992 was due to the increase in the revenues from the domestic base of the VAT, indicating a significant compositional change. M This is a common feature of VATs in developing countries, often the result of record-keeping inadequacies among owners of enterprises. Due to imperfect record keeping, most taxpayers are subjected to a basic rate that the government uses to assess them for their tax liabilities. Under such a system, there is no need for verification of records. This is the essence of ymesumptive taxation, or the ime fkrfitaire in contrast with the true VAT regime (tg). 22 47. Thus, even though there had been variations in the way the VAT was finally implemented during the SAL period, the spirit and intent of the reforms was achieved. A satisfactory transition was accomplished, and the basis for future reforms in the VAT laid out. Ditect-Taxation 48. The reform of income taxation followed the introduction of the VAT, with a slight lag. The condition related to the presentation of a satisfactory draft law prior to effectiveness required a waiver, because of delays in drafting the required changes.15 The direct tax reform affected both personal and corporate incomes, which put all these taxes under a simpler, unified direct tax, eliminating the double taxation of dividend income. It took effect for personal incomes earned in 1990, and extended to corporations beginning in 1991. The resulting law had a simpler structure than the one originally envisioned at the time of SAL negotiations, and it covered a wider tax base. The highest marginal tax rate on personal incomes was set at 35% and much below the original highest rate. Tax exemptions on interest income from treasury bonds were removed (as part of a measure supported by the SAL in the financial sector). Taxation of fringe benefits given to workers, heretofore, exempt from taxes was undertaken. The reversal of this policy was a major step forward in taxation of wages. 49. With respect to the corporate income tax, the plan originally envisioned a tax structure with four rates. The tax adopted made it also simpler, with a structure based on two rates - a normal rate of 35% and a low rate of 10% for agriculture, small enterprises, and special ventures, such as retail cooperatives. Dividend incomes were placed at the normal tax rate for corporations. 50. It is still early to assess the impact of these changes in the direct tax. Like the VAT, the reforms in direct taxation are more recent and, in fact, were delayed in implementation for one year. However, the changes undertaken to improve the direct tax structure exceeded the original expectations at the time of the SAL tiegotiation. With improved coverage and a simpler structure, direct taxation would be expected to yield more revenue and to have less distortionary impact. The reduction in average tax rates was expected to cause a fall in revenue, but compensatory effects on compliance were expected to improve collection. Moreover, the reduction in tax rates would make it more feasible to reduce the range of gt ierous incentives presently offered to favored economic activities, which create a big leakage in tax revenues. Another type of gain would be in terms of administration: lower tax rates with a simpler structure favored cheaper cost of tax administration. 51. The revenue effects of these changes had been positive. The revenues from direct taxes from 1990 to 1991 increased by 15%. This was brought about by changes in the tax rates. The overall effect was to reduce the average tax rate, but in the case of professionals, this raised the tax rate. The reduced tax rates had brought about a psychological impact and raised overall S 'The officials in charge of general taxation had been ambitious in their original schedule for undertaking tax reform. The work on the VAT and the direct tax reforms could not be undertaken at the same time, in view of the limited number of personnel handling these measures. The work on income taxation therefore had to be postponedby one year, contrary to the expectations that the reforms would be in satisfactory draft form prior to SAL effectiveness. Staff memorandum, UL Sunervision Mission: Back to Office ReMort November 2, 1989. 