Burundi - Forestry Project

Document of The World Bank FOR OFFICIAL USE ONLY ! Report No. 7839 PROJECT COMPLETION REPORT BURUNDI FIRST FORESTRY PROJECT (CREDIT 918-BU) JUNE 22, 1989 Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit - Burundi Francs (FBU) US$1.00 a FBU 90 (at appraisal) FBU 1 * US$ 0.011 WEIGHTS AND MEASURES Metric - BritishlUS Eauivalents 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 miles 1 square kilometer (kmz) 0.39 square miles 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) 8 2,204 pounds (lb) ABBREVIATIONS AIDR : International Association for Rural Development c : circa (approximately) CTFT : Centre Technique Forestier Tropical (French Tropical Forest Center DUB : Urban Development of Bujambura Project EEC : European Economic Community FAC : Fonds d'Aide et de Cooperation (French Aid & Cooperation Fund) FAO : Food and Agriculture Organization GOB Government of Burundi IDA International Development Agency ITAB Agricultural Technology Institute of Burundi MAI Mean Annual Increment NGO Non-Government Organization SAR Staff Appraisal Report UNDP : United Nations Development Program USAID U.S. Agency for International Development WFP : World Food Programme Fiscal Year Government of Burundi: 1 Jan. - 31 Dec. IDA : 1 July - 30 June.  FOR OMCIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. OEmce cd ODectkwG'w'a Opiratomi Ivakatrnn June 22, 1989 NNORANDUN TO THE EXCUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Burundi First Forestry Project (Credit No. 918-BU) Attached, for information, is a copy of a report entitled 'Project Completion Report on Burundi - First Forestry Project (Credit No. 918-BU)" prepared by the Africa Regional Office. No audit of this project has been made by the Operations Evaluatior- Department at this time. Attachment This document has a restricted distribution and may be Used by recipients only in tho performance of their official duties. Its contents maY not otherwise be disclosed without World Bank authorization.  FOR OMCIAL USE ONLY PROJECT COMPLrTION REPORT BURUNDI FIRST FORESTRY PROJECT (Cr. 918-BU) Table of Contents Page No. Preface .............................. O............ .......... .. i Basic Data Sheet . ............... . ..................... . .ii Evaluation Summary ............ .. .......... . .. ..... . . ... v PROJECT COMPLETION REPORT I. INTRODUCTION .............................. 1 II. FORESTRY SUBSECTOR ....... ............ ... ............ .... 2 A. Resources and Consumption ....... ............. 2 B. Forestry Department . ............... .. ............. 2 C. Forest Policy ........ ....... ..... ................. 2 D. Future Energy Needs . . .............. .... .......... 3 E. Alternative Energy Sources ..... ................... 3 F. Ongoing Forestry Projects ..#* .. ......*.......... 3 III. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL ..... 4 A. Origin ..# .................................... ............. 4 B. Appraisal Issues ......... ....... ............. 4 C. Targets and Goals ... ............. ................. 5 D. Board Approval ...... . ...... ... ..... ........ ...... 5 E. Benefits ....................................... .............. 6 F. Risks ........... ................. .......................... 6 IV. IMPLEMENTATION ...................... ....... ..... 7 A. Dates of Effectiveness and Completion ............ o 7 B. Main Components ........... ...................... .. 7 Rural Nurseries ............. .................. .7 Eucalyptus Plantations ......................... 9 Pine Plantation ... .............................. 11 Training Component ............................ 13 Studies and Trials .......... ... ...... .......... 14 Project Management Support . ..................... 15 C. Theft ...................................................... 15 D. Revolving Fund ...... . . .. . . ...................... ..... . 15 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Table of Contents (Cont'd.) Page No. V. OPERATING (BANK, FINANCIAL, INSTITUTIONAL) PERFORMANCE **...*...... too .......*...........o.... 16 A. Supervision ......I.... 4...0....0 ........0..... 16 B. Compliance with Credit Terms ...................... 16 Co Costs and Procurement ... ...... ....... ....... ....... 16 Do Expatriates ..... .o .. .......o...o........ 18 VI. ECONOMIC RE-EVALUATION ... . ................. .... o..#.. . 18 VII. ENVIRONMENTAL IMPACT ..... o..... .......... .....* .. .. *.. 18 VIII. ISSUES AND RECOMMENDED FOLLOW-UP ..... ............ o...o 19 A. Rural Nurseries ... . .... .......... ........ ..... ....... 19 B. Eucalyptus and Pine Plantations .......... o..... ... 19 C. Training ....................... * ...... ......... ............. 20 D. Studi68e******es ..o ... o...........o.o^^@@s ...... 20 IX. BURUNDI FORESTRY PROJECT PHASE II .................. 21 ANNEXES 1. Project Organization Chart ........................ 23 2. Key Indicator Table ...... _ .. ...... ... ....o.... 24 3. Economic Rate of Return Re-evaluation ............... 25 4. Files (Documents Issues) .............................. 29 COMENTS FRO THE BORROWER ANNEX 5: Comments from Ministere de l'samnagement, du tourisme et de l'environnement ................ 33 ANNEX 6: Comments from United Nations Development Programme .. . 35 MAP: IBRD 13869R: Forestry Project  PROJECT COMPLETION REPORT BURUNDI FIRST FORESTRY PROJECT (Cr. 918-BU) PREFACE 1. This project, the first in the forestry sector in Burundi, was identified in May 1977 and appraised in June-July 1978. Credit 918-BU for US$4.3 million equivalent was approved by the Board in May 1979 and became effective in October. The credit was closed in September 1985, more than one year later than scheduled, and was fully disbursed. 2. This Project Completion Report (PCR) was prepared by the Africa Regional staff and is based, inter alia, on the Borrower's PCR, the Staff Appraisal Report, the Loan Guarantee, and Project Agreements, supervision reports, correspondence between the Bank and the Borrower, and internal Bank memoranda. 3. This PCR was read by the Operations Evaluation Department (OED). The draft PCR was sent to the Borrower and to cofinanciers on February 23, 1989, for comments by April 13, 1989. Comments received are attached as Annexes 5 and 6. The Borrower's comments have been fully integrated into the text.  - ii - PROJECT COMPLETION REPORT BURUNDI FIRST FORESTRY PROJECT (Cr. 918-BU) BASIC DATA SHEET Actual or Appraisal current Key proiect data estimate estimate Total project costs (US$ million) 8.83 8.83 Underrun or overrun (Z) - - Credit amount (US$ million) 4.3 4.3 Disbursed 4*3 4.3 Date physical components completed Civil works March 1982 1984 Land development June 1984 March 1985 Proportion completed by above date (t) Civil works - iooz Land development _ 87Z Proportion of time underrun (-) or overrun (X) Civil works - 82% Land development _ 15% Economic rate of return (X) 20? 15? Eucalyptus fuelwood plantation 17? llZ Pine plantation 22? 