Document of The World Bank FOR OFFICIAL USE ONLY Report No. 10405 PROGRAM PERFORMANCE AUDIT REPORT MADAGASCAR INDUSTRIAL ASSISTANCE PROJECT (IDA CREDIT 1541-MAC AND SFA CREDIT A-7-MAG) FEBRUARY 27, 1992 MICROFICHE COPY Report No. 10405-MAG Type: (PPR) NAMISATO, / X31678 / T9 105/ OEDD2 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance o" their official duties. Its contents may not otherwise be disclosed without World Bank authorization. ABBRVIATIONS BNI Bankin'Ny Indostria (National Bank of Industry) ISAC Industrial Sector Adjustment Credit ITPAC Industrial and Trade Policy Adjustment Credit HIEK Ministry of Industry, Energy and Mines OGL Open General Licence PIP Public Investment Program PSAC Public Sector Adjustment Credit RIL Liberalized Import Regime SFA Special Facility for Sub-Saharan Africa SILI Liberalized Import System FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.5A Office of Directo-General Opeatms Evaluatkn February 27. 1992 MENORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Program Performance Audit Report on Madagascar - Industrial Assistance Project (IDA Credit 1541-MAG and SFA Credit A-7-MAG) Attached, for information, is a copy of a report entitled "Program Performance Audit Report on Madagascar - Industrial Assistance Project (IDA Credit 1541-MAG and SFA Credit A-7-MAG)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distributioki and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROGRAM PERFORMANCE AUDIT REPORT MADAGASCAR INDUSTRIAL ASSISTANCE PROJECT (IDA Credit 1541-MAG and SFA Credit A-7-MAG) TABLE OF CONTENTS Page No. PROJECT PERFORMANCE AUDIT REPORT PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i BASIC DATA SHEET . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 EVALUATION SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . v I. BACKGROUND....................... . . . 1 A. Historical Background of the Malagasy Economy.... . . .1 B. Evolution and Features of the Industzial Sector . . . . . 3 II. CONCEPT AND DESIGN OF THE ISAC. . ........... . . . . 6 A. Industry Constraints at the Time of the ISAC. .... . . 6 B. The New Industrial Development Strategy..... . . . . . 9 C. Objectives, Policy Instruments, and Conditionality . . . 10 III. IMPLEMENTATION EXPERIENCE AND OUTCOME. ......... . . . 12 A. Policy Package.............. ..... . . . 13 (1) Price Liberalization...... ..... . . . . . 13 (2) Public Investment Program (PIP) Reform..... . . . 14 (3) Import Liberalization and Export Promotion . . . . . 15 (4) Private Sector Encouragement. ....... . . . . . 17 B. BNI Investment Component . . . ............. 18 C. Technical Assistance Component..... .... . . . . . 20 IV. ASSESSMENT OF OUTCOME WITHIN THE OVERALL ADJUSTMENT PROGRAM . . 20 A. Assessment of Outcome . . . . . . . . . . . . . . . . . . 20 B. Impact on Macroeconomic Performance (1985-1989) . . . . . 26 C. Supply Response . . . . . . . . . . . . . . . . . . . . . 26 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. Table of Contents (cont'd) Page No. V. CREDIT DESIGN, UTILIZATION. AND SUPERVISION. ... . . . . . . 28 A. Design Evaluation . . . . . . . . . . . . . . . . . . . . 28 (1) Gradualism: The Governance Dimension and Social Considerations . . . . . . . . . . . . . . . . . . . 28 (2) Sequencing of Reforms . . . . . . . . . . . . . . . 30 (3) Conditionality Issues . . . . . . . . . . . . . . . 31 B. Disbursements . . . . . . . . . . . . . . . . . . . . . . 32 C. Supervision . . . . . . . . . . . . . . . . . . . . . . . 33 VI. SUSTAINABILITY OF THE ADJUSTMENT EFFORT........ . . . . 34 A. Economic Performance and Sustainability Assessment . . . 34 B. Social and Political Issues. ........ . . . . . . . 34 VII. CONCLUSIONS AND LESSONS OF EXPERIENCE......... . . . . 36 A. Summary Assessment . . . . . . . . . . . . . . . . . . . 36 B. Lessons of Experience . . . . . . . . . . . . . . . . . . 36 ANNEXES I. KEY ECONOMIC INDICATORS: 1980-1990. ........ . . . . . . 39 II. MATRIX: GOVERNMENT PROGRAM FOR INDUSTRIAL SECTOR.... . . . . 40 PROJECT COMPLETION REPORT A. INTRODUCTION.............. ...... . . . . . 45 B. BACKGROUND............... ...... . . . . . 45 C. PROJECT PREPARATION AND APPRAISAL...... .... . . . . . 46 D. INSTITUTIONAL PERFORMANCE.......... .... . . . . . 48 Price Liberalization . . . . . . . . . . . . . . . . . . . . . 48 Public Investment Program . . . . . . . . . . . . . . . . . . 48 Export Promotion . . . . . . . . . . . . . . . . . . . . . . . 49 Import Liberalization . . . . . . . . . . . . . . . . . . .49 Investment Code . . . . . . . . . . . . . . . . . . . . . . . 49 Rehabilitation Component . . . . . . . . . . . . . . . . . . . 50 Technical Assistance . . . . . . . . . . . . . . . . . . . . . 50 E. MONITORING AND IMPLEMENTATION......... .... . . . . 50 Table of Contents (cont'd) Page No. F. DISBURSEMENTS................ ..... . . . 51 G. CONCLUSIONS AND LESSONS LEARNED....... .... . . . . 51 PART I I A. COMMENTS ON THE WORLD BANK'S PROJECT COMPLETION REPORT . . . . 53 B. REPORT PREPARED BY THE BORROWER AND EXECUTING AGENCIES . . . . 54 1. Comments by the Ministry of Economy and Plan . . . . . . 54 2. Comments by the Ministry of Industry, Energy and Mining . 55 3. Comments by the National Bank of Industry (BNI) . . . . . 57 PROGRAM PERFORMANCE AVDIT REPORT MADAGASCAR INDUSTRIAL ASSISTANCE PROJECT (IDA Credit 1541-MAG and SFA Credit A-7-MAG) PREFACE 1. This is the Program Performance Audit Report (PPAR) on the Industrial Sector Adjustment Credit (ISAC) to Madagascar, involving IDA Credit 1541-MAG and Special African Facility Credit (SFA) A-7-MAG. The IDA Cred"t, in the amount of SDR 40.2 million, was approved in January 15, 1985 and *-< t disbursement was on Septembcr 22, 1986; the Credit was closed on Septembe. 1990, three years behind schedule. The Special African Facility Credit, i ne amount of SDR 18.9 million, was approved on December 6, 1985 on an absence of objection basis. Last disbursement and closing dates are the same as for the IDA Crcdit. 2. The PPAR was prepared by the Operations Evaluation Department (OED) and the Program Completion Report (PCR) was prepared by the Industry and Energy Operations Division of the South Central and Indian Ocean Department of the Africa Region. The PPAR is based on the attached PCR, the President's Report, the loan documents, economic and sector reports, loan and country files, President's Reports of follow-up credits, the latest CEM Report No. 9101, "Madagascar - Beyond Stabilization to Sustainable Growth," and Report No. 7784,"Madagascar - Adjustment in the Industrial Sector and an Agenda for Further Reforms." OED staff interviewed present and former Bank who had been associated with the ISAC or with economic work in Madagascar. 3. A mission was scheduled to visit the country but had to be canceled owing to civil disorders. Therefore, some of thp PPAR conclusions, particularly those dealing with sustainability, may have to be reconfirmed later. OED expects to conduct a field visit in connection with the forthcoming audit of ITPAC (Credit Nos. 1834-MAG and A-32-MAG). 4. The PCR provides a good account and assessment of the program experience with regard to preparation and appraisal; institutional performance; achievements of the reforms; credit monitoring and implementation; disbursement issues; and draws lessons from the experience. The PPAR discusses trends in the Malagasy economy and the evolution of industrial policy; summarizes the status quo ante and the critical constraints in the industrial sector that led to ISAC evaluates the implementation record of agreed upon actions, including compl,.entary actions undertaken with the support of follow up credits; assesses the outcome and the impact of the program on macroeconomic performance and industrial supply response within the framework of the overall adjustment effort of the Malagasy Government; assesses the quality of the design and policy package and che supervision effort; analyzes disbursement delays; determines key factors influencing the effectiveness and sustainability of the policy reforms; and draws lessons from the program experience. 5. The draft PPAR was sent to the Borrower for comments but none were received. - 111 - PROGRAM PERFORMANCE AUDIT REPORT MADAGASCAR INDUSTRIAL ASSISTANCE PROJECT (IDA Credit 1541-MAG and SFA Credit A-7-MAG) BASIC DATA SHEET Amounts (SDR million) As of April 30, 1991 Oriqnal Disbured canceled Repaid OutstandIng IDA 15414AG 40.20 40.19 0.01 0.00 40.19 SFA A-7MAG 18.90 '8.00 0.00 0.00 18.90 TOTAL 59.06 59.06 0.01 0.00 59.08 Original Credit Dates Actual or Re-etmated kiteing Memorandum 29 July 1983 29 July 1963 Lar of Development Policy 30 Nov 1984 30 Nov 1984 Negonlloe Nov 1984 Nov 19O Board Approval 15 Jan 1985 15 Jan 1985 Sgrature 22 Mar 1985 DA: 13 Dec 1986 SFA 22 Mar 198 IDA: 13 Dec 198 SFA Carit Eeomiveness Mar 1986 28 Aug 1985 IDA; 3 Mar 198 SFA 2nd Trano Rlosse Jan 198 22 Sep 1S9 Credit Closing 30 Sep 98 30 Sep 1990 CUULATIVE CREDIT DISBURSEMENT FY85 FYa6 FY87 FYe8 FY89 FY90 FY91 IDA 1541AAG Planned 5.33 32.86 39.80 40.20 40.20 40.20 40.20 Acual 0.00 14.24 23.12 30.97 38.42 40.16 40.19 SFA A-7MAG Planned 0.00 10.40 16.00 18-90 18.90 18.90 18.90 Actual 0.00 3.54 14.27 16.51 18.82 18.90 18.90 Total Planned 5.33 43.26 55.50 59.10 59.10 59.10 59.10 Actual 0.00 17.77 37.39 47.48 55.25 59.08 59.06 AdualTotalsa%PlannedTotal 0.00 41.0 67.0 80.0 93. 99. 99.0 MISSION DATA No. of No. of Siir Dab of MonthlYear Wees Paort Weeks Report Preparation 1983 3 3 9 29 JL 1983 AppraibulJPosAppraisal Jan*Feb 1984 4 4 16 25 Apr 1964 Supervwsin NA NA 3 50 MIsN Complsion NA NA NA NA FOLLOW-ON ADJUSTMENT OPERATIONS MADAGASCAR- Indufby and Trade Adjustment Credit (Credits 1844MAG and A.32-MAG) approved an June 30, 197 In the amount of usse3 mi"n. - v - PROCRAM PERFORMANCE AUDIT REPORT HADAGA%SCA INDUSTRIAL ASSISTANCE PROJECT (IDA C^adit 1541-MAG and SFA Credit A-7-MAG) EVALUATION SUMMARY Introduction 1. This is the audit of the Indus- began fully only in 1987 and was sup- trial Sector Adjustment Credit (ISAC) ported by IMF Stand-by loans and World to Madagascar, involving IDA Credit Bank operations in the agricultural, 1541-MAG in the amount of SDR 40.2 trade, and public sectors (paras. million, approved in January 15, 1985 1.03-07). and the Special African Facility Cred- it (SFA) A-7-MAG in the amount of SDR PbJectives 18.9 million, approved on December 6, 1985. 3. The purpose of ISAC was to sup- port the Government's strategy of 2. Madagascar, with a population of bringing about meaningful policy chan- approximately 11.5 million and a per ges in the industrial sector designed capita income of US$230 in 1989, is to increase its efficiency and produc- among the 12 poorest countries in the tivity, while beginning the process of world. After facing a sharp economic establishing an environment conducive recession in 1980-1982, the Government to efficient import substitution and began in 1983 the implementation of export production. These policy chan- economic reforms. Macroeconomic bal- ges would constitute important first ances improved and there was a modest steps towards promoting the efficien- recovery in growth; during the 1983- cy, openness, and export-orientation 1987 period GDP grew at an average of of the Malagasy industrial sector so 1.4 percent per year, still below that it could increasingly contribute population growth (para. 1.01). Even to the country's balance of payments though the stabilization effort that and to the growth of employment ard began in 1982 served to restore inter- output (para. 2.04). nal and external equilibria, structur- al problems remained in place. In 4. The credit supported policy 1984, the Government began emphasizing reforms designed to increase the effi- the underlying supply-side constraints ciency and productivity of the indus- faced by the economy. The authorities trial sector, and the capacity utili- started implementing structural ad- zation of selected efficient industri- justment measures, including the first al enterprises so as to enable them to zteps toward a general liberalization produce goods for the domestic market of most economic activities with ac- and for export. The credit provided tions such as those taken under the foreign exchange for imports of inputs ISAC. In practice, however, the ad- and spare parts, rehabilitation equip- justment program's implementation ment, and technical assistance. The - vi - reform package consisted of (para. Support Project (EMSAP-1989), the 2.06): Financial Sector and Private Enter prise Credit (1990), and the Environ- (1) price liberalization, ment Program (1990) (paras. 2.04 and (2) public investment program (PIP) 4.01-04). The implementation record refoim, for the various measures included in (a) import liberalization, the ISAC and the outcome of ISAC with- (4) export promotion, in the overall adjustment program are (5) private sector encouragement. described in detail in paras. 3.03- 3.22 and are summn&rized below: IMplementation Experience and Outcome 7. Between 1983 and 1989, in the 5. The original credit effective- context of the overall adjustment ness date was March 1985, but formal effort, the Malagasy Government suc- effectiveness was delayed to August ceeded in re oving most price con- 1985 since more time was needed to trols. In the industrial sector, the await the actual hiring of ti': consul- price liberalization effort supported tants for the preparation of the ex- by ISAC, was completed in 1989 through port action program. Second tranche reforms implemented under ITPAC. disbursement was postponed from Janu- Currently, only prices of products and ary 1986 to September 1986 owing to services provided by state monopolies delays in the preparation of the ex- are still controlled (paras. 3.04-05 port promotion program. Because of and 4.05). major problems with the investment component managed by BNI, credit clos- 8. While under ISAC a small number ing was delayed by three years. The of import prohibitions were removed closing date was originally set for and replaced by tariffs and a few September 30, 1987, but was extended export-promoting measures were imple- first to September 30, 1988, then to mented, substantial distortions re- March 30, 1989. Eventually the credit mained in the trade regime and a gen- was closed on September 30, 1990 (pa- eral anti-export bias of trade and ra. 3.01). macroeconomic policy still existed. Thus, particularly in the exchange and 6. It would not be appropriate to trade area. ISAC provided oply the assess results and draw lessons for groundwork while ITPAC and PSAC set in ISAC individually. ISAC was only motion a far-reaching program to re- intended to pave the way in the direc- form the instruments of trade Rolicv. tion of adjustment as a first step toward a more comprehensive reform 9. On the import side, in January process. It was followed by the Agri- 1988, the Malagasy Government elimi- cultural Sector Adjustment Credit nated almost all non-tariff barriers (ASAC-1986), the Industrial and Trade on imports (except for 94 items, which Policy Adjustment Credit (ITPAC-1987), were health and security-related and the Public Sector Adjustment Credit were reduced to 77 in March 1989). (PSAC-1988), and a planned Private All tariff exemptions were eliminated Sector Environment Adjustment Credit and nominal and effective Rrotection (PRIVAC) which is scheduled to be was lowered. In January 1988 specif- negotiated when the political situa- ic tariffs were substituted by I tion in Midagascar stabilizes. Fol- valore ones (paras. 3.09 and 4.07). low-up credits were complemented by non-adjustment operations such as the 10. On the exRort side. the liber- Economic Management and Social Action alization orocess began in 1986-1987 - vii - and accelerated during 19., and by Major Issuel September 1988, except for vanilla, all administrative export requirements 13. Conditionality was manageable were reduced to a single customs' and appropriate to the objectives declaration (paras. 3.10-13 and 4.08). pursued by ISAC. However, there were In order to further advance export two important 4ifferences in the way promotion and to complement import two second tranche conditions were liberalization actions, the Government stated in the irowident's Report and acted also on the exchange rate policy the legal conditions. First, while front and by 1989, the real effective the President's Report stated the exchange rate was approximately 40 second tranche condition as "decontrol percent of its 1982 --ye (paras. ^.14 of all industrial prices", the Devel- and 4.09). The cor- c7on of the FMG opment Credit Agreement specified overvaluation crea , x che conditions "free ex-factoXy prices." This gave fcr an overhaul of the foreign ex- the Government an option to maintain change re,ime and by July 1988 the controls on profit margins (para. Government widened the foreign ex- 3.04). Clearly, the spirit of the change policy reforms with the intro- condition was not met. Second, while duction of a full-fledged Geaeral the President's Report asked for the License System (OGL) through which iMRlementation of a satisfactory ex- unlimited foreign exchange became port incentive system, the legal con- available at the prevailing exchav e dition specified only the need to rate for imports of goods by any eco- adoRt a program of actions to promote nomic agent (para. 4.10). exports (para. 5.07). 11. While ISAC was successful in 14. SDR 5 million of the credit was improving the quality of the projects to be lent to the BNI for the financ- included in the 1985 and 1986 indus- ing of investment (rehat'.litation) trial PIP, in cutting the size of the projects through sub-loans to invest- program, and improving MIEM's project ment enteLprises. The BNI component appraisal capabilities, a more compre- was the least successful and BNI's hensive effort to rationalize the untenable financial situation was the overall national public investment main reason behind implementation program and to pay increased attention delays and three year delay in credit to efficiency and the social aspects closing (paras. 3.18-21). At the time of adjustment was pursued under ASAC, of ISAC serious constraints on finan- ITPAC. and PSAC. cial intermediation existed, of which the staff were not aware (para.2,02). 12. In the area of private sector Problems and delays in the implemen- encouragement, even though an invest- tation of the BN1 component of the ment code prepared under ISAC rep- credit and credit shortages consti- resented a subsuantial departure from tuting a key constraint to economic past negative attitudes towards the recovery, were caused by financial private sector and foreign investment, sector liquidity problems linked to it was not timely and flexible enough the need for public enterprise reform. to keep up with the changing economic The situation of the commercial banks' environment. Thus,in December 1989. portfolios, particularly of BNI, be- to attract private domestic and for- came known only during ISAC's imple- eian investment, the Government adopt- mentation thanks to effective supervi- ed-a newpinvestment code. sion. The financial health of the - viii - banking sector continued worsening up additional analytical work and gather to 1987, when JTPAC and PS&C began additional information and experience addressing both financial sector and to develop a more comprehensive ad- public enterprise issues which had jusement program. been left out of ISAC (paras. 4.14- 15). 16. The suRervision Lffort was Ade- guAte. Supervision was imaginative 15. A general design issue deals and understanding, and findings of with the appropriateness of a phased ISAC supervision missions were instru- sectoral approach as opposed to an un- mental in identifying outstandi%g front comprehensive adiustment opera- issues and thus helpful in designing tion. The gent3is of the ISAC as a subsequent operations. However, there first step on the road of adjustment, were deficiencies in the loan files, and its limited scope, derived from including a missing Implementati.n two considerations linked to the Summary (Form 590). Moreover, all 590 Malagasy's authorities political and forms were issued at least two months social constraints at the time. after the end of the supervision mis- First, even though the socialist Gov- sion involved. Probably due to these ernment was committed to reform in deficiencies, the PCR does not contain principle, it was not at the time details on the supervision effort. convinced of the "necessity or desir- According to a recent OED study on SAL ability of a comprehensive reform supervision and monitoring, deficien- program." Bank staff hoped that suc- cies in historical records appear to cessful implementation of ISAC would be a common problem which hampers the convince the Government to implement a maintenance of the Bank's institution- more vigorous reform package. Even al memory and thorough ex post evalua- though at the time of -SAC preparation tions of the operations concerned. governance issues were not yet explic- Finally, updates on the iMRlementatio itly considered, the governance dimen- of the TA coMonent were not adeguate. sion did affect both the design of ISAC and economic performance under ustainability the overall program. The Malagasy government in 1985 was not prepared to 17. The maintenance of macroeconomic risk implementing a comprehensive SAL stability and continued politi.1 type operation. However, the Govern- commitment and popular support for the ment felt secure enough to undertake reform process are the key determi- ISAC which was a first step in the nants of the sustainability of adjust- reform process but already contained ment measures. Economic nerformance measures which might have been unpopu- in the near future will be closely lar. Second, the country's economic tied to political economy consider- team was also attempting to strike a ations; recent political developments balance in the pace of policy reforms and the fragility of the current re- in order to improve efficiency in the gime have pushed the governance issue industrial sector without causin, to the forefront of any assessment of severe economic dislocation and possi- sustainability of the Malagasy adjust- ble social unrest. In conclusion, the ment effort (para. 6.01). phased approach was appropriate to the circumstances. First, positive re- 18. Notwithstanding political up- sults under ISAC stimulated heavals, the political commitment to a government's confidence and commitment continued adjustment effort seems to to the reform process. Second, it remain strong in 1991. On the econom- permitted Bank staff to carry out ic front both sides are in agre-ment - ix - on the need to continue liberalizing ation with close monitoring of credit the economy and eliminating remaining policies and domestic savings perfor- structural constraints to growth with mance. On the political side, commit- equity. In conclusion, if the politi- ment to reforms must continue to ex- cal situation does not interfere with ist, whichever political group remains economic processes, all the elements in power (para. 7.02). for sustained growth are in place and an optimistic assessment of the sus- 21. Th- review of the experience tainability of the overall adjustment with the ISAC offers instructive les- effort is possible (paras. 