Burundi - Second Forestry Project

DomamI(of The World Bank FOR OmCIAL USE ONLY Repdt No. 5368-3'S STAFF APPRAISAL REPORT BURUNDI SECOND FORESTRY REPORT May 17, 1985 Central Agriculture Division Eastern Africa Projects Department t1 s dsui a - idui i GM i .n be anus by h fie.s oly in _e pef.nuKc Of thr1w 1ac dw5m as m.kib myt meiwide be d6d em widwwl WvM Bak aUdDSKhtim.  BURUNDI CURRENCY EQUIVALENTS Currency Units = Burundese Franc (BuF) US$1.00 = BuF 120.0 BuF 100 = US$0.83 WEIGHTS AND MEASURES Metric System Metric = British/US Equivalents 1 meter (m) = 3.3 feet 1 cubic meter (m3) = 35.3 cubic feet 1 hectare (ha) = 100 ares = 2.47 acres 1 kilometer (km) = 0.62 mile 1 square kilometer (km2) 0.39 square mile (sq.mi.) 1 kilogram (kg) = 2.20 pounds (lb) 1 liter (1) = 0.26 U.S. gallon (gal) = 0.22 Imperial gallon (imp. gal) 1 ton (metric ton) = 2,204 pounds (lb) ABBREVIATIONS AND GLOSSARY AFVP - Association Franraise des Volontaires du Progras/ French Volunteers CTFT - Centre Technique Forestier Tropical/French Tropical Forest Center) EDF - European Development Fund/Fonds Europ$een de developpement (FED) FAC - Fonds d'Aide et de Cooperation/French Aid and Cooperation Fund ILO - International Labor Organization INCN - Institut National de la Conservation de la Nature/ National Institute for Nature Conservation ISABU - Institut des Sciences Agronomiques du Burundi/ National Agricultural Research Institute ITAB - Institut Technique Agricole du Burundi/Burundi Technical Institute of Agronomy MFCZN - Mission Francaise Crete Zaire Nil/French Forestry Zaire- Nile Divide Mission ONATOUR - Office National de la Tourbe/National Peat Development Agency SCM or - Stacked Cubic Meter (equivalent to approximately 0.6 cubic stere meter solid) SRD - Societe Regionale de Developpement/Regional Development Company UNDP - United Nations Development Programme USAID - United States Agency for International Development WFP - World Food Programme FISCAL YEAR Government of Burundi : January 1 - December 31  FOR OFFICL USE ONLY BURUNDI SECOND FORESTRY PROJECT Credit and Project Summary Borrower : Republic of Burundi Beneficiary : Ministry of Agriculture and Livestock, Forestry Department Amount : SDR 13.0 million (US$12.8 million equivalent) Terms : Standard IDA terms Project : The proposed Credit would finance the second phase of a long- Objectives term afforestation program and assist the Government in and establishing long-term production and marketing policies for Description the sub-sector. The main Project objectives are: (i) to address the problem of growing wood demand of the population; (ii) to reduce Burundi's need for imports of timber, timber substitutes and fuel; (iii) to contribute to protecting the natural environment; and (iv) to promote greater integration of cropping, forestry and livestock activities at farm level and to 'Launch an agro-forestry program. Ninety rural nurseries (including about 30 new ones) would produce plants for sale to the population. Agro-forestry would be promoted to address the rural population's needs for wood, fodder, shade and soil improvement. Pilot actions would be undertaken in a few plantation areas to integrate afforestation with livestock and agricultural improvement activities. About 3,500 ha of eucalyptus and 1,500 ha of pine plantations would be established in western Burundi; in addition 2,000 ha of woodlots would be established in communes all over the country. Benefits : The Project would benefit about 60,000 rural families. or about 8% of the farming community by increasing their supplies of fuelwood and poles. The eucalyptus and pine plantations would serve the urban population by meeting the needs for charcoal of about 60,000 people in Bujumbura. The pilot activities would primarily integrate better afforestation programs with agriculture and livestock. Agroforestry research would define better means to improve present farming systems. Finally, the Project would have a significant impact on employment. Risks : The proposed Project does not involve major or unusual risks. Limited land availability is the most important problem in establishing large and medium sized commercial plantations. To minimize this risk, legal arrangements necessary to ensure the availability of planting sites would be finalized before Board presentation. The principal This document has a retrited disrbution and may be used by rcipients only in the perfonnc of their ofric dutie Its wtents may not oherwise be disdosed wihout Wodd Bank authorization.  - ii - economic risk of the Project is that management performance will not reach expected standards, thus reducing productivity of plantations and nurseries. This risk would be reduced by the recruitment of a technical assistance team. Estimated Local Foreign Total Project Costs 1/ - -US$ million-- . Commercial plantations 3.5 1.3 4.8 . Rural nurseries & agroforestry 2.1 0.2 2.3 . Agroforestry research 0.0 0.2 0.2 - Maintenance of existing plantations 0.2 0.3 0.5 . Livestock and agriculture 0.6 0.9 1.5 improvement . Institutional strengthening 0.6 2.9 3.5 . Scholarships and specialized training - 0.2 0.2 Base Cost 7.0 6.0 13.0 Physical Contingencies 0.7 0.4 1.1 Price Contingencies 2.0 0.9 2.9 Total Project Costs 9.7 7.3 17.0 Financing Plan US$ Million Equivalent X of Total Local Foreign Total Financing IDA 8.1 4.7 12.8 75 FAC - 2.6 2.6 15 Government 1.6 - 1.6 10 Total 9.7 7.3 17.0 100 Estimated Disbursements IDA fiscal year: 1985/86 86/87 87/88 88/89 89/90 90/91 91/92 --US$ millions Annual 0.2 1.8 1.8 2.8 3.4 2.0 0.8 Cumulative 0.2 2.0 3.8 6.8 10.0 12.0 12.8 Economic Rate of Return: 24% on the commercial plantations component representing 50Z of Project cost. Map No: IBRD 18487 1/ Taxes are negligible  SECOND FORESTRY PROJECT STAFF APPRAISAL REPORT Table of Contents Page No. I. BACKGROUND A. Project Background ................................. 1 B. The Agricultural Sector ................*... ....... . 2 C. The Forestry Sub-sector Issues and Strategy ........ 5 II. GOVERNMENT SUPPORT AND POLICY FOR THE FORESTRY SUB-SECTOR A. Forestry Institutions ........ .... .................. 9 B. Forestry Projects ........... ...............* , 11 C. The Quest for Wood .... ................0................. 12 D. Government Strategy and Policy for the Sub-sector .. 15 III. THE PROJECT A. General Description ..... ...........e........ 18 B. The Project Areas .................................. 19 C . Detailed Features ...... .................. 20 D. Project Costs ........*.*...O .......... *.......................... 25 E. Financing ................... ,...................... 26 F. Procurement .......... .... 27 G. Disbursement ......... ... ..... ............. ......... 28 H. Accounts and Audit ................ 0........ ...... 29 I. Environmental Impact *- ....................... . . 30 IV. PROJECT IMPLEMENTATION A. Organization .. ... ........ .. ............... ......... 30 B. Management of Project Components ..................... 31 C. Annual Work Plans ... .............. . ...*.. ........... * . 33 D. Staffing .... ....................... *................... 34 E. Implementation and Monitoring ............. . ......... 35 V. TECHNICAL AND PRODUCTION ASPECTS A. Technical Aspects .................................... 35 B. Production Aspects .............. ................... 37  - ii- Page VI. MARKETS AND PRICES, FINANCIAL AND FISCAL ANALYSIS No. A. Marketing and Prices ........ ......................... 42 B. Financial Analysis .... ............................... 45 C. Fiscal Analysis ...... 46 VII. BENEFITS AND JUSTIFICATION A. Benefits ...... .. ...... 47 B. Justification ... ..... ... ....... 