Argentina - Agricultural Sector Loan Project

Doi The Worlk ... 1 ItS a I'FM OMCaAL ' IqewtNo. 9913 PROJECT COMPLETION REPORT ARGENTINA AGRICULTURAL SECTOR LOAN (LOAN 2675-AR) SEPTEMBER 26, 1991 Agriculture Operations Division Country Department IV Latin America and the Caribbean 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 UNIT - AUSTRAL (A) Rate at Appraisal: US$1 - A 1.0 Rate at Completion: US$1 - A 261.0 Glossarv of Abbreviations AGR Agriculture and Rural Development Department AGREP Agriculture and Rural Development Department, Economics and Policy Division BCRA Central Bank of Argentina BNA Banco de la Naci6n Argentina CEM Country Economic Memorandum CFI Federal Investment Council EC Executive Committee for Implementation of Federal Land Tax EEC European Economic Community FLT Federal Land Tax GOA Government of Argentina IICA Inter-American Institute for Agricultural Cooperation IMF International Monetary Fund INTA National Institute of Agricultural Technology JNC National Meat Board JNG National Grain Board HE Ministry of Economy OPS Operations Staff PPF Project Preparation Facility PRONAGRO National Agricultural Development Program SAGyP Secretariat of Agriculture, Livestock and Fisheries SOE Statement of Expenditure TCN National Court of Accounts UNDP United Nations Development Program ARGENTINE REPUBLIC FISCAL YEAR January 1 - December 31  FOR OMCAL USE ONMY THE WORLD BANK Washington, D.C. 20433 U.S.A. Ckg. Of D,tah.CARWA OpHatuns Evautm September 26, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on ARGENTINA - Agricultural Sector Loan (Loan 2675-AR) Attached, for information, is a copy of a report entitled "Project Completion Report on Argentina - Agricultural Sector Loan (Loan 2675-AR)" prepared by the Latin America and Caribbean Regional Office. 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 perfonnance of their official duties. Its contents may not otherwise be disclosed without World Bank auth,nzation.  FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT ARGENTINA AGRICULTURAL SECTOR LOAN (LOAN 2675-AR) TABLE OF CONTENTS Paze No. Preface . . . . . . . . . . . . . . * . . . . . . . . . . . Evaluation Summary . . . . . . . . . . . . .iii PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE I. BACKGROUND The Economy Prior to 1986 . . . . . . . . . . . . . .1 The Agricultural Sector. .. . . ....... . .. . 2 Constraints on Agricultural Growth . . . . . . .. . 3 Government Actions to Stimulate Agricultural Growth . 5 Agricultural Institutions and the Policy Relationship 5 Bank Group Operations in Argentina . . . . . . . . . . 6 II. IDENTIFICATION. PREPARATION AND APPRAISAL Introduction . . . . . . . . . . . . . . . . . . . . . . 7 Loan Preparation . . . . . . . . . . . . . . . . . . . 8 Loan Appraisal . . . . . . . . . . . . . . . . . . . . . 15 Objectives and Conditions of the Loan . . . . . . . . . . 20 Board Presentation . . . . . . . . . . . . . . . 22 III. IMPLEMENTATION Effectiveness and First Tranche Release . . . . . . . . . 22 Implementation Experience... 22 Release of the Second Tranche . . . . . . . . . . 25 Disbursement, Procurement and Audit . . . . . . . . 27 Compliance with Loan Conditions . . . . . . . . . . . . . 28 IV. ECONOMIC EVALUATION........ ...... 28 V. INSTITUTIONAL PERFORMANCE The World Bank . . . . . . . .. . . . . . . . . . . 28 The Borrower . . . . . . . . . . . . . . . . . . . . . . 30 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.  Page No. VI. CONCLUSIONS AND LESSONS Conclusions . . . . . . . . . . . . . . . . . . . . . 31 Loan Design . . . . . . . . . . . . . . . . . . . . . . 31 Lessons Learnt . . . . . . . . . . . . . . . . . . . . . 32 PART rI: PROJECT REVIEW FROH BORROWER'S PERSPECTIVE . . . . . 34 PART III: SUPPLEMENTARY INFORMATION AND STATISTICS Annex I Basic Data Sheet . . . . . . . . . . ... . . . ... 35 Annex II Compliance with Loan Conditions . . . . . . 37 Annex III Export Taxes and the Exchange Rate . . . . . 40 Annex IV Borrower's Statement Development Policy . . . . 50 Annex V Project-Related Economic Indicators . . . . . . 57 Map: IBRD 19235  PROJECT COMPLETION REPORT ARGENTINA AGRICUJLTURAL SECTOR LOAN (LOAN 2675-AR) PREFACE This Project Completion Report (PCR) reviews the Argentina Agricultural Sector Loan (2675-AR) for US$350 million. The loan was approved by the Board of Directors on April 3, 1986 and signed on June 5 of that year. The original Closing Date of June 30, 1988 was extended to June 30, 1989 to accommodate delays in completion of the technical basis for the land tax. Final disbursement was made on June 23, 1989. US$0.04 million was cancelled. The PCR was prepared by the Agriculture Operations Division, Department IV of the Latin America and the Caribbean Region. It is based on the President's Report, Loan Agreement, Supervision reports, correspondence between the Bank and Borrower, consultant studies, internal Bank studies, memoranda and interviews. The Borrower did not respond to the Bank's requests for an independent assessment of the operation. The Bank's draft PCR, Parts I and III, was sent to the Borrower for comments on March 27, 1991, but no response has been received.  - iii - PROJECT COMPLETION REPORT ARGENTINA AGRICULTURAL SECTOR LOAN (LOAN 2675-AR) EVAlIUATION SUMMARY Backfround and Objectives 1. The Economy. Argentina's decline from one of the richest countries in the world to one beset by seemingly intractable economic problems can be attributed t' a fundamental reorientation of the economy following the economic crisis provokeG by the Great Depression of the 1930s. For the next four decades the country's development strategy turned inwards, promoting rapid industrializa- tion financed largely by transfers from agriculture, the sector upon which early prosperity had been built. However, the exhaustion of import-substitution options led to increasingly frequent balance of payments crises and erratic economic growth. 2. Major efforts were undertaken in 1976 and again in 1978 to reverse the trend, but persistent inflation and failure to contain public expenditures undermined these programs, leading to extreme instability which, in turn, concentrated the efforts of decision-makers on short-terk, speculative operations which were unsustainable and perceived as such. A democratically-elected Government took office in late 1983, inheriting spiralling inflation and balance of payments difficulties stemming from unchecked growth of the fiscal deficit, mioguided exchange rate policies and excessive external borrowing. 3. In late 1984, the Government entered into a standby arrangement with the International Monetary Fund (IMF), but a deepening recession weakened its adherence to program targets and draw-down of the first tranche was suspended. Further deterioration led to the Austral Plan of June, 1985, a comprehensive adjustment program supported by a revised and more stringent IMF standby arrangement. Political and social support for the Plan was widespread, and the Bank believed the Government's initial success in stabilizing the economy augured well for improved economic performance. 4. Agriculture. Agriculture bore the brunt of the import-substitution strategy pursued by successive governments. Agricultural exporters variously faced export taxes, a lower effective exchange rate than other exporters, or inflated prices for their inputs. This pattern of discrimination against agriculture penalized the country in foregone production and export earnings, and more frequent balance of payments crises. In particular, the Government's  - iv - interchangeable use of export taxes and an over-valued currency to extract revenues from the sector has been a powerful disincentive to productive investment. 