172 Financial performance Good Institutional performance Good Actual or Original estimated Other proiect data plan Revisions actual First mention in files 10/30/76 Government's application - 10/25/77 Negotiations 03/26/76 mid-April 04/16/79 Board approval 05/24/79 Agreement date 06/01/79 Effectiveness date 09/05/79 10/22/79 10/22/79 DCA closing date 03/31/85 09/30/85 09/30/85 SAA (EEC's Spec.Action Acc.) closing date 12/31/82 12/31/83 12/31/83  - lli - Cumulative disbursements. IDAIEEC 1980 1981 1982 1983 1984 l9S5 IDA Appraisal estimate (USS million) .25 1.80 2.80 3.80 4.20 4.30 Actual disbursement (US$ million) .50 1.13 2.16 3.20 4.00 4.30 Actual as 2 of estimate 200 63 77 84 95 100 Date of final disbursement September 30, 1985 Appraisal estinte (USS dllion) .15 .40 1.20 - Actual disbrisemt (USS mlllion) .30 .48 .78 .90 Actual as % of estimte 2C0 120 65 Date of final disbirse'ent De,er 31, 1983 STAFF INPUTS (staff weeks) FYI8 FY79 FY860 FY81 FY82 FY88 FY84 FY8C FY88 FY87 FY88 TOTAL Preappral.al 9.9 .1 16.1 Approlsal 2.8 16.1 20.4 Negotiations 5.1 5.1 Supervision .7 10.1 6.1 9.4 6.3 19.6 1.2 2.8 1.4 48.6 Other .1 .4 .6 Subtotal 12.2 24.1 10.4 6.1 9.4 6.8 10.6 1.2 2.8 1.4 84.5 W.ssion data No. of Staffdays Spedalizationm Perforiunc Types of Mission Date person in field represented atrR Truid problems Identification 10/20/76 1 1 f 05/11/77 1 4 f 05/18/77 1 10 f P ation 01/16/78 1 1 o 03/13/78 2 28 f Appraisal 06/27/77 2 22 e, f Sub-total 66 Supervision 1 09/21/79 2 14 e, f 1 2 M II 03/02/80 2 12 e, f 1 2 III 10/29/80 2 14 e, f 1 2 M IV02/23/81 1 5 f 1 1 V 09/29/81 2 10 e, f 1 2 VI 03/27/82 1 7 f 1 2 VII 11/20/82 1 7 f 1 2 T VIII 05/19/83 1 4 e 1 2 T, F IX 05/22/84 1 9 f ' 2 T X 03/18/85 2 8 e, f 1 2 T Sut-total 90 TOTAL :156  - iv - Infomatim on Borrwer Borrwer Republic of Burund! DCA executing agency 1epar3ls of Waters and Forests (DWF), Mnistry of Agriculture and Livestodk SAC exeZting agency MA as Adinistrator estdblished with fwxi coniribnxd by the i Mber States of the Eurpean Ecwicdc Camumity Fiscal year of borrower Jamary I to Der 31 Currency exchange rate Name of currency Burundi Francs MFBu) Appraisal year average 1979 US $1.00 - FBu 90 Intervening years average 1980-85 US $1.00 a FBu 120 Completion year average 1986 US $1.00 - FBu 114 Follow-on project Name Seco ' Forestry Project Credit number 162(, 'J Amount (USS million) 12.8 Agreement date 08/13/85 /a Actual number of days worked on project. /b e - economist, f - forester; o - loan officer /c 1 - problem free or minor problems; 2 - moderate problems; 3 - major problems. j 1 - improving; 2 - stationary; 3 - deteriorating. /e F - financial; P - political; M - managerial; T - technical. Most critical problem from left to right.  PROJECT COMPLETION REPORT BURUNDI FIRST FORESTRY PROJECT (Cr. 918-BU) Evaluation Summary Introduction 1. The Burundi First Forestry Project, Credit 918-BU, was cnfinanced by the Government, IDA, FAC, EEC and the UNDP. Obiectives 2. The project had two main objectives: to strengthen the Forestry Department, and to promote tree planting. Implementation Experience 3. The project was implemented by the Forestry Department (FD) between 1979 and 1985. The physical targets were largely achieved. The quality of management and of output were good, and the project was rated '1-' throughout. The main implementation problems were occasional late availability of counterpart funds, land conflicts with the population, and the fact that the project was mainly being implemented by expatriates. The j latter two problems will be further dealt with under respectively 'Findings' and 'Sustainability". Results 4. To strengthen the Forestry Department, the project, with the help of UNDP technical assistance, trained some 80 middle-level forestry technicians at a local training institute. On the higher or professionrl level, the project was less successful, providing only one two-year scholarship to the Forestry Faculty at the University of Antananarivo in Madagascar, whereas four other scholarships were unused due to lack of candidates. The project also financed vehicles and a headquarters building for FD. 5. As regards tree planting, the project established 81 nurseries which, during the implementation period, produced seven million seedlings for sale in rural areas. With these seedlings, 2,700 ha of communal and 1,100 ha of individual plantations were established, which, over the next three decades, can be expected to produce more than a million cubic meters of wood, besides soil protection and other environmental benefits. The project also established industrial plantations: 2,000 ha of eucalyptus, mainly for charcoal and building poles, and 4,000 ha of pine, mainly for sawtimber. The industrial plantations can be expected to yield about two and a half million cubic meters of wood over the next three decades.  - vi- Sustainability 6. The project's main sustainability problem during the first phase was the fact that it was mainly being implemented by expatriates. However, in December 1988, about half-way through the second phase, the project was 'Burundizedu, the three remaining expatriates being nomiaated advisers. 7. Although the seedlings from the project nurseries are being sold to communes and individuals, not distributed free as in many other countries, the sales prices do not cover the cost of production, nor is it likely that such complete cost recovery can be achieved, except in the case of seedlings sold in urban areas. However, if FD becomes more active about marketing the output from its industrial plantations, the revenue, even though it would go to Treasury, could offset the cost of maintaining the nurseries. Findings 8. Obtaining land for the industrial plantations in a densely populated country like Burundi occasionally led to conflicts with farmers and pastoralists. For that reason, in the second phase of the project which is now being implement'd, the area of new industrial plantations has been reduced, while there is a greater emphasis on agro-sylvo pastoralism.  PROJECT COMPLETION REPORT BURUNDI FIRST FORESTRY PROJECT (Cr. 918-BU) I. INTRODUCTION 1.1 The Burundi Forestry Project Credit Agree_ent 918-BU was signed in October 1979. The Project was co-financed by IDA (US$4.30 million), FAC (US$1.86 million), EEC (U$1.2 million), the UNDP (US$0.70 million), and the Government of Burundi (US$0.77 million). The Project's estimated costs thus totalled US$8.83 million. The Project was the first stage of a long-term program to develop basic forestry services and plantations supplying fuelwood, building poles, and pine timber. The original components were: (i) establishment of 30 communal nurseries to produce eucalyptus seedlings; (2) the development of 2,0000 ha of short-rotation eucalyptus plantation; and (3) the development of about 5,000 ha pine plantation. Additional components were technical assistance, studies and training. 1.2 Overall there were no major revisions, and implementation differed little from the original proposals. The Project achieved nearly all of its original objectives. Especially well established is the rural nursery component. It created over twice the number of nurseries originally planned, although the numbet of seedlings per nursery was reduced by half, so the total number of seedlings sold approximately equalled the original target. Noteworthy is the success the Project had in its ability to sell, rather than merely give away, its seedlings. The Project was least successful in achieving its target of training Burundi nationals as professional foresters to take over the project management. 1.3 The Project was designed to address three long-term forestry development issues: a. The need for trained Burundi foresters. At the time of the Project's inception Burunei had only one trained forester. The Project achieved moderate success in addressing this issue by training 80 middle-level foresters. However, the goal of training enough high-level (professional) foresters in five years to replace the Project's e..patriate experts proved to be over-ambitious. b. The need for a forest policy, and for a price structure reflecting the real cost of forest products. This component of the Project was largely achieved in 1985 when a forest policy was prepared by the Government with the help of FAC, EEC, IDA and other development agencies. Because of the recent inception of the new forest policy, and the fact that the pricing structure is still not finalized, it is still too early to tell how effective the aeolicy and the pricing will be. c. The need for treeplanting, taking into account Burundi's keen competition for farmland, grazingland and forestland. The Project's rural nursery and plantation components were largely successful in meeting the targets in this respect.  -2- II. FORESTRY SUBSECTOR A. Resources and Consumption 2.1 Burundi's forests and plantations, which presently occupy about 7% or 200,000 ha of the country's total area, are being depleted far H more rapidly than they are being replaced, due partly to the weakness of the forest service. Wood consumption - estimates vary from 0.25 to 1.0 m3 per person per year - is primarily for fuel, especially in rural aracss, although polewood is also used there for construction of homes, and sawnwood is used in urban areas for construction and for furniture production. B. Forestry Department 2.2 The Forestry Department (Departement des Eaux et Forets) is part of the Ministry of Agriculture and Livestock (see Annexes 1 and 2). The Government allocates very little of its budget to the Forestry Department (US$190,000 in 1983). The latter is organized at provincial and arrondissement levels. Its effectiveness is hampered by lack of transport and supervision. Substantially more resources are needed if forestry objectives are to be realized. Presently, nearly all forestry investment is carried out through outside development agency financing. C. Forest Policy 2.3 The 200,000 ha of forested land in Burundi is made up of approximately 702 man-made plantations and 30% natural forests. It falls into four ownership categories: Government forests and plantations, religious communities' plantations, communal plantations and privately owned plantations. Utilization rights and maintenance duties of forestlands are often unclear, and market prices of forest products are often far below their replacement value. The Government is taking steps to address these problems. In 1980, the Government restructured the Forestry Department so as to be better able to control and enforce forest policy. Then, in 1985, with the help of FAC, the Project, and various other development agencies, the Government issued a new Burundi Forestry Code, redefined forest policy, and issued a new price structuring strategy. These documents will facilitate forest resource management in Burundi. 2.4 Charcoal is the preferred fuel of Bujumbura's population. It is normally produced and marketed by jobbers using traditional and highly inefficient techniques (yields of charcoal from wood using traditional methods - 10% on a weight basis). Bujumbura's annual demand for charcoal is estimated at 7,000 tons and traditional techniques require approximately 90,000 to 100,000 m3 of wood to meet this demand. Prices of wood commodities are difficult for the Government to control, especially because in Burundi, unlike in Kenya, the Government owns little of the country's plantations and therefore cannot control the price at the supply point (plantation), as the Kenyan Government can.  -3- D. Future Energy Needs 2.5 Estimates of Burundi's future energy needs are varied, but it is clear that the future demand for fuelvood and - though less important - construction wood cannot Le met solely from conventional plantations. The problem is that the theoretical requirement for plantations (approximately 50,000 ha in year 2,000) far exceeds both the capacity of the Forestry Department and suitable available public land resources. Thus, the size of conventional plantation projects will be determined more by what is possible than by what is desirable. For the rural population, the ideal solution would be for most households to have their own small woodlots to supply their fuelwood and construction wood needs. Assuming that the average rural household requires at least 2m3 of wood annually, then about two-tenths of one hectare (320 trees, or less if free-growing) would be sufficient to meet their needs more or less indefinitely. As for the urban population, a combination of more efficient charcoal-making and charcoal-using techniques along with alternative energy sources is probably the most realistic solution. E. Alternative Energy Sources 2.6 Peat seems to be the most attractive of the energy alternatives now known. Burundi has large unutilized peat resources which, in theory, could supply a large part of energy needs for many decades. However, considerable research and trials will be required before this potential can be exploited. Petroleum products would generally be prohibitively expensive for the rural population, but suspected large reserves within Burundi are now being investigated by Amoco. Thus, petroleum in the form of kerosene may one day prove to be an important and appropriate substitute, for the relatively wealthier urban population. However, for rural households it is difficult to envisage fuels other than wood and (with detriment to farm productivity) agricultural residues. F. Ongoing Forestry Projects 2.7 A number of forestry projects other than that financed by IDA and FAC are currently underway in Burundi. Belgium is financing an afforestation Project over 12,000 ha in south central Burundi (previously co-financed with Saudi Arabia). The EEC's European Development Fund will finance the afforestation of 3,200 ha at Rugazi and 6,000 ha near Gihofi. FAC is financing the planting of a 500 ha protection belt around the natural forest of the Zaire/Nile Divide in northern Burundi. Three IDA-financed rural development projects (Kirimiro, Ngozi III, and Muyinga) have forestry components. Most of these projects have a regional rather than a sectoral focus, and they generally concentrate on tree-planting through extension work.  -4- III. PROJECT IDENTIFICATION, PREPARATION AND APPRAISAL A. Origin 3.1 The first formal discussions for possible Bank financing for a Burundi Forestry Project took place in October 1976 (see Basic Data Sheet), when a Bank forester visited Bujumbura. An identification mission including an IDA forester and an FAO forestry consultant visited Burundi in May 1977. On the basis of the identification report, the Ministry of Agriculture and Livestock confirmed the Government's interest in the Project b" letter in October 1977 and requested Bank assistance in Project preparation. An IDA mission visited Burundi in March 1978 to assist the Government with preparation of a Project proposal. Details of the proposed investments, implementation procedures and major issues raised by the Project were reviewed with Government officials at the conclusion of the preparation mission and in letters to the Government. The Government formally requested IDA appraisal in a cable signed by the Prime Minister. The preparation mission prepared a report in Washington, but Government officials did not have an opportunity to review it prior to appraisal. B. Appraisal Issues 3.2 In reviewing the Project with senior Government officials, the following long-term forestry development issues were discussed: (a) Need for coordination of investments within the forestry sub-sector. Most of the money invested in Burundi's forest resources comes from outside agencies. (b) Need for manpower planning and training for the forestry sub-sector. The Project recognized that without trained foresters, Burundi has little hope of efficiently determining, implementing and monitoring its own forest resource policy. The Project's original intentions were to determine the number of foresters Burundi would need in the near future and then begin their tec.hnical training. (c) Lack of clear long-term forest policies, notably for pricing of forest products and control of existing and planned plantations. (d) Land use for forestry development. The keen competition for farmland and grazingland necessarily leaves forestland as a lower priority amongst the rural population.  -5- C. Targets and Goals 3.3 There were no substantial changes between the original Project proposal and the final Staff Appraisal Report. The Project was to be the first stage of Burundi's long-term program for the development of basic forestry services and plantations supplying fuelwood, building poles and timber, and establishing long-term production and marketing policies for its forestry sub-sector. The Project proposed in the Staff Appraisal Report comprised the following components, to be implemented over five years: (a) The establishment of nurseries in 30 of the 79 Communes of Burundi. These nurseries, which would be managed by the Forestry Department, would produce high quality seedlings of eucalyptus and other species to be sold to communal organizations or private farmers. The staff assigned to the nurseries would strengthen existing forestry extension services at the communal level by providing technical advice on, inter alia, planting and maintenance of seedlings. (b) Development of 2,000 ha of eucalyptus near Bujumbura as short-rotation plantations to supply charcoal, firewood and building poles for the population of Bujumbura. (c) Development of a 5,000 ha pine plantation near Bururi in the south to produce saw timber. (d) Training of Burundi forestry staff, principally through courses for higher and middle levels. (e) Technical assistance for Project management and on-the-job training for Burundi counterpart staff. (f) Studies and applied trials, notably a study of Burundi's energy use and needs, trials of more efficient charcoal production techniques and development of new designs for cooking stoves for the rural population. D. Board Approval 3.4 The Board approved the Project without discussion on May 24, 1979, and the Development Credit Agreement (918 BU) was signed between IDA and the Burundi Government in June 1, 1979. IDA, FAC, EEC, UNDP and the Burundi Government shared the costs of this US$8.83 million Project (see para. 1.1). The Project's actual costs equalled its estimated costs. However, the allocation to cost categories of the US$4.3 million IDA Credit was slightly modified during implementation. Distribution of Project costs are discussed more fully in para. 5.05.  -6- E. Benefits 3.5 The Project's principal production benefits were in the SAR seen to be seedlings for plantations in 30 communes (it was hoped that thereafter the Forestry Department would create similar rural nurseries in the remaining 49 communes), eucalyptus for fuelwood and building poles (the first harvest to be about year 9), and pine timber production (beginning about year 12, but major production not beginning before year 25). Other expected benefits would be training of forestry staff, strengthening existing institutions and assisting in development of clear long-term pricing and exploitation policies for forest resources. In the short-term, there were direct benefits to the rural population through the Project nurseries enabling them to develop their own plantations, and through increased employment opportunity in areas chosen for Project plantation development. Long-term benefits included increased soil fertility and soil stabilization and, especially for the urban population, lower costs of fuelwood and construction wood. F. Risks 3.6 The Project was relatively simple, appropriately addressing Burundi's forestry sub-sector needs. It was designed to require little reliance on non-project administrative services or farmers' involuntary support. Instead it appealed to the farmers' self-interest, either by paying them for their labor, or by providing them with Project-produced seedlings, i.e. it provided incentives to farmers. No unusual project risks were identified. Delays arising from material delivery problems would have no significant impact upon the Project's eventual benefits. The principal risk related to the rural nursery component was the uncertainty of whether farmers would pay for the seedlings. The Goverrment had been giving its agricultural seedlings (coffee, tea etc.) away free. During implementation, the Project worked with the Government to require payment for all seedlings, and it found the rural population quite willing and able to pay, especially when the Project increased its production of multiple-use and fruit tree species. Another risk identified was the possible lower-than-anticipated mean annual increment (MAI) of the plantations. Because soil conditions are relatively poorly known in Burundi, initial estimates were difficult to determine. During implementation, it appeared that the low soil fertility on large parts of the plantations will result in yields slightly lower than originally projected. However, through these experiences a better knowledge of local soil conditions and growth rates has been attained, which allows more accurate planning for future forestry investments.  -7- IV. IMPLEMENTATION A. Dates of Effectiveness and Completion 4.1 The Project Development Credit Agreement (918 BU) was signed in June 1979. The original completion date was September 1984. However, due to the late arrival of the three technical assistants, and their consequent inability to take advantage of the 1979/80 planting season, the completion date was extended to June 1985. This extension gave the Project five full planting seasons with which to work. B. Main Components Rural Nurseries 4.2 The Project's objective was to sell 60% of the seedlings to individuals and 402 to communes. Survival rates are generally much higher in individual woodlots (about 80% average survival) than in communal plantations (about 20Z). However, during the first year, only 19% of the seedlings were sold to individuals. Also, the production cost during the first year was far above the selling price of seedlings. Fortunately, the Project was flexible enough to make changes in the implementation methods suggested in the SAR. The Project Manager took effective steps to decrease production cost, increase selling price, and increase percentage of sales to individuals. The Project Manager's strategy was as follows: (a) increase revenues by diversifying production away from solely eucalyptus, to sell more pine, cypress, multiple-use and fruit tree species; (b) decrease transport costs to the consumer by creating a greater number of smaller nurseries; (c) decrease production costs by shifting from purely potted stock to over half bare- root stock in the nurseries; (d) increase the emphasis on the information and extension programs in the rural areas; and (e) work with the Government to have them sell rather than give away seedlings produced by the Ministry of Agriculture and Livestock. These modifications were successful, and resulted in higher survival and higher cost recovery. 4.3 According to the SAR, the development of rural nurseries capable of generating enough income to meet their production costs was to be one of the Project's most important indicators of long-term success. By the close of Phasa I, the Project nearly had the selling price in line with the production cost. If these rural nurseries become economically self-sufficient, private nurseries can be created and the rural population's access to tree seedlings will be improved. 