Madagascar - Industry and Trade Policy Adjustment and Public Sector Adjustment Credits Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report Nn. 14774 PERFORMANCE AUDIT REPORT MADAGASCAR INDUSTRY AND TRADE POLICY ADJUSTMENT CREDIT (CREDIT 1834-MAG AND A-32-MAG) AND PUBLIC SECTOR ADJUSTMENT CREDIT (CREDIT 1941-MAG) JUNE 30, 1995 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Currency Unit: Malagasy Franc (FMG) 1985 - US$1 = FMG 662 1986 - US$1 = FMG 676 1987 - US$1 = FMG 1069 1988 - US$1 FMG 1407 1989 - US$1 FMG 1603 1990 - US$1 = FMG 1494 1991 - US$1 = FMG 1835 1992 - US$1 FMG 1868 1993 - US$1 = FMG 1928 ABBREVIATIONS AND ACRONYMS ASAC - Agricultural Sector Adjustment Credit BFV - Commercial Bank (Banky Fampandrosoana ny Varotra) BNI - Industrial Development Bank (Bankin'ny Indostria) BTM - National Rural Development Bank (Bankin'ny Tantsaha Mpamokatra) EMSAP - Economic Management and Social Action Support Project EPZ - Export Processing Zone FMG - Malagasy Franc GDP - Gross Domestic Product ISAC - Industrial Sector Adjustment Credit ITPAC - Industry and Trade Policy Adjustment Credit OGL - Open General License PAR - Performance Audit Report PCR - Project Completion Report PIP - Public InVestment Program PSAC - Public Sector Adjustment Credit FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General June 30, 1995 Operations Evaluation MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Madagascar - Industry and Trade Policy Adjustment Credit (Credit 1834-MAG and A-32-MAG) and Public Sector Adjustment Credit (Credit 1941-MAG) Attached is the Performance Audit Report on the Madagascar Industry and Trade Policy Adjustment Credit (ITPAC, Credit 1834-MAG and A-32-MAG, approved in FY87) and the Public Sector Adjustment Credit (PSAC, Credit 194 1-MAG, approved in FY88), prepared by the Operations Evaluation Department. The main objective of these loans was to remove barriers to the efficient allocation of both public and private resources by liberalizing international and domestic trade, rationalizing the public enterprise sector, strengthening the public investment and budgetary formulation process, reforming the civil service, and restructuring the state-owned banking system. The bulk of the ITPAC reforms were successfully implemented: the exchange rate regime was reformed, import tariffs were reduced and rationalized and quantitative restrictions lifted, most controls on prices and profits were removed, and steps were taken to improve the public investment program and the condition of the financial sector. Real per capita GDP grew during the period 1988-90 for the first time since the early 1970s, while private investment and labor-intensive, non-traditional exports surged. The third tranche was delayed two and a half years, however, by difficulties with public enterprise reform, and some conditionality was finally shifted to PSAC. PSAC recorded some gains, most notably the introduction of private participation in the banking sector and reform of the budgetary process, but failure to resolve several fundamental issues led to delays in the second tranche and eventually to cancellation of the third tranche. In particular, the Government suspended the public enterprise reform program, adopted an unacceptable civil service statute, and failed to control public expenditures. A prolonged period of political upheaval during 1991-92, accompanied by reversals of some of the policies adopted earlier, also disrupted the adjustment process. The outcome of ITPAC is rated satisfactory, its sustainability uncertain, and its institutional development impact modest. The outcome of PSAC is rated unsatisfactory, its sustainability uncertain, and its institutional development impact negligible. These ratings are the same as those in the PCRs. Major lessons from the two projects include the need for Government participation in the design of an adjustment program to ensure internalization of the reforms; the importance of well-defined, meaningful performance indicators; the value of privatization as a tool for achieving sustainable reform of public enterprises; and the importance of political stability in sustaining the reform process. Comments from the Swiss Government emphasize the overoptimistic assessment of risks by the Bank, the weak oversight of financial management, and the importance of effective aid coordination. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. I FOR OFFICIAL USE ONLY Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Basic D ata Sheet................................................... 5 Evaluation Sum m ary ................................................ 1 1. Background ................................................. 15 2. The Adjustment Program ......................................... 17 ITPAC: Credit Objectives and Policy Instruments ...................... 18 ITPAC: Implementation Experience and Outcome ...................... 18 PSAC: Objectives and Policy Instruments .......................... 20 PSAC: Implementation and Outcome ............................. 21 Social Impact of Adjustment ................................... 26 Other Findings............................................ 29 3. Sustainability of the Adjustment Process and Lessons of Experience ... ............................................... 31 Sustainability . ............................................ 31 Lessons of Experience ....................................... 32 Annex 1: Key Indicators ............................................ 35 Annex 2: Comments by Federal Office for Foreign Economic Affairs, Switzerland ....... .36 This report was prepared by Alice Galenson (Task Manager), who audited the projects in February 1995. Debra Brummell provided administrative support. 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 wiLhout World Bank authorization.  3 Preface This is the Performance Audit Report (PAR) on the Madagascar Industry and Trade Policy Adjustment Credit (ITPAC), for US$83 million equivalent and on the Madagascar Public Sector Adjustment Credit (PSAC), for US$132 million equivalent. ITPAC was approved on June 30, 1987, and became effective on September 21, 1987. The Closing Date was extended one year and three months to March 31, 1991. Cofinancing in the amount of US$17 million equivalent was provided by a Japanese grant and Japanese and Saudi credits. PSAC was approved on June 29, 1988, and became effective on December 16, 1988. The third tranche, with a balance of US$45.6 million equivalent, was canceled. The Closing Date was extended two years and nine months to September 30, 1993. Cofinancing in the amount of US$50.4 million equivalent was provided by an African Development Fund credit and Swiss and Norwegian grants. The PAR is based on the Project Completions Reports,' prepared by the Africa Regional Office; the President's Reports; the loan documents; economic and sector reports; President's Reports, Staff Appraisal Reports and Performance Audit Reports of related credits; project files; and interviews of Bank staff, Government officials, and private operators. An OED mission visited Madagascar in February 1995. The mission discussed the origins, implementation and impact of the two Credits with current and former officials in the Ministries of Economy and Plan, Finance, Civil Service and Agriculture; the Central Bank and two commercial banks, BFV and BTM; and the President's office. The PCRs provide a good account of the program experience. The PAR builds on their findings and explores the wider context of adjustment in Madagascar, with particular attention to the relationship among the adjustment credits, the experience with the public enterprise restructuring and privatization component, the social impact of adjustment and the factors affecting sustainability. The outcome of ITPAC is rated satisfactory, and its sustainability uncertain. The institutional development impact was modest. The outcome of PSAC is rated unsatisfactory, and its sustainability uncertain. The institutional development impact was negligible. The draft PAR was sent to the Borrower and the cofinancing agencies for comment. Comments received from the Swiss Federal Office for Foreign Economic Affairs are included as an Attachment. 'The PCR for ITPAC was released to the Executive Directors and the President on April 15, 1992, and the PCR for PSAC on April 14, 1995. The Government was sent Parts I and III of the ITPAC PCR in September 1991, and of the PSAC PCR in March 1994, but did not prepare comments in either case.  5 BASIC DATA SHEET INDUSTRY AND TRADE POLICY ADJUSTMENT CREDIT (Credit 1834-MAG and A-32-MAG) Credit Position (Amounts in millions) As of December 31, 1994 Original Disbursed Cancelled Repaid Outstanding IDA 1834-MAG (US$) 16.00 16.92 0.00 0.00 18.11 SFA A-32-MAG (US$) 67.00 70.76 0.00 0.00 75.64 Sub-total (US$) 83.00 87.68 0.00 0.00 93.75 Saudi Loan (RIALS) 12.00 12.00 0.00 0.00 12.00 OECF Loan (YEN) 1,000.00 1,000.00 0.00 0.00 1,000.00 Japan Grant (YEN) 1,000.00 1,000.00 0.00 0.00 0.00 Sub-total (YEN) 2,000.00 2,000.00 0.00 0.00 1,000.00 Cumulative Estimated and Actual Disbursement (Amounts in millions) FY87 FY88 FY89 FY90 FY91 IDA 1834-MAG (US$) Planned 4.17 12.50 12.50 12.50 12.50 Actual 2.78 8.50 8.51 8.51 12.50 SFA A-32-MAG (US$) Planned 17.40 52.20 52.20 52.20 52.20 Actual 12.50 29.68 29.85 29.85 52.20 Total (US$) 15.28 38.18 38.36 38.36 64.70 Actual Total as Percent of 71 59 59 59 100 Planned Total OECF Credit (YEN) 250.25 475.77 1,000.00 Japan Grant (YEN) 1,000.00 Saudi Loan (RIALS) 6 Project Dates Original Actual Initiating Memorandum 10/30/86 Letter of Development Policy 06/05/87 Negotiations 04-05/87 Board Approval 06/30/87 Signature:IDA, SFA, Saudi Loan 07/10/87 01/20/87 Japanese Loan/Grant 07/10/87 Credit Effectiveness: IDA, SFA, Saudi Loan 09/21/87 Japanese Loan/Grant 04/19/88 First Tranche Release 09/21/87 Second Tranche Release 01/88 04/18/88 Third Tranche Release 07/88 12/27/90 Closing Date 12/31/89 03/31/91 Mission Data No. of No. of Staff Report Month/Year Weeks Persons Weeks Date Preparation 06/86 10/30/86 Appraisal 11-12/86 3 8 16 12/16/86 Post-Appraisal 02/87 1 3 3 03/02/87 03/87 2/3 2 5 04/03/87 Supervision 09/87 2 2 4 10/07/87 01/88 2 2 4 02/05/88 02/88 2 3 6 03/01/88 06-07/88 4 3 12 07/20/88 10/88 1 1 1 10/88 01-02/89 2 3 4 02/13/89 06/89 1 1 3 07/17/89 10-11/89 2 3 6 11/14/89 07/90 1 1 2 08/02/90 09/90 1 1 1 09/22/90 Completion 02-03/91 1 1 1 03/91 FOLLOW-ON ADJUSTMENT OPERATIONS Public Sector Adjustment Credit (Cr. 1941-MAG) approved on June 29, 1988. 