23 revenues because of an improved rate of payment. Thus, the initial fears about a fall in tax revenues did not materialize. Controlling subsidies 52. One aspect of fiscal reform centered on the need to hold down the level of expenditures for subsidies. Subsidies used up almost 11% of overall fiscal resources, and accounted for 3.5% of GDP. They consumed an inordinate amount of resources. The SAL called for a reduction of subsidies, by setting an overall target for them during the period of the loan. This meant that the Government could program the reduction of subsidies so long as it met a target cumulative level that was stated for the Seventh Plan period. In annual terms, this was an average of 190 million dinars. The strategy for reducing subsidies was designed to raise prices primarily on the less sensitive consumer items, especially those which were part of the smallest proportion of the consumption of the poor. This strategy worked for fertilizer subsidies and for barley and soya. On other items in the consumption basket of the poor - milk, sugar, and vegetable oils for soap-making - the subsidies were to be reduced faster than on the more directly sensitive consumption items such as wheat and cooking oil. 53. The program was linked Fig.10. CONSUMER SUBSIDIES in part to the Bv Ca.clory of Expnditu. liberalization of 32n 300 - markets, through 280 - the decontrol of 260 - prices and 240- distribution 20- o 200- margins. Even V on - though the ' o liberalization C '4 - program had been 120 - delayed in several 5 lo - aspects, the so reduction of 40 - consumer subsidies 20 - had been basically ' achieved. Fig. 10 _ _ _ achieved Fg10Cereals FePrv Ia lzersMe Cooking O1 is Suwar (EFlour presents the profile of the consumer subsidies, which have been held in check over time. The total level of consumer subsidies had been held in check to about the level of the overall fiscal deficit, at 3.8% of GDP. Adjustment of subsidies has therefore reached the quantitative targets agreed upon. 24 .Other Issue Social Aspects of Adjustment and Emoloyment 54. Reducing the social costs of adjustment was identified as an element of the adjustment program. But there was no specific action program designed to deal with this issue in the policy matrix."' Tunisia's programs dealing with the social sector are relatively well developed compared to many countries. Income distribution, although important, is not a major problem in Tunisia.17 55. The social costs of adjustment were discussed in connection with consumer subsidies. Elaborating on this issue: "The strategy envisaged by the authorities for reducing subsidies over the medium term is to raise prices primarily on the least sensitive items, those constituting the smallest part of the consumption of the poor.... The meast es would ensure the greatest redistributive effect within the overall ceilings."18 But if subsidies were to remain as instruments to help reduce the costs of policy changes on the poor, they would have to be much more effectively targeted. The incidence of subsidies in the past had been much more diffuse. A comprehensive study of the operations of the subsidies and the alternatives available to government was completed as part of the adjustment program. What remained was to improve the program of action in order to target subsidies more effectively towards the poor, while decontrolling some commodities and allowing their prices to adjust towards their market levels. However, because such a design was not elaborated in the SAL, there was no component of the adjustment measures which looked into this aspect closely during the SAL period. Had it been specifically made as an integral part of the program, there would have been some review of the social impact of the adjustment during loan supervision. 56. The burden of adjustment in Tunisia had fallen largely on wage earners in the urban sector. Fig. 11 shows the changes in real wages, using 1986 (year commencing structural adjustment program) as the base year. Average real wages in public administration, public enterprises, and the private non-agricultural sector fell from their levels of 1983 (year of wage freeze) and have not yet recovered, although they have remained steady since 1987. Nonwage income earners experienced rising real incomes, especially in agriculture. 57. Employment creation is an aspect of adjustment that is not fully or accurately tracked. Average employment increase has been estimated at around 26,400 per year during 1987-91. However, this is well- short of the 50,000 annual employment creation target of the 1' The President's Report, paras. 