95% of the Burundi population lives in rural areas and has little cash to pay for its fuelwood, which traditionally was cut or collected by its consumers without payment. Thus, the rural nursery program represented an appropriate attempt to make the rural population self-sufficient in  -8- fuelvood by making seedlings available at affordable prices. Since the rural nurseries are small (producing 40,000 to 60,000 seedlings per annum) and depend upon the surrounding population for their sales, they respond quickly to changes in demand for particular species. Also, the size and location of the nurseries enable the nurserymen to develop an effective extension system to directly respond to the surrounding population's concerns and strategies in meeting their energy and other forest product needs. 4.4 Originally the Project expected to create one nursery for each of 30 communes chosen from among the 79 communes in the country. Each nursery was to produce 100,000 eucalyptus seedlings per annum (E. grandis, E. tereticornis anc E. camaldulensis) and a few ornamantals. During implementation, project management found that it would be more appropriate to establish a denser network of smaller nurseries, so, with the concurrence of IDA, it actually established 81 nurseries of 40,000 to 60,000 seedlings each. It is hoped that as these nurseries become economically more self-sufficient, the program will continue to expand until all of Burundi's population has access to seedlings. 4.5 As for the types of seedlings plantedt although the Project found the farmers quite willing to pay for seedlings, they wtere more selective than anticipated. Consequently, instead of supplying only eucalyptus seedlings, the Project supplied pine, cypress and multiple- use species such as Grevillea, Acacia, Leucaena, Albizia and fruit trees. These products sold at higher prices than the eucalyptus. Tablo 1: ApDraisal Estimates v. Actual Achlvwment: Seedlinas YR 1 YR 2 YR S YR 4 YR 6 TOTAL APR /IXCTLI m ART IPMFCT WPFR T IF`'T APWR ACT Total no. of nurserles 10 21 20 46 80 46 80 82 80 81 J0 81 Ave. no. of *edlinge/nursery ('000) 60 25 60 84 100 49 200 88 100 85 - - He of communal plant tlon / 100 170 800 856 500 600 600 700 600 890 2100 2716 Na of lndividuals' plantatronn /2 150 60 450 177 750 280 900 258 900 881 8160 1098 Total ha 260 220 750 582 1250 880 1600 9608 1500 1221 6260 8811 LI APR u Appraised; ACT a Actual. /2 Assuming spacing of 2.6 a x 2.6 a (1,600 trues/ha) for communal plantations and 2.0 a x 2.0 m (2,006 tree for Individuals" plantations.  -9- Table 2: Other Actual Achlvements (SeedlInc) YR 1 YR 2 YR a YR 4 YR 6 Total-or Averago No. of *-edling* produced ('000) 654 1,585 2.288 2,028 2,849 9,284 X sold to commun*s 64 87 46 56 80 48 (ave.) X sold to Individuals 19 88 26 26 28 26 (ave.) X seedlings lost, stolen and unsold 28 80 80 19 26 26 Ave. cost/seedling (Fu) La 18.49 6.29 6.59 7.77 6.69 8.20 (ave.) Av. cost/seedling (FBu) /b - - - 5.70 6.70 5.70 (ave.) Ave. selling price/seedling (F8u) 1.0 1.6 1.6 2 Eue. 2 Euc. 2 Euc. 8 Pine 8 Ping 8 Pines 6 Fruit 5 Fruit 6 Fruit La Includes salaries during off seaton for permanent staff excluding expatriates. Lb Excludes off season costs for permanent staff. Not available for yers 1 to 8. Table 3: Species Percentages Species YR 1 YR 2 YR 3 YR 4 YR 5 Eucalyptus 93% 88S 87% 761 59% Grevillea and other multiple-use species 7% 122 13S 242 402 Fruit trees 02 02 02 02 12 Eucalyptus Plantations 4.6 The SAR proposed a 2,000-ha short-rotation eucalyptus plantation to supply the population of Bujumbura with firewood, charcoal and construction poles. The principal species suggested was E. grandis, and selected high quality seeds, unavailable in Burundi, were imported. The principal technical aspects of the plantation development were implemented as outlined in the SAR: manual ground preparation, seeding in one or more nurseries, manual planting and weeding. Plantations are scheduled to be coppiced at eight-yearly intervals with one seedling and three coppice rotations.  -10- 4.7 The SAR-estimated growth was as follows: Yr. Yield (m3/ha) MAI (m3lha) 8 (seedling rotation) 120 15 16 (coppice rotation 1) 120 15 24 (coppice rotation 2) 96 12 32 (coppice rotation 3) 80 10 The total MAI was thus estimated at (120 + 120 + 96 + 80)/32 - 13 m3 /ha. 4.8 The Project created just over the targeted 2,000 ha of eucalyptus plantations, with two slight modifications to the SAR. The first modification was in the site. The proposed site did not, in fact, measure a full 2000 ha - there had been a cartographic error during preparation. Also, the land in the proposed site turned out to be rockier and poorer than originally supposed. Consequently, three separate areas -- Mageyo (the original site, 30 km from Bujumbura), Gakara (55 km from Bujumbura) and Ryarusera (near Bugarama) -- had to be planted to achieve the planned total of 2000 ha. In choosing these three sites, the Project held to its decision not to select presently utilized farmlands so as to avoid land-use conflicts, although that decision resulted in placing the plantations on slightly marginal lands. 4.9 The second modification was that planting was not restricted to E. grandis, but included other eucalypts and a small area of pine. The principal species was E. grandis which grows fastest and gives good results at elevations between 1,400 to 2,200 m. E. tereticornis and E. urophylla were planted at the lower and middle elevations, and Pinus caribaea and P. kesiya were chosen for the poorest sites. The eucalyptus plantation component also achieved its targets for indirect afforestation activities: 57 km of new roads were opened and 208 km of existing roads were maintained; 98 km of firebreaks were constructed; 4 permanent buildings, 2 lodgings, 1 home/office, 1 store room, 8 traditional lodgings and various diverse service buildings were constructed (Key Indicator Chart, Annex 3). 4.10 The final results of the eucalyptus plantation will not, of course, be available until the final cutting. It should be noted, however, that the surface area target was met, the 95Z survival rate is better than originally estimated, and that appropriate species for each altitude and soil type have been identified. On the other hand, because three separate sites had to be used, a more extensive road network had to be built, along with the construction of more administrative buildings than the SAR targeted. The production rates will probably be slightly lower than expected because the soils are worse than originally estimated.  -11- 4.11 Plantation maintenance: the importance of initial weeding was underestimated. While such weeding increases production cost, the Project Manager felt that the optimum schedule would include two to three weedings the first year and one to two the second year. The increased growth rate of the trees due to weeding (reduced competition for water, nutrients and sunlight) apparently compensated for the additional cost of weeding. The high cost of fertilization, on the other hand, appears justified only if the treatment is used as a substitute for later weedings, permitting quicker cro.ni m-losure and associated weed suppression. Table 4: Eucalyptus Plantation: Surface Area Planted (ha) YR 1 YR 2 YR 3 YR 4 YL, 5 Total SAR target 100 00 400 600 600 2,000 Mageyo 200 195 10 - - 405 Gakara - 125 633 556 170 1,484 Ryarusera - - 152 152 Actual total 200 320 643 556 322 2,041 The SAR estimated 200 ha (1OZ) to be fertilized and 1,600 scedlings/ha to be planted. Actually 495 ha (24Z) were fertilized and 1,375 to 1,500 seedlings/ha were planted. Pine Plantation 4.12 The goal of a 5,000-ha pine plantation was not quite met: 4,034 ha were realized. The species used, the manual ground preparation and planting, and the housing and road construction were all implemented more or less as planned. There were, however, three main issues which arose during the implementation. First, there was a lack of laborers in the area (Bururi) when the Project began. After the first two years, however, enough laborers were attracted to the