Tanzania - Agricultural Adjustment Program Project (Vol. 1 of 3) : Main report

Docim.ent of The World Bank FOR OFFICIAL USE ONLY Repowt No. P-5200-TA MEMORANDUM AN)D RECONMENDATION OF THE PRESIDENT OF THE INTERN1IONAL DEVELOPMENT ASSOCIATION TO THE EXECUTIVE DIRECTORS ON A PROPOSED DEVELOPMENT CREDIT OF SDR 150.4 MILLION TO THE UNITED REPUBLIC OF TANZANIA FOR AN AGRICULTURAL ADJUSTMENT PROGRAM March 5, 1990 Thi document has a restricted #isrbutlon and may be used by recipients only in the perfonnance of thelr official dubies. Its oontents may not otherwise be disclosed without Worl Bank authoriation.  CURRENCY h14D E0uVJULNTUITSI Currency Unit - Tanzanian Shilling (TSh) Dec. 1985 US$1 TSh 17 Jon, 1986 US$1 * TSh 40 June 1987 US$1 a TSh 63 June 1988 US$1 * TSh 96 June 1989 US$1 * TSh 145 Feb. 1990 US$1 T tSh 193 FISCIAL Government - July I - Juno 30 MAIN ACRONYMS USED IN THE TEXT AISCO Agrieultural and Industrial Supply Company BOT Bank of Tanzania CRDB Cooperative and Rural Development Bank CSD Child Survival and Development Program DAFCO Tanzania Dairy Farming Company Dfl Dutch Guilder EEC European Economic Community EIB European Investment Bank ERP Econmic Recovery Programse ESAP Economic and Social Action Programme PAO Food and Agriculture Organization - UN IC Industrial Company (owns and operates cotton ginnerie.) ICO International Coffee Organization IDA International Development Association IFAD International Fund for Agricultural Development ILO International Labor Organization In International Monetary Fund IRTAC Industrial Rehabilitation and Trade Adjustment Credit (Cr. 1969-TA of November 1988) IUCN International Union for Conservation of Nature and Natural Resources LSTG Pounds Sterling MRC Multi Sector RehabilUtation Credit (Cr. 1741-TA of November 1986) NAFCO National Agriculture aad Food Corporation NAPOCO National Poultry Company NARCO National Ranching Company NBC National Bank of Commerce N1C National Milling Corporation (the grain parastatal) OGL Open General License Facility PCS Primary Cooperative Society (the first-tier cooperative society, within the RCUs) PER Public Expenditure Review PFP Pollcy Framework Paper PSAP Priority Social Action Programme RCU Regional (or subregional) Cooperative Unlon (the second tier cooperative society) SAF Structural Adjustment Facility SGR Strategic Grain Reserve SIDA Swedish International Development Authority SPPF Special Project Preparation Facility TANSEED Tsanania Seed Corporation TCHB (Cashewnut) Tanzania Cashewnut Marketing Board TCMB (Coffee) Tanzania Coffee Marketing Board TCMB (Cotton) Tanzania Cotton Marketing Board TFA Tanganylka Farmers Association TFC Tanzania Fertilizer Corporation UNDP United Nations Development Program UNFPA United Natlons Fund for Population Activities UNICEF United Nations Children's Fund  FOR OFCIL US ONLY TABLE OP CONTENTS C>WbIS AND PROGRA S11104RY . . . . . . . . . . . . . . . . . . -$ PART I. THEECONOMY . . . . . . . . . . . . . . . . . . . 1 A. Background . . . . . . . . . . . . . . . . . . . . . 1 B. The Economic Recovery Program . . . . . . . . . . . 2 C. Recent Developments and the Impact of the ERP . . . 3 D. Me4ium-Term Prospects . . . . . . ... . . . 4 PART 11. COUNTRY ASSISTANCE STRATEGY AND BANM GROUP OPElSRAONS 5 A. Country Assistance Strategy and Bank Objectives . . 5 B. Bank Group Operations . . . . . ...... . 7 Past Bank/IDA Lending . . . . . . . . . . . . . . . 7 Portfolio Performance . .............. 7 IFC Investments . . . . . . . . . . . . . . . . . . 8 C. External Capital Requirements . . . . . . . . . . . 9 D. Aid Coordination . . . . . . . . . . . . . . . . . . 9 Collaboration with the IMF . . . . . . . . . . . . . 9 Cooperstion with Other Donors . . . . . . . . . . . 10 PART III. THE AGRIULTUL SECTOR . . . . . . . . . . . . . . 10 A. Sector Structure . . . . . . .. . . . . . . ... 10 Production Base and Institutional Arrangements . . . 10 Institutional Arrangements for Marketing and Pricing 11 B. Sector Performance . . . . . . . . . . . . . . . . . 13 C. Sector Deveiopment Strategy . . . . . . . . . . . . 16 Resources and Constraints . . .... ..... . 16 Sector Objectives . . . . . . . . . . . . . . . . . 16 Sector Strategy . . .. . .. .*. . . . .. . .. . 17 This document has a ticted dstbuon and may be used by repenbyts ony in te pufonnce of theioffoc dutkists contents may not otherwe be dbcosed without World Dank authodrzaon.  PART IV. TIM AGRMICULTURA SCRTOR ADJUST_EN PRM. .. .. . 18 A. The Program's Objectives and Expected Achievements . 18 Objectives . . . . . . . . . . . . . . . . . . . . . 18 Expected Outcomes ... . .. ..................... 20 Expected Achievements ............... 21 Pace of Reform . . . . . . . . . . . . . . . . . . . 21 B. Proposed Policy Changes in Grain Pricing and Marketing 22 Changes Already Implemnted . . . . . .. . . . . . . 22 The Tasks Ahead . . . . . . . . . . . . . . . . . . . 23 The Strategic Grain Reserve . . ...... . . .. 24 C. Proposed Policy Changes in Export Crop1 Pricing and Marketing . . . . . . . . . . . . . . . . . . . . 25 Coffee . . . . . . . . . . . . . . . ... .... 25 Cotton . . . . . . . . . . . . . . . . . . . . . . . 28 Cashermuts . . .O O ........ ..... . * # ..... . . 29 D. Proposed Policy Changes in Input Supply, Distribution and Pricing . . . . . . . . . . . . . . . . . . 31 Seeds . . . . . . . . . . . . . . . . . . . . . . . 32 Fertilizers . .*. . . . . . . . . . *. . . . . . . .. 32 E. Restructuring of the Public Sector Expenditure Program in Agriculture . . . . . . . . 33 PART V. THE_PROPOSED CREDI ... ....... . . .9... 34 A. Rationale for IDA Involvement . . . . . . . . . . 34 B. Utilization of Proceeds . . . . . . . . . . .. . 35 Procurement . . . . . . . . . . . . . . . . . . . . . 35 Disbursements . . . . . . . . . . . . . . . . . . . . 36 Audits . . . . . . . . . . . . . . . . . . . 36 C. Monitoring and Evaluation . . . . . . . . . . . . . . 36 D. Technical Assistance . . . . . . . . . . . . . . . . 38 E. Benefits . . . . . . . . . . . . . . . . . . . . 38 Major Benefits . . . . . . . . . . . . . . . . . . . 38 Poverty Alleviation and Social Dimensions . . . . . 39 Food Security ..... . . . . . . . . . .. ... 40 Environment ...... . . . . . . . . * . .* . . 41  F. Program Rises . . . . . . . . . . . . . . . . . . . . 42 PAR? VI CONDISIONS POR MRMLESE OF SECOND TUANCfl . . . . . . 43 PART VII. RECOYHUNDATION . . . . . . . . . . . . . . . . . 44 ASWZ ANNEX 1 Economic and Social Data ANME 2 Bank Group Operations AMNEX S Agricultural Data ANNEX 4 Policy Matrix ANRX 5 Letter of Development Policy ANNEx 6 Technical Assistance; Terms of Reference and Budgets (not included in this volume)  TANZANIA AGRICULTURCL ADJUSTMENT PROGRAM CREDIT AND PROGRA SUMMARY B8rrowert United Republic of Tanzania Beneficiart: United Republic of Tanzania Amounts IDA SDR 150.4 million (US$200.0 million) The Netherlands Dfl 75.0 million (US$ 40.0 million) United Kingdom LSTG 12.5 million (US$ 20.0 million) The operation is supported under the Special Program of Assistance to Low-Income Debt-Distressed Countries in Sub-Saharan Africa. In eddition to the cofiuancing identified above, other donors have expressed their intention to participate in the finaning of this program at a later stage. Terms: Standard IDA terms; 40-year maturity with 10-year grace period Description: The proposed operation would be the third in a series of quiek-disbursing operations to support structural adjustment in Tanzania under the Governmen;'s recovery program. Under the macro-economic policy framework provided by the PFPs, the overall aims of the proposed operation are to remove the major marketing and pricing policy constraints to agricultural develo',ent in Tanzania, and to restructure its marketing institutions and procedures accordingly. The specific objectives are to: (i) Improve producer incentives through increased reliance on market forces in price formation; (1i) Improve efficiency of agricultural marketing and reduce marketing costs; (iii) ease the macroeconomic burden imposed by the current system, particularly with respect to credit expansion, inflation, and export performance, and (iv) reduce the cost of food grains to consumers, thus contributing to poverty alleviation. The envisaged changes would involve a dramatic reduction in Government Involvement in crop and input marketing and active encouragement of other channels, including cooperatives and other private entities, aad the substitution of Government- administered prices with a market-based pricing mechanism. To achieve these objectives, the proposed operation would support, inter alia, the following policy and institutional changes for food crops, the three major export crops (coffee, cotton and cashewnuts) and inputst (4) determination of prices for foodgrains by market forces, with support from Government in some regions; prices for export crops would continue to be determined through auction or  - it - tenders, but producers would now receive the full price fetched by their commodities; fertilizer subsidies would be reduced; (ii) liberalization of: grain marketing by allowing cooperatives and private traders to trade grain freely and terminating NNC grain trading on non-commercial basis; export crops by allowing cooperatives to reta:a ownership of their crops until final sale and terminating crop purchasing by the marketing boards; and inputs, by liberalizing seed production and marketing and by licensing private traders to sell fertilizers; (iiI) restructuring of crop processing facilities, by transferring coffee curing factories to producers; establishing industrial companies Which would cwn and operate the cotton ginneries, and utilizing, renting or selling idle cashewnut processing factories; Government will also consider closing the fertilizer factory at Tanga. The proposed operation would support Government in developing plans to restructure its agricultural investment portfolio and review the viability of its production-oriented agricultural parastatals. Benefits: The changes in pricing policy and marketing structures are expected to generate an improved set of incentives for producers. For export crops, the market-based pricing systems and the accompanying institutional changes are expected to result in producers receiving a much higher share of the export price. In addition, the new systems would better transmit quality premia to the producers. In reacting to these improved incentives, producers are expected to improve the efficiency of resource allocation, increase production, improve the quality of export crops, improve food supplies at reduced costs, and make the agricultural sector more responsive to changes in market conditions. The reduction in the Government's involvement in export crop marketing and tue requirement that NMC operates on a commercial basis should reCuce the marketing system demands on credit and fiscal resources at the macLo level. These outcomes would contribute towards dampening of inflationary pressures, reducing the balance of payments gap and facilitating the management of monetary and financial policies. Ultimately, Tanzania would enjoy enhanced economic dynamism and macro- economic stability, better food supplies and larger export returns, and reduced poverty. Risks: The implementation of the proposed Program and the achievement of the benefits which it should originate entail some risks. These refer to the pace of adjustment, the Government's capacity to implement the program, and the response of market agents. On the  - {As - first, Government has reiterated its comitment to the economic and social adjustment program and adopted the required decisions, particularly on exchange rate. Thus, the risk of a slack in the paee of adjustment is deemed unlikely. Political resistance to the implementation of the agricultural policy changes, which have already been approved by Government, can be ant4.cipated. However, the constituency in Tanania for reform In agricultural pricing and marketing Is strong, and experience with previous operat'ons shows that, once decisions are taken, implementation Is usually 4nencumbered. Finally, there is a risk that the private sector may not respond fast enough to the new incentive structare supported by the proposed operation. In particular, cooperatives may fail to operate commercially, private traders may take some time to expand to cover most required areas, and both may take time to build the managerial and financial capacity to perform their tasks. Disbursements: The Credit would be disbursed in two equal tranches, equivalent to SDR 72.57 million each, through the Open General License facility (OGL). The first tranehe would be available for disbursement upon Credit effectiveness; tne second tranche would be released upon satisfactory fulfillment of specified conditions, expected by December 1990. Up to SDR 5.26 million would be disbursed against the technical assistanee, monitoring and evaluation components. Aiiraisal Renort: This is a combined President's Report and Staff Appraisal Report.  I=TERATLIONAL DEVELOPM4NT ASSOCIATION MENORANDM4 AND UECOMM)ENDTION OF TUE PMESIDENT TO TEE EXECUTIVE DIRECTORS ON A PRoPOSEn DEYELOPMENT CREDIT TO THE UNITED REPUBLIC OF TANZANIA FOR AN AGRICULTURKL ADJUSTMENT PROGRAM 1. I submit the following report and recommendation on a proposed Credit of SDR 150.4 million (US$200.0 million) on standard International Development Association (IDA) terms with 40 years of maturity to the United Republic of Tanzania in support of the Government's Agricultural Adjustment Program. The Netherlands and the United Kingdom will also participate in the financing of this Program by contributing Dfl. 75.0 million and 12.5 million pounds sterling, respectively. 