Mozambique - Second Rehabilitation Credit Project

Document of The World Bank FOR OFFICIAL USE ONLY Repwt No. 11767 PROGRAM COMPLETION REPORT MOZANBIQUE SECOND REHABILITATION CREDIT (CREDIT 1841-MOZ) APRIL 2, 1993 Country Operations Division Southern Africa Department Africa Regional Office 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 EOUIVALENTS Currency Unit = Metical (pl. Meticais) Appraisal (March 1987): US$1 = 200 Meticais Completion (June 1991): US$1 = 2,746 Meticais ABBREVIATIONS AND ACRONYMS AGRICOM - Empresa de Comercialigao Agricola (State Enterprise for Agricultural Marketing) BDM - Banco de Mogambique (Bank of Mozambique) DANIDA - Danish International Development Agency CG - Consultative Group CNP - National Planning Commission ERP - Economic Rehabilitation Program GCPI - Gabinete de Coordenacao de Programas de Importagao (Bureau for Coordination of Import Programs) OECD - Organization for Economic Coordination and Development OPEC - Organization of Petroleum Exporting Countries PFP - Policy Framework Paper SAF - Structural Adjustment Facility (IMF Program) SFA - Special Facility for Africa SJF - Special Joint Financing SRC - Second Rehabilitation Credit UCPI - Unidade de Coordena,ao de Programas de Importaqao (Unit to Coordinate the Imports Program) UNDP - United Nations Development Program FISCAL YEAR OF THE BORROWER January 1 - December 31  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operaftons Evaluation April 2, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Mozambique Second Rehabilitation Credit (Credit 1841-MOZ) Attached is a copy of the report entitled "Project Completion Report on Mozambique - Second Rehabilitation Credit (Credit 1841-MOZ)" prepared by the Africa Regional Office. The original credit objectives of supporting policy and institutional reforms for restructuring the Mozambique economy and providing foreign exchange to finance essential imports needed for the rehabilitation of the economy were on the whole achieved, with the economy realizing an average annual GDP growth rate of 5.4 percent in 1987-89 after many years of sharp decline. The restructuring was part of a process begun in connection with an earlier IDA credit and has been continued subsequently, indicating that the changes made are being sustained. Institutional constraints slowed the procurement process, causing a delay in the closing date of the credit. However, the program helped generate additional balance of payments support from other donors for Mozambique that compensated for the delays in IDA disbursements. The PCR is of good quality and provides a candid assessment of the achievements and shortcomings of both the design and the implementation of the credit. It lacks, however, an assessment of the war impact on program results, a discussion of the social impact of the program, and a critical assessment of the adequacy of IDA's supervision. The Government of Mozambique prepared comments, albeit perfunctory, which are contained in Part II, generally concurring in the sections of the report prepared by IDA. The Operations Evaluation Department intends to carry out an audit of this credit. The audit will explore more thoroughly the validity of the IDA strategy underlying the program's design, the relationship between IMF conditionality and that of the IDA, the prospects for sustained progress in Mozambique following this operation, the social impact of the program, and lessons learned that may have applicability elsewhere. Attachment This document has a restricted distnbution and may be used by recipients only in the performance of their oflicial duties. Its contents may not otherwise be disclosed without World Bank authorization.   FOR OFFICIAL USE ONLY CONTENTS Preface ........................................................................................................................ Evaluation Summary .....................................................i PART I - REVIEW FROM BANK'S PERSPECTIVE ...........................................I A. The Economic Rehabilitation Program .................................................... 1 Background ....................................................1 Macroeconomic and Sectoral Objectives of the ERP .............................2 B. Role of IDA ....................................................3 Justification of IDA Support ....................................................3 Program Design ....................................................3 Complementarity between the SRC and the SAF Program .................... 4 Complementarity with Project Lending .................................................4 C. Implementation of Adjustment Program ....................................................4 Implementation and Results of the Adjustment Program ............ ...........4 Mobilization of External Aid ....................................................7 D. Lessons Learned for Subsequent Adjustment Programs ..............................8 Assessment of Risks ....................................................8 Sustainability ....................................................9 Lessons learned ....................................................9 E. Borroweres Performance ................................................... 10 Compliance with Credit Agreement ................................................... 10 Disbursements ................................................... 10 Audits ................................................... 11 Procurement ....................................................1.1 F. Bank Supervision ................................................... 12 PART II - REVIEW FROM BORROWER'S PERSPECTIVE ............................. 13 PART IH - STATISTICAL INFORMATION ................................................... 15 ANNEX 1: COMPLIANCE W1TH COVENANTS ............................................... 17 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.   