sites by increased wages and by supplementary food rations supplied by the World Food Programme jWFP). Second, the original site had less than the expected 5,000 ha of productive land, due to steep slopes and poor soils. The result is that about 251 of the plantation cannot be economically harvested and thus is classified as a 'protective'. While the pines will indeed add protection to the sites, the intention was to provide sawtimber. On these steep hillsides a less-costly investment than pine plantations would have been preferable, if possible using tree and shrub species more valuable as fodder and soil protection, and less inflammable. However, at present, the only species that bear up on the "protection forestry" sites of Vyanda are Callitris and Pinus eliottii. No multiple-use species suitable to these conditions are yet known. The last issue, land use conflicts, ties in with the second issue. The appropriation of the 5,000 ha of productive land was made difficult because herders saw the plantation land as their own. The herders have not vandalized any part of the plantation, but it is important that the Project resolve this land use conflict. The project is attempting to integrate grazing into some of the pine plantations. This is a good solution which should reduce weeding costs, maximize economic return  -12- from the plantation, *nd way encosrage good Project-herder relations. Phase II of the Project will continue to work towards the development of an economic and socially acceptable livestock-forestry program for the plantations. If such a program can be developed, it will become much easier to locate more land suitable for possible future pine plantation. In the SAR, par&. 3.06, it was expected that ultimately about 25,000 ha of pine would be planted in the Bururi region of southern Burundi. In retrospect, this *sems most unlikely, considering the scarcity of land and labor in this mainly cattle-raising region. Such a large area would in any case not be needed to sustain a sawmill. Even the 751 of the 4,034 ha which in the first phase were planted on harvestable sites can be expected to eventually yield about 18,000 m3 of sawlogs p.a., which is enough to foed quite a siseable sawmill. 4.13 The pine plantation component met Its indirect afforestation targets by opening 89 km of forest roads, 290 km of firebreaks and constructing six permanent :uildings, several lodging quarters, a workshop, one store room, one office, 16 locally constructed lodging quarters, four camps and three nursery offices (Key Indicator Chart, Annex 4). 4.14 Table 5: Pine Plantation: Targets and Results YR i YR 2 YR 3 YR 4 YR 5 Total Ha planned (SAR) 300 700 1,000 1,500 1,500 5,000 Ha planted (actual) 220 820 980 1,016 998 4,034 No. seedlings, '000 (SAR) 420 980 1,400 2,100 2,100 7,000 See Borrover'sNo. seedlings '000 (actual) 245 786 987 1,030 1,223 4,277 Comment Ha fertilized (actual) 30 45 169 325 - 567 The planting density varied from 1,000 seedlings per ha on poor soils to 1,250 seedlings per ha on the better sites. Fertilization took place on 141 of the plantation. The survival rate was 952, which was very good, and higher than the expected 902. Table 6t Species Distribution of Pine seedlings ('000) Total YR 1 YR 2 YR 3 YR 4 YR 5 No. P. patula 213.4 592.7 695 539.7 672 2,712 63 P. kesiya - 38.2 164.8 151.8 152.7 507 12 P. elliottii 31.6 122.4 79.9 101 93.6 429 10 P. caribaea and oocarpa - 32.7 32.9 237.5 227.3 530 12 Other - - 14.7 77.4 92.1 3 Total 245.0 786.0 987.0 1030.0 1223.0 4271.0 100  -13- Training Component 4.15 The SAR intended the project to train personnel at all levels. Eighty middle level agriculture-forestry technicians ("agronomes forestiers") were trained at ITAB between 1979 and 1984 with UNDP/FAO-paid teachers, as part of the Project. The training program Was designed by a UNDP specialist i-_ cooperation with Government officials, and was supervised by the Director General of the Ministry of Agriculture. The number of technicians thus trained corresponded to FAO's estimate of needs, and to the Project target. Only one high-level forester was trained under the Project (para. 4.17a below), but (a) there now exists a number of middle-level foresters from whom suitable candidates can be selected for further training, and (b) independently of the Project, the Institut Superieur d'Agriculture (ISA) in 1983 began professional training in Burundi of graduates in environmental management, including forestry. 4.16 The reason why only one high-level forester was trained under the project, was that qualified school-leavers were not interested in taking up the scholarships offered for studying forestry at the university level abroad. In any case, the Project's original objectives were too ambitious: it was unrealistic to expect that enough high-level foresters could be well enough trained in only five years to take over the project. This is especially apparent when one remembers that at the Project's inception Burundi had only one trained forester. 4.17 The following are the results of the training component: (a) Two scholarships for study outside Burundi: 1 scholarship of two years to Madagascar, professional level 1 scholarship of one year to Addis Ababa, technician level 4 other scholarships were available but lost due to lack of candidates (b) One training session to Kenya and Madagascar for three of the Project's forestry agents. (c) The Agricultural Technical Institute of Burundi (ITAB) trained 80 agro-forestry technicians. The first two classes received two years of forestry training (besides their agricultural training), the later classes three. (d) ITAB created a four-year professional resource management program. In light of the shift towards agro-forestry and livestock-forestry integration, this program is perhaps a more appropriate schooling for Burundi's foresters because of its integrated land-use techniques.  -14- (e) Short-term sessions: (i) ITAB students trained under Project supervision during school breaks. (ii) All levels of Project personnel received in-the- field training for numerous aspects of Project operations. (iii) Periodic training sessions for Project personnel of similar duties to encourage dialogue and promote the most efficient techniques. (f) Short-term study tours. Project personnel toured forestry projects in Tunisia, Rwanda, and China. Studies and Trials 4.18 The SAR recognized the importance of three principal studies: (a) to determine Burundi's Projected energy needs until the year 2000; (b) to develop more economical and efficient cooking stoves especially for the rural population; and (c) to determine a more efficient means to produce charcoal modeled after a method now widely used in Kenya and Uganda. 4.19 The first study consisted of two surveys in rural areas and one survey in an urban area. From 1979 to 1983 little progress was made in the study, and the subsequent work done by the Deputy Director of the Government statistical office proved too difficult to interpret and thereby of no use in the formulation of the 1985 forestry policy. Phase II should either reconduct the study through the Government Statistical Office or hire an outside consultant team to produce an accurate, useful estimate of Burundi's energy needs. 