27-28, discusses directly the elements of social programs, including Bank support through other operations, which have a bearing on reducing the cost of adjustment. In particular, these issues deal with social policies, related to consumer subsidies, social security and health care systems, family planning, education, training and employment, including small scale industry development. 0 The Gini coefficient, estimated at 40.1% in 1990, indicates that income inequality is not a major issue in Tunisia, having fallen from 44% in 1975. This estimate for Tunisia is midway between the Gini coefficients estimated for developed countries (USA, 36.8%; UK, 31.9%; France, 33.3%) and other middle income developing countries (e.g. (av. 1967- 85), Turkey, 57%; Thailand. 47%; Colombia, 45%; Mexico, 50%; Cote d'Ivoire, 55%). I President's Report, am. 52; see also 21m. 49 25 VIth medium term economic plan, and in respect of the original Fig. 11. REAL WAGES, 1973; 1980-91 expectations during adjustment. "" tO" Supply response from export- oAented enterprises should be felt postively by industries benefiting from adjustment,, but information on employment for these industrie. is not available. 58. Clues on the employment impact of adjustment . may be derived from statistics of * ;t . I. , . N if a, investment intentions. The 0 *not + MI Itg = aa. 0 ALL WVOM S4Mt A Mo)WMA. V=C. AM registry of investment intentions in - manufacturing for totally export- oriented enterprises fell in 1991: this was both in terms of value of investments and amount of new employment. On the other hand, intentions for projects serving the domestic sector rose. In overall terms, the value of investment intentions had remained almost the same in these two years, but the total employment expected from these intentions was lower for 1991.19 59. It would be premature to read any conclusions relevant to trends on investment and employment from these data. The investment intentions merely establish the existence of private future plans, and they have yet to be realized. The investment intentions in the second year have relatively higher capital requirements per unit of labor employed than the first year intentions, and this suggests an increasing capital intensity. Disburements 60. Failure to meet originally scheduled tranche release dates naturally delayed actual disbursements of SAL proceeds. Once the tranche release issue had been settled, %isbursements 19 The following statistics of investment intentions are shown below: Sector: Manufacturing Investment Employment to Be Intentions by Type of Market Number of Intended Investments Created Orientation Projects (In million dinars) (in man-years) 1990 1991 1990 1991 1990 1991 1446 880 319.0 300.3 62161 39397 Fully Export-Oriented Domestic, other than Fully Export- 2875 3527 735.3 768.8 36842 46842 Oriented Total 4321 4407 1054.1 1069.1 99003 86239 Source: Agency for the Promotion of Industry (A.P.I.). 26 were made in a straightforward fashion. The earlier adjustment loans had a similar experience. In the case of IMF's Extended Fund Facility, Tunisia delayed its use of the facility and applied for its use only in May 1991, as a result of pressures on balance of payments during the Gulf crisis. Even then, it requested for a reduction of the facility by one-third. The Bank's quick disbursing assistance linked to the adjustment program enabled Tunisia to fund its balance of payments shortfalls adequately. 61. The disbursements of the proceeds of the loan followed the Bank's policy. There was one exception, however; a contract for 20,000 tons of steel rods, valued in excess of US$5 million, was awarded without benefit of International Competitive Bidding (ICB). The price negotiated for the shipment was found reasonable by the Bank's procurement adviser.2 A similar contract for steel, representing a different batch of steel rods imports (for 15,000 tons), was awarded to a competing lower cost supplier. The resulting value was bUlow the threshold limit, and therefore it was exempted from ICB. IV. ASSESSMENT AND LESSONS OF EXPERIENCE Genera Assessment ad Susainabilt 62. The progress of adjustment in Tunisia has proceeded steadily. The macroeconomic framework in Tunisia was brought to a satisfactory balance by undertaking strong measures to reduce demand. This was accomplished by reducing aggregate demand in the economy through a number of measures -- cuts in public spending, in consumption and investment. This was accompanied by exchange rate devaluation and by controlling the rapid rise of wages through a wage freeze policy initially and then, through a modulated policy of allowing reasonable wage rate increases. The Government had anticipated the macroeconomic problems in the mid-1980s, and this probably accounted for the degree of understanding of the real steps needed to correct the macroeconomic imbalances. 