PROGRAM COMPLETION REPORT MOZAMBIQUE SECOND REHABILITATION CREDIT (Credit 1841-MOZ) PREFACE This is the Program Completion Report (PCR) for the Second Rehabilitation Credit in Mozambique, for which Credit 1841-MOZ in the original amount of SDR 69.0 million (IDA - SDR 54.5 million, Special Facility for Africa (SFA) - SDR 14.5 million) was approved on August 4, 1987. The Credit was closed on June 30, 1991, two and one half years behind schedule. The Credit's last disbursements were on May 14, 1991 (SFA) and June 13, 1991 (IDA). Undisbursed amounts of SDR 81,860 (IDA) and SDR 117,758 (SFA) were canceled by IDA on December 4, 1991. The Credit was cofinanced by the Swiss Joint Financing (16.9 million Swiss Francs) and Swiss Bilateral Financing (10.1 million Swiss Francs), the Government of Norway (30.0 million Norwegian Kroners), the Danish International Development Agency (DANIDA - US$2.8 million), and the Government of the Netherlands (7.3 million Dutch Guilders). The PCR was jointly prepared by the Borrower (Part II) and the Country Operations Division of the Southern Africa Department (Parts I and III). Parts I and III were sent to the Government and cofinanciers with a request for their comments. The Government's comments on the report are included in Part II. The PCR is based on the President's Report, the Development Credit Agreement, the Letter of Development Policy, supervision reports, correspondence between the Bank and the Borrower, internal Bank memoranda, and discussions with the relevant Government officials and Bank staff. The Program Completion Report (Parts I and III) was prepared by Ms. Carolina Machado and Mr. Rene Bonnel (AF6CO). The Task Manager at the time of appraisal and throughout implementation was Mr. Nils Tcheyan (now the World Bank's Representative in Mozambique). Mr. Praful Patel is the managing Division Chief and Mr. Stephen Denning is the Department Director.   PROGRAM COMPLETION REPORT MOZAMBIQUE SECOND REHABILITATION CREDIT (CREDIT 1841-MOZ) EVALUATION SUMMARY Objectives 1. The Second Rehabilitation Credit (SRC) was prepared to support the implementation of the Economic Rehabilitation Program (ERP), which was developed by Government in collaboration with the World Bank and the IMF in 1986-87. The key objectives of the ERP were to: (i) reverse the previous decline in production and restore a minimal level of consumption for the population, especially in the rural areas; (ii) curtail domestic financial imbalances and strengthen the country's external payments position; and (iii) establish the conditions for more rapid growth in the medium- to long-term when the security situation and other exogenous constraints would have eased. 2. The SRC formed part of a broader effort by the international community to assist Mozambique's efforts to reform economic management and establish a basis for sustainable growth. In 1987, an emergency donor meeting was organized; the JIF approved a SAF for Mozanbique totaling SDR 28.67 million; agreement on debt rescheduling was reached with the Paris and the London Clubs; and a Consultative Group meeting was held in July 1987 to review progress and firm up donors' support. 3. Against the above background, the main objectives of the SRC were to: (a) support specific policy and institutional reforms concerning the extemal sector (trade regime and foreign exchange allocation), pricing and distribution policies, fiscal policy, agricultural marketing, industrial pricing and efficiency, and transport sector efficiency; (b) provide foreign exchange to finance essential imports needed for the rehabilitation of the economy; and (c) assist the Government to develop an agenda for longer-term policy changes. Implementation Experience 4. The program was implemented largely as planned. The first IDA supervision mission in November/December 1987 reported on the liberal policy measures that were adopted by the Government. These and Government's compliance with covenants justified the release of the Credit's second tranche in February 1988.  5. The key policy objectives of the SRC were implemented as follows: (i) The foreign exchange allocation system was improved. The effective exchange rate was devalued by 77 percent in real terms in 1987; the management of external aid was strengthened with the establishment of the Bureau for the Coordination of Import Programs (GCPI) and the Commission for Extemal Economic Relations (CREE) in 1987 and 1988, respectively; and marketing was liberalized. (ii) Market-related price signals were enhanced through sharp increases in the prices of staples, ful pass-through of exchange rate adjustments and reduction of the number of products subject to centrally fixed prices. Increases of prices regulated through mark-ups were permitted without prior approval. (iii) A longer term policy agenda was developed and incorporated into Policy Framework Papers prepared annually with Government and the vIF. A key result was the initiation of a process to rationalize public expenditures. Results 6. Overall, the SRC achieved most of its key objectives. The main achievement of the SRC was to restore growth and minimum level of consumption despite the adverse effects of the civil war. The 1982-85 decline in GDP was reversed with the rate of economic growth averaging 5.4 percent p.a. in 1987-89, a rate significantly greater than the 2.7 percent initially projected. Much of this growth came from light industrial activity which rose because of improved availability of foreign exchange for raw material imports. Recovery was also apparent in the agricultural sector where production increased on average by about 5 percent per annum. 7. The macroeconomic framework underlying the SRC embodied a substantial degree of disequilibrium, but this imbalance was viewed as necessary for restoring growth and macroeconomic balances over the medium- to long-term. As a result, major improvements in the fiscal and external accounts were not expected. Nevertheless, substantial progress was achieved in the fiscal area. Government negative saving was drastically reduced; the budgetary deficit after grants was reduced from 15.3 percent of GDP in 1986 to 8.9 percent in 1989; and domestic bank financing of the fiscal deficit was drastically reduced from 12 percent of GDP in 1986 to -0.8 percent of GDP in 1989, which allowed repayments by the budget to the banking sector. In conjunction with credit restraint, inflationary pressures were reduced as the rate of inflation fell from 163 percent in 1987 to 42 percent in 1989.  8. Because of the priority attached to restoring economic growth and consumption levels, the current account deficit was expected to increase as a percentage of GDP, but the worsening turned out to be larger than foreseen. Mozambique's dependency on external aid rose substantially (the current account deficit before grants averaged 55.9 percent of GDP in 1988-89 compared to a projected average of 45.4 percent of GDP) as the responsiveness of exports to the modest growth of the economy was disappointing and imports rose more rapidly than projected. 