6.02-05). sons and suggestions which may provide useful insight for future Bank opera- Conclusions and Lessons of Exserience tions: 19. ISAC was essentially preparatory 0 In 1985 the Government was not to the more comprehensive reforms ready yet to implement a broad-based implemented in follow-up sector opera- structural adjustment effort. Thus, tions. ISAC was on the whole wel policy-based sector specific lending designed as a preparatory loan and the was used by the Bank as vehicle to Government participated in the devel- open dialogue with the authorities. opment of the conceptual framework of The phased approach was successful in ISAC as well as of the following sec- leading to full-fledged adjustment toral loans. Given the political beginning in 1987 thanks to the tnvironment at the time -- with a Government's commitment to the reform socialist regime skeptical of market process and to the appropriateness, mechanisms and a full-fledged liberal- timeliness, and, effectiveness of ization of the economy, and fearful of follow up operations. A hased sec- the political and social consequences toral androach might be necessary and of a full blown adjustment effort -- aRRropriate when a Governgent is not the preparation of phased sectoral ready for an u front comrehensive loans as opposed to the implementation adustment oReration. since the pro- of an up front adjustment loan was the cess itself creates a olicy dialogue appropriate choice (para. 7.01). which leads to refors 20. Overall the adjustment effort 0 During ISAC design there was implemented by the Malagasy Government awareness by Bank staff of the Gove- has been impressive and it is general- rnment's political constraints, which ly perceived as one of the success led to the design of a successful stories within Africa. However, the sequencing of sector adjustment opera- sustainability of the adjustment pro- tions, through an intense policy dia- cess will depend on macroeconomic as logue aimed at building a consensus. well a political developments. On The success of Madaascar's adustmint the macroeconomic area, certain condi- effort shows that. even under a so- tions must be maintained for the stru- cialist regime. olic reforms can be ctural reforms to have a lasting im- successful and sustained when the pact. Prospects for medium-term via- design ofthe rogram is internalized bility would be enhanced by a more and there is tolitical will, consensus aggressive exchange rate policy per- and commitment on the efart of the mitting the continuation of export authorities. diversification; by further import liberalization to enable a sustained lA nosteriori it is clear that a rise in imports to maintain the recov- more complete analysis of the finan- ery; by improving financial intermedi- cial sector at the time of ISAC's design would at - and not t( mentation - on 1 BNI as the implem rehabilitation cc it. Thus, the intermedi on lvzed in more d reform measures included in ISAC. * The Malagas witnessed an e: with the IMF ane Stand-bys and s especially on ex However, it appei were slow in kee exchange rate a an exchange ral liberalization e: by Bank operatiot * Even thou effort was ade PROGRAM PERFORMANCE AUDIT REPORT MADAGASCAR INDUSTRIAL ASSISTANCE PROJECT (IDA Credit 1541-MAG and SFA Credit A-7-MAG) I. BACKGROUND A. Historical Background of the Malagasy Economy 1.01 Madagascar, with a population of approximately 11.5 million and a per capita income of US$230 in 1989,11 is among the 12 poorest countries in the world. Agriculture accounts for about one third of GDP, industry accounts for 13 percent of GDP. The main staple is rice. Coffee, vanilla, and cloves are the main traditional exports, accounting for approximately 50 percent of export earnings. In the period from independence (1960) to 1972 Madagascar experienced a modest average GDP growth rate of approximately 2.7 percent per year. In the 1970-78 period real GDP grew on average only by 0.6 percent per annum and, given an average population growth rate of 2.8 percent per annum, by 1978 GDP per capita was almost 20 percent below the 1970 level. A sharp deterioration followed in 1980-1982, with a 9.7 drop in GDP in 1981. After 1983, with the beginning of the implementation of economic reforms, macroeconomic balances improved and there was a modest recovery in growth. During the 1983-1987 period GDP grew at an average of 1.4 percent per year, still below popri1ation growth. Adjustment efforts began in 1985 with the Industrial Sector Assistance Project (ISAC) under review. The tirst large-scale macroeconomic adjustment program was launched beginning in 1987 and was supported by IMF Stand-by loans and World Bank operations in the agricultural, trade, and public sectors. 1.02 Given recent political developments in Madagascar, it is important to provide a brief summary of political history since independence. Madagascar experienced a peaceful transition to democracy when - after De Gaulle's return to power in 1958 - the Malagasy population voted in a referendum to become an autonomous republic within the French post-colonial system. In 1960 Philibert Tsirabnana was elected the first Malagasy president and remained in power until 1972, when as a result of the deteriorating economic situation and massive anti- governmental demonstrations, Tsirabnana resigned handing power to his army commander, General Gabriel Ramantsoa. The new president pursued a close relationship with Eastern Bloc countries and after several failed coup attempts relinquished power in 1975 to Colonel Richard Ratsimandrava, who was murdered soon thereafter. Later that year Didier Ratsiraka, the foreign minister and navy commander, took power and still remains the head of Government to this date. Political developments in the summer of 1991 in Madagascar have shown the fragility of President Ratsiraka's position and the growing popular dissatisfaction with the current administration. 11 World Bank Atlas methodology. - 2 - 1.03 As emphasized by a recent Bank study, development policies followed from independence to 1983 led to the stagnation of agriculture, in "an attempt to shift to accelerated, mainly domestic market-based, industrialization, and expansionary macroeconomic policies."- This strategy resulted in escalating fiscal imbalances and unsustainable current account deficits in the balance of payments after 1977-1978. In particular, economic policies pursued after the 1972 military coup account for Madagascar persistent difficulties since the late 1970s. The Government which took office in 1972 emphasized state control of the economy: inward-looking policies aimed at achieving self-sufficiency in industry and agriculture were implemented, most of the larger private companies were nationalized, interventionist practices characterized by price controls and administrative regulations were adopted, and direct or indirect state control over the greater part of agricultural marketing was established. The Government which came to power in 1975 escalated, broadened, and codified such policies. GDP declined and Madagascar went from being a country self-sufficient in food to being an importer of rice, its main staple. 1.04 While through 1977 fiscal and balance of payments policies remained cautious, and public external debt was kept at a low level, ' beginning in 1978, in the face of stagnating economic activity, the Government tried to diversify the economy and accelerate development through a large expansion in public sector investment. The higher rate of public investment was financed through external borrowing and money creation. The volume of domestic capital formation increased by 74 percent in 1979. By 1980 it was about 60 percent above the average level of the 1975-1978 period; the share of gross domestic investment - mostly public - to GDP rose from 9 percent in 1978 to 15 percent in the 1979-1980 period. According to a Bank report "many of the projects selected were economically and financially non-viable; they therefore made little contribution to GDP, exports, or debt servicing capacity." 1 1.05 A dramatic deterioration in the balance of payments and the fiscal accounts ensued. Imports increased by almost 40 percent (in volume terms) in 1979 and with sluggish exports and deteriorating terms of trade, the external resource gap increased from about 5 percent of GDP in 1978 to approximately 16 percent in 1979 and 1980, while the exchange rate was becoming increasingly overvalued. The level of imports largely exceeded Madagascar's financing possibilities and, by 1980, arrears on external payments began accumulating and imports started decreasing. Since a high proportion of investments were undertaken within the public sector, the overall budget deficit rose from approximately 4 percent of GDP in 1978 to double digits in 1980-81. Inflation 1i Report No. 7784-MAG, Madagascar - Adjustment in the Industrial Sector and an Agenda for Further Reforms (Report 7784), October 29, 1990, para. 15. 1t PR No. P-4488-MAG, Industry and Trade Policy Adjustment Program (ITPAC-PR), June 5, 1987, para. 1.04. 1 Ibid., para. 1.05. -3- rose to 30 percent per year in 1981-1982.11 The Government, moreover, was increasingly relying on external resources for budget deficit financing aostly borrowing from commercial non-concessionary sources. Debt-service obligations rose from about 4 percent of export earnings in 1978 to over 70 percent in 1982. By 1982, real per capita GDP had fallen by an estimated 28 percent from its 1973 level. 1.06 In 1982, the Government began implementing a set of reforms, supported by IMF stand-by agreements, aimed at achieving stabilization by restraining aggregate demand. Madagascar's stabilization efforts included tax-increasing measures, expenditure c ts and an improved expenditure control system, bank credit ceilings to reduce inflationary pressures, increases in prices of public enterprises' goods and services, a sharp reduction in imports, and most importantly, specific actions on the exchange rate and consumer subsidies. The Malagasy Franc (FMG) had remained fixed at 50 FMG per French franc until 1982, even though Madagascar had left the French franc zone in 1973. In 1982, the FMG was devalued and pegged to a trade-weighted basket of currencies. Beginning in 1983 a flexible exchange rate policy was adopted and the quarterly rate was adjusted in line with domestic inflation, thus achieving a substantial real depreciation of the FMG. The nominal price of rice distributed through official channels more than doubled and consumer subsidies on imported rice were eliminated. As a result of the stabilization measures implemented the Government deficit was reduced from 15 percent of GDP in 1980 to 7 percent in 1982 and 5.4 in 1983. The resource gap was also cut down from over 16 percenit of GDP in 1980 to 9 percent in 1982 and 7 percent in 1983. Fiscal and monetary tightening contributed to reducing annual average inflation from over 30 percent in both 1981 and 1982, to 19 percent in 1983 (See table in Annex 1). 1.07 Even though the stabilization effort began in 1982 was quite successful in restoring internal and external equilibria, structural problems remained in place. In 1984, the Government began emphasizing the underlying supply-side constraints faced by the economy. The authorities started implementing structural adjustment measures, including the first steps toward a general liberalization of most economic activities with actions such as those taken under the ISAC. In practice, however, the adjustment program's implementation began fully only in 1987.6i B. Evolution and Features of the Industrial Sector 1.08 As indicated earlier, after independence the Malagasy authorities sought to diversify the structure of national production through the encouragement of industry. Their objective was to reduce the country's vulnerability to adverse fluctuations in agricultural commodity prices accompanied with the belief that industry would promote faster economic growth 11 Report No. 5454-MAG, Madagascar - Current Economic Situation and Prospects, October 25, 1984, paras. 5-7. i Report No. 9101-MAG, "Madagascar - Beyond Stabilization to Sustainable Growth", June 1991 (CEM 1991): para 21. -4- and increased employment opportunities. Three main themes emerged in Madagascar's industrial policy: * The reliance on an import substitution strategy to achieve rapid growth in manufacturing output. The main policy instrument was protection of domestic production from external competition. Originally, protection mechanisms were tariffs and import taxes. In the late 1970s, the Government began imposing import prohibitions, strict import licensing, and foreign exchange rationing; * A widespread nationalization of economic activity after 1972. At the time of their establishment, during the 1960s, most large manufacturing firms were owned by foreign private investors. Medium and small enterprises were mostly owned by local entrepreneurs. Beginning in 1972, however, banks, insurance companies, and large trading concerns were nationalized and the Government took over the marketing of key agricultural products. Throughout the 1970s the Government increased its involvement in the manufacturing sector; at first, by taking over foreign owned firms, later the Government brought under its control a large part of domesticall owned industry through nationalization or equity participation.- In 1982, out of a total of 362 industrial enterprises, 40 were entirely state owned. The Government had a majority interest in other 17 firms and a minority holding of at least 33 percent in another 14. These included most large scale operations; medium and small-scale firms remained mainly in private hands. Enterprises in which the Government had an interest of more than 50 percent accounted for an estimated 65 percent of sectoral output and a similar share of employment. Another 15 percent of output was contributed by firms in which the Government had a substantial minority interest;$' * The adoption by the Government of an interventionist attitude towards the economy. This trend was characterized by the intensification and expansion of central regulation and a corresponding reduction in the role of market forces in allocating resources, e.g., price controls were generalized and exchange controls were tightened. 1.09 As a result of these policies, the Malagasy manufacturing sector evolved with the following features: * Strong dependence on imported inputs and spare parts and, as a consequence. vulnerability to foreign exchange shortages: this dependence resulted from low production of intermediate goods due to weak inter-industry linkages. The high protection levels i PR, paras. 28-30. ' Ibid., para.30. - 5 - enjoyed by producers of consumer goods, and exemptions from tariffs and import taxes on investment and intermediate goods, discouraged the use of and/or investment in domestically produced inputs. IT, 1983, imports of intermediate goods (excluding petroleum) accounted for 25 percent of total imports, by 1986 intermediate goods represented over 30 percent of total imports (Annex 1). * High protection levels and administrative allocation of foreign exchange: traditionally custom duties and import taxes constituted the main vehicles of protectionist policy. In the early 1980s, due to worsening foreign exchange shortages, imports began to be subject to increasingly tight import prohibitions. This period was also characterized by quotas and administrative allocation of foreign exchange. * Low level of diversification: the two most important sub-sectors, food processing and textiles, accounted for more than 50 percent of manufacturing value added; beverages, cigarettes, garments, and leather goods accounted for an additional 20 percent of manufacturing value added. Other sub-sectors included household chemicals, construction materials, petroleum refining, and intermediate goods such as metal and paper products; * Heavy emphasis on the production of consumer goods: developments in the manufacturing sector were related to the import-substituting strategy implemented by the Malagasy authorities during the 1971s; thus, under a high level of protection from foreign competition, the domestic manufacturing sector became the supplier of many final consumer goods. Of 33 major products manufactured in Madagascar, 22 were consumer goods (food and tobacco products, garments and footwear) and simple household chemicals (soap, matches, batteries and candles), five were petroleum products, and only six were intermediate goods (pulp, paper, metal sheets, cement, nails and paints);!/ * Low levels of manufacturing exports: manufacturing exports - consisting mostly of sugar and molasses, cotton fabrics, essential oils, canned meat, and garments - accounted in 1983 for 15 percent of total exports and only 5 percent of manufacturing output; in 1986 manufactured exports represented only 13 percent of total exports, from a peak of 20 percent in 1975; * Inefficient production: as stressed by a 1986 Bank Report, a strong indication of the inefficiency of the sector was given by the finding that for 32 of 80 firms analyzed in 1985 (accounting for approximately 70 percent of industrial value added) the value of tradeable inputs (excluding capital) exceeded the value of output produced by the firm, both calculated at border prices. Thus, gi Ibid., para. 23. - 6 - these firms used up more foreign exchange than they ei-her saved (by import substitution) or generated (by exporting)." 01 1.10 Industry, which accounted for around 4 percent of GDP in 1960, reached approximately 15 percent of GDP in 1970 and, on average, remained at that level throughout the 1970s. Between 1970 and 1979 industry's growth rate fluctuated widely, and on average industry grew by only 0.8 per annum in real terms. Industry declined every year during the 1980-1982 period, with a sharp fall (22.5 percent) in 1981. Thus, during the first half of the 1980s, the share of industry in GDP declined from 14.3 percent in 1980 to around 11 percent in 1984 and 1985 (Table in Annex 1). The Malagasy manufacturing sector accounted for approximately two thirds of the industrial value added (excluding construction); according to a 1982 survey there were at the time 362 manufacturing enterprises, of which 100 accounted for 80 percent of output. II. CONCEPT AND DESIGN OF THE ISAC A. Industry Constraints at the Time of the ISAC 2.01 As discussed in chapter I industrial production fell by 22.5 percent in 1981 and 14.4 percent in 1982. The 1980-1982 drop in manufacturing production resulted from the fact that, after 1979, the emergence of a severe foreign exchange constraint (i.e., lack of imported inputs) and the deteriorating economic situation (i.e., lack of domestic demand) forced a great number of companies to slow down or stop production. These events brought to the surface the structural inefficiencies of the sector, showing that development following past industrial policies was unsustainable. 2.02 The diagnosis of the Malagasy industrial sector in 1984, when the ISAC was designed, focused on the following priority issues: * Heavy under-utilization owing to (a) acute shortages of domestically supplied agricultural raw materials and (b) lack of imported inputs and spare parts due to foreign exchange shortages and administrative allocation of foreign exchange. Between 1979 and 1982 imports of raw materials and spare parts dropped by approximately 40 percent in real terms. Imports of capital equipment were tied to on-going investment projects and external financing arrangements and could not be quickly reduced, and imports of food, particularly rice, actually increased thus, the most affected by the cuts were imports of raw materials, spare Loi ITPAC-PR, para. 2.08. 1 Ibid., para. 2.02. -7- parts, and non-food consumer goods. L21 Average capacity utilization in a number of firms, accounting for approximately three-quarters of manufacturing value added, dropped from about 75 percent in 1979 to 57 percent in 1981 and to 47 percent in 1982.