63. Even though the Government has succeeded in managing the macroeconomic balance, the complementary measures requiring structural policy changes in the sectors still need some degree of concerted push. The SAL followed two other sector structural adjustment loans, in which correcting the macroeconomic framework was a major element. But the progress in sector reforms require further deepening. This is most important in the case of reforms in trade and industry. 64. The Government's intentions with respect to liberalization of industry and trade, through the elimination of both quantitative restrictions and price controls, were announced at the beginning of the adjustment program and during SAL negotiation. However, there is only modest progress by the end of the VIIth Plan in 1991. This lack of depth of progress in the liberalization of trade and industry was due to a slow down of the liberalization process. Quantitative import restrictions and the continuation of many forms of price controls within domestic commerce still remain pervasive. As explained earlier, a strategic mistake in designing the program - both in terms of sequencing the reforms and in formulation of the conditions for Consultant memorandum to Staftf datod August 9, 1989. 27 monitoring progress - had led to some missed opportunities in strengthening the case for reforms. 65. Much remains to be achieved in creating a climate of business environment conducive to progress in opening the market for domestic competition, in spite of the passage of a law improving compeCdon. The necessary condition for improving domestic competition would require fufther progreSs in the liberalization. And even though the tariff structure is nominally structured so that protection rates have become more rationalized and the average rate has fallen, unless quantitative restrictions are eliminated, much of the distortions from protection within the economy would remain. As to domestic commerce, the presence of price controls restricts competitive behavior and creates a distortion in the incentive system. It also promotes a public attitude towards commerce which is not conducive to the development of competition. 66. In other areas of the structural adjustment program, additional progress of reforms would require additional adjustment lending beyond SAL, notably to support the reform of the public enterprise sector and privatization and to deepen reforms in the financial sector. The assessment of actual performance in these two areas of reforms would have to await a more intensive review. They constitute, however, two critical complementary reforms. 67. The area of economic policy that has been most successful is in tax reform. The limited administrative resources of the Government were made to implement two major reforms in the tax system, both within a relatively short period of time. Despite problems during the transition, the VAT is now an integral feature of the indirect tax system. By laying the foundations for improved implementation of the VAT during the few years of its adoption, the outlook for a more neutral tax base to raise government revenues has been strengthened. And even if the VAT is presently extended only to wholesale distribution, further experience would strengthen a movement towards its final application to the retail stage in the future. The direct tax system reforms were also adopted in relatively quick time, even though there was a slight delay. The reforms in the direct and indirect taxes were a demonstration that comprehensive tax reforms can be undertaken during a relatively short period of time. 68. Structural adjustment in Tunisia has a strong chance of succeeding. The Government has