9. In line with initial expectations, the SRC (in conjunction with the IMF SAF) played a major catalytic role in the mobilization of external aid. Disbursement of loans and grants rose from US$497 million in 1986 to US$644 million in 1989. The SRC also provided a vehicle for attracting general balance of payment support, which reached about US$200 million in 1989, most of it untied by sources of supply. Findings and Lessons Learned 10. Sustainability. The policy reforms adopted by the Government under the Economic Rehabilitation Program (ERP) have been sustained over the past five years. Particularly impressive has been the Government's continued determination to deepen the adjustment process initiated in 1987 and transform the system of economic management. On the basis of these policies, the Government successfully approached IDA and the IMF for continued balance of payments support. The SRC was followed by the third IDA Rehabilitation Credit (Credit 2021-MOZ), approved on May 18, 1989, and the Economic Rehabilitation Credit (Credit 2384-MOZ) approved by the Board in June 1992. Implementation of the IMF three-year SAF program also remained on-track. A three-year ESAF program was negotiated with the IMF in June 1990, and the third-year ESAF program was approved by the IMF Board in December 1992. 11. The focus of the SRC on reducing key macroeconomic distortions and restoring financial incentives at the producer level had a high payoff as economic growth accelerated in 1987-89. By 1990, however, it became clear that the initial objectives of the ERP of restoring macroeconomic balances and sustained economic growth would be a long-term process requiring: (i) continued policy reforms to reduce administrative management of the economy and increase the role of market forces in the allocation of resources; (ii) rehabilitation of the economic and social infrastructure to address long-term constraints to growth; and (iii) capacity building in parallel with substantial technical assistance to strengthen Government's capacity to plan for and manage economic development. 12. Lessons. The SRC thus brought to the fore the importance of deepening the adjustment program in the following four areas:  iv (i) Exchange rate and trade policy. Adjustment of the nominal exchange rate was viewed as a key policy instrument for eliminating distortions in the foreign exchange market and restoring price incentives. However, the gap between the parallel market exchange rate and the official exchange rate remained substantial and the successive devaluations led to high inflation. The lesson learned was that exchange rate adjustment had to be implemented in parallel with trade reform so as to bring parallel market transactions within the official market for foreign exchange. (ii) Licensing and trade restrictions. The SRC was focused on eliminating price distortions, but marketing restrictions and licensing regulations proved to be equally important in limiting the adjustment of the economy to the new structure of incentives. The SRC thus brought out the importance of removing licensing restrictions and obstacles to domestic trade, particularly in the case of agriculture. (iii) Enterprise restructuring. The initial approach focused on providing technical assistance to restructure state enterprises largely failed to revitalize the industrial sector. State enterprises remained dependent on budgetary subsidies, preferential access to credit and implicit subsidies provided through an overvalued exchange rate. The lesson learned was that a more radical approach involving privatization, management of state enterprises on commercial lines and increased competition was needed if state enterprises were to contribute to economic growth. (iv) Capacity building. In retrospect, there was a strong contrast between Government's sustained commitment to the reform program and its limited implementation capacity. In areas where implementation of the reform program involved modifying policy instruments such as prices, progress was rapid. In other areas where institutional reform was required, progress was uneven and slow. It was substantial in the fiscal area, but slow in the banking and enterprise sectors-where the institutional reforms proved to be more complex than foreseen. The key implication has been the need to focus on implementation and institutional issues more forcefully. As these are best addressed through investment or free standing technical assistance projects, the main lesson was the importance of integrating fully project lending in the design and implementation of structural adjustment.  PROGRAM COMPLETION REPORT MOZAMBIQUE SECOND REHABILITATION CREDIT (Credit 1841-MOZ) PART I - REVIEW FROM BANK'S PERSPECTIVE A. The Economic Rehabilitation Program Background 1. Mozambique is one of the poorest countries of the world with an annual per capita income of around US$70. Literacy rates, infant mortality rates, levels of acute malnutrition and life expectancy are among the worst in Sub-Saharan Africa. Many of Mozambique's problems can be attributed to structural weaknesses inherited from the colonial era and to instability within the Southern Africa Region. Years of colonial rule left Mozambique with an extremely poorly developed physical and social infrastructure and an acute shortage of trained manpower at all levels. Since 1980, the civil war has led to severe disruptions in production, an almost complete breakdown in the rural marketing system and widespread destruction of basic infrastructure. Over a third of health facilities and about one-half of all primary schools have been destroyed, and nearly one-third of the population has been displaced by the insecurity. 2. Inappropriate economic policies also contributed to Mozambique's poor performance. In the immediate years after Independence, economic management became increasingly centralized. Abandoned enterprises in agriculture, industry and services were taken over by Government to ensure the basic functioning of the economy. Production targets were established for many enterprises and agricultural inputs and outputs were increasingly subjected to administrative allocation. State farms were established and parastatals began to dominate agricultural marketing. Rigidities in the pricing and marketing system, combined with a concentration of resources in the state farm sector, severely constrained agricultural, especially smaDl holder production. 