48 C. Risks and Uncertainty ................................ 49 VIII. AGREEMENTS REACHED 50 Supporting Tables, Schedules, Charts and Annex Table: 1 - Forestry II Cost Table by Category 2 - Government Cash Flow with Project 3 - Economic Rate of Return - Flows of Costs and Benefits 4 - Financial Rate of Return - Flows of Costs and Benefits Schedule 1 Estimated Schedule of Disbursements Chart 1 Implementation Schedule Chart 2 Forest Department Organization Chart Annex 1 Statistics and Data on the Forestry St.b-sector Annex 2 Selected Documents and Data Available in the Project File Map IBRD 18487  BURUNDI SECOND FORESTRY PROJECT STAFF APPRAISAL REPORT I. BACKGROUND A. Project Background 1.01 The proposed Project would follow directly from the IDA-financed Forestry Project (Credit 918-BU) and was identified during the latter phases of implementation of the first project. In discussing future investments, it was agreed that a modified approach to afforestation should be adopted for a second phase project, involving greater emphasis on integration of forestry within smallholder farming systems. The reasons for this shift were that limited land availability was the main problem under the First Forestry Project, and that Burundi's farmers give the highest priority to improving crop and livestock enterprises. They are thus most likely to plant and maintain trees if it benefits their farm production overall. On this basis terms of reference for Phase II preparation were prepared in 1983 by the Burundi Government, reflecting IDA comments. The FAO/IBRD Cooperative Program was asked to assist with preparation of a Second Forestry Project. An FAO/IBRD-CP preparation mission visited Burundi in January/February 1984. It included an agronomist, a livestock specialist, a forester and an economist, all of whom were to address the recommended "agro-sylvo-pastoralV approach. Their report was received at the Bank in May 1984. An IDA mission composed of Messrs. Y. Gazzo and N. Brouard (IDA), J.C. Crouail, B. Ruet and P. Wood (consultants) visited Burundi in June/July 1984 to appraise the Project. The Fonds d'Aide et de Cooperation (FAC) was represented by Mr. C. Taupiac, forestry advisor attached to the Burundi Forest Department. 1.02 The Bank group is one of the major sources of development assistance in Burundi. To date IDA has financed six agricultural projects, of which three are completed, one is near completion (Forestry I, Credit 918-BU), and two others are being implemented. The Coffee Improvement Project (Credit 147-BU, US$1.8 million), the Second Coffee Project (Credit 593-BU, USS5.2 million), and the Fisheries Development Project (Credit 262-BU, US$6 million) have been completed. The general conclusion of the Project Completion Report, Performance Audit (Report No. 1531) and Impact Evaluation Report (Report No. 3314) on the First Coffee Project (at Ngozi) was that the project had achieved its objectives, although the quantitative results were somewhat below appraisal estimates. The Second Coffee Project (also at Ngozi) was to continue activities started under the First Project; in addition, it included a food crop promotion component. The Second Coffee Project PPAR (Report No. 4576) concluded that, overall, the project  -2- was a qualified success. The coffee component achievements were above appraisal estimates, while the food crop and livestock components were below estimates. This project was more affected than the first by the lack of trained and experienced staff, by insufficient and irregular Government counterpart funds, and by heavy dependence on foreign assistance. A Project Completion Report for the Fisheries Project was recently completed and a PPAR will be issued shortly. Poor management was the principal problem here and led to very disappointing results. Management problems were explained both by a weak parastatal agency and inadequate technical assistance in critical areas. The First Forestry Project (Credit 918-BU, US$4.3 million) is nearing completion; it is described in greater detail in paras 1.16 and 1.17. 1.03 The Kirimiro Project (Credit 1165-BU, US$19.3 million) and the Ngozi III Project (Credit 1192-RU, US$16 million) are under implementation. These two projects provide for a more integrated regional development approach than the two earlier projects for the Ngozi Region. They were designed to address key issues which emerged during the implementation of the First and Second Coffee Projects. As such, they involve substantial investments in institution-building, establishment and reinforcement of rural extension services, construction of roads and economic and social infrastructure, research, and a range of technical marketing and policy studies. They reflect lessons learned from previous Bank-supported projects and represent an attempt to respond to the Government's broader approach concerning rural development (para 1.08). Nonetheless, these two projects have experienced implementation difficulties caused by lack of qualified local personnel and of counterpart funds, and by uneven technical assistance performance. These problems are, however, being addressed and implementation has improved significantly. B. The Agricultural Sector 1.04 General. Agriculture is Burundi's most important economic activity. More than 90% of the population depends directly on agriculture which also provides the lion's share of foreign exchange earnings. During the 1970s, agriculture accounted for an average of 60% of GDP, but its share fluctuates with the performance of food crops and coffee production. 1.05 Burundi has a rich endowment for agriculture, but faces mounting pressure on its limited land resources from a growing population. Virtually all good land is cultivated (30% of the total land). Altitude and rainfall vary broadly (11 different natural zones are defined for land classification purposes) and most crops can be cultivated. On the other hand, the hilly topography renders modern farming difficult and has contributed to grave and growing soil erosion problems; control of soil erosion is a priority Government objective. Of land potentially available for agriculture, about 40% is under crops, 31% is used for grazing, 7% is  -3- forest and an estimated 20% is unclassified; the latter remains to be assessed in detail because it consists of steep slopes, temporarily unoccupied grazing land and soils of low quality. Burundi's farmers are primarily smallholders and there are about 700,000 farming units. Because of high population density (160 per km2 -- one of the highest in Africa) farm size is very small at 1 ha. Although all land in Burundi technically belongs to the State and is administered by local authorities (the communes), most land holdings are governed by a combination of traditional law and recent Government regulations. In practice crop land is held by individual families while all pasture land is communally held. Small areas are farmed commercially under more formal individual tenure arrangements. 