2. An Economic Memorandum on Tanzania (Report No. 5019-TA) was distributed to the Executive Directors in August 1984. A new Economic Memorandum is currently under preparation and will become available in mid- 19903 the Memorandum's preliminary findings are reflected in Part I of this report. Furthermore, three Policy Framework Papers (Reports No. SecM87- 1077, SecM88-1234 and SecM90-83) were distributed to the Executive Directors in October 1987, November 1988 and January 1990, respectively. PART I. THE ECONOMY A. Background 3. In the late seventies, Tanzania entered an economic obeline from which it is now recovering. The downturn was caused in part by external factors (including droughts, increases in oil prices, the collapse of the East African Community, and the war with Uganda) and in part by serious weaknesses in macro-economic policies and management. These included inadequate incentives and resource allocation to the agricultural sector, an overly ambitious and uneconomic industrialization strategy, excessive administrative controls over economic activity, and continued growth in the public sector. Between 1978 and 1982, GDP grew by only 0.42 per annum, with agricultural and industrial output declining by 2.9? and 16.6S per annum, respectively. Inflation accelerated to about 302, the fiscal situation deteriorated sharply with deficits averaging aboL. 162 of GDP, and exports and imports declined significantly. By 1982, import volumes were 24Z and 322 below their 1970 and 1978 levels, respectively. These  - 2 - problems were compounded by sharp deterioration in the industrial (including crop processing) and transport infrastructure. 4. In response to the rapidly deteriorating economic situation, the Government launched Economic Survival Plans in 1980 and 1981, and a Structural Adjustment Program in 1982, which fell short of the effort required. The 1984185 budget provided a new pragmatism in economic management. The exchange rate was devalued by a third, parastatal subsidies were cut, imports were liberalized, and some restrictions on agricultural marketing were eased. These policies had positive effects and encouraged the Government to consider a more comprehensive policy reform program. B. The Economic Recoverv Prosram 5. In June 1986, the Government launched the 1986187-1988189 Economic Recovery Programe (ERP), which was strongly supported by the international community including IDA and the International Monetary r'und (IMP). Under the ERP, exchange rate adjustments were accelerated and reforms were introduced in foreign exchange allocation, trade policy, price, fiscal and monetary policies, and interest rates, as well as in agriculture, industry and transport. In support of these measuress (i) an IMP standby arrangement of SDR64.2 million was approved in August 1986; (ii) a Multi-Sector Rehabilitation Credit (MRC) was approved in November 1986, which incorporated financing from IDA (US$50 million) and the African Facility (US$46.2 million), and attracted Special Joint Financing of US$33.8 million by the Federal Republic of Germany, Switzerland and the United Kingdom; (iii) a supplemental MRC credit comprising US$30 million and US$26 million of IDA and Africa Facility financing, respectively, was approved in January 1988 and attracted cofinancing of Us$4 million and 4 million pounds sterl4ng from Saudi Arabia and the United Kingdom, respectively, and (iv) an Industrial Rehabilitation and Trade Adjustment Credit (IRTAC) was approved in December 1988, with an IDA credit of US$135 million and cofinancing of US$63 million from the African Development Fund, United Kingdom, Netherlands and Switzerland. In addition, a Policy Framework Paper (PFP) was approved in October 1987 which allowed Tanzania access to the IMP's Structural Adjustment Facility (SAF). A second-year PPP was approved in November 1988, and a third-year PFP, which outlines the intended policies for implementation in the period 1989/90-1991192, in February 1990. Maintenance of proper macroeconomic policy, including an active exchange rate policy, would be regularly monitored during implementation of the Program and would be a condition for second tranche release (para 149). 6. Policy reform is proceeding well. The main reforms to date includes li) large exchange rate adjustment (from TSh 17/US$l in early 1986 to TSh 193 in mid-February 1990); (ii) establishment of the Open General License facility (OGL) to improve foreign exchange allocation; (iii) achievement of positive interest rates in real terms; (iv) removal of price controls (from 400 categories of goods down to 12); (v) real increases in producer prices for export crops and the beginning of  - 3 - agricultural market reforms, and (vi) establishment of fiscal and monetary targets. Most fiscal components of the ERP have been successfully implemented. C. Recent Developments and the Imoact of the ERP 7. The reforms carried out under the first phase of the ERP resulted in gradual improvement in the economy. They significantly expanded the availability of essential and other goods for the entire population, and created and allowed access to a videning array of economic opportunities. Real GDP, which stagnated and declined during she early- to mid-1980s, grew by 3.61 in 1986, exceeding the population growth rate (3.21) for the first time since 1980. GDP growth is estimated at 3.92 and 4.1S for 1987 and 1988, respectively, indicating continued real per capita growth and an acceleration in the pace of economic recovery. Data for 1989 are not yet available, but initial indications suggest growth continuing at the same rate. Exports are also up, growing by an estimated 101 in 1988. Moreover, much of the obvious growth in economic activity which can be observed throughout the country is taking place outside of the formal sector and is not captured in the GDP statistics. The most visible source of growth continues to be the agricultural sector, where overall production increased between four and five percent in both 1987 and 1988. This reflects continued increases in production of foodgrairs and some traditional export crops, as well as of non-traditional export crops and vegetable production, particularly adjacent to urbau areas. 8. There are several areas, however, where macroeconomic performance continued to be unsatisfactory. First, net international reserves continued to decline and arrears continued to increase due largely to export performance lagging behind expectations. Second, credit expansion was higher than anticipated due largely to structural and managerial problems in the financial and agricultural marketing systems (net domestic assets in the banking system in FY89 increased by TSh45 billion, substantially above the TSh27 billion SAP ceiling). Third, as a result, the inflation rate was persistently high (281 as of June 30, 1989, as compared with a target of 181). Lastly, the substantial nominal depreciation of the Shilling had not been adequate to offset relative price movements and other structural changes within the economy. These issues have been addressed under the third-year PFP (para 5). 9. At the same time, infrastructure has continued to deteriorate, making it difficult for the productive sectors, and especially for agriculture, to respond to the changes in policies and incentives which have begun to take place. This deterioration has been caused by insufficient priority attached to this sector in the past and consequent inadequate fuading levels for maintenance and investment; inefficient centrally-controlled institutional structures; poor management and inadiquate technical capacity, and a restrictive regulatory environment which has stifled the financial and operational management of parastatals.  - 4 - 10. The Government has continued and deepened the policy reform process started under the ERP, and expanded it by integrating social aspects and linking them more closely with economic adjustment under the Economic and Social Action Programme (ESAP). The 1989/90-1991192 ESAP stresses Government's strong conmitment to economic reform while strengthening efforts on social issues through poverty reduction and human resource development, especially among vulnerable groups. With the more explicit integration of the economic and social dimensions, poverty alleviation consideraticns would be prominent and pervasive in all aspects of the Programme, and not limited to the social sectors alone. The main elements of the ESAP ares (i) continued adjustment of the exchange rate; (ii) continued trade policy reform; (iii) public sector management reform, including restructuring of public expenditure and the framework for managing public sector enterprises; (iv) financial sector restructuring; (v) reform of agricultural pricing and marketing; (vi) industrial restructuring; (vii) rehabilitation and revitalization of infrastructure; (viii) rehabilitation of social service delivery capacity, which had fallen substantially during the years of economic decline; and (ix) lessening the process of environmental degradation and desertification. The ESAP has one overriding alms a sustainable improvement in the living conditions of the population at large, whose overwhelming majority is very poor. 11. The adjustment measures which Government started under the UP and is continuing now under the ESAP represent a significr-at break with the past. Therefore, the development and implementation of the refonm programs involve a painstaking process of consensus building. So far, the Government has managed to mobilize a growing consensus on ERP and ESAP. But even the more pragmatic elements in Government are constrained to operate within the legacy of past policies. This background, combined with the limited technical qnd administrative capacity of the Government and the protracted internal decision making process, helps define the pace of reform. Nevertheless, the Government made considerable progress in the past three years and as long as the consensus for change can be sustained, continued success in implementation of the reform program (now under the ESAP) may be expected. D. Medium-Term Prospects 12. Over the next decade, the Tanzanian economy is expected to grow on average at 4.5 percent per annum in real terms. Continued policy reform, better infrastructure, and institutional improvements, including improved agricultural marketing, are expected to lead to more efficient resource allocation. Per capita income is expected to grow at 1.2 - 1.5 percent per annum, and consumption at about 0.5 percent. Dependence on imports is expected to decline over time. Total export volume is projected to increase by about ten percent per annum over the 1990s. The growth in exports, combined with modest improvements in terms of trade, would enable exports to finance a growing share of imports, from an estimated 32 percent in 1988 to some 50 percent in the mid 19909. The current account deficit, however, would continue increasing in absolute terms well into the 1990s but would continue decreasing in relative terms, as a percentage of both  total exports and gross national product (current balance of payments data and projections up to 1997 are shown in Annex 1, Table 3). The adjustment program would thus continue to require an active exchange rate policy, to which the Government is committed, coupled with large amounts of external financing and debt alleviation to support sustained growth. PART II. COUNTRY ASSISTANCE STRATEGY AND BANA GROUP OPET?IORNS A. Country Assistance Stratest and Bank Objectives 13. The main challenges facing Tanzania are to sustain the policy and institutional reform program in order to remove the remaining distortions and put the economy on a sustainable growth path; rehabilitate its key infrastructure and services; improve social services, meet other basic needs, and increase standards of living; and address long-term development issues, especially the reduction in the population growth rate and effective protection of the environment. The Association's assistance strategy for Tanzania supports the Government to meet these challenges through four major and clciely related objectivest (a) to improve the efficiency and effectiveness of economic management and resource allocation., including a reduced and rationalized role of the public sector; (b) to reduce sectoral and physical constraints; (c) to address social and longer-term development issues, particularly population and environment; and (d) to mobilize adequate external resources. 14. The Association's first objective is to assist the Government to further design and implement specific policy and institutional reforms in the context of the ESAP to improve efficiency and effectiveness of economic manigement and resource allocation. The recently completed public expenditure review and the forthcoming country economic memorandum are expected to provide further guidance on the relative priorities for future resource allocations. Policy reform is further supported by IDA's quick- disbursing operations. The proposed Agricultural Adjustment Program supports measures in agricultural pricing and marketing policy which are expected to have a major impact on the country's macro-economic stability (including reduced demands on the credit and budget systems, enhanced support to the balance of payments, and increased agricultural production and income) and to increase incomes in the rural areas and reduce real cost of foodgrains in the urban areas. The litter would contribute to poverty alleviation. On the producers' side, 85 percent of the population lives in the rural areas, and rural incomes are generally lower than those in the urban areas. On the consumers' side, lower foodgrain prices would improve real incomes of urban groups, particularly of those with low incomes, which spend a larger share of them on food. 15. The second objective of IDA's assistance to Tanzania is to help in reducing sectoral and physical constraints and in addressing social and longer-term development issues. This is reflected in its lending program, which includes projects tot (i) preserve and extend the achievements of the last 25 years in education, health, nutrition, and provision of basic  - 6 - services to the population; (ii) support the rehabilitation of the country's infrastructure (particularly in transport, where the need is critical); and (iii) develop agricultural support services, particularly through the rehabilitation and strengthening of the research, extension, marketing and credit systems. 