Uganda - Economic Recovery Program

Document of The World Bank FOR OFFICIAL USE ONLY IICROFICHE COPY Report No. 10415-UG Type: (PCR) NAMISATO, / X31678 / T9 105/ OEi)D2 Report No. 10415 PROJECT COMPLETION REPORT UGANDA FIRST ECONOMIC RECOVERY CREDIT (CREDITS 1844-UG AND A-34-UG) MARCH 12, 1992 Country Operations Eastern 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 Period Currency Unit (annual average) USh per USS US$ per USh 1987/88 60 0.01667 1988/89 170 0.00589 1989/90 320 0.00313 1990/91 558 0.00179 GOVERNMENT FISCAL YEAR July 1 to June 30  THE WORLD BANK FOR omcuu USE ONLY Washington, D.C. 20433 U.S.A. Officei 4 fft"DectmGenval Operationm Evakhitmo March 12. 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on UGANDA - First Economic Recovery Credit (Credits 1844-UG and A-34-UG) Attached, for information, is a copy of a report entitled "Project Completion Report on Uganda - First Economic Recovery Credit (Credits 1844-UG and A-34-UG)," prepared by the Uganda Resident Mission, with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. 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 b* disclosed without World Bank authorization.  FOR OFFICAL USE ONLY ABBREVIATIONS AND ACRONYMS ADF African Development Fund AfDB African Development Bank BOU Bank of Uganda CIDA Canadian Intemational Development Agency CMB Coffee Marketing Board DANIDA Danish Intemational Development Agency EAU Economic Analysis Unit EEC European Economic Community ERC Economic Recovery Credit ERP Economic Recovery Program ESAF Enhanced Structural Arrangement Facility FEIL Foreign Exchange and Import Licensing Committee GDP Gross Domestic Product IDA International Development Association IMF International Monetary Fund KfW Kreditanstalt fuer Wiederaufbau VCs Letters of Credit LMB Lint Marketing Board MOF Ministry of Finance MPED Ministry of Planning and Economic Development MPSCA Ministry of Public Service and Cabinet Affairs NRM National Resistance Movement ODA Overseas Development Administration (United Kingdom) QOL Open General Licence PCR Project Completion Report PES Public Enterprise Secretariat PFP Policy Framework Paper PSRRC Public Service Review and Reorganization Commission PTA Preferential Trade Area RDP Rehabilitation and Development Plan SAF Structural Adjustment Facility SCF Special Credit Facility SDR Special Drawing Rights SIDA Swedish International Development Authority SIP Special Import Program SPA Special Program of Assistance UCB Uganda Commercial Bank UDC Uganda Dvelopment Corporation USh Uganda shilling 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.  PROJECT COPIJET.JOE REPORT FIRST ECONOMIC RECOVERY CREDII (CREDITS 1844UG AND A-34-UG) Table of Contents PREFACE ........................................................... i E'VALUATION SUMNMARY . iii PART I. PROJECT REVIEW FROM THE BANKS PERSPECTIVE. 1 A. Project Identity. 1 B. Project Background ......1................ I Economic and Political Developments Prior to ERC I. 1 Bank Group Strategy and Economic and Sector Work. 2 C The First Economic Recovery Credit. 3 Origins and Objectives. 3 Policy and Institutional Measures Supported by ERC I. 4 Implementation and Monitoring of Measures Supported by ERC I 4 Evaluation of Performance .......... ......................... 10 Other Aspects of ERC I ............ ......................... 13 Co-financing of ERP ............. .......................... 14 The Results of the ERP ............ ......................... 14 D. The Main Lessons Learned ........................................ 15 PART IL PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE ..... ... 17 A. Background, Evolution and Development of the Program ..... ............ 17 B. Implementation of the Program ..................................... 17 C. Evaluation of the Program ......................................... 20 PART Im. STATISTICAL INFORMATION ................................. 23 Table 1 CreditPosition ........................................... 23 Table 2 Cumulative Estimated and Actual Disbursements ................. 23 Table3 Project Dates ............................................ 23 Table 4 Staff Inputs ............. ................................ 24 Table 5 Mission Data ............................................ 24 Table 6 Other Project Data .................................... 24  Table of Contents (cont'd) Table 7 GDP at Factor Cost, 1985-1990, at Constant (1987) Prices ........ .. 25 Table 8 GDP in Constant 1987 Prices - Percentage Annual Growth Rates .... 26 Table 9 Production and Exports of Principal Agricultural Products ... ....... 27 Table 10 Production of Selected Goods ............ ................... 28 Table 11 Index of Industrial Production ........................... 29 Table 12 Balance of Payments ................. ..................... 30 Table 13 Ugan ia: Central Government Budgetary Operations, 1988/99-199091 ................... ....................... 31 Table 14 Monetary Survey, 1987-1990 ........ ......................... 32 Table 15 Structure of Interest Rates 1986-1990 ......... .. ............... 33 Table 16 Exchange Rates .. ............... 34 Table 17 Kampala Cost of Living Index Middle Income Group ............ .. 35 Table 18 New Consumer Price Index: Kampala ......... .. ............... 36 Table 19 Average Market Prices for Selected Consumer Goods, Kampala ... 37 Table 20 Producer Prices for Export Crops .......... .. ................ 39 Table 21 Consumer Prices of Petroleum Products ........ .. .............. 40 Table 22 Disbursement of ERC 1 to 30 June 1991 .......... ............. 41 $1  I - RT~-1 PROJECT COMPLETION REPORT UGANDA FIRST ECQ0NOMIC RECOVERY CREDIT (CREDITS 1844-UG AND A-34-UG) PREFAC 1. This is the Project Completion Report (PCR) for the First Economic Recoveiy Credit (ERC I) in Uganda, for which IDA Credit 1844-UG and African Facility Credit A-34-UG, in the amounts of SDR 50.9 million and SDR 18.8 million respectively, weir pp d on September 15, 1987. In the case of the IDA Credit the Board approved thr- plements as follows: SDR 1.3 million in Februaly 1989, SDR 19 million in April 1989 ana SDR 1.2 million in November 1989. The Board approved a second tranche of SDR 10 million under the African Facility Credit on December 18, 1989. The credits were fully disbursed; the last disbursement was in August 1991. Nevertheless, the closing date was extended to June 30, F 1992 in order to allow the Government more time to utilize small balances remaining on the accounts in respect of funds provided by two of the co-financiers. 2. The PCR was prepared by the Uganda Resident Mission (Preface, Evaluation Summary, Part I and Part IIl and by the borrower (Part II). 3. Preparation of this PCR was based on, inter alia, the PresiJent's Report, the credit agreements, supervision reports, correspondence between the Bank and the Borrower and internal Bank memoranda.  PROJECT COMPLION REPORT UGANDA FIRST ECONOMIC RECOVERY CREDIT (CREDITS 1844-UG AND A-34-UG) EVALUATION SUMMARY aclairound f. The Ugandan economy suffered massive damage throughout the 1970s owing to mismanagement, political turmoil and civil war. Both the human capital and the physical infrastructure were neglected, destroyed and alowed to decay. Producer incentives became severely eroded. The outcome of this turn of events was a 20 percent decline in real GDP during the 1970s. The situation changed for the better between 1981 and 1984 as the second Obote regime introduced, with the support of the Bank, the Fund and other donors, a wide range of policy reforms aimed at promoting economic recovery. However, before they had had time to work, these efforts were overwhelmed by an intensified civil war which took a heavy toil on human life and property, especiaDy in the Luwero triangle. ii. The National Resistance Movement (NRM) which assumed power at the end of the civil war in January 1986 inherited an economy which was in ruins. The war had disrupted crop production and marketing. The roads were in an acute state of disrepair; there was a severe shortage of trucks and other transport equipment; manufacturing activity had virtually ground to a halt; water and electric power supplies were at best erratic. Foreign exchange reserves were depleted. Moreover, large parts of the north and east of the country continued to be plagued by insecurity. In responding to these challenges the Government initialy took an interventionist stance and adopted expausionary fiscal and monetary policies. Notwithstanding tentative signs of economic recovery, the results of these policies were disastrous, as evidenced by the sharp acceleration of inflation (to over 300 percent) and the substantial deireciation of the Ugandan Shilling in the parallel market. It did not take the Government long to realize that its interventionist approach to economic management had failed. It then decided to try the market-oriented policies advocated by the Bank and the Fund. Objectives iii. The objectives of the Economic Recovery Program (ERP) were to: bring about internal financial stability and lower the rate of inflation; reduce the imbalances in the external accounts; and promote growth. Given the severe imbalances in the economy the program accorded high priority to stabilization. It also included structural measures aimed at improving producer incentives, increasing capacity utilization in industry, improving the allocation and use of public resources, raising the standard of public administration and rehabilitating the infrastructure.  - iv- liglementation Experience iv. The Government took a number of strong adjustment measures up-front (in May 1987). These included the devaluation of the Shilling by 77 percent iP foreign currency terms, hefty increases in the producer prices of export crops and big increases in the pump prices of petroleum products. Other notable measures taken under ERC I were the rapid dismantling of price and distribution controls, further large devaluations of the Shilling, the introduction of a limited Open General License (OGL) system of allocating foreign exchange, the creation of a Special Import Program (SIP) for foreign exchange allocation on a first-come first-served basis and the preparation of verified staff lists and a list of approved and filled positions which would form the basis for an Establishment Register. Implementation was, however, very disappointing in the area of stabilization: the fiscal deficit and monetary targets were exceeded by a wide margin, resulting in accelerating inflation well into the second half of 1988. Moreover, the nominal exchange rate was kept fixed for long periods or adjusted only marginally. Weak ii plementation was a reflection of two factors: a lack of the capacity to carry out some of the agreed actions an-- i..sufficient understanding and commitment to the program. v. For its part, IDA szriously underestimated the extent to which the human resource and data base as well as management systems in the public service had been destroyed by years of political and military turmoil; hence some of the heavy demands made on the Government (for example, IDAls expectations regarding the preparation of a public investment program and the verification of the ownership of the Custodian Board properties could not be met). The program was largely formulated by Bank and Fund staff and pushed through the highest echelons of the Government, leaving many Cabinet ministers and senior and middle level public servants ignorant of, or hostile to, the program in general and the adjustment measures in particular. Of course, it was the Government's decision to keep the dialogue withini a narrow circle of officials; this decision reflected the Government's initial distrust of persons who had served previous regimes. When the program, especially the stabilization component, ran into serious difficulties soon after it was launched (because of lack of fiscal discipline coupled with delays in the flow of external resources), IDA and the Fund had no option but to accept large deviations from the original performance targets while intensifying the policy dialogue with the Govermment. This sympathetic stance paid off in that the dialogue succeeded in forestalling total abandonment of the program. Results VL In spite of the implementation problems encountered, ERC I, together with the first annual arrangement under the IMFs Structural Arrangement Facility (SAF), initiated a policy dialogue between Govermnent and IDA which has borne fruit. Thus, the Government came to realize the futility of price controls and eliminated these controls except on petroleum products; it accepted the imperative of economic and financial stabilization and therefore made a more determined effort to bring aggregate expenditure under control; it introduced an export retention scheme and extended it to all exports except coffee, in partial recognition of the need for an exchange rate that is conducive to external competitiveness; and it accepted the need for divestiture  - v - of public enterprises. The OGL, and later the SIP, were instrumental in generating a strong supply response, especially in the industrial sector. With increased activity in other sectors, GDP grew by over 6 percent per annum during the life of ERC I. Sustainabilit vii. The Government did not show a strong commitment to the economic stabiiization measures supported by ERC I until well into 1989. There was, for example, a tendency to question the efficacy of devaluation, the incentive effects of higher producer prices (especially for coffee farmers) and the benefits of monetary restraint. As the costs of the failure to implement the reforms vigorously continued to mount (as shown by the persistwnce of high inflation, the depreciation of the Shilling in the parallel market e.t.c.) the Government gradually came round to accept the need for stabilization and structural adjustment. A subsequent operation, the Second Economic Recovery Credit (ERC II), has allowed the Government to build upon the reforms initiated under ERC I. Without ERC IT, * was little or no chance of the reforms supported by ERC I being sustained. Findines and Lessons Learned viii. Government did not, at the outset, make a strong effort to generate popular understanding of, and support for, the policies it had adopted; hence the widespread perception in the country that what the Government was implementing were Bank and Fund policies. For its part, IDA was handicapped by the lack of up-to-date economic and sector work and therefore failed to appreciate fully the weaknesses in public administration. As it turned out, the ability of the key economic institutions to implement agreed policies and programs depended almost entirely upon the performance of a handful of individuals. Weak implementation capacity was compounded by the absence of co-ordination among the relevant institutions. The problem of poor co-ordination was only partially solved through the establishment of an Economic Monitoring Committee chaired by the Finance Minister. Although an effort was made under ERC I to start the process of addressing weaknesses in public administration, it became more and more apparent that, to make an impact, that effort must go much further to fill critical gaps in staffing, generate management information and restore management processes as well as address problems of co-ordination across ministries.  PROJECT COMPLEllON REPORT UGANDA FMRST ECONOMIC RECOVERY CREDOT (CREDITS 1844-UG AND A-34-UG) PART I. PROJECT REVIEW FROM THE BANK'S PERSPECI A. Project Identity - Project Name: First Economic Recovery Credit - Credit Nos.: 1844-UG and A-34-UG - RVP Unit: Africa Region, Country Department I - Country: Uganda - Sector: Macroeconomic B. Project Background Economic and Political Developments Prior to ERC I 1. The 1970s were characterized by unprecedented political repression and gross economic mismanagement in Uganda. Thousands of educated Ugandans were either killed or forced to flee into exile. The economic and social infrastructure and productive plants deteriorated progressively for lack of maintenance. The 1978-1979 civil war which culminated in the overthrow of the Amin regime inflicted further damage on the economy. As a result of these developments real GDP declined by 20 percent in the 1970s. The second Obote Government which came to power in 1980 adopted a comprehensive package of stabilization and structural adjustment measures, with the support of the Bank, the Fund and other donors. The Government also appeared to give the country a semblance of political stability. However, its legitimacy was chalienged right from the outset. This challenge soon turned into a guerilla war against the regime. The regime was overthrown in a coup d' etat in July 1985 but the civil war continued until January 1986 when the NRM assumed power. 2. The NRM inherited an economy that was in ruins. The war had severely disrupted the production and marketing of crops and livestock as farmers fled the fighting and the soldiers plundered whatever they could lay their hands on, especially in the Luwero triangle which bore the brunt of the fighting. Industrial enterprises remained idle for lack of spare parts, intermediate inputs and utilities such as power and water. The roads were in an acute state of disrepair and the stock of goods and passenger vehicles was severely depleted. The power and water supply and telecommunications systems were barely functioning. Education and health services had been badly disrupted by the wvar. The economy was beset by serious imbalances, particularly between the demand for and supply of foreign exchange and between Government revenue and expenditure. Public administration was in a shambles. Worse still,  -2- there was, in the north and east, continued fighting between the National Resistance Army (NRA) and forces opposed to the new Government. 3. The NRM Government's first response to the daunting problems that it inherted was to intervene heavily in the economy. The Government monopolized the internal and external trade in a number of commodities (sugar, salt, wheat flour e.t.c.) and controlled the prices of these and other commodities. The Shilling was revalued and the official exchange rate fxed at one-third the parallel market rate. While maintaining a very overvalued e-hange rae the Government raised crop producer prices significantly. The producer price or robusta coffee, for example, was increased by 81 percent. Desirable as this price increase was, it contributed to the erosion of the tax revenue from this important source. (Coffee tax revenue in 1986/87 was 1.7 percent of GDP compared to 4.9 percent in the previous year). This was at a time when the Government had doubled budgetary outlays and the capacity to collect other taxes had all but collapsed. 