demonstrated adequate control of the macroeconomic environment. However, a sufficient condition for ultimate success would depend on how the authorities would restructure policies affecting trade and industry, to make it more competitive, both for exports and for domestic commerce. Success would depend on undertaking further advance in liberalization, beyond the intended scope that the Government had announced within SAL and in the medium term plan. However, the policy pronouncements indicate that the limited objectives, as indicated in the slowed liberalization process, would continue. Role of the Bn 69. The Bank supported the efforts of the Government to stave off balance of payments crisis and to facilitate the structural policy ,hanges required to undertake a change in economic direction, with economic growth as the important objective. By providing the necessary balance of payments financing through adjustment lending, the Bank allowed the authorities to meet essential external payments when they were due. With the appropriate 28 measures taken by the Government to improve the macroeconomic situation, the impending economic crisis became less onerous in impact. 70. In spite of the limited accomplishments so far of the liberalization program, it is likely that most of the measures that were accomplished would have been impossible to implement without the support of the Bank. If the appropriate sequencing of the liberalization measures had been correctly pursued and the design of monitoring progress in liberalization had been better undertaken, it would have been possible to advance the liberalization program further. Lessons of Experience 71. The implementation of SAL reveals important issues of design and adjustment strategy. The major lessons of experience are discussed below: 72. Political economy of reforms. Difficult structural reforms can be carried out with speed only if sufficient national consensus in support of them is present. This is the main lesson from the slowed trade and industry liberalization program. The Government was unable to implement its commitment to undertake the programmed liberalization measures, because there was not sufficient national consensus to back it up. The technocratic team which promoted the agenda of economic liberalization was not successful in convincing industry and the rest of Government on the benefits to be derived from liberalization. This probably partly stemmed from the presentation of the adjustment program. The reformers failed to sponsor their program of reform within a package of benefits and costs to help foster political consensus on liberalization. And the reform program was perceived by industrial interests with a strong channel within government as a package of painful measures. 73. In connection with transitional costs of liberalization, industries which are grossly inefficient would not survive competition. But there are industries that could survive the change in regime. The failure to identify those industries and to provide them with some temporary support to survive the changing situation strengthened resistance against reform and made it more difficult to win support from industry. 74. Sequencing the liberalization process: tariffication of import protection is made more difficult if quantitative restrictions on imports are not removed ahead of tariff rate restructuring. The removal of quantitative restrictions should be undertaken ahead of tariff reduction. Doing it the other way - reduce tariffs ahead of eliminating quantitative restrictions - - was the approach attempted in Tunisia. This sequence of policy change meant that the potential gains in tariffs from protection were lost by Government (to the extent that the protective rate exceeded the new, lowered tariff rate). 75. At the same time, the rents from quantitative restrictions were retained and, in turn, strengthened those whose interests were to oppose the liberalization program. This had the consequence of further strengthening the position of those holding quotas and had the indirect effect of strengthening the already imbedded resistance to reform coming from quota holders. 