7 BASIC DATA SHEET PUBLIC SECTOR ADJUSTMENT CREDIT (Credit 1941) Credit Position (Amounts in millions) As ofDecember 31, 1994 Credit Original Disbursed Cancelled Repaid Outstanding IDA 1941-MAG (US) 125.00 76.76 45.58 0.00 83.32 IDA 1941-1-MAG (US$) 1.40 1.37 0.00 0.00 1.59 IDA 1941-2-MAG (US$) 1.20 1.29 0.00 0.00 1.30 IDA 1941-3-MAG (US$) 1.70 1.63 0.00 0.00 1.74 IDA 1941-4-MAG (US$) 1.30 1.38 0.00 0.00 1.45 IDA 1941-5-MAG (US$) 1.37 1.41 0.00 0.00 1.43 Sub-total 131.97 83.84 45.58 0.00 90.83 Swiss Grant (SwF) 10.00 10.00 0.00 0.00 0.00 ADF (UA) 30.00 Norwegian Grant (NKr) 20.00 Cumulative Estimated and Actual Disbursements (Amounts in millions) FY89 FY90 FY91 FY92 FY93 FY94 IDA 194 1-MAG (US$) Planned 50.60 39.90 Actual 22.40 2.35 31.82 0.17 0.61 0.16 IDA 1941-1-MAG (US$) Actual - 1.10 - - - - IDA 1941-2-MAG (US$) Actual - - 0.90 - - - IDA 1941-3-MAG (US$) Actual - - - 1.20 - - IDA 1941-4-MAG (US$) Actual - - - - 1.00 - IDA 1941-5-MAG (US$) Actual - - - - - 0.99 Total Actual 22.40 3.45 32.72 1.37 1.61 1.15 Swiss Grant (SwF) Actual 1.94 8.06 ADF (UA) Norwegian Grant (NKr) 8 Project Dates OriginalVPlanned RevisedlActual Dates Credit Dates Preparation Mission 10/20/87 Initiating Memorandum - 01/15/88 Appraisal 01/31/99 01/31/88 Negotiations 04/88 05/02/88 Letter of Development Policy - 05/88 Board Approval 06/88 06/29/88 Signing of Credit Agreement - 07/18/88 Effectiveness 08/88 12/16/88 Effectiveness of 1941-1 (1st IDA reflow) - 08/25/89 Effectiveness of 1941-2 (2nd IDA reflow) - 07/23/90 Second tranche release 05/89 03/91 Third tranche release 12/89 Cancelled Effectiveness of 1941-3 (3rd IDA reflow) - 05/29/91 Effectiveness of 1941-4 (4th IDA reflow) - 11/30/92 Effectiveness of 1941-5 (5th IDA reflow) - 07/15/93 Closing Date 12/31/90 09/30/93 Bank Staff Inputs FY86 FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 Total Identif/Preparation 3.7 12.8 16.5 Appraisal 103.5 103.5 Negotiations 36.4 36.4 Supervision 59.9 40.1 28.1 28.8 43.9 200.8 Project Completion Report 10.0 10.0 Total 3.7 12.8 139.9 59.9 40.1 28.1 28.8 43.9 10.0 367.2 9 Mission Data No. of No. of Staff Report Month/Year Weeks Persons Weeks Date Preparation 10/87 3 9 18 12/02/87 Appraisal 01/88 3 6 15 02/26/88 Supervision 10/88 3 8 15 12/14/88 02/89 4 6 16 03/17/89 06/89 3 4 10 07/11/89 10/89 3 7 18 11/21/89 12/89 2 2 4 12/22/89 03/90 1 1 1 04/03/90 04/91 1 2 2 05/07/91 01/92 2 2 4 02/04/91 03/92 3 6 8 05/17/92 06/92 3 1 2 07/14/92 08/92 2 4 2 09/18/92 09/92 2 6 8 09/18/92 12/92 3 2 4 01/07/93 Completion 08/93 3 7 4 08/30/93  11 Evaluation Summary Background 1. With a per capita GDP of US$230, Madagascar is one of the poorest countries in the world. For many years, the country pursued a strategy of industrial import substitution, with a major public sector role and pervasive controls on private sector activity. The late 1970s saw a series of large, economically non-viable public investments which, together with deteriorating terms of trade, led to a decline in real GDP and a large external debt burden. Efforts at financial stabilization began in the early 1980s, followed by structural adjustment, starting with the industrial sector in 1985 and the agricultural sector in 1986. 2. As it became clear that sectoral reforms by themselves were unlikely to lead to a sustained supply response, the Government turned to broader reforms. The Industry and Trade Policy Adjustment Credit (ITPAC, US$100 million equivalent, including cofinancing) was approved on June 30, 1987 and declared effective on September 21, 1987. The second tranche was released in April 1988, three months late, and the third tranche was released 2 1/2 years late, in December 1990; the Closing Date was extended to March 31, 1991. The Public Sector Adjustment Credit (PSAC, US$180 million, including cofinancing) was approved on June 29, 1988 and declared effective on December 16, 1988. The second tranche was released in March 1991, 22 months after the originally scheduled date, and the third tranche, with a balance of US$45.6 million equivalent, was canceled. The Closing Date was extended to September 30, 1993. Objectives 3. ITPAC aimed at liberalizing international and domestic trade by: * devaluation and establishment of a non-discriminating foreign exchange allocation regime; * elimination of quantitative restrictions and simplification and reduction of the tariff structure; and * elimination of most price and profit margin controls. Complementary reforms were designed to: * rationalize the public investment program; * restructure and/or privatize 10 poorly performing public enterprises; and * clean up the portfolios of the state banks, while preparing action programs to encourage competition and improve Central Bank supervision. 12 4. PSAC attempted to improve the efficiency of the allocation and use of public resources by: * strengthening the formulation and monitoring of public expenditure and investment programs and initiating civil service reform; * rationalizing the public enterprise sector through a program of liquidation, divestiture or restructuring; and * restructuring the banking system, with private participation. PSAC also included monitoring of the trade reform agreed under ITPAC, removal of export restrictions and formulation of a health sector policy. Other measures to alleviate the short term negative social impact of adjustment, along with technical assistance, were included in the companion Economic Management and Social Action Support Project (EMSAP). Implementation Experience 5. The bulk of ITPAC was successfully implemented. The overvaluation of the exchange rate was corrected; access to foreign exchange was freed; quantitative restrictions on imports were eliminated and a four-year program of tariff reform adopted; most price controls and remaining profit controls were lifted; satisfactory public investment programs were agreed for the relevant years; and the condition of the financial sector was substantially improved. The third tranche was delayed by two and a half years, however, due mainly to difficulties with the public enterprise reform component, and uncompleted conditionality was eventually shifted to PSAC. 6. Implementation of PSAC was more problematic. Some components were successful-- the continuation of tariff liberalization, export liberalization, the introduction of private participation in the banking sector, and the reform of the budgetary process--but other aspects of the program encountered serious delays, and failure to resolve certain fundamental issues eventually led to cancellation of the third tranche. In particular, the Government suspended the public enterprise reform program, adopted an unacceptable civil service statute, and was unable to control public expenditures. In addition, standard procedures were bypassed in allocating some of the foreign exchange under the project. Results 7. The price and trade liberalization carried out under ITPAC led to significant improvements in the economy, including growth in real per capita GDP during the period 1988-90 for the first time since the early 1970s. Industrial capacity utilization rose and non- traditional exports surged. 8. The public enterprise reform showed far less favorable results. Despite the liquidation or privatization of some 70 public enterprises, at least 100 commercial and financial enterprises remain under state ownership. They account for about 14 percent of GDP and a quarter of modern sector employment, and they dominate many sectors of the economy. In 1991, these enterprises received net transfers from the Government and the banking system 13 equal to 8 percent of GDP and had total indebtedness equivalent to 25 percent of GDP. Although the banking sector is now partially private, the portfolios of the two majority state- owned banks are once more in serious trouble, with major repercussions for the economy. 9. The social impact of adjustment appears to have been largely beneficial, particularly for the rural poor. Certain groups were targeted for assistance through EMSAP, but lengthy delays in that project meant that no results could be felt during the active life of PSAC. 10. Meanwhile, economic liberalization had led to a demand for political liberalization, which erupted in mid-1991 with civil disturbances and political turmoil that lasted for 18 months and which, while relatively peaceful, disrupted economic activity and derailed the reform process. Some policy reversals occurred, the currency once more became significantly overvalued, real GDP fell, and the elected Government which emerged in 1993 was not convinced of the need for adjustment. New liberalization steps have since been taken, but structural imbalances remain, and serious talks have only just resumed between the Government and the World Bank and IMF. Sustainability 11. Even before the political events of 1991, sustainability of the adjustment process was not assured. Growth of GDP and exports, as well as revenue mobilization, were lower than expected, and lax credit policies weakened the financial sector. The suspension of adjustment during the long political transition, and the failure to resume reform once the new Government was in place, cast further doubt on the sustainability of the reforms achieved to date. 12. The latest economic projections for Madagascar target 6 percent annual growth of GDP by the year 2000, a rate never even approached during the last 20 years. The forecast is based on a strategy of export promotion linked to production diversification and private sector development. Achievement of these objectives would require firm government commitment to a renewal of the adjustment effort, including a disengagement of the public sector from productive and commercial activities, an improved environment for private sector investment, privatization of the remainder of the banking sector, and elimination of remaining barriers to free markets, along with macroeconomic stability. 13. This is an ambitious agenda, and the record of past reform efforts does not inspire confidence in its feasibility. However, some of the elements--trade liberalization and improved budgetary processes--are already in place, and the success of these reforms demonstrated that strong government commitment can bring results. Internalization of the reforms by the Government will be essential for any future reforms. In addition, the Government must signal its intentions unambiguously to the private sector if it is to generate the required supply response. Findings and Lessons 14. The experience of reform under ITPAC and PSAC offers a number of lessons. Trade liberalization and budgetary reform succeeded because of intensive Bank efforts and close Government participation, leading to internalization, whereas reform of public enterprises and 14 the civil service suffered from a lack of understanding and commitment. The enterprise-by- enterprise approach used in ITPAC and expanded under PSAC allowed conditionality to be met by focussing on small, unimportant enterprises, and did not generate an agreement on principles. A more effective approach would have defined the principles of reform and applied them first to the enterprises most burdensome to the economy. Active Government participation in preparing the component, along with up-front actions, would have helped establish commitment to reform--or demonstrated its absence right at the beginning. Closer attention by the Bank to warning signals during preparation and implementation might have allowed actions to increase the likelihood of success or a decision to proceed more cautiously with PSAC. 15. Well defined indicators are essential to monitoring progress. The conditionality for trade reform was based on explicit quantitative measures, such as specific reductions in maximum and average tariffs. Public enterprise conditionality, on the other hand, called for "satisfactory progress" in liquidation, divestiture and restructuring, but these actions were not well-defined, and it was difficult to distinguish among them or to determine whether they had actually taken place. 16. Projects aimed at reducing poverty were designed to accompany the adjustment program, but the Government did not publicize them widely, and, in the end, they were substantially delayed. Greater efforts are needed to ensure that the social components of an adjustment program are carried out in a timely manner. In the present environment of hostility to further adjustment, it is more important than ever to develop a cohesive safety net and to publicize it well. 17. Privatization is a powerful tool for preventing the reversal of reforms to public enterprises. Cleaning up the bank portfolios in Madagascar did not prevent the excesses from recurring; had all the banks been privatized, they would have been far less likely to permit the credit expansion that set off a liquidity crisis in 1991. 