5. Influences on the Loan. Several important factors influenced the nature and processing of the loan. First, the promise inherent in fast-disbursing loans coupled with policy change and the fact that several such loans were already underway or peniing in other countries in tLa Region created considerable interest in extending such lending to Argentina. Second, the decision to build a project lending pipeline in Argentina was subsequently reinforced by the explicit requirement under the Baker Plan for a major Bank role in assisting indebted countries to develop their way out of debt. Lastly, little recent agricultural sector work existed on which to construct an informed dialogue and thus the Country Economic Memorandum (CEM) of 1984 became the principal blueprint, largely pre-selecting the policy issues and loan concept. 6. The Loan Concept and Preparation. The CEM argued that the entire Argentine farming system would be transformed by replacing agricultural export taxes and import tariffs on inputs by other revenue measures, preferably a federal land tax (FLT). The latter would be developmentally sound, would be easy to collect, and would transfer the tax burden from exporters to the domestic producers of high-value products. The CEM did not discuss the complementarity between export taxes and the exchange rate in discriminating against agriculture or the central significance of a stable exchange rate as an agricultural incentive. 7. Loan preparation took almost two years and was characterized by persistent debate over the loan's feasibility and timeliness, an inadequate analytical base and an apparent inability to determine the critical path for achieving loan objectives. Resource transfer was a major preoccupation which undermined the structural objectives and reduced the seriousness of the dialogue with the Government. It is also evident that the loan's structural requirements lacked a rational linkage to the IMF's stabilizc.tion efforts, and implicitly involved the Government in trying to fulfill conflicting demands. 8. Despite the complexity of loan objectives and the need for unequivocal government support, the Bank's substantive dialogue was limited to the Secretariat of Agriculture, Livestock and Fisheries (SAGyP). The Ministry of Economy (ME) and its Secretariat of Finance had little interest in structural reform. While not opposing a federal land tax in principle, they were strongly averse to relinquishing the secure source of income represented by export taxes, and were thus not committed to the loan concept and objectives. The Bank gave principal responsibility for the planning and implementation of the reform program to SAGyP despite that agency's lack of control over the policy instruments and resources most affecting the sector. By contrast, ME's formal role was largely peripheral despite it being the agency most affected by the proposed changes. 9. Loan Objectives. The Loan was des;igned to increase agricultural production and exports through a structural change in federal taxation away from  export taxes (retenciones), which discourage the production of exportable crops, to a production-neutral federal land tax which would improve incentives and lead to a more intensive use of agricultural resources. The loan is described in the President's Report as supporting the following policy reforms and complementary activities under the Government's economic stabilization program: (a) a substantial reduction in agricultural export taxes; (b) fiscal measures to maintain the fiscal deficit at satisfactory levels, including the introduction of a FLT; (c) modified regulations and tariffs pertaining to imported inputs; and (d) complementary studies and institutional support. 10. Loan Conditions. The tranche release conditions were as follows: (a) First Tranche. The first tranche of US$170.5 million was to be released upon loan effectiveness (and before June 30, 1986) based on: (i) establishment of an Executive Committee; (ii) re-designation of tractors and agricultural machinery to the "prior approval" list for imported goods; and (iii) Government's execution of subsidiary agreements with the National Grain Board (JNG) and National Meat Board (JNC) for complementary studies and institutional support activities. Disbursement under the loan for studies and institutional support was contingent upon presentation to the Bank of acceptable terms of reference. (b) Second Tranche. Disbursement of the second tranche of US$170.5 million (forecast for January, 1987) required: (i) reduction of agricultural export taxes to a level not to exceed 70Z of the rates in force on May 1, 1985 or other levels to be jointly agreed; (ii) implementation of fiscal measures sufficient to compensate for revenue losses from the export tax reduction, without increasing the overall sector tax burden compared to its May 1, 1985 level; (iii) designation of tractors and agricultural machinery to the "automatic entry" classification for imported goods, and removal of the 102 import surcharge on agricultural chemicals; and (iv) satisfactory progress in the execution of the agreed land tax implementation schedule, and in all stud'es financed under the loan. Implementation Experience 11. The first tranche was released on September 24, 1986, five and a half months after Board presentation due to difficulties in complying with a number of conditions of effectiveness. 12. Loan implementation was hampered by political and macroeconomic impediments, some of which might have been predicted. The following summarizes the main aspects: (a) The Federal Land Tax. Legislation to establish a FLT remained stalled in the Congressional committees to which it was first referred for debate. Evidence suggests that a progressive breakdown of the Plan Austral and deteriorating conditions in the rural sector  - vi - rapidly eroded the Executive's willingness to pursue the project and provided the opportunity for earlier opposition to re-surface. Farmer groups argued that reductions in export tax rates at that time could not compensate them for sharply lower world commodity prices or in any way be accepted as a quid pro qur for future introduction of a land tax. While the Bank believed it had satisfactory legal assurances from the GOA that export tax rates would be reduced permanently, farmers ncted that the prerogative of ME to manipulate rates remained intact, and reached the opposite conclusion. The FLT was perceived by farmers as additional to expc-t taxes, not as a replacement. (b) Export Taxes. First-stage reductions of export taxes on major agricultural and agro-industrial products by mid-1986 were marginally below agreed levels, reflecting the Bank's considerable effort to secure real movement on the issue of rates. Subsequent reductions by 'ate 1986 to zero (in the case of grains) and to low levels (in the case of oilseeds) far exceeded loan requirements, but were neces- sitated by the fall in international commodity prices to historical lows, in real terms. These latter reductions were consistent with historical patterns of export tax implementation and had little to do with the loan. Indeed, after the Spring Plan of August, 1988, the Government abandoned the IMP's exchange rate regimen, adopting a multi-tiered rate designed specifically to discriminate against agriculture. In May, 1989 (i.e., before loan closing), with the adoption of a unified exchange rate, export taxes were reinstated at high levels. (c) Studies. The five studies and two programs oZ institutional support were completed successfully. The special taxation study, which combined preparation of the technical basis for the FLT with a study to determine the appropriate level of taxation for agricultural activities and possible alternative measures to the land tax, was also completed but with delays. The closing date of the loan was extended for one year to permit its completion. These high-quality studies, elements of which have won several international prizes for technical excellence, represent the single success story under this loan. (d) Import Liberalization. Tractors and agricultural machinery were transferred to the "automatic entry" list for imports after considerable delays related to depressed conditions in the domestic industry. However, deficit reduction requiremenits of a new IMP arrangement in early 1988 meant that the Government was unable to remove the import surcharge on agricultural chemicals, which instead was raised. 