4.20 The rural stove trials were entrusted to International Association for Rural Development (AIDR). After manufacturing 1,000 wood-burning clay stoves all of which cracked, AIDR abandoned the study, concluding that the Burundi clay is unsuitable, regardless of the type of kiln used. Another IDA Project, Urban Development of Buiumbura (DUB) is developing a fuel saving charcoal stove for upper income arban groups, and USAID is developing a peat-burning stove, mainly for institutions like schools, hospitals etc. As the Project's stove component for the rural population went poorly, it was not continued in the second phase. The Ministry of Energy was deemed a more suitable agency to carry out this type of work than the Ministry of Agriculture. However, the Project should offer its resources to NGOs and/or other groups and aid them in the dissemination of appropriate wood-burning stoves for rural areas. The Project already has a significant infrastructure established in several rural areas into which other stove projects could feed to facilitate the dissemination of new stove technology in the countryside. Such programs might evolve around the Project's rural nursery progrsm, for example. (In Niger a rural stove Project successfully disseminated the new technology by "paying" rural stove builders with seedlings instead of cash.)  -15- 4.21 Charcoal-making trials using imported steel to manufacture an improved metal kiln were carried out in 1984 by an FAO consultant. The trials were very successful and the consultant trained a few charcoal- makers in this and other techniques. The most popular method, the "casa-ance" method, requires little capital investment, is very similar to traditional techniques, and increases charcoal yields by 202. 4.22 The Project also funded a feasibility study for the Kayongozi protection afforestation project. Because it would have a negative financial rate of return, it was at the time deemed unsuitable for Bank financing, but the economic return was not estimated. Proiect Management Support 4.23 When the project began, it shared one office with another project, and the Forestry Department in general was housed in very cramped quarters. During Phase I the project financed the construction of buildings for Project headquarters at Bujumbvra, which are today part of a functional new Forest Department headquarters, as foreseen in para. 3.07 of the SAR. C. Theft 4.24 Two cases involving money losses occurred. Firstly, approximately US$20,000 was stolen in a hold-up of a Project vehicle carrying monthly salaries. Reasonable prudence in delivering the money had been observed. Secondly, paysheet irregularities occurred over a two-month period on one of the plantations. This resulted in the prompt d&smissal of three staff members. D. Revolving Fund 4.25 Under Section 3.01 (b) of the Development Credit Agreement (918 BU), the Government was to provide a revolving fund with a value of US$200,000. This was done. The Government, however, failed to replenish the fund every three months as agreed, occasioning late payment of salaries or delays in purchase of local materials. During implementation, the Project created a foreign currency revolving fund or special account, using the US$200,000 previously unallocated under Category 5 of the IDA Credit 918 BU (para. 5.25). IDA replenished this fund regularly and thereby maintained a liquid cash reserve for the IDA reimbursenent of local purchases and salaries.  -16- V. OPERATING (BANK. FINANCIAL. INSTITUTIONAL) PEWFORMANCE A. SuHprvision 5.01 The Bank's performance in the supervision of the Project was satisfactory. Missions to Burundi vwre conducted approximately every six month. thereby lsimting the effect of any problms. The working relationship between the Bank and the Borrower was good. The Bank's flexibility in allowing the Project Manager to modify the rural nursery component was a significant factor in that component's success. The accounting and auditing procedures were effective. The Project was relatively simple in design, and strong support by the Government and the Forest Department resulted in a relatively problem-free and successful Project. B. Compliance with Credit Terms 5.02 The Borrower satisfactorily complied with the terms of the Development Credit Agreement. The creation and maintenance by IDA of a second rovolving fund (under section 3.01 (b) of the Development Credit Agreement 918 BU) worked out well for the implementation period of the Project. During Phase II, the Project should work to ensure that the Borrower will be able to maintain a liquid cash roserve sufficient to cover daily operating expenses once IDA funding support ends. C. Costs and Procurement 5.03 The Forest Department within the Ministry of Agriculture, Livestock and Rural Development was the executing agency. It obtained and accounted for funds and materials without major issue. During pre- Project negotiations, it was agreed to allocate EEC funding specifically towards civil works expenses in order to allow tho Project to purchase vehicles and materials from non-EEC countries. This modification saved the Government money and time in procurement, delivery and maintenance of vehicles and materials. 5.04 The actual costs of the Project equalled the estimated costs and totalled US$8.83 million. The costs were shared as follows: FBu US$ Major S (million) (million) Component IDA 49 390.0 4.30 Other FAC* 21 167.4 1.86 Technical assistance EEC 13 108.0 1.20 Buildings and studies UNDP 8 63.6 0.70 Training Govt. of Burundi 9 72.4 0.77 10% of local expenditures Total 100 801.4 8.33 * Comprising FF 9.4 million from FAC itself and FBu 16 million from AFVP (French Volunteers for Progress Association).  -17- 5.05 The allocation of the US$4.3 million IDA Credit was modified as follows in 1984: US$ US$ Category (Appraisal) (Actual) I. 100% of foreign exchange 900,000 940,000 expenditures or 90% of local expenditures for vehicles, equipment and materials II. 90% of total expenditures 1,800,000 2,250,000 for civil works including road works and direct cost of plantation development III. 90% of total operating costs 800,000 810,000 of Project implementation IV. 100% of foreign exchange 300,000 100,000 expenditures or 90% of total experditures for Project preparation work and monitoring and evaluation V. Unallocated 200.000 200.000 * Total 4,300,000 4.300.000 * This amount was assigned to a second Revolving Fund Account (para. 5.27). This fund was allowed to deplete to zero by completion date in order to utilize all US$4.3 million allocated.  -18- D. Expatriates 5.06 The expatriate experts from CTFT were largely responsible for the success of the Project due to their aboility to maintain good dialogue between the Government, Project staff and IDA. The use of the four French "volontaires du progres was highly effective, and the FAO consultant hired for the charcoal-making study worked well. VI. ECONOMIC RE-EVALUATION 6.01 The actual costs of the eucalyptus and pine plantation components - the only ones to be subjected to economic analysis in the SAR - were somewhat higher that anticipated, due to dispersed planting sites in the case of the eucalypts and higher than expected labor costs in the case of the pines. Also, the average increment is somewhat lower, due to poorer than expected land in the case of the eucalypt plantations and steeper than expected land (non-harvestable) in the case of the pine plantations. The overall rate of return of these two components is therefore somewhat lower than estimated in the SAR. 6.02 The rates of return estimated in the SAR and in the present PCR are as follows (details are set out in Annex 6 below, and especially in the Table Notes which follow it): Eucalyptus Pine Total plantations plantations SAR 17% 22% 20% PCR 11% 17% 15% Although the above re-evaluated economic rates of return, varying from 11 to 17 percent, are merely estimates - only 7 years have elapsed of the 37-year period of analysis, and harvesting in the project plantations has not yet begun - the indications so far are that the project will be worth while from the economic point of view. 6.03 As regards the project's effect on income distribution, it can be expected that most of the benefits will go the lower income groups, because: (a) unskilled rural labor for plantation establishment was a major cost item; (b) fuelwood, charcoal, and building poles, which are all related to basic needs, will be some of the major project outputs. This is even more the so in the case of the rural treeplanting component, the rate of return of which was not calculated in the SAR or here. VII. ENVIRONMENTAL IMPACT 7.01 The Project had several beneficial environmental impacts. The trees planted within the framework of the pine plantation and the rural nursery component improved protection against wind and water erosion of  -19- the soil. The poles and timber produced from the plantations will soon be available which will decrease the need of cutting the increasingly rare local species. Also, the increased fuelwood supply will decrease household use of agricultural residue for fuel, thereby improving soil fertility and increasing crop yields. 