3. The combination of increased insecurity and inappropriate domestic policies was disastrous for the Mozambican economy. By 1986, real GDP had fallen to around two-thirds of the 1980 level. Paralel markets emerged and widened, and goods and foreign exchange were, respectively, selling for up to 10 and 40 times their official equivalents. In the external accounts, the value of exports and imports shrunk by nearly two-thirds and one-third, respectively. Moreover, capital inflows declined  2 while debt service payments escalated resulting in a substantial accumulation of arrears. 4. In 1984, the Govemment undertook a reform program-the Economic Action Program-to address these problems, and it requested IDA assistance in support of its program. The first IDA Rehabilitation Credit was approved in June 1985 with the objective of providing financial assistance to rehabilitate and maintain those investments and economic services which were expected to generate some immediate production response. 5. The remedial measures taken in 1984-86 helped restore modest economic growth (2.3 percent) in 1986, but they did not address critical macroeconomic issues related to exchange rate, prices and credit policies. As such they were insufficient to address Mozambique's substantial internal and extemal imbalances. Accordingly, in January 1987, the Government introduced a far more comprehensive program-the Economic Rehabilitation Program (ERP)-intended to strengthen growth prospects and reduce financial imbalances in the economy. Macroeconomic and Sectoral Objectives of the ERP 6. The general objectives of the ERP were to: (i) reverse the decline in production and restore a minimal level of consumption and income for the population, especially in rural areas; (ii) curtail domestic financial imbalances and strengthen the country's external payments position; and (iii) establish the conditions for more rapid and more efficient economic growth in the medium- to long-term, when the security situation and other exogenous constraints would have improved. 7. The ERP was designed with the objective of raising GDP growth to about four percent p.a. by 1989. Recovery was expected to be centered on the agricultural sector, which would provide the needed increase in rural incomes and the foundation for a restoration of previous export levels. To achieve this goal, policy and institutional issues were to be addressed in key agricultural, industrial, transport, and energy sectors, with the objective of improving the quality of planning and policy making, and ensuring that public expenditures were focused on the economy's priority requirements. Centralized administrative controls were to be reduced and market forces were to play a greater role in the allocation of resources. Accordingly, key macroeconomic policies, such as exchange rate and external trade policies, budgetary, pricing, marketing and credit policies, were to be modified over the medium-term within the framework of annual Policy Framework Papers in conjunction with substantial extemal assistance.  3 B. Role of IDA Justification of IDA Support 8. IDA assistance was timely. The Government had been reassessing its macroeconomic policy to reverse the economic decline of the early 1980s. But the measures taken in 1984-86 under the Economic Action Program had proven insufficient for reversing the severe economic decline. Accordingly, the Government decided to deepen the reform process. At the Government's request, a Bank mission went to Mozambique in November/December of 1986 to assist in the preparation of an Economic Rehabilitation Program (ERP) for the period 1987-89. Bank staff worked in close cooperation with the Bank of Mozambique (BDM) and the Minister of Finance who was taking a lead in the preparation of the ERP. The assistance provided in the elaboration of the ERP led to the formulation of an adjustment program which was supported by the SRC. Program Design 9. The major objectives of the SRC were to: (i) support specific policy and institutional reforms in the following areas: (a) the external sector, including the trade regime and foreign exchange allocation; (b) pricing and distribution policies; (c) fiscal policies; (d) agricultural marketing and producer prices; (e) industrial pricing and efficiency; and (f) transport sector efficiency; (ii) provide foreign exchange to finance essential imports needed for the rehabilitation of the economy; and, (iii) assist the Government to develop an agenda for longer-term policy changes. 10. The SRC was approved by the Board of Executive Directors in August 1987. It included financing from IDA of SDR 54.5 million, from the IDA-administered Special Facility for Africa (SFA) of SDR 14.5 million, and grants from the Swiss Special Joint Financing (SJF) of 16.9 million Swiss Francs, and from the Swiss Bilateral Financing of 10.1 million Swiss Francs, all signed in September-October 1987. In addition, co-financing was provided by the Government of Norway (30 million Norwegian Kroners), signed in March 1988, and by the Danish International Development Agency (DANIDA) (US$2.8 million) and the Ministry of Development Cooperation of the Netherlands (7.3 million Dutch Guilders), both signed in March 1989.  4 Complementarity between the SRC and the SAF Program 11. Mozambique became a member of the IMF in September 1984. Technical assistance on monetary policy, central banking and fiscal affairs was provided in 1985-87, but Mozambique did not made use of the Fund's ordinary resources until mid-1987 when the three-year Structural Adjustment Facility (SAF) was approved. As both the SRC and the SAF program were designed to support the implementation of the ERP, there was a strong complementarity between the two programs. SAF policy measures were particularly focused on: * Exchange rate policy: the exchange rate was to be substantially devalued in real terms and further adjusted as needed. * Budgetary expenditures: enterprise subsidies were to be substantially reduced and current expenditures were to be contained. - Revenue policy: fiscal reform was to be implemented to raise fiscal revenues. * Credit policy: ceilings on credit expansion were to be set and interest rates adjusted. Technical assistance was to be provided to help revise the banking system's accounts and lay the basis for the separation of the Bank of Mozambique's commercial and central banking functions. Complementarity with Project Lending 12. Successful implementation of some of