13. Second Tranche Release. The second tranche was released on June 22, 1988 following protracted and unusually contentious debate during which fundamental aspects of the loan's design and feasibility were once again  - VIA - questionekL. Acknowledging tne impossibility of imposing an FLT, the Bank agreed to accept flternative revenues proposed by the GOA as compensatlon for export tax reductions, and insisted or compliance with the condition affecting imports of tractors and machinery, which was seen as a key indication of the Government's seriousness in pursuing a much broader program of trade liberalization then under discussion. The Board of Directors approved a decision to waive two loan conditions (i.e., FLT implementation and removal of the import surcharge on egricultutral chemicals). Economic Impact 14. The loan provided a large, infusion of foreign exchange but achieved no structural reform on the central issue of agricultural taxation. The unpredictable mixture of over-valued c, rrency, multiple exchange rates and export taxes persisted following the Spring Plan of August 1988 and continued to affect producers' confidence and their ability to plan and make investment decisions. The production response during the perio of low export tax rates and high real value of the currency (early 1986 to August 1988) was therefore low; producers clearly and correctly interpreted these moves as mere stabilization measures and not as the first stage of any long-term effort to improve agricultural incentives. Sustainabilitv 15. The preparation phase of the loan did not establish the conditions necessary tor sustainability. First, Government support for the loan concept and objectives was weak and the Bank's efforts to construct the essential dialogue were not sufficiently forceful or comprehensive. Second, loan design lacked a strategy for locking in export tax reductions and removing their discretionary use, leaving ME free to manipulate rates and causing producers to perceive and resist the FLT as an additional measure. Lastly, pending elections and continued macroeconomic instability subverted official interest in or willingness to pursue structural reform. Conclusions and Lessons 16. Decisions made during both preparation and implementation of the loan were too often motivated by a need to transfer resources, which subordinated structural reform objectives. The reform package was not well designed and it was superimposed on a difficult political and macroeconomic situation, pre- disposing the loan to failure. The principal lessons relating to this specific loan and its design are as follows: (a) The fiscal emphasis of the Bank's approach to reducing export taxes was too narrow, e.g., the industrial protection implications reflected in differential tax rates between the raw and processed product, which, in the case of oilseeds and their processed products, represents a transfer from farmers to processors of around US$200 million per year, did not receive the close attention they merited;  - viii - (b) Export tax rate reductions should have been locked in and their discretionary use controlled through legislative and/or otl,r means, to prevent the oscillation which has been, and continues to be, so powerful a disincentive to producers. Such action would have greatly improved the chances for legislative passage of the FLT; (c) The anticipated disburseamnt of the loan (as presented in the President's Report) was overly optimistic and incompatible with a realistic schedule for achieving its main conditions, making problems with second tranche release inevitable; (d) Concentration on one fiscal element in the absence of a coherent and stable macroeconomic environment, especially with respect to exchange rate policy, greatly reduced the likelihood of achieving loan objectives; (e) Passage of the land tax legislation should have been a condition of negotiations. The extreme riskiness of legislation, its central significance to the reforms sought, and the size of the loan, made requirement for its passage up-front an entirely reasonable condition; (f) In the absence of more significant achievements on the policy reforms than were actually accomplished, a more appropriate course of action might h.-i been to have cancelled the loan; (g) The institutional scheme was unrealistic. The primary impetus and responsibility for planning and implementing the reform program should have come from ME, not SAGyP, since it was the agency most affected by the reforms sought; (h) Loan withdrawal responsibilities and procedures should have been established at negotiations with full BCRA involvement, understanding and agreement; 17. In addition, a number of generic issues should be highlighted: (a) Sector adjustment must be preceded by and must fit into a broad adjustment program which realistic evaluation shows is working Close coordination of macreconomic and sectoral lending is essential; (b) Adjustment lending requires the unequivocal support of Government, leading agencies and important personalities. The Bank needs to address dialogue development in a more sophisticated manner if it continues to design politically-sensitive operations, and should not be afraid to halt loan processing if =mmunication with Government is clearly inadequate;  -ix (c) Greater "front-loading" of conditionality is needed to achieve as much as possible early in the loan and mitigate the tendency to give away the second tranche when conditionality is not met; (d) Legislation is an exceptionally risky form of conditionality for a fast-disbursing loan, unless it is achieved as a condition of negotiations. Government is generally reluctant to support measures that are difficult to reverse; the vagaries of the political process cannot reasonably be tied to a schedule; and, since the Bank is precluded from intervening directly in the political process, it becomes largely an observer; (e) Sustaining policy reform is a fundamental and well-recognized weakness of adjustment lending. The imperative for the Bank, if it is to continue to make these loans, is to develop a broad strategy, including appropriate incentives and mechanisms, for establishing commitment and sustainability. Legal covenants ace an ineffective substitute for this process; (f) Clearly-defined objectives are the sine qua non of adjustment lending. In retrospect, it is evident that this particular operation suffered from a lack of clear definition between its resource transfer and policy reform objectives; and (g) Agricultural Sector Loans (AGSALs) are not the only loans which have sought to reduce or eliminate export taxes. Many have encountered the inherent problem that export taxes are a comparitively simple way of raising revenue in countries where other forms of taxation are not effectively implemented. Thus, the lesson for the Bank is that export tax reform can only be part of a broad and workable reform of the overall tax system.  