1.06 Traditional subsistence farming dominates agricultural production; production systems are characterized by high labor intensity and the use of traditional methods and rudimentary tools (mainly hoes and machetes). Farming systems are relatively complex, with substantial intercropping and multiple crops grown on most farms. Farmers use organic fertilizer, as available, but little chemical fertilizer; chemicals are used on coffee and cotton for insect control. Animal traction is largely unknown. Fallow periods, particularly in the highlands, have been decreasing because of population pressure, and this has resulted in overcultivation and soil deterioration. Forest land has steadily declined as farmers bring new land into cultivation. Most holdings have a few trees (either wild or cultivated), growing scattered on poor, unproductive patches or steep slopes. On the average, not more than 5-10 ares per farm (5-10% of farm size) are under trees. Food crops are the mainstay of the subsistance economy and also represent the single most important source of money income for farmers (mainly through the sale of banana beer). Food crops cover about 92% of cultivated land and cash crops about 8%. Traditional food crops include beans, peas, cassava, potatoes and sweet potatoes, maize, sorghum, groundnuts and bananas. Roughly 85% to 90% of the production of these foodstuffs is for farm consumption. Information about production, yields, areas planted, cultivation techniques and producer prices (for the 10% to 15% that is marketed) is poor but important efforts are underway to improve the statistical coverage. Non-traditional food crops include rice, wheat, palm oil and sugar, most of which have been introduced relatively recently. 1.07 Agriculture is the most important source of Burundi's foreign exchange earnings. Its share of foreign exchange revenues has averaged 95% over the past decade, with coffee contributing an average 90%, and cotton, tea and hides accounting for most of the remainder. Due to climatic conditions coffee's production fluctuates considerably from year to year. Most coffee produced is exported. Burundi's coffee export quota, determined by the International Coffee Agreement for 1982/83 and 1983/84, was 25,800 tons. Production has fluctuated between a low of 17,000 tons in 1978 and a high of 45,000 tons in 1982. Cotton production averages 6,000  - 4 - tons/year. Cotton exports have fluctuated between a high of 3,630 tons in 1978 and a low of 800 tons in 1981. Tea production is shared between industrial plantations (1,750 has and smallholders (3,250 ha). The value of tea exports is roughly the same as cotton. Agriculture accounted for about 10-12% of total import expenditures over the last decade, most of it for food items, and only a small part for agricultural inputs. 1.08 Sector performance. The last three Five-year Development Plans (1968-72, 1973-77 and 1978-82) have emphasized increased food crop production and cash crop development to provide foreign exchange resources and additional revenues to the Government. The 1978-82 plan proposed a new strategy aimed at increasing incomes and improving living conditions in the rural areas through a set of measures including food crop and livestock development, erosion control and cash crop diversification. A stronger cooperative movement and the creation of Regional Development Companies (SDR) to implement integrated rural development projects were to constitute the major vehicles for achieving the plan's targets. Actual performance, in agriculture as in other sectors, has fallen short of the plans' ambitious objectives although there has been important progress in some areas. Although statistics are scanty, it appears that food production is barely keeping up with population growth. 1.09 Issues. Burundi's recent agricultural development experience indicates that some important issues must be addressed if the broad strategic objectives of increasing production and improving rural living conditions are to be achieved. These issues can be grouped into three major categories: production, institutional and financial issues. Major efforts to upgrade productivity within traditional farm systems will be essential if agricultural production is to be increased. Priority areas for action are adaptive research and improving the production and distribution of inputs. Agricultural research is a pre-requisite and should focus on measures to control erosion and to improve soil fertility. The responsibilities of agricultural institutions, notably Regional Development Companies (SRDs) and parastatals need to be clarified, their objectives defined and their financial impact on the national budget reviewed. Agricultural service and investment management, including SRD and parastatal management, need to be strengthened. Insufficient numbers of trained, experienced Burundi staff is a key problem, but more explicit attention is needed in planning and implementing services and above all investment projects, to institution building, and measures to ensure that efforts are sustained over time. A host of complex issues turn around financial resources available for agriculture and their management. Agriculture's share in budget resources is low and has declined. Over the past five years the Government has allocated only 10-25% of its investment budget and about 5% of the recurrent budget to agriculture. These allocations are insufficient to support a dynamic investment program and basic farmer services. External financing is not reflected in budget  figures, but even so agriculture's share of estimated total investment declined from 30% in the early 1970s to 15% today. Credit available for agriculture is confined essentially to coffee marketing and processing operations. In short investment is far below what it should be and a key priority is to expand Burundi's ability to plan and manage investments and services for agriculture. An important related issue is cost recovery from agriculture. The Government is committed to encouraging recovery of costs for beneficiaries, to help render services and institutions self-supporting, but a coherent policy and mechanisms to implement this policy have yet to be worked out. C. The Forestry Sub-sector Issues and Strategy 1.10 Forestry development was long neglected in Burundi's development efforts. The most dramatic symbol of the low priority to forestry is the fact that to date there is no qualified national forester in the Forestry Department. Since the late 1970s, however, there has been a sharp focus on forestry issues prompted largely by unmistakable evidence that forest resources were being cut down much faster than they were replaced, with a real threat that most forests could disappear within a decade, with potentially devastating ecological consequences. There has also been a growing appreciation of the links between dwindling forest resources and the decline in soil fertility and soil erosion problems, one of the most critical issues for Burundi's agriculture: as trees become scarcer, farmers have increased use of agricultural residues for fuel instead of using them on crops. Denuded slopes in Burundi's hilly landscape are prone to erosion. The new focus on forestry has led to an impressive investment program, which has been relatively successful in stepping up the rate of plantations established, protecting natural forests and launching a new nursery program to supply seedlings to farmers for individual wood lots. The time has come now, however, to define more clearly Burundi's forest development strategy and to develop the institutional framework which will ensure its implementation. An essential element of this strategy must be an integrated approach to Burundi's forestry sector, including measures to address the critical problem of the place of forestry in Burundi's 700,000 small farms, supply of fuelwood to the growing urban population, supply of timber resources and protection of the environment. An important policy paper, which sets out a strategic framework and specific action measures, has been recently approved by the Government (para 2.14). 