the key policy reforms of the SRC depended on institutional strengthening and enterprise restructuring. These reforms were supported through IDA project lending. The Economic and Financial Management Technical Assistance Credit (1989) was to provide the institutional strengthening required under the SRC, while the Small and Medium Enterprise Development Credit (1989) and the Industrial Enterprise Restructuring Credit (1989) would provide financing for the rehabilitation of enterprises and help support strengthening of the financial sector. C. Implementation of Adjustment Program Implementation and Results of the Adjustment Program 13. Overall, most of the objectives of the SRC were achieved. The key achievement of the SRC was to restore economic growth and minimum level of consumption despite the continuing difficulties created by the civil war. Relative to the earlier 1982-85 decline of GDP, the establishment of economic growth averaging 5.4 percent in 1987-89 was a substantial achievement.  5 14. The restoration of an appropriate incentive structure was a key focus of the SRC program, given the emphasis on the agricultural sector and small holders. To restore financial incentives, fixed prices of commodities and services were raised manifold;' the number of products subject to such pricing was reduced by 20 percent in 1987; increases of prices regulated through mark-ups were permitted without prior approval; and most utility charges were raised substantially. In the agricultural sector, as part of a pricing and marketing reform aimed at stimulating production and improving incomes of rural producers, the Government removed controls on the prices of fruits and vegetables and some minor animal products, and it decided that the prices of some products would be fixed at the provincial rather than at the national level so as to allow for greater flexibility in price setting. In the industrial sector, a study of 25 industrial enterprises and 15 agricultural enterprises was completed. An action plan was defined to improve the performance of the sector. Enterprises were given greater flexibility in setting their own prices and Government subsidies to enterprises were drastically reduced. 15. Measures were also taken to improve the foreign exchange allocation system. The effective exchange rate was devalued in January 1987 by 77.3 percent in real terms, with the result that the ratio of the parallel exchange rate to the official exchange rate fell from 40 to 2.2 by end-1987. However, the crawling peg policy implemented in 1988-90 failed to achieve much additional results as the effective exchange rate did not depreciate further in real terms and exchange rate distortions remained large. 16. With a view to reduce the scope of administrative allocations in the exchange and trade systems, the Government decreased the number of products that were traded through one trading corporation only to five major import groups (petroleum, pharmaceuticals, agrochemicals, cereals, electricity) and one major export commodity (sugar). In addition, importation of goods up to US$500 was allowed automatically when importers have their own source of foreign exchange. To improve the matching of donor import support funds with the needs of the economy and establish priorities for the use of foreign exchange, the Bureau for the Coordination of Import Programs (GCPI) and the Commission for External Economic Relations were established in 1987 and 1988, respectively. The export retention scheme, introduced in 1984, was expanded to include, by end-1986, around 100 large producers who could earn foreign exchange. In January 1987, it was further broadened to include all exporters of non-traditional products. 17. The adjustment framework underlying the SRC embodied a substantial degree of macroeconomic disequilibrium, but this imbalance was viewed as necessary to the 1 Petroleum product prices were increased sixfold and electricity tariffs increased threefold; road, sea, and air transport tariffs were increased by between 200 and 400 percent.  6 restoration of growth and macroeconomic balance in the medium to long term. As a result, major improvements in the magnitude of the macroeconomic imbalances were not expected. Nevertheless, the reduction of external imbalances proved more difficult than expected. External deficits worsened more than expected (the current account deficit before grants of the balance of payments averaged 55.9 percent of GDP in 1988-89 compared to a projected average of 45.4 percent of GDP). As the 1987 debt rescheduling agreement failed to provide the needed debt relief (the debt service ratio after debt rescheduling averaged 30.7 percent in 1988-89), arrears were incurred. 18. On the other hand, macroeconomic adjustment was substantial in the fiscal area. The revenue base was strengthened through tax reform and recurrent expenditures were contained through cuts in enterprise subsidies and reduction of wages in real terms. As a result, fiscal revenues rose from 13.3 percent of GDP in 1986 to 23.4 percent of GDP in 1989 and the current deficit shrank from 12.1 percent of GDP to 2.1 percent of GDP during the same period. To stimulate economic growth, budgetary investment was increased sharply from 5.6 percent of GDP in 1986 to 22.2 percent of GDP in 1989 (Table 1). A review of public expenditures was initiated in mid-July 1988 and completed in 1989, which led to the preparation of a three-year investment plan and the prioritization of public expenditures. 19. Progress was also achieved in reducing excess liquidity and inflationary pressures as the rate of monetary expansion was kept significantly below that of nominal GDP, while interest rates were raised from a maximum of 6 percent for deposits and 10 percent for loans to 20 percent and 35 percent, respectively. However, the separation of the central and commercial banking functions of the Bank of Mozambique took longer than expected with the separation of the Bank of Mozambique occurring in early 1992 only. Separation and reconciliation of the accounts proved particularly complex and time consuming.  