PROJECT COMPLETION REPORT ARGENTINA AGRICULTURAL SECTOR LOAN (LOAN 2675-AR) PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE I. Backaround The Econo2M Prior to 1986 1.01 Argentina was among the richest countries in the world in the early decades of the twentieth century, but by the 1980s its per capita income had fallen far below that of many countries with similar resource endowment. The reasons for this relative decline have their origin in the 1930s. The Great Depression severely curtailed exports of the traditional agricultural products which had been the foundation of Argentina's earlier development. The ensuing economic crisis provoked a reformulation of the country's development strategy which, for the next four decades, became in varying degrees inward-looking and protectionist, promoting industrialization financed largely by transfers from agriculture. However, as easy import-substitution options were exhausted, balance of payments crises became more frequent and economic growth more erratic. The political pendulum between nationalist populist governments emphasizing autonomous development and market-oriented reformists stressing an open economy and world view damaged the productive and financial structure. Increased public sector spending placed growing strains on fiscal stability, while price, credit and interest rate controls inhibited saving and investment and distorted the allocation of resources. 1.02 A marked change of direction occurred in 1976, beginning with an austere stabilization program accompanied by a relaxation of price, marketing and financial controls, major devaluations of the peso, sharp reductions in the fiscal d4eficit and improved export incentives. The initial results were encouraging. However, the failure to reduce inflation as expected provoked further major changes in macro-economic policy in 1978, involving exchange rate adjustments, import tariff reductions and almost complete abolition of remaining controls on credit, interest rates and foreign exchange transactions. Despite these measures, fiscal deterioration from growing expenditures for debt service, defense and domestic security, and increased public sector wages fed the inflationary pressures which the exchange rate and tar_.ff policies were intended to diminish. Thus, policies pursued to stabilize the economy and improve economic efficiency resulted in extreme instability which, in turn, concentrated the efforts of decision-makers on short-term, speculative operations. The policies adopted were generally unsustainable and were perceived as such. 1.03 Between 1979 and 1983 there were five changes of leadership in Argentina, the consequences of which were extreme economic and political  -2- uncertainty which, combined with the unstable international environment, led to the cessation of commercial bank lending. A democratically-elected Government took office in late 1983, inheriting spiralling inflation and balance of payments difficulties stemming from unchecked growth of the fiscal deficit, erratic and misguided exchange rate policies, and uncontrolled external borrowings by both the public and private sectors. 1.04 Towards the end of 1984, the Government entered into a 15 month standby arrangement with the IMF designed to lower the rate of inflation and achieve a balance of payments position enabling Argentina to meet its external obligations while initiating domestic recovery. The Government sought to comply with the conditions of the IMF arrangement despite a deepening recession, but its adherence to program targets weakened and, in early 1985, the IMF suspended the first conditional draw-down until completion of the 1985 first-quarter review of the program. A significant deterioration in econxmic conditions by the second quarter of 1985, with severe imbalances in the domestic and external sectors and very high inflation, forced the Government to adopt drastic measures designed to prepare the country for a major change in economic policy. 1.05 The Government instituted the Austral Plan in June 1985. This was a comprehensive adjustment program involving an immediate and substantial increase in fiscal revenues, large reductions in public expenditures, a temporary price freeze, and monetary reform designed to neutralize inflationary expectations. The program was supported by a revised and more stringent standby arrangement (covering 1985/86) with the IMF and commercial bank creditors. In addition to the Plan, the Government formulated a medium-term growth strategy for 1985-1989, based on an improved investment climate and the restoration of adequate export incentives. The overall program received wide political and social support despite its austerity, and itf, initial success in stabilizing the economy permitted draw-down of the first tranche under the IMF agreement. The Bank believed that the Plan augured well for improved economic performance if efforts to reduce the fiscal deficit could be sustained. The Azricultural Sector 1.06 Agriculture and agro-industry have been crucial in Argentina's economic development and have traditionally been relied upon to lead the country out of economic crises. Crop production grew rapidly from the early 1970s through the adoption of improved seed, chemicals and mechanization which raised yields and produced high and relatively risk-free returns to investment. Primary exports rose from about 70% of total exports in the late 1970s to above 802 in 1983/84. Argentina was, at What time, supplying about 5% of annual world exports of wheat and around 12% of maize and sorghum, ranking respectively fifth and second among the major world suppliers. 1.07 At the time of loan preparation, cereals (chiefly wheat, maize and sorghum) and oilseeds (soybeans and sunflower seed) were the main crop activities, accounting for almost 90% of the total crop area. Almost all the area under these crops was cultivated under rainfed conditions, with little use of fertilizer. The Bank believed that there was substantial scope for greater  -3- use of fertilizer and other inputs and for investment in infrastructure and equipment to enhance production, if prices could be improved. By 1985, the agricultural sector wac contributing about 15% of GDP, 17% of employment, 75% of foreign exchange earnings and between 14Z and 20% of the Federal Government's tax revenues, chiefly in the form of export taxes. Constraints on ARricultural Growth 1.08 Pricinx Policies. Agriculture bore the brunt of the import substitution strategy pursued by successive governments. Agricultural exporters variously received a lower effective exchange rate than other exporters, received prices seriously reduced by the imposition of export taxes, and paid inflated prices for their inputs. Despite technological advances, a well-developed input distribution system and a competitive marketing environment, agriculture was growing at a rate well below its potential with vast areas of arable land either lying idle or being utilized without modern inputs, penalizing the economy in foregone production and export earnings, more frequent balance of payments crises and depleted soils.'