16. The third objective of IDA assistance to Tanzania is to help the Government to address social and longer-term development issues, particularly population and environment. Other donors are also actively involved in this task. With regard to population growth, the Government is rarticularly concerned about its capacity to provide social services to a rapidly expanding population, given its limited human and budgetary resources. Therefore, a National Population Policy, already approved by the Cabinet, has been prepared with the support of the U.N. Fund for Population Activities (UNFPA) and the Inte+--.ational Labor Organization (ILO). Among its key objectives, the Policy aims to: (i) integrate population variables into planning; (ii) improve the standard of living of the population through protection of the environment and enhanced provision of basic human needs; and (iii) achieve lower population growth rates through birth rate reduction by voluntary regulation. IDA has provided the Government with Special Project Preparation Facility funds to carry out workshops to build national consensus around the population policy. Furthermore, the proposed FY90 IDA-supported Health and Nutrition project would strengthen the capacity of the Government to deliver health and family planning services, and to implement the national population policy. 17. While poptlation densities are low throughout most of the country, population growth rates are high and increasing population pressure on resources is beginning to be felt in: (i) some already densely populated areas (e.g., Kilimanjaro); (ii) some low potential areas; and (iii) around national parks and game reserves. Rapidly growing urban areas are also putting pressure on periurban land resources and on natural forest lands. Therefore, measures are required to reduce the rate of population growth and to protect natural resources, particularly wildlife, forest resources and agricultural land. 18. IDA is working closely with other donors to support the Government to address environmental issues. The Government, utilizing the technical expertise of the International Union for Conservation of Nature and Natural Resources (IUCN) with financing from the Swedish International Development Authority (SIDA), is carrying out an environmental assessment. This is intended to clarify the issues and to analyze the current resource status and prospects and the role of population growth in resource conservation. The assessment is expected to lead to the preparation of a National Conservation Strategy, which would identify areas at risk, propose and rank protecting and developing activities, and quantify resources required for its implementation. Over time, increases in agricultural yields stemming from IDA- and donor-supported rehabilitation of the research and extension services and improved input supply systems should reduce the pressure on land resources. Since rapid population growth increases the speed at which forest resources, and particularly woodfuel resources, are depleted, the Government with donor support prepared a  Tropical Forest Action Plan; a donor conference was held in December 1989 to discuss the implementation strategy of the Plan. The forthcoming forest resource management project, to be supported by IDA (see para 60) is an integral part of that 3trategy. 19. The fourth objective of IDA is to assist the Government in mobilizing external resources in the form and amounts required to finance the expenditures which are essential complements to the policy reform program. The increases in commitments from bilateral donors in recent years have been very encouraging, but for many of them Tanzania is already their largest recipient, and the scope for further expansion appears limited. Maintaining IDA commitments is therefore essential. B. Bank Group Operations Past Bank/IDA Lending 20. Tanzania joined the Bank Group in 1962. Beginning with an IDA credit for education in 1963, 74 IDA credits and 19 Bank loans amounting to US$1,613 million have been approved for Tanzania. In addition, Tanzania was a beneficiary of 11 loans totalling US$244.8 million which were extended for the development of the common services and development bank operated jointly by Tanzania, Kenya and Uganda through the former East African Community. (Annex 2 shows Bank Group operations in Tanzania). 21. Bank group lending in Tanzania has focused on agriculture, transport and communications, industry, and education and manpower development. Since FY81, Bank Group lending has been primarily for the rehabilitation and use of existing productive facilities and the expansion of infrastructure and services (such as power generation and education facilities) of long-term use to the economy. Projects have been designed to minimize new demands on the Government's recurrent and foreign exchange budgets and, since 1986, to support the ERP. They have included technical assistance for institution building and for better maintenance and use of existing capital facilities, and have supported policy reforms aimed at improved resource allocation and use. 22. The Bank Group has provided assistance on five occasions in support of Government efforts to deal with its balance of payments difficulties. The first such credit was made in 1974, and the second in 1977. The third, an Export Rehabilitation Program Credit for US$40 million (Credit 1133-TA, April 1981), assisted in financing much needed agricultural inputs, but the overall economic environment did not substantially change and hence the desired recovery of agricultural exports did not take place. The fourth and fifth credits, MRC and IRTAC, were briefly described in para 5 above. Portfolio Performance 23. The deteriorating econcmic environment of the past decade had a significant impact on the performance of the Bank Group's portfolio in  - 8 - Tanzania. Project implementation was adversely affected by the distorted policies of the past, by foreign exchange difficulties, and by management, marketing and infrastructural constraints. Recently, there has been improvement in project implementation, inclu"'_ng the agricultural sector. Bank Group disbursements, which were as low as US$70 million in FY86, rose to US$111 million in FY87 and to US$132 million in FY89. 24. Implementation of the quick-disbursing adjustment operations is progressing well. Besides real exchange rate devaluation, one of the major policy changes was to make, starting in February 1988, foreign exchange available automatically upon application for high priority import categories through an Open General License (OGL) system. The list of goods eligible was expanded in February 1989 and again in June 1989 and now covers a range of imported raw materials, intermediate goods and spate. parts and other items such as some drugs and pharmaceuticals. During the first ten months of 1989, Import licenses granted under the OGL have averaged some US$15 million per month, which is equivalent to about 122 of total imports. 25. The Operations Evaluation Department has issued a country study on World Bank/Tanzania Relations 1961-1987 (Report No. 8329, January 16, 1990). Overall, the study found that Tanzania and the Bank had given high priority to the productive sectors in the early seventies while the importance of an appropriate macroeconomic environment was underestimated. In contrast, issues of macroeconomic management were given a proper emphasis in the eighties. In the case of agriculture, the Bank ector studies in the 19708 and 1980s pointed out that pricing problems and the inefficiency of the parastatal institutions were undermining GDP growth, export expansion and project implementatinn. The record of effectiveness of project lending in the agriculture and rural development sector was not good, however. The highest project failure rate was in the mid-seventies, which coincided with a rapid build-up in lending. Besides inappropriate macroeconomic environment, severely constrained absorptive capacity was a factor in poor project implementation. The study's main recommendations on lending relate to the need to better gauge the size of the lending program and the complexity of projects relative to the country's absorptive capacity, to improve the linkages between economic and sector work and lending, to ensure strong government sense of ownership, and to avoid the creation of new institutions and enclave units. These recommendations have been taken into account in the current Bank assistance strategy for Tanzania and in the lending program, including the proposed agricultural adjustment operation. IFC Investments 26. IFC has carried out five operations in Tanzania. The first two, in 1960 and 1964, involved the Kilombero Sugar Company, with a loan of US$4 million and equity of US$0.7 million. In 1978 and 1979 loans for US$1.3 million each were made to the Highland Soap and Allied Products Limited, and Metal Products Limited, where IFC took equity positions of US$0.4 million and US$0.2 million, respectively. In 1985, IFC made a US$5.6 million loan to Amboni Limited for a sisal rehabilitation project. Total  - 9 - loans are US$12.2 million, almost fully disbursed, of which US$4.3 million is still outstanding. (Annex 2 shows the status of IFC operations in Tanzania). C. External CaRital Reauirements 27. The total external financing requirements for the 1990-92 period are expected to be about US$6.06 billion, including import requirements of US$4.72 billion and scheduled debt service, supplier credit repayments and reserves build-up of US$1.34 billion. The resources expected to be available to finance these requirements include US$2.08 billion in official exports, US$0.59 billion in private transfers, and US$2.59 billion in donor aid and multilateral financing (including the US$200.0 million for the proposed credit). (Annex 1, Table 4). 28. Tanzania is now receiving only IDA credits, but for much of the 1970s it was classified as a blend country. As of January 30, 1990, the principal outstanding to the Bank Group was $1,423 million, comprising $1,265 million to IDA and $158 million to the Bank. Annual debt servicing is $57 million, the bulk of which ($46 million) relates to Bank lending. With disbursements of $132 million in FY89, the net flow of Bank Group funds to Tanzania is positive. However, because of the low level of export earnings relative to total foreign exchange outgoings, the retention of a proportion of foreign earnings by exporters, the limited fungibility of aid flows and the need to finance oil imports, the amount of free resources left for debt servicing is very limited. Until export earnings are substantially stronger, the tight payments position is likely to persist. D. Aid Coordination 29. Since the economic reform program will continue to require substantial amounts of external financing, a key aspect of IDA's assistance strategy is to help Tanzania mobilize external financing. Over the past year, the Government has begun to take a more active stance, including the strengthening of the external assistance section of the Ministry of Finance and arranging regular meetings of donors to discuss both the overall progress of the ERP and the rehabilitation programs for particular sectors. However, Government capacity needs substantial strengthening in order to establish mechanisms for effective aid coordination. The Association -rill continue to assist the Government's efforts in aid coordination, using formal channels such as Consultative Group meetings and sectoral donor meetings as well as informal channels. Collaboration with the DMF 30. Staff of IDA and the IMf have collaborated in assisting the Tanzanian authorities in implementing the ERP since its launching in 1986, and in reviewing macroeconomic and sectoral developments, aiming at consistency between policy instruments and the objectives of stability, economic restructuring and growth. The PFP outlines the macroeconomic and  - 10 - general policy directions for this proposed operation. The policy reforms to be supported under this operation have been discussed and agreed vith the IMF. Cooperation with Other Donors 31. The Association will continue to collaborate closely with other donors in preparing and financing quick-disbursing adjustment operations and investment projects, and in addressing population, environment, and poverty alleviation issues. In the case of adjustment lending, donors are being enccuraged to untie their assistance so that they can support the overall move from administrative allocation of foreign exchange to allocation through the OGL facility. 