-i Moreover, there were substantial falls in the output of some basic consumption goods; soap and matches production fell by more than 50 percent, candles and corrugated metal sheets by over 30 percent;1 * Import restrictions. Industry in Madagascar was highly protected through custom duties, import taxes, import prohibitions, and administrative allocation of foreign exchange. Import duties had been generally low and uniform with only four rates (0, 5, 10, and 15 percent), while import taxes had been relatively high and had varied over a wide range up to 100-150 percent. In several cases, finished industrial products had the same or even lower duty rates than raw materials and/or intermediate goods used as inputs. Import taxes were relatively high and varied over a wide range of up to 150 percent. Import taxes were generally low on raw materials, higher on intermediate goods, and in the upper range for most finished products. On balance, then, average tariff protection (including all taxes and duties) was, in the mid-1980s, at 112 percent for consumer goods, 37 percent for intermediate goods, and 34 percent for 40 capital goods,- not taking into account the impact of import prohibitions. At the time of the ISAC (i.e., 1984) all imports - except those needed by manufacturers to produce exports - were subject to import restrictions and were classified in two categories: (1) those which were prohibited, and (2) those subject to a detailed annual licensing program. Even though at the time of ISAC design no effective protection estimates were availabl1, this system of import controls and quantitative restrictions, combined with the existing structure of import taxes, granted a hiFh level of effective protection to import substitution industries; * Price controls. Even though price controls had existed in Madagascar since the 1960s, their use was intensified and generalized in the late 1970s in an attempt to control the pricing behavior of protected local industry. In the early 1980s, controls were extended to all industrial goods to defuse inflationary i Report No. 5154-MAG, Madagascar - Current Eccnomic Situation and Prospects, October 25, 1984, paras. 10, 31. .i PR, para. 21. Li Ibid., para. 30. i Ibid., para.45 and ITPAC-PR, para.2.04. Li Ibid., paras. 45-46. - 8 - pressures and guard against the possibility of excessive price increases in the absence of competitive markets. Any price increase had to be authorized by the Government on the basis of cost plus a set margin for each sub-srsctor; authorizations were given approximately one month after the application. The system was complex, it aggravated shortages and distortions, impeded industrial efficiency, discouraged output increases, and in some cases had affected the financial health of manufacturing firms.L' Moreover, Government attempts to hold selling prices below market- clearing levels had a negative impact on both the agricultural and industrial sectors and led to increased imports at a time of balance of payments crisis. Finally, price controls, together with the fall in demand, made it increasingly difficult to pass on rising costs due to low capacity utilization, despite the fact that most enterprises were enjoying a high level of protection from foreign competition * Ineffective public investment program in industry (PIP). Poor project selection was a major problem particularly after the 1978- 1980 investment boom. A number of large scale investment operations (e.g., Zeren urea plant, Mamisoa vegetable oil factory, oil-palm refinery, and several tanneries) were undertaken without a careful assessment of their feasibility or likely returns. At the time of ISAC appraisal, out of 21 industrial projects under execution, 16 had been inadequately prepared, were likely to face significant technical and financial problems, and their viability was doubtful; * Export incentives and regulations. Up to 1982-1983 Government policies were repressing rather than encouraging exports due to the seriously overvalued exchange rate and the implementation of an array of procedural impediments. Export incentives available under an ordinance enacted in 1973 had never become operational for lack of specific legislation required for its implementation. Moreover, the regulatory side of the ordinance had restricted freedom of exporting, e.g., each exporter had to hold an exporter's card and each export shipment had to be authorized; * Investment incentives limitations. The Investment Code enacted in 1973 had a series of limitations: (1) lacked precision and was open to widely different interpretations; eligibility criteria were vague and the nature and extent of benefits were decided upon by the Government without apparent consistency; (2) several of the main incentives were made redundant by other Government regulation, U Ibid., paras. 47 and 54. Report No. 5996-MAG, The Democratic Republic of Madagascar Country Economic Memorandum, March 18, 1986, Appendix B, para. 32. .1 PR, para 33. -9- e.g., high protection through import taxes made some investments profitable even without tax relief, meaning that the Government might have been unnecessarily forgoing fiscal revenues; (3) gave greater advantages to new credits as opposed to investments for expansion, rehabilitation, or modernization; (4) included generous accelerated depreciation provisions, and easy access to tax-free imported equipment, favoring capital intensive techniques, which was not consistent with the employment-generation objectives of the Government,Li and * Constraints on financial intermediation and public enterprises' issues. Madagascar's financial system comprised the Central Bank of Madagascar, three national commercial banks specialized by sectorU - of which the State was the sole shareholder, a postal checking system, two insurance companies, and an investment fund. The financial situation of the three commercial banks was untenable and their portfolios were in need of major write-offs. This was mainly a consequence of their funding of public enterprises which had fallen into serious financial difficulties and had been able to survive thanks to credit from the three state banks (crowding out the private sector). By 1985 the Government controlled 167 companies. The majority of these enterprises received Government subsidies and suffered from managerial, technical, and financial problems. After the 1983 stabilization program sharply reduced subsidies, the banking system kept these enterprises alive at the cost of building up a very high percentage of non-performing loans. 22 B. The New Industrial Development Strategy 2.03 Even though stabilization efforts initiated in 1981-1982 were relatively successful in restoring internal and external equilibria, underlying structural problems remained in place. Moreover, given the fiscal adjustment already achieved in 1982 and 1983, the Government felt that further budget reductions were likely to be slower and their balance of payments impact smaller. Therefore, starting in 1984, the Government, with the support of a fourth IMF stand-by arrangement, emphasized the liberalization of pricing and marketing arrangements in agriculture to improve production and export performance. At the time, the Government also recognized that "the revitalization of inustry on an efficient basis would make an important contribution to overall economic recovery."Li The Government's strategy for achieving these industrial policy -' Ibid., paras. 48-49. 1- The National Bank for Industry (BNI), the National Bank for Agriculture (BTM), and the National Bank for Trade (BFV), created and nationalized in 1977 as a result of the consolidation of the existing commercial and development banks. L CEM 1991, paras. 68-69 and 84. Ul PR, para. 52. 1 10 - objectives involved the adoption of actions in three policy areas: (1) raise the level of capacity utilization of the sector to promote production of consumer goods and exports; (2) undertake initiatives aimed at improving efficiency and resource allocation in the public sector; and, (3) provide foreign exchange to efficient enterprises to import inputs necessary for a higher level of industrial production. C. Objectives. Policy Instruments, and Conditionality 2.04 The purpose of ISAC was to support the Government's strategy of bringing about meaningful policy changes in the industrial sector designed to increase its efficiency and productivity, while beginning the process of establishing an environment conducive to efficient import substitution and export production. These policy changes would constitute important first steps towards promoting the efficieticy: openness, and export-orientation of the Malagasy industrial sector so tI.; Zt could increasingly contribute to the country's balance of payments are -t the growth of employment and output.L4 ISAC was the first in a series )f sequenced sector adjustment operations. It was followed by the Agricultural Sector Adjustment Credit (ASAC-1986), the Industrial and Trade Policy Adjustment Credit (ITPAC-1987), the Public Sector Adjustment Credit (PSAC-1988), and a forthcoming Private Sector Environment Adjustment Credit (PRIVAC) which is scheduled to be negotiated when the political situation in Madagascar stabilizes. Follow-up credits were complemented by non-adjustment operations such as the Economic Management and Social Action Support Project (EMSAP-1989), the Financial Sector and Private Enterprise Credit (1990), and the Environment Program (1990). 2.05 The ISAC was jointly financed by IDA (SDR 40.2 million) and a supplemental credit from the Special Facility for Sub-Saharan Africa (SFA) (SDR 18.5 million). The credit supported policy reforms designed to increase the efficiency and productivity of the industrial sector, and the capacity utilization of selected efficient industrial enterprises so as to enable them to produce goods for the domestic market and for export. The Credit provided foreign exchange for imports of inputs and spare parts, rehabilitation equipment, and technical assistance. 2.06 The reform package consisted of: (a) price liberalization, intended to make the Malagasy economy more responsive to market forces so as to increase its efficiency, dynamism, and productivity through the progressive elimination of price controls on most industrial goods; (b) public investment program (PIP) refor, to rationalize the PIP process through the establishment of criteria for project selection and the strengthening of project appraisal capability; (c) import liberalization, to improve industrial efficiency through greater competition from abroad and to encourage exports by eliminating import prohibitions on 20 percent of products; (d) export promotion, to increase the outward-orientation of the Malagasy economy and as a way to break out of the foreign exchange constraint holding back the recovery and long-term development of the industrial sector through the adoption of measures designed to stimulate exports, including the elimination of all export taxes, and (e) private sector 1i Ibid., para. 77. - 11 - encouragement, to foster an active role of the private sector in the manufacturing sector through the elaboration of a new investment code. 2.07 Concerning exchange rate management, there was close collaboration with the IMF. In the early 1980s, the overvaluation of the Malagasy franc required establishmen4. of strict exchange controls, quantitative barriers to imports, and administrative allocation of foreign exchange. Beginning in April 1982 - date until which it was pegged to the French franc - the FMG became pegged to trade- weighted baiket of currencies and devalued several times. In 1984 the Government adopted a flexible exchange rate policy wh ih led, by 1986, to a 58 percent nominal depreciation and to a 29 percent depreciation in real effective terms.L1 Thus, at the time of ISAC design, overvaluation did not appear to be a major issue since the real exchange rate stood at approximately the same level as in 1978, a Tear judged by Bank staff as one of reasonable external balance for the country.- However, Bank staff acknowledged that given the 1985 terms of trade deterioration and increasing debt service ratios, further real devaluations might be required. At the time of ISAC an exchange rate management strategy for 1984-1985 was developed by the Government, with the help of IMF staff calling for a depreciation of the FMG at a pace equal to domestic inflation. 27 2.08 The credits provided foreign exchange (SDR 33.5 million from the IDA credit and all of the SFA credit) to efficient industrial enterprises for raw materials and spare parts imports to increase their capacity utilization and promote the production of exports and consumer goods. The project included an investment component (SDR 5 million) for the financing of equipment for the rehabilitation of sejected enterprises to be channeled through the National bank for Industry (BNI). The Credit also included a technical assistance component (TA) (SDR 1.5 million) which provided technical support to the Ministry of Industry, Energy and Mining (MIEM). The TA component would provide technical support for many of the policy reforms the Malagasy authorities were planning to introduce in later years. The TA component would achieve the following objectives: (1) improve data collection, processing, and analysis at the MIEM; (2) reinforce the Ministry's capability to analyze investment projects; (3) prepare an action program for export promotion and import liberalization and reinforce the MIEM's capability to analyze and formulate policies in the area of trade and incentives; and, (4) assist the Ministry in upgrading and refurbishing an existing building to serve as its headquarters. 2.09 The proceeds for importation of inputs would be disbursed in two tranches conditional on sat? Zactory implementation of the following actions under ISAC, while disburseme. -s under the rehabilitation and technical assistance components were linked only to progress in the implementation of these two programs. L GEM 1991, para. 36. EU PR, para. 18. L7 Idem. - 12 - (1) Actions to be taken before the release of the first tranche (credit effectiveness): (a) Remove import prohibitions on five percent of the goods whose importation is currently prohibited and make foreign exchange available for their importation; (b) Agree % the size and content of the 1985 Public Investment Program in Industry; (c) Hire consultants for the preparation of the export promotio and import liberalization action programs; (d) Eliminate export taxes for all manufactured goods and exonerate exporters of manufactured goods from all indirec taxes on raw materials and spare parts used as inputs o export productio.t. (2) Actions to be taken before the release of the second tranche: (a) Decontrol prices of all industrial goods with the exceptio of 31 products (i.e., 30 percent of value added); (b) Agreement on the size and composition of the 1986 Industria PIP; (c) Apply EPI system to all exporters of manufactured goods; (d) Implement a satisfactory export incentive system; (e) Submit the new investment code to the National Assembly follwing consultation with the Bank; (f) Remove import prohibitions on a further 15 percent of the goods whose importation is currently prohibited and make foreign exchange available for their importation. A policy matrix of the ISAC to Madagascar is presented in Annex II. The matrix lists issues to be addressed in all the areas of the reform package, describes the measures which were to be taken under the ISAC, and the implementation record of such measures. III. IMPLEMENTATION EXPERIENCE AND OUTCOME 3.01 The original credit effectiveness date was March 1985, but formal effectiveness was delayed to August 1985 since more time was needed to await the actual hiring of the consultants for the preparation of the export action program (para 3.12). Second tranche disbursement was postponed from January 1986 to September 1986 owing to delays in the preparation of the export promotion - 13 - program. Because of major problems with the investment component managed by BNI (paras. 3.18-21), credit closing was delayed by three years. The closing date was originally set for September 30, 1987, but was extended first to September 30, 1988, then to March 30, 1989. Eventually the credit was closed on September 30, 1990. 3.02 This section discusses the implementation experience of the various components of the credit and provides a preliminary assessment of outcome of the ISAC policy package in itself. Chapter IV provides a more comprehensive assessment of performance in the areas included in ISAC within the overall adjustment program of the Malagasy Government. A. Policy Package (1) Price Liberalization 3.03 The Government had agreed to free prices of manufactured goods represanting 70 percent of value added in manufacturing before the release of the secono tranche of the ISAC. At the time of Board presentation the Government had already freed prices of goods representing 35 percent of manufacturing value added. By July 1986 an additional 35 percent of ex-factory prices of goods had been decontrolled, leaving only the prices of 31 industrial products (approximately 30 percent of manufacturing value added) under control. By September 1986, the Government had further reduced the number of goods under price control to ten, representing less than 15 percent of manufacturing value added." 3.04 Notwithstanding the apparent successful implementation of price liberalization measures under ISAC, in reality, until late 1987, the Government freed only ex-factory prices. However, it continued to control retail prices and profit margins,- adding a fixed margin to costs. This system tended to defeat, at least in part, the purpose of the liberalization effort. These controls were maintained with the aim of curbing inflationary pressures resulting from the protection of monopolistic local industry existing in the import- substitution environment." 3.05 Since the implementation of ISAC, significant piogress has been made in price liberalization. By late 1986, the number of goods under price control was further reduced to ten. In 1987, all profit margin controls were eliminated by the Government as one of ITPAC's effectiveness conditions, even though maximum price guidelines remained on all but 16 goods. All profit margins controls were eliminated between 1987 and 1989. Eventually, in February 1989, all price Li Memo September 18, 1986, para. 6. !i Regarding this issue, while the President's Report stated the second tranche condition as "decontrol of all industrial prices", the Development Credit Agreement specified "free ex-factory prices". The implications of this difference are discussed in Chapter V, para. 5.07. Lt Draft ITPAC-PCR, para.23. - 14 - controls were removed and a free market system was achieved for the production and sale of all goods (para. 4.05).11 (2) Public Investment Program (PIP) Reform 3.06 A Public Investment program (PIP) in industry consisting only of projects whose viability had been established was agreed for 1985 and 1986. The size of the PIPs was reduced as compared to the unsustainable levels of previous years and the quality of the projects was improved. For 1985 the Ministry of Industry (MIEN) had initially p:oposed an industrial PIP of US$37 million, of which 80 percent would be for rehabilitation and 20 percent for completing ongoing projects. Eventually, by applying the principle of including only projects of established financial, economic, and technical viability, a US$10.6 million PIP was agreed upon, of which 40 percent would fund the completion of seven ongoing projects and the remaining 60 percent would be used to rehabilitate six existing public sector enterprises. The realized PIP for 1985 was US$8.2 million. The authorities also agreed to consult with the Bank before starting any new industrial project for which total investment was in excess of US$4 million. For 1986 the MIEN proposed an industrial PIP of US$47 million. Once more, by applying the principle of including only projects of demonstrated financial, economic, and technical viability, a PIP amounting to US$17.8 million was agreed upon, of which 70 percent would be used to rehabilitate existing public sector enterprises. 3.07 At the time of the 1984-1985 PIP review sixteen on-going projects were considered economically dubious and specialized consultants carried out PPF financed studies to establish their viability. The most important conclusions of these studies were that the Zeren urea plant project - the single most important industrial public investment in Madagascar - and the Mamisoa vegetable oil factory were determined to be economically and financially non-viable. Bank staff recommended the Government not to commence operations with Zeren and Mamisoa, since they could only be operated at a loss. At first, the Government challenged the consultants' findings but eventually agreed on the need to search for the best alternative, including liquidation, moth-balling, or some recovery of the investment cost.32 3.08 Under ISAC, the quality of the projects included in the 1985 and 1986 industrial PIP improved, the size of the program decreased, and MIEM's project appraisal capabilities also improved. The average annual PIP for the 1984-1986 period averaged US$14 million per year, below the unsustainable levels of approximately US$80 million a year since 1979.1 The dropping of several projects of dubious viability alleviated the public sector's demand for bank credit, diminishing the crowding out of the private sector.- The role of the 1" CEM 1991, para. 93 and Draft ITPAC-PCR, para. 23. 2t Memo April 9, 1985, para. 6. 1 ITPAC-PR, para. 3.09. 