/ 1.09 Price discrimination focussed primarily on the traditional agricultural exports (cereals, oilseeds and beef), but its consequences affected the entire sector. The combination of direct and indirect nominal protection was strongly negative for the entire period 1960-1985: agriculture was discriminated against by between 40-50% (i.e., product prices would have been between 80-100% higher in the absence of these distortions). Over the long-term, direct measures of discrimination (i.e., export taxes) were higher when a competitive rate of exchange or high international prices prevailed (para 14).1' 1.10 The policies outlined above attempted to satisfy the following non- agricultural objectives: (a) maintaining low food prices in urban markets; (b) protecting domestic industry; and (c) raising fiscal revenues. At the time of loan preparation (1984/85), most agricultural exrorts were taxed at rates up to 25%. Export taxes acted as a price stabilization device which permitted higher real wages in the politically-active urban sector at the expense of rural producers. Intervention in the sector was conservative, aimed at stabilizing farm incomes rather than prices, and it served to neutralize long-run changes in productivity." Between 1981 and 1983, export taxes rose steadily; by 1983 they represented about 20% of Federal fiscal revenues, being the third most important source of revenue after value-added and fuel taxes. The tax reaped about US$1 billion per year in 1983 and 1984 and around US$1.3 billion in 1985 (some 1.8% 11 IBRD, Economic Memorandum on Argentina, Report No. 4979-AR, June 22, 1984. 2/ IBRD, Agricultural Sector Review, Report No. 7733-AR, Volume I, June 30, 1989. In purely fiscal terms, it is not easy to demonstrate that the agricultural sector pays more than other sectors. Evasion of direct taxes (wealth, income, etc.) is significant. The broader picture, however, shows a pattern of discrimination. 31 Adolfo C. Sturzenegger, A Comparative Study of the Political Economy of Agricultural Pricing Policies, Argentina, January, 1988.  -4- of GDP). Soybeans, maize and wheat accounted for over 90% of export tax revenues collected. Buenos Aires, Santa Fe and C6rdoba provinces contributed over 75% of those revenues. 1.11 Exiort Taxes -- Origins and Impact. Taxes on exports have a colonial pedigree based on a tributary or extractive relationship between Argentina and its principal collaborating powers and, in microcosm, Buenos Aires and the hinterland. Historically, they were levied not because the producer evaded other taxes (although this had become a powerful motivation for their continuance) but because they are legitimized in the Argentine Constitution and the producer has always paid them. In fact, export taxes as they are now administered derive from laws passed in the 1930s giving the Government broad, formal taxing authority. In contrast to other taxes, however, which require congressional action to modify, export taxes are the prerogative of HE acting through its Secretariat of Finance. Differentiated semantically as retenciones, they are used by the Secretariat of Finance to retain part of the country's foreign exchange earnings to cover current expenses. While their consumption subsidy function (para 12) was deeply-entrenched, increasingly in the 1980s arguments to retain them on price stabilization grounds have tended to mask their more dominant fiscal purpose. 1.12 In a country where tax evasion is common, export taxes have several desirable characteristics: (a) they are easy to collezt; (b) changes in export tax rates can be enacted by the Executive without congressional approval, providing flexibility in varying fiscal revenues in the short term; and (c) they are politically attractive in reducing consumer prices and providing a cost advantage to processors of agricultural raw materials. While such advantages do not compensate for the adverse effects export taxes have on agricultural production and exports, they are vital to an understanding of the GOA's tenacious attempt to keep them. 1.13 On the negative side, export taxes have been a serious impediment to agricultural growth because they have: (a) penalized productive investment; (b) created unfavorable input/output price relationships, discouraging the use of inputs such as fertilizers; (c) caused intra-sectoral distortions in resource allocation by applying different and changing rates of taxes to raw and processed agricultural products; and (d) caused resource allocation out of alignment with international market signals by preventing producer access to export parity prices. 1.14 Export Tax or Exchange Rate. Agricultural investment has been discouraged not only by the high overall level of export taxe3 but also by the price uncertainty associated with the Government's interchangeable use of export taxes and an over-valued currency. Export taxes have been lowered when the currency has been over-valued, and increased when devaluations have taken place (Annex III). For example, over the 1977-81 period, when the Argentine peso climbed almost 50% against the US dollar, export taxes on wheat were reduced from 482 to zero. From 1982-1984, when peso devaluations took place, the wheat export tax was raised in stages from 10% to 24%. Restoration of an efficient exchange rate has almost invariably been accompanied or followed by increased export tax  rates designed to confiscate "windfall profits", a punitive pattern which was re- established with renewed vigor even before Ln. 2675-AR was closed.4/ 1.15 Input Usafe. Throughout this period, Argentine farmers have also been discouraged from adopting some high-yield technologies due to their high cost relative to policy-depressed output prices. Until early 1984, import tariffs on agricultural inputs were substantial, ranging from 25% for urea and veterinary medicines to 38% for most plant protection chemicals and tractor components. Argentina was a striking example of a relatively advanced agricultural producer with a generally low level of agricultural chemical usage. Government Actions to Stimulate Agricultural Growth 1.16 Under the Austral Plan, the Government developed a four-pronged program to stimulate agricultural p.oduction and exports based on: (a) significant reductions in export taxes and their substitution by alternative and less damaging fiscal tools (then-recent export tax increases were to be a purely temporary component of short-term economic stabilization); (b) reducing tariffs on imported agricultural inputs (even though reducing the fiscal deficit under the Austral Plan required a temporary, across-the-board tariff increase on all imports); (c) phasing out all quantitative import restrictions; and (d) a commitment to maintaining export-competitive exchange rate adjustments. 1.17 Complementing the Government's program, SAGyP produced a medium-term National Agricultural Development Program (PRONAGRO), whose sectoral objectives were defined as increased production of cereals and oilseeds; expanded agricultural exports; developing regional economies, especially in the non-Pampa zone; generating employment; and encouraging natural resource conservation. Complementary actions were to include gradual replacement of export taxes with a federal land tax based on unimproved market value; eliminating import tariffs on agricultural inputs; making herbicide and fertilizer use more economical; and broadening lines of credit for private investment in the sector. Aaricultural Institutions and the Policy Relationship 1.18 SAGyP, the Secretariat (now Sub-Secretariat) of Agriculture, Livestock and Fisheries, is a dependency of ME. This has meant that policies affecting the sector have traditionally been based largely on ME (and its Secretariat of Finance) initiatives, with SAGyP acting as technical adviser rather than policy- maker and planner. Sectoral policies have, as a result, often been subordinated to short-term macroeconomic expediences authorized by ME. Within the public sector apparatus, therefore, SAGyP does not reflect the importance which the sector has in the economy, especially in relation to exports.5/ if IBRD, OD cit. Since international price fluctuations have an impact on the CPI and real wages, successive governments have sought to neutralize these changes through the use of exchange rate manipulations and export taxes. 