32. Regarding the development and implementation of sector rehabilitation programs, IDA is working together with other donors to assist the Government to prepare comprehensive programs for each of the major sectors and subsectors; to help organize donor conferences where such programs are discussed, and to assist the Government in mobilizing external assistance to finance agreed programs. In this context, a donor conference on transport was held in December 1987, and meetings on education, health and possibly power are being envisaged. PART III. THl AGRICULTURAL SECTOR A. Sector Structure Production Base and Institutional Arranaements 33. Agriculture is by far the most important sector in the Tanzanian economy, and its development is critical for overall development, macroeconomic stability, and poverty alleviation. It accounts for more than half of the country's GDP, more than 80X of recorded export earnings, and almost 902 of employment. Host of the poorest of the poor are in the rural areas. The most important food crop is maize, followed by rice, cassava, sorghum, millet, pulses and sugar. The main export crop is coffee, followed by cotton, cashewnuts, tea, tobacco, sisal, cocoa and cloves. Cattle are the most important livestock, with sheep and goats in secondary roles. 34. Ecologically the country is very diverse, having nine different physiographic regions and some twenty agro-ecological zones. In general terms, the areas of soils with high fertility levels constitute only a small portion of the total land area and largely comprise soils of alluvial, colluvial or volcanic origin. Elsewhere, the soils are typical of tropical Africa, with low levels of nutrients and low organic matter content. The potential productivity of such soils is low unless inorganic fertilizers are applied and sound rotational practices employed. Rainfall is adequate in about one fifth of the country (about 21? of the country can expect, with 90X probability, more than 750 mm of rainfall) but low and  - 11 - unreliable in other areas. The central part of the country is the driest, receiving less than 500 mm of rainfall per year with evaporation exceeding rainfall during nine months of the year. 35. Yields are generally low due to a combination of massive use of unimproved local varieties and inadequate or inappropriata use of fertilizers. Overall, between 1978 and 1985, real agricultural GDP grew at an average of 1.0? per annum. This resulted in increasing reliance on imported (mostly donated) food and in a significant reduction in Tanzania's share in world markets. For example, the country's share in total world trade for some of its major export crops dropped as follows from 1975-80 to 1984-86 in value terms: coffee, from 1.51? to 1.18S; cotton, from 0.98? to 0.66?; tea, from 1.222 to 0.97?, and tobacco from 0.68? to 0.34?. 36. Agricultural production is mainly dependent on smallholders. The predominant farming strategy is subsistence agriculture (practiced by 80S- 90? of the population) aimed at ensuring that the family unit is self- reliant for its basic food requirements, selling surplus production only to provide for essential cash needs. While other systems such as monocultural export crops, irrigated agriculture and pastoralism are practiced, they represent a relatively minor part of the agricultural scene, even if their importance seems greater because they include the export crops. Estates are an important component of the tea, wheat and sugar industries, and make some contribution to coffee and rice production. Only in the sisal industry is production confined to estates. 37. There are basically four categories of smallholder farnings (i) the homestead farm, maintained primarily to meet individual family needs; (ii) the village collective farm, which is owned by the village and operated by village authorities; (iii) the block farm, which is basically a private farm subdivided into individual plots but farmed under the supervision of the village authorities; and (iv) private farming. Additionally, there are four kinds of large-scale commercial farming: (i) state farms, owned by the state directly or through parastatals; (11) foreign private farms, owned by companies or individuals, specializing in certain crops, such as tea; (iII) national private farms, owned by expatriates resident in Tanzania, national companies and associations; and (iv) joint ventures. Individual smallholdings, which occupy most of the cultivated land (83? of the farms on the mainland are of less than 2 ha, while farms of over 20 ha represent only 0.1?), have customary rights of usufruct. In the course of the villagization programme in the late seventies, the traditional right to allocate land to individuals was transferred from tribal elders to the village leaders. Nevertheless, passing on land use rights through inheritance still prevails. Institutional ArranRements for Marketing and Pricin. 38. Prior to Independence in 1961, Tanzania had a rural marketing system operated by itinerant traders and rural shopkeepers who were supplied by, and delivered agricultural products to, wholesalers located mostly in the major towns. It was a traditional trading system whereby traders provided inputs and credit in addition to purchasing produce.  - 12 - Ivmediately after Independence, private trade in agricultural commodities and inputs was banned and the internal marketing of agricultural products was reorganized as a three-tier system with primary cooperative societies at the bottom, secondary cooperative societies, called unions, in the middle, and the National Agricultural Products Board (NAPB) at the top. NAPB appointed the cooperative unions and their affiliated primary cooperative societies as its sole agents. Thus, from about 1964/65 when the NAPB and its cooperative agencies were established, Tanzania's agricultural marketing became essentially a compulsory single-channel system with primary village societies as the only buyers and the NAPB as sole exporter or distributor for local consumption. This system led to inefficiencies and reduced the incentive for farmers to increase the quantity and quality of their products. 39. A new system was introduced in 1973174. Not much cnanged, however. The marketing system remained a compulsory single-channel system, with village societies as only buyers and a newly established crop authority for each crop as monopoly exporter and distributor for the local market. The new system exhibited many of the weaknesses of the old system, often in -ore pronounced form. The village societies were cooperatives imposed from the top, instead of organized from the grassroots up. Farmers' allegiance to the societies was weak; so was the discipline and authority of the new crop purchasing entities. Officially set producer prices were determined as a residual after all costs of the marketing system had been deducted from an assumed sales price. The farmer was thus completely divorced from any marketing signals that would provide incentives for crop production. Producer pricos were established with little attention to the impact of intercrop competition. 40. In 1985, the cooperatives were re-established, and marketing boards replaced most of the crops authorities. Thus, the current three- tier marketing system was established, with considerable Government control. 41. For grain crops, the National Milling Corporation (NMC) sits at the apex of the marketing channel. Until recently, it held a monopoly on the wholesale trade of maize and rice. It buys from the Regional (or Sub- Regional) Cooperative Unions (RCUs), which each group a number of Primary Cooperative Societies (FCSs). The PCSs are the sole legal outlets for farmers selling rice and maize. Farmers producing coffee, cotton and cashewnuts face similar single-channel monopsonistic marketing channels. Harketing boards are the sole buyer and exporter of crops that are assembled, transported and in some cases processed by the cooperative societies. The private sector is only allowed to operate in maize and rice, and in minor food and export crops marketing. Farmers dealing with the marketing structure face unremunerative prices, unpredictable purchasiag periods and delayed payments. 42. Input marketing follows patterns similar to those for crops. Until 1986, input supply was confined to the relevant parastatals and the cooperative societies. Usually, inputs were imported by a crop authority or a parastatal (e.g., Tanzania Seed Corporation (TANSEED) for seeds), sold  - 13 - on credit from CRDB to the cooperative societies, and then distributed to the farmers. Other retailers such as the Tanganyika Farmers Association (TEA, a private firm), the Agricultural and Induetrial Supply Company (AISCO, a parastatal) and some regional trade companies (RTCs) buy inputs from the relevant parastatals, or occasionally import them, and sell them to farmers. They operate from their national or regional headquarters with outlets in only nine of the 20 mainland regions and no presence in the villages. In 1987188, the RCUIPCS channel handled 752 of all fertilizers, 361 of certified seeds and, on behalf of the crop marketi:Lg boards, 96X of all agrochemic1,s. As a result, input quality is variable. availability is unreliable (particularly regarding timing), and prices do not reflect economic costs. 43. For most crops, the Government publishes a price scale that includes prices for each level in the marketing channel, namely the producer, the PCS, the RCU, aLd the relevant parastatal. A cost-plus approach is used in establishing these prices whereby the marketing costs at each level in the marketing chain are added to arrive at the price to be paid at that level. The prices at each level are pan-territorial, thereby failing to take into account regional differences in transport costs. The prices for export crops at the producer, PCS, and RCU level also fail to properly distinguish quality and reflect changes in international supply and demand conditions. Input prices are controlled and in the case of fertilizer include a large implicit subsidy (TFC has not been required to pay to Treasury the worth of the donated fertilizers it distributes). Input prices are not differentiated regionally and in most cases do not reflect marginal costs. 44. Consumer prices for food grains are set independently of this cost-plus system, often with serious financial consequences for the NM. For example, for maize, the consumer price was set at about the same level as NWC's into-store price over the last few years. As a result, all the costs of the PCSs and RCUs are covered, but all the operational and transport costs of NMC are not, being translated into losses. Similarly, in selling export crops, the marketing boards receive the relevant international prices. In many instances, particularly before the recent Shilling devaluations, export earnings were not large enough to cover the sum of the shares of the producers, PCSs, RCUs, and the marketing board, with the boards absorbing the ensuing losses. B. Sector Performance 45. Throughout the first decade following Independence, crop production expanded rapidly, due largely to well-functioning extension and input delivery systems. The volume of marketed coffee, cotton, cashewnuts and tobacco doubled between 1960/61 and 1966/67 and doubled again by 1973/74 for cashewnuts and tobacco. Self-sufficiency in food-grain production was achieved by the late 1960s. However, the adoption of several policies in the 19709 and early 19809 (including villagization, dissolution of cooperatives, establishment of parastatal export and food crop marketing monopolies, closure of private shops, confinement of supply  - 14 - of agricultural inputs to a few agencies, and encouragement of heavy industry at the expense of smallholder agriculture) led to a steep decline in export crop production and stagnating food production. Agricultural growth rates plummeted, averaging only 1.02 per annum between 1978 and 1985, well below the population growth rate of 3.2X p.a. Between 1976177 and 1985/86. he production of major export crops declined by 82 per annum, and agricultural export revenues declined by 572 frow a peak of US$426.1 million in 1977 to US$184.1 million in 1985. 46. Over the last six years, the Government introduced changes, albeit limited, in pricing and marketing policies as well as in the structure of marketing institutions. In 1984, limits on intra-regional transport of grains were relaxed. In 1987, under the MRC, the special permits to move food grains within the country were abolished, producer prices were substantially increased for both food and export crops, export of minor crops was liberalized, and some flexibility was introduced into input supply. The real price of maize recovered -ubstantially, and the real price of paddy was revised to its highest level in the past two decades. These reforms were in line with the overall emphasis of the ERP on liberalizing the economy and aimed at reducing Government involvement in agricultural marketing, encouraging the development of other marketing channels, and increasing the role of the private sector, including co- operatives. 47. As a result of these policy changes, and with the coincidence of good weather, the growth rate of agricultural output accelerated. After having grown at 2.12 per annum in 1980-84, it grew at 3.52 in 1985, 4.4S in 1987 and an estimated 4.82 in 1988. The initial focus of reform was on maize marketing and the response was encouraging. Tanzania moved from a maize deficit to a maize surplus situation, including minor maize exports, and food availability throughout the country has generally improved. The private sector is now supplying more than half of Dar es Salaam's maize requirements. 