1.11 Burundi's forest resources are estimated at about 200,000 ha, or nearly 7% of the total land area. This is a rough estimate because forest resources (like other land use patterns) are not systematically assessed and changes are not monitored. Some important efforts to improve data on forest resources are, however, underway and deserve a high priority in future action programs (paras 2.16 and 2.18). About 140,000 ha of forests are man-made plantations, while 60,000 ha are protected forests, including the remnants of dense tropical mountain forest along the Zaire/Nile crest  - 6 - divide in the northwest (about 40,000 ha), small areas of savanna woodland (15,000 to 20,000 ha) in the south, and forest galleries along water courses and in steep gorges. Forest plantations are owned by the Government, communes, religious communities and individuals. Productivity of plantations varies widely, but annual increments are generally fairly low and overall wood production is far below demand so that total forest resources are declining. 1.12 Information on wood consumption and demand is fragmentary and unreliable, although better data on wood and charcoal sold in the main towns have become available in recent years. Burundi relies on a variety of energy sources, including hydroelectric and thermal power, but wood products remain a vital source of energy, both in rural and urban areas. Both for energy needs and o.her uses of wood, demand, measured both in terms of theoretical requirements and the evidence of markets, far outstrips current supply. Notably, the need for firewood is estimated at a minimum of one million m3/year (assuming a per capita annual wood consumption, low by international standards, of 0.25 m3/year), while current supply and hence consumption is far less (perhaps one half). Projections for future demand show a steadily rising trend; by the end of the century demand is likely to double. Estimated demand for timber is 130,000 m3 (including 2,000 m3 of legal imports), compared to an annual production of 10,000 m3. Per capita consumption of fuelwood and charcoal in towns, especially Bujumbuta, is higher than in rural areas due to the unavailability of alternative fuels such as agricultural residues and higher cash incomes (para 2.09). The rural population is increasingly using agricultural residues to meet their fuel needs and in some areas agricultural wastes are the primary fuel. This contributes directly to declining fertility and crop yields, which in turn increase pressure on the land and leads to shortened fallow periods, soil degradation and falling crop production. 1.13 The wood shortage is clearly reflected in rapid increases in charcoal prices: from BuF 150/bag in 1970 to about BuF 750 in 1984 (BuF 20,000/ton). This price increase (500X) is much higher than the consumer price index increase (250%) for that period. Production cost in 1984 was estimated at BuF 300/bag, including 33% for salaries, 59% for wood purchase and 8% for bags. Inefficient charcoal making is a serious problem as much wood is wasted in its production. This has particular importance because of the growing Bujumbura market for charcoal (Bujumbura, with 5.5% of the population, accounts for 15% of total fuelwood consumption). Charcoal is produced and marketed by small-scale entrepreneurs. Traditional methods are used in its production (yields are around 10%); there are few incentives for economies as charcoal-making fees are based on the number of bags produced and not on area of wood cut or wood volume. Better charcoal production methods have been introduced and tested through the First Forestry Project (para 2.07), but there is an urgent need to promote their application.  1.14 Since 1978 the Government has initiated several large afforestation projects to replenish its dwindling forest resources. During the period from 1978 to 1984 a total of 19,000 ha of new plantations were established with external financing. This represents about 45% of the total area planted between the same dates. While these projects have been well implemented, their impact has been blunted both by the absence of a well articulated, long-term development strategy and by the weakness of institutions in the forestry sector. 1.15 Sub-sector issues and strategy. Continued support for development in the forestry sub-sector is an important element of the Government's development strategy; however, its financial contribution to the subsector is low and a substantial increase from present levels is needed to turn strategic objectives to reality (para 2.05). Objectives are clear: new afforestation projects should contribute to soil erosion control and they should not use land suitable for agriculture; emphasis should shift from large-scale plantations to farm forestry to the extent that farmers interest and technical packages emerge; agro-forestry should be developed to promote farm forestry activities; and state plantations and natural forests should be protected and weil managed. These are priority objectives of the Government Fourch Development Plan (1983-87) (para. 2.14) and it is also at the center of recent policy papers and of the recently approved forestry policy paper (paras 2.15 to 2.18). To achieve these objectives a long-term investment and institution-building program is required. The long-term planting program should include commercial plantations, protection plantations and rural nurseries/agroforestry programs to meet expected demands for different types of woods. Institutional development efforts must focus above all on training of high level technicians at the Forestry Department. Clearer mechanisms for management of existing and future plantations, creation of planning and monitoring mechanisms, and well defined systems for pricing of wood products are also needed. Additional priority areas for action include establishing a comprehensive research program on forestry and agroforestry, improving charcoal production methods, and improving links between forestry programs and other agricultural or livestock programs in high population density areas. 1.16 Previous Bank involvement in the sub-sector. The first IDA Forestry Project (Credit 91S-BU) represented the first stage of a long-term program to develop basic forestry services and plantations supplying fuelwood, building poles, and timber. It was appraised in July 1978, approved by the Executive Directors in May 1979, and the US$4.3 million credit became effective in October 1979. Total costs were estimated at US$8.8 million. EEC financed a US$1.3 million special action credit and FAC the technical assistance. The World Food Program (WFP) also provided food rations to plantation workers in the Project's remote afforestation areas. The project objectives were: to establish rural nurseries in 30 communes, and 2,000 ha of eucalyptus and 5,000 ha of pine plantations; to finance technical assistance for project management and training for local staff; to study energy needs and conduct applied trials, and in particular  to improve charcoal production. Benefits expected from the project were: increased supplies of fuelwood, building poles and charcoal to the local population. The project was due to be completed in September 1984, but completion has been delayed to June 1985, which will allow a full five planting seasons and help to ensure a smooth transition to the second project. 1.17 Project implementation has generally been successful. It is ahead of schedule in several of its physical objectives and most objectives should be reached by the end of the investment period. By June 1985, approximately 6,150 ha will have been planted, including 2,050 ha of eucalyptus and 4,100 ha of pine. About 60 new nurseries have been established, one per commune and are operating, presently selling 25,000year of seedlings to farmers. Problems encountered by the First Forestry Project (para 2.07) have included limited land availability at some of the plantation sites (Mageyo), resentment by the local population and temporary labor shortage in others (Vyanda), lack of candidates for training (at the university level) and inadequate and unreliable Government funding (replenishment of revolving fund). The chronic shortages of Government funds were overcome by the estab'.ishment of a revolving fund financed by credit funds. 