18. It is essential to carry out the divestiture process in a transparent manner. In Madagascar, the initial privatizations took place with little openness, leading to suspicion and a loss of Government credibility and undermining the entire program. In particular, the privatization process raised fears of excessive foreign participation in the economy and of the potential for unbalancing ethnic relationships. Creation of a privatization fund to facilitate broader Malagasy participation could have alleviated these fears. 19. Discussions between the Government and the Bank are now focussing on the prerequisites for resumed adjustment lending. A proposed new Adjustment Credit would apply lessons from the past to the critical areas of private sector development and macro- economic and public sector management. Conditionality would be limited to a few key measures in each area. The strategy calls for a series of one-tranche operations, with all conditionality to be accomplished before Board approval; during each set of negotiations, agreement would be reached on monitorable actions that could trigger the subsequent operation. A follow-up project to EMSAP would promote complementary social actions. 15 1. Background 1.1 Following independence in 1960, Madagascar experienced a decade of modest growth, at about 3 percent p.a. A military coup in 1972 brought in a government that favored state control of the economy, characterized by a strategy of import substitution, with widespread nationalization of economic activity, strengthening and extension of central regulation and price controls, government control over most agricultural marketing, and a corresponding reduction in the role of market forces in allocating resources.2 The Government that came to power in 1975 escalated, broadened and codified such policies. In the late 1970s, faced with stagnating economic activity, the Government tried to diversify the economy and accelerate development through a rapid expansion in public sector investment. The resulting large, uneconomic projects were financed mainly by external borrowing and money creation. Imports of domestically produced goods were prohibited, and quantitative restrictions and tariffs were imposed on the remainder of imports. 1.2 This environment had a major negative impact on the structure of the economy. Heavy protection implied little or no competition from abroad, and the regime for foreign exchange allocation, along with other bureaucratic licensing procedures, constrained domestic competition. Exports were discouraged by restrictions on goods deemed to be of vital national interest. This inward-looking trade regime led to small monopolistic domestic markets. The Government responded by imposing controls on prices and profit margins, thereby distorting price signals and furthering the misallocation of resources. Marketing monopolies were established for cereals, agricultural inputs, basic commodities and transportation. Managers of public enterprises failed to consider economic costs, large factories and agricultural units operated at well below capacity, and a capital-intensive technology was adopted which required imported inputs and spare parts whose purchase aggravated the foreign exchange situation. Smallholders were neglected, and the private sector stagnated. Many of the nationalized and most of the newly created public sector enterprises survived only with subsidies and the help of credit from the three state banks? 1.3 This situation, together with deteriorating terms of trade and stagnant domestic revenues, led to a severe financial and economic crisis: by 1980, the fiscal and current account deficits had reached 15 percent of GDP, and external public debt had grown to US$1 billion, nearly one-third of GDP. Debt service obligations rose from about 4 percent of export earnings in 1978 to 72 percent in 1982, and inflation reached levels of 30 percent p.a. By 1982, due also in part to a population growth rate of close to 3 percent p.a., real per 'he discussion in this section is based in part on the following World Bank reports: "Madagascar: Adjustment in the Industrial Sector and an Agenda for Further Reforms," Report No. 7784-MAG. October 29, 1990; "Madagascar: Industry and Trade Policy Adjustment Program," President's Report No. P-4488-MAG, June 5, 1987; "Madagascar: Industrial Assistance Project," Program Performance Audit Report No. 10405, February 27, 1992; and "Madagascar: Agricultural Sector Adjustment Credit," Performance Audit Report No. 13081, May 23, 1994. 'One indication of the pervasive inefficiency is the finding in 1985 that for 32 of 80 firms analyzed (representing 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 (Report P-4488). Madagascar also shified from being self-sufficient in food to being an importer of rice, the main staple. 16 capita GDP had fallen an estimated 28 percent from its 1973 level; today Madagascar remains one of the world's poorest countries, with a per capita GDP of US$230. 1.4 Efforts at financial stabilization began in the early 1980s, supported by IMF Stand-by agreements. Policies focussed on demand management, including tax increases, expenditure cuts and control systems, bank credit ceilings, increases in the prices of goods and services from public enterprises, a sharp reduction in imports, and actions on the exchange rate and consumer subsidies. The Malagasy franc (FMG) had been fixed at 50 FMG to the French franc; in 1982 it was devalued and pegged to a trade-weighted basket of currencies. Beginning in 1983 a flexible exchange rate policy was adopted, with adjustments based on the rate of domestic inflation. The nominal price of rice distributed through official channels more than doubled and consumer subsidies on imported rice were eliminated. These measures brought the budget deficit down to 5 percent of GDP and the current account deficit to 9 percent of GDP by 1983. Inflation fell to around 10 percent by the mid-1980s, and GDP grew at a modest rate of 1.4 percent p.a. during the period 1983-87 (still a decline in per capita terms). 1.5 While necessary and effective, stabilization did little to address the underlying problems of the economy. The reduction in the budget deficit was achieved in part by replacing explicit subsidies to public enterprises with credit from the nationalized banking system, which itself developed serious financial difficulties; and the improvement in the current account of the balance of payments resulted from a decline in imports, rather than any increase in exports. Agricultural output fluctuated: production of some food crops grew significantly, but rice continued to be imported; coffee exports stagnated due to low producer prices, shortages of inputs and transportation bottlenecks; cotton production grew rapidly in response to increased producer prices, but the production of other industrial and export crops stagnated or declined due to inappropriate pricing and marketing policies which reduced incentives to producers and traders. Industrial output declined sharply between 1980 and 1984 and grew modestly in 1985 and 1986, due to external financing of inputs and parts. Nontraditional exports fell sharply, in part because of the still overvalued exchange rate. Transport costs escalated due to a deterioration of the network, and low administered tariffs discouraged the provision of services. Further policy reforms were needed to induce the hoped-for supply response. 17 2. The Adjustment Program 2.1 The initial approach to adjustment was sectoral, given the socialist Government's fear of a more comprehensive reform program and too sudden a break with the past. The Industrial Sector Adjustment Credit (ISAC), in 1985, was intended to lay the groundwork for further reform. It supported the progressive removal of price controls on the majority of industrial goods (not actually completed until 1989); rationalization of the public investment program; import liberalization for basic necessities and goods produced by local monopolies; export promotion, including the elimination of all export taxes and streamlining of export procedures; and adoption of a more private sector friendly investment code (which was, however, never implemented). 2.2 The reforms led to some increase in capacity utilization of existing plants, but output remained constrained by continuing erosion of domestic demand and lack of export competitiveness. Although foreign exchange allocation to the least efficient firms--those with negative value added at world prices--was reduced, resources were not reallocated to more efficient firms. Price controls were replaced by maximum profit margins. Meanwhile, supervision of a line of credit component of ISAC revealed serious constraints on financial liquidity linked to the condition of the parastatal sector: in 1987, public enterprises absorbed over 50 percent of bank credit, crowding out the private sector, and over half of the loans were non-performing. Probably the most valuable contribution of ISAC was the opportunity it provided for a collaborative effort between the Government and the Bank in undertaking the detailed analytical work that supported the ultimately successful trade reform. 2.3 ISAC was followed in 1986 by the Agricultural Sector Adjustment Credit (ASAC) which, above all else, supported the liberalization of the production and marketing of rice. Rice became freely traded, producer prices increased (as did prices to consumers, who are no longer subsidized), imports were drastically reduced, and private mills began to replace state- owned companies. Other reforms were less successful, and after ten years of stabilization and adjustment, agricultural sector performance remains disappointing: between 1984 and 1990, output grew by only 1.5 percent p.a. The growth came mostly from rice; traditional export crops, not liberalized and facing poor prospects in international markets, stagnated or regressed. A road rehabilitation program was initiated to remove bottlenecks in the transport sector, and the road transport industry was deregulated, but inadequate infrastructure remains a major problem. 2.4 As it became evident that sectoral policy reforms by themselves were unlikely to provoke a sustained supply response, and buoyed by its success in the liberalization of rice and other markets, the Government moved to broaden the reforms, with support from the Industry and Trade Policy Adjustment Credit (ITPAC) in 1987 and the Public Sector Adjustment Credit (PSAC) in 1988, as well as from IMF Stand-by loans. 