48. Up to 1986, these overall growth rates resulted from diverging intra-sectoral trends. Food crop oroduction was inadequate but its growth rate was not substantially lower than that of population growth. Since 1987, its growth has outpaced population growth, resulting in maize surpluses. In terms of individual crops, maize has expanded from 1.65 million tons in 1982/83 to 2.36 million tons in 1986187 and an estimated record of 3.13 million tons in 1988/89. Paddy increased only slightly in 1987/88, but it had grown by almost 100,000 tons a year over the previous three years, reaching 6C4,000 tons in 1986/87, a figure more than double the average of the previous decade; production in 1988/89 reached new heights at 718,000 tons. Production of pulses and vegetables has also reached record levels. Wheat (a relatively minor crop in Tanzania) reached a new record as well, at 97,000 tons, higher than the previous 88,000 tons (1972/73) and 90,000 tons (1980/81). Total food production in 1988/89 is estimated to be 18Z larger than in the previous year. (Annex 3, Table 1). 49. In contrast to fooderops, the output of most of Tanzania's exPort crons stagnated through the late 19709 and early 1980s, or fell from the  - 15 - peak levels of the early 19709. In addition to the overvalued currency and serious processing and transport bottlenecks, ptoducer prices for several export crops fell in real terms. In the past two years, production trends have been reversed for some crops but others remain stagnant or continuu to decline. Thus, registered exports of coffee, tea, and tobacco, as well as their quality, have continued to fall, and pyrethrum and cardamom have performed poorly. The decline in cashewyut and sisal production appears to have bottomed out. Cotton production more than doubled since 1982183, to the equivalent of some 400,000 bales or more in each of the last three seasons. There has been an increase in the production of "minor" export crops as well. 50. The dete:ioration in agricultural performance in the late 1970s and early 1980s (paras 44, 47 and 48 above) had a devastating impact on the national economy and macro-economic stability. First, the 4ecline in agricultural exports through this period contributed to a widening balance of payments gap that reduced Tanzania's import capacity and made it increasingly dependent on foreign capital flows to meet import requirements. Second, inefficiencies and financial problems in the state- run marketing monopolies led to substantial losses. The c'ontinuing financial demands of the NMC and crop marketing boards are a principal force pushing credit expansion (new credit to parastatals in FY89 was about TSh 9.0 billion, as compared with a target of TSh 2.0 billion), while their unpaid overdrafts constitute a substantial portion of the total credit assets of the National Bank of Commerce (NBC) and CRDB. Every four or five years, Government had to assume liability for these overdrafts, which then became a claim on the budget. Finally, excessive credit expansion helped to spur inflationary pressures. 51. In the specific case of food grains, NMC is saddled with large and persistent losses due mostly to internal inefficiencies, high operational costs, poor financial structure and inappropriate Government pricing and marketing policies, including cost-plus pricing, compulsory purchases, and pan-territorial prices. Accumulated losses as of June 1988 were TSh 10 billion, equivalent to 41 of CDP. The reforms of the last three years have helped NMC's position considerably, particularly the steps to reduce NMC's role in maize and now rice marketing, but considerable market controls and distortions remain. 52. Similarly, the cooperative unions and marketing boards dealing with major export crops operate on a cost-plus basis which entails no incentive for efficiency or cost-effectiveness, and no accountability. As a result of internal inefficiencies, high operational costs, poor financial structure and, in some cases (particularly, cotton), low international prices, most of the marketing boards have run high losses. As of June 1988. the combined bank overdrafts of the marketing boards dealing with coffee, cotton and cashewnuts totalled TSh 10 billion, as large as those of NOC (or, another 4Z of the GDP). Board losses have also been periodically picked up by Treasury. 53. In an effort to compensate for increasing marketing costs and to reduce their losses, most marketing boards exerted downward pressure on  - 16 - producer prices. In the early 1980s, low producer prices had becoz3 a major disincentive for farmers to produce export crops. For example, cashewnut producers, who received up to 702 of export price in the early 1970s, received less than 301 throughout much of the 19809. The Government agreed under the MRC to increase most export crop producer prices to at least 602 of export prices, but the revised prices were subsequently eroded by further devaluations. (Annex 3, Table 3). C. Sector Develogment Strateav Resources and Constraints 54. Agricultural development will remain the engine of growth of the national economy. Tanzania has adequate land potential for increasing the production of both food staples and export crops. Progressive utilization of this potential would enable the country to attain food self-sufficiency on a permanent basis--already achieved for maize in the last three years-- and, at the same time, to obtain additional foreign exchange resources through expanding production of export crops. 55. However, critical bottlenecks to full recovery of agricultural production and exports remained, including: (i) weak agricultural services, including poor research, extension, credit and input-delivery systems; (ii) shortcomings in the marketing of agricultural exports, domestic food crops and agricultural inputs; (iii) inadequate and deteriorating transport facilities, and (iv) inappropriate pricing and macroeconomic policies (particularly on exchange rate) that contribute to poor farmer incentives and lead to further deterioration in the rural infrastructure. Sector Obiectives 56. The Government attaches a high priority to the development of the agricultural sector. Its main objectives are: (i) to maintain food security, ensuring at least self-sufficiency in maize and rice, and to improve nutrition; (ii) to increase crop export earnings; (iii) to improve rural incomes and their distribution and contribute to poverty alleviation; and (iv) to develop a stable agricultural sector, conserve its resource base, and preserve the environment. Subsidiary objectives are to reduce the economic cost of food marketing, to reduce its financial burden on the budget and credit system, and to allow for lower consumer prices r4ile preserving adequate producer prices. 57. In the long term, tI,e sector is expected to increase its sustainability through the use of improved agricultural practices; Introduction of conservation measures including agroforestry and the discouragement of cropping in high-slope terrain; proper use of pesticides and increased use of integrated pest management techniques; and better use of fertilizers including organic ones. The sector is expected to utilize lmproved, though affordable, technology with improved access to modern inputs, and to be well linked to the main input and output markets. It is  - 17 - envisaged that agricultural prices and output volume and structure xill be largely determined by market forces. As a result, the sector should generate all food required domestically and additional inputs for industrial processing, and export substantially larger amounts of today's 'export* crops, maize and maybe cassava and beans, and flowers, fruits, vegetables and foliage. The sector would thereby generate substantial income and employment in the rural areas, although its share in total employment would be decreasing. Migrations between rural areas, stimulated by improved transportation and marketing systems, increased monetization, and better access to consumer goods, may increase population density in currently under-populated high-potential areas. 58. Over the years, the country should aim at reducing the economic cost of the required foodstuffs and exports. This requires increasing productivity through technological change fostered by improving research and extension and increasing use of modern inputs, and concentrating incremental production in the most favored areas, in terms of trchnical potential and location. This would permit addressing poverty and income- related food insecurity while preserving adequate farmer incentives and ensuring a steady flow of export earnings. Sector Strateay 59. Efforts to achieve these objectives focus on augmenting production in the smallholder sector under rainfed conditions. (In view of the large amounts of land resources with a potential for rainfed crop development and the high cost of irrigation development and limited potential for it, irrigation development is not seen as a major factor in agricultural development in Ttnsania. Although large farms are, or may become, important in the case of some export crops, the vast majority of crop production is in smallholder hands, and experience on large-scale farm production of food grains is not encouraging. Thus, large-scale farming is not seen as a major factor either). 60. This smallholder/rainfed strategy requires an enhanced flow of new technologies to producers and an improved policy framework. The main priorities over the next five years are to significantly reform, rehabilitate, modernize and expand the country's research, extension, and credit systemss to protect the environment, and to reform agricultural input and output pricing and marketing systems. (i) Regarding the former, IDA credits to support the rehabilitation of the research and the extension systems, and for a tree crops project, were approved in December 1988 and March and June 1989, respectively. Drawing from past experience, these projects are aimed at strengthening the core planning, implementation, management and monitoring functions of specific institutions. In the agricultural financing area, the Agricultural Export Rehabilitation project provides funds for the rehabilitation of export crops while strengthening the capacity of the Cooperative and Rural Development Bank (CRDB) to identify, appraise and supervise specific projects. The reform and modernization of the credit system would be further pursued at two levels: an adjustment operation (Industry and Finance Restructuring) would address its reform at the national level, while an agricultural credit project  . i8 - would address the sector-specific requirementss both operations are currently under preparation. (ii) In the area of natural resources, a forest resource management project is being prepared. It would deal with institutional arrangements to manage the forestry sector and with legislation and policies to prevent over-exploitation of native forests; *.is would include pilot components to develop better systems to manage these forests. The project would also address land issues by reviewing land tenure arrangements, land legislation and regulations, land titling and land use planning. It would also support the introduction of more efficient ways of using woodfuel in tobacco curing. An environment technical assistance grant has been approved by IDA to assist with project preparation. A project-specific environmental assessment is being prepared. (iii) The proposed operation would address major changes in the grain and export-crop marketing systems and policies, reducing marketing casts, increasing the efficiency of marketing agents, and allowing lower- cost agents to operate in the markets, while at the same time improving the availability of food supplies and its reliability and reducing its real cost to consumers. New input supply policies are being implemented under projects assisted by the International Fund for Agriculturdl Development (IFAD) and the Food and Agriculture Organization (FAO)IEEC. These new policies and structures would be expanded Lationwide under the proposed operation. 61. The second most important priority over the next five years is the rehabilitation and expansion of the trunk and feeder road systems, and of the railways, which would improve the availability and reliability of traneport for agricultural inputs and outputs while reducing costs. Following a sector review and close Government/donor collaboration, comprehensive programs have been prepared to address Tanzania's transport problems. Most significant is the recently appraised Integrated Roads Project, which will finance the rehabilitation and maintenance of key trunk, feeder and access roads. All 16 donors involved in the roads subsector have agreed to cofinance the project. In addition, a ports project is scheduled for Board presentation on February 27, 1990, and a railways project (FY91) was preappraised in July 1989. Third, efforts should be made to make state farms more efficient and profitable; a review to determine their profitability and development potential and identify those activities which should be discontinued and assets which should be divested would be conducted under the proposed operation (see paras 117 and 118). PART IV. TMB AGRICULTURAL SECTOR ADJUSTMENT PROGRUg A. The Program's Objectives and Expected Achievements Obiectives 62. The overall aims of the proposed operation are to remove the major policy constraints to agricultural development in Tanzania and restructure accordingly its marketing and pricing institutions and procedures, and to  - 19 - contribute to macro-economic stability by addressing the problems in the marketing system which generate major pressures on aggregate variables. The immediate beneficiaries are the agricultural producers, who would receive better incentives for production and higher incomes for their effort, particularly in the case of export crops, where the difference between what Tanzania receives for its exports and what the producers ultimately receive would be reduced. At the aggregate level, the operation would support macroeconomic stability and trade reform measures through agricultural marketing restructuring consistent with more stable credit requirements and continued fiscal prudence; continued active exchange rate management, consistent with enhanced farmer incentives, and continued broadening access to foreign exchange; increased generation of foreign exchange through expanding agricultural exports, and public expenditure reforms in agriculture. The ultimate beneficiary is thus the nation at large, which would enjoy enhanced economic dynamism and macro-economic stability, better food supplies and larger export returns, and reduced poverty. 