1.18 Sector and sub-sector lending strategy. The key challenges for agricultural development in the decade ahead are to tackle the issues of changing farming systems in the face of mounting population pressures, and to develop Burundi's capacity to plan and manage services and investments. A dynamic research program geared to small farmer needs and closely linked to farmer services, notably extension and input supply, is a top priority. Agroforestry, the integration of trees in farming systems, has been neglected but will play a key role. Institution-building also poses vital problems; well-focused training programs are essential and care must be taken to ensure that programs and services are tailored to available financial and managerial resources. Burundi's high (ard increasing) population density poses special issues which must feature in agricultural strategies. This dilemna of rising population pressure on land must guide research and extension programs. Host urgently it dictates a dynamic action program to address the related problems of dwindling forest resources and soil erosion and exhaustion of soil fertility. An integrated approach, starting at farm level with soil conservation, better use of inputs and judicious use of trees on the farm, but also including production plantations wherever land is available and good management of plantations and systematic protection of natural fcrests, is needed. The proposed project is designed to support such an intrgrated approach to the forestry sub-sector and thus has an imp4.ctant and central role in the Bank's operational strategy for Burundi.  CHAPTER II GOVERNMENT SUPPORT AND POLICY FOR THE FORESTRY SUB-SECTOR A. Forestry Institutions 2.01 General and local administration. The Forestry Department ("Eaux et Forfts") is one of four departments of the Directorate General of Agriculture, in the Ministry of Agriculture and Livestock. The Department's director is also responsible for fisheries and wildlife, and parks. The department consists of two sub-directorates: forests and fisheries. The responsibilities of the sub-directorate of forests include: implementation of the Government's forest policies, management of reserved natural forests, management of man-made plantations, control of exploitation of communal plantations, and management of Government-control sawmilling. Forest reserves became the responsibility of the National Institute for Nature Conservation (INCN) which falls directly under the Presidency, by the decree of May 3, 1980. The Department nonetheless maintains a general mandate for coordinating all forestry activities, including natural forests. The Forestry Department staff includes about 100 civil servants (52 technicians and 13 assistant technicians) representing 20% of the personnel of the Ministry of Agriculture - excluding contractual employees. The department's current organization is not adapted to the major challenges it faces in implementiAg forestry policies and programs, notably, responsibilities of various units are not clearly defined, and the field level organization is limited and weak. The forest policy paper addresses this issue, and calls for a reorganization of the department, both at the central and field level. The proposed project would support this reorganization and institution-building effort (para 4.01). 2.02 The number one problem for the Forestry Department is lack of qualified staff. Burundi has no trained high-level Burundi forester and there is a marked shortage of lower level staff; there are, however, adequate numbers of trained mid-level staff. The Department has no deputy-director and some services are currently run by technical advisers. The high level staff are all agricultural engineers with no forestry qualifications. Thirteen provinces out of 15 have one or two forest agronomes-, the remaining two provinces sharing the services of five assistants. Forest "agronomes- all come from the Burundi Technical Institute of Agronomy (ITAB). Assistant foresters come from the Kigamba School, while guards and monitors are trained on the job. Another important problem facing the Department is a severely limited means of transport. Most equipment is old, inadequate and poorly maintained because of lack of funds. Field staff have barely any means of transport. 2.03 Efforts to strengthen forestry extension are badly needed. At present the forestry extension service is small and narrowly focused, although national tree planting campaigns and individual actions are having  - 10 - a valuable impact. Cooperation between the forestry and agricultural extension services is limited. The Forestry Department has tried to promote tree planting, but lacks a well-defined plan and has no means of monitoring progress. Serious understaffing has been one of the main problems, and many communes are without an assistant forester. Training and supervision of extension workers are generally inadequate and little effort has been made so far to teach the rural population to plant and manage woodlots, or how to save cooking fuel. Additional financing will be required to have a significant impact through training programs, e.g., by making greater use of audio-visual and othez methods. 2.04 Forestry research has been almost non-existent until recently; however, work on several forest research topics, including agroforestry, is expected to start soon. Forestry research is the responsibility of the National Agricultural Research Institute (ISABU), which depends directly on the Ministry of Agriculture, but it is only a minor activity at the Institute. The orientation of work exclusively toward cash crops (coffee, tea and cotton) is being modified to include more research on food crops and forestry. Trials undertaken over the past 20 years are currently at a standstill for lack of personnel. The only ongoing forestry Zaire-Nile Divide Mission (MFCZN). This project was initiated in 1976 and is financed by French bilateral aid (FAC). FAC is also planning to start an agroforestry research program shortly, as well as the funding of a research specialist and equipment. A large number of trials have been carried out in different ecological zones, covering many important topics. Among the problems requiring specific attention are live firebreaks, leguminous crops, the effects of fertilization on cultivation, the preparation of volume tables and the study of production costs. The FAC project will tackle these problems and, in the short and medium term this represents an effective approach to addressing priority forestry research issues. 