'ro finance the resulting deficit the Government accumulated new external payments arrears and borrowed heavily from the central bank. This expansionary jiscal stance, coupled with a 238 percent increase in credit to the private sector (excluding coffee), transportation bottlenecks and an acute shortage of foreign exchange, stoked the fires of inflation; by December 1986 the year-on-year rate of inflation had reached 356 percent compared to 130 percent twelve months earlier. 4. The sharp acceleration of inflation summed up best the failure of the interventionist economic policies pursued by the Government in 1986/87. To its credit the Govemment drew the right lessons from this experience. The basic lesson learnt was that, rather than improving supply, econc iic controls accentuated the shortage of goods and services. Having reached this concusion, the Government then set about examining an alternative direction for economic policy. Bank GrouR Strateg and Economic and Sector Work 5. The Bank Group's strategy was tailored to the peculiar circumstances of Uganda, a country whose infrastructure had been destroyed and the economy ceverely dislocated by a succession of civil wars. On-going traditional investment projects i;; more or less come to a standstill for a variety of reasons, including insecurity and the poor performance of the civil service. One element of the Bank's strategy was, therefore, to assist the Government to remove infrastructure bottlenecks affecting production and marketing. This meant the resumption of the implementation of already approved projects at the earliest opportunity and the identification and preparation of new investment projects. The second element of the strategy was to provide quick-disbursing funds in support of reforms aimed at restoring macroeconomic stability and improving the structure and functioning of markets and public institutions. Aside foom the rehabilitation of infrastructure, these reforms held the key to sustained economic recovery. 6. The economic and sector work underpinning ERC I was summarized in the 1985 Country Economic Memorandum entitled: Progress Towards Recovery and Ptospects for Development. The macroeconomic framework underpinning ERC I was developed jointly with tne Government and the IMF in the context of the Policy Framework Paper (1987/88- 1989190). Sector work needed updating in all areas because the Bank had been inacJve in Uganda for a number of years. In agriculture the gap in the Bank's knowledge of the sector  -3 - was filled to some extent by the reports of the Agricultural Task Forces covering marketing, inputs, credit, research and other areas. The Task Forces were supported by the Bank C: The First Economic Recovery Credit Orieins and Obiectives 7. The NRM Government came to power very suspicious of the Bank (and the IMF) and the market-oriented policies that these institutions advocate. It was also irked by the fact that the two institutions had provided much financial assistance to the second Obote regime. Relations between the Bank and Uganda therefore remained cool for most of 1986. Towards the end of that year Bank staff held preliminary discussions with Government officials conceming a possible package of economic reforms that Uganda could adopt. In January 1987 the Government, having agreed with the outline of the policy package, formally requested the Bank to provide quick-disbursing funds and technical assistance and to convene a Consultative Group meeting. At about the same time the Government expressed interest in the use of resources from the Fund-s Structural Adjustment Credit (SAF). The stage was thus set for dialogue and co-operation between the new Government and the Bank and the Fund. The staff of the two institutions then collaborated with the overnment in preparing the first Policy Framework Paper (PFP). The PFP was approved by the Committee of the Whole on May 26, 1987. A Consultative Group meeting took place in June 1987 and gave strong support to the Economic Recovery Program (ERP) that was announced by the Government on May 15, 1987. The Paris Club met on June 18, 1987 and responded favorably to Uganda's request for debt relief ERC I was appraised in May-June 1987, negotiated in July and approved by the Board on September 15, 1987. 8 The basic objective of ERC I was to support the Government's Economic Recovery Program. The aims of the ERP were to: (i) bring about internal financial stability and lower the rate of inflation; (ii) reduce the imbalances in the external accounts; (iii) raise capacity utilization in industry; (iv) rehabilitate infrastructure and productive plants; (v) restore producer incentives; (vi) re-establish discipline, accountability and efficiency in the public service, and (vii) improve public sector resource mobilization and allocation. In furtherance of these aims the Goverment implemented at the outset the following important policy measures: (a) the introduction of a new currency equivalent to 100 old Shillings, coupled with a 30 percent tax on currency and bank deposits held by the public; (b) the devaluation of the Shilling by 77 percent in foreign currency terms, from UShl4 per US dollar to USh6O; (c) increases ranging from 150 percent top 400 percent in the producer prices of the traditional export crops, and (d) big increases in the prices of petroleum products.  - 4- 9. This was an important beginning to reform but it was clear that much more needed to be done to stabilize the economy and create conditions conducive to sustained growth. The exchange rate was still grossly overvalued; there was a huge excess demand for foreign exchange; inflation remained very high; there were many price and distribution controls; and public sector institutions exhibited glaring weaknesses. ERC I was therefore designed to broaden and deepen the above policy initiatives, in conjunction with the SAF which was approved by the Fund's Board in June 1987. Policy and Institutional Measures Supported by ERC I 10. The measures supported by ERC I fall into the foDlowing three broad categories: (a) actions directed at restoring macroeconomic stability; (b) actions aimed at improving producer incentives and giving markets a greater role in the allocation of resources, and (c) measures geared towards strengthening public sector management. 11. At the time of launching the ERP issues of short-term stabilization loomed ven large and were in fact accorded the highest priority. The stabiiza jon measures supported by ERC I were basicaly agreed in the context of the PEP and the ',AF arrangement with the IMF. These measures may be summarized thus: the budget deficit was to be kept around 3.5 percent ane, even more importantly, there was to be a progressive reduction in Government indebtedness to the banking system; monetary and credit policies were to be tightened considerably; and the Shilling was to be devalued in the light of domestiz and international economic developments. The Government agreed to reinforce these measures by instituting structural reforms aimed at improving tax administration, budget formulation and expenditure control. 12. The second set of actions supported by ERC I was aimed at promoting economic recovery through improved producer incentives and greater reliance on market forces. In this area the Government agreed: to review and increase as necessary the producer prices of the traditional export crops (coffee, tea, cotton and tobacco); to ensure prompt payments to the farmers; to liberalize agricultural marketing; to Umit price controls to a handful of items; to improve the system of foreign exchange allocation; to keep the exchange rate competitive; to speed up the preparatory work on the disposal of the properties under the Custodian Board; and to review and simplify industrial licensing. 13. In view of the devastated state of public institutions, actions directed at strengthening public sector management also featured prominently in the ERP. Specifically, the Government undertook to improve planning and budgeting, rehabilitate the civil service; and to initiate public enterprise reforms, including divestiture. Implementation and Monitoring of Measures Sugr -1ed by ERC I 14. Economic Stabilization. Ile implementation of the measures aimed at restoring macroeconomic stability got off to a very poor start. The clearest manifestation of this was  -5- the announcement of a budget for 1987/88 which was much more expansionary than what had been agreed with the Fund in the context of the first annual arrangement under the SAF. The budgetary allocation for defence, understandably in reflection of the unsettled security situation prevailing in the country, was particularly large. Unplanned outlays for the rehabilitation of hotels needed to host the 1987 Preferential Trade Area (PTA) summit inflated Government expenditure further. The fiscal problem was compounded by the emergence of big shortfalls in tax revenue early in the year. Given the seriousness of the emerging fiscal problems, the issue of a letter by the Finance Minister to the IMF setting out specific adjustments to the budget was made a condition of Board presentation of the credit. 