2t PCR, para. 18. - 15 - TA component in achieving these improvements remains unclear due to lack of details in supervision reports on TA's implementation. Nevertheless, mission reports' acknowledged that neither at the level of sectoral planning nor at the level of national investment coordination had there been significant progress in project selection and the preparation of PIPs. Clearly a more comprehensive effort to rationalize the public investment efiort and to focus on the efficiency and social aspects of adjustment was needed. Follow-up actions supported by ITPAC and PSAC attempted to address this issue during the 1987-1990 period. (3) Import Liberalization and Export Promotion 3.09 Concerning imports, under ISAC, the Malagasy Government removed prohibitions on the importation of basic necessities and goods that were produced by local monopolies representing 20 percent of the number of prohibited goods. Before effectiveness, 7 of the 146 categories of prohibited goods became importable and import licenses together with US$1 million in foreign exchange were made available. These goods included edible oils and fats, soap, candles, writing paper and toilet paper and represented about 5 percent of the list of prohibited goods.' An additional 15 percent reduction was implemented before the release of the second tranche. In addition, a study on effective protection levels in Madagascar was completed under the technical assistance component of the ISAC (see para. 18). 3.10 Concerning export promotion, beginning in 1982 with the devaluation of the FMG the Government began implementing policies geared toward the gradual elimination of the existing anti-export bias. The authorities began allowing manufacturers to import goods needed for the production of exports and a scheme known as the EPI account (export, parts, inputs) - whereby exporters were assured access to part of their net export receipts to pay for their spare parts and inputs used in the production of exports - was established for specific firms beginning in March 1983. During ISAC preparation the Malagasy authorities started streamlining export procedures by automatically renewing exporters' cards (which had acted as an export license), by permitting manufacturers to export a range of goods rather than only specific items, and dropping, for all exporters with a previous export record, Government clearance prior to each consignment. 7 Export promotion measures supported by the ISAC included (a) the elimination of export taxes applicable to all manufactured products and the exoneration of exporters of manufactured goods from all indirect taxes on raw materials and spare parts used in the production of exports and (b) the preparation and implementation of an action program to promote exports, including the extension of the EPI system to all exporters of manufactured goods. 3.11 Export taxes applicable to all manufactured products and to a number of agricultural products were suppressed by the 1985 Finance Law. However, ambiguities remained concerning the exoneration of exporters of manufactured Li Memo April 9, 1985, para. 8. 3 Memo, April 9, 1985, para. 9. L PR, para. 61. - 16 - goods from all indirect taxes on raw materials and spare parts used in the production of exports. The first supervision mission (March 18-25, 1985) ascertained that laws enabling exporters to be exempted from the two indirect taxes applicable to inputs - the TUT1i and the TCI191 - already existed within the Tax Code ("Code g6n6ral des imp6ts"). Evidence was produced by the Government indicating that some of the exemptions included in the Tax Code were effectively applied (e.g. TUT tax credit). But in other cases exporters did not appear to have benefitted from exemptions, such as the tax on inputs of exported products not normally subject to the TCI - as in the case of wood products - which could be exempted on the basis of individual Ministerial Decisions. Bank staff assessed that this could be explained by a lack of knowledge of the relevant legislation among exporters and difficulties in dealing with the bureaucratic aspects of obtaining the exemption decision. Similarly, because of practical difficulties in identifying which spare parts were used in export production and in what proportions, spare parts had not been exempted from indirect taxes. Nevertheless, Bank staff concluded that since the necessary laws already existed and the remaining difficulties referred only to their effective application this condition of effectiveness could be considered as being met. The mission, however, recommended that implementation issues be addressed in the overall context of the preparation of the action program for export promotion.40 3.12 The EPI foreign exchange retention facility was extended to all exporters of manufactured goods covering all their foreign exchange needs for the production of export goods. The preparation and implementation of an action program to promote exports, however, ran into several delays. During the March 1985 supervision mission, Bank staff determined that ever though consultants had been invited to present proposals for the export promotion program, formal effectiveness had to await the actual hiring of the consultants for the preparation of the program," which had been delayed due to changes within the MIEM. At the time of the following supervision mission (June 23-July 10, 1986) the MIEM expressed its intention of implementing certain export promotion measures recommended by a draft export promotion study -such as the extension of the EPI to all manufacturing exports, the application of a uniform retention rate of 35 percent, the removal of a limit on the maximum absolute amount of foreign exchange that an exporter could retain, and the creation of a national commission for exports - but had not yet developed a comprehensive and detailed system to promote exports.- Moreover, Bank staff felt that the Government was not giving enough priority to the need to further depreciate the FMG, even though the IMF was monitoring the evolution of the exchange rate. The adoption of the action program was delayed due to the fact that the Government had not yet Le "Taxe Unique sur le Transactions" at "Taxe de Consommation Int6rieure" L Memo, April 9, 1985, para. 12-13. i Memo, April 5, 1985, para. 2. L1 Memo April 11, 1986, para.14. - 17 - received the final version of the consultant's report as well as to the lack of inter-ministerial consensus on the proposed measures.31 3.13 Eventually, in August 1986, the Government introduced an export promotion program that guaranteed exporters the necessary foreign exchange for their imported inputs and simplified exporting administrative procedures. More specifically the program: (a) guaranteed the foreign exchange required for spare parts and material inputs to all economic agents for the part of their expected export output, provided that the export output exceeded 20 percent of overall turnover per year; (b) guaranteed producers exporting 50 percent or more of their output foreign exchange for their entire production, including the part that was sold in the domestic market; (c) exempted the latter two categories of exporters from a 10 percent fee levied on all other economic agents requesting foreign exchange; (d) abolished the requirement that the value of exports reach a minimum amount and that export prices be approved; set up an inter-ministerial committee (Industry and Commerce) to oversee the simplification of export procedures and to generate ideas for further encouragement to exporters.! 3.14 Within the context of IMF Stand-by agreements, the Government also took actions pertaining to the exchange rate regime. On August 1986 the FMG was devalued by 20 percent in foreign currency terms against the basket of currencies to which it was pegged. The Government also agreed under the sixth IMF Stand-by to continue a policy of quarterly exchange rate adjustment based on the evolution of domestic inflation.5' 3.15 As discussed in the PCR, even though these actions represented a step in the right direction, in 1987 major structural issues relating to the Malagasy exchange and trade policy remained to be addressed. Outstanding issues included: import prohibitions and a cumbersome licensing system, high and dispersed levels of effective protection, foreign exchange shortages and the maintenance of foreign exchange controls, and a limited export response given persisting distortions in the overall incentives' structure.6 A more comprehensive and far reaching effort to address trade and exchange regime issues and tariff reform were pursued under ITPAC and would also be supported under the coming PRIVAC. (4) Private Sector Encouragement 3.16 With the objective of resuming a more active role of the private sector in the industrial sector a new Investment Code, satisfactory to IDA, was approved by the National Assembly before the release of the second tranche of the ISAC. The code constituted an improvement over the previous code by providing additional guarantees against nationalization, by allowing the repatriation of dividends and profits, and by reducing special treatment for particular classes 13 Telex, April 29, 1986. j Memo, October 6, 1986, para. 3. i Memo, September 18, 1986, para. 3. i' PCR, paras. 19-20. - 18 - of investors.- Investments for new enterprises as well as for capacity expansion were eligible to receive benefits under the Code, which included two regimes, the "priority agreement" and the small and medium scale national enterprise". The "priority agreement" would apply to all "large" investments for a period between four and seven years. It granted a total exemption from import duties on capital goods; a profit tax holiday during the first years of the agreement, progressively phased out during the last three years; repatriation of dividends to non resident foreign owners, and guarantee of a stable tax regime during the length of the agreement. The "small and medium scale national enterprise" regime would apply to the enterprises which were not large enough to qualify to the previous regime. Those enterprises, however, had to have a majority Malagasy owned capital and a management under the control of Malagasy people to be eligible to this regime, which could not in any case exceed five years. It granted the same benefits as the previous one with the exception of the repatriation of profits. Finally, a "partnership" regime was also created for those private entities wishing to undertake a project jointly with the State. The benefits under this regime were to be decided on a case by case basis.L81 3.17 By early September 1987 - when a supervision mission visited Madagascar - while the Investment's Code applicability decree had been promulgated by the National Assembly (before the release of the second tranche), the Code had not been published or enacted yet. For the Investment Code to become a non- discretionary law with transparent applicability, the publication of the adopted implementing decree was required. The applicability decree was eventually published in the Official Gazette only on September 21, 19871f and was enacted only in May 1988. By then the code had been rendered obsolete by the changing economic environment and a new, more open investment code was promulgated under PSAC in December 1989 partly as a result of the continuing policy dialogue between the Government and the Bank.So B. BNI Investment Component 3.18 SDR 5 million of the credit was to be lent to the BNI for the fInancing of investment (rehabilitation) projects through sub-loans to investment enterprises. Sub-loans were restricted to investment projects with a total cost not exceeding US$ 500,000. According to the PCR!i "the BNI component was the least successful, due mainly to the unfavorable investment climate and BNI's slowness in processing applications for sub-loans and its parlous financial situation.-' Particularly at the beginning of implementation process, projects i Ibid., para. 21. M Memo, April 9, 1985, para. 15. Mi Memo,October 7, 1987, para. 21. Lt PCR, para. 5. Li Ibid., para v. 1-t Ibid., para. 22. - 19 - proposed by BNI were more expansion than truly rehabilitation and this resulted in implementation delays. Nonetheless, BNI's financial situation was the main reason behind implementation delays. 3.19 Bank staff became aware of BNI's untenable financial position only during a July 1985 mission, when staff realized that BNI was having serious liquidity problems. By the time the next supervision mission visited Madagascar, in February 1986, "BNI's portfolio situation in general remained obscure ... the year end 1985 financial data (balance sheet and income statement) was not yet available, BNI's management was also quite reluctant to disseminate much other information to the supervision mission."Lu Moreover, the mission realized that existing rehabilitation projects were not coordinated by BNI, as they were supposed to be, but were instead carried out by the Government. Studies to rehabilitate certain enterprises apparently existed but were not made available to the Bank mission and at the end of the mission it remained unclear who was actually in charge of coordinating these rehabilitation programs.L' 3.20 At the end of the February supervision mission Bank staff agreed with the authorities that among other, (a) an auditor would finalize an on-going portfolio update by April 1986, and (b) BNI would hire an international audit firm by April 1986 to carry out the audit of the 1985 financial accounts, including a comprehensive review of BNI's accounting systems and procedures. If BNI did not comply with these and other steps agreed during the mission to clear up BNI's portfolio situation and the terms and status of rehabilitation projects, no extension of the commitment deadline would be considered.25 During the June-July, 1986 supervision BNI's financial situation remained difficult: a number of large public sector enterprises were contributing to the steady deterioration of BNI's portfolio; the non-performing portion of the portfolio increased by 32 percent during 1985 and by mid-1986 represented about half of BNI's total portfolio. BNI still did not prepare a budget, projections, or a sound pipeline of sub-projects. Nevertheless, BNI's component cow,iitment date was extended and ISAC's closing date which was originally set foi 'ptember 30, 1987 was also extended to September 30, 1988.- By July 1988, and despite a start at rehabilitating selected public sector enterprises which were responsible for most of BNI's difficulties, half of BNI's portfolio was still not performing. 3.21 During supervision missions it became clear that the steps taken under ISAC would be by far insufficient to reintroduce efficiency and viability to BNI and that further action would need to be pursued within a financial sector operation since BNI problems were common to the other two commercial banks and could not be solved within the context of ISAC alone. 2 Memo, April 11, 1986 para. 18 and 23. 2 Ibid., para. 25. 25' Ibid., para. 30. 2U Memo June 8, 1987, para. 3. - 20 - C. Technical Assistance Component 3.22 SDR 1.5 million had been allocated to the MIEM to (a) reinforce its data collection, analysis, and evaluation of Investment projects; (b) prepare an action program for export-promotion and a study of Madagascar's import system; and, (c) rehabilitate and upgrade a building to serve as MIEM headquarters. A data bank for the MIEM was created and the MIEM building was rehabilitated. The trade regime study undertaken under the TA component provided estimates of effective protection levels and dispersion which were used as groundwork for the design of ITPAC. The study concluded that in 1986-1987 tariffs were providing a high and uneven pattern of effective protection. The average effective protection, granted by tariffs, for 68 representative products was about 160 percent while specific tariffs varied between -5 percent and over 1,000 percent.- Nonetheless, effective protection levels were underestimated since the did not take into account import prohibitions and exchange controls. Additional details on the TA component's implementation are unavailable due to lack of detailed supervision of the TA effort. IV. ASSESSMENT OF OUTCOME WITHIN THE OVERALL ADJUSTMENT PROGRAMLi A. Assessment of Outcome 9' 4.01 As emphasized by the PCR,- it would not be appropriate to assess results and draw lessons for ISAC individually. ISAC was only intended to pave the way in the direction of adjustment as a first step toward a more comprehensive reform process and was followed relatively quickly by other adjustment operations.L- In fact, the process leading to ISAC "served as a useful vehicle between the Government and the Bank Group on a number of far- reaching policy changes affecting a varibty of macro and industrial sector issues."6 Acknowledging the limitations of ISAC as a preparatory credit and recognizing that its outcome must necessarily be assessed within the framework of the overall Malagasy adjustment effort, ISAC results will be evaluated according to two criteria: (a) whether the stated objective of the ISAC was a-'iieved and (b) whether priority issues in the industrial sector, discussed in Chapter II, were indeed addressed either by ISAC or by subsequent operations. 4.02 The stated objective of the ISAC was to support the Government's strategy of bringing about meaningful policy changes in the industrial sector 1U ITPAC-PR, para. 3.20. 1t This chapter draws on CEM 1991, PRIVAC-PR, and Report No. 7784. L PCR, para. 10. i Ibid., para. 29. t PR, para. 77. - 21 - designed to increase its efficiency and productivity, while beginning the process of establishing an environment conducive to efficient import substitution and export production (para. 2.04). These policy changes would constitute important first steps towards promoting the efficiency, openness, and export-orientation of the Malagasy industrial sector so that it could increasingly contribute to the country's balance of payments and to the growth of employment and output.L1 4.03 ISAC did indeed provide the impetus and groundwork for the implementation of follow-up credits, which taken as a whole represent a fairly comprehensive adjustment effort. After the implementation of the ISAC major outstanding issues still remained to be addressed by follow-up operations, including high and uneven protection, exchange rate overvaluation, foreign exchange controls, public enterprises' issues, credit policy (including a review of savings mobilization, interest rate structure, and credit allocation), rehabilitation of portfolios of all three banks, establishment of sound write-off procedures, assessment of type of financial services needed, establishment of bank supervisory entities. Therefore, ISAC was followed by a subsequent operation in the industrial sector, the Industry and Trade Policy Adjustment Program (ITPAC - Credits 1834-MAG and A-32-MAII) in FY87, separate adjustment operations in agriculture (ASAC, Credits 1691-MAG and A-16-MAG) in FY86, in the public sector (PSAC, Credit 1941-MAG) in FY88, and a proposed private sector environment adjustment credit (PRIVAC) which is scheduled to be negotiated in late 1991. The adjustment effort was complemented by non-adjustment operations in the social sector with the support of the Economic Management and Social A tion Support Project (EMSAP, Credit 2125-MAG) in FY89, in the financial system ( 2EX, Credit 2104-MAG) and in the environment (Credit 2125-MAG) in FY90. Notwithstanding the follow-up effort, the appropriateness of the decision taken by the Bank to disaggregate structural adjustment lending into a series of subsequent sector operations, rather than on a comprehensive SAL, is an issue linked to the design of the credit and is discussed in Chapter V. 4.04 Beginning with ISAC and continuing in earnest in 1987, the Malagasy Government embarked on a process of general liberalization of most economic activities. With respect to the industrial sector, the adjustment process focused on: (1) abolition of controls on prices as well as on industrial and commercial profit margins; (2) elimination of import prohibitions and quantitative restrictions for protective purposes; (3) adoption of a market-determined foreign exchange allocation regime; (4) simplification of the administrative exporting procedures; Lt Ibid., para. 77. Li Report 7784, para. 12. - 22 - (5) the rationalization of the overall PIP process; and, (6) the stimulus of private domestic and foreign investment. In addition, an active and more competitive exchange rate management policy was pursued throughout the adjustment period with the collaboration of the IMF, which beginning in 1980 supported the Government's stabilization effort with several Stand-by arrangements and special facilities. 4.05 Between 1983 and 1989, in the context of the overall adjustment effort, the Malagasy Government succeeded in removing virtually all price control. Price liberalization began with reforms in the agricultural sector freeing the producer price of rice in 1983 and the consumer price of rice in 1985. Between 1986 and 1988 the Government liberalized the prices of all agricultural commodities with the exception of vanilla, cotton, sugarcane, and tobacco. Road transport tariffs were also freed in 1985. In the industrial sector, the price liberalization effort supported by ISAC, was completed in 1989 through reforms implemented under ITPAC (paras. 3.04-05). Currently, only prices of products and services provided by state monopolies are still controlled.- For natural monopolies and telecommunication services the Government has agreed with the Bank on a new pricing system based on the long term marginal cost of the services. The liberalization of vanilla, air transport, and petroleum will be pursued under the proposed PRIVAC. 'According to World Bank reports, the inflationary impact of price liberalization has been limited because market prices already reflected scarcity premia and because the price liberalization coincided with devaluations, tariff reductions, and the removal of quantitative restrictions.65 4.06 While under ISAC a small number of import prohibitions were removed and replaced by tariffs and a few export-promoting measures were implemented, substantial distortions remained in the trade regime and a general anti-export bias of trade and macroeconomic policy still existed. Thus, particularly in the exchange and trade area. ISAC provided only the groundwork while ITPAC and PSAC set in motion a far-reaching program to reform the instruments of trade Rolicy.