7 TABLE 1: KEY MACROECONOMIC INDICATORS, 1986-91 1/ 1986 1987 1988 1989 1990 1991 Real Growth (% pa.) GDP Growth Rate 2.3 5.3 5.5 5.4 1.3 2.7 Consunption per Capita -3.3 -0.4 2.0 3.7 -4.1 -2.0 Gross InvestrnenttGDP 2/ 9.2 24.0 33.4 35.5 38.1 42.1 Domestic Savings/GDP -1.1 -12.0 -16.4 -16.6 -11.7 -9.9 Budgetary Data (% of GDP) Budgetary Revenues 12.9 16.2 19.9 23.5 22.2 23.5 Government Expenditures 30.9 37.5 45.2 49.0 51.7 50.3 Overall Deficit before Grants -17.6 -21.1 -25.3 -25.5 -29.5 -26.8 Overall Deficit after Grants -15.3 -11.8 -11.3 -8.9 -12.6 -6.0 Shares of GDP (%) Exports (G&NFSYGDP 3.5 12.1 15.1 15.4 15.9 23.3 Export Growth Rate 3/ 4.0 18.8 6.8 6.5 14.6 34.9 Imports (G&NFS)IGDP 14.3 48.2 65.0 67.5 65.7 75.4 Current Account Deficit Before Grants -15.1 -47.8 -52.9 -58.8 -54.7 -58.9 After Grants -9.9 -26.8 -22.6 -28.9 -23.6 -21.2 Debt Service Ratio Before debt relief 247.5 227.5 190.6 178.4 169.4 146.4 After Debt Relief 247.5 20.9 37.2 24.2 28.4 28.5 Real Effective Exchange Rate (depreciation-) 22.2 -77.3 1.8 0.3 -0.5 -16.7 Inflation(%p.a.) 4/ 38.7 163.3 50.1 42.1 49.2 34.3 I/ Due to inadequacies in national accounts, estimates on investment and savings are distorted. The increase in ratios to GDP reflects the effects of major exchange rate adjustment on the external conponents of investment relative to a less faster increase in donestic inflatini, and therefore, GDP. 2/ Investment aependitures include about 20 percent of recurrent expenditures. 3/ Goods and non-factor sevices in cun-ent dollars. 4/ Annual average. Mobilization of External Aid 20. In conjunction with annual Consultative Group meetings, the SRC has played a catalytic role in the mobilization of extemal financing. Substantial amounts of external assistance have been available by bilateral donors, both as cofinancing of IDA credits and as parallel operations. In aggregate, gross inflows of loans and grants increased from US$497 million in 1986 to US$644 million in 1989. Despite the cessation of aid from the former Soviet Union and other Eastern European countries  8 (amounting to about US$100 million), disbursement of aid increased further to US$718 million in 1992. 21. In addition to project and balance of payments support, Mozambique received substantial debt relief. Following the 1987 debt rescheduling with the Paris Club and the London Club, Mozambique concluded a new rescheduling on Toronto terms in June 1990 covering the period 1990-92. A number of donors have converted their outstanding loans into grants or canceled their loans, thus bringing closer the prospect for balance of payments viability. About 64 percent of Mozambique's commercial debt was retired at a discount in December 1991, utilizing funds from the IDA Special Debt Reduction Facility and donor financial support. D. Lessons Learned for Subsequent Adjustment Programs Assessment of Risks 22. At the time of appraisal (President's Report of July 16, 1987), the assessment was that there were two main risks to the successful implementation of the SRC and the ERP: (i) the limited implementation capacity of the Government; and (ii) the negative impact of the insecurity situation on production. As the reforms required a fundamental revision in the system of economic management affecting all levels of the productive structure, it was felt that the limited number of skilled civil servants could make it difficult to develop and implement a comprehensive reform program. In addition, the insecurity in the country created additional uncertainty concerning both Government's ability to implement the program as intended and the prospective responsiveness of the economy to the reform. 23. These risks were rightly assessed. Apart from the constraints brought about by the prevailing insecurity in Mozambique, the limited number of managers with the skills required by the new system of economic management strained the speed with which some of the program components were implemented. This affected particularly the Government's capacity to formulate macroeconomic policies, which improved less than expected. Macroeconomic management through monetary and exchange rate policies required Government to have timely and accurate information on prices, monetary and credit expansion. But data-gathering has remained limited and slow, thus inhibiting the design of appropriate macroeconomic policies. In addition, the disbursement of external assistance proceeded slower than expected, because of lack of familiarity on the part of borrowers and enterprises with the procurement rules of IDA and other donors.  9 Sustainability 24. The policy reforms initiated under the SRC have been sustained over the past five years. Particularly impressive has been the Government commitment to implement the various policy measures agreed with the Bank and the IMF, and more recently to accelerate the pace of adjustment. On the basis of these policies, the Government adjustment program was supported by the Third Rehabilitation Credit and the Economic Recovery Credit in 1989 and 1992, respectively. Implementation of the IMF SAF program also remained on track and a follow-up three-year ESAF arrangement was approved in 1990. 25. The focus of the SRC on reducing key macroeconomic distortions, restoring financial incentives at the producer level and mobilizing internal and external resources proved to be generally adequate in achieving the key objective of restoring economic growth: GDP growth accelerated from 2.3 percent in 1986 to 5.4 percent p.a. on average in 1987-89, which exceeded the projected rate of GDP growth of 2.7 percent p.a. Nevertheless, economic growth was fragile and macroeconomic imbalances remained substantial. The fragility of Mozambique's economic growth became most apparent in the early 1990s as the economy was subject to a series of large extemal shocks-including the cessation of external aid from the former Soviet Union, a severe drought in 1992 and continued insecurity in rural areas. On balance, nearly all the fall in the rate of GDP growth to 0.9 percent p.a. in 1990-92 can be accounted for by external shocks. Lessons learned 26. The experience gained with the SRC indicated that Mozambique's structural adjustment would be a long-term process requiring: (i) continued policy reforms to reduce administrative management of the economy and increase the role of market forces; (ii) rehabilitation of the economic and social infrastructure to address long- term constraints to growth; and (iii) capacity building in parallel with substantial technical assistance to strengthen Government's capacity to plan and manage economic development. 27. The SRC thus brought to the fore the importance of modifiying the design of the adjustment program in the following four areas: (i) Exchange rate and trade policy. The nominal exchange rate was continuously devalued in 1988-90, but the real effective exchange rate remained nearly constant. The lesson learned was that exchange rate adjustment had to be implemented in parallel with trade reform so as to bring parallel market transactions within the official market for foreign exchange.  