5/ IBRD, Argentina: Agricultural Sector Review, Report No.7733-AR, Volume I, June 30, 1989.  - 6 - 1.19 Political instability has caused the rapid turn-over of Secretarie8 of Agriculture who have tended to be either technically-oriented and independent of sectoral pressures, trying to design policies to improve sectoral incentives, or representatives of the sectoral interests themselves acting as a channel for the agrarian lobby. The resulting inconsistency of agricultural policy-making has contrasted sharply with the constancy of export taxes. This more permanent form of intervention has been due to the presence of pressure groups which have limited the autonomy of government and placed strong restrictions on the instrumentation of economic policy.i' 1.20 This relationship had important implications for the preparation and subsequent implementation of the Agricultural Sector Loan. Its core concept, the reduction of export taxes and their substitution by a FLT, represented a profound restructuring of the tax regime initiated and advocated by SAGyP, not ME, and resulted from a long-standing and hitherto sterile debate between the Government and interest groups as to how best to tax agriculture. To SAGyP, export taxes were a malicious measure counteracting other efforts on behalf of agriculture, and needing eventual abolition as part of a long-term rationalization of agricultural inc ntives. Importantly, SAGyP did not view the land tax as a substitute for export taxes, but rather as a separate and beneficial measure which would bring order to the tax system, i.e., have a range of collateral benefits. It was supported in this effort by several influential farmer groups including CONINAGRO (the producer cooperative movement) and the Federaci6n Agraria (medium and small farmers), and the policy issue was part of the platform of the Radical Party of President Alfonsin. 1.21 However, in its quest for genuine sectoral reform, SAGyP's trajectory was very different from ministries whose -ocus was the economy at large, who saw agriculture as the comparatively successful sector, and who had little interest in, or commitment to, structural change. Evidence suggests that the Bank did not fully understand, or was unwilling to address, the implications of this fundamental conflict, which found SAGyP isolated in the Executive Branch and the Bank with only a partially-developed dialogue (mostly with SAGyP) on which to base major change. Bank Grouv Operations in Argentina 1.22 Bank lending to Argentina up to 1985 had been sporadic, the result of recurring macroeconomic and sectoral difficulties. Between 1979 and 1985, non- agricultural lending comprised ten loans totalling US$1.045 billion, focussed on 61 Adolfo C. Sturzenegger, o2 cit. Sturzenegger describes conflicting interests between the economic team of the Government (Minister of Economy and his Treasury, Finance and Domestic Trade secretaries) and the Secretary of Agriculture. Since export taxes have been seen as a virtually irreplaceable instrument to accomplish the objectives of decision-makers, there has been a constant conflict between governments, who always preferred high export taxes, and rampean interests who wanted them reduced or eliminated. As noted above, the more technically-oriented Agriculture Secretaries sought policies to reconcile Pampean development with government purposes.  -7- major infrastructure projects and credit for the industrial and hydrocarbon sectors. Bank operations in the agriculture sector since the mid-1960s had been limited to three projects. The Balcarce Livestock Development Project (Ln.505- AR), approved in 1967, was designed to encourage the adoption of new technology in pasture production and management and in animal health practices. Implementation spanned 12 years and results were mixed due to high inflation, economic and sector policies non-conducive to sectoral investment, and the cattle cycle. An Agricultural Credit Project (Ln.1564-AR), approved in 1978, was to provide medium- and long-term credit for on-farm investments. However, an unfavorable investment climate and large inflows of competitive low-cost capital from foreign commercial banks caused its cancellation in 1980 without disbursement. Lastly, the Grain Storage Project (Ln.1521-AR), approved in 1978 and reformulated in 1983, aimed at expanding national grains storage capacity, and at upgrading rail and port grain transport and handling facilities. By 1985, disbursements were still only 19% of the loan and implementation was hampered by serious institutional and financial problems. II. Identification. Preparation and Appraisal Introduction. 1.23 Several important internal considerations spurred the Bank's development of this loan. First, policy-based lending was still a comparatively new mechanism in the Bank. The dual possibilities of rapid disbursement supporting major policy change while alleviating acute foreign exchange shortages created considerable enthusiasm for such lending. In the Region, Ecuador, Uruguay, Panama and Brazil already had AGSALs underway or pending and Argentina seemed ready for such an operation. A new democracy in Argentina lent added impetus. Secondly, the Baker Plan announced in October, 1985, committed the Bank to a US$9.0 billion share in assisting debtor nations develop their way out of debt. Thirdly, agricultural lending in Argentina had been modest and sporadic (para 22) and the Region had taken a decision to develop a lending pipeline. No recent agricultural sector work existed, however, on which to construct an informed dialogue, and thus the Country Economic Memorandum (CEM) of June, 1984 became the principal blueprint, largely pre-selecting the policy issues and loan concept. 1.24 Three policy areas were identified by the CEM as being crucial to economic stabilization: (_) the chronic causes of the public sector deficit, especially tax administration; (b) export promotion; and (c) weaknesses in the financial system. Concerning the agricultural sector, the CEM noted that yields were low and that much could be done to improve cost/benefit ratios to promote fertilizer use, especially on grains. Price stability was needed to promote large-scale adoption of improved technologies. The CEM's arguments for export tax and import tariff removal were persuasive. It asserted that the entire Argentine farming system would be transformed, and that overall agricultural production would rise by at least 10-15X (grains by up to 50%), the result of  -8 more intensive use of plant protection materials and farm machinery, as well as more productive croppino, patterns. 