15. The Government attempted mid-course corrections in August 1987 and January 1988 by, among other things, lowering some of the original expenditure estimates and increasing the pump prices of, and the duty on, petroleum products. These efforts proved largely unsuccessful. The result was that the deficit on a commitment basis increased significantly. In reflection of this development, net bank claims on Government increased by USh7.8 billion instead of declining by USh2 billion as programmed under the SAF. The 1988/89 fiscal program fared no better, even though the revenue outturn was very much in line with the original budget estimates. The actual deficit on a commitment basis was 54 percent higher than the SAF target and for the second year in a row the Government had heavy recourse to bank borrowing to finance the deficit. This time the Government's indebtedness to the banking system went up by USh12.3 billion instead of declining by USh5 billion as envisaged under the SAF. The original budget for 1989/90 also deviated significantly from the targets agreed to in the context of the first annual arrangement under the ESAF. A revised budget had to be worked out in October 1989 with the assistance of IDA and Fund staff. 16. The main reason for the large deviations from the SAF/ESAF fiscal targets at the estimates stage was the tendency to overestimate expenditure, especially extemally-financed expenditure. As far as the fiscal outcomes were concerned, the deviations from the planned levels of revenue and expenditure were due mainly to the Government's failure to control expenditure, especially defence expenditure, and to collect revenue, particularly coffee tax revenue. Another contributory factor was weak monitoring of the program in general and the budget in particular; until the Economic Analysis Unit (EAU) in the Ministry of Fmance succeeded in creating the necessary database, it was left largely to the Fund and the Bank to identify slippages in budget implementation and to propose corrective actions. 17. Increased Government borrowing from the banking system during the first two years of the ERP served to undermine the monetary and credit policies agreed to under the SAF arrangement. At the same time credit to the private sector was allowed to expand much faster than programmed. Credit to the private sector rose by 200 percent in 1987/88 and by 248 percent in the following year. In December 1988, in an effort to ensure prompt payment to the coffee farmers, the Government transferred the responsibility for the provision of crop finance from the commercial banks to the Bank of Uganda (BOU). This led to an explosive increase in credit provided to the Coffee Marketing Board (CMB), the co-operative unions and the primary societies; crop finance outstanding thus rose from USh 4.5 billion in June 1988 to USh 19.5 billion in June 1989 and increased further to USh 25.0 billion by the end of December 1989. In reflection of these credit developments, broad money increased by 231 percent in 1987/88 compared to a target of 40 percent and by a further 125 percent in 1988/89 as against a program target of 65 percent.  -6 - 18. The credit stance adopted by the authorities was very passive. This stance made it possible for the Uganda Commercial Bank (UCB) and the Co-operative Bank to expand their lending well beyond the dictates of banking prudence. Indeed, with the acquiescence of the monetary authorities, these two banks were able to run up huge overdrafts with the central bank. At the beginning of the ERP the Government had in fact made matters worse by reducing interest rates from levels which were already highly negative in real terms. Deposit rates were reduced from a range of 10-35 percent to 7-22 percent while the maximum lending rate was reduced from 42 percent to 30 percent. In spite of the persistence of high inflation (well in excess of 200 percent per annum), the Government kept interest rates unchanged until July 1988 when the rates were increased across the board by 10 percentage points. Further increases of 5 and 10 percentage points in deposit rates and lending rates, respectively, followed in March 1989. The rates remained frozen at the March 1989 level for the next fifteen months. Although this was a period of rapidly declining inflation, real interest rates were still quite negative up until December 1989. 19. The Government was equally passive in the realm of exchange rate policy for much of the period covered by ERC I. There was clear evidence of growing excess demand for foreign exchange: the parallel market exchange rate depreciated rapidly during the first year of the ERP. The rate dropped from USh1O6 per US dollar in July 1987 to USh 451 in June 1988. In that period the cumulative increase in prices was 133 percent. The Government, however, chose to maintain the official exchange rate established in May 1987 until the beginning of July 1988 when the Shilling was devalued by 60 percent in foreign currency terms. Except for two small devaluations between December 1988 and March 1989, the Government maintained its passive stance on the exchange rate until October 1989. A 41 percent devaluation in foreign currency terms was announced on October 24, 1989. This marked the beginning of a more active exchange rate policy, as shown by more or less regular monthly devaluations of the Shilling. Between May 1987 and December 1989 the Shilling was devalued by 84 percent in foreign currency terms. 20. Very little progress was made in implementing the structural measures which were indispensable to the attainment of macroeconomic stability. The three main areas of concem were tax administration, budget formulation and monitoring as well as expenditure control. Studies which were to provide the basis for actions aimed at improving the tax structure and tax administration were either delayed or the reports turned out to be unsatisfactory. Still, the Government took some steps towards improving the tax structure: it extended the coverage of the sales tax to more locally-produced goods; lowered the corporate tax rate; eliminated the customs duty exemption for' a number of products and imposed a 10 percent minimum tariff on most imports; and kept petroleum taxation in line with exchange rate developments. If the budget estimates were any guide, the Government continued to place great emphasis on the revenue from the export duty on coffee. The actual results were, however, very disappointing. In 1987/88 actual coffee tax revenue amounted to USh6.3 billion. In the following year USh19 billion (original estimate) was expected from the coffee duty but only USh5.4 billion was collected. The sharp fall in international coffee prices was partly responsible for the low tax receipts from coffee. However, the problem was due largely to poor accounting and payment for coffee barter shipments, of which the Government itself was the main beneficiary. The Government also failed to address the issue of tax administration in any serious way. As a result, the tax departments continued to operate under severe handicaps as  far as work incentives, transport, office accommodation, technical forms, office supplies and other complementary inputs were concerned. 21. Improvements in budget format and processes did not proceed as far as had been expected. Again the problem was partly due to delays in carrying out the necessary studies. The Government was very slow in responding to proposals put forward by a consu!tant engaged to develop a program for budget reform. However, the Government established a new unit, the Economic Analysis Unit, in the Ministry of Finance (MOF) in 1987. Within a very short period of time this unit succeeded in developing a fairly comprehensive database which has been invaluable in the preparation and monitoring of the budget, principally the recurrent budget and in monitoring other macroeconomic developments. This in turn paved the way for the creation of an Economic Monitoring Committee chaired by the Finance Minister. The Ministry of Planning and Economic Development was not so successful in compiling data on project aid and the development budget. Consequently, data on commitments and the utilization of project aid remained subject to wide margins of error. 22. With regard to expenditure control, the Government relied on a system of requisitioning funds which was aimed at vetting each item of expenditure in order to prevent overspending and reduce corruption. The corollary of this was that only a small proportion of the budget was subject to automatic releases. Each requisition for funds had to be vetted by officials of three different departments namely, Budget, Treasury Office of Accounts and the Auditor General. The system was thus very cumbersome. Moreover, not only did it not prevent overspending where it mattered most, that is, in respect of defence expenditure and unbudgeted expenditure, but also it starved essential economic and social services of funds. 