2.05 The forestry sub-sector has been relatively privileged in terms of investment levels in recent years, and a number of large investment projects have been launched. The key problems are poor coordination among project efforts, the absence of a coherent, carefully vetted investment program for the sub-sector, and chronic difficulties in providing Government counterpart funds for these investments. An additional, related problem is the very low level of resources going to support recurrent costs of the forestry services, which seriously impede efforts to strengthen management of forestry policies and programs and has implications for the sustainability of programs launched. We conclude that Government budget allocations to the sub-sector have been insufficient to meet the challenge of developing forestry resources and operating on-going programs and services well, and substantially more resources are needed if forestry objectives are to be realized. The share of forestry in recurrent budget expenditures has varied little over the past decade (the share of agriculture has ranged between 3-5% during the same period). The recurrent  - 11 - budget is almost entirely absorbed by salaries and wages. Of the BuF 18 million (US $150,000) allocated in 1983, BuF 10 million and BuF 8 million were earmarked for contract staff and regular staff, respectively. The share of forestry in investment budget expenditures is also small, representing less than 10% of the total investment budget for agriculture (BuF 54 million; US$450,000 in 1983). These figures, however, have limited significance because almost all investments are carried out under donor financed projects (which do not figure in the budget) to which the Government adds a small contribution. B. Forestry Projects 2.06 Current forestry projects. During the Third Five-Year Plan period (1978-82), major afforestation projects established approximately 12,600 ha of new plantations: between 1978 and 1984 the total stood at 18,800 ha. Area Planted 1978-84 (ha) Timber Fuelwood Protection Total - Externally-financed Projects 8,500 2,000 8,300 18,800 - Government-financed Plantations 4,400 5,600 - 10,000 - Communal Plantations 100 11,800 - 11,400 13,000 19,400 8,300 40,700 A number of other forestry projects are currently underway in Burundi, including the IDA-financed project (para 1.16). Belgium and Saudi Arabia are co-financing an afforestation project over 12,000 ha in Mugamba-Bututsi, in south-central Burundi. The European Development Fund (EDF) will finance the afforestation of 3,200 ha at Mumirva and 6,000 ha near Gihofi. FAC is financing the planting of a 500 ha protection belt around the Zaire-Nile divide in northern Burundi. Two IDA rural development projects (paras 1.03 and 1.04) have forestry components. An FAO-UNDP project (which was jointly carried out with the First IDA-financed Forestry project) has already trained approximately 35 forestry technicians. Most of these projects have a regional rather than a sectoral focus and they generally concentrate on tree-planting rather than on extension work, conservation programs or training. The communes have been engaged in afforestation programs undertaken in the context of communal works, with mixed results. National, essentially public relations campaigns have been initiated, such as the proclamation of 1979 as "l'Annee de l'Arbre' (tree planting year), in an attempt to reduce forest destruction and to promote individual tree planting. An important element of this national effort has been establishment of Government nurseries to produce seedlings for sale to the population. Public response has been favorable and the demand for seedlings has grown regularly as the number of nurseries has increased.  - 12 - 2.07 The First Forestry Project (para 1.16), is described in Chapter I. Some specific issues, however, merit further discussion since they have implications for design of the second Project. Training of technicians was an important component; these activities were successful but need to be expanded at the national level. 1While training efforts at first focused on mid-level technicians, emphasis now needs to be put on training higher level technicians. Another important issue is project management arrangements. Project management was handled expertly and efficiently, but it is noteworthy that it was handled mainly by expatriates. FAC financed a technical assistance team from the Centre Technique Forestier Tropical (CTFT). In addition to their main plantation activities these expatriates have provided valuable on-the-job training to national project staff. However, because of the shortage of Burundese agronomists, transfer of responsibilities to the local foresters at the Project management level was slower than planned and concerted efforts to ensure that management responsibility is assumed by Burundi nationals are needed in the next few years. Better charcoal production methods have been introduced under the First Forestry Project and traditional fuel-conserving stoves are being experimented with. Traditional charcoal yields are around 10%: efficient methods can double this figure. The challenge is now to pursue experimental demonstrations to find ways to transfer new technologies to charcoal-makers. Generally the rural nurseries have been a success but for logistic reasons their number had to be doubled while the average annual production per nursery was halved. This led to a higher average production cost of seedlings than expected. Obviously steps should be taken to ensure this is not repeated as the program is expanded to cover Burundi's 90 communes. The changes introduced were intended to mitigate local hostility to the planting of large tracts of trees on the extensive grazing lands, particularly in the southern part of the country, at Vyanda. This laudable objective can in future be better met by careful siting of nurseries. C. The Quest for Wood 2.08 It is clear that wood fuel will remain the main source of energy for household purposes for some time to come; this is because of the high cost of imported fuel and the low purchasing power of the population. Therefore it is vital that efforts focus on increasing wood supplies and the efficiency of their use rather than on substitution for wood products. Consumption of timber is relatively limited in comparison to fuelwood, as are official wood imports. The need for timber is nonetheless important and need to be addressed. 2.09 The rural population uses firewood and agricultural residues to cook food and for their other energy needs. Agricultural residues cover an estimated 15% of energy needs. The fuelwood scarcity has an impact on food crop production as the rural population increasingly uses agricultural residues to meet fuel needs. This reduces soil fertility and crop yields,  - 13 - which in turn leads to shortened fallow periods, soil degradation and falling production. Deforestation resulting from the energy needs of the population, without alternative energy sources, is accelerating the process of soil erosion and there is good reason to fear that consequences of this deforestation are worsening. Marketed firewood supplies the principal towns and rural institutions such as missions a-d hospitals, and rural industries such as tea factories. A recent survey showed that 87% of households in Bujumbura use purchased charcoal as their main cooking fuel, the remainder using mostly purchased wood. According to the same survey, the amount spent per month per person on charcoal for :ooking was about BuF 250 (US$2). This is equivalent to 10-15% of the amount spent on food. The main potential substitutes for charcoal and firewood are electricity, imported kerosene and domestic peat. In terms of useful energy for cooking, electricity costs roughly the same as charcoal, and higher income urban families are using it; however, the cost of electric appliances is too high for most of the population and the prospects for their reduction through bulk procurement and/or local manufacturing for a regional market are not very promising at least for the near future. Even without taking the cost of kerosene stoves into account, the cost of imported petroleum products (US$85/barrel c.i.f. Bujumbura) prohibits the use of kerosene for cooking. Efforts have been made to promote domestic peat as a cooking fuel, but consumer acceptance has been poor. This is due to: (i) the fact that people are accustomed to cooking with charcoal; (ii) difficulties in peat combustion; (iii) the high ash content of the local peat; and (iv) the price of peat not being competitive with that of charcoal. Household use of peat may become possible in the future once the peat industry has developed further. The conclusion is inescapable that Burundi will remain dependent on wood for the foreseeable future. 