18 ITPAC: Credit Objectives and Policy Instruments 2.5 ITPAC's primary goal was to remove barriers to the efficient allocation of resources by liberalizing international and domestic trade through reform of the exchange rate regime, removal of import prohibitions, import tariff reform, and additional price decontrol. Complementary reforms were designed to rationalize the public investment program, restructure and/or privatize some of the many poorly performing public enterprises, and improve the banking system. More specifically, ITPAC contained the following components: a) Exchange Rate Regime. Significant devaluation, leading to the establishment of a non-discriminating foreign exchange allocation regime and introduction of an Open General License (OGL) system covering all imports. b) Tariff Regime. Elimination of quantitative restrictions and simplification and reduction of the tariff structure, beginning with the establishment of maximum tariffs of 80 percent and minimum tariffs of 5 percent, and agreement on a four year program to reduce average nominal protection from 46 to 35 percent, with a dispersion around the average reduced from 45 to 15 percent. c) Price Decontrol. Elimination of most remaining price controls and all profit margin controls. d) Public Investment Program (PIP). Agreement on the content of the PIP (for all sectors) for 1987, 1988 and 1989 (technical assistance was provided to improve the preparation, control and supervision of the PIP). e) Public Enterprises. Establishment of an improved framework for managing public enterprises and formulation of an action plan for restructuring the sector, to include closing or selling nonviable enterprises and criteria for rehabilitating and possibly privatizing viable ones; agreement on rehabilitation plans for 10 enterprises, in addition to 16 previously agreed (funds were included for studies of the 10 enterprises). f) Financial Sector. Preparation of action programs to enhance competition and permit private participation in the banking sector; audits and portfolio restructuring of three state-owned banks; agreement on systems to improve the efficiency of supervision by the Central Bank; and a study of the need for further financial sector reforms. ITPAC: Implementation Experience and Outcome 2.6 Most of the agreed reforms were carried out in a relatively timely manner. The overvaluation of the exchange rate was corrected, access to foreign exchange was freed, quantitative restrictions on imports were eliminated and a four-year program of tariff reform adopted, remaining profit controls were lifted (although some price controls continued until 1989), satisfactory PIPs were agreed for the relevant years, portfolio restructuring plans were implemented for the three state-owned commercial banks, steps taken toward private sector participation in the banking sector, and new rules and procedures were adopted for Central 19 Bank supervision. Problems quickly emerged, however, in the public enterprise reform component. 2.7 Trade liberalization was the cornerstone and the major achievement of ITPAC. This component was well prepared, building on detailed studies and discussions between the Bank and the Government over a period of several years, bolstered by continuity in staffing on the Bank side, and leading to the eventual internalization of the reforms by the Government; evidence of this internalization is the fact that the Government resisted efforts by a number of firms to restore their protection and that the rationalization of tariffs has continued even beyond the agreed targets at a time when other reforms were floundering. Setbacks occurred- -the OGL was suspended for a time when shortages of foreign exchange reemerged--but the trade regime in effect today goes farther than was envisioned at the time of ITPAC, with the tariff structure recently reduced and narrowed further to a range of 10 to 30 percent. On the basis of a study carried out under ITPAC, an export processing zone (EPZ) was established in 1990 which allows firms exporting 100 percent of their output to operate on a completely "off-shore" basis, i.e., free of import duties, exchange controls and most taxes. 2.8 By mid-1988 a supply response began to emerge. The years 1988-90 showed significant improvements in the economy, including growth in real per capita GDP (though at only 0.5 percent p.a.) for the first time since 1971. The industrial sector responded to the liberalized trade regime: industrial capacity utilization rose and non-traditional exports surged. The latter have continued to do well; exports of garments, for example, rose from US$19 million in 1992 to an estimated US$52 million in 1993, 15 percent of merchandise exports. Well over 100 operations have been approved for EPZ status, all in the private sector, creating an estimated 20,000 jobs by the end of 1993 (5 percent of total formal sector employment).' 2.9 The most problematic aspect of ITPAC was the public enterprise reform component, which encountered obstacles almost immediately, foreshadowing the even more serious problems of PSAC. During ITPAC preparation (in March 1987), Bank staff noted that the public enterprise component had to rely on an enterprise-by-enterprise approach because they could not agree with the Government on a global approach; eight months later, in the context of PSAC preparation, another memo raised the same issue and questioned whether the Government was committed to reform. A memo describing the consultants' reports that had been submitted on seven enterprises (February 1988, five months after ITPAC effectiveness), pointed out that they contained no definite action plans and no mention of future ownership, and predicted that the Government would not be receptive to either privatization or liquidation. A back-to-office report on a mission to discuss second tranche release (March 1988) stated that after two intense weeks of discussions, the mission had been able to persuade the Government of the financial and technical merits of the public enterprise component, begging the question of why this had not been accomplished earlier. The release of the second tranche, originally scheduled for January 1988, was delayed until April, and was possible only then with a waiver putting off completion of action plans for the rehabilitation 'Me 1991 CEM ("Madagascar: Beyond Stabilization to Sustainable Growth," Report No. 9101-MAG, June 1991) was optimistic about the country's prospects: the main scenario, which assumed a consolidation and expansion of policy progress, forecast that GDP growth would stabilize at 6 percent p.a., while even the lower, alternative scenario for the future showed some continued per capita growth. 20 of 7 of the 10 enterprises until December 1988. Third tranche release, originally scheduled for July 1988, was delayed nearly 2 1/2 years, until December 1990, largely because of continuing problems with this component.10 At that time, remaining conditionality for public enterprises was shifted to PSAC. 2.10 Most of the 10 enterprises slated for reform under ITPAC were unprofitable and held little interest for the private sector. In the end, 6 were liquidated and 3 sold to private investors, but as the PCR points out, the reluctance shown by the Government to carry out even this modest first step in reducing its direct role in the economy cast doubt on its commitment to liberalization and discouraged private investors. Madagascar's legal environment was not conducive to privatization, particularly with foreign participation: obstacles abounded in land ownership, dividend payment, capital repatriation, foreign currency denominated local bank accounts, and employee retention requirements. The new Investment Code approved under ISAC was never implemented; a revised code, implemented in December 1989, remained complex and discriminatory. The absence of a well functioning capital market further discouraged private participation, particularly by the bulk of Malagasy citizens. Even had the atmosphere been more favorable to reform, however, the impact would have been limited by the fact that the selection of enterprises ignored some of the worst performers. In retrospect, it is clear that this component was unlikely to achieve its ultimate objectives; the experience of PSAC confirms that this was the case. PSAC: Objectives and Policy Instruments 2.11 PSAC focussed on improving the efficiency of allocation and use of public resources by strengthening the formulation and monitoring of the public expenditure programs, initiating reform of the civil service, rationalizing the public enterprise sector, and restructuring the banking system. It also followed up on the trade reforms initiated under ITPAC. The specific components were: a) Public Expenditure Policies. Definition of detailed priorities for current and development expenditures and agreement on three-year public expenditure and investment plans; formulation of strategies for the main economic and social sectors; strengthening of the institutional basis for public expenditure programming; formulation of a consolidated budget including all sources and uses of funds; and a census of the civil service, followed by a plan of action and timetable for reform. b) Public Enterprises. A three year action plan to rationalize the public enterprise sector, to be undertaken in three phases, and including the liquidation, divestiture or restructuring of 51 enterprises. c) Financial Sector. Continuation and extension of ITPAC conditionality, including adoption of a new banking law permitting participation of private capital in existing banks and providing for the establishment of new private commercial banks and other financial '*Declaration of PSAC effectiveness in December 1988 may have taken much of the pressure off the Government to meet the conditions for the third tranche of ITPAC. 21 institutions; and action plans to complete the portfolio clean-up and restructuring of the three existing banks, including private participation (with private majority control in at least one case). d) Export Promotion and Trade Liberalization. Publication of a law formalizing the removal of most administrative export restrictions; agreement on an action plan to restructure agricultural export marketing; and monitoring of the tariff rationalization plan agreed under ITPAC. e) Social policies. Formulation of a comprehensive health sector strategy. Other measures to alleviate the short term social costs of adjustment, along with technical assistance to support the policy reforms, were to be supported by the parallel Economic Management and Social Action Support Project (EMSAP), which was also approved in 1988.11 PSAC Implementation and Outcome 2.12 The goal of PSAC--to improve the efficiency of use of public resources--was, and remains, relevant to the economic development of Madagascar. Some aspects of PSAC-- export liberalization and reform of the budgetary process--were very successful and have laid the groundwork for improved growth and efficiency. Other components, however, were less well designed and failed to bring about the hoped for changes. Taken as a whole, and despite the accomplishment of some if its goals, PSAC must be judged unsuccessful in achieving its primary objective. 2.13 As noted above, even during the preparation of PSAC there were signs of incipient problems. An internal Bank memo commented at one point that the Minister of Finance wanted to delay the next phase of adjustment in order to have some breathing space to consolidate ongoing activities, yet two months later a telex announcing a preparation mission referred to the ambitious accelerated timetable. The back-to-office report from an identification/preparation mission (September 1987) recorded the Government's strong opposition to the introduction of private banks or the opening up of the banking system to private sector equity participation; it also noted that with elections coming up in 1988 and 1989, the Government could not take steps such as closing a major public enterprise that might increase the level of unemployment. Another back-to-office report, in December 1988, on the eve of PSAC effectiveness, expressed serious concern with emerging problems, particularly with public enterprise divestiture, and warned that they could delay the second tranche. 