63. In more specific terms, the objectives of the proposed operation are tot (i) improve producer incentives through increased reliance on supply and demand forces in price formation; (ii) improve the efficiency of agricultural marketing and reduce marketing costs; (iii) ease the macroeconomic burden imposed by the current system, particularly with respect to credit expansion, inflation, and export performance, and (iv) reduce the cost of food grains to consumers, thus contributing to poverty alleviation. To achieve these objectives, the proposed operation would support policy and institutional changes in the current pricing and marketing systems for food crops, the three major export crops (coffee, cotton and cashewnuts) and inputs. Following up on the Public Expenditure Review, the proposed operation would support Government to develop plans to restructure its agricultural investment portfolio and review the viability of its production-oriented agricultural parastatals. A detailed list of the policy changes to be undertaken under the proposed operation is in the Policy Matrix in Annex 4. 64. The basic thrust of the proposed policy and institutional changes is to move away from Government-administered monopsonistic marketing channels and pricing regimes towards more competitive multi-channel marketing systems with prices largely determined by market forces. The envisaged changes would involve a dramatic reduction in Government involvement in crop and input marketing and the active encouragement of other channels, including a much larger role for the cooperatives and other private entities. 65. The basic policy objectives are similar for all crops, but the actual changes are specific for each of them, taking into account their different characteristics and requirements. In view of the sensitive political and food security issues related to food crops, the proposed institutional and pricing reforms recognize the need for a continued direct role for Government. This role, however, would be much smaller than at present and more oriented towards full cost recovery. NHC would still buy and sell crops, but it would do so on a commercial basis in competition  - 20 - with the private sector, including cooperatives. The proposed policy for export crops, in contrast, would aim at the complete phasing out of direct Government intervention in marketing and pricing. The parastatal marketing boards would no longer be involved in direct buying and selling operations, but limited to providing basic market services. For inputs, the proposed measures envisage a continued direct role for Government in the short term, in the context of a more competitive multi-channel system that includes the private sector. 66. The proposed pricing reforms aim at replacing the system of Government-administered prices with a market-based pricing mechanism. Government would still announce producer prices, but these would only serve as indicative prices and no Government intervention would be undertaken to defend them. In the exceptional case of maize, Government would maintain a limited market presence by supporting prices in remote areas, subject to budget availability. Market prices would be allowed to prevail over a wide range, with the Government playing a passive price stabilization role through the operation of the Strategic Grain Reserve/buffer stock. For inputs, pricing measures would aim to ensure that farmers pay for their inpttJ in accordance with marginal pricing principles. Expected Outcomes 67. The changes in pricing policy and marketing structures are expected to generate an improved set of incentives for producers. In turn, producers, by reacting to these improved incentives, are expected to improve the efficiency of resource allocation, increase production, improve the quality of Tanzania's export crops, make the agricultural sector more responsive to changes in international market conditions, and improve food supplies at reduced costs. Regional variation in prices based on differential transport costs would replace panterritorial prices and as a result, resource allocation and cropping patterns would be expected to shift in line with internal comparative advantage. 68. For export crops, the market-based pricing systems and the accompanying institutional changes are expected to result in producers receiving a much higher share of the export price. In addition, the new systems would better transmit price differentials for quality to the cooperative societies and individual producers, thereby increasing the incentives for improving export quality. The prices producers would receive would more accurately reflect changing supply and demand conditions on international markets, thereby improving the sector's ability to shift resources to take advantage of emerging market opportunities. At the same time, producers would face a more uncertain pricing regime as any variability in international prices would be transmitted back to them. However, the benefits of generally higher prices expected to accrue to producers should outweigh any disadvantage resulting from increased exposure to price variability in international markets. 69. The reduction in the Government's role in export crop marketing and the requirement that NMC operates on a commercial basis should also reduce the demands of the marketing system on the credit and fiscal  - 21 - resources at the macro level. These outcomes would contribute towards dampening of inflationary pressures, reducing the balance of payments gap and facilitating the management of monetary and financial policies, thus reinforcing the pace of economic reform and helping to strengthen the political commitment for change. Expected Achievements 70. In the case of food crops, successful implementation of the proposed policy changes - including the ones already implemented, see below - should remove most of the institutional constraints for the development of a multi-channel system with market-determined prices. The next stages would be to monitor the evolution and performance of the market (prices at the producer and consumer levels; behavior of the different market operatorst NMC, RCUs, PCSs, and private traders; grain flows between regions, and the like) and to evaluate the policy changes and their on- going effects in order to identify required adjustments to fine-tune its evolution. Such monitoring and evaluation would be carried out under the proposed operation (see paras 128 to 132). 71. In the case of export crops, the proposed changes would remove the most important single constraint to the development of a multi-channel marketing system, namely, the exclusive right of the boards to market their designated commodities. All RCUs - and the PCSs, farmers' associations and estates willing to do so - would be allowed to market their export cnmmodities on their own account. Also, private traders would be allowed to do so in the case of cashewnut marketing. Export commodities would be sold to foreign buyers through auctions or tender procedures, and the RCUs, PCSs and other sellers would receive the full price fetched by their commodities less legitimate marketing cost. Evaluation of the performance of the new system crop by crop, and particularly, assessment of the role and efficiency of the new participants in the market, would permit the identification and preparation of the following steps required to achieve a full-fledged multi-channel market for each export crop. One such step would be the reinforcement of the cooperative system, including the restructuring of some of the RCUs. 72. In the case of inputs, the planned restructuring should lead over the next few years to an effective seed industry, based on a multiplicity of companies openly competing with each other. In the case of fertilizers, in contrast, the massive subsidy - which materializes through TFC's handling of donated fertilizer products - and the panterritorial pricing policy make the participation of other market operators financially implausible. Besides its economic and financial benefits, the phased reduction in the subsidy would facilitate the emergence of other fertilizer importers and distributors. In the meanwhile, TFC is in the process o-7 licensing selling agents, who would buy from TFC and sell to producers on a cash basis while TFC retains the wholesale distribution function.  - 22 - Pace of Reform 73. Some of the intended changes would proceed rapidly, and many of their effects would be perceived over a short period. For example, as described below, open market maize supplies to the major cities have grown rapidly, and NMC losses and credit demands have been substantially reduced over the last two years (para 78). Others, in contrast, will require more tins. In some cases, the need for additional time would be due to growth reqairements. For instance, it would take some time for the RCUs to develop the managerial and financial capacity to operate commercially and take advantage of market opportunities, and for private firms to develop commercial networks to supply a full line of equipment and inputs to producers. In other cases, Government is cautious about the pace and scope of market liberalization because of concerns thats (i) overly frequent institutional changes could be disruptive; (ii) radical changes could undermine the recently re-established cooperative movement; and (iii) private traders may revert to old malpractices, including underpayment to farmers and under-invoicing of exports. B. Proposed Policy Changes in Grain Pricina and Marketing 74. In reference to the current problems (described in paras. 39 to 41; 44; 50 and 51 above), the policy reforms in grain marketing aim at liberalizing grain marketing, including termination of NMC trading in maize and rica on non-commercial terms, and enhanced roles for traders, farmer groups and the cooperative societies. Maize and rice prices would be largely determined by market forces. Subject to budget availability, Government would support maize producer prices in remote areas, and would retain a maize reserve for food security purposes. Chan&es Already Imnlemented 75. Government has already implemented many of the proposed policy changes. Maize and paddy/rice marketing was decontrolled (Odeconfinedw) at the primary society level in June 1988 and September 1989, respectively. Thus, RCUs and PCSs are now allowed to sell maize and paddy to other RCUs or PCSs, to traders and to any other market operators. NMC is no longer required to buy the maize or paddy offered by the RCUs. Any interested party is free to set up and operate maize, paddy and wheat mills. These reforms imply a major change in the role of NMC, from mandatory to discretionary participation in maize and paddylrice marketing. 76. Since June 1988, maize prices are largely determined by market forces. Government supports maize prices only in the most remote regions by buying maize through NKC on its own account at a pre-set price (TSh 13.90/kg for the 1988189 season, the same price used for procurement for the Strategic Grain Reserve, SGR). Maize producer prices in the rest of the country are neither supported nor controlled by Government. As a result, panterritorial maize prices have in fact disappeared. 77. The SGR was replenished in 1988 and an independent stock verification was carried out in December 1988. The SGR is now managed by a Board of Trustees, which has retained NMC as its agent to maintain the  - 23 - stocks for a fee limited to NMC's direct cost of handling, storing and transporting SGR stocks. Net additions to the SGR would be limited to what can be financed from the budget. 78. These policy changes are already showing results. Private traders have further expanded their role in maize marketing; open market supplies to Dar es Salaam are estimated to have grown from 35,000 tons in 1984185 to 70,000 tons in 1987f88, an amount already larger than current NHC sales, and have continued to grow since. RCUs and PCSs have developed an interest in marketing their produce. NMC has reduced substantially its market participation (e.g., maize purchases for the 1988/89 season were 124,000 tons, down from 225,000 tons in the 1987/88 season; the program for 1989/90 calls for purchasing 90,000 tons); fresh NMC demands on the credit system in 1988/89 (about TSh 3 billion) were smaller than in 1987/88 (TSh 6.3 billion), a reduction of almost two-thirds in real terms; the program for 1989/90 calls for additional demands of TSh 1.0 billion. The country is fully supplied with maize, and some maize has been exported. During most months in 1988, open market consumer prices were lower than the official consumer price (TSh 17/kg). Further, during the second semester of 1989, the new maize grading system was introduced, and NKC and RCU staff is being trained in grading procedures. The Tasks Ahead 79. To complete implementation of the policy package to be supported by the proposed operation, several actions would be adopted in the near future. They are discussed below. 