23. Economic Recoverv. The actions required to bring about economic recovery focused on improving producer incentives, increasing reliance on market forcms and rehabilitating the economic and social infrastructure. With regard to the traditional export crops (coffee, cotton and tea), the Government undertook to review producer prices twice a year on the basis of a methodology agreed with IDA, and to adjust prices so as to maintain producer incentives. The implementation of this policy, especially in the case of coffee, was at best half-hearted. First, the Government postponed the price increases due in October/November 1987. Tlhen it announced in January 1988 producer prices which did not adequately compensate the farmers for the inflation that had occurred. The producer price for robusta coffee (kiboko), for example, was increased by only 21 percent even though consumer prices had gone up by nearly 80 percent in the intervening period. Tea and cotton fared better; their respective producer prices were increased by 100 and 68 percent. Further producer price increases were announced in July 1988 (107, 100 and 150 percent, respectively, for coffee, tea and cotton). Again the new prices were not high enough to restore incentives to the real levels established in May 1987. The Government continued to raise, albeit inadequately, the producer prices of tea and cotton but froze the robusta kiboko price at the July 1988 level for the next two years. In fairness to the Government it should be stated that, given the apparently overriding concern with the attainment of macroeconomic stability, IDA and the Fund adopted a rather ambivalent position on the question of adequate producer prices for coffee farmers. 24. There were other problems in the agricultural sector. For one thing, the expected improvements in the efficiency of the agricultural marketing agencies did not materialize. This meant that the farmers were not paid promptly or regularly for produce delivered to the  -8- marketing agencies. In the case of coffee the situation deteriorated to the point where, by the last quarter of 1988, the farmers were not being paid at all. The Government intervened in December 1988 to remedy the situation. It directed the Bank of Uganda to take over from the commercial banks responsibility for the provision of credit to the CMB and to provide or guarantee credit to the co-operative unions and the primary societies. The result was a credit explosion which had disastrous consequences for the money supply and inflation. 25. With regard to the marketing of agricultural products, the Government acted to remove restrictions on the inter-district movement of produce and to end the Produce Marketing Board's monopoly on the export of food crops. It also announced that the co- operative unions were free to sell lint directly to the textile mills, provided the unions cleared up their indebtedness to the Lint Marketing Board (LMB). This announcement had no practical effect because the unions had no means of settling their debts to LMB. The LMB itself was expected to reduce its interest in the oil and soap industry to a minority position. This was not done. 26. As mentioned earlier, the Government only began to pursue an active exchange rate policy towards the end of 1989. The result was that the Shilling continued to be severely overvalued. This constituted - big disincentive to exports. In cognizance of this, the Government introduced in 1988 a scheme which allowed the exporters of non-traditional products to retain 100 percent of export proceeds. In March 1989 the scheme was extended to all exports except coffee. Exporters were free to use the proceeds to import goods under the 'No Funds Involved" import Ucense system but could not sell the proceeds to another party. Subject to this restriction, the retention scheme in effect allowed exporters to convert non-coffee receip.s into local currency at the parallel market exchange rate rather than the official rate. 27. The Government moved very quickly to dismantle price and distribution controls. Statutory controls on the prices of beer, soft drinks, soap, sugar, salt, hoes, cooking oil and other goods were phased out by the end of 1988; only the prices of petroleum products continue to be controlled by the Government The monopoly rights enjoyed by state trading organizations such as Foods and Beverages Limited and Produce Marketing Board in the importation and/or distribution of certain commodities were also abolished. 28 In an effort to improve the allocation of foreign exchange and assure certain manufacturing firms regular supply of foreign exchange, the Government introduced a limited Open General Licensing system in January 1988. The OGL covered the raw materials and spare parts requirements of firms producing the following commodities: soap, beverages, textiles, blankets, mattresses, tobacco and cement. The OGL remained limited, as the Government, pleading scarcity and uncertain supply of foreign exchange, refrained from expanding the system as originally envisaged. Instead of expanding the OGL, the Government introduced in December 1988 a Special Import Program under which foreign exchange is provided to users on a first-come first-served basis. SIP was intended to speed up the utilization of import support funds. Under ERC I the Government was expected to create a Special Credit Facility (SCF) to assist industrial firms which faced problems in raising local cover to buy foreign exchange. The Government and IDA had second thoughts about the need for such a credit facility. No action was therefore taken to set up the SCF. However, the Government made ad hoc arrangements with UCB which met the local cover  -9- requirements of a selected number of firms which were predominantly state-owned enterprises. 29. Under ERC I the Government also agreed to speed up the verification of the ownership of Custodian Board properties which have long cast a shadow over the investment climate and Uganda's commitment to the inviolability of private property rights. Verification would pave the way for the return of the properties to the former owners or disposal through sale or other means. In an effort to give momentum to the verification exercise and related matters the President appointed a full-time Deputy Minister in charge of the Custodian Board. Some progress was made towards the verification of claims to, and the status of, the properties but it was difficult to ascertain the exact scope of work that was done. 30. The Government had undertaken to review, by December 1987, the system of industrial licensing, including the relevant legislation. This was not done. Rather, the Government initiated work leading to the drawing up of a new investment code. Public Sector Manaiement. The Government took a number of measures aimed at strengthening public sector management. The main measures are enumerated here. (i) As part of the 1988/89 budget it announced a 25 percent cut in the wage bill for group employees. (ii) It prepared verified staff lists for the traditional civil service, with the exception of the Teaching Service and the Local Authorities. (iii) It compiled a list of approved positiors, together with filled posts, and initiated action towards the publication of an Establishment Register. (iv) The staff lists were linked to each ministry's or department's payroll; in this way the Government was able to weed out a number of ghost employees, with resulting budgetary savings estimated at USh1.2 billion in 1989/90. (v) A functional review of the portfolios of 14 of the key ministries was also completed. (vi) The Government revived the Public Accounts Committee of Parliament to scrutinize the Auditor General's report on financial reports prepared by the Ministry of Finance. (viii) A number of public servants, including some very senior ones, were dismissed or prosecuted for indulging in corrupt practices. (ix) Tne Government appointed a Public Service Review and Reorganization Commission (PSRRC) in 1989 to examine the structure, organization, staffing, remuneration, management and other aspects of the public service and make recommendations. 31. The Government did not act as quickly as expected to complete the studies which were to provide the basis for improvements in central procurement practices. Nevertheless, a Directorate of Central Purchasing was established; it made practically no impact on Government procurement. 32. In the area of parastatal reform the Government took the following actions: it established the Public Industrial Enterprises Secretariat (PIES) in the Ministry of Industry and Technology to assist in restoring financial discipline and improving the management of the parastatals; it classified the parastatals into three groups: those to be retained, those to be divested and those to be closed; and it created the Public Enterprise Secretariat (PES) under the Ministry of Finance to spearhead the process of parastatal divestiture. The PES initiated a Divestiture Design Study, diagnostic studies of BOU, UCB, the Co-operative Bank and a number of other public enterprises, as well as a Public Sector Administrative Reform and Planning Study. A comprehensive restructuring plan was to be drawn up for the Uganda  - 10- Development Corporation (UDC) and approved by the Corporation's Board. This did nc t happen because the factual and analytical basis for such a plan did not exist. 