7.02 On the other hand, the new roads established by the Project will result in increased soil erosion, especially on the steeper slopes, if they are not properly maintained. And there is also a risk that the eucalyptus plantations impede herb growth as has happened in several other countries. The thin canopy cover naturally formed by eucalypts, combined with the scant herbaceous soil cover, can actually result in increased soil erosion. VIII. ISSUES AND RECOMMENDED FOLLOW-UP 8.01 Overall, the Project represented an important first step in developing a long-term program in Burundi to create basic forestry services and begin more rational forest management. The Project began Phase II in 1985, and it is vital that the implications of each component of Phase I are recognized. A. Rural Nurseries 8.02 The selling price of the seedlings is approaching the production cost. The implications of Phase I for Phase II are the following: (a) Coatinue work to decrease seedling production costs by decreasing thefts and gifts of seedlings, through closer monitoring of sales. Methods to give more responsibility and increased incentives for nurserymen to run profitable nurseries should be investigated. (b) Increase income to the nurseries by growing higher percentages of the more valuable multiple-use and fruit tree species. (c) Continue monitoring of consumer demands and expand extension services further into rural areas. (d) Assimilate new discoveries of appropriate agro-forestry and livestock-forestry techniques into the extension programs to encourage economically, socially and ecologically sound multiple-use strategies.  -20- B. Eucalyptus and Pine Plantations 8.03 The implications of Phase I for Phase II are the following: (a) Increase the development of a livestock-forestry program for the plantations. If successful, this will allow plantations to be created on better sites without co-.flict between the Project and traditional user groups (herders). (b) Don't reforest lands where reforestation is uneconomic. On those lands which are so steep that they obviously will be uneconomical to harvest, no known fodder species or soil/site protective shrub/herb species will survive. (c) For the pine plantation, determine ard establish the minimum number of harvestable hectares necessary to support a sawmill. (d) Monitor the MAI of the plantations, and on the basis of the new figures re-estimate from time to time the rate of return, to assess the justification for additional plantations. C. TraininR 8.04 The Project was successful in training Burundi's first middle level foresters, but not in the training of high level foresters. Phase I has the following implication for Phase IIt (a) Increase the emphasis on training high level foresters and incorporate into their training short and long-term visits to neighboring countries' forestry Projects. Also, incorporate management, administrative and specialized business training into the program for higher level foresters. Ensure that every expatriate in Phase II has a suitably qualified Burundian counterpart who participates fully in project work. D. Studies 8.05 The Project did not achieve its objectives with regards to the study program except in the charcoal techniques study. The implications for Phase II are the following: (a) Incorporate the new charcoal-making techniques into the eucalyptus plantation harvests. (b) Continue to disseminate information on the use of the new charcoal-making techniques. (c) Help NGOs and other organizations disseminate and monitor their newly developed cooking stoves. Encourage these organizations to work towards developing a more efficient, low-cost and easily fabricated rural wood burning stove.  -21- (d) Reconduct the study to determine Burundi's future energy needs using either the Government Statistical Office or an outside consulting agency. IX. BURUNDI FORESTRY PROJECT PHASE II 9.01 The Burundi Forestry Project Phase II continues much of the work of Phase I. The goals and implementation methods for Phase II reflect the lessons learnt from Phase I. The purpose of Phase II is to: (a) speed up reforestation attempts on individual farms while reducing the development of new large-scale plantations (because of the land availability problems experienced during Phase I); (b) continue development of viable agro-forestry and sylvo-forestry techniques; (c) strengthen the Forestry Department's capacity to manage investment programs and staff training, and to train charcoal-makers in the more efficient techniques developed during Phase I; and (d' continue development of long-term pricing strategies for forest products. 9.02 Phase II will attempt to accomplish the above goals over five years (1986-90) through the following components: (a) Development of 90 Government-run nurseries (an average of more than one per commune) .icluding maintenance of 60 existing nurseries and establislhment of about 30 new ones, each producing 40,000 to 50,000 seedlings (3 to 4 million altogether) per year; development of forestry extension scrvices and a pilot agro-forestry scheme; and establishment of up to 2,000 ha of demonstration fuelwood, pole, fodder and fruit plots in communes throughout the country (up to 5 ha per year close to each rural nursery). (b) Establishment by the Government of: (i) 3,500 ha of eucalyptus fuelwood plantations; and (ii) 1,500 ha of pine sawlog plantations. (c) The protection, maintenance and exploitation of existing plantations established under Phase I. (d) The initiation of a program to promote integrated sylvo- pastoral activities in a few selected plantation areas with high population densities. (e) Institutional strengthening of the Forest Department through the creation of a planning and mon.itoring unit and two new regional forest inspectorates including support for additional staff, buildings, equipment and running costs and expatriete technical assistance and strengthening of the Project management unit.  -22- (f) An agro-forestry training program including foreign scholarships for foresters at university level and local training, and studies, trials and training consultancies in agro-forestry, charcoal making, forest exploitation and wood utilization. ANNEXES Annex 1 - Project Organization Chart Annex 2 - Key Indicator Table Annex 3 - Map of Project Area Annex 4 - Economic Rate of Return Re-evaluation Annex 5 - Files (Documents issued)  UNGANIGRU DU PRWD FlIET WE I NINISTF DB L'AhOI- |JLTRUE ET DE L'ZZVAA DIRECTION GE1ALE DE L'ADRICULIURU DIUCEC DU DEPA PM \ ~DMi SM ET sOE- PII~I ID )R_M-AOIL IRECTEllR DU RJ | / COfT.13ILIS 4 , \ \ | Vol. du Progrs Norno Agronoge AgronaDme Agronome Agroname V. du Progr0 V. Progtbs Resp. hoa:.pti Aos istant NoniteurX Capita + 115