2.10 Consumption of timber is small by comparison to fuelwood (130,000 3 in 1983). Bujumbura accounts for about 45% of timber consumption. Most of the wood comes from the Zaire/Nile Crest divide (103,000 m3) while the rest comes from State and communal plantations (25,000 m3). Wood is also widely used as a fuel in industrial or small-scale operations (e.g. tea processing, brick making, bakeries, lime kilns and fish drying). 2.11 Official wood imports are limited: there may be some unofficial imports but in very small quantities. An equivalent of US$300,000 in wood products are imported annually; this represents about 0.2% of the value of all imports in 1982, and 2,000 m3 of sawn timber. These imports cover about 1.5% of the present demand. There is some potential for imports of construction timber or wood products from Tanzania and Zaire. This might assist Burundi in meeting critical short-term wood and fuel requirements, but the costs of imported wood products would be high and it seems unlikely that long-term requirements could be met through imports. Considerable quantities of metal are imported and substitute for wood in the modern construction sector. Petroleum products represented about 15X of the value of imports in 1982; the percentage for metal products is unknown. Burundi exports virtually no wood or wood products.  - 14 - 2.12 Marketing and pricing. Most charcoal is produced and marketed by small scale entrepreneurs. Wood is cut and charcoal produced in situ. Production techniques employed are exclusively traditional (burning in earthen pits), and they are inefficient. The general scarcity of wood has resulted in periodic shortages of charcoal, with sharply increasing prices. The Government, concerned by illegal forest cutting and the reduction in forest area, has restricted charcoal production in many areas. Thus, charcoal is coming from increasingly afar, with consequent increases in prices (para 1.13). Wood for firewood and poles is normally marketed by small entrepreneurs who employ woodcutters. If trees are ou government or communal land a cutting permit and payment of a communal tax are normally required. Timber logs are normally pit sawn by skilled sawyers and the market price is relatively high (BuF 18,000/m3 of sawn wood at Bujumbura in 1983). Burundi's only 'modern' processing equipment is an old portable band-sawmill at Bugarama. This mill breaks down frequently due to inadequate supervision, poor operation and lack of spare parts. Its annual production is a small fraction of its rated capacity of 500 m3 per annum. It processes mostly cypress thinnings from trees 15 to 20 years old. 2.13 At present the pricing of wood and wood products in Burundi is not well organized. The Government has not yet defined a coherent policy on pricing of forest products, but is taking steps to articulate and implement one (para. 2.16). The problem is complex. For trees on private land", including all trees planted by farmers, no tax or payment is involved, and by tradition farmers gather and cut wood wherever it-is available. In principle utilization of areas of natural forest, all Government plantations, and commanal plantations is strictly regulated. and either a permit fee or a communal tax is payable. In practice it is difficult, if not impossible, for the Government to control cutting of trees and collect fees except on large plantation blocks. Cutting permits range from BuF 1,800 to BuF 2,500 per lot of 10 to 15 trees. This is well below the cost of replacing trees and the level of permit fees is clearly too low. The Government, however, hesitates to increase rates in the fear that it would increase illegal cutting and marketing of forest products. The staff of the forestry department is limited and plantations are widely scattered, with many in such poor condition that they scarcely warrant control efforts. Illegal cutting is for all these reasons widespread. The situation is complicated by confusion of records, so that ownership of many small plantations and tree stands is uncertain. Further, there are many unknowns about what is actually happening, as no effective monitoring system exists. It is very clear that forest resources are dwindling and there is mounting demand for wood products, but it is difficult to quantify exactly how much is being cut or to project trends with any real accuracy. This complex of issues is being addressed in the Forestry Policy Paper, and the Government is committed in principle to rationalizing the present situation and developing a more effective cost recovery system for forest production (Chapter VI).  - 15 - D. Government Strategy and Policy for the Sub-sector 2.14 Strategy. The Government has launched large afforestation programs since 1978 (para 1.14) to arrest the pace of deforestation, and minimize its consequences. Objectives are broad and cover not only forest resources per se but also related issues of soil fertility and soil erosion control. These programs are expected to continue under the 1983-87 Fourth Plan period. The objectives for the Government's forest policy, as stated in the Plan, are as follows: (i) The protection and management of existing natural forests: protection against encroachment, brush fires and grazing animals will be assisted by the creation of a buffer zones. (ii) In new projects priority will be given to afforestation by the state of hilltops and hillsides unsuitable for agriculture and of low pasture value, mainly for soil and water conservation. Ciii) Communal plantations will be of sufficiently large size to ensure the best use of available staff. (iv) Individual and collective plantations will be supported by rural nurseries, supplies of seedlinxgs and good extension staff. (v) Agro-forestry in high density population areas will be improved. (vi) The introduction of anti-erosion hedges_on pastures will be complemented by the introduction of soil-improving and fodder trees. (vii) Wood uses. A study will establish wood needs, local, regional and national market projections for firewood, poles and timber, and the organization of the wood industry. Planned allocations to the forestry sub-sector are BuF 2.4 billion, or 2% of the total and 10% of the share of agriculture. This would represent, in annual terms, six times the 1983 Government allocation to the sub-sector. These financial objectives are somewhat over-ambitious, as are the overall afforestation objectives (para 2.15); even a partial achievement of objectives will depend on the willingness of external agencies to finance the lion's share of the program.  - 16 - 2.15 The Burundi Government has stated as its long-term objective that it will seek to extend the forest cover to 500,000 ha, or 20% of the total area of Burundi. Afforestation is considered the best way of protecting the land in an agro-silvo-pastoral context. New forestry projects are planned in areas of low demographic pressure and on land with no clear potential for agriculture or livestock. Large scale plantations, which will cover some 30,000 ha by 1987, are thus to be geared to the dual objectives of protection and production. Pine plantations will produce timber and firewood, callitris will be useful for protection and will produce fuelwood, with some timber; and eucalyptus will produce mainly firewood and timber. To reduce the risks of large-scale monoculture, silvicultural research will aim to broaden the range of useful species. 