1.25 The CEM recommended gradual replacement of agricultural export taxes and import tariffs on inputs by other revenue measures, preferably a land tax, thereby producing rapid and significant increases in the production of Argentina's principal export crops. In order to be an effective policy instrument and to have these desired effects, however, its burden on the taxpayer had to be considerable. It concluded that a land tax was preferable on developmental grounde to an agricultural income tax or value-added tax, would be easier to collect, and would transfer the burden of agricultural taxation away from those producing for export and cause it to fall more heavily on domestic producers of high-value products. Surprisingly, the CEM made no mention of the interchangeable use of export taxes and the exchange rate (para 14 and Annex III) or the central significance of a stable exchange rate policy to an export- dependent agricultural sector. 1.26 Land Taxes in Argentina. Taxes on rural property in Argentina have been viewed periodically as instruments of agricultural policy. Not easily enacted, they have proved even harder to implement. A preliminary draft law of 1964 sought to reform farm taxation by changing the method of calculating farm income, but was not approved. In 1969, efforts were made to use a proportional land tax levied jointly by the national and provincial governments, in conjunction with income taxes, to reward efficient agricultural producers and penalize those who under-utilized their land. After a few years, this system was abandoned. Again, in 1973, the Congress approved a tax on the potential net income of land but it was never implemented. A similar draft law in 1974 did not proceed. Rural property taxes devolved once more exclusively to the provinces, where they had become a familiar though modest revenue tool. In a number of provinces, notably Buenos Aires, C6rdoba and Entre Rios, well-designed and technically well-managed legal and fiscal cadastere exist which were seen as good models for a federal system. In general, however, provincial land tax administration was poor, collections low and evasion widespread.7' Loan Preparation 1.27 An exploratory mission in mid-1984 identified a credit project and an extension/research project; it was felt that both could be co-financed with the Inter-American Development Bank (IDB) in 1985, if arrangements could be made rapidly. However, the IDB was active in Argentina at this time and ultimately proceeded independently with two similar operations which made no demands on the GOA with regard to sectoral reforms. Thus, neither opportunity materialized for the Bank. The new democratic government under President Alfonsin was str"- ?ling to formulate policy in a deteriorating economy, little sector work ex sted on which to base a dialogue, and the results of discussions with the Government were 7/ Under agricultural and livestock emergency laws (Emerpencia Agropecuaria), the Secretary of Agriculture can declare a district to be in a "state of emergency", resulting in tax forgiveness for up to five years. Evidently, up to 102 of rural districts are declared tax exempt in this manner each year.  -9- inconclusive. Nevertheless, it was tentatively concluded that the Bank's short- term strategy for agricultural development in Argentina could focus on projects with short gestation periods, rapid disbursement and export bias. 1.28 The early germ of an AGSAL envisaged a modified investment program with no policy conditionality, focussing on financing imports of yield-augmenting inputs, e.g., fertilizers, agro-chemicals and farm machinery. An AGSAL was seen to avoid uncertainties associated with institutional credit at that time. The CEM went much further, however, urging policy dialogue and reforms as necessary conditions for Bank lending. The logical step up from removal of import levies was removal or reduction of export taxes, which clearly needed more time, study, and close consultation with the IMF because of the revenue implications. Further exploratory missions in 1984 produced the outline of an AGSAL with policy conditionality to be aimed at promoting increased production and exports. 1.29 In September, 1984, during informal discussions for an AGSAL, Argentine officials asked the Bank to finance a short-term agricultural recovery program, citing bold recent measures to reduce distortions. These included almost total removal of fertilizer import levies and other import taxes, major reductions in ad valorem taxes, and maintenance of positive real interest rates on agricultural credit despite very high inflation. Dialogue was seen focussing on the removal of export taxes on grains and their substitution by an alternative revenue mechanism. Staff warned of likely delays in implementing such reform, which might necessitate a series of loans. 1.30 Issues Paper. Noting the "long-term commitment" in Argentina to eliminating export taxes on grains, the Issues Paper (November, 1984) outlined an AGSAL based on export tax substitution with the inclusion of studies designed to guide medium-term policy and institutional reform. Export tax reform would take longer than the implementation period of the loan, but could be agreed based on a time-bound program. It was decided to leave the matter of an over-valued currency out of discussions with the Government, but to ensure that loan conditionality underscored exchange rate compliance with the existing IMF agreement. 1.31 The Preparation Mission of November, 1984 concluded that export taxes should be replaced as a source of fiscal revenue, preferably with a land tax, within the context of a complete review of the tax system. It was evident, however, that the GOA was divided on the loan concept. First, the Ministry of Economy claimed it had not formally approved the mission, and was both unprepared and reluctant to discuss substantive issues. Mission plans to meet with the Minister of Economy fell through and senior level staff were not available. Second, during the mission, much-publicized policy reforms were retracted. Most importantly, export taxes were raised to help offset the 6.82 devaluation effected in late October, under the IMF standby arrangement. GOA officials (and the IMF) stressed that these measures were purely temporary, necessary to alleviate an immediate fiscal crisis. 1.32 Just before the November mission, SAGyP had released a draft of the National Agricultural Development Program (PRONAGRO) for 1984-87 (para 17). It  - 10 - was a coherent policy statement, but was unfortunately explicit in retaining the concept of a price-stabilizing export tax, stating: "A stabilization of agricultural prices will be promoted through a flexible policy of export taxes in order to regulate variations that the country, as a price taker, does not control. The flexible use of export taxes means using them as an instrument of incomes policy for the sector and avoiding their utilization for only fiscal purposes". The Bank's Aide Memoire stressed the importance of full substitution to prevent reinstatement, pointing out that while it was technically possible to maintain both taxes, continued discretion in export tax use reduced incentives and depressed farmer confidence. 