33. In June 1988 the Government rolled over the Rehabilitation and Development Plan (RDP) as expected, covering the period 1988/89-1991/92. The revised plan was endorsed by IDA. However, the RDP continued to suffer from a number of weaknesses. First, the database for the preparation of the plan was very deficient. Secondly, the plan was essentially a compilation of externally-financed projects. Thirdly, economic criteria played little role in project selection. Fourthly, the link between the RDP and the development budget was at best tenuous. Evaluaffon of Performance 34. The conditions relating to the release of the second tranche were only five. However, the policy matrix contained approximately thirty other measures which the Government, through the letter of development policy, had undertaken to implement. The type and number of conditions attached to this operation reflected IDA's perception of not only the problems facing the country but also the Government's capacity to implement the required actions. Given the fact that IDA had been inactive in Uganda for a number of years and in view of the civil war and the generally chaotic conditions that had prevailed in Uganda, IDA staff's knowledge of the state of the economy and implementation capacity was bound to be patchy in many respects. Looked at in this light, the loose wording of some of the conditions ( e.g. "substantial progress" or "initiate") was justified. In this context it should be noted that ERC I was prepared in response to an urgent request from the Government. IDA was anxious to assist Uganda but this meant that the time available for effective dialogue with the Government was limited. At the same time, such dialogue was necessary, since a big and influential section within the Government still favored a controlled economy. Worse still, the capacity to analyze various policy options and propose appropriate measures was woefully lacking in Government. These two considerations raised questions, right from the outset, about the extent to which the Government had internalized the poiicies and measures to which it was committing itself 35. The ERC I conditions relating to economic stabilization were, as is customary with macroeconomic adjustment operations, conditions agreed between the Government and the Fund in the context of the SAF arrargements. Both the Bank and the Fund recognized that stabilization was an essential prerequisite to sustained economic growth. For this reason, the fiscal benchmarks under the 1987/88 SAF were quite detailed. There were, for example, quarterly benchmarks on recurrent expenditure and tax revenue. Such degree of specificity was, however, inappropriate in a situation where information systems had collapsed and much of the required data was lacking. A great deal of effort was thus put into establishing benchmarks which were practically unmonitorable. 36. The participation of a large number of Government agencies (BOU, Ministry of Co- operatives and. Marketing, Ministry of Commerce, Ministry of Industry, Ministry of Public Service and Cabinet Affairs e.t.c.) was required for the fulfillment of the conditions attached to ERC L There was, however, no mechanism for co-ordinating the efforts of these agencies. The responsibility for implementing and monitoring the various actions therefore rested  ahncst entirely with MOF. Given its own serious internal weaknesses, MOF was simply not capable of performing this role. 37. Performance of Government. The Government's performance in implementing ERC I was generally disappointing, particularly in 1987/88. Prior to Board approval of this operation the Government had taken a number of strong measures up-front. (See the May 15 1987 measures summarized in paragraph 8). This was, naturally, taken by IDA and the Fund to be a good demonstration of the Government's commitment to the ERP. However, the Government's actions soon called into question its resolve to implement the economic reforms. The first sign of this was the announcement of a budget which deviated substantially from the SAF targets. In addition, the Government reduced interest rates when it should have raised them. The bottom line was that the Goverment made practically no effort to meet the SAP benchmarks during the first year of ERC I. The stabilization objective was clearly not taken seriously by the Government. For, large sections of the Government clung to the view that there was no correlation between high and rising inflation and excessive credit and monetary expansion, fuelled mainly by the printing of money to finance the budget deficit. Besides pursuing expansionary fiscal policies, the Govemment maintained an overvalued exchange rate (USh6O per US dollar) throughout the fist year of the program. In this regard it should be noted that the SAF was rather vague on exchange rate policy; it only called for a mid-year review. For its part, the Govermment showed little or no understanding of the role that the exchange rate could play in improving Uganda's international competitiveness and bolstering fiscal revenues. With regard to petroleum prices and crop producer prices there was, during the first year of ERC I, a greater willingness to take the required actions. Also the OGL was made operational in January 1988. 38. The Government's performance started to improve in some respects during the second year of ERC I. A 40 percent currency devaluation in July 1988 was followed by a further 25 percent devaluation in two steps (in December 1988 and March 1989). But the Government again froze the exchange rate until October 1989. Other positive developments were increases in interest rates (although they remained substantially negative in real terms); the introduction of the SIP; the dismantling of most price and distnbution controls; the introduction of the 100 percent export retention scheme and its extension to all exports except coffee; and the establishment of the Public Service Review and Reorganization Commission (PSRRC). On the negative side: the Government continued to have problems in controlling expenditure, especially defence expenditure; introduced a credit arrangement for coffee marketing which was highly inflationary; refused to grant coffee farmers further price increases; merely announced that co-operative unions can sell lint directly to the textile mills; failed to review the OGL on a quarterly basis, let alone expand it; made no meaningful effort to address the weaknesses in tax administration; allowed the issue of the disposal of the Custodian Board properties to drag on; and made little progress on public enterprise reform. 39. Poor budget formulation and execution were among the most persistent weaknesses exhibited by the Government. For three years in a row (1987/88-1989/90) the original revenue and expenditure estimates had to be re-worked so that they would more or less conform with the understandings reached with IDA and the Fund in the context of the PFP and the SAF/ESAF. The problem could have been avoided or minimized if the Government had consulted closely with the Fund and IDA prior to finalizing the draft budget estimates. The subsequent revisions were very damaging to the credibility of the budget process. Budget  - 12 - imnplementation also left a lot to be desired. Expenditure control did improve as time went by but the manner in which that control was exercised created its own problems. In particular, defence expenditure was allowed to grow enormously while the other sectors were being severely squeezed. The Government was rather slow in implementing the conditions laid down in Schedule 3 to the Development Credit Agreement for the release of the second tranche. Actually, at the time of review of the tranche release conditions the Government seemed to have been surprised by IDA's insistence on the production of evidence of fulfillment of each of the conditions. In the event the Government just managed to comply with. the second tranche conditions. 40. Performance of IDA. In designing ERC I there was a temptation on the part of IDA to address as many of the macroeconomic and structural problems affecting the country as possible; hence the wide-ranging measures included in 4he letter of development policy and the policy matrix. This broad approach carried the risk of dissipating the Government's energies over a wide front. Of course, IDA and the Fund were well aware that the most pressing need was for stabilization and the program correctly placed great emphasis on this. -lowever, neither IDA nor the Fund was able, from the outset, to convince Government of the need to accord the highest priority to the restoration of macroeconomic stability. Prerequisites for macroeconomic stability included expenditure discipline, stepped-up domestic revenue mobilization and improved export performance. These things did not happen. The result was that the Government practically abandoned the stabilization components of the ERP in 1987/88 and 1988/89. To their credit, rather than turning their backs on Uganda, IDA and the Fund intensified their dialogue with the Government. It took a great deal of perseverance, over a two-year period, to get the Government to perceive the benefits of successful stabilization and to focus on this aspect of the program. The dialogue was conducted mainly in the course of preparing a follow-on operation, ERC II. The preparation of ERC II started as early as September 1988. 41. IDA did not fully appreciate the extent to which key economic institutions like MOF, MPED, MPSCA and BOU had been weakened during the past decade and a half of economic and political turmoil. As a result it tended to overestimate the Government's capacity to implement policy and institutional measures. The problem was compounded by the paucity of accurate and timely data needed to formulate policies appropriate to Uganda's circumstances and to monitor the implementation of those policies. Neither IDA nor the Fund focused on the data and related problems early enough. As a result the formulation of the macroeconomic framework as well as the monitoring of the program was plagued by severe data problems. 