80. Maize Price Support Levels. Production of maize has been larger than domestic consumption in the last three years. Therefore, Tanzania has moved from being a net maize importer to a marginal exporter. Since this situation is likely to continue in the future, market prices would move progressively from levels which reflect import costs to prices which would allow Tanzania to export at a profit. The price at which Government acquires maize for the SGR was increased by 152 in nominal terms (to TSh 15.90/kg) in 1989/90, resulting in a reduction by 112 in real terms. Thereafter, the SGR maize price would be set at a level at least 15? below the NMC into-store price prevailing in the region. Price changes will be closely monitored to check whether sufficient incentives are provided for Tanzanian producers to continue generating a maize supply adequate for food security purposes. Government would set indicative producer prices for maize starting with marketing season 1990/91. The cooperative unions would procure and sell maize on commercial terms, having the flexibility to decide upon the prices at which they will actually sell or purchase. 81. Liberalization of Grain Marketina Channels. Maize and paddy/rice trading at the villages, traditional markets, where producers have the right to sell and traders to buy, would continue, thus ensuring liberalization to the producer level. Traders will have the right to get from the authorities of any village approval to set up buying posts at the village level.  - 24 - 82. Decontrol of Maize Consumer Price. The maize consumer price would no longer be controlled. Appropriate instructions to NMC have been issued by Government in this regard. 83. NMC restructuring. The volume of NMC commercial operations, plus management of the SGR, is much smaller than the volume of operations NMC had when it was mandatory for it to purchase all maize offered by the RCUs. Thus, NMC must now be downsized to fit its reduced responsibilities. Further, NMC's financial structure is inappropriate, with large unsecured overdrafts and several non-performing assets. Therefore, an action plan acceptable to IDA for NMC's financial and operational restructuring would be approved by Goverument by June 1990; its implementation would be completed before the release of the second tranche (para 149.i). The restructuring would include, inter alias (i) each crop would be handled on comercial basis as a separate cost center and under a separate account; short-term cross-subsidization between crop accounts would only be utilized in special circumstances to achieve specific social objectives; (ii) other assets (particularly those not related to grain marketing) would remain in NUC hands only if they are profitable or be divested, if required, to pay off NMC debts/overdrafts not otherwise secured by stocks; and (iii) NMC activities intended to attain social objectives (such as price support and food security) would be undertaken only on Government's account and financed out of the budget. Starting in 1989190, NMC operates commercially, meaning NMC purchases each crop only when it expects to do so profitably, and is free to price maize and other crops to the consumer according to commercial considerations. 84. Complementary Measures. Success in transferring grain marketing responsibilities to the cooperatives, farmer associations and private traders requires improving cooperative performance, market facilities, information flows, and linkages among the different market operators. Actions designed to improve the commercial capability and efficiency of both cooperatives and private traders, and to improve market performance and facilitate the flow of grains from surplus to deficit regions have been identified. In addition, IDA and donor-supported programs would be prepared to enhance the cooperatives' managerial capacities. It has been determined that some cooperative unions are in need of financial restructuring; a study on their financial situation and proposals for putting them on a sounder financial basis would be undertaken. The Strategic Grain Reserve 85. Revised SGR rules of operation would regulate the acquisition and disposal of stock, stock rotation, and day to day management; and provide procedures to ensure SGR's financial viability. The SGR would be operated as both a food reserve and a tool for price support. It would thus contain two elementss the core SGR and the buffer stock. The SGR would be established from opening stocks plus acquisitions for price support purposes and for the SGR itself. The SGR core stock has been set at 100,000 tons of maize. This figure, which would cover the expected requirements of a three-month period, was estimated on the basis of average monthly NMC sales over the current decade using 1.5 standard deviations to  - 25 - define a medium confidence level. The upper limit for the buffer stock has been set at 50,000 tons (on top of the 100,000 tons of the core SGR, for a total of up to 150,000 tons). 86. SGR stocks would be released for purposes of stock rotation (either to the domestic or export markets), to provide supplies to agencies involved in emergency supply to food deficit areas, to serve under-supplied markets, and to reduce stocks in excess of the target SGR. Rules for stock rotation and emergency relief are straightforward. Determination of whether markets are properly supplied or under-supplied would be done on the basis of price indicators rather than volume indicators. Releases to serve under-supplied markets would be triggered by two trigger pricess one "may sell' and one 'must sell", which would be defined each year In reference to the average consumer price during the preceding harvesting season. Market prices rising above those ceilings would be interpreted as reflecting under-supplied markets. Depending on market prices, excess stocks may be released to the domestic markets or exported. Therefore, the operation of the SQR/buffer stock would be basically price-driven, instead of being utilized to affect market prices. Government intention is to use the SGR in a rather passive way, basically as a physical food stock for emergency situations rather than as an active market intervention tool. C. ProDosed Policy Chanaes in Exnort Crop Pricing and Marketing 87. In reference to the current problems (described in paras. 39-40, 43-44, 50 and 52-53 above), the policy reforms in export crop marketing aim at liberalizing export crop marketing channels for coffee, cotton and cashewauts, which together account fer US$200 million of exports per year, equivalent to 802 of total agricultural exports and 50 of all exports. Farmer incentives would be improved by better linking the prices producers receive to actual export prices. These improvements would be accomplished through the removal of the marketing boards from directly owning, purchasing and selling coffee, cotton and cashewnuts. The RCUs would be allowed to retain the ownership of these commodities until these are sold, and to receive the price they fetch (less marketing costs), allowing them to capture quality as well as enhanced marketing and transport efficiency premia. PCSs, estates and farmers' associations willing to sell their commodities under their own name through the auction or tender procedures would be granted the same right. This right would be available to all of them at their own option. Legislation amending the marketing board Acts to permit the implementation of the proposed policy reforms was passed by Parliament on January 26, 1990. Coffee 88. Coffee is Tanzania's most important export crop. It is grown by some 420,000 farmers (mostly smallholders plus a few estates) and generates between 48,000 and 55,000 tons per year, with a slight upwards trend over the last 20 years. Total registered exports have been worth some US$140 million per year (but only US$100 million in the last two years), equivalent to 55? of total agricultural exports and 35S of total exports.  - 26 - In spite of increasing production, coffee revenues are likely to be lower in the current year due to dramatic price reductions as a result of the collapse of the international coffee agreement. 89. Farmers sell their coffee to the PCSs. It is then channeled through the RCUs to the Tanzania Coffee Marketing Board. The Board sells the coffee at a fortnightly auction, -which the Board is also reeponsible for operating. Farmers get a down payment upon delivery of their coffee to the PCS, an interim payment one year later, and a final payment between 18 months and two years after harvesting. On average over the last 10 years, these payments have respectively represented 75S, 101, and 151 in nominal terms of total annual revenue received by farmers (equivalent to 822; 8.51 and 9.51 in real terms, respectively, assuming a 301 inflation rate). Tanzania Coffee Marketing Board (TCMB Coffee) also imports inputs and channels them to coffee growers through the RCU-PCS system. This process leads to untimely supplies of inputs, often of the wrong type, and to an expensive *cost recovery" procedure through mandatory deductions from coffee producer prices. 90. The Government has already approved the most essential features of the new marketing regime. The Board would continue managing the coffee auction. Producers would send their coffee directly to the auction floor under their own name, and get the full price their coffee receives. One coffee curing plant (Moshi) has already been returned to the relevant cooperative unions and farmer associations. Construction of two other plants was recently completed and they would be transferred to the relevant cooperative unions after at least one year of operation. TCMB (Coffee) vould continue providing ultimate quality control and market intelligence, and representing Tanzania in the International Coffee Organization (ICO). It could also offer its services as shipping/forwarding agent for the cooperative unions in competition with other agents registered in Tanzania. As in the case of NKC, the new TCMB functions would make some of its organizational units and staff redundant, and would end its requirements for credit. Accordingly, an action plan acceptable to IDA to reduce TCMB (Coffee) staffing and operating budget to reflect its reduced function would be approved by Government by July 1990, and its implementation would be completed before the release of the second tranche (para 149.i). 91. Input supoly. The current system through which inputs are force- fed to farmers and their costs recovered through a blanket deduction from coffee prices (i.e., zero marginal cost pricing) would be discontinued starting with the 1990/91 crop season, which starts in October 1990. Producers would be free to procure their inputs from any source of their choice (including the current TCMB-RCU channel) and would pay on cash or credit for the actual value of the inputs acquired, and not through output price deductions. TCMB would retain its input supply activities, but its de facto monopoly enforced through the blanket price deduction would be ended, thus opening the way for other suppliers to participate. This change would permit a major increase in the real coffee producer price, which would reflect the full value of coffee with effect from the 1991/92 marketing season, which starts in October 1991.  - 27 - 92. Private sector participation. The RoUs, PCSo, farmer associations and estates would be allowed at their own option to retain the ownership of their coffee, to send it to the auction floor under their own name, and to receive directly any quality or efficiency premium their coffee may fetch. 93. Improved auction Drocedures. Currently TCMB, through its Export Division, purchases from itself some 302 of the coffee sold at auction, creating a negative image among other bidders. Thus, eliminating the Export Division, or turning it into an independent entity, might strengthen the auction and make it more transparent and more attractive to buyers. On the other hand, there may be some advantages for Tanzania in retaining the Export Division, such as reducing the risk of collusion among buyers and generating profits for TCMB. Moreover, turning the Export Division into a separate entity would imply additional cost and run counter to Government's policy to reduce the number of parastatals. Therefore, a decision satisfactory to the Association on the future of the Export Division would be adopted based on the reconmmendations of a study on the coffee auction, which would include an assessment of the impact of the Division's operations on the auction's performance, to be conducted before the release of the second tranche. The coffee auction study would also review the benefits and losses accruing to Tanzania from the coffee deliveries to pre- determined buyers required under the Coffee Pre-export Financing Facility. If the study were to indicate that losses outweigh benefits, the Government would seek to eliminate the clause in the Facility agreement requiring specified coffee deliveries outside the auction (para 149.ii). 