29 76. Importance of monitorability of progress in liberalization. A methodology that is not easily monitorable is subject to frictions during economic policy dialogue, because of lack of understanding. An example was the methodology designed to determine progress of the liberalization program. Although conceptually attractive in presenting some progress of liberalization by measuring it with respect to overall production level that was being freed from economic controls, the difficulty of the monitoring procedure probably delayed the achievement of the targets. In this connection, the use of a positive list of commodities under quantitative restrictions -- to be reduced progressively in stages -- would be more implementable as a reform conditionality. 77. Price liberalization can proceed more effectively under a regime promoting internal competition. The price liberalization process, as negotiated, was somewhat limited in terms of depth. A large degree of price controls remained within the economy. Because economic agents operated under a regime of dirigiste structure;, through government administered pricing policies, competitive pricing was hampered institutionally. So long as price controls remain as a feature of domestic commerce, the implementing machinery for promoting competition would be hampered by price regulatory pressures. 78. Tax reform. The experience of Tunisia in instituting two major sets of tax changes within a reasonably short period of time indicates that with adequate preparation and careful study of the issues, it is possible to undertake a comprehensive reform of the tax system. There are lessons of experience that could be cited in connection with the shift towards VAT in indirect taxation and the reduction of the average rates and their simplification in respect to direct taxation. In the case of VAT, the preparatory work prior to the shift to VAT was significant in making VAT accepted by the general public. The presence of a countervailing set of temporary consumption taxes, within the tax structure during the shift to VAT, was significant in avoiding revenue losses from VAT. 30 ANNEX I Page 1 of 8 TABLE 1: MACROECONOMIC EVOLUTION DURING THE VIlTM PLAN PERIOD SAL PORECASTS AND REALIZATIONS 1986 Basis Average Roa (in millions Growth Ratos of dinars) 1987-91 (in % per year) (1) SAL Forecasts Actual (2) (3) WDP 7021.0 3.9 4.2 Agriculture 933.0 4.1 6.7 Energy 519.0 -4.4 -1.7 Others 5569.0 4.5 4.9 Consumption 7825.0 3.2 3.2 Investns 2207.0 1.5 3.5 Expor* (GNPS) 3123.0 5.8 7.6 Ipors (ONPS) 3849.0 2.8 5.2 1986 ratios lncreases from 1986 as to 1991 cn Groas Invetmets/GDP 23.5 -1.6 0.6 Gov't Invcstmnts/GDP 4.1 -1.2 1.0 Public Enterprises Inv./GDP 8.2 -2.5 Goverament Expenditurs/GDP 37.3 6.1 -7.1 Govornment Deflit/GDP -5.5 4.0 1.5 Expot/GDP 30.8 5.2 8.5 Imort/GDP 38.0 1.2 5.8 External Current Account/GDP 8.0 4.3 -3.5 Debt Outstanding 58.6 -3.8 -6.4 Disbursed/GDP 28.2 -1.0 -4.7 Debt Service/Exp. GNPS Real Effetive Exh. Rate 93.6 -5.6 -12.0 ('80.100) 100.0 -18.0 -13.5 SMIG Deftted by CPI 100.0 15.1 23.8 SMIG 48 hts. Monthly Noninal Somrme Col. (1) SAL Pfeident Repor or revised daua by MinW~y of Pa~ning cl. @ SAL P~ednt Roprt COl ( Minis y of Pnnin Noa: Is the m~hod of calcuag awage groef aes the sar~ o. n tahe Prsid t Rort Takbe 1, p.7. nd i the Tåble prpured by d Miusuy of Paning In ky 1992? Diffeenc ibeowen SAL fosts and actui do ot k erco nt betwee grind h prtes usinatthln pper partof thotalesand ado of lnM tatol GDPinthe ower part, 31 Pag. 2 of 8 TABLE 2: EXPORT EARNINGS IN MILLION DINARS AT 1990 PRICES 1986 1987 1988 1989 1990 1991 1992 Ptroleum & Products 560 520 504 633 535 518 441 Phospbates & Producta 416 471 514 523 454 460 505 AgiCulturo & 248 302 296 302 348 496 428 Processed 637 692 769 936 1,179 1,235 1,371 Textils & Lather 159 177 224 249 302 353 389 Mochanical & 84 129 181 314 272 193 231 Electrical Others Total Goods 2,103 2,290 2,488 2,956 3,089 3,255 3,364 Tourism 483 672 1,236 925 828 585 957 Other Services 537 578 576 632 676 664 779 Total Services 1,020 1,251 1.S11 1,557 1,503 1,249 1,735 Grand Total 3,123 3,540 4,299 4,513 4,592 4,504 5,100 Source: Ministry of Planning. TABLE 3: REAL ~El1RN TO LABOR O M'SIDE AGRICULTURE I/ 1973 1980 1981 1982 1983 1984 i15 1986 1987 198 1989 1990 1991 SMIG 2/ 46.4 73.3 82.0 98.1 10.0 92.1 85.6 89.4 86.3 90.0 80.3 81.4 7.8 &ubie AdmInisr. 93.9 99.3 190.0 95.5 89.6 89.6 87.4 85.8 89.3 87.9 89.1 AiIWagU yamess 67.4 85.3 89.1 95.6 100A 94.2 90.7 OB.4 83.1 82.1 82.8 82.t 83.7 Ncn-Wage E,n~es Exchling Agri. 41 40.0 105.5 109.5 95.9 100.0 99.0 98.6 102.0 104.0 120.5 12B.4 133. 