2.14 The implementation experience of PSAC was rockier than that of ITPAC; the program encountered serious delays, and failure to resolve certain fundamental issues eventually led to cancellation of the third tranche. Meanwhile, economic liberalization had ""Democratic Republic of Madagascar: Economic Management and Social Action Project," Staff Appraisal Report No. 7410- MAG, October 27, 1988. 22 led to a growing popular sentiment for political liberalization, which erupted in mid-1991 with civil disturbances and political turmoil that lasted for 18 months. A transitional coalition Government was put in place in January 1992 and a new Constitution adopted. Presidential elections were held in February 1993, and the opposition candidate elected; legislative elections were held in July 1993, and a new prime minister and cabinet were in office by the end of August. 2.15 While remarkably nonviolent, the political transition disrupted economic activity and halted the reform process. GDP fell by 7 percent in 1991 and grew by only 1 percent in 1992 and 2 percent in 1993; the budget deficit rose to over 10 percent of GDP. A growing shortage of foreign exchange led to the reimposition of import restrictions and foreign exchange rationing. The elected Government that finally emerged in 1993 was not convinced of the need for adjustment: foreign borrowing rose, on unfavorable terms and for questionable purposes. In May 1994, the Government resumed stabilization and adjustment efforts, floating the exchange rate and abolishing import prohibitions, but its commitment to reform has continued to waver, and serious talks have only just resumed between the Government and the World Bank and IMF. Thus it remains to be seen whether the adjustment process can be put back on track. 2.16 While it had a major negative impact on the economy, and probably would have derailed even a successful reform program, the lengthy period of political uncertainty cannot be held solely accountable for the failures of PSAC, which was originally scheduled to close in December, 1990. As noted above, problems, particularly with the key public enterprise component, surfaced immediately. Technical assistance for preparation of some of the reforms proved difficult to mobilize. Release of the second tranche was delayed nearly two years, to March 1991, and even then became possible only through the waiver of one condition (completion of a proposal for a new budgetary system), which was shifted to the third tranche. In September 1993, the third tranche was canceled when it became clear that the new Government would not meet three remaining conditions: a) Even after reducing the list of remaining enterprises to be divested to 11, this component could not be achieved: the transitional Government had suspended the public enterprise reform program and it was never resumed. b) The 1993-95 public investment/expenditure plan contained expenditures 21 percent above the level that would have been acceptable to the Bank. c) While the Government had frozen all hiring and agreed with the Bank on an action plan for reforming the civil service, the new civil service statute actually adopted in 1993 contained provisions that subverted the goals of the action plan. 2.17 Public enterprise refonn, a central focus of PSAC (originally called public enterprise reform and adjustment), was its least successful component. The nationalizations and heavy public investment of the 1970s had resulted in a large parastatal sector, comprising about 170 public enterprises by 1987. These enterprises, many of them inefficient and loss-making, imposed a heavy fiscal burden on the Government (the majority received subsidies from the budget), and generated a major misallocation of resources in the economy. The stabilization program of the early 1980s had reduced the subsidies, but the Government-owned banking 23 sector kept the enterprises alive, putting itself at risk, and crowding out the private sector. Privatization was expected to reduce these burdens by bringing in experienced and autonomous private managers, with technical and marketing know-how, who would take a significant commercial risk and have a major stake in their success. It would also send a clear signal of state disengagement to domestic and foreign private investors and encourage expanded private investment. Enterprises that could not succeed commercially would be closed. 2.18 The Government was not fully in accord with this scenario. Preparation of the privatization component was less thorough than that of the trade reform: the studies that were done did not clearly demonstrate the costs of badly managed public enterprises, and while the Government may have recognized the need to rid itself of nonperforming enterprises, neither it, nor the public, understood why it should privatize those that appeared to be operating adequately. The Government therefore proposed a list of enterprises which included many that were empty shellsl2 and others that would not have been viable under the liberalized trade regime and would have closed down in any case. Instead of insisting that the exercise concentrate on the most burdensome enterprises (10 to 15 in all), the Bank accepted the Government's list. While it was useful to clarify the status of these enterprises, doing so did not matter very much. Despite the seemingly large numbers of enterprises involved, the privatization program was too modest in scope to have a major impact on the budget, the banking system or the economy in general. 2.19 Other design problems emerged: * Reform was not grounded in a clear, well publicized statement of policy regarding the roles of the public and private sectors, and there were no guidelines or criteria to evaluate public enterprises for eventual liquidation, divestiture, or restructuring within the public sector. * Too many Government agencies were involved in the transactions, often causing conflicts and delays; even after the creation of an independent Government agency in 1990, the program lacked effective leadership and management. * A number of legal and administrative issues complicated the process, including procedures for judicial review, cross-ownership of assets among enterprises, decisions on which shares would be reserved for employees, questions about the responsibility for existing debts and arrears to suppliers, and an outmoded and rigid bankruptcy law which blocked liquidations. * National savings were insufficient to buy all the enterprises, particularly given the absence of financial instruments to facilitate widespread participation, but the Government was ambivalent about selling them to foreign investors. 12One estimate puts at nearly half the proportion of enterprises on the Government's list that existed in name only and were not in operation. 24 Finally, conditionality was vague, calling for "satisfactory progress" in public enterprise reform, without defining the criteria to be used for judging this progress; much time was spent simply debating whether various actions were satisfactory or not. (In contrast, the conditionality for trade reform was specific and easily monitored.) 2.20 Second tranche release took place nearly two years late, in March 1991, based inter alia on "satisfactory progress of public enterprise reform," but three months later an analysis of this component" identified a long list of problems: * There was little internalization by the Government of the reform objectives - the impetus came rather from Bank conditionalities. * Little attention was being given to the design and implementation of rehabilitation and commercialization action plans for enterprises that were remaining state- owned. * The list of enterprises to be liquidated or privatized for the third tranche conditionality was unclear and difficult to monitor. * It was hard to get a clear picture of what had already taken place, there being no consistent definitions of liquidation and privatization. * There was no centralized information system and no reliable information on, for example, the number of public enterprise staff that had been laid off or the level of remaining Government shareholding. * The liquidation of some enterprises had been going on for more than two years. * Most privatizations had been achieved through the dilution of Government equity rather than outright sale. * Negotiations with the private sector were hampered by Government reluctance to welcome foreign investment, lack of access by foreigners to real estate, and lack of transparency. Initially, for example, there was no competitive bidding, and information about sales was not made public; questions about the validity of a number of sales diminished the credibility of the Government and impeded the continuation of the program. Reforms designed to make the process more competitive and open were adopted only in 1992, too late to have any impact. 2.21 In September 1992, the Bank, acknowledging that the transition Government lacked a mandate to make decisions, but insisting on some signs of commitment, agreed to reduce public enterprise conditionality to the divestiture of 11 of the most important ones and to extend the closing date. However, when the elected Government that came to power the "Madagascar - CASEP (Public Enterprises Component) - Back to Office Report," June 28, 1991. 25 following June did not move to fulfill the final conditions, the third tranche of PSAC was canceled. 2.22 The public enterprise reform program did have an impact, but much less than had been expected." As of March 1993, 70 public enterprises had reportedly been divested, including 40 liquidations and about 30 privatizations. In practice, however, many of the recorded liquidations may remain incomplete. Most of the divestitures were in agriculture, agro-processing, manufacturing and mining. Payments received totalled FMG 14 billion by 1991, very small compared to the estimated FMG 3,500 billion remaining in Government hands. (Few, if any, divestitures took place after 1991.) As of 1993, at least 100 commercial and financial enterprises remained under state ownership. They accounted for about 14 percent of GDP and employed about 100,000 people, or a quarter of modern sector employment. While many are medium-sized enterprises, in manufacturing and agriculture, state holdings are concentrated in a small number of very large enterprises, often with monopoly/oligopoly control over their sectors: nine firms accounted for 80 percent of public enterprise value added in 1991. The public sector continues to dominate many areas of the economy, including agro-processing, mining, several major manufacturing industries, utilities, a significant segment of the trading sector and hotel industry, infrastructure and the financial sector (the two state-owned banks hold two-thirds of banking system assets, and the insurance industry is a state-owned monopoly). In 1991, the latest year for which data are available, public enterprises received net transfers from the Government and the banks equal to 8 percent of GDP and had total indebtedness equivalent to 25 percent of GDP. 2.23 The budgetary reform component of PSAC was successful, and, as in the case of ITPAC, the success resulted from prolonged collaboration between the Government and the Bank, supported by technical assistance funds under EMSAP and sustained by continuity of staff on the Bank side, which fostered the internalization of the reforms. Budgetary reform had the added advantage that it did not threaten any entrenched interests. Although seriously delayed, a new budget system was adopted in mid-1992 which included the functional analysis of expenditures and the consolidation of all expenditures and revenues. New implementation procedures were adopted in early 1993, fully meeting PSAC objectives. Even after cancellation of the third tranche, the Government continued to improve th6 budget system and has developed the capacity to maintain the system without external assistance. Unfortunately, the improved budget process could not ensure that expenditures would be held to satisfactory levels. 