10 (ii) Trade and licensing restrictions. Although the SRC was designed to address price distortions, marketing restrictions and licensing regulations proved to be equally important in limiting the adjustment of the economy to the new structure of incentives. The SRC thus brought out the importance of policy reforms at the district level (where restrictions are applied). (iii) Enterprise restructuring. The initial approach focused on providing technical assistance to restructure state enterprises largely failed to produce the expected outcome. Enterprises remained dependent on imports and implicit subsidies provided through an overvalued exchange rate. The lesson learned was that a more radical approach involving privatization, management of state enterprises on commercial lines and increased competition was needed. (iv) Capacity building. In retrospect, the SRC has highlighted the contrast between Government's strong and sustained commitment to the reform program and its limited implementation capacity. In areas where implementation of the reform program involved modifying policy instruments such as prices, progress was rapid. In other areas where institutional reform was required, progress was much more uneven. It was substantial in the fiscal area, but slow in the banking and enterprise sectors -where the institutional reforms proved to be more complex than foreseen. The key implication has been the need to focus on implementation and institutional issues more forcefully. As these are better addressed through investment or free standing technical assistance projects, the main lesson was the importance of integrating fully project lending in the design and implementation of structural adjustment. E. Borrower's Performance Compliance with Credit Agreement 28. Conditionality for release of the Credit's second tranche is described in Annex 1. The first supervision mission in November/December 1987 reported on the liberalization measures adopted by the Government and on the studies, reviews and implementation plans that were or had been carried out as requested under the SRC. All these were considered as justifying the release of the second tranche. Disbursements 29. The total of SDR 69.0 million from IDA and the Special African Facility (SAF) was to be disbursed in two tranches. During the first two years of the Credit, the actual rate of disbursements was much in line with the one estimated at appraisal. Afterwards, disbursements lagged and this was caused by several factors, including:  I1 Length of time needed to negotiate letters of credit (from three to six months), which often delayed the submission of the documentation required by the World Bank. As most letters of credit were being drawn against the special account (revolving fund), funds were tied up for long periods of time thus preventing the timely replenishment of the special account. This in turn caused a temporary lack of funds to finance new imports. * Enterprises' lack of own funds. As importers applied for domestic credit only once their request for foreign exchange had been approved, delays were incurred. * Lack of coordination between the Bank of Mozambique, as the interlocutor of IDA in administering disbursements, and the Bureau for Coordination of Import Programs (GCPI) in the Ministry of Commerce which, as the executor in the utilization of the funds to be disbursed, issues the import licenses and oversees procurement procedures. 30. As a consequence of the delays in disbursements, the closing date of the SRC was extended three times. The Credit closed officially on June 30, 1991. The remaining undisbursed amounts under the IDA credit (SDR 81,860) and SFA (SDR 117,758) were canceled. The Dutch grant was used in its entirety. The remaining balance from the DANIDA grant was reallocated to the ongoing Third Rehabilitation Credit. Audits 31. Audits of the expenditures under the Credit were not done on a yearly basis or on time, as required by the Credit Agreement. A report prepared by the Department of Inspection of the Ministry of Finance was submitted for fiscal year 1988, but it was not fully satisfactory as some of the supporting documentation was not present such as the number of enterprises benefiting from the Credit. IDA accepted this first audit of accounts, but indicated that the audit authority was not sufficiently independent of the Credit beneficiary to meet the audit covenant in the future. 32. Because of the number of overdue audit reports for IDA projects, the Bank of Mozambique has hired a consulting firm of auditors to do aDl outstanding audits other than for the energy sector projects, which were audited separately. Audit of the SRC was submitted for expenditures up to end-1990; a final audit on the SRC is due shortly. Procurement 33. Procurement was limited to eligible imports on the basis of a negative list in accordance with Schedule 2 of the Development Credit Agreement and foDowing  12 IDA guidelines. Procurement under the African Facility Credit and Swiss Grant followed the same procedure, but was limited to the BDA Part II member countries and any other country meeting procurement criteria in the Resolution establishing the Special Facility for Sub-Saharan Africa. Purchases under contracts of more than US$2.0 million equivalent were to be procured through international competitive bidding. 34. Several factors affected procurement adversely, namely: * The procurement guidelines of IDA and donors were new to the newly created Bureau for the Coodination of Import Programs (GCPI), and enterprises benefiting from the imports program were not familiar with the use of "technical specifications" in procuring certain goods, having previously bought only directly from traditional suppliers. * The general procurement practices of the private sector were, in a number of cases, not adequate for donor support. Simplification and harmonization of procurement guidelines had therefore to be agreed with donors and introduced. F. Bank Supervision 35. The first Bank supervision, at the end of 1987, looked into compliance of conditionality for release of the second tranche. The second one was done in conjunction with the IDA Third Rehabilitation Credit and looked into issues of procurement and delays in submission of audit reports. The third and fourth missions in 1990 and 1991, respectively, focused on procurement and disbursement issues and on obtaining information for this report. The task manager was appointed Resident Representative and, thereafter, supervision began on a regular basis, without requiring formalized supervision missions.  