-L6 4.07 On the import side, in January 1988, the Malagasy Government eliminated almost all non-tariff barriers on imports (except for 94 items, which were health and security-related and were reduced to 77 in March 1989). The authorities instituted a temporary 30 percent surcharge on some previously prohibited imports to ease adjustment in the sectors concerned, primarily textiles and meat. The surcharge was reduced to 10 percent in 1989 and was then eliminated in January 1990. All tariff exemptions were eliminated and nominal and effective protection was lowered. A program of tariff rationalization was introduced in January 1988, si CEM 1991 para. 93, PRIVAC-PR, para.54, and Report 7748, para. 79. L CEM 1991, para 93. and Report No. 7784, para. 80. S1 Additional details and an in-depth analysis of the outcome and outstanding issues of the reform process in the industrial sector are provided in Report 7784. - 23 - when specific tariffs were substituted by ad valorem ones and tariff categories were reduced from 69 (ranging from 0 to 1,000) to 21, ranging from 5 to 110. In 1990, the number of rates was further reduced to 14 ranging from 5 to 80 percent. In January 1991 the Government adopted a new tariff schedule ranging from 5 to 60 percent, and in 1992, supported by PRIVAC, the Government has expressed the intention of reducing the number of tariff rates to 5, ranging from 10 to 50 percent. The overall (un-weighted) mean tariff rate dropped from 46 percent in 1988 and 41.5 in 1989 to 32 percent in 1990 and is programmed to further decrease to about 29 percent in 1991. The standard deviation of the tariff schedule also decreased form 28.6 9ercent in 1988 to 20 percent in 1990, and is expected to be 17 percent in 1991.1- The 1989-1991 tariff reforms significantly contributed to the reduction of the dispersion of the tariff distribution, eliminating cascading to a very large extent. Even in the absence of an effective protection study, clearly the tariff reduction has significantly contributed to bring down effective protection rates. 4.08 On the export side. the liberalization process began in 1986-1987 and accelerated during 1988.1' In 1986-1987, supported by ISAC and ITPAC, prior authorization to export, prior export price controls, minimum value requirements and trial periods for exporters were abolished, aid exporters' cards were issued more freely. However, until the implementation o. PSAC, the existence of tight administrative controls limited the effectiveness of these liberalization measures. In January 1988, under PSAC, the state stabilization fund's monopoly on exports of coffee, pepper, and cloves was abolished. Exporters' cards and export declaration requirements were eliminated. In September 1988, the Government published a new export law enabling exporters to freely export their products, except for vanilla, at prices negotiated directly by exporters and importers. All mandatory quality controls and certifications of exportable goods by the authorities were abolished. In addition, under ISAC export taxes had been abolished on all but three traditional export products (coffee, vanilla, and cloves) and in 1990 the implicit multiple levies on coffee and cloves exports were merged into single explicit variable ad valorem taxes.- Thus, except for vanilla, all administrative export requirements were reduced to a single customs' declaration. 4.09 In order to further advance export promotion and to complement import liberalization actions, the Government acted also on the exchange rate policy front.- During 1983-1986 the nominal effective exchange rate depreciated by a cumulative 60 percent, corresponding to approximately 30 percent in real terms. Notwithstanding these devaluations, at the time of ITPAC prparation (i.e., early 1987), significant excess demand for foreign exchange persisted and the parallel market rate was about 50 percent higher (in foreign currency terms) than the LU CEM 1991 para. 39, Draft PRIVAC-PR, para. 42 and draft ITPAC-PCR, para. 22. Lt GEM 1991, para. 40. Lt Ibid., para. 58. LGEM 1991, para. 37. - 24- official rate.ZL; By end-1987 the authorities further devalued the FMG by a cumulative 43 percent in nominal trade-weighted foreign currency terms. In 1988 the FMG was further depreciated by a total 26 percent. Thus, by 1989. the real effective exchange rate was approximately 40 percent of its 1982 level (Table in Annex 1).4' On January 10, 1991 - as an up front action for PRIVAC - and to improve the country's deteriorated external balances, the Government devalued the FMG by 13 percent. To avoid future real appreciations of the FMG the authorities have indicated their intention to implement periodic corrections of the exchange rate in line with the inflation rate differential between Madagascar and its major trading partners. 4.10 The correction of the FMG overvaluation created the conditions for an overhaul of the foreign exchange regime. In January 1987, supported by ITPAC, the Government took a first step toward the implementation of a foreign exchange allocation system that did not discriminate between activities and end users. The import liberalization phase of the ITPAC introduced and developed a market-based system of foreign exchange allocation, the Liberalized Import Regime (LIR). In July 1988 the Government widened the foreign exchange policy reforms with the introduction of a full-fledged General License System (OGL) through which unlimited foreign exchange became available at the prevailing exchange rate for imports of goods by any economic agent. In February 1991 - during discussions on the PRIVAC-supported reform program - the Government also expanded automatic access to foreign exchange for invisibles. Further actions to be taken under PRIVAC would attempt to increase the flexibility and automaticity of the OGL system and address other limitations of the system.L 4.11 In conclusion, the 1970s and 1980s strong anti-export bias of the trade regime has been being dismantled, particularly after the cumulative devaluations and the replacement of import licensing with progressively reduced tariffs. To further promote exports, the Government has recently implemented further actions: in December 1989 the authorities promulgated a free-trade zone law offering generous fiscal incentives. Moreover, under the proposed PRIVAC, to complement reforms of the foreign exchange system (para. 4.10) ani uhe investment (para. 4.13) and free-trade zone legislation, the Government has irIcated its intention to publish and disseminate clear administrative guidelines for exporters to benefit from the existing mechanisms (temporary admission, drawback and bonded warehouse) allowing tax-free access to imported inputs.!S_ 4.12 While ISAC was successful in improving the quality of the projects included in the 1985 and 1986 industrial PIP, cutting the size of the program, and improving MIEM's project appraisal capabilities, a more comprehensive effort Lt ITPAC-PR, para. 3.13. Lt Ibid., para. 10. Zt CEM 1991, para. 38. 1t Ibid., para. 37-41. 2i Draft PPIVAC-PR, para. 45. - 25 - to rationalize the overall national oublic investment program and to pay incrsed attention to efficiency and the social aspects of adJustment was pursued under ASAC. ITPAC. and PSAC. Under ASAC an agreement was reached on the 1986 agricultural PIP. Under ITPAC, the overall PIP's preparation, control, and supervision was improved and agreement was reached on the 1987 and 1988 national PIP. Agreement on the 1989 and 1990 PIP was reached during the review of the ITPAC and PSAC implementation progress. The technical assistance and material requirements to support the budgetary reform process were financed through the EMSAP. Under PSAC, PIP's monitoring and programming was further improved with the implementation in 1989 of the first annual investment budget in the context of a three-year rolling PIP comprising all sources of financing with the objective of integrating project selection into comprehensive three-year public expenditure programs. Since then the authorities have prepared successive three- year rolling PIPs and PEPs for the health, education, transport and mining, energy and industry sectors. The draft PRIVAC aims at further improving the programming and monitoring of public expenditures, by seeking agreement on the 1992-94 and 1993-95 PIPs, PEPs, and the 1992 and 1993 budgets. 4.13 In the area of private sector encouragement, even though the investment code prepared under ISAC represented a substantial departure from past negative attitudes towards the private sector and foreign investment, it was not timely and flexible enough to keep up with the changing economic environment. Thus, in December 1989. to attract private domestic and foreign investment, the Government adopted a new investment code. Under PRIVAC the Government has indicated its intention to adopt the implementation decree for the new investment code, which is to be published before negotiations.L1 4.14 At the time of ISAC serious constraints on financial intermediation existed (para 2,02). Problems and delays in the implementation of the BNI component of the credit (paras. 3.18-21), and credit shortages constituting a key constraint to economic recovery, were caused by financial sector liquidity problems linked to the need for public enterprise reform. Unfortunately, the situation of the commercial banks' portfolios, particularly of BNI, became known only during ISAC's implementation thanks to effective supervision. The financial situation of the three commercial banks was untenable and their portfolios were in need of major write-offs, mainly reflecting substantial non-performing loans to non-viable public enterprises which had been able to survive only thanks to credit from the three state banks. The banking system kept these enterprises alive at the cost of building up a high percentage of non-performing loans and seriously limiting its capacity to meet the financial intermediation needs of a liberalizing economy. 4.15 The financial health of the banking sector continued worsening up to 1987, when ITPAC and PSAC began addressing both the financial sector and public enterprise issues which had been left out of ISAC. On the public enterprise front,1.1 between 1988 and end-1990, dissolution or divestiture actions were initiated for 71 out of 167 public enterprises and fully completed for 34 of Li Ibid., paras. 46-48. L' CEM 1991, paras. 68-75. - 26 - them; restructuring plans were formulated for the six largest agricultural and commercial parastatals; the textile industry is currently going from almost entirely state owned to fully privatized. On the financial side, a ceiling on credit to the public enterprises was included in IMF programs. As a result of these measures, the share of public enterprises in commercial banks' credit decreased from 54 percent in 1983 to less than one third in 1990. The two credits also supported, during 1988-1989, the authorities' successful implementation of a major clean-up of the commercial banks' portfolios.- In May 1988, a comprehensive new banking law was enacted which terminated the State monopoly and opened the domestic banking system to foreign capital. BNI has been privatized and the capital of another state bank has been opened to foreign investors for a 25 percent minority participation. In addition, PRIVAC would support further reforms aiming at increasing the efficiency and effectiveness of financial intermediation.Li B. Impact on Macroeconomic Performance (1985-1989) 4.16 Even though the first large-scale macroeconomic adjustment program was launched in 1987, the Malagasy Government, throughout the 1980s, implemented several IMF and World Bank-supported stabilization and adjustment programs, which contributed to reducing domestic and external financial imbalances. After 1984, the fiscal deficit levelled off at approximately 3.5 - 4.0 percent of GDP, and by the end of 1987 fiscal accounts had been restored to the situation existing prior to the 1978-79 investment boom. Monetary creation slowed down, and by 1989 inflation had decreased to 10 percent, compared to a peak of over 30 percent in 1981 and 1982. The external current account deficit (including net official transfers) dropped to 5 percent of GDP, from a peak of 14 percent of GDP in 1980. The exports to GDP ratio increased from 8.8 percent in 1983 to 13 percent in 1989. Gross reserves reached a peak of 7.7 months of imports from a situation of almost no reserves at the beginning of the 1980s, and imports expanded for the first time since 1980. C. SuRply ResponseL1 4.17 In 1985-1986, under ISAC, limited reforms were implemented. During 1987-1988, with ITPAC and PSAC, reforms began in earnest. After a transition period - when the exchange rate was still overvalued and a full-fledged import liberalization effort had not yet begun - the overall economy began responding to the adjustment effort in mid-1988. The mobilization and use of domestic resources improved and there was an increase in capacity utilization, leading to a real GDP growth rate averaging 3.8 per year during 1988-1989, compared to dpproximately zero growth on average between 1980 an, 1987 (see Table in Annex Lt CEM 1991, paras. 84-88. 1' Ibid., paras. 62-65. Lt GEM 1991, paras. 27 and 31-33. L Ibid., paras. 28-30. - 27 - 1 for details on macroeconomic indicators). During 1988 and 1989 real GDP per capita growth was also positive, after a dismal record throughout the 1980s. 4.18 The initial reaction of industrial production to the implementation of reforms was limited. Industrial output grew on average by approximately 4 percent during 1986-1987, and by only 2 and 1 percent in 1988 and 1989, respectively. However, in 1990 industrial sector growth accelerated to over 6 percent and private industrial sector confidence improved. Between 1987 and 1990, industrial capacity utilization rose to 60 percent on average, as compared to earlier rates averaging 45 percent. The garment sub-sector was particularly impressive, with its capacity utilization raising from 50 percent in 1986 to 85 percent in 1988. Since short-term increases in capacity utilization are driven by export demand, 1987-1990 increases were primarily the result of the increasing profitability of exports as a result of several devaluations beginning in 1987 and of the plentiful availability of foreign exchange with the adoption of LIR and later of the OGL allocation system. 4.19 However, according to Bank assessments of the adjustment effort, "the most impressive change in the economy induced by the trade and price liberalization has been the increased outward orientation of the historically import-substituting industrial sector. The successful macroeconomic stabilization and, in particular, the devaluations and the liberalization of the import and export procedures have stimulated private sector development in the garment and knitwear and the food processing industries.,i2 Export of manufactures increased from 13 percent of total exports in 1986 to approximately 23 and 25.5 percent of total exports in 1988 and 1989, respectively. July 1990 discussions with the private sector - conducted during the preparation of Report 7784 - confirmed a widespread supply response to the liberalization measures, exemplified by an intensification of firm entry in labor intensive activities, an observable increase in the number of takeovers and, substantial restructuring in industrial activities that existed prior to the liberalization. Ui report 7784, paras. 96-100. - 28 - V. CREDIT DESIGN. UTILIZATION. AND SUPERVISION A. Design Evaluation (1) Gradualism; The Governance Dimension and Social Considerations 5.01 The theoretical underpinnings for ISAC were jointly established by the Malagasy Government and the Bank, in close cooperation and coordination with the IMF. The Government of Madagascar, fearing the economic, political, and social implications of the up front implementation of a comprehensive adjustment program, requested IDA's support for a sequenced approach, of which ISAC would be the first step. ISAC was on the whole well designed as a preparatory loan and, as stressed by the PCR,8' the reforms adopted under ISAC were steps in the right direction even though further reforms were needed in order to generate a significant supply response. 5.02 A general design issue deals with the appropriatenees of a phased sectoral approach as opposed to an up front comprehensive adjustment operation. While ISAC was indeed followed by subsequent operations (para. 4.03) the appropriateness of the decision taken by the Bank to disaggregate structural adjustment lending into a series of subsequent sector operations, rather than on a comprehensive SAL should be commented upon. The genesis of the ISAC as a first step on the road of adjustment, and its limited scope, derived from two considerations linked to the Malagasy's authorities political and social constraints at the time. 5.03 First, as stressed by the PCR,14 even though the socialist Government was committed to reform in principle, it was not at the time convinced of the "necessity or desirability of a comprehensive reform program." ISAC, therefore, was seen from the beginning as a the first step on the path to reform and attempted to: "strike a balance between the need for far-reaching reforms and the Government's fear of the unknown by starting the adjustment process with a program that was significant enough to lay the foundations for recovery and the resumption of growth and at the same time not too far reaching as to fail to command reasonable consensus."L5 Bank staff hoped that successful implementation of ISAC would convince the Government to implement a more vigorous reform package. Eventually, the overall adjustment program - supported by several IMF stand-by loans - consisted of several sector loans. Even though at the time of ISAC preparation governance issue were not yet explicitly considered, the governance dimension did affect i PCR, para. iii. _i Ibid., para. 8. 5i Ibid., para. 8. - 29 - both the design of ISAC and economic performance under the overall program.L6 The Malagasy Government in 1985 was not prepared to risk implementing a comprehensive SAL type operation. The Government, however, felt strong enough to undertake ISAC which was a first step in the reform process but already contained measures which might have been unpopular. The overall adjustment program was designed jointly by the Malagasy authorities and Bank staff not without difficulty given the skepticism and distrust of the socialist regime over market mechanisms. There was an effort by the authorities at building a consensus on liberalization measures, which was particularly difficult within a socialist economy with a socialist Constitution. 5.04 Second, the country's economic team was also attempting to strike a balance in the pace of policy reforms in order to improve efficiency in the industrial sector without causing severe economic dislocation and possible social unrest. Notwithstanding Madagascar's outstanding social issues, there is no evidence of a social impact analysis at the time of ISAC design, while both ITPAC and PSAC do discuss possible social implications of their implementation. However, this credit was considered from the beginning as a first step towards eliminating distortions in the economy, thus enabling growth with equity. It was anticipated that the social dimensions of adjustment would be the focus of subsequent IDA credits. 5.05 In conclusion, as stressed by the PCR,L- the phased approach was appropriate to the circumstances: "First, it allowed government confidence and commitment in the reform process to be slowly built up. Both the Government and important interest groups were at best skeptical of the need and desirability for adjustment at the time the process was begun, and at worst they were actively opposed. Positive experience under ISAC allowed these political obstacles to be gradually overcome. Second, it gave time for additional analytical work to be carried out and information and experience to be built up within the Bank, thus allowing a more comprehensive program to be developed. This latter factor was particularly important in light of the limited experience, at the time the project was first being planned, with this type of adjustment operations. The danger with such an approach is that 'adjustment fatigue' might set in, and commitment to further reform would lag. This does not appear to have happened in this case. There was some limited backtracking on tariff reforms and delays in the public enterprise privatization process i Nooter, R.H. and R.A. Stacy, Progress on Adjustment in Sub-Saharan Africa: Implications for Future Lending Strategies, October 15, 1990, IBRD internal draft: paras. 37-40. Li PCR, paras. 30-33. - 30 - during implementation of ITPAC, but the program was soon set back on track. The other significant drawback of the gradualist approach is that the funds could go to firms (including public sector firms) still operating in a highly distorted environment. This problem was partidularly acute for the investment component of the Credit, but also applied to the general imports component. Doubts were expressed by the loan committee that the funds would be used efficiently with important distortions remaining. The BATA subloan, for example, was used for equipment which later proved inappropriate in the liberalized regime. In a sense, the BNI's slowness in processing applications proved to be a blessing in disguise, since subprojects which were eventually approved were conceived by their promoters and implemented under the more liberalized system established under ITPAC. These subprojects seem to have been reasonably efficient. The complicated tests to ensure 'adequate value added' and 'additionality' are another example of the problems caused by the partial approach. They were, however, useful to illustrate the need for a transparent non-administrative resource allocation system and, in any case, they were short-lived." (2) Sequencing of Reforms 5.06 The issue of sequencing is closely related to the choice of a phased approach to adjustment. The PCR/ raises the issue of whether the adjustment process should have started with a more rigorous foreign trade reform first, followed later by domestic reform. Such an approach, however, was not deemed politically feasible at the time. Moreover, even though external competition policies are generally introduced prior to internal competition, this is not necessarily a dogma. In fact, as stressed by a recent OED study on trade policy reform,L9 "the effectiveness of trade reform is strengthened when it is accompanied by internal competition policies in industrial, financial and other sectors. It not only improves the economy's flexibility to respond to external competition but it also prevents domestic regulatory policies from eventually being used as anti-trade devices." 