2.24 Another important achievement of PSAC was the removal of export controls and the abolition of state export monopolies for all but one product (vanilla), carried out largely up- front as conditions of Board presentation and credit effectiveness. 2.25 The experience with financial sector reform was mixed. ITPAC conditionality had been fully met, and the portfolios of the three state-owned banks were cleaned up by the end of 1989. PSAC conditionality went further and was also fulfilled: the Banking Law of May "Information on progress of the privatization program is taken from "Madagascar: New Horizons - Building a Strategy for Private-Sector, Export-Led Growth - A Private Sector Assessment," Report No. 14385-MAG, February 28, 1995 (green cover). The expected impact of this component had not been quantified in terms other than numbers of enterprises. 26 1988, enacted in the context of PSAC negotiations, established supervisory guidelines for the Central Bank and allowed for the participation of private capital in existing banks, as well as the creation of new private financial institutions. A new private bank started operations in 1989, and in 1991 foreign banks acquired shares in two of the three state-owned banks (majority in BNI and minority in BFV). In retrospect, however, it is clear that these reforms did not go far enough. In 1990 and 1991, a breakdown in Central Bank supervision and in the internal information and control systems of BTM, the remaining wholly state-owned bank, permitted a large and uncontrolled expansion of credit to the private sector, mostly to finance imports, fueling a foreign exchange liquidity crisis and constituting a major factor in the suspension of the OGL import system in September 1991. Problems in BTM eventually spilled over to the other majority state-owned bank, BFV. This experience afforded an important lesson, that the clean up of the banking system should have been linked with institutional and administrative safeguards that would prevent a recurrence of the problem. The lesson has been assimilated: the Financial Institutions Development Technical Assistance Project" supports improved banking supervision and the privatization of the remaining majority state-owned banks, and the latter is a priority of future adjustment lending. 2.26 Civil service reform was another area where Government commitment was missing. This component experienced major delays, starting with difficulties in agreeing on a proposal and selecting a consultant for the first step--a census of the civil service--and continuing through realization of the census (which led to the elimination of about 3 percent of the civil service identified as "ghosts") and development of an action plan for reform, finally approved in April 1992. Contrary to the agreed action plan, however, the new Government adopted a civil service statute in 1993 which included a system of seniority, rigid classifications and additional benefits that had substantial fiscal implications; this became one of the grounds for cancellation of CASEP's third tranche. Social Impact of Adjustment 2.27 Echoing Government concerns,7 the President's Report for ITPAC" considered the social impact of adjustment from two viewpoints: the possible loss of employment in the short term from the closure of firms and the inflationary impact of devaluation. Studies of the efficiency of Malagasy industries had indicated that the problem lay more in inefficient lines of production than in totally inefficient firms, so it was expected that labor could be "Total imports rose by 45 percent in 1990 (in nominal dollar terms), while traditional exports continued to fall, due to bad weather and falling world prices. Delays in aid disbursements (including the ITPAC third tranche and the PSAC second tranche) exacerbated the foreign exchange shortage. By the end of the year, reserves had fallen from over 7 months worth of imports to less than 2 months. The exchange rate was allowed to appreciate significantly. ("Madagascar: Beyond Stabilization to Sustainable Growth," Report No. 9101-MAG, June 1991.) "Staff Appraisal Report No. 11682-MAG, April 28, 1993. "In a letter to the President of the World Bank, before the Board presentation of ITPAC, the President of Madagascar raised questions about the possible impact of adjustment on inflation and employment, and pointed out the need for measures in the areas of medicine and housing to alleviate the social repercussions. These themes were picked up again in the context of PSAC. "Report No. P-4488, June 5, 1987. 27 reallocated to a large extent within firms. (The report did not address the possibility of massive lay-offs from public enterprises, since ITPAC contained only a relatively small public enterprise component.) Employment was also expected to expand in agriculture and in small- scale operations or self-employment in response to adjustment. As for inflation, an analysis of the import content of the consumer price index for low income urban groups indicated that the inflationary impact of devaluation would be modest. Factors such as increased competition and expanded capacity utilization in efficient activities were expected to offset any upward pressure on prices. Experience with the 46 percent devaluation of the FMG in 1987 bore out this analysis. Finally, the report noted the substantial social costs from the downward economic trends that could be expected to continue in the absence of adjustment. 2.28 During PSAC preparation, a considerable degree of attention was given to measures that could alleviate any potential social cost of the program. The President's Report noted that Madagascar's social problems originated mainly from the persistent decline in per capita income, and suggested that the next phase of adjustment should address these problems, cushion the specific impact of adjustment on certain groups and meet more effectively the general social objectives of development." The Letter of Development Policy outlined the Government's social policies in a number of areas, including health, nutrition, employment generation and population issues. However, to avoid delays to PSAC, these measures were separated from the adjustment operation and grouped, along with technical assistance, in the Economic Management and Social Action Project (EMSAP). 2.29 EMSAP included a variety of components: * an emergency malaria control program; * expansion of family planning services; * pilot projects in food security; * financing of labor-intensive road works in targeted areas; * a redeployment program for workers laid off as a result of public enterprise reform; * establishment of capacity to undertake and monitor social programs; and * short term technical assistance, studies, training and other support to the adjustment process. 2.30 EMSAP was extremely demanding on both local and Bank resources: the components were only loosely defined, and each Ministry that was involved managed its own components, with no central coordination.' Success of such a design would have required intensive "Report No. P-4780-MAG, June 10, 1988. 20A central coordination unit was established in 1993. 28 coordination and supervision by the Bank, neither of which was forthcoming (the Task Manager for EMSAP changed several times during its implementation). Furthermore the project did not offer a coherent strategy to combat poverty and to reassure the people that the Government was concerned with their well-being. The project experienced extensive delays and did not really get underway until several years after the start of PSAC; only 25 percent of the funds had been disbursed by the time PSAC closed. Thus although attention was paid to the idea of alleviating the social costs of adjustment, in reality, few measures of substance were actually carried out until five years after the start of the project they were meant to accompany. 2.31 One group targeted for special assistance was the employees of public enterprises scheduled to be closed or privatized. The Government established a fund to cover the required severance pay for laid-off workers, and EMSAP included a fund to assist in their "reinsertion" in the work force through training and assistance in identifying and initiating new activities. However, rather than publicize the availability of the latter fund, its managers were practically forbidden to contact the personnel of public enterprises until after a decision had been taken for their disposition. Thus a powerful tool for reassuring workers and forestalling unrest was foregone. Lengthy delays in initiating reemployment assistance--up to four years--meant that many of the workers concerned had already spent their severance pay and did not have the means for investing in the start-up of a new activity. Altogether, about 3,000 people (less than half that expected), from about 15 enterprises, were assisted by this program through 1994. 2.32 Given the lack of growth in Madagascar's economy and the interruption to the adjustment process, it is difficult to identify the social impact of the reforms that did take place. On balance, however, the adjustment process appears to have had a positive impact on the bulk of Madagascar's population. Exchange rate depreciation, trade liberalization and decontrol of agricultural prices benefitted the agricultural sector and small farmers, who make up the bulk of the poor. The urban poor were hurt by the elimination of rice subsidies during the stabilization in the early 1980s, but the trade liberalization of the later 1980s appears to have benefitted not only small farmers, but also the urban poor and nonfarm rural poor. Before 1986, wide fluctuations of consumer prices and periodic rice shortages had hurt consumers; following the liberalization of rice markets, rice became available in adequate quantities throughout the year, and price fluctuations remained moderate. The greatest burden of adjustment may have fallen on the urban rich, who had previously enjoyed rents derived from the now greatly reduced trade restrictions.2' 2.33 There is little evidence that the political upheavals of 1991 were linked to the social costs of adjustment, and, in fact, none of the parties taking part in the ensuing elections advocated an end to the adjustment effort. Nonetheless, more timely action on a number of fronts could have helped to cushion certain groups from negative effects and might have led to a more successful implementation of the reform program. The current efforts to resume adjustment are being sharply criticized by certain groups, at least partly on social grounds, "Dorosh, Paul A. "Structural Adjustment, Growth, and Poverty in Madagascar: A COE Analysis." Cornell Food and Nutrition Policy Program, Monograph 17, February 1994. 