42. IDA was very responsive to Uganda's external financing requirements. It took a relatively short period (less than twelve months) for ERC I to be to processed and approved. IDA took a gamble by going ahead with the credit soon after the Government announced a budget which deviated substantially from the SAF targets. During the life of ERC I Uganda's external financing requirements increased mainly because of the sharp fall in coffee prices following the collapse of the quota system of the Intemational Coffee Agreement in July 1989. Moreover, during the first two years there were significant slippages in program implementation, particularly the measures required to restore macroeconomic stability and  - 13 - bring down inflation. These slippages necessitated a delay in the processing of the next adjustment operation; that is, ERC II. To bridge the resulting funding gap IDA provided additional financing in the form of (a) two IDA reflows totalling SDR 2.5 million, (b) a supplemental of SDR 19.0 million to ERC I and (c) a supplemental of SDR 10.0 million to the African Facility credit. These supplemental credits played a key role in sustaining the policy dia!ogue, maintaining an adequate flow of imports into the country and preventing the excessive depreciation of the Shilling on the parallel foreign exchange market. IDA also convened three meetings of the Consultative Group and intensified its efforts to mobilize additional resources for Uganda within the framework of the SPA. Other Aspects of ERC I 43. Procurement. The use of ERC I funds was limited to eligible imports on the basis of a negative list. The bulk of the funds was used to procure imports under the OGL and SIP following normal commercial practice. The full amount (SDR 14 million) allocated for imports of petroleum products was used. 44. Disbursements. The IDA and Africa Facility credits (inclusive of the supplemental credits), totalling SDR 101.2 million, were fully disbursed. The original closing date of September 30, 1989 had to be extended to June 30, 1991 in order to allow for the disbursement of the supplements and co-financing to the credits. The original amounts were disbursed in two tranches. The first tranche of SDR 38.7 million was disbursed upon credit effectiveness which occurred within one month of Board approval. The release of the second tranche of SDR 31 million was delayed by about two months for two reasons. Firstly, IDA sought clarification and more evidence regarding the fulfillment of the tranche conditions. Secondly, IDA wanted to be certain that the 1988/89 budget incorporated measures which would bring the ERP back on tracl. 45. Utilization. BOU was the Government's agency responsible for the allocation of ERC I funds to various users. Disbursements were paid into two special accounts (one for the IDA credit and the other for the AF credit) which BOU opened abroad. The utilization of the funds was plagued by a number of problems. First, the decision to restrict the OGL to a handful of firms meant continued reliance upon cumbersome and often arbitrary administrative procedures for the allocation of the bulk of the funds. This problem was solved by introducing the SIP in December 1988. Secondly, the minimum amount (US$200,000) eligible for processing under the special commitment procedure proved to be too large- in Uganda's circumstances; the figure was therefore reduced to US$75,000. Thirdly, a number of companies were unable to raise the local cover for the foreign exchange allocated to them. These were companies which could neither generate the necessary funds internally nor borrow from the commercial banks. Fourthly, bottlenecks in the coDlection of documentation needed to support requests for replenishment resulted in delays in replenishing the special accounts. This problem was obviated to some extent by increasing, in March 1988, the initial deposit under the IDA credit. Fifthly, there were delays in committing some of the funds promised by co-financiers. Finally, credit utilization suffered from a general lack of familiarity with IDA procurement and disbursement procedures. 46. Accounts and Audit. BOU did not have the capacity to maintain adequate records and accounts in respect of the expenditures financed out of the proceeds of the credits. This  - 14 - was not peculiar to these credits; the accounting function had collapsed completely throughout BOU. Only one set of accounts and audit was produced by the Government in November 1989; IDA had insisted that the requisite set of accounts be produced before ERC II was presented to the Board. Co-financigg of ERP 47. ERC I played an important role in mobilizing additional balance of payments support for Uganda. Nine donors co-financed the program. These were AfDB, SIDA, Switzerland, Saudi Arabia, CIDA, Norway, DANIDA, ODA (UK) and Finland. Together these donors provided approximately US$120 million of quick-disbursing funds for the program. With the exception of the AfDB, IDA administered the funds provided by these donors. The program was also supported by the Federal Republic of Germany, Japan, Austria, the EEC and the USA. Debt service arrears to some of the donors resulted in delays in the commitment of funds or disrupted disbursements. The Results of the ERP 48. The ERP showed mixed results during the first two and half years; i.e. during the life of ERC I. On the one hand, much progress was made in rehabilitating the infrastructure, particularly the trunk roads; there was a rapid revival of economic activity; and important steps were taken towards creating the appropriate climate increased production and development. On the other hand, the Government only began to make real progress towards restoring fiscal and monetary stability during the second half of 1989. 49. Real GDP growth averaged 6.7 percent a year during 1987-1989, compared to a target of 5 percent. As a result per capita GDP grew by nearly 4 percent a year. These high rates of growth were made possible by the following factors: the return of peace and security to most parts of the country; significant increases in crop producer prices at the beginning of the program; the rehabilitation of the major trunk roads, public utilities and other infrastructure; the elimination of price, distribution and other controls; and the availability of import support at a time of sharp declines in export earnings. ERC I contributed to the creation of conditions which facilitated the strong recovery of output. 50. During the life of ERC I the Government made very slow progress in stabilizing economic conditions and bringing down the rate inflation. The rate of inflation was programmed, optimistically, to fall to 30 percent by the end of the first year of the ERP. As it turned out the year-on-year rate did not drop below 100 percent until June 1989; the 30 percent target was finally achieved in June 1990. The co-existence of high rates of GDP growth and severe macroeconomic imbalances appeared for a while to give the Government the false impression that the imbalances did not matter. 51. The results of the ERP in respect of the external accounts were also disappointing. High deliveries of coffee to the CMB in 1987/88 and 1988/89, accompanied by a rapid build- up of stocks, and the attendant crop rmance problems tempted the Government into denying the farmers further price increases. The delayed effects of this policy began to be felt in lower deliveries of coffee in 1989/90. On top of this, international coffee prices collapsed in July 1989. The net result was a drastic decline in coffee export earnings (from US$286  - 15 - million in 1987/88 to US$167 million in 1989/90) and a growing dependence on external assistance. Taken together with the fact that other exports did not recover as expected, this meant that Uganda's balance of payments situation became more prewarious than ever before. D. The Main Lessons Len 52. The foregoing review has highlighted a number of lessons which are important for future adjustment operations. These lessons touch on the following areas: Government commitment; program design; disbursements; institutional capacity; co-financing; and results. 53. The experience of the first two years of the ERP showed how difficult it was to gauge Government commitment to the program. IDA and the Fund had beein successful in convincing tlre Government to take several important measures up-front. After taking the initial actions the Government's commitment slackened appreciably. One lesson to be drawn from this is that the dialogue with the Government must, from the outset, emphasize the fact that reform is a continuing process. Another lesson is that the dialogue itself must go beyond the narrow circle of top politicians and officials to involve a broader spectrum of decision makers and implementors. For, the greater the number of people who understand and support the policies the greater the prospects of sustaining the. reforms. Moreover, some people may oppose the Government's economic policies out of a pique at being excluded from the discussions leading up to the adoption of those policies. 