121.4 A~grcultual Sector 83.6 121.2 118.6 109.6 100.0 119.9 146.2 121.9 12,0 121.5 138.7 186.8 216.7 C.P.l. 80. 91.8 100.0 108.6 116.8 124.0 134.2 143.8 155.0 165.5 17.9 S~urce: Ministry of Planning. 1/ Defla~ed by ft cUmr pri~ ~ dex (CP). 2/ sa1.a, minimum ~iutier saai gulm~ed mamma indunrial wago,. 31 Avrage ways in pubqle admin^istin pulic aueprise and private aC-Sricultural ecor. 32 ANNEXI rbbk 4. koan~ ac ~s Pi im im im im im im im im m 1990 1991 im ln Rd Terms: WDP 5.5 0.5 4.7 5.7 5.6 .1.4 6.7 0.1 3.7 7.6 3.5 5.9 C-~supdn 7.4 3.8 1.9 6.0 2.9 1.0 1.2 .03 3.9 6.6 4.0 4.1 Fat inveae 14.2 8.2 -4.7 .8.3 -8.3 .17.8 -10.1 124 19.1 2.9 10.6 10.6 DomWek n e 9.5 2.5 2.9 6.7 -1.4 -3.8 0.6 -2.7 7.7 11.0 0.9 5.0 E ~t ~ Goods & NF8 3.5 6.9 0.9 2.1 3.3 5.2 13.4 21.4 5.0 1.8 4.9 13.2 E r-~f~ 20.2 7.2 4.1 .8 1.7 13.1 19.17 16.6 17.3 14.7 14.7 6.1 ~ ~port of Gooda & NPS 13.0 0.9 -2.2 5.1 13.0 -2.1 -3.5 16.1 i4.6 8.9 -7.7 8.9 Red Eff~Ldv ~ e 1u~ngRta -2.9 1.1 -2.8 -1.6 -3.4 -21.0 .5.3 0.5 .2.2 0.3 2.5 in Nmiall Twra CPi 9.0 13.6 8.9 8.6 7.6 6.2 8.2 7.2 7.7 6.6 7.8 ODP 17.6 15.4 14.4 13.5 10.7 1.6 13.9 8.6 11.2 13.8 10.4 13.5 Money (M2 20.2 21.6 11.1 11.17 14.4 4.9 14.9 11.5 15.5 7.7 5.8 Domnm* Cmi« 29.6 25.6 21.0 15.9 16.2 8.3 8.8 4.4 22.5 7.7 8.2 GWrant 16.4 2.1 15.4 27.8 22.8 13.2 14.9 -6.1 49 28.1 5.5 Ea-m 31.8 26.4 21.8 14.4 15.2 7.5 7.8 6.1 26.9 5.4 8.5 Pwrclage of DP Goaenk~ Rr~ 29.3 32.0 31.6 34.0 31.9 31.8 29.6 30.0 29.3 27.3 27.1 26.3 Overnm EspudiMe 34.6 37.9 30.2 40.1 36.9 31.3 32.7 33.6 33.1 31.1 31.1 29.0 Def~oi b~1nae gM •5.2 40 .7.6 47 -4.9 -5.5 -3.1 *4.6 -3.8 -3.8 4.0 -2.7 Eport 41.4 369 3$A 3.9 32.6 30.8 35. 41.9 44.0 41.8 39.3 41.9 49.8 47.4 44.0 45.0 38.7 38.0 36.0 41.5 48.0 50.0 43.8 44.9 Curan Acount Defii -7.7 -9.4 -7.4 10.9 -7.1 -8.0 1.0 1.0 -3.3 •5.3 -4.2 -2.6 Publio 1avsu * 5.1 5.1 5.6 5.3 5.5 4.9 4.1 4.1 4.5 5.4 5.6 53 Grea ~ invunnan 32.3 31.7 29.6 32.0 26.6 23.5 25.6 19.4 27.6 26.6 23.2 23.4 Oes Domoido Savings 23.9 21.1 21.0 2.3 A0.4 16.2 19.6 19.8 18.6 19.3 18.7 0.4 Deb ~taadng Diab~d 42.8 464 50.1 51.0 58.9 66.9 10.2 66.7 68.0 61.5 d 59.1 Ratio and Indics a Speified Deok Service Rati 15.2 16.2 19.2 22.5 24.7 28.2 28.2 21.9 21.7 23.8 27.0 21.7 påbL over Pri~. tnvuOnn 18.9 19.2 23.5 19.7 26.0 26.3 24.5 27.2 24.6 25.5 31.6 29.3 Tms of Trade (1980 - 100) 105.3 103.0 99.0 96.3 93.4 82.0 84.2 83.4 84.3 84.0 82.7 83.0 Read Effö~dvs Ranny Rata 99.3 98.8 9.4 96.2 95.1 81.6 70.5 69.7 69.0 66.9 68.9 (1980 - 1007) Souce Miisy f Pumngandsff w~OLna * DomSetic demand equl n- pi 1ie ihIlmn pl changs in stoack . " Du~ of u yr y oer Decmbe ef p~evios yar basd on 1MP uegbi mevised in 1992 (ance 32.7%.11aly 20.8%. Gernuny 15.0%. &~gim 6.8%. USA 5.1%. Spai. 4.3%. Nulbarlands, 3.2%. Algeri 2.8%, Grece 2.2%. U.K. 2.0%. Japan 1.0%, Tr~cey 1.5%. and lendl 1.3%). " Defied a nvestmt by Ctral Goveman t (1985 mmard P.5636 Tisia). 33 ANNEX I Page 4 of 8 Table S. TRADE AND INDUSTRY IABERAlUZAThON: QRs ant PRICE DECONTROL (Percentage level Planal or AWamnd) 1[BERAlIZATION OF QUANT(TATIVE RESTRICTFONS PRICES DISTRIBUTON MARGINS MEDIUM TERM MEDIUM TERM MEDIUM TERM PERIOD SAL PLAN ACTUAL SAL PLAN ACTUAL SAL PLAN ACTUAL 0/d/88 5.2 5.2 60.0 10.0 30109188 12.9 31103189 70.0 20.0 30/09/89 28.3 12.5 70.0 25.0 10.0 31/03/90 36.0 71.7 30.0 31/06/90 73.8 35.0 30.8 30/09/90 43.7 73.4 40.0 30/12/90 48.5 19.8 31/03/91 51.4 21.9 45.0 30/09/91 75.0 59.1 75.0 50.0 31/03/92 66.7 30/09/92 75.0 Source: Preaident's Report and Supervision report. TAiL8 243001T02RS OF NORfINAL PROTCnON RAT4is. DIS40GREG4?ED BY SOLrRcE OFD7iFECl5* NOMINAL PROTECTION RATEs NMINAL ?R~ECTION RATis (NPM) T-nffs & Twifli & Tw^iff A1 ALpn Tex Tdal TWiffM Al Lmpat Tex T4el ridv, 'stln~ P~oedion Di~aam. 1mpa Only Cm~ P~Dcedi.n aImp NmT CHANGE FROM 1989 A989 DECOPOSIT1ON 1990 DECM~ITION DCM ON PodTaOrmps T1 T2 T3 T4 TS NPR Ti T2 T3 T4 TS NPR TI' 12 T3* T4° TS' NPM* Prènury Mimnee Mialo W 0.51% 448% 0.00% 0.00% 20.35% 19.70% 0.52% 4&42% 0.00% -13% 18.67% d0% 401% 0.0% 0.00% -3.3% -3.6% PnrOMy gne Mo. .t.lae 1.%1 -0.41% 0.00% -0.37% 23.28% 24.20% 0.S8% -59% 0.00% -2.72% 21.47% 0.70% 0.02% 418% 0.00% -2.35% -1.81% Enex8y P1odexts. 