29 and more visible efforts in the past to deal with the social impact might have given the present Government greater credibility in its efforts to reassure the public. Other Findings 2.34 The audit for the Agricultural Sector Adjustment Credit (ASAC) identified a need for better coordination among successive adjustment operations, particularly when they are handled by different divisions within the Bank. While not questioning the validity of the reforms themselves, the audit pointed out that the major devaluation concurrent with Board approval of ITPAC reduced demand for fertilizers funded under ASAC and that the new licensing procedures for allocating foreign exchange introduced under the former slowed down ASAC disbursements, which were based on a more cumbersome bidding procedure. Only after a two-year delay in the release of the second tranche of ASAC were its disbursement mechanisms modified.' Better coordination between the two Bank divisions handling these projects might have speeded up the necessary changes in ASAC. 2.35 Similarly, the public enterprise component of PSAC, while less far-reaching than intended, did have some unintended results. The liquidation of project agencies involved in two of the Bank's agricultural projects resulted in the collapse of farmer organizations and a total lack of maintenance of irrigation systems; the result has been a decline in irrigated area and agricultural production, as well as increased opposition by the public to privatization. In one case, the agency was abolished before civil works had been completed, resulting in poor supervision and quality of work? 2.36 In 1992 a much delayed annual audit revealed a major irregularity in the use of PSAC funds during the period December 1989 to September 1990: US$9 million had been drawn directly from the Central Bank in favor of one local importer, bypassing OGL procedures, and only $50,000 equivalent had been reimbursed in local currency. The Government replaced the missing funds, and in 1993 arrested those said to be responsible; they are presently on a status of "provisional liberty." More stringent application of procurement and auditing guidelines by the Bank could have uncovered this abuse much earlier. 2.37 In retrospect it is clear that the risk analysis, particularly for PSAC, was inadequate. The President's Report noted that "The risk of policy slippage is alleviated by the Government's five-year record of satisfactory implementation of agreements with the Bank and the Fund. Also, the risk is minimized by the important measures already taken by the Government before Board presentation." However, problems with the centerpiece of PSAC, the public enterprise reform, had appeared even before Board presentation, in the context of ITPAC implementation. The President's Report expressed the belief that the reform process had been internalized, but subsequent events demonstrated that this was not the case. A more cautious approach would have entailed more detailed preparation of this component; more '"Madagascar: Agricultural Sector Adjustment Credit," Performance Audit Report No. 13081, May 23, 1994. '"Madagascar: Mangoky Agricultural Development Project, Study of the Plain of Antananarivo TA Project, Second Mangoro Forestry Project, Lac Alaotra Rice Intensification Project, and Cotton Development Project," Performance Audit Report No. 12646, December 30, 1993. 30 intensive discussions and more precise definitions of conditionality might have either revealed the lack of Government commitment or helped to generate that commitment, as had occurred in the case of the tariff reform under ITPAC. A second category of risk, that of insufficient Government administrative and institutional capacity was to be minimized through the provision of technical assistance in EMSAP. Technical assistance projects, however, often lag behind the projects they are designed to support, often because they are not accorded as high a priority for supervision. Such was the case with EMSAP, which was also delayed because of Government reluctance to accept foreign technical assistance. The lack of timely technical assistance contributed to delays in PSAC, particularly in the budgetary and civil service reforms components; and the design of EMSAP itself overburdened the Government's administrative capacity. 31 3. Sustainability of the Adjustment Process and Lessons of Experience Sustainability 3.1 The audit of the first adjustment operation, ISAC, noted that "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 ... have pushed the governance issue to the forefront of any assessment of sustainability of the Malagasy adjustment effort."I Written two years ago, at a time when the transition Government was still in place, this statement remains valid. 3.2 Even before the political events of 1991, sustainability of the adjustment process was far from assured. While a supply response did begin to emerge in 1988, it was not as strong as had been expected. GDP growth during 1988-90 was 3.5 percent p.a., compared to a projected 4 percent. Per capita income and consumption actually fell during the period, exacerbating the poverty situation. Revenue mobilization was lower than expected--tax revenues reached only 11 percent of GDP, compared to the forecast of 16.6 percent--resulting in a fiscal deficit of over 10 percent of GDP instead of the projected 2.4 percent. In retrospect, as the PCR for ITPAC points out, trade reform should have been linked to domestic tax reform. Exports grew, but at 6 percent p.a., not the projected 10 percent. Lax credit policies contributed to a surge of imports in 1990 and a current account deficit of 13 percent of GDP. The PCR for PSAC notes that "Altogether, the economic performance during 1988-90 demonstrated that adjustment can work in Madagascar. However, it also showed that reforms had not been strong enough to eliminate structural imbalances in the fiscal and external accounts. Also the policy backtrack of the early 1990s showed that the improvement was fragile, and that the structural adjustment stance was not deeply internalized." Events during and after 1991 confirm this assessment. 3.3 Some lasting reforms have emerged from the adjustment process, most notably the trade and exchange rate liberalization and the reform of the budgetary process. These reforms are necessary and significant and help to set the stage for future growth. However, they are not sufficient for that growth to take place. The latest economic report for Madagascar' proposes a sustained growth scenario, with a target of 6 percent annual growth by the year 2000, based on a strategy of export promotion, diversification and private sector development. The recent private sector assessment presents a policy agenda for such a strategy, calling for a shift in the roles of the public and private sectors whereby the public sector would disengage '"Madagascar: Industrial Assistance Project," Program Performance Audit Report, No. 10405, February 27, 1992, p. 34. 25"Madagascar: A Strategy for High Growth and Poverty Alleviation--An Economic Strategy Note," Report No. 13274-MAG, June 29, 1994. 32 from productive and commercial activities and reform the environment for private sector investment. Based on the historical record of adjustment in Madagascar, these projections appear to be very optimistic. However, careful preparation, with active Government involvement at every step, offers a chance for success. In addition, adoption of a program that sends a strong, unambiguous message to the private sector about the Government's intentions will be needed to unleash private investment.' Lessons of Experience 3.4 The trade policy reforms of ITPA C and the budgetary process reforms of PSA C succeeded because of intensive work by Bank staff over several years, during preparation and supervision, in conjunction with Government teams, by the end of which the Government had internalized the reforms. Consistency of Bank staff and their close involvement in these two components were important elements of success. The absence of this process in several other components contributed to their failure. While the Letter of Development Policy referred to the need to rationalize the public enterprise sector, the Government apparently viewed this primarily as an agreement to divest itself of nonperforming public enterprises; it was never convinced of the need to privatize the seemingly successful enterprises. The Bank should have paid closer attention to the warning signals that emerged at several points during preparation and implementation of the two projects. Likewise, the civil service reform never got off the ground, and in fact despite agreement on an action plan, a new civil service statute adopted in 1993 included a number of counterproductive measures. It is not at all clear that the current Government embraces widespread understanding and support for the idea of continued adjustment, and internalization of any reform efforts will be a critical factor in their success. 3.5 Perhaps reflecting its own uncertainty, the Government failed to explain the benefits of reform to the people, or to inform them of measures taken to alleviate the social costs of adjustment. Although projects aimed at reducing poverty were designed to accompany the adjustment program, there was no attempt by the Government to explain them to the public. Employees of public enterprises were not broadly informed of the assistance available to them under EMSAP. In any case, the social actions were all delayed substantially. Greater efforts are needed to ensure that the social components of an adjustment program, whether contained in a separate project or not, are carried out in a timely manner. In the present environment of hostility to further adjustment, it is more important than ever to develop a cohesive safety net and to publicize it well. 3.6 The approach taken by PSAC to public enterprise reform proved largely ineffective. Instead of developing an agreed framework for privatization, and applying it initially to a few key enterprises, the conditionality consisted of a long list of public enterprises to be restructured, divested or closed. The list, prepared by the Government and accepted by the Bank, contained many enterprises that were insignificant in terms of their cost to the economy or the public sector (many in fact had already ceased to operate), while it omitted some of the 'A private sector environment adjustment credit was brought to the point of negotiations, but abandoned when the Government collapsed in 1991. 33 biggest problems. Supervision of this component became a matter of counting enterprises rather than examining the impact of their closing or reform. The public enterprise component might have been more effective had it concentrated on defining an approach to reform and applying it to the enterprises that were the greatest burden to the economy.' 3.7 Well defined indicators are essential to the monitoring of conditionality. The conditionality for trade reform was based on explicit measures such as the removal of quantitative restrictions and specific reductions in maximum and average tariffs and in the dispersion around the average. Public enterprise conditionality, on the other hand, called for "satisfactory progress," based on the liquidation, divestiture or restructuring of certain numbers of enterprises. These terms were not well-defined, however, with the result that it was not always possible to distinguish among them (liquidation and sale of assets as opposed to divestiture, for example) or to determine whether they had actually taken place. 3.8 Privatization is a powerful tool for preventing the reversal of reforms to public enterprises. Madagascar's financial sector offers a good example: cleaning up the bank portfolios could not prevent the excesses from recurring, but had all the banks been privatized, they would have been far less likely to permit the credit expansion that set off the liquidity crisis in 1991. Privatization of the remaining two state-controlled banks is, in fact, an objective of the ongoing Financial Institutions Development TA project and will be a priority in future lending. 