2.16 Policy. With assistance from bilateral and multi-lateral agencies (FAC and EDF in particular), the Government has ix'tiated important efforts for the development of the forestry sub-sector based on an integrated development of agriculture, livestock and forestry. Over the past six years a series of policy documents have been prepared and reviewed and action to implement them has begun. Within the past year a broader policy document (summarized below in paras 2.17 and 2.18) has been drafted for the purpose of defining clear, well articulated Government forestry objectives and providing a rational framework for future projects, including those financed by external donors. This document (which has now been fully endorsed by the Government) is the first comprehensive document on Burundi's forestry sub-sector. 2.17 The Forestry Policy Review provides a good overview of the complex problems of forestry and proposes a strategic framework for addressing them. It is broad ranging in scope, addressing both issues of wood supply in all its forms (sawnwood, poles, fuelwood, charcoal) and threats to the environment (erosion, disappearance of natural fauna and flora). It underlines the importance of dynamic, active measures to increase wood supplies and protect the environment, arguing that action here is a key to Burundi's future economic independence. Proposals for action cover the broad policy environment, where reliance on private initiative is emphasized and the importance of cost recovery; they also include institutional measures and reorganization proposals (reinforcement of the Forestry Department and better linkages between forestry and crop and livestock development), and finally a framework for investments in forestry services, plantations, and wood processing and marketing. The review underlines the key importance of measures to ensure adequate incentives, given the critical importance of private farmers and entrepreneurs throughout the chain of wood-related activities, from tree growing to charcoal and timber marketing. Some key themes evoked in the review, which represent an agenda for action in the next few years, include  - 17 - (a) the need to improve knowledge of resources available and comsumption levels; (b) the need for more efficient management of existing resources; (c) measures aimed at increasing productivity of plantations and forests; (d) more dynamic efforts to encourage farmers to grow trees, including research and better extension services; (e) development of private wood processing, including provision for credit; (f) development and promotion of more efficient wood processing, especially for charcoal; and (g) more effective conservation programs aimed at preserving natural forests and plant and animal life. Key elements in the strategy will include training of personnel, development of effective and relevant research programs, and establishment of mechanisms to apply laws and policies affecting forests. The Review recognizes clearly that the challenge is to make the best use of existing resources, as resource constraints will be a continuing reality. 2.18 The proposed Project is thus set against a backdrop of an intensive review of policy issues for forestry, which sets out a clear and comprehensive development strategy and an ambitious agenda for policy analysis and action. It also follows the recent approval of a new forestry law, which was based on needs identified through earlier policy reviews and which sets out a better legal framework for sector management. The challenge now is to translate policy and law into action. The Policy Review has now been finalized and approved by the Government (including the President). It has benefitted from important contributions by several external agencies involved in the sector, including IDA. Further discussions on implementation of the proposed measures and periodic review of progress and assessment of the need for adjustments would be pursued in the course of implementation of the Project. Specific areas and issues to be addressed include the institutional arrangements for improvement of sector management (e.g. reorganization of the Forestry Department), training programs, application of cost recovery measures and definition and implementation of regulatory measures related to the new forest law. In addition, an annual review of the proposed investments in the forestry sector would be conducted. Assurances to this effect were obtained at negotiations. The proposed Project has been designed explicitly to support implementation of policy and institutional measures, and the institution building and pilot agroforestry and research components, in particular, are key to finding long-term new solutions to the forestry dilemma.  - 18 - III. THE PROJECT A. General Description 3.01 The proposed Project would be the second phase of a long-term program to develop basic forestry services and promote tree planting by farmers, and to establish plantations supplying fuelwood, building poles and timber. The Project would support the Government's strategy for forestry development, notably in speeding the pace of reforestation on individual farms and in larger scale plantations. It aims at laying the groundwork for future development of agroforestry. The Project has important institutional development objectives, notably in strengthening the Forestry Department, particularly its capacity to manage investment programs and staff training. It includes support for new research programs in agroforestry and pilot schemes to promote an integrated sylvo-pastoral approach, and training of charcoal makers. The Project would provide support for the continuing efforts to establish long-term pricing and marketing strategies and mechanisms to implement them. Thus the Project proposes to address important issues which have come to light in the course of implementation of the first project, particularly in the financial, institutional (management and training) and policy areas (paras 1.16 and 2.07). The proposed Project would comprise, over a period of five years, the following components: (a) development of 90 Government-run rural nurseries (about one per commune) including future development of 60 existing nurseries and establishment of about 30 new nurseries, each producing 40,000-50,000 seedlings (3-4 million altogether) per year; development of forestry extension services and a pilot agroforestry scheme; and establishment of 2,000 ha of demonstration fuelwood, pole, fodder and fruit plots in communes throughout the country (5 ha per year close to each rural nursery); (b) establishment by the Government of: - 3,500 ha of eucalyptus fuelwood and pole plantations at Gakara, Bukinanyana and Mabanda, mainly to supply charcoal to Bujumbura; and - 1,500 ha of pine timber commercial plantations at Vyanda and Vugizo; (c) the protection, maintenance and silvicultural treatment of existing plantations established under the Phase I project (about 2,000 ha of eucalyptus and 4,000 ha of pine); (d) the initiation of a program to promote integrated sylvo-pastoral activities, on a pilot basis, in a few selected plantation areas with high density population;  - 19 - (e) institutional strengthening of the Forestry Department through the creation of a planning and monitoring unit, and two new regional forest inspectorates, includ