5.07 Given the close link between financial intermediation and industrial sector reforms, a second set of sequencing issues relate to the need for a prompt analysis of the status of the financial system and to the timing of financial t Ibid., para. 34. ' Report No.9527, World Bank Support for Trade Policy Reform, April 22, 1991, paras. 4.62-4.65. - 31 - sector reforms. The industrial restructuring process can succeed only if adequate financial resources, supplied by an efficient financial intermediation system, are available at reasonable cost. When the financial system is in deep distress and when most banks in the system have large non-performing portfolios, financial sector reforms should be tackled in parallel to other industrial sector reforms, since further lending through unsound financial intermediaries would only worsen the problem.2' During ISAC preparation no detailed analysis of the financial sector situation was performed. At the time of the President's Report Bank's staff assessment of BNI financial situation was that "overall, BNI's portfolio is of acceptable quality ... the financial position of BNI is sound.11l Only during supervision did Bank staff become aware of financial sector issues. 5.08 In discussions with the Region it became clear that at the time of ISAC preparation Bank staff accepted BNI numbers at face value and that the loan classification available at the time led to the optimistic conclusions presented in the President's Report. BNI's real financial situation became clear only when loans were later reclassified enterprise by enterprise. Nonetheless, the lack of financial sector work in the design phase was made up by effective supervision and financial sector follow-up measures taken under ITPAC and PSAC. 5.09 Even though foreign exchange controls were in place at the time of ISAC the freeing-up of foreign exchange allocation was not pursed until after 1987 under ITPAC. Tnis was explained by the strong skepticism of the socialist Government towards market mechanisms, especially in such a delicate area, at the time of ISAC design. (3) Conditionality Issues 5.10 Finally, conditionality was manageable and appropriate to the objectives pursued by ISAC. However, there were two important differences in the way two second tranche conditions were stated in the President's Report and the legal conditions. First, while the President's Report stated the second tranche condition as "decontrol of all industrial prices," the Development Credit Agreement specified "free ex-factory prices."L' The Region acknowledges that the Presiaeit's Report was not specific enough since at the time of ISAC design Bank staff had not focused on the differences between various price controls. This gave an opportunity to the Government to maintain controls on profit margins (para. 3.04). Clearly, the spirit of the condition was not met. Second, while the President's Report asked for the implementationI of a satisfactory export qoQ' Lieberman, I. Industrial Restructuring, Policy and Practice, , Policy and Research Series, No.9, Industry and Energy Department, 1990: 21. _' PR, Annex IV, paras. 14 and 17. 2i PR, para. 95. 2t Development Credit Agreement, Schedule 4. 2i PR, para. 95. - 32 - incentive system, the legal condition specified only the need to adoDt!- a program of actions to promote exports. Moreover, the Investment Code was not flexible enough to keep up with the changing economic environment (para. 3.17). Finally, the condition calling for the exemption of indirect export taxes was redundant since the Tax Code already contained such exemptions. (para. 3.11). B. Disbursements 5.11 The credit was jcintly financed by IDA (credit 1541-MAG, for SDR 40.2 million) and a supplemental credit from the Special Facility for Sub-Saharan Africa (SFA) (Credit A-7-MAG, for SDR 18.5 million). The proceeds for importation of inputs would be disbursed in two tranches, while the rehabilitation and technical assistance components were not tranched. The industrial imports component of the project - which represented 84 percent of the ISAC and 100 percent of the SFA Credit, for a total of SDR 52.2 million - consisted of the provision of foreign exchange for the importation of intermediate goods and spare parts, and materials for eligible enterprises through the three commercial banks, i.e., BNI (National bank for Industry), BTM (National Bank for Agriculture), and BFN (National Bank for Trade). The list of eligible enterprises was established on the basis of value-added eligibility criteria, i.e. eligible enterprises had to have a ratio of value added to the value of their inputs, both calculated at border price, at least equal to 50 percent. Foreign exchange allocated to eligible firms from the proceeds of the Credit had to be additional to the amount that would otherwise have been allocated without the Credit. 5.12 The quick-disbursing component of the credit was channeled through the Central Bank, which then sold the foreign exchange to eligible enterprises through the three commercial banks. In order to accelerate disbursements a Special Account, i.e. a revolving fund, was established at the Central Bank. At the beginning disbursements were slower than expected due to the six month delay in Credit effectiveness resulting from late hiring of consultants to prepare the export program and as a result of problems associated with the use of the Special Account. Disbursements began catching up when a supervision mission (February 4-20, 1986) assessed that the exclusive reliance on the Special Account coupled with the use of that account to guarantee imports instead of pay!ng for them outright was the major cause of slower than expected disbursements. As a result of consultations between Bank staff in charge of the project, the Loan Department, and the Malagasy authorities, it was decided to use Procedure V in addition to the Special Account. Disbursements under the Special Account would from then on be made under a general and irrevocable guarantee of reimbursement issued by the Bank to a French bank chosen by the Malagasy Government.2 This change resulted in faster disbursements. 5.13 SDR 5 million of the credit was to be lent to the BNI for the financing of investment (rehabilitation) projects through sub-loans to investment 9i Development Credit Agreement, Schedule 4. - Memo, April 11, 1986, para. 4. - 33 - enterprises. Because of BNI financial problems and lack of coordination by June 1986, only 20 percent of the US$5 million assigned by ISAC to this component had been committed for two rehabilitation projects. The commitment date expired June 30, 1986 and after some actions to address BNI portfolio situation were taken, the commitment date was extended to March 1987.27 Due to BNI's difficulties in implementing the investment component ISAC's closing date which was originally set for September 30, 1987, was extended to September 30, 1988. Eventually the credit was closed in September 30, 1990 with only 53 percent of the planned amount disbursed to the BNI component and the remaining amount reallocated to the fast-disbursing component. Finally, regarding the third component, of the SDR 1.5 million that had been allocated to the MIEM for technical assistance purposes US$ 400,000 were reallocated to the general imports component. ' Unfortunately, few details are available of the TA component performance and of reasons of this reallocation of funds due to lack of continued supervision of this component. C. Supervision 5.14 The supervision effort was adequate. Supervision was imaginative and understanding, and findings of ISAC supervision missions were instrumental in identifying outstanding issues and thus helpful in designing subsequent operations. Such as in the case of BNI financial difficulties, when Bank supervision missions acknowledged the impossibility of solving them within the framework of ISAC and recognized the necessity of taking care of such problems within the implementation of financial sector and public enterprise reform, which were later supported by both ITPAC and ISAC. There was close coordination and cooperation between the Bank and the IMF, and the parallel implementation of stand-by Fund operations assured close monitoring of macroeconomic management, particularly exchange rate policy. Although the IMF had primary responsibility on exchange rate management, the World Bark did follow exchange rate developments closely during supervision. 5.15 However, there were deficiencies in the loan files, including a missing Implementation Summary (Form 590). Moreover, all 590 forms were issued at least two months after the end of the supervision mission involved. Probably due to these deficiencies, the PCR does not contain details on the supervision effort. According to a recent OED study on SAL supervision and monitoring,2- deficiencies in historical records appear to be a common problem which hampers the maintenance of the Bank's institutional memory and thorough ex post evaluations of the operations concerned. Finally, updates on the implementation of the TA component were not adequate. 1 Memo June 8, 1987, para. 3. Li PCR, para. 24. L Report No. 9711, Effectiveness of SAL Supervision and Monitoring, June 26, 1991, para. 105-106. - 34 - VI. SUSTAINABILITY OF THE ADJUSTMENT EFFORT 6.01 The maintenance of macroeconomic stability and continued political commitment and popular support for the reform process are the key determinants of the sustainability of adjustment measures. Economic performance in the near future will be closely tied to political economy considerations; recent political developments and the fragility of the current regime have pushed the governance issue to the forefront of any assessment of sustainability of the Malagasy adjustment effort. A. Economic Performance and Sustainability Assessment 6.02 Macroeconomic developments in 1990, while confirming the upward trend of the real sector, demonstrated also the impact of exogenous factor on the Malaga&y economy and raised the issue of financial sustainability.0i In 1990, real GDP grew by 3.5 percent atimulated by strong growth in manufacturing activities and in trade ard tourism related services, notwithstanding a poor performance of the agricultural sector due to adverse weather conditions. Real per capita GDP growth was positive for the third consecutiv' year. Other economic indicators, however, did not fare as well. A severe foreign exchange liquidity crisis - which brought down reserves to less that two months of imports, their lowest level since 1985 - was precipitated by a series of factors. Imports increased by 45 percent (in nominal US dollar terms) due to the appreciation of the real effective exchange rate and an excessive expansion of credit provided by the only remaining fully state-owned commercial bank. The impact of the rise in imports on external accounts was magnified by a 6 percent decline in merchandise export earnings, due to deteriorating terms of trade for coffee, cloves, and vanilla. These events were magnified by a substantial shortfall in balance of payments support owing to delays in the implementation of policy reforms agreed with donor agencies. As a consequence, the current- account deficit worsened from 8.8 percent of GDP in 1989 to 11.8 in 1990. The inflation rate increased three percentage points to reach 12 percent. 6.03 The reserve position of the country has improved in the first quarter of 1991 thanks to a cumulative 20 percent devaluation of the FMG, a severe tightening of credit policy since November 1990, and disbursement of balance of payments support. Overall, indicators for the first months of 1991, suggest that the 1990 financial crisis was an isolated episode and has been overcome without a lasting negative impact on the adjustment and process and growth.1fl B. Social and Political Issues 6.04 Recent political upheavals in Madagascar have been triggered by changes in the Constitution and in the electoral law; the opposition is seeking to achieve a more open systam and not necessarily to change economic policies. More specifically, riots have not been directly linked to the social costs of Bank 0i For details on developments in 1990 see CEM 1991 paras. iv and 95-101. 0 Ibid., para v. - 35 - supported adjustment policies. Although no quantitative analysis of the social impact of the adjustment efforts is available, Bank staff yses show that policies that since 1983 have stopped GDP decline and have pen ced positive GDP per capita growth beginning in 1987, have had different effects on the various segments of the population: "For instance, rapid shifts in the urban/rural terms of trade, together with a sharp decline in the real wages of civil servants, while beneficial for the bulk of the Malagasy population, have adversely affected urban dwellers, and especially the poorest among them, worsening already dismal living conditions. Other groups, such as public enterprise employees, while not in absolute poverty, will also be affected by the policy measures."i The Economic Management and Social Action Project (EMSAP, 1988) addressed in part these issues and other social dimensions of adjustment. 6.05 Notwithstanding political upheavals, the political commitment to a continued adjustment effort remains strong in 1991. On the economic front both sides are in agreement on the need to continue liberalizing the economy and eliminating remaining structural constraints to growth with equity. This commitment has been demonstrated on the Government side by the policy dialog .; with Bank staff leading to the preparation of a comprehensive package of reforms under a the proposed PRIVAC. At the same time, the opposition has expressed its intention, if it gains control of the Executive, to proceed even faster than the current regime with the adjustment effort. In conclusion, if the political situation does not interfere with economic processes, all the elements for sustained growth are in place and an optimistic assessment of the sustainability of the overall adjustment effort seems justified. This assessment reflects largely the result of discussions with Bank staff since the planned audit mission had to be canceled owing to civil disorders. Therefore, this sustainability assessment may have to be reconfirmed during a field visit OED expects to conduct in connection with the forthcoming audit of ITPAC. Li-" EMSAP-PR, paras. 3. - 36 - VII. CONCLUSIONS AND LESSONS OF EXPERIENCE A. Summary Assessment 7.01 ISAC was essentially preparatory to the more comprehensive reforms implemented in follow-up sector operations. ISAC was on the whole well designed as a preparatory loan and the Government participated in the development of the conceptual framework of ISAC as well as of the following sectoral loans. Given the political environment at the time - with a socialist regime skeptical of market mechanisms and a full-fledged liberalization of the economy, and fearful of the political and social consequences of a full blown adjustment effort - the preparation of subsequent sectoral loans as opposed to the implementation of an up front adjustment loan was the appropriate choice. 7.02 Overall the adjustment effort implemented by the Malagasy Government has been impressive and it is generally perceived as one of the success stories within Africa.1 However, the sustainability of the adjustment process will depend on macroeconomic as well as political developments. On the macroeconomic area, certain conditions must be maintained for the structural reforms to have a lasting impact. Prospects for medium-term viability would be enhanced by a more aggressive exchange rate policy permitting the continuation of export diversification; further import liberalization to enable a sustained rise in imports to maintain the recovery; stronger financial intermediation with close monitoring of credit policies; and, improvements in domestic savings performance. On the political side, commitment to reforms must continue to exist, whichever political group remains in power. B. Lessons of Experience 7.03 The review of the experience with the ISAC offers instructive lessons and suggestions which may provide useful insights for future Bank operations: (i) In 1985 the Government was not ready yet to implement a bioad- based structural adjustment effort. Thus, policy-based sector specific lending was used by the Bank as vehicle to open dialogue with the authorities. The phased approach was successful in leading to full-fledged adjustment beginning in 1987 thanks to the Government's commitment to the reform process and to the appropriateness, timeliness, and, effectiveness of follow up operations. The phased approach opened a general policy dialogue with the Government and many reforms were implemented during the identification/preparation of projects. Not all achievements or policy actions were part of the conditionality package of the various credits. ITPAC and ISAC were fairly comprehensive programs which achieved more than 1i RaJcoomar, S., "Madagascar: Crafting Comprehensive Reforms", Finance and Development, September 1991: 46-48 and Nooter, R.H. and R.A. Stacy, Progress on Adjustment in Sub-Saharan Africa: Implications for Future Lending Strategies, October 15, 1990, IBRD internal draft. - 37 - sectoral reforms since they both included policy conditions having an impact across sectors and because much was accomplished even before effectiveness through the ongoing policy dialogue and negotiations with the authorities. Thus, much more was achieved than what is shown by the conditionality itself; they were closer to SAL operations than to SECALs. Although it has been found that in general it is more effective to proceed with a comprehensive up-front adjustment effort, a phased approach was appropriate and necessary in the case of Kadaga§caL. This approach was indeed suitable for a country where not only the Government was not ready for a full fledged SAL, but also because following a long interruption in Bank involvement in the country. the stock of economic and sector knowledge was not considered sufficient to mount a more comprehensive reform effort. (ii) During ISAC design there was awareness by Bank staff of the Government's political constraints, which led to the design of a successful sequencing of sector adjustment operations, through an intense policy dialogue aimed at building a consensus. The success of Madagascar's adjustment effort shows that, even under a socialist regime. Rolicy reforms can be successful and sustained yhen the design of the program is internalized and there is valitical will, consensus and commitment on the part of the authorities, (iii) A posteriori it is clear that a more complete analysis of the financial sector at the time of ISAC's design would have raised doubts ex ante - and not too late during implementation - on the appropriateness of BNI as the implementing agency for the rehabilitation component of the credit. Such analysis would also have identified the issues of financial situation and efficiency of the banking system, and public enterprises as key constraints to be addressed up front during the adjustment process. Thus, the status of financial intermediation should have been analyzed in more detail even if specific reform measures were not necessarily included in ISAC. (iv) The Malagasy adjustment process witnessed an effective cooperation with the IMF and a complementarity of Stand-bys and sectoral operations, especially on exchange rate issues. However, it appears that IMF actions were slow in keeping up with necessary exchange rate adjustments to maintain an exchange rate appropriate to the liberalization effort being suported by Bank operations, (v) Even though the supervision effort was adequate, there was a lack of detailed supervision of the TA component of the ISAC. In hybrid loans all the supervision effort seems to concentrate on the policy package rather than the less visible TA component, Thus, hybrid loans should perhaps go to the Board as a package - 38 - with one loan containing the policy package and a separate one dealing with TA efforts. (vi) Two of the second tranche conditions were spelled out in a slightly different manner in the President's Report and the Credit Agreement. Supervision missions determined the Borrower's compliance based on the fulfillment of legal conditions which appear to be diluted as compared to the President's Report wording. This potential issue was also found to be common to other adjustment operations by the already quoted OED supervision report,-V which recommends that since legal conditions are usually emphasized and better implemented. Management should ensure that the most critical actions constitute the legal conditions in the credit or loan agreement, gu Report No. 9711, Effectiveness of SAL Supervision and Monitoring, June 26, 1991, para. 66-69. - 39 - ANNEX 1. MADAGASCARe RET ECONCHIC INDICATORS, 1960-1990 1980 1981 1962 1983 1984 1985 1986 1987 1968 1989 1990 me.meme.m.. .sm... um........m.m.m nam.m.m.m..m.s SW...e.m.mm..m..m..m.eammgassee. National Accounts (Growth Rates, 2) Real GDP 0.8 (9.7) (1.8) 0.9 1.7 1.1 2.0 1.2 3.4 4.1 3.5 Real ODP per capita (1.9) (12.3) (4.4) (1.2) (3.5) 0.4 (3.3) (2.0) 0.2 0.9 0.5 Real Industry sector (2.7) (22.5) (14.4) 1.4 13.9 1.3 3.6 4.9 1.9 0.9 6.3 (As I of Current GDP) Gross Domestic Investment 15.0 11.5 8.5 8.4 8.6 8.5 9.0 10.1 13.3 13.3 16.9 Industry 14.3 12.6 11.9 12.0 11.5 11.9 11.6 12.0 11.7 11.3 11.6 Balance of Paymentes (In Months of CIF Imports) Total Gross Reserves minus Gold 0.1 0.5 0.4 0.6 1.1 1.5 3.5 5.8 7.1 7.7 1.6 (Growth Rates, 2) Exports (fob) 10.8 (23.9) (1.4) (3.3) 8.8 (13.6) 13.0 (5.9) (9.7) 14.1 (5.3) Imports (fob) 15.3 (33.1) (11.6) (16.2) (7.0) (4.6) (1.4) (4.8) (1.0) 0.7 45.6 (In Million US Dollars) Exports (fob) 436 332 327 310 337 291 329 310 260 319 302 Imports (fob) 764 511 452 378 352 336 331 315 312 314 458 Trade Balance (327.4) (176.9) (124.3) (68.3) (14.8) (44.5) (1.8) (3.3) (32.3) 4.6 (156.0) Current Account Balance at (599.2) (424.7) (371.9) (307.9) (261.6) (244.1) (244.7) (256.6) (263.4) (216.6)(364.7) (As I of Current GDP) Exports (fob) 10.8 9.2 9.3 8.8 11.5 10.2 10.1 12.1 11.5 12.9 9.8 Resource Balance (16.4) (11.3) (9.4) (7.0) (4.9) (6.2) (4.7) (5.4) (6.7) (4.5) (8.7) Current Account Balance a/ (14.8) (11.8) (10.5) (8.8) (8.9) (6.5) (7.5) (10.0) (10.8) (8.8) (11.6) (As 2 of TotA Cu Imports) Imports of intermediate goods b/ 22.0 18.3 19.5 25.0 27.2 27.8 30.2 29.8 25.8 23.9 23.7 (As % of Total Exports) Exports of manufactures 12.5 16.6 14.5 15.5 11.0 13.6 13.3 18.7 23.2 25.5 n.e. Fiscal Accountas (As I of Current GDP) Government Deficit (cash basis) (14.8) (12.1) (7.1) (5.4) (3.9) (3.8) (3.3) (3.5) (3.5) (4.2) (0.9) Prices and Exchange Rates Inflation (CPI - annual average*) 18.3 30.7 31.7 19.3 10.0 10.5 14.5 15.0 26.8 9.0 12.0 Nominal Effective Exchange Rate (1980-100) cl 100.0 89.6 78.3 69.3 56.9 50.6 40.3 22.9 17.0 15.7 15.3 % change (-.depreciation) (10.4) (12.5) (11.5) (17.9) (11.1) (20.3) (43.3) (25.5) (7.6) (2.6) Real Effective Exchang* Rate (1980*100) cl 100.0 105.6 112.1 111.0 94.6 89.0 79.9 50.9 45.5 44.0 44.6 2 change (--depreciation) 5.6 6.2 (1.0) (14.8) (5.9) (10.2) (36.3) (10.6) (3.2) 1.6 Terms of Trade (1960*100) 100.0 78.5 60.3 91.6 95.1 61.4 100.7 8.7 81.4 57.8 30.3 % change (0.3) (21.5) 2.3 14.3 3.6 (14.4) 23.7 (12.0) (8.2) (29.0) (13.0) emos GDP (Million US$) 4.042 3,595 3,526 3,511 2,939 2,857 3,239 2,565 2.443 2,472 3,097 -am.me.mmmm.m..n. memen.e.mme....men.......... Sourcest IHF - International Financial Statistics Tearbook World Bank Report No. 7784. "Madagascar - Adjustment in the Industrial Sector and ao Agenda for Further Reforms, October 19901 World Bank Report No. 9101, "Hadagascar - Beyond Stabilization to Sustainable Growth, June 1991. Draft President's Report. *Private Sector Environment Adjustment Credit". August 15. 