3.9 The initial privatizations took place with little openness, a situation which led to suspicion and a loss of Government credibility and contributed to the breakdown of the program. It is essential to carry out the divestiture process in a transparent manner so that everyone concerned believes they are being treated fairly and given an equal opportunity to compete. In particular, the privatization process raised fears of excessive foreign participation in the economy and of the potential for unbalancing ethnic relationships. Creation of a privatization fund to facilitate broader Malagasy participation in the privatization process could have alleviated these fears. 3.10 Discussions between the Government of Madagascar and the Bank are now focussed on the prerequisites for resuming the adjustment program. An initiating brief prepared in May, 1994 listed conditions for Board approval of a new Adjustment Credit. Progress has been uneven, but some measures, including the floating of the FMG, the removal of import restrictions and an increase in petroleum prices, have been achieved. Picking up the unfinished agenda from ITPAC and PSAC, the Credit would concentrate on private sector development and macro-economic and public sector management, with conditionality limited to a few key measures in each area. Learning from the past, public enterprise reform would be pursued initially through the introduction of competition in a few key sectors (vanilla, petroleum products and air transport), and broader reform would be sought through a process that would foster internalization by the Government. Financial sector reform would seek majority private sector participation in at least one of the two remaining state-controlled banks and private management of the other. To encourage the private sector, the Credit would "For a discussion of design issues for privatization programs, see World Bank, Operations Evaluation Department, "World Bank Assistance to Privatization in Developing Countries", Report No. 13273, August 19, 1994. 34 focus on measures to liberalize markets, remove barriers to entry, streamline the incentives system and simplify the legislative and regulatory framework. Given the mixed record of the past, the strategy calls for a multi-year reform package, supported by a series of one-tranche operations, with all conditionality to be accomplished before Board approval. At each stage, negotiations would specify monitorable actions to be completed before the next operation. A follow up to EMSAP is being prepared to promote complementary social actions. 35 MADAGASCAR KEY INDICATORS 3/3/94 15:07 Est. 1985 1986 1987 1988 1989 1990 1991 1992 1993 Real growth rate: Gross Domestic Product (GDP) 1.1 2.0 1.2 3.4 4.1 3.1 -6.8 1.0 2.1 Gross Domestic Income (GDY) 0.3 3.5 -0.3 2.6 3.1 0.2 -8.7 0.8 3.6 Real per capita growth rate: Gross Domestic Product -1.8 -1.1 -1.9 0.2 0.8 -0.1 -9.6 -1.8 -0.6 Total consumption 1.7 -7.6 -3.6 -4.4 -1.7 1.3 -4.1 -5.6 1.0 Private consumption 1.6 -8.0 -4.3 -4.1 -2.1 2.1 -3.6 -5.9 2.1 Debt and Debt Service (LT+IMF+ST): Total DOD (US$ mil) 2,472.2 2,925.6 3,945.2 3,996.0 3,423.3 3,651.5 3,715.3 3,818.4 3,979.5 DOD/GDP 83.0 102.2 177.5 177.5 131.0 116.3 138.8 130.7 119.6 Debt service (US$ mil) 1/ 155.8 296.0 313.6 330.6 206.1 464.6 169.9 111.0 86.9 Debt service/ExporLs 43.4 72.6 71.7 79.7 42.9 88.1 34.7 22.0 17.7 Debt service/GDP 5.5 9.1 12.2 13.5 8.3 15.1 6.4 3.7 2.6 Interest burden (LT+IMF+ST): Interest paid (US$ mil) 73.1 78.6 110.8 76.7 130.4 173.8 49.7 53.8 27.7 Interest/Exports 20.4 19.3 25.3 18.5 23.0 32.9 10.2 10.7 5.6 Interest/GDP 2.6 2.4 4.3 3.1 4.4 5.6 1.9 1.8 0.8 Gross Investment/GDP 8.5 9.0 10.1 13.3 13.4 17.0 8.2 11.3 12.3 ICOR (5 years ending year shown) 11.4 6.1 4.7 4.1 3.9 13.8 14.2 19.4 26.1 Domestic Savings/GDP 1.3 6.9 5.9 8.4 10.0 6.0 -0.8 2.8 3.6 BOP Resource Balance/GDP -7.7 -2.8 -5.2 -6.1 -4.0 -10.7 -7.3 -7.8 -8.6 National Savings/GDP -1.9 3.5 1.2 3.7 5.3 5.6 -0.5 3.7 5.2 BOP Current Account Balance/GDP -10.8 -6.3 -10.0 -10.8 -8.7 -13.4 -9.9 -8.8 -8.8 Marginal Domestic Savings Rate -115.6 668.9 94.7 1387.4 57.7 991.8 98.8 860.1 25.2 Marginal National Savings Rate -33.6 633.8 207.3 187.1 53.7 -66.5 88.5 995.5 45.2 Government Investment/GDP 6.0 5.4 7.0 6.9 9.7 7.9 5.9 7.6 8.1 Government Savings/GDP -0.4 -0.9 -0.2 -1.4 -3.1 0.0 -4.2 -4.3 -2.6 Private Investment/GDP 2.6 3.6 3.1 6.4 3.7 6.9 4.6 3.7 4.3 Private Savings/GDP -1.5 4.4 1.4 5.1 8.4 5.6 3.7 8.0 7.8 Government revenues/GDP 13.3 12.7 15.3 13.8 12.7 13.3 9.3 10.9 10.3 Government expenditure/GDP 20.8 20.0 23.9 24.1 25.5 21.2 20.1 23.5 21.6 Overall Budget deficit/GDP 2/ -7.5 -7.3 -8.6 -10.3 -12.6 -7.8 -10.6 -12.4 -11.0 Primary deficit/GDP -3.6 -2.9 -2.7 -3.2 -5.1 -2.1 -4.9 -5.6 -6.4 Consumer Prices, growth rate 31 9.6 16.2 27.2 24.4 12.3 15.4 14.9 13.2 9.5 GDP Deflator growth rate 10.5 14.2 23.0 21.2 12.0 11.4 14.4 12.6 12.6 Real Exchange Rate Index (1987=100) 148.9 141.2 100.0 85.8 84.9 96.1 87.5 92.8 100.7 Terms of Trade Index (1987= 100) 112.1 132.6 100.0 95.6 81.2 74.5 73.7 69.2 66.8 Exports GNFS, volume growth rate 4/ -3.2 0.0 2.9 -8.2 16.9 12.0 4.8 2.0 -6.1 Exports GNFS/GDP 11.6 12.0 16.5 16.3 18.0 15.9 17.3 16.7 15.5 Imports GNFS, volume growth rate 4/ 9.5 -20.9 0.3 -9.6 0.9 33.0 -14.9 0.6 3.0 Impors GNFS/GDP 18.9 14.1 20.7 21.2 21.4 26.9 26.2 25.2 24.2 BOP Current Account Balance (US$ mil) 5/ -310.0 -204.6 -256.6 -264.2 -216.3 -411.7 -264.6 -261.8 -293.9 Net reserves (US$ mil) -135.8 -88.7 -53.2 33.4 80.0 -55.5 -41.6 -18.3 3.1 Gross reserves (in months of imports) 0.8 2.1 3.0 3.6 3.9 1.0 1.2 1.2 1.3 1/ Actual service paid after rescheduling. 2/ Including interest on rescheduling agreements. 3/ Private consumption deflator (National Accounts). 4/ National accounts. 5/ Excluding official grants.  37 Annex 2 Telefax Federal Office for Foreign Ecommic Affairs Bundeshaus Ost 3003 Berne crne, 15th June. 1995 M 0041-31-324 08 11 (M& Fm 0041-31-324 09 62 QW Mr. M Penalver Division Chief Our ref., 3/85-MG - niVegf Country Policy, Industry and Finance Division, OED The World Bank Number of pages: 2 (including cover sheet) Washingtop D.C. Fax no.: 001-202-522 31 24 Madagascr / Public Sector Adjusmnent Credit (PSAC) : Project Completion Report and Performance Audit Report Dear Mr. Penalver, I acknowledge with thanks receipt of the two abovc-mentioned documents. The prepara- tion of the PAR by the Operations Evaluation Department is a very crucial exercise to ensure that the past experiences are properly rcviewed and analyzed in order for them to be capitalized upon in the preparation and implementation of new operations. The Federal Office for Foreign Economic Affairs (FOFEA) therefore very much welcomes the submission of these two reports on the PSAC for Madagascar. FOFEA is fully aware of the political events which happened at the time of implementa- tion of the PSAC and which definitely affected itN performance. However, as mentioned in the PAR, these external factors do not suffice to fully explain the rather dismal overall performance of the project (rated in PAR as unsatisfactory with uncertain sustainability and negligible institutional development). I basically agree with the analysis, findings and recommendations presented in the PAR. I also fully support the main lessons identified. However, I believe that more emphasis should be given to the critical review of the inadequate risk analysis, not only at the inception of the project, but also at later stages of implementation : the PAR is not providing many elements to better understand why signals of caution (given by staff members of the Bank at the preliminary stage for example) were not properly taken into account, and what corrective measures - in the Annex 2 38 decision maldng process to initiate or continue a project - are currently, or should be, considered by the Bank. Moreover, I would like to request that two additional points are also considered and analyzed in the comext of the Performance Audit of the project : I would appreciate to see in the PAR a qualitative and critical analysis, in terms of performance and accountability, of the actual disbursement mechanism used for PSAC during the OGL period and after (OGL was suspended by nid-1991). Moreover the question of the tinacceptably long delay for the submission of audit reports by the Government should be addressed in more details, particularly taking into considefation that about 25 % of the first tranche was allocated to one importer only, by bypassing the standard disbursement procedures and that such a substantial irregularity was discovered by the Bank only in 1992 : what lessons were learned by the Bank in the fields of procurement and auditing, which are of high concern to the cofinanciers of the Bank's projects ? 2 there seems to have been a major underestimation of the Government's administrative and institutional capacity to Implement the reform prograinme. In this regards, I would appreciate to have some more analysis on the lessons learned from that experience, and the potential for better coordination with bilateral donors (particularly the cofinanciers of such projects) for them to provide additional grant ressources for the very purpose of financing the strenghtening/developing the Government's capacity to better manage adjustmcnt projects. 3 Finally, two small remark on the Basic Data Sheet : (i) it could be interesting to know why the credit position (p.7 of PAR) of ADF and Norwegian grant are filled with "n.a." (ii) the table on project dates (p.8 of PAR) could include the Information regarding the signature and effectiveness of the cofinanciers' loan/credit. FOFEA is following with particular interest and attention the development of the general and political situation in Madagascar. Switzerland Is currently contributing to PRODIF, the Central Bank rehabilitation project. Additional economic cooperation will be possible only when a sound macroeconomic policy framework is designed and agreed upon between the Government and the International Financial Institutions. At this stage, we shall definitely, also for bilateral economic support measures, capitalize on the experiences and lessons learned during the P'SAC. I shall therefore be very grateful if you could kindly send a copy of the finalized PAR, as soon as available. Thanking you for your kind cooperation, I remain, With best regards. elopment Policy Division Michel Mordasini Chicf, Balance of Payments Operations and Debt Reduction Fede-l Ofie for Foreign Economic Affairs 39 Annex 2 Endnotes 1. This is an important point. The PAR stresses the fact that warning signs were not taken seriously enough (paras. 14, 2.9 and 2.13). Staff in charge of PSAC apparently judged the problems to be within an acceptable range of risk. 2. The Bank granted extensions requested by the Government for submission of the 1990 audit, and therefore discovered only belatedly the irregularity in the use of funds. Once it was discovered, the Government replaced the missing funds and arrested several people alleged to be responsible. More stringent application of procurement and auditing conditionality would have uncovered the abuse much sooner. 3. Adequate resources were available through EMSAP for strengthening the Government; the problem was in utilizing those funds (para. 2.30). EMSAP did finally get underway, but only several years after the start of PSAC. 4. The Bank administered the Swiss grant, but not the others. It therefore has information only on the former.   IMAGING Report No: 14774 Type: PCR