1.33 The issue of export tax elimination versus reduction is exceedingly important, and the Bank's attitude on the issue shifted over time. There is no discussion of what elimination really entailed; removing them from the statute books through congressional action meant something quite different from, say, reducing them to zero with Goverrment agreement to keep them there. Evidence suggests that SAGyP saw elimination as a distant, ideal goal, with substantial reductions equating with reform in the medium term. SAGyP's real focus, along with producer groups, was to remove discretion in export tax use, which was seen as needing legislation. The Initiating Memorandum (para 36) included this as part of the evolving loan strategy, but no concrete expression followed in loan design.Y8 1.34 Despite these set-backs, the missior. recommended the Bank's favorable consideration of a sector loan to Argentina if a policy framework could be agreed with the GOA. The loan would be around US$200 million for two years, authorized in tranches, and serviced through a revolving fund. Som" 601 would be disbursed against fertilizer imports. Loan size had to be large enough to permit a significant export tax reduction, not only to motivate policy change but to substitute directly for tax revenues foregone. A second loan, or even a series of loans, was envisaged to complete the process. 1.35 Loan processing was gathering a momentum not matched by the quality of the dialogue or by the extent of commitment in Argentina. Telexes of the period convey a sense of the Bank pressing the GOA to provide a formal, written request for the loan. However, economic policy formulation and implementation were in disarray at the end of President Alfonsin's first year in office, macroeconomic conditions were again deteriorating and the economic team had just been replaced. 1.36 The Initiatint Memorandum. The principal elements of the loan strategy required: (a) a rational statement of agricultural policy (PRONAGRO Of The issue of elimination was not raised again until the Bank began to assess its options for release of the second tranche, once passage of the FLT became unlikely.  11 - being described as a first, imperfect step); (b) total abolition of export taxes on agricultural commodities as soon as a viable substitute could be found (with a complementary study on feasible substitutes within the context of an overall taxation review);91 (C) GOA commitment to rational exchange rate policies beyond 1985 (the IMF arrangement called for a foreign exchange policy that generated a trade surplus, but only covered 1985); and (d) selective removal of restrictions on imported inputs. The two essential steps to land tax enactment were to be an updating and upgrading of the cadaster and congressional passage of the required legislation. Importantly, the Initiating Memorandum (IM) stated that reintroduction of export taxes was to be by congressional approval only. 1.37 Loan funds were to be disbursed to cover revenues lost from export tax reductions, i.e., direct substitution. Guarantees of follow-on loans would be needed to see the GOA through a program lasting four years or more. The proposed first loan amount had almost doubled since the mission of November, 1984 -- from US$200 million to US$360-370 million -- in order tos (a) motivate GOA to carry out major agricultural policy reform; and (b) permit a significant and early reduction in export tax rates. A second loan of about US$265 million would follow to complete the process, unless the FLT became partly operational in the interim. 1.38 The strategy assumed that the fear of serious economic repercussions would deter the Government from backsliding on its program to build producer confidence. This assumption was dubious in view of evidence that producer confidence had had short shrift for decades and benefits foregone from misguided policies had long been an accepted cost of economic management. Similarly, a public statement of policy intent was expected to bind the GOA to sustained reform, even though widely-heralded measures had been revoked under fiscal pressure just six months earlier. 1.39 Risk Assessment. The strategy was acknowledged to be risky in the following respects: (a) four years might be inadequate to secure the reforms sought, miring the Bank in a protracted and costly commitment; (b) uneven progress in fvture IMF/Argentine agreements could delay second tranche release and jeopardize processing and approval of a second sector loan; (c) budgetary pressures might force the GOA to raise export tax rates again; and (d) the US and the EEC might radically alter their agricultural support programs, adversely affecting international markets in which Argentina was operating. 1.40 Reaction of the IMF. The IMF criticized key aspects of the strategy as follows: (a) It rejected firmly the substitution of borrowed funds for fiscal revenues, viewing it as extremely risky especially if implementation were to become protracted. Export taxes might be economically 9/ The land tax was seen as "probably the best substitute" because of its positive impact on production. However, studies to confirm this would be carried out simultaneously with a complex series of actions to enact the land tax and prepare its cadastral base.  - 12 - inefficient, but reducing them before instituting a new tax was believed to be unjustifiable, especially in a country with chronic public deficit problems, and tax reform could be interrupted at any time. Further, the agricultural sector would object strongly to the introduction of a land tax after even partial relief from export taxes, seeing it as an additional tax on the sector; (b) Once reduced to some rational level, the export tax rate needed to be locked in to avoid the historical oscillation which had discouraged producers, and set at uniform rates for all products. Basing rates on relative supply elasticities, as proposed by the Bank, was believed technically difficult to formulate and hard to explain to officials and to the public; and (c) It viewed unfavorably the Bank's suggested collaboration on an exchange rate strategy, citing jurisdictional parameters between the two institutions (which meant that the Bank had to fall back on much weaker understandings in loan conditionality, and hope for the best). 1.41 Loan Viability and Timeliness. The evolving discussion and the IMF's reaction cast doubt on the fundamental practicality of the central concept in the Argentine context. The Bank continued to assert (despite IMF disagreement) that the time was opportune for t&x reform, citing GOA and private support. While it was agreed that the substitution of a land tax for export taxes would make Argentina highly unusual in the developing world, it was believed its latent potential to change rapidly, given the will, made it atypical of most developing countries. However, IMF objections to the concept of using loan funds to substitute for lost fiscal reveslues made it imperative that alternative interim taxes be found to maintain farmers' tax burden at then-current levels. Whether these levels were appropriate or optimal was not kro,wn, and whether such interim measures should have any specific relation to agriculture was not discussed. 1.42 Realizing that its knowledge of the proposed loan's wider fiscal and legal context was inadequate and that the magnitude of the tax changes proposed would raise questions, the Bank included studies of a revised agricultural tax system with submission of terms of reference for the study a condition for disbursement. However, conducting studies to determine fundamental matters affecting policy changes which were themselves the central feature of the program to be supported by the proposed loan merely iilustrated the weak analytical base upon which the loan was built. 1.43 The evolving design envisaged a two-year operation of about US$300 million, to be disbursed against general imports (even though the trend in sector  - 13 - loans at the time was to finance imports directly related to the sector)1