54. With regard to desigu, ERC I highlighted the need t6: give Government officials every opportunity to participate in the design of the program so that sooner or later the Government can "own" the program; focus on a few actions which are critical to the restoration of macroeconomic stability; pay more attention to implementation capacity, especially in the sequencing of structural measures; set performance targets which are not wholly unrealistic (e.g. the first-year inflation target of 30 percent). 55. The disbursement of ERC I fundi taught us three lessons. First, much more effort is needed to make the Government ager responsible for the allocation of funds fully 4.'rnversant with IDA procurement anu disbursement procedures. Secondly, tied aid is, by its ndture, not quick-disbursing; mixing this type of aid with untied aid leads to the overestimation of the amount of import support that can be utilized in a given period. Thirdly, donors who ask IDA to administer their funds must accept that the pace of disbursement of those funds would be determined by IDA procedures and BOU's ability to comply with the required procedures. While still demanding accountability for the use of their funds, some donors seemed to have lost sight of this point and became rather frustrated at th. slow pace of utilization of the funds.  - 16 - 56. In the area of institutional capacity, one lesson of ERC I is that it is easy to overestimate the Government's capacity to implement and monitor the required actions. In reality, the program depended, for its success, on just a handful of grossly-overworked officials. Moreover, the elementary information systems necessary for sound policy forniulation and monitoring did not exist. The second lesson is that rebuilding institutional capacity takes a long time; the key actions in this area must therefore commence very early. A third lesson is that while studies may weil be needed to map out strategies and develop action plans for institutional reform, such studies can become an excuse for inaction. 57. As far as co-financing is concerned, the lesson to be drawn from ERC I is that, aside from the Government's willingness to stay the course in implementing economic reforms, factors such as respect for human rights and moves towards democratization can help to unlock additional donor resources. Given Uganda's long history of brutality and political chaos, the issue of governance could not be divorced from economic reform. 58. Perhaps, the most important lesson of ERC I is that results, in the form of an early supply response, hold the key to the sustainability of structural adjustment programs. The quick rebound of economic activity was a major selling point for the reforms. The Government was thus in a stronger position to persevere with the fight against inflation and implement other reforms.  - 17 - PART H. PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE A. Baclonnd. Evolution and Development of the Program 59. After the change of Government in January 1986, the NRM Government came up with a program of objectives, policies and actions designed to achieve overall recovery of the country's economy (dubbed the Economy Recovery Program). The Government approached a number of possible donors, and in this connection wrote a formal letter dated 1 July 1987 to the International Development Association (IDA) requesting it to help in financing urgently needed imports for the program by way of an import support credit. 60. The Association agreed to assist in financing the program. The Association had extended program assistance to the country before by way of three reconstruction credits, but this was the first recovery credit to the country and was a successor program to the 3 earlier programs. 61. Negotiations between the Government and the Association for the credit took place in Washington in mid-July 1987. The credit was approved by IDA's Board in mid-Septenhe:r 1987 for an amount initially of SDR 50.9 million. An African Facility Credit of SDR i8.8 million was also approved at the same time. The original closing date was September 30, 1989. 62. The IDA credit was later enhanced by supplements/reflows totalling SDR 21.5 million to reach SDR 72.4 million. The African Facility Credit was also enhanced by a supplement of SDR 10 million to SDR 28.8 million. 63. Under the agreements the Government undertook to institute, among other things, the following five major policy reforms and actions: - To prepare and adopt a public investment program for 1988/89 through 1990,91; - to revise the May 1987 producer prices for export crops; - to adopt an Open General Licensing system to give priority industries access to foreign exchange for eligible imports; - to restructure the public enterprise sector including retention, liquidation, complete or partial privatization of some public enterprises, and - to complete the verification of ownership of a significant number of the properties falling under the Departed Asians' Property Custodian Board. B. Implementatlon of the Program 64. Despite the post-war nature of the period in which the program was initiated and despite the continuing instability in some pasts of the country, implementation was quite  - 18 - successful. The program, however, was rather slow in the initial stages and its closing date had to be successively extended from 30th September 1989 to 30th June 1991. 65. Immediately following the signing of the credits, a Foreign Exchange and Import LAcensinj Committee (FEIL) composed of persons drawn from various Ministries and the Bank of Uganda was set up to allocate the funds; the special accounts for the IDA and African Facility credits were opened and applications for both the authorized allocation or initial deposit and retroactive financing were lodged with IDA. However, authorized signatures took time to send to IDA and this was not done until 5th November 1987. The legal opinion of the Attorney-General which takes a special format and can only be provided after signature of the agreements (yet it is a condition precedent to effectiveness) also took time to send to IDA. Getting administrative agreements for the co-financing signed, provision of legal opinions and specimen signatures thereon also took time owing to lack of close coordination between various agencies at the early stages of the program. The appointment of an overall ERC Co-ordinator in 1989 helped to make things easier subsequently. 66. The credits became effective in October 1987 and both retroactive funds and the initial advances were released in October and November respectively. 67. Initially allocation of credit funds to eligible importers was done administratively, except for OGL companies. Lack of local cover hampered or slowed down the pick-up of the funds; this in turn greatly slowed down implementation of the program. Thus, although the allocation of foreign exchange started as early as 18th November 1987, the first Letter of Credit (L/C) was not opened until January 1988. In some cases the time span between allocation and pick up ran into 6-12 months. The ever-depreciating value of the shilling also compounded matters with respect to the amount of local cover that had to be put up as time went by. 68. Government found a partial remedy by scrapping administrative allocation and instituting a new system on 5th December 1988 whereby any importer with local cover automatically got foreign exchange for commodities on a positive list. This innovative system greatly speeded up the taking up of funds. 69. The initial advance of US$10 million to the special account for the IDA credit was too low to support speedy implementation in view of the turnaround time of supporting documentation for replenishing the special account. IDA agreed to increase the initial advance to US$19 million in April 1988. The Bank of Uganda put up a temporary US$5 million collateral security with the depository bank to help finance confirmation of L/Cs in- between replenishments. All these actions facilitated smoother and speedier utilization of the funds. 70. A lesson to be drawn here is that the initial advance should be large enough to permit uninterrupted opening and ronfirmation of L/Cs. Disbursement would be faster if IDA could replenish the special accounts against evidence of firm commitment of funds i.e. against confirmed irrevocable L/Cs rather than shipping documents and payment advice. The African Development Fund (AfDF), for example, can and does replenish against commitment i.e. on the basis of value of confirmed L/Cs. So does KfW.  - 19 - 71. With IDA as the lead-donor, the program was subsequently supported by AfDF, SIDA, Switzerland, Saudi Arabia, CIDA, Norway, DANIDA, UK-ODA and Finland: these together provided a total of about US$120 million, all in the form of import-support funds which, with the IDA contribution of approximately US$132 million, brought total funds committed under the operation to about US$252 million. 72. The contribution of the various donors is shown in the table below. FUND CURRENCY AMOUNT US$ IDA 1844 UG (0-4) SDR 72.40 95.00 IDA African Facility SDR 28.80 37.40 (0 & 1) SIDA (1 & 2) SEK 60.00 9.70 Swiss Grant SFR 10.00 7.00 Saudi Fund Grant SR 17.00 4.50 Norwegian Grant NOK 20.00 3.00 DANIDA Grant USD 6.00 6.00 UK Grant