70% 0.34% 0.00% 0.00% 0.00% 12.04% 10.80% 0.36% -42% 0.00% 0.00 10.74% 490% 0.02% -042% .00% 0.00% -.30% Vfinnel~d P"oda ..41% Gæ% -0.15% 0.00% 429% 25.56% 24.40% 00% -415% 0,00% 454% 18.31% -1.00% 0.00% 0.00% 0.00% 425% -7.25% Agiui Equipe, :440 0.46% 406% 3.87% -5.63% 23.24% 25.00% 0.45% 0,00% IR% -9.32% 20.01% 0.40% 4.01% 0.06% 0.01% -3.69% 3.23% ind Equipmebe 22.60% 0.59% -0.26% 5.30% -2.69% 25.35% 2210% 0.58% 425% 4.89% ..67% 24.65% -0.50% -01% 0.01% 441% 0.01% 4L% poed Pd~as 37.t0% 0.84% -4.30% 0.00% -0.75% 35.39% 37.50% 1.17% -3.43% 0.00 -2.76% 2.48% 410% 33% 0.87% OØS 4t01% -91% Other cosump9en Prodoas .80% 0.86% -2.77% 0.00% 0.10% 34.99% 37.20% 0.87% 3.49% 00% 4.25% 33.33% 0.40% 0.01% 4«2% 0.00 4% -1.% WE~HTED AVERA E 29.80% 0.71% 1.55% 0.95% 4.69% 32.32% 29.20% 0.74% -1.61% 0.87% -3.45% 25.75% -40% 0.03% .3.16% 40~ -23% 457% Tariffi Only - TI Pr~cain ~ale befed øn cut~ms dis and feem Taiff* & r Cber Cuetoms Lees - T2 Prodian n bed onted nd <mer comn« dios atoms 4dtion Impon Pedian -T3 Pr~edioø m~ ;nd, ed-in.3 owaptio taeme Tx Disaination -T4 Pretdin raee indoding ar~d nl pr aion to1 tkx sale TddI P~-~din" -TS Pr~edion salon indode tedffs, cutom fe, aM«sumption txes, pr ~mnd fixerm~l taxes Souro: 106ke0 fr Qm ~ Eonosms. Tom»i ut 00 - 35 ii j e'0 0~ .1!!aI * * 0 9 i. - 36 ANNEX I Page 7 of 8 Table . VALUE ADDED TAX AND OTHER CONSUMFITON TAXES SINCE VAT INRODUCTION IN1988 (Million Dinars) 1988 1989 1990 1991 1992 VAT REVENUES, excl. tobacco Domestic 213.9 191.1 232.9 279.6 341.8 Import 307.6 334.5 360.0 360.1 415.1 VAT REVENUES, tobacco Domestic 30.1 18.6 23.8 22.9 20.2 Import 10.0 10.5 12.1 16.4 18.6 TOTAL VAT REVENUES 561.6 554.7 628.8 679.0 795.7 % of Import Base 56.6% 62.2% 59.2% 55.4% 54.5% % Yearly Increase in Revenues -1.2% 13.4% 8.0% 17.2% CONSUMPION TAX REVENUES Domestic, Miscellaneous 10.3 10.9 12.6 7.9 8.2 Imported vehicles 10.9 7.9 13.6 Imported, other than Vehicles 14.9 32.4 11.9 10.3 11.2 TOTAL VAT AND CONSUMPTION TAX 586.8 598.0 664.2 705.1 828.7 % Based on Imports 56.7% 63.1% 59.5% 56.0% 55.3% % Yearly Increae in Revenues 1.9% 11.1% 6.2% 17.5% YEARLY CHANGE, CPI 6.4% 7.4% 6.8% 8.2% Source: Ministry of Finance, Tunisia 37 ANNEX I Page 8 of 8 TABLE 9. VALUE ADDED TAX: COMPOSITION OF TAXPAYERS PERCENTAGE COMPOSITION Category of Taxpayer 1/7/88 31/12/89 31/12/90 31/12/91 Subgroup Main Groups Individuals 57,356 69,197 74,078 100.0% 80.2% VAT Regime na 2,456 3,520 5,627 7.6% Regime Forfaitaire 40,000 54,900 65,377 68,451 92.4% Professionals na 5,884 6,598 6,939 7.5% Corporations na 8,646 10,865 11,397 12.3% TOTAL TAXPAYERS 53,000 71,886 86,660 92,414 100.0% 100.0% VAT Regime 13,000 16,986 20,983 23,963 25.9% Regime Forfaitaire 40,000 54,900 65,377 68,451 74.1% Source: Ministry of Finance, Tunisia TABLE 10. EPENDTITIRES OF 77E CAISSE GENERALE CEMR4LE (In Million Dinas) Expenditure 1987 1988 1989 1990 1991 1992 Cereals 110.0 116.0 200.0 180.0 165.0 165.0 Fertilizers 17.7 10.0 21.0 17.0 11.5 5.0 Cooking Oils 20.0 25.0 25.0 35.0 43.0 40.0 Sugar 15.0 17.0 35.0 34.5 18.0 13.5 Wheat 8.5 15.0 35.2 27.0 16.0 24.3 Paper for School, culture 4.5 3.3 7.5 8.5 11.0 12.2 Subsidies for Groups 2.6 0.7 0.7 0.6 0.7 0.7 Reimb. to Treasury 5.0 5.0 5.0 5.0 5.0 5.0 Aid to School Children 0.5 0.7 1.5 1.2 1.8 Potato 1.8 2.9 2.2 2.8 2.5 1.2 Other expenses 2.7 1.5 7.4 10.0 6.3 TOTAL 188.2 194.9 333.8 319.3 283.8 275.0 Source: Ministry of Finance 38 ANNEX II Page 1 of 1 Monsieur Mark Baird ,Chef de Division Politiques Economiques Industrie et Finance ,l à la Banque Mondiale 0 B J E'T Observation sur le Projet de Rapport dEvaluation Rétrospective du Projet Republique Tunlsienne- Prêt à l'Ajustement Structurel (PAS)• RE ERExCE: Votre lettre du 7 Ma 1993. Monsieur, J'ai llionneur daccuser réception du projet de rapport Reputique umsienne-Pr$t à l'Ajustement Structurel transmis par votre courrier du 7 Mai 1993 La partie " Réforme Fiscale du projet de rapport ne soulève pas dobjetton de ma part. Toutefois, il convient de noter comme complément au point 41 de la page 22 . qu'une liste de produits compoitant 34 sous positionstarifairera été retirée du tableau dui - b tt de consommation par les lois de finances pour la gestion 1992 et 1993. Cette tistE colliporte notamment les produits suivants : produits alimentaires, machines à laver. tisus et v$tement en soie, broderies, verreries pour mosaique, montres, ¢afé soiuble. bijouterie de fantaisie... Veuillez agt4er, Monsieur le chef de Division, l'expression de tnd ·rnsidération distinguée . Le Directeur G4néral du Contr8le Signé: Mohamed HAJ MAN UR