94. Rules for 'allocating' the ICO auota amona the different producers. No 'allocation' problem exists while the International Coffee Agreement quota system is suspended. Rules for allowing the different producers to sell through the auction or otherwise to quota and non-quota markets, however, would become essential to operate the new marketing arrangements if a new International Coffee Agreement is signed. The problem is as follows. As of April 1989, Tanzania had a quota of 40,200 tons. Coffee sold to quota markets earned higher prices. The quota premium was higher for arabicas; prices in non-quota markets were between 40-602 below those in quota markets as compared with only 30-35? for good robustas. From a national efficiency point of view, therefore, it is better to ship arabicas to quota markets and robustas to non-quota markets. Under the marketing system to be changed now, TCMB pooled all coffees, allocated them to both markets, and absorbed the quota premia. Under the proposed system, coffees belonging to individual RCUs, PCSs, farmer associations and estates would be sold to either quota or non-quota markets, leading to an unfair appropriation of the quota premia to those who manage to have their coffee sold to quota markets. Therefore, the introduction of the new system requires rules which, while protecting the transmission of quality premia to individual producers, permit a fair distribution of the quota premia - which is a national and not an individual asset - among all coffee producers. Rules to this effect would be adopted by either Government or an association (or rerresentative meeting) of coffee producers within three months of the teinstatement of an ICO quota system.  - 28 - Cotton 95. Cotton is the second largest export crop, generating US$50 million a year, or 20X of agricultural exports and 12.52 of total exports. Cotton revenues have increased lately due to, first, increased production, and later, improved prices. 96. Farmers sell their seedcotton to the PCSs, which sell it to the RCUs. The RCUs currently own and operate the ginneries and sell the lint to Tansania Cotton Harketing Board (TCMB Cotton), which sells it to foreign buyers through tenders. The ginneries have neither financial nor commercial autonomy, being subsumed into general RCU administrative procedures. nor the freedom to decide whose seedcotton they are going to process. Farmers get the Government-set producer price in a single payment upon delivery. Payment premia for quality are minimal. 97. Cotton marketing would undergo some major reforms, both in its export segment and in its domestic segment. On the export one. the Government would introduce improvements into the current tender procedure administered by TCMB (Cotton) to give it the transparency the Government, producers and buyers would like to see. Domestic textile mills would have full access through the tender to all lint grades and types. On the domestic segment, TCMB would stop buying lint from the producers. RCUs would retain the ownership of their cotton and sell lint directly through the improved tender procedures. Individual PCOS willing to retain the ownership of their cotton until final sale would be granted the same ownership rights accorded to the RCUs at their own option. 98. TCMB (Cotton) would continue providing ultimate quality control and market intelligence and representing Tanzania in international cotton bodies. As in the case of NMC and TCMB (Coffee), an action plan acceptable to IA to reduce TCMB (Cotton) staffing and operating budget to reflect its reduced function and to restructure its financial situation would be approved by Government by May 1990, and its implementation would be completed before the release of the second tranche (para 149.i). 99. Industrial comnanies. In order to improve management and operations of the ginneries, the Government would adopt the policy that ginneries will be operated by industrial companies (IC), which will be allowed to have a wide range of shareholders (depending on new equity or investment required, some of them may not have majority or any RCU participation). The ICs would be incorporated under the Companies Ordinance Cap. 212. They would operate independently of the RCUs, with complete managerial, financial and commercial autonomy, with each ginnery functioning as a separate cost center. At least one company (one in the MWanza region being the most likely candidate) would be established by June 1990. ICs to manage at least 50 of the effective cotton ginning capacity in Tanzania would be established before the release of the second tranche (para 149.i1i). These companies, or any other investors, may establish and operate new ginneries without any restrictions. Both RCUs and PCSs may send their seedcotton to any ginnery they so wish, a decision to be made on coamercial principles. A study would be undertaken with the objective of  - 29 - enhancing the ICs' efficiency and income generating capacity. The study would determine the various options available to achieve these twin objectives. The terms of reference include among other things modalities for facilitating wide shareholding, mobilizing of private capital, and purchasing (including modalities for payment) and processing seedcotton and selling lint on the c-mpanies' own account. The study would be completed by December 1990. The Government would adopt an action plan satisfactory to IDA to facilitate the implementation of appropriate actions for enhancing the IC's efficiency and income generating capacity before the release of the second tranche (para 149.iii). 100. TCMB (Cotton) as anent. The RCUs (and PCSs, as well as the ICs) may need an agent to bring their lint to the tender procedure; shipping/forwarding agents would be required to deliver the contracts afterwards. TCMB would be allowed to offer such agency services in competition with other agents registered in Tanzania; the RCUs and other lint sellers would be free to select the agent of their choice. 101. Two-tier payment. The Government would introduce a two-tier payment system for seedcotton during the 1990191 marketing season, which starts June 1990. Government would continue announcing a producer price. which would be the minimum each RCU and PCS can pay to the producers. Each RCU and PCS would buy seedcotton from them at prices no lower than the producer price; any difference remaining after selling the crop would be distributed to the producers. 102. The proposed policy change in cotton is consistent with the ginery rehabilitation program being implemented with the cooperation of the Dutch Government, and with the specific ginnery rehabilitation projects in *wAnza (supported by the European Investment Bank), Tabora (United Kingdom), and Mara and Shinyanga (Netherlands). Cashewnuts 103. Cashew is the main cash crop of Southern Tanzania (Mtwara and Lindi regions) and, to a lesser extent, of the Coast and Tanga regions. It is grown by an estimated 280,000 smallholders. Production declined dramatically from 145,000 tone in 1973/74 to about 24,000 tons in 1987188. This decline was due to an unfortunate combination of a fungal disease, an insect pest, the villagization program, a halving in real producer prices, and changes in cashewnut marketing, which led to substantial delays in both nut collection and payment. Nevertheless, cashewnuts is still the fourth largest export crop, and there is a large potential for expansion once the current technological and policy constraints are overcome. The Cashew and Coconut Tree Crops project, supported by IDA (Credit 2050-TA. of July 28, 1989), addresses many of the technological problems, while the proposed operation would focus on the marketing and pricing issues. The proposed policy change in cashewnut marketing is fully consistent with the requirements of the Tree Crops proje~t. 104. Theoretically, cashewnut marketing should work as described above for coffee and cotton. Producers' sole outlet are the PC$s which, through  _ 30 - the intermediation of the RCUs, sell to the relevant marketing board, Tanzania Cashewnut Marketing Board (TCMB Cashewnut), which is the sole exporter and processor. Three features make the cashewnut case different, however. First, in contrast with coffee and cotton, the RCUs in the cashew areas are weak and do not undertake actual nut marketing functions, making it difficult in the short term to adopt policy changes similar to those adopted for coffee and cotton whereby the RCUs would become the major marketing agents. Second, currently the price differential between raw and processed nuts is not large enough to generate a positive added value in processing. As a result, nine of the tvelve TOMB cashewnut processing factories are closed. Finally, the marketing margins, including both actual costs and rents, are extremely high, resulting in producer prices that up to June 1989 were only 302 of the export price. 105. Government has already implemented the proposed chdnge in pricing policy. In August 1989, the Government raised the producer price from TSh 40/kg to TSh 84 (a 1102 iucrease in nominal terms, equivalent to 632 increase in real terms), thus raising it from the equivalent of 302 to 752 of the current export price (and about 652 of the anticipated average price for the year). In future, producer price for cashewnut would be equivalent to at least 65S of export prices. 106. Regarding external marketing, the current private tender system would be improved, along the same lines as the lint tender; it would continue to be administered by TCMB (Cashewnut). Eventually, farmers would receive prices obtained through the tender based on the quality of their nuts less actual marketing costs. During the transition period, producer prices would continue to be announced to reduce farmers' risks. Regarding processing, a study to determine the best possible use of the idle cashewnut factories would be completed by December 1990. An action plan acceptable to IDA to utilize, rent or sell the currently idle cashewnut processing plants would be approved by the Governmet; its implementation would be satisfactory to IDA in accordance with the agreed timetable as of the release of the second tranche (para 149.iv). 107. To improve domestic marketing of raw nuts, other marketing channels, including private sector involvement, would be introduced. On the one hand, Governmet would encourage the establishment of joint ventures involved in cashewnut production and processing. These would be licensed to buy nuts from outgrowers in the neighboring areas. On the other hand, TCMB and the RCUs have the authority to license buyers, and would take active steps to do so. Procedures for licensing traders would be started early in 1990. Licenses would be provided upon request to all bona-fide applicants. A trader would not be precluded from getting licenses to operate in several areas, and different traders would be allowed to compete in the same areas. They would buy nuts from producers at prices no lower than the floor price, and from primar-es and unions, and sell to TCMB at freely negotiated prices until October 1991, when TCMB's role would be redefined to become the tender administrator and be responsible for ultimate quality control and market intelligence. From then on, RCUs, PCSs and traders would sell the cashewnuts directly through the tender. An action plan acceptable to IDA to reduce TCMB (Cashewnuts)  - 31 - staffing and operating budget to reflect its reduced function would be approved by Government by September 1990, and its implementation would be completed before the release of the second tranche (para 149.i). D. Proposed Policy Changes in Ingut SuPpl1. Distribution and Pricina 108. Availability of quality seeds remains a major constraint for increasing productivity of Tanzanian agriculture and for efficiaat use of fertilizers. Seeds are produced and imported solely by TANSEED. Major issues regarding availability and distribution of seeds are closely linked to the institutional and organizational aspects of TANSEED. Agrochemicals were chosen and imported by the Crop Authorities - and then by the Marketing Boards - and passed on through the RCUIPCS channel, without any choice for the farmers. 109. All fertilizers currently available are provided by donors. They are imported aud distributed by the Tanzania Fertilizer Corporation (TFC). TYC delivers/sells the fertilizers at its regional depots at the same (panterritorial) price. Currently this price covers all transportation and handling costs, but leaves nothing to repay the fertilizers' value back to Treasury. As a result, farmers are paying prices which are less than one third of the fertilizers' full cost at normal international prices. The implicit fertilizer subsidy and the panterritorial pricing introduce distorted incentives for effective input distribution and use, and preclude the development of a multi-channel national distribution system. The vast majority of fertilizer sales are currently carried out on credit (from CRDB through the cooperative societies), but PAO has estimated that no more than 402 of producers need credit to acquire fertilizers. 110. Two steps towards input supply liberalization were adopted recently. First, to promote an effective multi-channel input marketing system in the IFAD-supported Southern Highlands project area, all eligible input suppliers were licensed, exempted from price controls, and freed from administrative interference in their commercial affairs. Suppliers are permitted to purchase inputs directly from source, including from abroad. Second. the supply and distribution of agrochemicals was fully deconfined undet MRC. Major chemical companies are importing them, and all relevant agrochemicals are already ircluded in the OGL list. ill. The proposed adjustment program would support the restructuring and commercialization of seed production and distribution, and further deconfinement of fertilizer distribution at the regional level. The long- term objective would be for cooperat' unions, farmer associations, large estates, and licensed traders to freei