13 PART I - REVIEW FROM BORROWER'S PERSPECTIVE 38. The objective of the SRC was to support the policy reforms of the Government of Mozambique, particularly in the areas of prices, marketing, fiscal policy including public expenditures, extemal trade, allocation of foreign exchange, and support to the import program, smallholders and industrial production. 39. Some of the problems which arose during the implementation of the program were promptly overcome once more experience was gained. 40. One of the problems mentioned in Part I of the present report is the incapability of the institutions to deal with the procedures of the World Bank, in particular GCPI and BDM due to: (i) weak familiarization with procurement requirements; (ii) difficulties in interpreting and implementing the procedures recommended by the World Bank; (iii) deficient coordination between the coordinator and executing entities of the credit, particularly as concems the mechanisms of disbursement of funds; and (iv) weak capacity to monitor disbursements by BDM, due to the lack of periodical reports by the coordinator of the project. 41. The closing of the Credit was delayed due to the difficulties experienced by the executing agencies in following the required procedures. It was not easy for BDM to verify the process of release of funds and prepare all the work related to the disbursements. On the other hand, GCPI had been recently created and its staff was not duly prepared to deal with World Bank procurement procedures. The same constraint applied to enterprises, which were not familiar with the use of technical specifications in procurement. This has caused delays in the disbursement of funds, and it has been agreed that technical assistance would be necessary for future IDA operations. For all those reasons, the closing date of the Credit had to be extended three times until June 30, 1991. Al these factors were addressed in follow-up IDA adjustment credits. 42. As a result of all the various constraints mentioned above, audits of the SRC was not done on a yearly basis, as required. The audits were performed only in 1991, for the accounts related to 1990. An agreement has been reached between the Government and IDA as concerns the audit of the 1991 accounts.  14 43. Despite these problems, the SRC has led to the formulation of more liberal policy measures on the part of the Government in implementing the Economic Rehabilitation Program (ERP). 44. The lessons learned during the implementation of this Credit indicate that a reinforcement of the institutions has become a priority, and talks are being held with the World Bank for that purpose, and provisions are being set in place in all sectoral operations.  15 PART Ell: STATISTICAL INFORMATION I - PROGRAM TIMETABLE Original Date Actual Date Initiating Memorandum Feb-27-87 Feb-27-87 Appraisal Mar-16-87 Mar-16-87 Negotiations Jun-22-87 Jun-22-87 Letter of Development Policy Jun-26-87 Jun-26-87 Board Approval Aug-4-87 Aug-4-87 Credit Agreement Sep-8-87 Oct-5-87 Effectiveness Oct-22-87 Oct-22-87 Second Tranche Release Feb-4-88 Feb4-88 Credit Closing Dec-31-88 Jun-30-91 II - MISSION DATES Date No. of Weeks Persons Staff Weeks Report Date Preparation Nov-86 6 7 42 Dec-30-86 Appraisal Mar-87 3 6 18 Jun-16-87 Supervision Nov-87 2 5 10 Nov-28-87 Supervision May-89 1 2 2 Jun-28-89 Supervision Nov-90 1 1 1 Dec-21-90 Supervision Jul-91 1 1 I Sep-10-91  16 III - CREDIT DISBURSEMENTS (SDR Million) Cumulative IDA Cumulative SFA SDR Million % of Total SDR Million % of Total FY88 33.1 60.7 8.9 61.7 FY89 48.7 89.4 9.0 62.0 FY90 54.4 99.8 11.5 79.4 FY91 54.4 99.9 14.3 99.2 VI - FINANCING (SDR Million) IDA Special Total Credit Facility for Africa Initial amnount 54.4 14.5 68.9 Actual amount 54.5 14.5 69.0 Disbursed (as of 6/30/91) 54.4 14.4 68.8 Cancelled (as of 6/30/91) 0.1 0.1 0.2 Repaid (as of 6/30/91) 0.0 0.0 0.0 Outstanding (as of 6/30/91) 54.4 14.4 68.8  COMPLIANCE WITH COVENANTS ANNEX I (For Release of Second Tranche) (Page 17) COVENANTS ACTION TAKEN A. PRODUCTS ON FIXED PRICE LIST Reduce number of product groups on Prices of several products were no longer fixed but fixed price list from 46 to 37. were moved to the conditioned price system. Furnish IDA with plan of actions for Complied with. further reductions from 37 to 32. B. FOREIGN EXCHANGE Establish improved allocation system which Ministry of Trade established GCPI (Bureau to shall, inter alia, improve the matching of Coordinate Import Programs) to maintain up to date donor assistance with the needs of the information on import requirements, assist in economy and establish priorities for the use mobilizing donor aid, and coordinate procurement. of foreign exchange taking into account the overall economic efficiency of the operations of proposed users. Review export retention scheme and complete Review was focused on improving existing foreign study of interim trading measures to improve exchange retention system and establishing system the flexibility of trade. whereby individuals with access to foreign exchange could import goods automatically. Submit action plan to implement results of Complied with. study. C. IMPORT/EXPORT Reduce number of products which must Restrictions on paper, fishing gear, textiles, tentalite be imported through only one trading and citrus were removed. corporation from 11 to 5 and the number of export commodities from 3 to 1. D. ENTERPRISE RESTRUCTURING Complete study of 25 industrial and 15 Study was completed. Report also reviewed in detail agricultural enterprises. rehabilitation plans for five major enterprises. Submit plan of action for restructuring of Action plan was submitted. these enterprises.  COMPLIANCE WITH COVENANTS ANNEX I (For Release of Second Tranche) (Page 18) COVENANTS ACTION TAKEN E. MARKETING Complete study of fruits and vegetables Thorougb study was completed. marketing system. Review the domestic allocation system by Govenmment confirmed that review was on track; which goods are administratively allocated. some decisions were made concerning the coverage of the system. In 1987, some products (shoes, fish, salt) were removed from the system.