1991. Notess al Excluding official transfers. b/ Excluding petroleum. c/ The NEER and REEK are based on the following basket of currenciess FF 412 US$ 302; Ot 122; TEN 112; UK 62. ANNE 11 Page 1 -lof 2 THE ISAC AN THE GOVERNMENT ADJUSTMENT PROGRAH EFFECTIVENESS SECOND TRANCHE IMPLE11ENTATION FOLLOW- POLICY AREA OBJECTIVES CONDITIONS CONDITIONS RECORD UP CREDITS 1. PRICE To make the Malagasy Prices of goods Only 31 products (repre- Met. However, ITPAC LIBERALIZATION economy more responsive to representing 35% value senting aboug 302 of only ex-factory market forces so as to added freed from control industrial value added) to prices were increase its efficiency, remain under price freed. Profit dynamism and productivity, control. For these margin controls Progressive elimination of products controlled prices were maintained. price controls on most to be brought into line industrial goods. with world prices 11. EXPORT Increase the outward Prepare Action Program to Implement satisfactory Delayed, then ITPAC PROMOTIt_ or4entation and efficiency promote exports-through, Action Program to promote Met. However, of the Nalagasy econa,y for example, improved exports action program PSAC through the promotion f export earnings retention was only I exports. scheme, duty drr<wback madopted" not system, etc. *implemented" at a the time of Export taxes on all manu- tranche ITFAC factured goods eliminated; disbureement. exporters of manufactured goods exonerated from all Export earnings rt.ention NET indirect taxes on raw scheme to !e extended to materials and spare parts all manufacturing used in the production of exporters covering all exports--both conditions of their foreign exchange effectiveness needs for the production of export goods. II:. IMPORT Progressive import Undertake a study of the Remove prohibitions on a MET ITPAC LIBERALIZATION liberalization in order to trade regime. Also, remove further 152 of goods improve industrial Import prohibitions on 52 current prohibited PSAC efficiency through greater of currently prohibited competition from abroad Soods (priority to he given and to encourage exports. to incentive and monopoly goods) before effectiv"- ness. Foreign exchange also to be made available to permit actuai importa- tion of these goods. ANNEX 11 Page 2 of 2 THE ISAC AND THE GOVERNMENT ADJUSTMENT PROGRAM EFFECTIVENESS SECOND TRANCUE IMPLEMENTATION FOLLOW-UP POLICY AREA OBJECTIVES CONDITIONS CONDITIOVS RECORD CREDITS IV. PUBLIC Rationalize the PIP Government/IDA agreement on Government/1DA agreement MET ITPAC INVESTMENT process, establish 1985 industry PIP - as on 1986 industry PIP PSAC PROGRAM IN criteria for project condition of effectiveness ASAC INDUSTRY selection, strengthen project appraisal Strengthen Project capability Appraisal capability at HIEH V. PRIVATE Encourage private sector Continue program of Present new Investment MET PSAC SECTOR to resume active role in involving private sector in Code, which would reflect ENCOURAGEMENT the industrial sector Public Enterprises IDA comments to National Assembly I -- 43 - PROJECT COMPLETION REPORT MADAGASCAR INDUSTRIAL SECTOR ADJUSTMENT CREDIT (Credits 1541-MAG and A.7-MAG) June 24, 1991 Industry and Energy Operations Division South Central and Indian Ocean Department Africa Region - 45 - MADAGASCAR INDUSTRIAL SECTOR ADJUSTMENT CREDIT (Credit 1541-HAG and A-7-14AG) PROJECT COMPLETION REPORT PART I A. Introduction 1. The Inductrial Sector Adjustment Credit (ISAC, Credit 1541-MAG of SDR 40.2 million and A-7-MAG of SDR 18.9 million)' was the first in a series of adjustment operations in Madagascar. It was designed to aid the Malagasy Government in its efforts to reverse the country's economic decline by adopting stabilization and liberalization measures. 2. ISAC's main objective was to bring about policy change- designed to increase the efficiency and productivity of the industrial sector and to boost the capacity utilization of selected industrial enterprises. Though ISAC was a hybrid operation, the bulk of the Credit (SDR 52.6 million, or 89 percent of total funds) was for a quick-disbursing component which provided foreign exchange for imported inputs and spare parts, while smaller components provided funding for enterprise rehabilitation and technical assistance. Policy reforms supported by the project fell into five areas: (a) price liberalization; (b) export promotion; (c) import liberalization; (d) improvement of the public investment program; and (e) encouragement of the private sector. 3. ISAC represented a significant break with Madagascar's past policies, but it was seen from the beginning as only the first step on the long road to adjustment. The Credit has since been followed by the Industrial and Trade Policy Adjustment Credit (ITPAC, Credit 1834-MAG and A-32-MAG) in FY87, which deepens and broadens the reform process begun under ISAC by supporting a market-oriented foreign exchange allocation regime, liberalizing foreign and domestic trade by lifting quantitative restrictions and price controls, and introducing a rationalized tariff structure. Separate adjustment operations are also being carried out in the agriculture (ASAC, Credits 1691-MAG and A-16-MAG), the public sector (PSAC, Credit 1941-MAG) and the financial system (Apex I, Credit 2104-MAG). B. Background 4. Madagascar. with a population of 10.9 million and a GDP per capita of US$190 in 1988, is among the poorest countries in the world. Almost 90 percent of the labor force is employed in the agricultural sector, which contributes about 40 percent of GDP. Industry accounts for only about 16 percent of GDP. 1/ The formal name of the project, as stated on the loan document, was 'Industrial Assistance Project.' However, the name ISAC, or its French equivalent CASI, is used almost exclusively in Bank files and in correspondence with the borrowers. It will, therefore, be used here. - 46- 5. The persistent economic difficulties encountered by Madagascar since the late 1970s can be traced to the economic policies pursued after 1972, which entailed an unprecedented expansion of the public sector and of government intervention in the economy. Foreign firms were nationalized and a large number of new public enterprises were created. Public investwent rose particularly rapidly after 1978, when the government embarked on a policy of massive investment. In the belief that national objectives could better be met by direct allocation of resources than by market signals, a policy of extensive intervention at all levels of the economy was initiated. A rigid and cumbersome system of price controls was instituted. Domestic firms were given high levels of protection, initially through the tariff system, and subsequently through administered foreign exchange allocation, tight import licensing requirements, quantitative restrictions and prohibitions. 6. This policy environment created an unfavorable climate for industry, discouraging private sector investment, sending incorrect signals that oriented economic activity towards inefficient import substitution. It also encouraged excessive use of imported inputs, discouraged exports and permitted inefficiencies to persist. Furthermore, to finance the higher rate of public investment, the government had to resort to borrowing abroad and to inflationary financing. This resulted in rapidly growing fiscal and current account deficits that peaked in 1980 at 18 percent of GDP, while the debt service ratio reached 72 percent in 1982. The exchange rate became increasingly overvalued and there was an inevitable and marked annual decline in GDP of 3.6 percent between 1980 and 1983. 7. The government initiated a program of macroeconomic stabilization, supported by the IMF, in 1982. Though progress was made towards restoring internal and external balance, the underlying supply-side problems remained. The government, therefore, while continuing its stabilization efforts, embarked on an adjustment program in 1984. C. Project Preparation and Appraisal 8. Initially, the project was envisioned as a second line of credit to BNI, the government-owned National Bank of Industry. Given the extent of the distortions in the Malagasy economy, however, a DFC-style operation was quickly discarded. It was also decided in early 1983 not to pursue a full structural adjustment operation. Though committed to reform in principle, the government was not at the time convinced of the necessity or desirability of a comprehensive reform program. The project, therefore, aimed at striking a balance between the need for far-reaching reforms and the government's fear of the unknown by starting the adjustment process with a program that was significant enough to lay the foundations for recovery and the resumption of growth and at the same time not too far reaching as to fail to command reasonable consensus. 9. The principal elements of the reform package adopted under ISAC were measures to begin the process of liberalization of domestic markets through the removal of price controls. Measures to reduce administrative constraints to exports and to begin the process of liberalization of imports were also included. Considerable debate took place as to whether more should - 47 - be done, especially on the reform of the foreign trade regime; but, in the absence of a significant exchange rate adjustment, import liberalization could not have been realistically more extensive. Ultimately, however, a more vigorous reform program was resisted by the government. Indeed, despite the limited reach of the policy reforms envisaged under ISAC (relative to the extent of the distortions), the possibility of internal resistance to the reforms was identified as the main risk facing the project. 10. It was stressed from the beginning that ISAC should be seen only as a first step in the reform process; further adjustment measures would be required. It was hoped that successful implementation of ISAC would convince the government to proceed with a more vigorous reform package. Funds were included as part of the ISAC package for studies to identify major constraints to efficient industrial development, analyze the incentives system, and make recommendations for further policy reform. Adjustment would thus be phased, with successive phases building on the success of previous phases and the results of studies carried out under them. 11. ISAC sought to boost capacity utilization in industry by making foreign exchange available for imports of inputs. It was hoped that relatively rapid improvements in output arising from the infusion of foreign exchange would ease the economic situation and lead to an increased commitment to the reform process. Increases in capacity utilization would also help mitigate possible adverse employment effects of the adjustment process, and provide incentive goods to stimulate increased agricultural production. Reflecting its origins as a DFC-style operation, ISAC retained a small investment component of SDR 5 million (8 percent of total funds), to be used to finance enterprise rehabilitation projects. 12. Work on the project began during 1983. Appraisal missions took place in January and June 1984. The Credit was approved by the Board on January 15, 1985, and signed on March 22, 1985. The IDA Credit was declared effective on August 28, 1985; the SFA Credit on March 3, 1986. The second tranche was released on September 22, 1986. 13. The project was jointly financed by IDA (Credit 1541-MAG, for SDR 40.2 million) and the Special Facility for Sub-Saharan Africa (Credit A-7-MAG, for SDR 18.9 million). SDR 33.7 million of the IDA Credit and all of the SFA Credit were allocated to finance industrial imports; this component was divided into two tranches of SDR 27.6 million (17.2 million from IDA and 10.4 million from SFA) and of SDR 25.0 million (16.5 million and 8.5 million), respectively. The balance of the IDA Credit went to the BNI for rehabilitation projects (SDR 5 million), for TA and consultant studies (SDR 1.3 million), and for civil works (SDR 0.3 million). 14. The quick-disbursing component of the Credit was channelled through the Central Bank, which then sold the foreign exchange to eligible enterprises through the commercial banks. The list of eligible enterprises was established on the basis of value-added eligibility criteria, calculated at international prices (para. 25). SDR 5 million of the total Credit was to be onlent to the National Bank for Industry (BNI) for the financing of - 48 - rehabilitation projects. Finally, SDR 1.5 million was made available to the Ministry of Industry, Energy and Mines (MIEM), which was responsible for the implementation of studies and technical assistance. D. Institutional Performance 15. Policy reforms supported by the project fell into five areas: (a) price liberalization: progressive elimination of price controls on industrial goods representing about 70 percent of industrial value added, so that only 31 products would remain under price control by second tranche release; (b) reform of the public investment program: agreement on a program to rationalize the industrial PIP; (c) export promotion: adoption of measures designed to stimulate exports, including the elimination of all export taxes; (d) import liberalization: elimination of import prohibitions on a total of 20 percent of goods subject to them; and (e) private sector encouragement: elaboration of a new investment code. Price Liberalization 16. A rigid and cumbersome price control system had been applied to all industrial goods since the late 1970s. It had resulted in shortages and distortions, and had either perpetuated industrial inefficiency or inflicted financial hardship on enterprises. Ex-factory prices of goods representing 35 percent of manufacturing value added were decontrolled prior to Board presentation; another 35 percent were decontrolled as a condition of second tranche release. This left only 31 products subject to price control; a number which was later further reduced to 10 goods, representing 15 percent of manufacturing value added. However, although administrative setting of prices was terminated, restrictions on profit margins remained. Although in practice these restrictions were looser than previous controls, they prolonged a sense of control over the private sector. Controls over profit margins were eliminated at ITPAC effectiveness in September 1987. Public Investment Program 17. A massive and generally ill-conceived Public Investment Program (PIP) had been one of the principal causes of Madagascar's economic and financial crisis in the early 1980s. Under ISAC, the government was to agree with the Bank on industrial PIPs for 1985 and 1986, stop further investment in existing projects until their viability could be determined, and obtain the Bank's agreement for any new projects expected to cost in excess of US$4 million. - 49 - 18. Agreement on industrial PIPs of USA10.6 million and US$17.8 million for 1985 and 1986, respectively, was reached. This represented a vast improvement over the unsustainable level of US$80 million attained in 1979. Perhaps equally important, the quality of projects included in the PIPs increased considerably. Although several projects of dubious viability remained, the worst offenders were dropped. In addition to the reduction in waste, these measures alleviated the public sector's demand for bank credit, thus diminishing the crowding out of the private sector. A more complete overhaul of the public sector still remained to be accomplished, however. Work on rationalizing the management of public sector resources has continued under ITPAC and PSAC. Export Promotion 19. In addition to the general anti-export bias of trade and macro- economic policy (particularly the seriously overvalued exchange rate), regulatory restrictions had contributed to discouraging exports. ISAC included a few export-promoting measures, such as abolition of export taxes and of export licensing requirements, and creation of a foreign exchange retention scheme for exporters. These measures were carried out as planned, but did not prove sufficient to generate an export response, given the continuing distortion in the overall structure of incentives. Import Liberalization 20. During the 1970s, imports had been subject to increasingly tight import controls and prohibitions. Under ISAC, a small number of prohibitions were lifted, and replaced by tariffs. Imports of the goods concerned soon took place, sometimes in considerable quantities. Although this represented a step in the right direction, it was clearly meant as a relatively small first step towards full liberalization of the trade regime and substantial distortions remained. Fulfilling this expectation, a far-reaching program of import liberalization has since been undertaken under ITPAC. Investment Code 21. The investment code enacted in 1973 to provide fiscal and other incentives had lacked precision and was applied in a discretionary manner by the government. A revised investment code taking into account the Bank's suggestions was presented to the National Assembly in 1985. Though this code remained relatively hostile to private entrepreneurship, it constituted a considerable improvement over the previous code: additional guarantees were provided against nationalization; the repatriation of dividends and profits was allowed for (though no mechanism was created to ensure this); and special conventions for particular classes of investors were reduced, leaving only a distinction between small- and medium-scale enterprises and large enterprises. Although the code was presented to the National Assembly on time, the implementing decree was not enacted until Hay 1988. By then, the changing economic environment had rendered the code obsolete. A new, more open, investment code was promulgated under PSAC in December 1989. - 50 - Rehabilitation Component 22. The BNI component was the least successful, due mainly to the unfavorable investment climate and BNI's slowness in processing applications for subloans and its parlous financial state. BNI had previously managed a DFC-style line of credit reasonably effectively, and was considered a sound institution at time of appraisal, but progress during implementation was much slower than expected. Only SDR 0.9 million had been committed by the June 30, 1986 commitment deadline. Extension of the project component was only agreed to following an audit of the BNI's portfolio and adoption of acceptable action plans to rehabilitate three enterprises which together accounted for about 30 percent of the non-performing portfolio (these measures were pursued further under ITPAC). Restrictions on use of the funds were relaxed several times. Even so, commitment rates remained low. Further, by limiting the use of the investment component funds to rehabilitation purposes, the Credit did not encourage entry of new enterprises. Ultimately, almost half the funds allocated to this component were reallocated to the general industrial imports component. Under subsequent Credits, efforts have been made to address the weakness of the domestic banking system and to encourage entry through a less restrictive regulatory and financial environment. 23. The success of the subloans that did get approved was mixed. Some loans, such as that to SOPEBO, an export-oriented fishing firm, was used effectively to help upgrade equipment needed to expand export activities. The BATA shoe factory loan, on the other hand, was used for equipment which later proved unsuitable for producing at competitive prices. In time, the company's performance, however, improved considerably. Technical Assistance 24. The technical assistance component of the Credit included support for the creation of a data bank for the MIEM and funding for various s udies which later operations drew upon. This component generally proceeded smoothly. Some additional funds (US$400,000) were reallocated to this component from the general imports component. E. Monitoring and Implementation 25. In addition to ensuring that policy conditionality was respected, a serious effort was made to assess the impact of the reform on the economy. Some results are summarized in Section VII and in the report cited in footnote 2, para. 30. Supervision was also required to verify that funds provided for imports were used appropriately. Eligibility criteria were established to avoid the foreign exchange being used by inefficient enterprises. These stipulated that the only eligible firms would be those whose ratio of value added to the value of inputs, measured in border prices, at least equalled 50 percent. An initial list of 80 enterprises was drawn up during negotiations. This list was subsequently expanded several times. However, several of the firms which were originally considered eligible were subsequently found to be non-viable. This was understood from the beginning as an interim and by no means perfect system of allocating foreign exchange that would be discontinued once a market-based system was in place. Once a - 51 - non-discriminating foreign exchange allocation regime was established in early .988, the eligibility list was eliminated. At this point, all remaining ISAC funds were allocated through the Open General Licence system (OGL). 26. The foreign exchange provided to eligible firms under ISAC was to be additional to the foreign exchange that they would otherwise have been allocated. This condition was included to avoid foreign exchange freed up by use of ISAC funds being reallocated to ineligible, inefficient firms. A formula was devised to determine whether this principle of additionality was respected. Provision of funds by the government fell somewhat short of what was required, but the condition was considered to have been met in view of the constraints in terms of overall availability of foreign exchange and uncertainties in the data. This complication was also resolved once funds begun to be allocated through the OGL. F. Disbursements 27. Use of project funds was delayed by difficulties in establishing Credit withdrawal procedures. New arrangements were made during a February 1986 mission, resulting in an increase in the pace of commitments and disbursements. Delays in the release of the second tranLhe resulted in a backlog of eligible expenses accumulating. Disbursements resumed after tranche release, but the pace remained slow. The closing date, originally set for September 30, 1987, was extended three times, first to September 30, 1988 and then to March 30, 1989. Since funds remained after the latter date had passed, and the Credit had not been formally closed, the closing date was once again extended, in March 1990, to September 30, 1990. 28. Only SDR 2.7 million were ultimately disbursed under the BNI component, or only 53 percent of the planned amount. Moreover, 34 percent of total disbursements were against a single subproject, which, because its size exceeded the maximum limit, was approved on an exceptional basis in 1988 (two years after the original commitment date for this component). The technical assistance and public works components proceeded essentially as planned, with some additional funds being reallocated to the TA component from the general imports component. G. Conclusions and Lessons Learned 29. It is difficult to assess results and draw lessons for ISAC individually, since it was only a first step in the direction of adjustment and was followed relatively quickly by other adjustment operations. The overall program of adjustment, which has included also ITPAC, ASAC, PSAC, and several IMF stand-by loans, is however considered relatively successful in the Sub-Saharan African context. - 52 - 30. Substantial adjustment has already taken place in the industrial sector.2 The manufacturing sector has increased its outward orientation considerably. Almost all subsectors have experienced higher capacity utilization. Significant entry of new firms has occurred, particularly in export-oriented activities. New export activities have been characterized by labor-intensive production technologies. The garment subsector has been particularly dynamic, with a threefold increase in production and significant increases in total employment since 1987. Much of the response has, however, been due to increases in outward orientation, so the measures adopted under ITPAC probably played a more important role. Without the initial impetus provided by ISAC, however, ITPAC may not have occurred. 31. The phased approach adopted in this sequence of operations appears to have been well suited to the circumstances for two reasons. First, it allowed government confidence and commitment in the reform process to be slowly built up. Both the government and important interest groups were at best skeptical of the need and desirability for adjustment at the time the process was begun, and at worst they were actively opposed. Positive experience under ISAC allowed these political obstacles to be gradually overcome. Second, it gave time for additional analytical work to be carried out and information and experience to be built up within the Bank, thus allowing a more comprehensive program to be developed. This latter factor was particularly important in light of the limited experience, at the time the project was first being planned, with this type of adjustment operations. 32. The danger with such an approach is that 'adjustment fatigue' might set in, and commitment to further reform would lag. This does not appear to have happened in this case. There was some limited backtracking on tariff reforms and delays in the public enterprise privatization process during implementation of ITPAC, but the program was soon set back on track. 33. The other significant drawback of the gradualist approach is that the funds could go to firms (including public sector firms) still operating in a highly distorted environment. This problem was particularly acute for the investment component of the Credit, but also applied to the general imports component. Doubts were expressed by the loan committee that the funds would be used efficiently with important distortions remaining. The BATA subloan, for example, was used for equipment which later proved inappropriate in the liberalized regime. In a sense, the BNI's slowness in processing applications proved to be a blessing in disguise, since subprojects which were eventually approved were conceived by their promoters and implemented under the more liberalized system established under ITPAC. These subprojects seem to have been reasonably efficient. The complicated tests to ensure 'adequate value added' and 'additionality' are another example of the problems caused by the partial approach. They were, however, useful to illustrate the need for a transparent non-administrative resource allocation system and, in any case, they were short-lived. 2/ A detailed analysis of the adjustment process in the Halagasy industrial sector can be found in "Madagascart Adjustment in the Industrial Sector and an Agenda for Further Reforms," Report No.7784-MAG, October 1990. - 53 - 34. An issue that is particularly acute in the context of such a phased approach is that of sequencing. This is because of the possibility that removing one distortion may not be welfare-improving if other distortions remain. Nor is it the case that all starting points will advance the reform process equally. A case could be made that the adjustment process should not have started with domestic reform, but rather with more vigorous foreign trade reform, which might have introduced compet,-ion into the domestic markets more rapidly and speeded the pace of adjustment. Such an approach was not, however, deemed politically feasible at the time. 35. An issue raised by all early sectoral operations concerns the advisability of extending balance of payments support to a particular sector rather than to the economy as a whole. This clearly risks directing available foreign exchange in ways that might not be optimal. (Now, all sector operations provide general balance of payments support.) In the case of ISAC, this problem was alleviated--but not eliminated--by a parallel agricultural sector adjustment operation. Of course, the introduction of the OGL, towards the end of the life of ISAC, alleviated this problem fully. 36. The period during which ISAC was being formulated also saw the beginning of a close cooperation between the Bank and the IMF in their operations in Madagascar. A high level of frequent communications was developed between the staff of the two institutions. This was quickly extended to include a core team from the government. The effects of such cooperation were not so evident on ISAC, but were critically important for ITPAC, which followad closely after ISAC. PART II A. Comments on the World Bank's Project Completion Report 37. The Industrial Sector Adjustment Credit (ISAC, Credit 1541-MAG for SDR 40.2 million and Credit A-7-MAG for SDR 18.9 million) marked the start of a new direction in Madagascar's economic policy. The economy badly needed the Credit during its external and internal financial stabilization, primarily to provide the necessary foreign exchange for imported inputs to be used in production and spare parts. Although it was very difficult to simultaneously carry out a policy to restart growth in production and reestablish internal and external balance, the Government was clearly committed to undertaking a series of reforms to increase the performance of the industrial sector. ISAC was the first step on that road. 38. With the start of the implementation of ISAC in 1985, the introduction of the Liberalized Import Regime (RIL) about mid-1987 and of the Liberalized Import System (SILI) in July 1988, private investments were again being made despite certain long-term persistent structural rigidities. Based on the 1985 Investment Code, 100 companies wera approved between 1985 and 1988, of which 65Z were in the industrial sector, most of them rehabilitated enterprises. - 54 - 39. Nevertheless, difficulties were encountered as mentioned in the Bank's PCR, but the phased approach to implementing the Credit was well suited to a difficult economic environment in which the reforms could be only a starting point for a market-oriented economy. The impetus generated by ISAC, supplemented by other macroeconomic measures, the total liberalization of imports and the establishment of the Investment Code of 1988, resulted in a steady growth of the production capacity of industrial enterprises. B. Report Prepared by the Borrower and Executing Agencies 1. Comments by the Ministry of Economy and Plan Introduction 40. The Industrial Sector Adjustment Credit (ISAC, Credit 1541-MAG for SDR 40.2 million and Credit A-7-MAG for SDR 18.9 million) was a hybrid credit that marked the beginning of the structural adjustment process in Madagascar. The project had four components: - quick disbursement for imports of production inputs and spare parts: 89Z - rehabilitation of industrial enterprises: 82 - technical assistance: 3% - public works: 1%. The Credit was approved by the Bank's Board of Directors on January 15, 1985 and became effective on August 28, 1985. The closing date was extended three times, ultimately being September 30, 1990. Objectives 41. ISAC's objective was to bring new vitality co the industrial sector, which had been weakened by long-standing stru,-ural rigidities, and to bring about economic policy changes designed to ! zrease the efficiency and eventual competitiveness of the industrial sector. The credit was a means of increasing supply during the difficult macroeconomic stabilization period necessary before the structural adjustment mentioned earlier could begin. Implementation Experience 42. Based on the reports prepared by the executing agencies (MIEM, BNI, Central Bank), the project was successful. The economic policy reforms supported by the project, namely, price liberalization, export promotion, import liberalization, rationalization of the public investment program, and encouragement of the private sector, were begun but were difficult to implement for a host of reasons: - ISAC is the first step on the road to reform; - 55 - - The recommended reforms could have had the major effects sought with a regard to supply but the conditions for implementing the measures were not ir. place; i.e., the existence of a climate favorable to a recovery in production; - A macroeconomic stabilization policy could not be pursued at the same time that supply was expected to respond. Restrictions on demand, especially imports (RIL, SILI), although necessary, limited the increase in supply. The weakness of industrial development brought in its wake a s,ries of operating characteristics: multiple external dependencies, particularly for capital goods, inputs and means of production, insufficiency of backward-forward linkages, limited absorption capacity, relatively low level of technical know-how. 43. The executing agencies ran into various problems, in particular: delays in the utilization of the funds, inability of the industrial enterprises to award a minimum order of US$52,000 per supplier for imports of spare parts, and the significant exchange risks borne by the industrial enterprises. On this latter problem, although the exchange risk was borne by the State in the subsequent APEX credit, it is not yet evident that industrial development is going in the desired direction. 44. In point of fact, because of the need to develop a favorable climate for production in particular, training should ultimately create a cadre of national entrepreneurs capable of promoting joint ventures with foreign investors. Thus, although the credit was not regularly and fully utilized, this does not mean that the absorption capacity of the economy is limited. 2. Comments by the Ministry of Industry, Energy and Mining 45. Prior to the involvement of the World Bank, MIEM had already established certain criteria for the allocation of the foreign exchange required by the industrial and mining enterprises for importing inputs and the spare parts needed for their operations. Those criteria were ranked by priority. - First priority: Exporting enterprises - Second priority: Enterprises manufacturing basic necessities and intermediate goods - Third priority: Other enterprises. In other words, the scarce foreign exchange were already being used conscientiously, and the enterprises clearly understood the need for such action despite their initial reticence. Moreover, the list of selected enterprises was almost identical to that consistent with priorities one and two above. - 56 - 46. As regards the delay in utilizing the funds for which the closing date was extended three times: (a) When the first tranche of Credit 1541-MAG was released, priority was given to the selected enterprises with import licenses but whose foreign ex.change account transactions were blocked by the Central Bank. Accordingly, import licenses had to be checked pending Central Bank approval of such transactions by those enterprises before they could be incorporated in ISAC 1541-MAG. (b) The required rinimum order of US$52,500 per supplier delayed the filing of additional applications, especially for imports of specific parts, as US$52,500 is not a negligible amount for the small and medium enterprises, which had had financial difficulties following the 1981 crisis. (c) There were delays with the replenishment of the special accounts and application procedures. 47. Utilization of ISAC (Part A) required the establishment of a special "Monitoring Committee," composed of representatives of BCM, BNI, BTM, BFV, MFB, MIEM and MC, who met every Wednesday. Duties: - Assess the situation regarding: - status of account - pending applications - disbursements situation - Raise problems - Identify action to be taken - Act As a result: - An overrun of the overall amount was avoided by establishing a reserve taking into account fluctuations in SDR/US$ exchange rates and changes in freight costs. - The credits not used by certain selected enterprises, which preferred to use the SILI not subject to the fixed rate of 5r, could be reallocated to other interested and efficient entities. - Virtually all ISAC funds were utilized. 48. Lastly, ISAC and ITPAC played an undeniably important role in the new outward orientation (export-oriented market production). Furthermore, other credits are now under way as part -f structural adjustment: ITPAC, PSAC, APEX, etc.; ISAC's experience should be used to avoid the same errors. What our enterprises lack now is working capital to meet the constraints of liberalization. - 57 - 3. Comments by the National Bank of Industry (BNI) 49. our comments refer to the part with which we are directly involved; i.e., the component involving the financing of rehabilitation projects, and in particular para. 22. 50. First of all, the statement that the above-mentioned component was the least successful, owing mainly to BNI's slowness in processing applications for subloans, requires an explanation. In point of fact, as we already indicated in the Project Completion Report for IDA Credit 977-MAG, the exchange risk penalizes clients and was already causing financial difficulties for many anterprises. The persistence of that problem, together with the rest of the unfavorable investment climate prevailing at the time, resulted _'A clients continuing to manifest their reluctance especially to contract loans denominated in other currencies. It was moreover for that reason that at the same time we were able to satisfactorily use other sources of funds such as the long-term refinancing facility of the Caisse Centrale de Coop6ration Economique (the main advantage of which is the fact that the enterprises do not bear the exchange risk, and under certain conditions can be granted a relatively low preferential rate). 51. In addition, BNI alone accounted for more than 60% of that line of credit to be used by all local banks. The exchange risk problem, moreover, also forced Soci6td JB, one of the most prosperous firms in the country, to seek cancellation of the loan that had been accorded it under IDA Credit 1541-MAG. 52. Furthermore, as regards the volume of lending under the BNI component, the ultimate overall amount was SDR 3,246,925 (i.e., 65% of the funds allocated to it) and this, after the withdrawal of Soci6td JB as well as the rejection by IDA of the request from REFRIGEPECHE-0UEST for partial financing of a trawler in the amount of SDR 1,300,000, which we were ul:imately able to negotiate under the Caisse Centrale line discussed above. 53. Lastly, we would stress the measures taken by BNI to promote and support small and medium enterprises through the establishment of venture capital accumulated from our earnings that enabled us to increase the funds for projects with firms such as RUBIS, GAMO and L'ELEVEUR, to give only a few examples, in the form of joint ventures, and without which those project would not have been able to meet the criteria for loans under IDA Credit 1541-MAG, and, consequently, could never have been financed. 54. The next two tables summarize the indicators for each credit operation financed under the BNI component of Credit 1541-MAG. CREDIT 1541-MAG, COMPONENT B AT JANUARY 30, 1991 in SDRs (Credits denominated in Malagasy francs) Outstanding Client Amount granted Amount utilized Repayments balance at April Comments (FMG) 30, 1991 (FMG) 31 - SHE 52,000.00 SDR 28,312.95 SDR 26,297,126 28,297,126 Regular repayments, good 52,594,252 FMG prospects 32 - RUBIS SA 48,555.00 SDR 47,010.46 SDR 23,346,993 7 250,000 Regular repayments thanks to On 95,096,993 FMG the efforts of the promoters despite difficulties related to liberaiization B3 - GAMO CI 99,210.00 SDR 86,104,48 SDR Regular repayments, good 177,198,517 FMG 39,377,449 13?.921,068 prospects B4 - FTT 8,000.00 SDR 7,961.86 SDR 6,517,234 7.775,848 Regular repayments 16,293,082 FMG 121,000.00 SDR 114.939.87 SDR - 236,849,266 Project in startup phase, B5 - L'ELEVEUR 236,849,26 G good prospects Total 328,765.00 284,278.62 SDR 96,038,802 471,993,308 Equivalent 578,032,110 PMG CREDIT IDA 1541-MAG, COMPONENT 3 AT APRIL 30, 1991 (Credits denominated in SDRe) Client Amount granted Amount utilized Repayments Outstanding balance Coments (SDR) (SDR) at April 30, 1991 1 - ZRAOMA 500,000 483,596.48 700,000.00 383,596.48 Normal progress 2 - BATA 486,700 446,980.44 746,980.44 0 Repaid in full 3 - Soci6tE JB 420,000 - - - Credit canceled 4 - SF0I 420,000 417,766.67 76,360.00 341,406.67 Norwal progress 5 - SAMIMAD 401,460 402,514.06 60,584.00 241,930.06 Project in startup phase 6 - SOPEBO 1,110,000 1,030,299.19 1,730,299.19 0 Repaid in full TOTAL 3,338,160 2,781,156.84 1,714,223.63 966,935.21 1
Группа Всемирного банка · Project Performance Assessment Report
Madagascar - Industrial Assistance Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Project Performance Assessment Report
Страна
Мадагаскар
Источник
Всемирный банк